This paper studies the generalization capability of feedforward neural networks (FNN).The mechanism of FNNs for classification is investigated from the geometric and probabilistic viewpoints. It is pointed out that th...This paper studies the generalization capability of feedforward neural networks (FNN).The mechanism of FNNs for classification is investigated from the geometric and probabilistic viewpoints. It is pointed out that the outputs of the output layer in the FNNs for classification correspond to the estimates of posteriori probability of the input pattern samples with desired outputs 1 or 0. The theorem for the generalized kernel function in the radial basis function networks (RBFN) is given. For an 2-layer perceptron network (2-LPN). an idea of using extended samples to improve generalization capability is proposed. Finally. the experimental results of radar target classification are given to verify the generaliztion capability of the RBFNs.展开更多
Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering.Recent deep learning technique has been receiving more interest due to the g...Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering.Recent deep learning technique has been receiving more interest due to the great potential to deal with pore pressure prediction.However,most of the traditional deep learning models are less efficient to address generalization problems.To fill this technical gap,in this work,we developed a new adaptive physics-informed deep learning model with high generalization capability to predict pore pressure values directly from seismic data.Specifically,the new model,named CGP-NN,consists of a novel parametric features extraction approach(1DCPP),a stacked multilayer gated recurrent model(multilayer GRU),and an adaptive physics-informed loss function.Through machine training,the developed model can automatically select the optimal physical model to constrain the results for each pore pressure prediction.The CGP-NN model has the best generalization when the physicsrelated metricλ=0.5.A hybrid approach combining Eaton and Bowers methods is also proposed to build machine-learnable labels for solving the problem of few labels.To validate the developed model and methodology,a case study on a complex reservoir in Tarim Basin was further performed to demonstrate the high accuracy on the pore pressure prediction of new wells along with the strong generalization ability.The adaptive physics-informed deep learning approach presented here has potential application in the prediction of pore pressures coupled with multiple genesis mechanisms using seismic data.展开更多
Rolling force for strip casting of 1Cr17 ferritic stainless steel was predicted using theoretical model and artificial intelligence.Solution zone was classified into two parts by kiss point position during casting str...Rolling force for strip casting of 1Cr17 ferritic stainless steel was predicted using theoretical model and artificial intelligence.Solution zone was classified into two parts by kiss point position during casting strip.Navier-Stokes equation in fluid mechanics and stream function were introduced to analyze the rheological property of liquid zone and mushy zone,and deduce the analytic equation of unit compression stress distribution.The traditional hot rolling model was still used in the solid zone.Neural networks based on feedforward training algorithm in Bayesian regularization were introduced to build model for kiss point position.The results show that calculation accuracy for verification data of 94.67% is in the range of ±7.0%,which indicates that the predicting accuracy of this model is very high.展开更多
Geological stratification interpretation is a critical task in oil and gas exploration,aiming to delineate subsurface formations based on well logging data and provide a structural framework for drilling and reservoir...Geological stratification interpretation is a critical task in oil and gas exploration,aiming to delineate subsurface formations based on well logging data and provide a structural framework for drilling and reservoir development.Traditional stratification methods rely heavily on manual interpretation,which is subjective,laborintensive,and often inconsistent,making it inadequate for complex geological settings.To address these limitations,this study proposes a fine-scale stratification method based on a Multi-Layer Perceptron(MLP),incorporating both intra-well and inter-well stratigraphic constraints into the model architecture.The proposed approach begins with principal component analysis(PCA)to reduce the dimensionality of logging parameters while retaining key geological features.Selected input features include well location,depth,drilling time,gamma ray logs,and lithology logs.An MLP model is constructed,and a custom loss function integrates stratigraphic consistency both within single wells and across multiple wells to improve formation boundary prediction.Furthermore,the study introduces a spatial segmentation strategy based on well locations to evaluate both interpolation performance within known areas and extrapolation capability to unseen regions.A case study in a coalbed methane block of the Ordos Basin demonstrates the effectiveness of the method.The model achieves a prediction accuracy of up to 95.04%in stratigraphic regions similar to the training data.Even when applied to extrapolated areas with well distances of approximately 1500–2000 meters from the nearest training point,the model maintains an accuracy of 85.36%.These results indicate that the proposed method not only delivers high precision in familiar formations but also generalizes well to new drilling areas.In conclusion,the MLP-based stratification model developed in this study reduces reliance on expert knowledge and exhibits strong performance in both precision and generalization.It provides a practical and reliable tool for automated stratigraphic interpretation and can support the planning of infill wells and the development of new blocks.展开更多
