Deep Learning(DL)model has been widely used in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR)and has achieved excellent performance.However,the black-box nature of DL models has been the f...Deep Learning(DL)model has been widely used in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR)and has achieved excellent performance.However,the black-box nature of DL models has been the focus of criticism,especially in the application of SARATR,which is closely associated with the national defense and security domain.To address these issues,a new interpretable recognition model Physics-Guided BagNet(PGBN)is proposed in this article.The model adopts an interpretable convolutional neural network framework and uses time–frequency analysis to extract physical scattering features in SAR images.Based on the physical scattering features,an unsupervised segmentation method is proposed to distinguish targets from the background in SAR images.On the basis of the segmentation result,a structure is designed,which constrains the model's spatial attention to focus more on the targets themselves rather than the background,thereby making the model's decision-making more in line with physical principles.In contrast to previous interpretable research methods,this model combines interpretable structure with physical interpretability,further reducing the model's risk of error recognition.Experiments on the MSTAR dataset verify that the PGBN model exhibits excellent interpretability and recognition performance,and comparative experiments with heatmaps indicate that the physical feature guidance module presented in this article can constrain the model to focus more on the target itself rather than the background.展开更多
BACKGROUND To investigate the preoperative factors influencing textbook outcomes(TO)in Intrahepatic cholangiocarcinoma(ICC)patients and evaluate the feasibility of an interpretable machine learning model for preoperat...BACKGROUND To investigate the preoperative factors influencing textbook outcomes(TO)in Intrahepatic cholangiocarcinoma(ICC)patients and evaluate the feasibility of an interpretable machine learning model for preoperative prediction of TO,we developed a machine learning model for preoperative prediction of TO and used the SHapley Additive exPlanations(SHAP)technique to illustrate the prediction process.AIM To analyze the factors influencing textbook outcomes before surgery and to establish interpretable machine learning models for preoperative prediction.METHODS A total of 376 patients diagnosed with ICC were retrospectively collected from four major medical institutions in China,covering the period from 2011 to 2017.Logistic regression analysis was conducted to identify preoperative variables associated with achieving TO.Based on these variables,an EXtreme Gradient Boosting(XGBoost)machine learning prediction model was constructed using the XGBoost package.The SHAP(package:Shapviz)algorithm was employed to visualize each variable's contribution to the model's predictions.Kaplan-Meier survival analysis was performed to compare the prognostic differences between the TO-achieving and non-TO-achieving groups.RESULTS Among 376 patients,287 were included in the training group and 89 in the validation group.Logistic regression identified the following preoperative variables influencing TO:Child-Pugh classification,Eastern Cooperative Oncology Group(ECOG)score,hepatitis B,and tumor size.The XGBoost prediction model demonstrated high accuracy in internal validation(AUC=0.8825)and external validation(AUC=0.8346).Survival analysis revealed that the disease-free survival rates for patients achieving TO at 1,2,and 3 years were 64.2%,56.8%,and 43.4%,respectively.CONCLUSION Child-Pugh classification,ECOG score,hepatitis B,and tumor size are preoperative predictors of TO.In both the training group and the validation group,the machine learning model had certain effectiveness in predicting TO before surgery.The SHAP algorithm provided intuitive visualization of the machine learning prediction process,enhancing its interpretability.展开更多
Forecasting landslide deformation is challenging due to influence of various internal and external factors on the occurrence of systemic and localized heterogeneities.Despite the potential to improve landslide predict...Forecasting landslide deformation is challenging due to influence of various internal and external factors on the occurrence of systemic and localized heterogeneities.Despite the potential to improve landslide predictability,deep learning has yet to be sufficiently explored for complex deformation patterns associated with landslides and is inherently opaque.Herein,we developed a holistic landslide deformation forecasting method that considers spatiotemporal correlations of landslide deformation by integrating domain knowledge into interpretable deep learning.By spatially capturing the interconnections between multiple deformations from different observation points,our method contributes to the understanding and forecasting of landslide systematic behavior.By integrating specific domain knowledge relevant to each observation point and merging internal properties with external variables,the local heterogeneity is considered in our method,identifying deformation temporal patterns in different landslide zones.Case studies involving reservoir-induced landslides and creeping landslides demonstrated that our approach(1)enhances the accuracy of landslide deformation forecasting,(2)identifies significant contributing factors and their influence on spatiotemporal deformation characteristics,and(3)demonstrates how identifying these factors and patterns facilitates landslide forecasting.Our research offers a promising and pragmatic pathway toward a deeper understanding and forecasting of complex landslide behaviors.展开更多
The application of machine learning in alloy design is increasingly widespread,yet traditional models still face challenges when dealing with limited datasets and complex nonlinear relationships.This work proposes an ...The application of machine learning in alloy design is increasingly widespread,yet traditional models still face challenges when dealing with limited datasets and complex nonlinear relationships.This work proposes an interpretable machine learning method based on data augmentation and reconstruction,excavating high-performance low-alloyed magnesium(Mg)alloys.The data augmentation technique expands the original dataset through Gaussian noise.The data reconstruction method reorganizes and transforms the original data to extract more representative features,significantly improving the model's generalization ability and prediction accuracy,with a coefficient of determination(R^(2))of 95.9%for the ultimate tensile strength(UTS)model and a R^(2)of 95.3%for the elongation-to-failure(EL)model.The correlation coefficient assisted screening(CCAS)method is proposed to filter low-alloyed target alloys.A new Mg-2.2Mn-0.4Zn-0.2Al-0.2Ca(MZAX2000,wt%)alloy is designed and extruded into bar at given processing parameters,achieving room-temperature strength-ductility synergy showing an excellent UTS of 395 MPa and a high EL of 17.9%.This is closely related to its hetero-structured characteristic in the as-extruded MZAX2000 alloy consisting of coarse grains(16%),fine grains(75%),and fiber regions(9%).Therefore,this work offers new insights into optimizing alloy compositions and processing parameters for attaining new high strong and ductile low-alloyed Mg alloys.展开更多
Low-temperature hydrogenation of silicon tetrachloride(STC)is an essential step in polysilicon production.The addition of CuCl to silicon powder is currently a commonly used catalytic method and the silicon powder act...Low-temperature hydrogenation of silicon tetrachloride(STC)is an essential step in polysilicon production.The addition of CuCl to silicon powder is currently a commonly used catalytic method and the silicon powder acts as both a reactant and a catalyst.However,the reaction mechanism and the structure-activity relationship of this process have not been fully elucidated.In this work,a comprehensive study of the reaction mechanism in the presence of Si and Cu_(3)Si was carried out using density functional theory(DFT)combined with experiments,respectively.The results indicated that the ratedetermining step(RDS)in the presence of Si is the phase transition of Si atom,meanwhile,the RDS in the presence of Cu_(3)Si is the TCS-generation process.The activation barrier of the latter is smaller,highlighting that the interaction of Si with the bulk phase is the pivotal factor influencing the catalytic activity.The feasibility of transition metal doping to facilitate this step was further investigated.The Si disengage energy(E_(d))was used as a quantitative parameter to assess the catalytic activity of the catalysts,and the optimal descriptor was determined through interpretable machine learning.It was demonstrated that d-band center and electron transfer play a crucial role in regulating the level of Ed.This work reveals the mechanism and structure-activity relationship for the low-temperature hydrogenation reaction of STC,and provides a basis for the rational design of catalysts.展开更多
