As the core component of inertial navigation systems, fiber optic gyroscope (FOG), with technical advantages such as low power consumption, long lifespan, fast startup speed, and flexible structural design, are widely...As the core component of inertial navigation systems, fiber optic gyroscope (FOG), with technical advantages such as low power consumption, long lifespan, fast startup speed, and flexible structural design, are widely used in aerospace, unmanned driving, and other fields. However, due to the temper-ature sensitivity of optical devices, the influence of environmen-tal temperature causes errors in FOG, thereby greatly limiting their output accuracy. This work researches on machine-learn-ing based temperature error compensation techniques for FOG. Specifically, it focuses on compensating for the bias errors gen-erated in the fiber ring due to the Shupe effect. This work pro-poses a composite model based on k-means clustering, sup-port vector regression, and particle swarm optimization algo-rithms. And it significantly reduced redundancy within the sam-ples by adopting the interval sequence sample. Moreover, met-rics such as root mean square error (RMSE), mean absolute error (MAE), bias stability, and Allan variance, are selected to evaluate the model’s performance and compensation effective-ness. This work effectively enhances the consistency between data and models across different temperature ranges and tem-perature gradients, improving the bias stability of the FOG from 0.022 °/h to 0.006 °/h. Compared to the existing methods utiliz-ing a single machine learning model, the proposed method increases the bias stability of the compensated FOG from 57.11% to 71.98%, and enhances the suppression of rate ramp noise coefficient from 2.29% to 14.83%. This work improves the accuracy of FOG after compensation, providing theoretical guid-ance and technical references for sensors error compensation work in other fields.展开更多
Piezo actuators are widely used in ultra-precision fields because of their high response and nano-scale step length.However,their hysteresis characteristics seriously affect the accuracy and stability of piezo actuato...Piezo actuators are widely used in ultra-precision fields because of their high response and nano-scale step length.However,their hysteresis characteristics seriously affect the accuracy and stability of piezo actuators.Existing methods for fitting hysteresis loops include operator class,differential equation class,and machine learning class.The modeling cost of operator class and differential equation class methods is high,the model complexity is high,and the process of machine learning,such as neural network calculation,is opaque.The physical model framework cannot be directly extracted.Therefore,the sparse identification of nonlinear dynamics(SINDy)algorithm is proposed to fit hysteresis loops.Furthermore,the SINDy algorithm is improved.While the SINDy algorithm builds an orthogonal candidate database for modeling,the sparse regression model is simplified,and the Relay operator is introduced for piecewise fitting to solve the distortion problem of the SINDy algorithm fitting singularities.The Relay-SINDy algorithm proposed in this paper is applied to fitting hysteresis loops.Good performance is obtained with the experimental results of open and closed loops.Compared with the existing methods,the modeling cost and model complexity are reduced,and the modeling accuracy of the hysteresis loop is improved.展开更多
A novel algorithm for constructing support vector machine regression ensemble is proposed. As to regression prediction, support vector machine regression （SVMR） ensemble is proposed by resampling from given training...A novel algorithm for constructing support vector machine regression ensemble is proposed. As to regression prediction, support vector machine regression （SVMR） ensemble is proposed by resampling from given training data sets repeatedly and aggregating several independent SVMRs, each of which is trained to use a replicated training set. After training, several independently trained SVMRs need to be aggregated in an appropriate combination manner. Generally, the linear weighting is usually used like expert weighting score in Boosting Regression and it is without optimization capacity. Three combination techniques are proposed, including simple arithmetic mean, linear least square error weighting and nonlinear hierarchical combining that uses another upper-layer SVMR to combine several lower-layer SVMRs. Finally, simulation experiments demonstrate the accuracy and validity of the presented algorithm.展开更多
The construction method of background value is improved in the original multi-variable grey model （MGM（1,m）） from its source of construction errors. The MGM（1,m） with optimized background value is used to elimin...The construction method of background value is improved in the original multi-variable grey model （MGM（1,m）） from its source of construction errors. The MGM（1,m） with optimized background value is used to eliminate the random fluctuations or errors of the observational data of all variables, and the combined prediction model together with the multiple linear regression is established in order to improve the simulation and prediction accuracy of the combined model. Finally, a combined model of the MGM（1,2） with optimized background value and the binary linear regression is constructed by an example. The results show that the model has good effects for simulation and prediction.展开更多