Accurate chiller performance prediction is crucial for improving the energy efficiency of heating,ventilation,and air conditioning(HVAC)systems.Data-driven models commonly used to enhance chiller performance often rel...Accurate chiller performance prediction is crucial for improving the energy efficiency of heating,ventilation,and air conditioning(HVAC)systems.Data-driven models commonly used to enhance chiller performance often rely on sparse data collected under restricted conditions.These models must extrapolate beyond their training data in practical applications,but they generally lack the generalization capability needed for reliable predictions outside their training range.Additionally,their limited interpretability hampers understanding of the physical processes affecting chiller performance,complicating fault identification and performance optimization.To address these issues,this study embeds physical neurons in physics-informed neural networks(EP-PINNs)to enhance chiller performance prediction.By leveraging prior physical knowledge,physical neurons are introduced and embedded into the neural network,forming a neural network architecture with intrinsic physics-based information flow.Simultaneously,simplified physical loss terms are used to guide the training process.The proposed EP-PINNs were applied to predict the performance of four different chillers,and the results demonstrated their high prediction accuracy.Compared to data-driven models,the EP-PINNs exhibited significantly improved generalization capability and interpretability.These advantages highlight the practical value of EP-PINNs in HVAC equipment performance prediction.展开更多
The synthetic minority oversampling technique(SMOTE) is a popular algorithm to reduce the impact of class imbalance in building classifiers, and has received several enhancements over the past 20 years. SMOTE and its ...The synthetic minority oversampling technique(SMOTE) is a popular algorithm to reduce the impact of class imbalance in building classifiers, and has received several enhancements over the past 20 years. SMOTE and its variants synthesize a number of minority-class sample points in the original sample space to alleviate the adverse effects of class imbalance. This approach works well in many cases, but problems arise when synthetic sample points are generated in overlapping areas between different classes, which further complicates classifier training. To address this issue, this paper proposes a novel generalization-oriented rather than imputation-oriented minorityclass sample point generation algorithm, named overlapping minimization SMOTE(OM-SMOTE). This algorithm is designed specifically for binary imbalanced classification problems. OM-SMOTE first maps the original sample points into a new sample space by balancing sample encoding and classifier generalization. Then, OM-SMOTE employs a set of sophisticated minority-class sample point imputation rules to generate synthetic sample points that are as far as possible from overlapping areas between classes. Extensive experiments have been conducted on 32 imbalanced datasets to validate the effectiveness of OM-SMOTE. Results show that using OM-SMOTE to generate synthetic minority-class sample points leads to better classifier training performances for the naive Bayes,support vector machine, decision tree, and logistic regression classifiers than the 11 state-of-the-art SMOTE-based imputation algorithms. This demonstrates that OM-SMOTE is a viable approach for supporting the training of high-quality classifiers for imbalanced classification. The implementation of OM-SMOTE is shared publicly on the Git Hub platform at https://github.com/luxuan123123/OM-SMOTE/.展开更多
Because of overfitting and the improvement of generalization capability (GC)available in the construction of forecasting models using artificial neural network (ANN), a newmethod is proposed for model establishment by...Because of overfitting and the improvement of generalization capability (GC)available in the construction of forecasting models using artificial neural network (ANN), a newmethod is proposed for model establishment by means of making a low-dimension ANN learning matrixthrough principal component analysis (PCA). The results show that the PC A is able to construct anANN model without the need of finding an optimal structure with the appropriate number ofhidden-layer nodes, thus avoids overfitting by condensing forecasting information, reducingdimension and removing noise, and GC is greatly raised compared to the traditional ANN and stepwiseregression techniques for model establishment.展开更多
文摘This paper studies the generalization capability of feedforward neural networks (FNN).The mechanism of FNNs for classification is investigated from the geometric and probabilistic viewpoints. It is pointed out that the outputs of the output layer in the FNNs for classification correspond to the estimates of posteriori probability of the input pattern samples with desired outputs 1 or 0. The theorem for the generalized kernel function in the radial basis function networks (RBFN) is given. For an 2-layer perceptron network (2-LPN). an idea of using extended samples to improve generalization capability is proposed. Finally. the experimental results of radar target classification are given to verify the generaliztion capability of the RBFNs.