Nonfungible tokens(NFTs)have become highly sought-after assets in recent years,exhibiting potential for profitability and hedging.The large and lucrative NFT market has attracted both practitioners and researchers to ...Nonfungible tokens(NFTs)have become highly sought-after assets in recent years,exhibiting potential for profitability and hedging.The large and lucrative NFT market has attracted both practitioners and researchers to develop NFT price-prediction models.However,the extant models have some weaknesses in terms of model comprehensiveness and operational convenience.To address these research gaps,we propose a multimodal end-to-end interpretable deep learning(MEID)framework for NFT investment.Our model integrates visual features,textual descriptions,transaction indicators,and historical price time series by leveraging the advantages of convolutional neural networks(CNNs),adopts integrated gradient(IG)to improve interpretability,and designs a built-in financial evaluation mechanism to generate not only the predicted price category but also the recommended purchase level.The experimental results demonstrate that the proposed MEID framework has excellent properties in terms of the evaluation metrics.The proposed MEID framework could help investors identify market opportunities and help NFT transaction platforms design smart investment tools and improve transaction volume.展开更多
The widespread adoption of tunnel boring machines(TBMs)has led to an increased focus on disc cutter wear,including both normal and abnormal types,for efficient and safe TBM excavation.However,abnormal wear has yet to ...The widespread adoption of tunnel boring machines(TBMs)has led to an increased focus on disc cutter wear,including both normal and abnormal types,for efficient and safe TBM excavation.However,abnormal wear has yet to be thoroughly investigated,primarily due to the complexity of considering mixed ground conditions and the imbalance in the number of instances between the two types of wear.This study developed a prediction model for abnormal TBM disc cutter wear,considering mixed ground conditions,by employing interpretable machine learning with data augmentation.An equivalent elastic modulus was used to consider the characteristics of mixed ground conditions,and wear data was obtained from 65 cutterhead intervention(CHI)reports covering both mixed ground and hard rock sections.With a balanced training dataset obtained by data augmentation,an extreme gradient boosting(XGB)model delivered acceptable results with an accuracy of 0.94,an F1-score of 0.808,and a recall of 0.8.In addition,the accuracy for each individual disc cutter exhibited low variability.When employing data augmentation,a significant improvement in recall was observed compared to when it was not used,although the difference in accuracy and F1-score was marginal.The subsequent model interpretation revealed the chamber pressure,cutter installation radius,and torque as significant contributors.Specifically,a threshold in chamber pressure was observed,which could induce abnormal wear.The study also explored how elevated values of these influential contributors correlate with abnormal wear.The proposed model offers a valuable tool for planning the replacement of abnormally worn disc cutters,enhancing the safety and efficiency of TBM operations.展开更多
The 2016–2022 monitoring data from three ecological buoys in the Wenzhou coastal region of Zhejiang Province and the dataset European Centre for Medium-Range Weather Forecasts were examined to clarify the elaborate r...The 2016–2022 monitoring data from three ecological buoys in the Wenzhou coastal region of Zhejiang Province and the dataset European Centre for Medium-Range Weather Forecasts were examined to clarify the elaborate relationship between variations in ecological parameters during spring algal bloom incidents and the associated changes in temperature and wind fields in this study.A long short-term memory recurrent neural network was employed,and a predictive model for spring algal bloom in this region was developed.This model integrated various inputs,including temperature,wind speed,and other pertinent variables,and chlorophyll concentration served as the primary output indicator.The model training used chlorophyll concentration data,which were supplemented by reanalysis and forecast temperature and wind field data.The model demonstrated proficiency in forecasting next-day chlorophyll concentrations and assessing the likelihood of spring algal bloom occurrences using a defined chlorophyll concentration threshold.The historical validation from 2016 to 2019 corroborated the model's accuracy with an 81.71%probability of correct prediction,which was further proven by its precise prediction of two spring algal bloom incidents in late April 2023 and early May 2023.An interpretable machine learning-based model for spring algal bloom prediction,displaying effective forecasting with limited data,was established through the detailed analysis of the spring algal bloom mechanism and the careful selection of input variables.The insights gained from this study offer valuable contributions to the development of early warning systems for spring algal bloom in the Wenzhou coastal area of Zhejiang Province.展开更多
In the face of the unrelenting challenge posed by earthquakes-a natural hazard of unpredictable nature with a legacy of significant loss of life,destruction of infrastructure,and profound economic and social impacts-t...In the face of the unrelenting challenge posed by earthquakes-a natural hazard of unpredictable nature with a legacy of significant loss of life,destruction of infrastructure,and profound economic and social impacts-the scientific community has pursued advancements in earthquake early warning systems(EEWSs).These systems are vital for pre-emptive actions and decision-making that can save lives and safeguard critical infrastructure.This study proposes and validates a domain-informed deep learning-based EEWS called the hybrid earthquake early warning framework for estimating response spectra(HEWFERS),which represents a significant leap forward in the capabilities to predict ground shaking intensity in real-time,aligning with the United Nations’disaster risk reduction goals.HEWFERS ingeniously integrates a domain-informed variational autoencoder for physics-based latent variable(LV)extraction,a feed-forward neural network for on-site prediction,and Gaussian process regression for spatial prediction.Adopting explainable artificial intelligence-based Shapley explanations further elucidates the predictive mechanisms,ensuring stakeholder-informed decisions.By conducting an extensive analysis of the proposed framework under a large database of approximately 14000 recorded ground motions,this study offers insights into the potential of integrating machine learning with seismology to revolutionize earthquake preparedness and response,thus paving the way for a safer and more resilient future.展开更多
Purpose:This study aims to integrate large language models(LLMs)with interpretable machine learning methods to develop a multimodal data-driven framework for predicting corporate financial fraud,addressing the limitat...Purpose:This study aims to integrate large language models(LLMs)with interpretable machine learning methods to develop a multimodal data-driven framework for predicting corporate financial fraud,addressing the limitations of traditional approaches in long-text semantic parsing,model interpretability,and multisource data fusion,thereby providing regulatory agencies with intelligent auditing tools.Design/methodology/approach:Analyzing 5,304 Chinese listed firms’annual reports(2015-2020)from the CSMAD database,this study leverages the Doubao LLMs to generate chunked summaries and 256-dimensional semantic vectors,developing textual semantic features.It integrates 19 financial indicators,11 governance metrics,and linguistic characteristics(tone,readability)with fraud prediction models optimized through a group of Gradient Boosted Decision Tree(GBDT)algorithms.SHAP value analysis in the final model reveals the risk transmission mechanism by quantifying the marginal impacts of financial,governance,and textual features on fraud likelihood.Findings:The study found that LLMs effectively distill lengthy annual reports into semantic summaries,while GBDT algorithms(AUC>0.850)outperform the traditional Logistic Regression model in fraud detection.Multimodal fusion improved performance by 7.4%,with financial,governance,and textual features providing complementary signals.SHAP analysis revealed financial distress,governance conflicts,and narrative patterns(e.g.,tone anchoring,semantic thresholds)as key fraud indicators,highlighting managerial intent in report language.Research limitations:This study identifies three key limitations:1)lack of interpretability for semantic features,2)absence of granular fraud-type differentiation,and 3)unexplored comparative validation with other deep learning methods.Future research will address these gaps to enhance fraud detection precision and model transparency.Practical implications:The developed semantic-enhanced evaluation model provides a quantitative tool for assessing listed companies’information disclosure quality and enables practical implementation through its derivative real-time monitoring system.This advancement significantly strengthens capital market risk early warning capabilities,offering actionable insights for securities regulation.Originality/value:This study presents three key innovations:1)A novel“chunking-summarizationembedding”framework for efficient semantic compression of lengthy annual reports(30,000 words);2)Demonstration of LLMs’superior performance in financial text analysis,outperforming traditional methods by 19.3%;3)A novel“language-psychology-behavior”triad model for analyzing managerial fraud motives.展开更多
