Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using d...Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.展开更多
Accurate and robust detection of wax appearance(a medium-to high-molecular-weight component of crude oil)is crucial for the efficient operation of hydrocarbon transportation.The wax appearance temperature(WAT)is the l...Accurate and robust detection of wax appearance(a medium-to high-molecular-weight component of crude oil)is crucial for the efficient operation of hydrocarbon transportation.The wax appearance temperature(WAT)is the lowest temperature at which the wax begins to form.When crude oil cools to its WAT,wax crystals precipitate,forming deposits on pipelines as the solubility limit is reached.Therefore,WAT is a crucial quality assurance parameter,especially when dealing with modern fuel oil blends.In this study,we use machine learning via MATLAB’s Bioinformatics Toolbox to predict the WAT of marine fuel samples by correlating near-infrared spectral data with laboratory-measured values.The dataset provided by Intertek PLC-a total quality assurance provider of inspection,testing,and certification services-includes industrial data that is imbalanced,with a higher proportion of high-WAT samples compared to low-WAT samples.The objective is to predict marine fuel oil blends with unusually high WAT values(>35℃)without relying on time-consuming and irregular laboratory-based measurements.The results demonstrate that the developed model,based on the one-class support vector machine(OCSVM)algorithm,achieved a Recall of 96,accurately predicting 96%of fuel samples with WAT>35℃.For standard binary classification,the Recall was 85.7.The trained OCSVM model is expected to facilitate rapid and well-informed decision-making for logistics and storage when choosing fuel oils.展开更多
Focusing on strip steel surface defects classification, a novel support vector machine with adjustable hyper-sphere （AHSVM） is formulated. Meanwhile, a new multi-class classification method is proposed. Originated f...Focusing on strip steel surface defects classification, a novel support vector machine with adjustable hyper-sphere （AHSVM） is formulated. Meanwhile, a new multi-class classification method is proposed. Originated from support vector data description, AHSVM adopts hyper-sphere to solve classification problem. AHSVM can obey two principles： the margin maximization and inner-class dispersion minimization. Moreover, the hyper-sphere of AHSVM is adjustable, which makes the final classification hyper-sphere optimal for training dataset. On the other hand, AHSVM is combined with binary tree to solve multi-class classification for steel surface defects. A scheme of samples pruning in mapped feature space is provided, which can reduce the number of training samples under the premise of classification accuracy, resulting in the improvements of classification speed. Finally, some testing experiments are done for eight types of strip steel surface defects. Experimental results show that multi-class AHSVM classifier exhibits satisfactory results in classification accuracy and efficiency.展开更多
Aiming at the problems of the traditional method of assessing distribution of particle size in bench blasting, a support vector machines （SVMs） regression methodology was used to predict the mean particle size （X50...Aiming at the problems of the traditional method of assessing distribution of particle size in bench blasting, a support vector machines （SVMs） regression methodology was used to predict the mean particle size （X50） resulting from rock blast fragmentation in various mines based on the statistical learning theory. The data base consisted of blast design parameters, explosive parameters, modulus of elasticity and in-situ block size. The seven input independent variables used for the SVMs model for the prediction of X50 of rock blast fragmentation were the ratio of bench height to drilled burden （H/B）, ratio of spacing to burden （S/B）, ratio of burden to hole diameter （B/D）, ratio of stemming to burden （T/B）, powder factor （Pf）, modulus of elasticity （E） and in-situ block size （XB）. After using the 90 sets of the measured data in various mines and rock formations in the world for training and testing, the model was applied to 12 another blast data for validation of the trained support vector regression （SVR） model. The prediction results of SVR were compared with those of artificial neural network （ANN）, multivariate regression analysis （MVRA） models, conventional Kuznetsov method and the measured X50 values. The proposed method shows promising results and the prediction accuracy of SVMs model is acceptable.展开更多
Support vector machines (SVMs) are utilized for emotion recognition in Chinese speech in this paper. Both binary class discrimination and the multi class discrimination are discussed. It proves that the emotional fe...Support vector machines (SVMs) are utilized for emotion recognition in Chinese speech in this paper. Both binary class discrimination and the multi class discrimination are discussed. It proves that the emotional features construct a nonlinear problem in the input space, and SVMs based on nonlinear mapping can solve it more effectively than other linear methods. Multi class classification based on SVMs with a soft decision function is constructed to classify the four emotion situations. Compared with principal component analysis (PCA) method and modified PCA method, SVMs perform the best result in multi class discrimination by using nonlinear kernel mapping.展开更多
Support vector machines （SVMs） are combined with the artificial immune network （aiNet）, thus forming a new hybrid ai-SVM algorithm. The algorithm is used to reduce the number of samples and the training time of SV...Support vector machines （SVMs） are combined with the artificial immune network （aiNet）, thus forming a new hybrid ai-SVM algorithm. The algorithm is used to reduce the number of samples and the training time of SVM on large datasets, aiNet is an artificial immune system （AIS） inspired method to perform the automatic data compression, extract the relevant information and retain the topology of the original sample distribution. The output of aiNet is a set of antibodies for representing the input dataset in a simplified way. Then the SVM model is built in the compressed antibody network instead of the original input data. Experimental results show that the ai-SVM algorithm is effective to reduce the computing time and simplify the SVM model, and the accuracy is not decreased.展开更多