Removal of cloud cover on the satellite remote sensing image can effectively improve the availability of remote sensing images. For thin cloud cover, support vector value contourlet transform is used to achieve multi-...Removal of cloud cover on the satellite remote sensing image can effectively improve the availability of remote sensing images. For thin cloud cover, support vector value contourlet transform is used to achieve multi-scale decomposition of the area of thin cloud cover on remote sensing images. Through enhancing coefficients of high frequency and suppressing coefficients of low frequency, the thin cloud is removed. For thick cloud cover, if the areas of thick cloud cover on multi-source or multi-temporal remote sensing images do not overlap, the multi-output support vector regression learning method is used to remove this kind of thick clouds. If the thick cloud cover areas overlap, by using the multi-output learning of the surrounding areas to predict the surface features of the overlapped thick cloud cover areas, this kind of thick cloud is removed. Experimental results show that the proposed cloud removal method can effectively solve the problems of the cloud overlapping and radiation difference among multi-source images. The cloud removal image is clear and smooth.展开更多
In order to optimize plastic viscosity of 18 mPa·s circulating micro-bubble drilling fluid formula,orthogonal and uniform experimental design methods were applied,and the plastic viscosities of 36 and 24 groups o...In order to optimize plastic viscosity of 18 mPa·s circulating micro-bubble drilling fluid formula,orthogonal and uniform experimental design methods were applied,and the plastic viscosities of 36 and 24 groups of agent were tested,respectively.It is found that these two experimental design methods show drawbacks,that is,the amount of agent is difficult to determine,and the results are not fully optimized.Therefore,multiple regression experimental method was used to design experimental formula.By randomly selecting arbitrary agent with the amount within the recommended range,17 groups of drilling fluid formula were designed,and the plastic viscosity of each experiment formula was measured.Set plastic viscosity as the objective function,through multiple regressions,then quadratic regression model is obtained,whose correlation coefficient meets the requirement.Set target values of plastic viscosity to be 18,20 and 22 mPa·s,respectively,with the trial method,5 drilling fluid formulas are obtained with accuracy of 0.000 3,0.000 1 and 0.000 3.Arbitrarily select target value of each of the two groups under the formula for experimental verification of drilling fluid,then the measurement errors between theoretical and tested plastic viscosity are less than 5%,confirming that regression model can be applied to optimizing the circulating of plastic-foam drilling fluid viscosity.In accordance with the precision of different formulations of drilling fluid for other constraints,the methods result in the optimization of the circulating micro-bubble drilling fluid parameters.展开更多
For imbalanced datasets, the focus of classification is to identify samples of the minority class. The performance of current data mining algorithms is not good enough for processing imbalanced datasets. The synthetic...For imbalanced datasets, the focus of classification is to identify samples of the minority class. The performance of current data mining algorithms is not good enough for processing imbalanced datasets. The synthetic minority over-sampling technique(SMOTE) is specifically designed for learning from imbalanced datasets, generating synthetic minority class examples by interpolating between minority class examples nearby. However, the SMOTE encounters the overgeneralization problem. The densitybased spatial clustering of applications with noise(DBSCAN) is not rigorous when dealing with the samples near the borderline.We optimize the DBSCAN algorithm for this problem to make clustering more reasonable. This paper integrates the optimized DBSCAN and SMOTE, and proposes a density-based synthetic minority over-sampling technique(DSMOTE). First, the optimized DBSCAN is used to divide the samples of the minority class into three groups, including core samples, borderline samples and noise samples, and then the noise samples of minority class is removed to synthesize more effective samples. In order to make full use of the information of core samples and borderline samples,different strategies are used to over-sample core samples and borderline samples. Experiments show that DSMOTE can achieve better results compared with SMOTE and Borderline-SMOTE in terms of precision, recall and F-value.展开更多
The thermal induced errors can account for as much as 70% of the dimensional errors on a workpiece. Accurate modeling of errors is an essential part of error compensation. Base on analyzing the existing approaches of ...The thermal induced errors can account for as much as 70% of the dimensional errors on a workpiece. Accurate modeling of errors is an essential part of error compensation. Base on analyzing the existing approaches of the thermal error modeling for machine tools, a new approach of regression orthogonal design is proposed, which combines the statistic theory with machine structures, surrounding condition, engineering judgements, and experience in modeling. A whole computation and analysis procedure is given. Therefore, the model got from this method are more robust and practical than those got from the present method that depends on the modeling data completely. At last more than 100 applications of CNC turning center with only one thermal error model are given. The cutting diameter variation reduces from more than 35 μm to about 12 μm with the orthogonal regression modeling and compensation of thermal error.展开更多