Dear Editor,Health management is essential to ensure battery performance and safety, while data-driven learning system is a promising solution to enable efficient state of health(SoH) estimation of lithium-ion(Liion) ...Dear Editor,Health management is essential to ensure battery performance and safety, while data-driven learning system is a promising solution to enable efficient state of health(SoH) estimation of lithium-ion(Liion) batteries. However, the time-consuming signal data acquisition and the lack of interpretability of model still hinder its efficient deployment. Motivated by this, this letter proposes a novel and interpretable data-driven learning strategy through combining the benefits of explainable AI and non-destructive ultrasonic detection for battery SoH estimation. Specifically, after equipping battery with advanced ultrasonic sensor to promise fast real-time ultrasonic signal measurement, an interpretable data-driven learning strategy named generalized additive neural decision ensemble(GANDE) is designed to rapidly estimate battery SoH and explain the effects of the involved ultrasonic features of interest.展开更多
BACKGROUND The early diagnosis rate of pancreatic ductal adenocarcinoma(PDAC)is low and the prognosis is poor.It is important to develop an interpretable noninvasive early diagnostic model in clinical practice.AIM To ...BACKGROUND The early diagnosis rate of pancreatic ductal adenocarcinoma(PDAC)is low and the prognosis is poor.It is important to develop an interpretable noninvasive early diagnostic model in clinical practice.AIM To develop an interpretable noninvasive early diagnostic model for PDAC using plasma extracellular vesicle long RNA(EvlRNA).METHODS The diagnostic model was constructed based on plasma EvlRNA data.During the process of establishing the model,EvlRNA-index was introduced,and four algorithms were adopted to calculate EvlRNA-index.After the model was successfully constructed,performance evaluation was conducted.A series of bioinformatics methods were adopted to explore the potential mechanism of EvlRNA-index as the input feature of the model.And the relationship between key characteristics and PDAC were explored at the single-cell level.RESULTS A novel interpretable machine learning framework was developed based on plasma EvlRNA.In this framework,a two-layer classifier was established.A new concept was proposed:EvlRNA-index.Based on EvlRNA-index,a cancer diagnostic model was established,and a good diagnostic effect was achieved.The accuracy of PDACandCPvsHealth-Probabilistic PCA Index-SVM(PDAC and chronic pancreatitis vs health-probabilistic principal component analysis index-support vector machine)(1-18)was 91.51%,with Mathew’s correlation coefficient 0.7760 and area under the curve 0.9560.In the second layer of the model,the accuracy of PDACvsCP-Probabilistic PCA Index-RF(PDAC vs chronic pancreatitis-probabilistic principal component analysis index-random forest)(2-17)was 93.83%,with Mathew’s correlation coefficient 0.8422 and area under the curve 0.9698.Forty-nine PDAC-related genes were identified,among which 16 were known,inferring that the remaining ones were also PDAC-related genes.CONCLUSION An interpretable two-layer machine learning framework was proposed for early diagnosis and prediction of PDAC based on plasma EvlRNA,providing new insights into the clinical value of EvlRNA.展开更多
AIM:To investigate the associations between urinary dialkyl phosphate(DAP)metabolites of organophosphorus pesticides(OPPs)exposure and age-related macular degeneration(AMD)risk.METHODS:Participants were drawn from the...AIM:To investigate the associations between urinary dialkyl phosphate(DAP)metabolites of organophosphorus pesticides(OPPs)exposure and age-related macular degeneration(AMD)risk.METHODS:Participants were drawn from the National Health and Nutrition Examination Survey(NHANES)between 2005 and 2008.Urinary DAP metabolites were used to construct a machine learning(ML)model for AMD prediction.Several interpretability pipelines,including permutation feature importance(PFI),partial dependence plot(PDP),and SHapley Additive exPlanations(SHAP)analyses were employed to analyze the influence from exposure features to prediction outcomes.RESULTS:A total of 1845 participants were included and 137 were diagnosed with AMD.Receiver operating characteristic curve(ROC)analysis evaluated Random Forests(RF)as the best ML model with its optimal predictive performance among eleven models.PFI and SHAP analyses illustrated that DAP metabolites were of significant contribution weights in AMD risk prediction,higher than most of the socio-demographic covariates.Shapley values and waterfall plots of randomly selected AMD individuals emphasized the predictive capacity of ML with high accuracy and sensitivity in each case.The relationships and interactions visualized by graphical plots and supported by statistical measures demonstrated the indispensable impacts from six DAP metabolites to the prediction of AMD risk.CONCLUSION:Urinary DAP metabolites of OPPs exposure are associated with AMD risk and ML algorithms show the excellent generalizability and differentiability in the course of AMD risk prediction.展开更多
The optimization of reaction processes is crucial for the green, efficient, and sustainable development of the chemical industry. However, how to address the problems posed by multiple variables, nonlinearities, and u...The optimization of reaction processes is crucial for the green, efficient, and sustainable development of the chemical industry. However, how to address the problems posed by multiple variables, nonlinearities, and uncertainties during optimization remains a formidable challenge. In this study, a strategy combining interpretable machine learning with metaheuristic optimization algorithms is employed to optimize the reaction process. First, experimental data from a biodiesel production process are collected to establish a database. These data are then used to construct a predictive model based on artificial neural network (ANN) models. Subsequently, interpretable machine learning techniques are applied for quantitative analysis and verification of the model. Finally, four metaheuristic optimization algorithms are coupled with the ANN model to achieve the desired optimization. The research results show that the methanol: palm fatty acid distillate (PFAD) molar ratio contributes the most to the reaction outcome, accounting for 41%. The ANN-simulated annealing (SA) hybrid method is more suitable for this optimization, and the optimal process parameters are a catalyst concentration of 3.00% (mass), a methanol: PFAD molar ratio of 8.67, and a reaction time of 30 min. This study provides deeper insights into reaction process optimization, which will facilitate future applications in various reaction optimization processes.展开更多
System upgrades in unmanned systems have made Unmanned Aerial Vehicle(UAV)-based patrolling and monitoring a preferred solution for ocean surveillance.However,dynamic environments and large-scale deployments pose sign...System upgrades in unmanned systems have made Unmanned Aerial Vehicle(UAV)-based patrolling and monitoring a preferred solution for ocean surveillance.However,dynamic environments and large-scale deployments pose significant challenges for efficient decision-making,necessitating a modular multiagent control system.Deep Reinforcement Learning(DRL)and Decision Tree(DT)have been utilized for these complex decision-making tasks,but each has its limitations:DRL is highly adaptive but lacks interpretability,while DT is inherently interpretable but has limited adaptability.To overcome these challenges,we propose the Adaptive Interpretable Decision Tree(AIDT),an evolutionary-based algorithm that is both adaptable to diverse environmental settings and highly interpretable in its decision-making processes.We first construct a Markov decision process(MDP)-based simulation environment using the Cooperative Submarine Search task as a representative scenario for training and testing the proposed method.Specifically,we use the heat map as a state variable to address the issue of multi-agent input state proliferation.Next,we introduce the curiosity-guiding intrinsic reward to encourage comprehensive exploration and enhance algorithm performance.Additionally,we incorporate decision tree size as an influence factor in the adaptation process to balance task completion with computational efficiency.To further improve the generalization capability of the decision tree,we apply a normalization method to ensure consistent processing of input states.Finally,we validate the proposed algorithm in different environmental settings,and the results demonstrate both its adaptability and interpretability.展开更多