Accurately estimating the State of Health(SOH)and Remaining Useful Life(RUL)of lithium-ion batteries(LIBs)is crucial for the continuous and stable operation of battery management systems.However,due to the complex int...Accurately estimating the State of Health(SOH)and Remaining Useful Life(RUL)of lithium-ion batteries(LIBs)is crucial for the continuous and stable operation of battery management systems.However,due to the complex internal chemical systems of LIBs and the nonlinear degradation of their performance,direct measurement of SOH and RUL is challenging.To address these issues,the Twin Support Vector Machine(TWSVM)method is proposed to predict SOH and RUL.Initially,the constant current charging time of the lithium battery is extracted as a health indicator(HI),decomposed using Variational Modal Decomposition(VMD),and feature correlations are computed using Importance of Random Forest Features(RF)to maximize the extraction of critical factors influencing battery performance degradation.Furthermore,to enhance the global search capability of the Convolution Optimization Algorithm(COA),improvements are made using Good Point Set theory and the Differential Evolution method.The Improved Convolution Optimization Algorithm(ICOA)is employed to optimize TWSVM parameters for constructing SOH and RUL prediction models.Finally,the proposed models are validated using NASA and CALCE lithium-ion battery datasets.Experimental results demonstrate that the proposed models achieve an RMSE not exceeding 0.007 and an MAPE not exceeding 0.0082 for SOH and RUL prediction,with a relative error in RUL prediction within the range of[-1.8%,2%].Compared to other models,the proposed model not only exhibits superior fitting capability but also demonstrates robust performance.展开更多
Open networks and heterogeneous services in the Internet of Vehicles(IoV)can lead to security and privacy challenges.One key requirement for such systems is the preservation of user privacy,ensuring a seamless experie...Open networks and heterogeneous services in the Internet of Vehicles(IoV)can lead to security and privacy challenges.One key requirement for such systems is the preservation of user privacy,ensuring a seamless experience in driving,navigation,and communication.These privacy needs are influenced by various factors,such as data collected at different intervals,trip durations,and user interactions.To address this,the paper proposes a Support Vector Machine(SVM)model designed to process large amounts of aggregated data and recommend privacy preserving measures.The model analyzes data based on user demands and interactions with service providers or neighboring infrastructure.It aims to minimize privacy risks while ensuring service continuity and sustainability.The SVMmodel helps validate the system’s reliability by creating a hyperplane that distinguishes between maximum and minimum privacy recommendations.The results demonstrate the effectiveness of the proposed SVM model in enhancing both privacy and service performance.展开更多
The total nitrogen(TN)is a major factor contributing to eutrophication and is a crucial parameter in assessing surface water quality.Accurate and rapid methods are crucial for determining the TN content in water.Herei...The total nitrogen(TN)is a major factor contributing to eutrophication and is a crucial parameter in assessing surface water quality.Accurate and rapid methods are crucial for determining the TN content in water.Herein,a fast,highly sensitive,and pollution-free approach is proposed,which combines ultraviolet(UV)absorption spectroscopy with Bayesian optimized least squares support vector machine(LSSVM)for detecting TN content in water.Water samples collected from sampling points near the Yangtze River basin in Chongqing of China were analyzed using national standard methods to measure TN content as reference values.The prediction of TN content in water was achieved by integrating the UV absorption spectra of water samples with LSSVM.To make the model quickly and accurately select the optimal parameters to improve the accuracy of the prediction model,the Bayesian optimization(BO)algorithm was used to optimize the parameters of the LSSVM.Results show that the prediction model performs well in predicting TN concentration,with a high coefficient of prediction determination(R^(2)=0.9413)and a low root mean square error of prediction(RMSE=0.0779 mg/L).Comparative analysis with previous studies indicates that the model used in this paper achieves lower prediction errors and superior predictive performance.展开更多
Abstract Accurate aerodynamic models are the basis of flight simulation and control law design. Mathematically modeling unsteady aerodynamics at high angles of attack bears great difficulties in model structure determ...Abstract Accurate aerodynamic models are the basis of flight simulation and control law design. Mathematically modeling unsteady aerodynamics at high angles of attack bears great difficulties in model structure determination and parameter estimation due to little understanding of the flow mechanism. Support vector machines （SVMs） based on statistical learning theory provide a novel tool for nonlinear system modeling. The work presented here examines the feasibility of applying SVMs to high angle.-of-attack unsteady aerodynamic modeling field. Mainly, after a review of SVMs, several issues associated with unsteady aerodynamic modeling by use of SVMs are discussed in detail, such as sele, ction of input variables, selection of output variables and determination of SVM parameters. The least squares SVM （LS-SVM） models are set up from certain dynamic wind tunnel test data of a delta wing and an aircraft configuration, and then used to predict the aerodynamic responses in other tests. The predictions are in good agreement with the test data, which indicates the satisfving learning and generalization performance of LS-SVMs.展开更多
In microarray-based cancer classification, gene selection is an important issue owing to the large number of variables and small number of samples as well as its non-linearity. It is difficult to get satisfying result...In microarray-based cancer classification, gene selection is an important issue owing to the large number of variables and small number of samples as well as its non-linearity. It is difficult to get satisfying results by using conventional linear sta- tistical methods. Recursive feature elimination based on support vector machine (SVM RFE) is an effective algorithm for gene selection and cancer classification, which are integrated into a consistent framework. In this paper, we propose a new method to select parameters of the aforementioned algorithm implemented with Gaussian kernel SVMs as better alternatives to the common practice of selecting the apparently best parameters by using a genetic algorithm to search for a couple of optimal parameter. Fast implementation issues for this method are also discussed for pragmatic reasons. The proposed method was tested on two repre- sentative hereditary breast cancer and acute leukaemia datasets. The experimental results indicate that the proposed method per- forms well in selecting genes and achieves high classification accuracies with these genes.展开更多