基金funded by the National Natural Science Foundation of China(General Program:No.52074314,No.U19B6003-05)National Key Research and Development Program of China(2019YFA0708303-05)。
文摘Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering.Recent deep learning technique has been receiving more interest due to the great potential to deal with pore pressure prediction.However,most of the traditional deep learning models are less efficient to address generalization problems.To fill this technical gap,in this work,we developed a new adaptive physics-informed deep learning model with high generalization capability to predict pore pressure values directly from seismic data.Specifically,the new model,named CGP-NN,consists of a novel parametric features extraction approach(1DCPP),a stacked multilayer gated recurrent model(multilayer GRU),and an adaptive physics-informed loss function.Through machine training,the developed model can automatically select the optimal physical model to constrain the results for each pore pressure prediction.The CGP-NN model has the best generalization when the physicsrelated metricλ=0.5.A hybrid approach combining Eaton and Bowers methods is also proposed to build machine-learnable labels for solving the problem of few labels.To validate the developed model and methodology,a case study on a complex reservoir in Tarim Basin was further performed to demonstrate the high accuracy on the pore pressure prediction of new wells along with the strong generalization ability.The adaptive physics-informed deep learning approach presented here has potential application in the prediction of pore pressures coupled with multiple genesis mechanisms using seismic data.
基金Project(2004CB619108) supported by National Basic Research Program of China
文摘Rolling force for strip casting of 1Cr17 ferritic stainless steel was predicted using theoretical model and artificial intelligence.Solution zone was classified into two parts by kiss point position during casting strip.Navier-Stokes equation in fluid mechanics and stream function were introduced to analyze the rheological property of liquid zone and mushy zone,and deduce the analytic equation of unit compression stress distribution.The traditional hot rolling model was still used in the solid zone.Neural networks based on feedforward training algorithm in Bayesian regularization were introduced to build model for kiss point position.The results show that calculation accuracy for verification data of 94.67% is in the range of ±7.0%,which indicates that the predicting accuracy of this model is very high.
基金National Key R&D Program(Geological Resource Intelligent Prediction and Exploration Development Geological Risk Assessment Theory Models and Methods,Project No.2022YFF0801202)for providing significant research fund-ing and technical support。
文摘Geological stratification interpretation is a critical task in oil and gas exploration,aiming to delineate subsurface formations based on well logging data and provide a structural framework for drilling and reservoir development.Traditional stratification methods rely heavily on manual interpretation,which is subjective,laborintensive,and often inconsistent,making it inadequate for complex geological settings.To address these limitations,this study proposes a fine-scale stratification method based on a Multi-Layer Perceptron(MLP),incorporating both intra-well and inter-well stratigraphic constraints into the model architecture.The proposed approach begins with principal component analysis(PCA)to reduce the dimensionality of logging parameters while retaining key geological features.Selected input features include well location,depth,drilling time,gamma ray logs,and lithology logs.An MLP model is constructed,and a custom loss function integrates stratigraphic consistency both within single wells and across multiple wells to improve formation boundary prediction.Furthermore,the study introduces a spatial segmentation strategy based on well locations to evaluate both interpolation performance within known areas and extrapolation capability to unseen regions.A case study in a coalbed methane block of the Ordos Basin demonstrates the effectiveness of the method.The model achieves a prediction accuracy of up to 95.04%in stratigraphic regions similar to the training data.Even when applied to extrapolated areas with well distances of approximately 1500–2000 meters from the nearest training point,the model maintains an accuracy of 85.36%.These results indicate that the proposed method not only delivers high precision in familiar formations but also generalizes well to new drilling areas.In conclusion,the MLP-based stratification model developed in this study reduces reliance on expert knowledge and exhibits strong performance in both precision and generalization.It provides a practical and reliable tool for automated stratigraphic interpretation and can support the planning of infill wells and the development of new blocks.