The potential toxicity of ionic liquids(ILs)affects their applications;how to control the toxicity is one of the key issues in their applications.To understand its toxicity structure relationship and promote its green...The potential toxicity of ionic liquids(ILs)affects their applications;how to control the toxicity is one of the key issues in their applications.To understand its toxicity structure relationship and promote its greener application,six different machine learning algorithms,including Bagging,Adaptive Boosting(AdaBoost),Gradient Boosting(GBoost),Stacking,Voting and Categorical Boosting(CatBoost),are established to model the toxicity of ILs on four distinct datasets including Leukemia rat cell line IPC-81(IPC-81),Acetylcholinesterase(AChE),Escherichia coli(E.coli)and Vibrio fischeri.Molecular descriptors obtained from the simplified molecular input line entry system(SMILES)are used to characterize ILs.All models are assessed by the mean square error(MSE),root mean square error(RMSE),mean absolute error(MAE)and correlation coefficient(R^(2)).Additionally,an interpretation model based on SHapley Additive exPlanations(SHAP)is built to determine the positive and negative effects of each molecular feature on toxicity.With additional parameters and complexity,the Catboost model outperforms the other models,making it a more reliable model for ILs'toxicity prediction.The results of the model's interpretation indicate that the most significant positive features,SMR_VSA5,PEOE_VSA8,Kappa2,PEOE_VSA6,SMR_VSA5,PEOE_VSA6 and EState_VSA1,can increase the toxicity of ILs as their levels rise,while the most significant negative features,VSA_EState7,EState_VSA8,PEOE_VSA9 and FpDensityMorgan1,can decrease the toxicity as their levels rise.Also,an IL's toxicity will grow as its average molecular weight and number of pyridine rings increase,whereas its toxicity will decrease as its hydrogen bond acceptors increase.This finding offers a theoretical foundation for rapid screening and synthesis of environmentally-benign ILs.展开更多
Liquid rocket engine(LRE)fault diagnosis is critical for successful space launch missions,enabling timely avoidance of safety hazards,while accurate post-failure analysis prevents subsequent economic losses.However,th...Liquid rocket engine(LRE)fault diagnosis is critical for successful space launch missions,enabling timely avoidance of safety hazards,while accurate post-failure analysis prevents subsequent economic losses.However,the complexity of LRE systems and the“black-box”nature of current deep learning-based diagnostic methods hinder interpretable fault diagnosis.This paper establishes Granger causality(GC)extraction-based component-wise multi-layer perceptron(GCMLP),achieving high fault diagnosis accuracy while leveraging GC to enhance diagnostic interpretability.First,component-wise MLP networks are constructed for distinct LRE variables to extract inter-variable GC relationships.Second,dedicated predictors are designed for each variable,leveraging historical data and GC relationships to forecast future states,thereby ensuring GC reliability.Finally,the extracted GC features are utilized for fault classification,guaranteeing feature discriminability and diagnosis accuracy.This study simulates six critical fault modes in LRE using Simulink.Based on the generated simulation data,GCMLP demonstrates superior fault localization accuracy compared to benchmark methods,validating its efficacy and robustness.展开更多
As industrial systems become increasingly complex,the significant research interest has been devoted to intelligent fault diagnosis approaches leveraging deep learning.However,existing methods still face two critical ...As industrial systems become increasingly complex,the significant research interest has been devoted to intelligent fault diagnosis approaches leveraging deep learning.However,existing methods still face two critical challenges in practical applications:1)the extracted features often fail to maintain robustness in nonstationary conditions;2)deep neural networks generally exhibit a black box nature,offering limited interpretability in their feature extraction process.To solve the above issues,an interpretable wavelet Kolmogorov-Arnold convolutional Long Short-Term Memory(WKAConvLSTM)is proposed,which mainly consists of two key components:1)a wavelet Kolmogorov-Arnold kernel(WKAK)with learnable scale and translation parameters is designed and then embedded into convolutional layers to enable the extracted spatial features interpretable;2)a multi-head attention-enhanced Long Short-Term Memory(MHA-LSTM)is proposed to effectively capture crucial temporal dependencies in sequential data.In order to verify its effectiveness,the proposed model is tested on bearing and gearbox datasets under complex conditions,including noise interference,nonstationary operating conditions,and data class imbalance.The experimental results demonstrate that it not only achieves superior diagnostic accuracy compared with advanced baseline models but also enhances the interpretability of the extracted features.展开更多
Despite significant progress in the Prognostics and Health Management(PHM)domain using pattern learning systems from data,machine learning(ML)still faces challenges related to limited generalization and weak interpret...Despite significant progress in the Prognostics and Health Management(PHM)domain using pattern learning systems from data,machine learning(ML)still faces challenges related to limited generalization and weak interpretability.A promising approach to overcoming these challenges is to embed domain knowledge into the ML pipeline,enhancing the model with additional pattern information.In this paper,we review the latest developments in PHM,encapsulated under the concept of Knowledge Driven Machine Learning(KDML).We propose a hierarchical framework to define KDML in PHM,which includes scientific paradigms,knowledge sources,knowledge representations,and knowledge embedding methods.Using this framework,we examine current research to demonstrate how various forms of knowledge can be integrated into the ML pipeline and provide roadmap to specific usage.Furthermore,we present several case studies that illustrate specific implementations of KDML in the PHM domain,including inductive experience,physical model,and signal processing.We analyze the improvements in generalization capability and interpretability that KDML can achieve.Finally,we discuss the challenges,potential applications,and usage recommendations of KDML in PHM,with a particular focus on the critical need for interpretability to ensure trustworthy deployment of artificial intelligence in PHM.展开更多
Intelligent fault diagnosis technology plays an indispensable role in ensuring the safety,stability,and efficiency of railway operations.However,existing studies have the following limitations.1)They are typical black-...Intelligent fault diagnosis technology plays an indispensable role in ensuring the safety,stability,and efficiency of railway operations.However,existing studies have the following limitations.1)They are typical black-box models that lacks interpretability as well as they fuse features by simply stacking them,overlooking the discrepancies in the importance of different features,which reduces the credibility and diagnosis accuracy of the models.2)They ignore the effects of potentially mistaken labels in the training datasets disrupting the ability of the models to learn the true data distribution,which degrades the generalization performance of intelligent diagnosis models,especially when the training samples are limited.To address the above items,an interpretable few-shot framework for fault diagnosis with noisy labels is proposed for train transmission systems.In the proposed framework,a feature extractor is constructed by stacked frequency band focus modules,which can capture signal features in different frequency bands and further adaptively concentrate on the features corresponding to the potential fault characteristic frequency.Then,according to prototypical network,a novel metric-based classifier is developed that is tolerant to mislabeled support samples in the case of limited samples.Besides,a new loss function is designed to decrease the impact of label mistakes in query datasets.Finally,fault simulation experiments of subway train transmission systems are designed and conducted,and the effectiveness as well as superiority of the proposed method are proved by ablation experiments and comparison with the existing methods.展开更多
基金co-supported by the National Natural Science Foundation of China(No.62001507)the Youth Talent Lifting Project of the China Association for Science and Technology(No.2021-JCJQ-QT-018)+1 种基金the Program of the Youth Innovation Team of Shaanxi Universitiesthe Natural Science Basic Research Plan in Shaanxi Province of China(No.2023-JC-YB-491)。
文摘Deep Learning(DL)model has been widely used in the field of Synthetic Aperture Radar Automatic Target Recognition(SAR-ATR)and has achieved excellent performance.However,the black-box nature of DL models has been the focus of criticism,especially in the application of SARATR,which is closely associated with the national defense and security domain.To address these issues,a new interpretable recognition model Physics-Guided BagNet(PGBN)is proposed in this article.The model adopts an interpretable convolutional neural network framework and uses time–frequency analysis to extract physical scattering features in SAR images.Based on the physical scattering features,an unsupervised segmentation method is proposed to distinguish targets from the background in SAR images.On the basis of the segmentation result,a structure is designed,which constrains the model's spatial attention to focus more on the targets themselves rather than the background,thereby making the model's decision-making more in line with physical principles.In contrast to previous interpretable research methods,this model combines interpretable structure with physical interpretability,further reducing the model's risk of error recognition.Experiments on the MSTAR dataset verify that the PGBN model exhibits excellent interpretability and recognition performance,and comparative experiments with heatmaps indicate that the physical feature guidance module presented in this article can constrain the model to focus more on the target itself rather than the background.