Miniature air quality sensors are widely used in urban grid-based monitoring due to their flexibility in deployment and low cost.However,the raw data collected by these devices often suffer from low accuracy caused by...Miniature air quality sensors are widely used in urban grid-based monitoring due to their flexibility in deployment and low cost.However,the raw data collected by these devices often suffer from low accuracy caused by environmental interference and sensor drift,highlighting the need for effective calibration methods to improve data reliability.This study proposes a data correction method based on Bayesian Optimization Support Vector Regression(BO-SVR),which combines the nonlinear modeling capability of Support Vector Regression(SVR)with the efficient global hyperparameter search of Bayesian Optimization.By introducing cross-validation loss as the optimization objective and using Gaussian process modeling with an Expected Improvement acquisition strategy,the approach automatically determines optimal hyperparameters for accurate pollutant concentration prediction.Experiments on real-world micro-sensor datasets demonstrate that BO-SVR outperforms traditional SVR,grid search SVR,and random forest(RF)models across multiple pollutants,including PM_(2.5),PM_(10),CO,NO_(2),SO_(2),and O_(3).The proposed method achieves lower prediction residuals,higher fitting accuracy,and better generalization,offering an efficient and practical solution for enhancing the quality of micro-sensor air monitoring data.展开更多
The endpoint parameters are very important to the process of EAF steel-making, but their on-line measurement is difficult. The soft sensor technology is widely used for the prediction of endpoint parameters. Based on ...The endpoint parameters are very important to the process of EAF steel-making, but their on-line measurement is difficult. The soft sensor technology is widely used for the prediction of endpoint parameters. Based on the analysis of the smelting process of EAF and the advantages of support vector machines, a soft sensor model for predicting the endpoint parameters was built using multiple support vector machines （MSVM）. In this model, the input space was divided by subtractive clustering and a sub-model based on LS-SVM was built in each sub-space. To decrease the correlation among the sub-models and to improve the accuracy and robustness of the model, the sub- models were combined by Principal Components Regression. The accuracy of the soft sensor model is perfectly improved. The simulation result demonstrates the practicability and efficiency of the MSVM model for the endpoint prediction of EAF.展开更多
Engine spark ignition is an important source for diagnosis of engine faults.Based on the waveform of the ignition pattern,a mechanic can guess what may be the potential malfunctioning parts of an engine with his/her e...Engine spark ignition is an important source for diagnosis of engine faults.Based on the waveform of the ignition pattern,a mechanic can guess what may be the potential malfunctioning parts of an engine with his/her experience and handbooks.However,this manual diagnostic method is imprecise because many spark ignition patterns are very similar.Therefore,a diagnosis needs many trials to identify the malfunctioning parts.Meanwhile the mechanic needs to disassemble and assemble the engine parts for verification.To tackle this problem,an intelligent diagnosis system was established based on ignition patterns.First,the captured patterns were normalized and compressed.Then wavelet packet transform（WPT） was employed to extract the representative features of the ignition patterns.Finally,a classification system was constructed by using multi-class support vector machines（SVM） and the extracted features.The classification system can intelligently classify the most likely engine fault so as to reduce the number of diagnosis trials.Experimental results show that SVM produces higher diagnosis accuracy than the traditional multilayer feedforward neural network.This is the first trial on the combination of WPT and SVM to analyze ignition patterns and diagnose automotive engines.展开更多
Gears alternately mesh and detach in driving process, and then workingconditions of gears are alternately changing, so they are easy to be spalled and worn. But becauseof the effect of additive gaussian measurement no...Gears alternately mesh and detach in driving process, and then workingconditions of gears are alternately changing, so they are easy to be spalled and worn. But becauseof the effect of additive gaussian measurement noises, the signal-to-noises ratio is low; theirfault features are difficult to extract. This study aims to propose an approach of gear faultsclassification, using the cumulants and support vector machines. The cumulants can eliminate theadditive gaussian noises, boost the signal-to-noises ratio. Generalisation of support vectormachines as classifier, which is employed structural risk minimisation principle, is superior tothat of conventional neural networks, which is employed traditional empirical risk minimisationprinciple. Support vector machines as the classifier, and the third and fourth order cumulants asinput, gears faults are successfully recognized. The experimental results show that the method offault classification combining cumulants with support vector machines is very effective.展开更多
SVMs（support vector machines） is a new artificial intelligence methodology derived from Vapnik＇s statistical learning theory, which has better generalization than artificial neural network. A Csupport vector classi...SVMs（support vector machines） is a new artificial intelligence methodology derived from Vapnik＇s statistical learning theory, which has better generalization than artificial neural network. A Csupport vector classifiers Based Fault Diagnostic Model （CBFDM） which gives the 3 most possible fault causes is constructed in this paper. Five fold cross validation is chosen as the method of model selection for CBFDM. The simulated data are generated from PW4000-94 engine influence coefficient matrix at cruise, and the results show that the diagnostic accuracy of CBFDM is over 93 % even when the standard deviation of noise is 3 times larger than the normal. This model can also be used for other diagnostic problems.展开更多