基金supported by the National Natural Science Foundation of China(No.22441020).
文摘Accurate chiller performance prediction is crucial for improving the energy efficiency of heating,ventilation,and air conditioning(HVAC)systems.Data-driven models commonly used to enhance chiller performance often rely on sparse data collected under restricted conditions.These models must extrapolate beyond their training data in practical applications,but they generally lack the generalization capability needed for reliable predictions outside their training range.Additionally,their limited interpretability hampers understanding of the physical processes affecting chiller performance,complicating fault identification and performance optimization.To address these issues,this study embeds physical neurons in physics-informed neural networks(EP-PINNs)to enhance chiller performance prediction.By leveraging prior physical knowledge,physical neurons are introduced and embedded into the neural network,forming a neural network architecture with intrinsic physics-based information flow.Simultaneously,simplified physical loss terms are used to guide the training process.The proposed EP-PINNs were applied to predict the performance of four different chillers,and the results demonstrated their high prediction accuracy.Compared to data-driven models,the EP-PINNs exhibited significantly improved generalization capability and interpretability.These advantages highlight the practical value of EP-PINNs in HVAC equipment performance prediction.
基金Project supported by the National Natural Science Foundation of China(No.61972261)the Natural Science Foundation of Guangdong Province,China(No.2023A1515011667)+1 种基金the Key Basic Research Foundation of Shenzhen,China(No.JCYJ20220818100205012)the Basic Research Foundation of Shenzhen,China(No.JCYJ20210324093609026)。
文摘The synthetic minority oversampling technique(SMOTE) is a popular algorithm to reduce the impact of class imbalance in building classifiers, and has received several enhancements over the past 20 years. SMOTE and its variants synthesize a number of minority-class sample points in the original sample space to alleviate the adverse effects of class imbalance. This approach works well in many cases, but problems arise when synthetic sample points are generated in overlapping areas between different classes, which further complicates classifier training. To address this issue, this paper proposes a novel generalization-oriented rather than imputation-oriented minorityclass sample point generation algorithm, named overlapping minimization SMOTE(OM-SMOTE). This algorithm is designed specifically for binary imbalanced classification problems. OM-SMOTE first maps the original sample points into a new sample space by balancing sample encoding and classifier generalization. Then, OM-SMOTE employs a set of sophisticated minority-class sample point imputation rules to generate synthetic sample points that are as far as possible from overlapping areas between classes. Extensive experiments have been conducted on 32 imbalanced datasets to validate the effectiveness of OM-SMOTE. Results show that using OM-SMOTE to generate synthetic minority-class sample points leads to better classifier training performances for the naive Bayes,support vector machine, decision tree, and logistic regression classifiers than the 11 state-of-the-art SMOTE-based imputation algorithms. This demonstrates that OM-SMOTE is a viable approach for supporting the training of high-quality classifiers for imbalanced classification. The implementation of OM-SMOTE is shared publicly on the Git Hub platform at https://github.com/luxuan123123/OM-SMOTE/.
基金This work is sponsored by the Ministry of Science and Technology of China Project "2004 DIB3J122"
文摘Because of overfitting and the improvement of generalization capability (GC)available in the construction of forecasting models using artificial neural network (ANN), a newmethod is proposed for model establishment by means of making a low-dimension ANN learning matrixthrough principal component analysis (PCA). The results show that the PC A is able to construct anANN model without the need of finding an optimal structure with the appropriate number ofhidden-layer nodes, thus avoids overfitting by condensing forecasting information, reducingdimension and removing noise, and GC is greatly raised compared to the traditional ANN and stepwiseregression techniques for model establishment.