基金Supported by National Key Research and Development Program,No.2022YFC2407304Major Research Project for Middle-Aged and Young Scientists of Fujian Provincial Health Commission,No.2021ZQNZD013+2 种基金The National Natural Science Foundation of China,No.62275050Fujian Province Science and Technology Innovation Joint Fund Project,No.2019Y9108Major Science and Technology Projects of Fujian Province,No.2021YZ036017.
文摘BACKGROUND To investigate the preoperative factors influencing textbook outcomes(TO)in Intrahepatic cholangiocarcinoma(ICC)patients and evaluate the feasibility of an interpretable machine learning model for preoperative prediction of TO,we developed a machine learning model for preoperative prediction of TO and used the SHapley Additive exPlanations(SHAP)technique to illustrate the prediction process.AIM To analyze the factors influencing textbook outcomes before surgery and to establish interpretable machine learning models for preoperative prediction.METHODS A total of 376 patients diagnosed with ICC were retrospectively collected from four major medical institutions in China,covering the period from 2011 to 2017.Logistic regression analysis was conducted to identify preoperative variables associated with achieving TO.Based on these variables,an EXtreme Gradient Boosting(XGBoost)machine learning prediction model was constructed using the XGBoost package.The SHAP(package:Shapviz)algorithm was employed to visualize each variable's contribution to the model's predictions.Kaplan-Meier survival analysis was performed to compare the prognostic differences between the TO-achieving and non-TO-achieving groups.RESULTS Among 376 patients,287 were included in the training group and 89 in the validation group.Logistic regression identified the following preoperative variables influencing TO:Child-Pugh classification,Eastern Cooperative Oncology Group(ECOG)score,hepatitis B,and tumor size.The XGBoost prediction model demonstrated high accuracy in internal validation(AUC=0.8825)and external validation(AUC=0.8346).Survival analysis revealed that the disease-free survival rates for patients achieving TO at 1,2,and 3 years were 64.2%,56.8%,and 43.4%,respectively.CONCLUSION Child-Pugh classification,ECOG score,hepatitis B,and tumor size are preoperative predictors of TO.In both the training group and the validation group,the machine learning model had certain effectiveness in predicting TO before surgery.The SHAP algorithm provided intuitive visualization of the machine learning prediction process,enhancing its interpretability.
基金supported by the Postdoctoral Fellowship Program of CPSF(Grant No.GZB20230685)the National Science Foundation of China(Grant No.42277161).
文摘Forecasting landslide deformation is challenging due to influence of various internal and external factors on the occurrence of systemic and localized heterogeneities.Despite the potential to improve landslide predictability,deep learning has yet to be sufficiently explored for complex deformation patterns associated with landslides and is inherently opaque.Herein,we developed a holistic landslide deformation forecasting method that considers spatiotemporal correlations of landslide deformation by integrating domain knowledge into interpretable deep learning.By spatially capturing the interconnections between multiple deformations from different observation points,our method contributes to the understanding and forecasting of landslide systematic behavior.By integrating specific domain knowledge relevant to each observation point and merging internal properties with external variables,the local heterogeneity is considered in our method,identifying deformation temporal patterns in different landslide zones.Case studies involving reservoir-induced landslides and creeping landslides demonstrated that our approach(1)enhances the accuracy of landslide deformation forecasting,(2)identifies significant contributing factors and their influence on spatiotemporal deformation characteristics,and(3)demonstrates how identifying these factors and patterns facilitates landslide forecasting.Our research offers a promising and pragmatic pathway toward a deeper understanding and forecasting of complex landslide behaviors.
基金funded by the National Natural Science Foundation of China(No.52204407)the Natural Science Foundation of Jiangsu Province(No.BK20220595)+1 种基金the China Postdoctoral Science Foundation(No.2022M723689)the Industrial Collaborative Innovation Project of Shanghai(No.XTCX-KJ-2022-2-11)。
文摘The application of machine learning in alloy design is increasingly widespread,yet traditional models still face challenges when dealing with limited datasets and complex nonlinear relationships.This work proposes an interpretable machine learning method based on data augmentation and reconstruction,excavating high-performance low-alloyed magnesium(Mg)alloys.The data augmentation technique expands the original dataset through Gaussian noise.The data reconstruction method reorganizes and transforms the original data to extract more representative features,significantly improving the model's generalization ability and prediction accuracy,with a coefficient of determination(R^(2))of 95.9%for the ultimate tensile strength(UTS)model and a R^(2)of 95.3%for the elongation-to-failure(EL)model.The correlation coefficient assisted screening(CCAS)method is proposed to filter low-alloyed target alloys.A new Mg-2.2Mn-0.4Zn-0.2Al-0.2Ca(MZAX2000,wt%)alloy is designed and extruded into bar at given processing parameters,achieving room-temperature strength-ductility synergy showing an excellent UTS of 395 MPa and a high EL of 17.9%.This is closely related to its hetero-structured characteristic in the as-extruded MZAX2000 alloy consisting of coarse grains(16%),fine grains(75%),and fiber regions(9%).Therefore,this work offers new insights into optimizing alloy compositions and processing parameters for attaining new high strong and ductile low-alloyed Mg alloys.
基金supported by Hubei Three Gorges Laboratory Open Innovation Fund Project(SC231002)CFD Simulation to Explore the Mass and Heat Transfer Laws of Thermal Decomposition of Mixed Salt Organic Compounds Project(2021YFC 3201404).
文摘Low-temperature hydrogenation of silicon tetrachloride(STC)is an essential step in polysilicon production.The addition of CuCl to silicon powder is currently a commonly used catalytic method and the silicon powder acts as both a reactant and a catalyst.However,the reaction mechanism and the structure-activity relationship of this process have not been fully elucidated.In this work,a comprehensive study of the reaction mechanism in the presence of Si and Cu_(3)Si was carried out using density functional theory(DFT)combined with experiments,respectively.The results indicated that the ratedetermining step(RDS)in the presence of Si is the phase transition of Si atom,meanwhile,the RDS in the presence of Cu_(3)Si is the TCS-generation process.The activation barrier of the latter is smaller,highlighting that the interaction of Si with the bulk phase is the pivotal factor influencing the catalytic activity.The feasibility of transition metal doping to facilitate this step was further investigated.The Si disengage energy(E_(d))was used as a quantitative parameter to assess the catalytic activity of the catalysts,and the optimal descriptor was determined through interpretable machine learning.It was demonstrated that d-band center and electron transfer play a crucial role in regulating the level of Ed.This work reveals the mechanism and structure-activity relationship for the low-temperature hydrogenation reaction of STC,and provides a basis for the rational design of catalysts.