Support vector machines and a Kalman-like observer are used for fault detection and isolation in a variable speed horizontalaxis wind turbine composed of three blades and a full converter. The support vector approach ...Support vector machines and a Kalman-like observer are used for fault detection and isolation in a variable speed horizontalaxis wind turbine composed of three blades and a full converter. The support vector approach is data-based and is therefore robust to process knowledge. It is based on structural risk minimization which enhances generalization even with small training data set and it allows for process nonlinearity by using flexible kernels. In this work, a radial basis function is used as the kernel. Different parts of the process are investigated including actuators and sensors faults. With duplicated sensors, sensor faults in blade pitch positions,generator and rotor speeds can be detected. Faults of type stuck measurements can be detected in 2 sampling periods. The detection time of offset/scaled measurements depends on the severity of the fault and on the process dynamics when the fault occurs. The converter torque actuator fault can be detected within 2 sampling periods. Faults in the actuators of the pitch systems represents a higher difficulty for fault detection which is due to the fact that such faults only affect the transitory state(which is very fast) but not the final stationary state. Therefore, two methods are considered and compared for fault detection and isolation of this fault: support vector machines and a Kalman-like observer. Advantages and disadvantages of each method are discussed. On one hand, support vector machines training of transitory states would require a big amount of data in different situations, but the fault detection and isolation results are robust to variations in the input/operating point. On the other hand, the observer is model-based, and therefore does not require training, and it allows identification of the fault level, which is interesting for fault reconfiguration. But the observability of the system is ensured under specific conditions, related to the dynamics of the inputs and outputs. The whole fault detection and isolation scheme is evaluated using a wind turbine benchmark with a real sequence of wind speed.展开更多
Based on support vector machines, three modeling methods, i.e., white-box modeling, grey-box modeling and black-box modeling of ship manoeuvring motion in 4 degrees of freedom are investigated. With the whole-ship mat...Based on support vector machines, three modeling methods, i.e., white-box modeling, grey-box modeling and black-box modeling of ship manoeuvring motion in 4 degrees of freedom are investigated. With the whole-ship mathematical model for ship manoeuvring motion, in which the hydrodynamic coefficients are obtained from roll planar motion mechanism test, some zigzag tests and turning circle manoeuvres are simulated. In the white-box modeling and grey-box modeling, the training data taken every 5 s from the simulated 20°/20° zigzag test are used, while in the black-box modeling, the training data taken every 5 s from the simulated 15°/15°, 20°/20° zigzag tests and 15°, 25° turning manoeuvres are used; and the trained support vector machines are used to predict the whole 20°/20° zigzag test. Comparisons between the simulated and predicted 20°/20° zigzag tests show good predictive ability of the proposed methods. Besides, all mathematical models obtained by the proposed modeling methods are used to predict the 10°/10° zigzag test and 35° turning circle manoeuvre, and the predicted results are compared with those of simulation tests to demonstrate the good generalization performance of the mathematical models. Finally, the proposed modeling methods are analyzed and compared with each other in aspects of application conditions, prediction accuracy and computation speed. The appropriate modeling method can be chosen according to the intended use of the mathematical models and the available data needed for system identification.展开更多
Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statisti...Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statistical methods. However, neural networks have inherent drawbacks, such as local optimization solution, lack generalization, and uncontrolled convergence. A relatively new machine learning technique, support vector machine （SVM）, which overcomes the drawbacks of neural networks, is introduced to provide a model with better explanatory power to select ideal supplier partners. Meanwhile, in practice, the suppliers＇ samples are very insufficient. SVMs are adaptive to deal with small samples＇ training and testing. The prediction accuracies for BPNN and SVM methods are compared to choose the appreciating suppliers. The actual examples illustrate that SVM methods are superior to BPNN.展开更多
Since the efficiency of photovoltaic(PV) power is closely related to the weather,many PV enterprises install weather instruments to monitor the working state of the PV power system.With the development of the soft mea...Since the efficiency of photovoltaic(PV) power is closely related to the weather,many PV enterprises install weather instruments to monitor the working state of the PV power system.With the development of the soft measurement technology,the instrumental method seems obsolete and involves high cost.This paper proposes a novel method for predicting the types of weather based on the PV power data and partial meteorological data.By this method,the weather types are deduced by data analysis,instead of weather instrument A better fault detection is obtained by using the support vector machines(SVM) and comparing the predicted and the actual weather.The model of the weather prediction is established by a direct SVM for training multiclass predictors.Although SVM is suitable for classification,the classified results depend on the type of the kernel,the parameters of the kernel,and the soft margin coefficient,which are difficult to choose.In this paper,these parameters are optimized by particle swarm optimization(PSO) algorithm in anticipation of good prediction results can be achieved.Prediction results show that this method is feasible and effective.展开更多
基金The work described in this paper was fully supported by a grant from Hong Kong Metropolitan University(RIF/2021/05).