基金supported by the National Key Research and Development Program of China(Project No.2022YFC3320800)the National Natural Science Foundation of China(Project No.72571210).
文摘Nonfungible tokens(NFTs)have become highly sought-after assets in recent years,exhibiting potential for profitability and hedging.The large and lucrative NFT market has attracted both practitioners and researchers to develop NFT price-prediction models.However,the extant models have some weaknesses in terms of model comprehensiveness and operational convenience.To address these research gaps,we propose a multimodal end-to-end interpretable deep learning(MEID)framework for NFT investment.Our model integrates visual features,textual descriptions,transaction indicators,and historical price time series by leveraging the advantages of convolutional neural networks(CNNs),adopts integrated gradient(IG)to improve interpretability,and designs a built-in financial evaluation mechanism to generate not only the predicted price category but also the recommended purchase level.The experimental results demonstrate that the proposed MEID framework has excellent properties in terms of the evaluation metrics.The proposed MEID framework could help investors identify market opportunities and help NFT transaction platforms design smart investment tools and improve transaction volume.
基金support of the“National R&D Project for Smart Construction Technology (Grant No.RS-2020-KA157074)”funded by the Korea Agency for Infrastructure Technology Advancement under the Ministry of Land,Infrastructure and Transport,and managed by the Korea Expressway Corporation.
文摘The widespread adoption of tunnel boring machines(TBMs)has led to an increased focus on disc cutter wear,including both normal and abnormal types,for efficient and safe TBM excavation.However,abnormal wear has yet to be thoroughly investigated,primarily due to the complexity of considering mixed ground conditions and the imbalance in the number of instances between the two types of wear.This study developed a prediction model for abnormal TBM disc cutter wear,considering mixed ground conditions,by employing interpretable machine learning with data augmentation.An equivalent elastic modulus was used to consider the characteristics of mixed ground conditions,and wear data was obtained from 65 cutterhead intervention(CHI)reports covering both mixed ground and hard rock sections.With a balanced training dataset obtained by data augmentation,an extreme gradient boosting(XGB)model delivered acceptable results with an accuracy of 0.94,an F1-score of 0.808,and a recall of 0.8.In addition,the accuracy for each individual disc cutter exhibited low variability.When employing data augmentation,a significant improvement in recall was observed compared to when it was not used,although the difference in accuracy and F1-score was marginal.The subsequent model interpretation revealed the chamber pressure,cutter installation radius,and torque as significant contributors.Specifically,a threshold in chamber pressure was observed,which could induce abnormal wear.The study also explored how elevated values of these influential contributors correlate with abnormal wear.The proposed model offers a valuable tool for planning the replacement of abnormally worn disc cutters,enhancing the safety and efficiency of TBM operations.
基金the Zhejiang Provincial Natural Science Foundation of China(No.LY21D 060003)the Project of State Key Laboratory of Satellite Ocean Environment Dynamics,Second Institute of Ocean-ography,MNR(No.SOEDZZ2103)+1 种基金the National Natural Science Foundation of China(No.42076216)the Open Research Fund of the Key Laboratory of Marine Ecological Monitoring and Restoration Technologies,MNR(No.MEMRT202210)。
文摘The 2016–2022 monitoring data from three ecological buoys in the Wenzhou coastal region of Zhejiang Province and the dataset European Centre for Medium-Range Weather Forecasts were examined to clarify the elaborate relationship between variations in ecological parameters during spring algal bloom incidents and the associated changes in temperature and wind fields in this study.A long short-term memory recurrent neural network was employed,and a predictive model for spring algal bloom in this region was developed.This model integrated various inputs,including temperature,wind speed,and other pertinent variables,and chlorophyll concentration served as the primary output indicator.The model training used chlorophyll concentration data,which were supplemented by reanalysis and forecast temperature and wind field data.The model demonstrated proficiency in forecasting next-day chlorophyll concentrations and assessing the likelihood of spring algal bloom occurrences using a defined chlorophyll concentration threshold.The historical validation from 2016 to 2019 corroborated the model's accuracy with an 81.71%probability of correct prediction,which was further proven by its precise prediction of two spring algal bloom incidents in late April 2023 and early May 2023.An interpretable machine learning-based model for spring algal bloom prediction,displaying effective forecasting with limited data,was established through the detailed analysis of the spring algal bloom mechanism and the careful selection of input variables.The insights gained from this study offer valuable contributions to the development of early warning systems for spring algal bloom in the Wenzhou coastal area of Zhejiang Province.
基金the financial support from the Chilean National Research and Development Agency(Agencia Nacional de Investigación y Desarrollo,ANID)through Fondo Nacional de Desarrollo Científico y Tecnológico(FONDECYT)Regular 1240503Fondo de Valorización de la Investigación(FOVI)230030 projectsthe financial support from the ANID through FONDECYT Reg-ular 1240501.
文摘In the face of the unrelenting challenge posed by earthquakes-a natural hazard of unpredictable nature with a legacy of significant loss of life,destruction of infrastructure,and profound economic and social impacts-the scientific community has pursued advancements in earthquake early warning systems(EEWSs).These systems are vital for pre-emptive actions and decision-making that can save lives and safeguard critical infrastructure.This study proposes and validates a domain-informed deep learning-based EEWS called the hybrid earthquake early warning framework for estimating response spectra(HEWFERS),which represents a significant leap forward in the capabilities to predict ground shaking intensity in real-time,aligning with the United Nations’disaster risk reduction goals.HEWFERS ingeniously integrates a domain-informed variational autoencoder for physics-based latent variable(LV)extraction,a feed-forward neural network for on-site prediction,and Gaussian process regression for spatial prediction.Adopting explainable artificial intelligence-based Shapley explanations further elucidates the predictive mechanisms,ensuring stakeholder-informed decisions.By conducting an extensive analysis of the proposed framework under a large database of approximately 14000 recorded ground motions,this study offers insights into the potential of integrating machine learning with seismology to revolutionize earthquake preparedness and response,thus paving the way for a safer and more resilient future.
基金supported by the 2021 Guangdong Province(China)Science and Technology Plan Project“Research and Application of Key Technologies for Multi-level Knowledge Retrieval Based on Big Data Intelligence”(Project No.2021B0101420004)the 2022 commissioned project“Cross-border E-commerce Taxation and Related Research”from the State Taxation Administration Guangdong Provincial Taxation Bureau,China.
文摘Purpose:This study aims to integrate large language models(LLMs)with interpretable machine learning methods to develop a multimodal data-driven framework for predicting corporate financial fraud,addressing the limitations of traditional approaches in long-text semantic parsing,model interpretability,and multisource data fusion,thereby providing regulatory agencies with intelligent auditing tools.Design/methodology/approach:Analyzing 5,304 Chinese listed firms’annual reports(2015-2020)from the CSMAD database,this study leverages the Doubao LLMs to generate chunked summaries and 256-dimensional semantic vectors,developing textual semantic features.It integrates 19 financial indicators,11 governance metrics,and linguistic characteristics(tone,readability)with fraud prediction models optimized through a group of Gradient Boosted Decision Tree(GBDT)algorithms.SHAP value analysis in the final model reveals the risk transmission mechanism by quantifying the marginal impacts of financial,governance,and textual features on fraud likelihood.Findings:The study found that LLMs effectively distill lengthy annual reports into semantic summaries,while GBDT algorithms(AUC>0.850)outperform the traditional Logistic Regression model in fraud detection.Multimodal fusion improved performance by 7.4%,with financial,governance,and textual features providing complementary signals.SHAP analysis revealed financial distress,governance conflicts,and narrative patterns(e.g.,tone anchoring,semantic thresholds)as key fraud indicators,highlighting managerial intent in report language.Research limitations:This study identifies three key limitations:1)lack of interpretability for semantic features,2)absence of granular fraud-type differentiation,and 3)unexplored comparative validation with other deep learning methods.Future research will address these gaps to enhance fraud detection precision and model transparency.Practical implications:The developed semantic-enhanced evaluation model provides a quantitative tool for assessing listed companies’information disclosure quality and enables practical implementation through its derivative real-time monitoring system.This advancement significantly strengthens capital market risk early warning capabilities,offering actionable insights for securities regulation.Originality/value:This study presents three key innovations:1)A novel“chunking-summarizationembedding”framework for efficient semantic compression of lengthy annual reports(30,000 words);2)Demonstration of LLMs’superior performance in financial text analysis,outperforming traditional methods by 19.3%;3)A novel“language-psychology-behavior”triad model for analyzing managerial fraud motives.
基金supported by the National Natural Science Foundation of China(62373224,62333013,U23A20327)the Natural Science Foundation of Shandong Province(ZR2024JQ021)
文摘Dear Editor,Health management is essential to ensure battery performance and safety, while data-driven learning system is a promising solution to enable efficient state of health(SoH) estimation of lithium-ion(Liion) batteries. However, the time-consuming signal data acquisition and the lack of interpretability of model still hinder its efficient deployment. Motivated by this, this letter proposes a novel and interpretable data-driven learning strategy through combining the benefits of explainable AI and non-destructive ultrasonic detection for battery SoH estimation. Specifically, after equipping battery with advanced ultrasonic sensor to promise fast real-time ultrasonic signal measurement, an interpretable data-driven learning strategy named generalized additive neural decision ensemble(GANDE) is designed to rapidly estimate battery SoH and explain the effects of the involved ultrasonic features of interest.
基金Supported by Talent Scientific Research Start-up Foundation of Wannan Medical College,No.WYRCQD2023045.
文摘BACKGROUND The early diagnosis rate of pancreatic ductal adenocarcinoma(PDAC)is low and the prognosis is poor.It is important to develop an interpretable noninvasive early diagnostic model in clinical practice.AIM To develop an interpretable noninvasive early diagnostic model for PDAC using plasma extracellular vesicle long RNA(EvlRNA).METHODS The diagnostic model was constructed based on plasma EvlRNA data.During the process of establishing the model,EvlRNA-index was introduced,and four algorithms were adopted to calculate EvlRNA-index.After the model was successfully constructed,performance evaluation was conducted.A series of bioinformatics methods were adopted to explore the potential mechanism of EvlRNA-index as the input feature of the model.And the relationship between key characteristics and PDAC were explored at the single-cell level.RESULTS A novel interpretable machine learning framework was developed based on plasma EvlRNA.In this framework,a two-layer classifier was established.A new concept was proposed:EvlRNA-index.Based on EvlRNA-index,a cancer diagnostic model was established,and a good diagnostic effect was achieved.The accuracy of PDACandCPvsHealth-Probabilistic PCA Index-SVM(PDAC and chronic pancreatitis vs health-probabilistic principal component analysis index-support vector machine)(1-18)was 91.51%,with Mathew’s correlation coefficient 0.7760 and area under the curve 0.9560.In the second layer of the model,the accuracy of PDACvsCP-Probabilistic PCA Index-RF(PDAC vs chronic pancreatitis-probabilistic principal component analysis index-random forest)(2-17)was 93.83%,with Mathew’s correlation coefficient 0.8422 and area under the curve 0.9698.Forty-nine PDAC-related genes were identified,among which 16 were known,inferring that the remaining ones were also PDAC-related genes.CONCLUSION An interpretable two-layer machine learning framework was proposed for early diagnosis and prediction of PDAC based on plasma EvlRNA,providing new insights into the clinical value of EvlRNA.
基金Supported by the National Key Research and Development Program of China(No.2022YFC2502800)the National Natural Science Foundation of China(No.82171076)the Shanghai Municipal Education Commission(No.2023KJ05-67).
文摘AIM:To investigate the associations between urinary dialkyl phosphate(DAP)metabolites of organophosphorus pesticides(OPPs)exposure and age-related macular degeneration(AMD)risk.METHODS:Participants were drawn from the National Health and Nutrition Examination Survey(NHANES)between 2005 and 2008.Urinary DAP metabolites were used to construct a machine learning(ML)model for AMD prediction.Several interpretability pipelines,including permutation feature importance(PFI),partial dependence plot(PDP),and SHapley Additive exPlanations(SHAP)analyses were employed to analyze the influence from exposure features to prediction outcomes.RESULTS:A total of 1845 participants were included and 137 were diagnosed with AMD.Receiver operating characteristic curve(ROC)analysis evaluated Random Forests(RF)as the best ML model with its optimal predictive performance among eleven models.PFI and SHAP analyses illustrated that DAP metabolites were of significant contribution weights in AMD risk prediction,higher than most of the socio-demographic covariates.Shapley values and waterfall plots of randomly selected AMD individuals emphasized the predictive capacity of ML with high accuracy and sensitivity in each case.The relationships and interactions visualized by graphical plots and supported by statistical measures demonstrated the indispensable impacts from six DAP metabolites to the prediction of AMD risk.CONCLUSION:Urinary DAP metabolites of OPPs exposure are associated with AMD risk and ML algorithms show the excellent generalizability and differentiability in the course of AMD risk prediction.
基金supported by the National Natural Science Foundation of China(22408227,22238005)the Postdoctoral Research Foundation of China(GZC20231576).
文摘The optimization of reaction processes is crucial for the green, efficient, and sustainable development of the chemical industry. However, how to address the problems posed by multiple variables, nonlinearities, and uncertainties during optimization remains a formidable challenge. In this study, a strategy combining interpretable machine learning with metaheuristic optimization algorithms is employed to optimize the reaction process. First, experimental data from a biodiesel production process are collected to establish a database. These data are then used to construct a predictive model based on artificial neural network (ANN) models. Subsequently, interpretable machine learning techniques are applied for quantitative analysis and verification of the model. Finally, four metaheuristic optimization algorithms are coupled with the ANN model to achieve the desired optimization. The research results show that the methanol: palm fatty acid distillate (PFAD) molar ratio contributes the most to the reaction outcome, accounting for 41%. The ANN-simulated annealing (SA) hybrid method is more suitable for this optimization, and the optimal process parameters are a catalyst concentration of 3.00% (mass), a methanol: PFAD molar ratio of 8.67, and a reaction time of 30 min. This study provides deeper insights into reaction process optimization, which will facilitate future applications in various reaction optimization processes.