文摘Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.
基金Newcastle University and EPSRC(Grant No.2020/21 DTP:ref.EP/T517914/1).
文摘Accurate and robust detection of wax appearance(a medium-to high-molecular-weight component of crude oil)is crucial for the efficient operation of hydrocarbon transportation.The wax appearance temperature(WAT)is the lowest temperature at which the wax begins to form.When crude oil cools to its WAT,wax crystals precipitate,forming deposits on pipelines as the solubility limit is reached.Therefore,WAT is a crucial quality assurance parameter,especially when dealing with modern fuel oil blends.In this study,we use machine learning via MATLAB’s Bioinformatics Toolbox to predict the WAT of marine fuel samples by correlating near-infrared spectral data with laboratory-measured values.The dataset provided by Intertek PLC-a total quality assurance provider of inspection,testing,and certification services-includes industrial data that is imbalanced,with a higher proportion of high-WAT samples compared to low-WAT samples.The objective is to predict marine fuel oil blends with unusually high WAT values(>35℃)without relying on time-consuming and irregular laboratory-based measurements.The results demonstrate that the developed model,based on the one-class support vector machine(OCSVM)algorithm,achieved a Recall of 96,accurately predicting 96%of fuel samples with WAT>35℃.For standard binary classification,the Recall was 85.7.The trained OCSVM model is expected to facilitate rapid and well-informed decision-making for logistics and storage when choosing fuel oils.
文摘Focusing on strip steel surface defects classification, a novel support vector machine with adjustable hyper-sphere （AHSVM） is formulated. Meanwhile, a new multi-class classification method is proposed. Originated from support vector data description, AHSVM adopts hyper-sphere to solve classification problem. AHSVM can obey two principles： the margin maximization and inner-class dispersion minimization. Moreover, the hyper-sphere of AHSVM is adjustable, which makes the final classification hyper-sphere optimal for training dataset. On the other hand, AHSVM is combined with binary tree to solve multi-class classification for steel surface defects. A scheme of samples pruning in mapped feature space is provided, which can reduce the number of training samples under the premise of classification accuracy, resulting in the improvements of classification speed. Finally, some testing experiments are done for eight types of strip steel surface defects. Experimental results show that multi-class AHSVM classifier exhibits satisfactory results in classification accuracy and efficiency.
基金Foundation item:Project (2006BAB02A02) supported by the National Key Technology R&D Program during the 11th Five-year Plan Period of ChinaProject (CX2011B119) supported by the Graduated Students' Research and Innovation Fund of Hunan Province, ChinaProject (2009ssxt230) supported by the Central South University Innovation Fund,China
文摘Aiming at the problems of the traditional method of assessing distribution of particle size in bench blasting, a support vector machines （SVMs） regression methodology was used to predict the mean particle size （X50） resulting from rock blast fragmentation in various mines based on the statistical learning theory. The data base consisted of blast design parameters, explosive parameters, modulus of elasticity and in-situ block size. The seven input independent variables used for the SVMs model for the prediction of X50 of rock blast fragmentation were the ratio of bench height to drilled burden （H/B）, ratio of spacing to burden （S/B）, ratio of burden to hole diameter （B/D）, ratio of stemming to burden （T/B）, powder factor （Pf）, modulus of elasticity （E） and in-situ block size （XB）. After using the 90 sets of the measured data in various mines and rock formations in the world for training and testing, the model was applied to 12 another blast data for validation of the trained support vector regression （SVR） model. The prediction results of SVR were compared with those of artificial neural network （ANN）, multivariate regression analysis （MVRA） models, conventional Kuznetsov method and the measured X50 values. The proposed method shows promising results and the prediction accuracy of SVMs model is acceptable.
文摘Support vector machines (SVMs) are utilized for emotion recognition in Chinese speech in this paper. Both binary class discrimination and the multi class discrimination are discussed. It proves that the emotional features construct a nonlinear problem in the input space, and SVMs based on nonlinear mapping can solve it more effectively than other linear methods. Multi class classification based on SVMs with a soft decision function is constructed to classify the four emotion situations. Compared with principal component analysis (PCA) method and modified PCA method, SVMs perform the best result in multi class discrimination by using nonlinear kernel mapping.
文摘Support vector machines （SVMs） are combined with the artificial immune network （aiNet）, thus forming a new hybrid ai-SVM algorithm. The algorithm is used to reduce the number of samples and the training time of SVM on large datasets, aiNet is an artificial immune system （AIS） inspired method to perform the automatic data compression, extract the relevant information and retain the topology of the original sample distribution. The output of aiNet is a set of antibodies for representing the input dataset in a simplified way. Then the SVM model is built in the compressed antibody network instead of the original input data. Experimental results show that the ai-SVM algorithm is effective to reduce the computing time and simplify the SVM model, and the accuracy is not decreased.
基金funded by the Pyramid Talent Training Project of Beijing University of Civil Engineering and Architecture under Grant GJZJ20220802。
文摘Accurately estimating the State of Health(SOH)and Remaining Useful Life(RUL)of lithium-ion batteries(LIBs)is crucial for the continuous and stable operation of battery management systems.However,due to the complex internal chemical systems of LIBs and the nonlinear degradation of their performance,direct measurement of SOH and RUL is challenging.To address these issues,the Twin Support Vector Machine(TWSVM)method is proposed to predict SOH and RUL.Initially,the constant current charging time of the lithium battery is extracted as a health indicator(HI),decomposed using Variational Modal Decomposition(VMD),and feature correlations are computed using Importance of Random Forest Features(RF)to maximize the extraction of critical factors influencing battery performance degradation.Furthermore,to enhance the global search capability of the Convolution Optimization Algorithm(COA),improvements are made using Good Point Set theory and the Differential Evolution method.The Improved Convolution Optimization Algorithm(ICOA)is employed to optimize TWSVM parameters for constructing SOH and RUL prediction models.Finally,the proposed models are validated using NASA and CALCE lithium-ion battery datasets.Experimental results demonstrate that the proposed models achieve an RMSE not exceeding 0.007 and an MAPE not exceeding 0.0082 for SOH and RUL prediction,with a relative error in RUL prediction within the range of[-1.8%,2%].Compared to other models,the proposed model not only exhibits superior fitting capability but also demonstrates robust performance.
基金supported by the Deanship of Graduate Studies and Scientific Research at University of Bisha for funding this research through the promising program under grant number(UB-Promising-33-1445).
文摘Open networks and heterogeneous services in the Internet of Vehicles(IoV)can lead to security and privacy challenges.One key requirement for such systems is the preservation of user privacy,ensuring a seamless experience in driving,navigation,and communication.These privacy needs are influenced by various factors,such as data collected at different intervals,trip durations,and user interactions.To address this,the paper proposes a Support Vector Machine(SVM)model designed to process large amounts of aggregated data and recommend privacy preserving measures.The model analyzes data based on user demands and interactions with service providers or neighboring infrastructure.It aims to minimize privacy risks while ensuring service continuity and sustainability.The SVMmodel helps validate the system’s reliability by creating a hyperplane that distinguishes between maximum and minimum privacy recommendations.The results demonstrate the effectiveness of the proposed SVM model in enhancing both privacy and service performance.
基金supported by the National Natural Science Foundation of China(Nos.32171627 and 62105252)the Science and Technology Research Program of Chongqing Municipal Education Commission(No.KJZD-M202200602)the Hangzhou Science and Technology Development Project(No.202204T04).
文摘The total nitrogen(TN)is a major factor contributing to eutrophication and is a crucial parameter in assessing surface water quality.Accurate and rapid methods are crucial for determining the TN content in water.Herein,a fast,highly sensitive,and pollution-free approach is proposed,which combines ultraviolet(UV)absorption spectroscopy with Bayesian optimized least squares support vector machine(LSSVM)for detecting TN content in water.Water samples collected from sampling points near the Yangtze River basin in Chongqing of China were analyzed using national standard methods to measure TN content as reference values.The prediction of TN content in water was achieved by integrating the UV absorption spectra of water samples with LSSVM.To make the model quickly and accurately select the optimal parameters to improve the accuracy of the prediction model,the Bayesian optimization(BO)algorithm was used to optimize the parameters of the LSSVM.Results show that the prediction model performs well in predicting TN concentration,with a high coefficient of prediction determination(R^(2)=0.9413)and a low root mean square error of prediction(RMSE=0.0779 mg/L).Comparative analysis with previous studies indicates that the model used in this paper achieves lower prediction errors and superior predictive performance.
文摘Abstract Accurate aerodynamic models are the basis of flight simulation and control law design. Mathematically modeling unsteady aerodynamics at high angles of attack bears great difficulties in model structure determination and parameter estimation due to little understanding of the flow mechanism. Support vector machines （SVMs） based on statistical learning theory provide a novel tool for nonlinear system modeling. The work presented here examines the feasibility of applying SVMs to high angle.-of-attack unsteady aerodynamic modeling field. Mainly, after a review of SVMs, several issues associated with unsteady aerodynamic modeling by use of SVMs are discussed in detail, such as sele, ction of input variables, selection of output variables and determination of SVM parameters. The least squares SVM （LS-SVM） models are set up from certain dynamic wind tunnel test data of a delta wing and an aircraft configuration, and then used to predict the aerodynamic responses in other tests. The predictions are in good agreement with the test data, which indicates the satisfving learning and generalization performance of LS-SVMs.