文摘System upgrades in unmanned systems have made Unmanned Aerial Vehicle(UAV)-based patrolling and monitoring a preferred solution for ocean surveillance.However,dynamic environments and large-scale deployments pose significant challenges for efficient decision-making,necessitating a modular multiagent control system.Deep Reinforcement Learning(DRL)and Decision Tree(DT)have been utilized for these complex decision-making tasks,but each has its limitations:DRL is highly adaptive but lacks interpretability,while DT is inherently interpretable but has limited adaptability.To overcome these challenges,we propose the Adaptive Interpretable Decision Tree(AIDT),an evolutionary-based algorithm that is both adaptable to diverse environmental settings and highly interpretable in its decision-making processes.We first construct a Markov decision process(MDP)-based simulation environment using the Cooperative Submarine Search task as a representative scenario for training and testing the proposed method.Specifically,we use the heat map as a state variable to address the issue of multi-agent input state proliferation.Next,we introduce the curiosity-guiding intrinsic reward to encourage comprehensive exploration and enhance algorithm performance.Additionally,we incorporate decision tree size as an influence factor in the adaptation process to balance task completion with computational efficiency.To further improve the generalization capability of the decision tree,we apply a normalization method to ensure consistent processing of input states.Finally,we validate the proposed algorithm in different environmental settings,and the results demonstrate both its adaptability and interpretability.
基金funded by Research Platforms and Projects for Higher Education Institutions of Department of Education of Guangdong Province in 2024(2024KTSCX256)2023 Guangdong Province Higher Vocational Education Teaching Quality and Teaching Reform Project(2023JG080).
文摘The potential toxicity of ionic liquids(ILs)affects their applications;how to control the toxicity is one of the key issues in their applications.To understand its toxicity structure relationship and promote its greener application,six different machine learning algorithms,including Bagging,Adaptive Boosting(AdaBoost),Gradient Boosting(GBoost),Stacking,Voting and Categorical Boosting(CatBoost),are established to model the toxicity of ILs on four distinct datasets including Leukemia rat cell line IPC-81(IPC-81),Acetylcholinesterase(AChE),Escherichia coli(E.coli)and Vibrio fischeri.Molecular descriptors obtained from the simplified molecular input line entry system(SMILES)are used to characterize ILs.All models are assessed by the mean square error(MSE),root mean square error(RMSE),mean absolute error(MAE)and correlation coefficient(R^(2)).Additionally,an interpretation model based on SHapley Additive exPlanations(SHAP)is built to determine the positive and negative effects of each molecular feature on toxicity.With additional parameters and complexity,the Catboost model outperforms the other models,making it a more reliable model for ILs'toxicity prediction.The results of the model's interpretation indicate that the most significant positive features,SMR_VSA5,PEOE_VSA8,Kappa2,PEOE_VSA6,SMR_VSA5,PEOE_VSA6 and EState_VSA1,can increase the toxicity of ILs as their levels rise,while the most significant negative features,VSA_EState7,EState_VSA8,PEOE_VSA9 and FpDensityMorgan1,can decrease the toxicity as their levels rise.Also,an IL's toxicity will grow as its average molecular weight and number of pyridine rings increase,whereas its toxicity will decrease as its hydrogen bond acceptors increase.This finding offers a theoretical foundation for rapid screening and synthesis of environmentally-benign ILs.
文摘Liquid rocket engine(LRE)fault diagnosis is critical for successful space launch missions,enabling timely avoidance of safety hazards,while accurate post-failure analysis prevents subsequent economic losses.However,the complexity of LRE systems and the“black-box”nature of current deep learning-based diagnostic methods hinder interpretable fault diagnosis.This paper establishes Granger causality(GC)extraction-based component-wise multi-layer perceptron(GCMLP),achieving high fault diagnosis accuracy while leveraging GC to enhance diagnostic interpretability.First,component-wise MLP networks are constructed for distinct LRE variables to extract inter-variable GC relationships.Second,dedicated predictors are designed for each variable,leveraging historical data and GC relationships to forecast future states,thereby ensuring GC reliability.Finally,the extracted GC features are utilized for fault classification,guaranteeing feature discriminability and diagnosis accuracy.This study simulates six critical fault modes in LRE using Simulink.Based on the generated simulation data,GCMLP demonstrates superior fault localization accuracy compared to benchmark methods,validating its efficacy and robustness.
基金supported by the National Natural Science Foundation of China(No.52505107)Yunnan Fundamental Research Projects(NO.202501CF070179).
文摘As industrial systems become increasingly complex,the significant research interest has been devoted to intelligent fault diagnosis approaches leveraging deep learning.However,existing methods still face two critical challenges in practical applications:1)the extracted features often fail to maintain robustness in nonstationary conditions;2)deep neural networks generally exhibit a black box nature,offering limited interpretability in their feature extraction process.To solve the above issues,an interpretable wavelet Kolmogorov-Arnold convolutional Long Short-Term Memory(WKAConvLSTM)is proposed,which mainly consists of two key components:1)a wavelet Kolmogorov-Arnold kernel(WKAK)with learnable scale and translation parameters is designed and then embedded into convolutional layers to enable the extracted spatial features interpretable;2)a multi-head attention-enhanced Long Short-Term Memory(MHA-LSTM)is proposed to effectively capture crucial temporal dependencies in sequential data.In order to verify its effectiveness,the proposed model is tested on bearing and gearbox datasets under complex conditions,including noise interference,nonstationary operating conditions,and data class imbalance.The experimental results demonstrate that it not only achieves superior diagnostic accuracy compared with advanced baseline models but also enhances the interpretability of the extracted features.
基金Supported in part by Science Center for Gas Turbine Project(Project No.P2022-DC-I-003-001)National Natural Science Foundation of China(Grant No.52275130).
文摘Despite significant progress in the Prognostics and Health Management(PHM)domain using pattern learning systems from data,machine learning(ML)still faces challenges related to limited generalization and weak interpretability.A promising approach to overcoming these challenges is to embed domain knowledge into the ML pipeline,enhancing the model with additional pattern information.In this paper,we review the latest developments in PHM,encapsulated under the concept of Knowledge Driven Machine Learning(KDML).We propose a hierarchical framework to define KDML in PHM,which includes scientific paradigms,knowledge sources,knowledge representations,and knowledge embedding methods.Using this framework,we examine current research to demonstrate how various forms of knowledge can be integrated into the ML pipeline and provide roadmap to specific usage.Furthermore,we present several case studies that illustrate specific implementations of KDML in the PHM domain,including inductive experience,physical model,and signal processing.We analyze the improvements in generalization capability and interpretability that KDML can achieve.Finally,we discuss the challenges,potential applications,and usage recommendations of KDML in PHM,with a particular focus on the critical need for interpretability to ensure trustworthy deployment of artificial intelligence in PHM.
基金supported in part by the National Key R&D Program of China under Grant 2022YFB4300601in part by the State Key Laboratory of Advanced Rail Autonomous Operation under Grant RAO2023ZZ003.
文摘Intelligent fault diagnosis technology plays an indispensable role in ensuring the safety,stability,and efficiency of railway operations.However,existing studies have the following limitations.1)They are typical black-box models that lacks interpretability as well as they fuse features by simply stacking them,overlooking the discrepancies in the importance of different features,which reduces the credibility and diagnosis accuracy of the models.2)They ignore the effects of potentially mistaken labels in the training datasets disrupting the ability of the models to learn the true data distribution,which degrades the generalization performance of intelligent diagnosis models,especially when the training samples are limited.To address the above items,an interpretable few-shot framework for fault diagnosis with noisy labels is proposed for train transmission systems.In the proposed framework,a feature extractor is constructed by stacked frequency band focus modules,which can capture signal features in different frequency bands and further adaptively concentrate on the features corresponding to the potential fault characteristic frequency.Then,according to prototypical network,a novel metric-based classifier is developed that is tolerant to mislabeled support samples in the case of limited samples.Besides,a new loss function is designed to decrease the impact of label mistakes in query datasets.Finally,fault simulation experiments of subway train transmission systems are designed and conducted,and the effectiveness as well as superiority of the proposed method are proved by ablation experiments and comparison with the existing methods.