基金Project supported by the National Basic Research Program (973) of China (No. 2002CB312200) and the Center for Bioinformatics Pro-gram Grant of Harvard Center of Neurodegeneration and Repair,Harvard Medical School, Harvard University, Boston, USA
文摘In microarray-based cancer classification, gene selection is an important issue owing to the large number of variables and small number of samples as well as its non-linearity. It is difficult to get satisfying results by using conventional linear sta- tistical methods. Recursive feature elimination based on support vector machine (SVM RFE) is an effective algorithm for gene selection and cancer classification, which are integrated into a consistent framework. In this paper, we propose a new method to select parameters of the aforementioned algorithm implemented with Gaussian kernel SVMs as better alternatives to the common practice of selecting the apparently best parameters by using a genetic algorithm to search for a couple of optimal parameter. Fast implementation issues for this method are also discussed for pragmatic reasons. The proposed method was tested on two repre- sentative hereditary breast cancer and acute leukaemia datasets. The experimental results indicate that the proposed method per- forms well in selecting genes and achieves high classification accuracies with these genes.
文摘Miniature air quality sensors are widely used in urban grid-based monitoring due to their flexibility in deployment and low cost.However,the raw data collected by these devices often suffer from low accuracy caused by environmental interference and sensor drift,highlighting the need for effective calibration methods to improve data reliability.This study proposes a data correction method based on Bayesian Optimization Support Vector Regression(BO-SVR),which combines the nonlinear modeling capability of Support Vector Regression(SVR)with the efficient global hyperparameter search of Bayesian Optimization.By introducing cross-validation loss as the optimization objective and using Gaussian process modeling with an Expected Improvement acquisition strategy,the approach automatically determines optimal hyperparameters for accurate pollutant concentration prediction.Experiments on real-world micro-sensor datasets demonstrate that BO-SVR outperforms traditional SVR,grid search SVR,and random forest(RF)models across multiple pollutants,including PM_(2.5),PM_(10),CO,NO_(2),SO_(2),and O_(3).The proposed method achieves lower prediction residuals,higher fitting accuracy,and better generalization,offering an efficient and practical solution for enhancing the quality of micro-sensor air monitoring data.
基金Item Sponsored by National Natural Science Foundation of China (60374003)
文摘The endpoint parameters are very important to the process of EAF steel-making, but their on-line measurement is difficult. The soft sensor technology is widely used for the prediction of endpoint parameters. Based on the analysis of the smelting process of EAF and the advantages of support vector machines, a soft sensor model for predicting the endpoint parameters was built using multiple support vector machines （MSVM）. In this model, the input space was divided by subtractive clustering and a sub-model based on LS-SVM was built in each sub-space. To decrease the correlation among the sub-models and to improve the accuracy and robustness of the model, the sub- models were combined by Principal Components Regression. The accuracy of the soft sensor model is perfectly improved. The simulation result demonstrates the practicability and efficiency of the MSVM model for the endpoint prediction of EAF.
基金supported by University of Macao Research Grant,China (Grant No. RG057/08-09S/VCM/FST, Grant No. UL011/09-Y1/ EME/ WPK01/FST)
文摘Engine spark ignition is an important source for diagnosis of engine faults.Based on the waveform of the ignition pattern,a mechanic can guess what may be the potential malfunctioning parts of an engine with his/her experience and handbooks.However,this manual diagnostic method is imprecise because many spark ignition patterns are very similar.Therefore,a diagnosis needs many trials to identify the malfunctioning parts.Meanwhile the mechanic needs to disassemble and assemble the engine parts for verification.To tackle this problem,an intelligent diagnosis system was established based on ignition patterns.First,the captured patterns were normalized and compressed.Then wavelet packet transform（WPT） was employed to extract the representative features of the ignition patterns.Finally,a classification system was constructed by using multi-class support vector machines（SVM） and the extracted features.The classification system can intelligently classify the most likely engine fault so as to reduce the number of diagnosis trials.Experimental results show that SVM produces higher diagnosis accuracy than the traditional multilayer feedforward neural network.This is the first trial on the combination of WPT and SVM to analyze ignition patterns and diagnose automotive engines.
基金This project is supported by 95 Pandeng Preselect Project (No.PD9521908)and 973 Project(No.G199802320).
文摘Gears alternately mesh and detach in driving process, and then workingconditions of gears are alternately changing, so they are easy to be spalled and worn. But becauseof the effect of additive gaussian measurement noises, the signal-to-noises ratio is low; theirfault features are difficult to extract. This study aims to propose an approach of gear faultsclassification, using the cumulants and support vector machines. The cumulants can eliminate theadditive gaussian noises, boost the signal-to-noises ratio. Generalisation of support vectormachines as classifier, which is employed structural risk minimisation principle, is superior tothat of conventional neural networks, which is employed traditional empirical risk minimisationprinciple. Support vector machines as the classifier, and the third and fourth order cumulants asinput, gears faults are successfully recognized. The experimental results show that the method offault classification combining cumulants with support vector machines is very effective.
文摘SVMs（support vector machines） is a new artificial intelligence methodology derived from Vapnik＇s statistical learning theory, which has better generalization than artificial neural network. A Csupport vector classifiers Based Fault Diagnostic Model （CBFDM） which gives the 3 most possible fault causes is constructed in this paper. Five fold cross validation is chosen as the method of model selection for CBFDM. The simulated data are generated from PW4000-94 engine influence coefficient matrix at cruise, and the results show that the diagnostic accuracy of CBFDM is over 93 % even when the standard deviation of noise is 3 times larger than the normal. This model can also be used for other diagnostic problems.
文摘Support vector machines and a Kalman-like observer are used for fault detection and isolation in a variable speed horizontalaxis wind turbine composed of three blades and a full converter. The support vector approach is data-based and is therefore robust to process knowledge. It is based on structural risk minimization which enhances generalization even with small training data set and it allows for process nonlinearity by using flexible kernels. In this work, a radial basis function is used as the kernel. Different parts of the process are investigated including actuators and sensors faults. With duplicated sensors, sensor faults in blade pitch positions,generator and rotor speeds can be detected. Faults of type stuck measurements can be detected in 2 sampling periods. The detection time of offset/scaled measurements depends on the severity of the fault and on the process dynamics when the fault occurs. The converter torque actuator fault can be detected within 2 sampling periods. Faults in the actuators of the pitch systems represents a higher difficulty for fault detection which is due to the fact that such faults only affect the transitory state(which is very fast) but not the final stationary state. Therefore, two methods are considered and compared for fault detection and isolation of this fault: support vector machines and a Kalman-like observer. Advantages and disadvantages of each method are discussed. On one hand, support vector machines training of transitory states would require a big amount of data in different situations, but the fault detection and isolation results are robust to variations in the input/operating point. On the other hand, the observer is model-based, and therefore does not require training, and it allows identification of the fault level, which is interesting for fault reconfiguration. But the observability of the system is ensured under specific conditions, related to the dynamics of the inputs and outputs. The whole fault detection and isolation scheme is evaluated using a wind turbine benchmark with a real sequence of wind speed.
基金financially supported by the National Natural Science Foundation of China(Grant No.51279106)the Special Research Fund for the Doctoral Program of Higher Education of China(Grant No.20110073110009)
文摘Based on support vector machines, three modeling methods, i.e., white-box modeling, grey-box modeling and black-box modeling of ship manoeuvring motion in 4 degrees of freedom are investigated. With the whole-ship mathematical model for ship manoeuvring motion, in which the hydrodynamic coefficients are obtained from roll planar motion mechanism test, some zigzag tests and turning circle manoeuvres are simulated. In the white-box modeling and grey-box modeling, the training data taken every 5 s from the simulated 20°/20° zigzag test are used, while in the black-box modeling, the training data taken every 5 s from the simulated 15°/15°, 20°/20° zigzag tests and 15°, 25° turning manoeuvres are used; and the trained support vector machines are used to predict the whole 20°/20° zigzag test. Comparisons between the simulated and predicted 20°/20° zigzag tests show good predictive ability of the proposed methods. Besides, all mathematical models obtained by the proposed modeling methods are used to predict the 10°/10° zigzag test and 35° turning circle manoeuvre, and the predicted results are compared with those of simulation tests to demonstrate the good generalization performance of the mathematical models. Finally, the proposed modeling methods are analyzed and compared with each other in aspects of application conditions, prediction accuracy and computation speed. The appropriate modeling method can be chosen according to the intended use of the mathematical models and the available data needed for system identification.
文摘Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statistical methods. However, neural networks have inherent drawbacks, such as local optimization solution, lack generalization, and uncontrolled convergence. A relatively new machine learning technique, support vector machine （SVM）, which overcomes the drawbacks of neural networks, is introduced to provide a model with better explanatory power to select ideal supplier partners. Meanwhile, in practice, the suppliers＇ samples are very insufficient. SVMs are adaptive to deal with small samples＇ training and testing. The prediction accuracies for BPNN and SVM methods are compared to choose the appreciating suppliers. The actual examples illustrate that SVM methods are superior to BPNN.
基金supported by the National Natural Science Foundation of China(61433004,61473069)IAPI Fundamental Research Funds(2013ZCX14)+1 种基金supported by the Development Project of Key Laboratory of Liaoning Provincethe Enterprise Postdoctoral Fund Projects of Liaoning Province
文摘Since the efficiency of photovoltaic(PV) power is closely related to the weather,many PV enterprises install weather instruments to monitor the working state of the PV power system.With the development of the soft measurement technology,the instrumental method seems obsolete and involves high cost.This paper proposes a novel method for predicting the types of weather based on the PV power data and partial meteorological data.By this method,the weather types are deduced by data analysis,instead of weather instrument A better fault detection is obtained by using the support vector machines(SVM) and comparing the predicted and the actual weather.The model of the weather prediction is established by a direct SVM for training multiclass predictors.Although SVM is suitable for classification,the classified results depend on the type of the kernel,the parameters of the kernel,and the soft margin coefficient,which are difficult to choose.In this paper,these parameters are optimized by particle swarm optimization(PSO) algorithm in anticipation of good prediction results can be achieved.Prediction results show that this method is feasible and effective.