A multivariable regression(MVR) approach is proposed to identify the real power transfer between generators and loads.Based on solved load flow results,it first uses modified nodal equation method(MNE) to determine re...A multivariable regression(MVR) approach is proposed to identify the real power transfer between generators and loads.Based on solved load flow results,it first uses modified nodal equation method(MNE) to determine real power contribution from each generator to loads.Then,the results of MNE method and load flow information are utilized to determine suitable regression coefficients using MVR model to estimate the power transfer.The 25-bus equivalent system of south Malaysia is utilized as a test system to illustrate the effectiveness of the MVR output compared to that of the MNE method.The error of the estimate of MVR method ranges from 0.001 4 to 0.007 9.Furthermore,when compared to MNE method,MVR method computes generator contribution to loads within 26.40 ms whereas the MNE method takes 360 ms for the calculation of same real power transfer allocation.Therefore,MVR method is more suitable for real time power transfer allocation.展开更多
As the solutions of the least squares support vector regression machine （LS-SVRM） are not sparse, it leads to slow prediction speed and limits its applications. The defects of the ex- isting adaptive pruning algorit...As the solutions of the least squares support vector regression machine （LS-SVRM） are not sparse, it leads to slow prediction speed and limits its applications. The defects of the ex- isting adaptive pruning algorithm for LS-SVRM are that the training speed is slow, and the generalization performance is not satis- factory, especially for large scale problems. Hence an improved algorithm is proposed. In order to accelerate the training speed, the pruned data point and fast leave-one-out error are employed to validate the temporary model obtained after decremental learning. The novel objective function in the termination condition which in- volves the whole constraints generated by all training data points and three pruning strategies are employed to improve the generali- zation performance. The effectiveness of the proposed algorithm is tested on six benchmark datasets. The sparse LS-SVRM model has a faster training speed and better generalization performance.展开更多
A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the mai...A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the main influence on day-ahead price, avoiding the strong correlation between the input factors that might influence electricity price, such as the load of the forecasting hour, other history loads and prices, weather and temperature; then GRNN was employed to forecast electricity price according to the main information extracted by PCA. To prove the efficiency of the combined model, a case from PJM (Pennsylvania-New Jersey-Maryland) day-ahead electricity market was evaluated. Compared to back-propagation (BP) neural network and standard GRNN, the combined method reduces the mean absolute percentage error about 3%.展开更多
In this article, authors introduce a method to assess local influence of obser- vations on the parameter estimates and prediction in multivariate regression model. The diagnostics under the perturbations of error vari...In this article, authors introduce a method to assess local influence of obser- vations on the parameter estimates and prediction in multivariate regression model. The diagnostics under the perturbations of error variance, response variables and explanatory variables are derived, and the results are compared with those of case- deletion. Two examples are analyzed for illustration.展开更多
High-precision filtering estimation is one of the key techniques for strapdown inertial navigation system/global navigation satellite system(SINS/GNSS)integrated navigation system,and its estimation plays an important...High-precision filtering estimation is one of the key techniques for strapdown inertial navigation system/global navigation satellite system(SINS/GNSS)integrated navigation system,and its estimation plays an important role in the performance evaluation of the navigation system.Traditional filter estimation methods usually assume that the measurement noise conforms to the Gaussian distribution,without considering the influence of the pollution introduced by the GNSS signal,which is susceptible to external interference.To address this problem,a high-precision filter estimation method using Gaussian process regression(GPR)is proposed to enhance the prediction and estimation capability of the unscented quaternion estimator(USQUE)to improve the navigation accuracy.Based on the advantage of the GPR machine learning function,the estimation performance of the sliding window for model training is measured.This method estimates the output of the observation information source through the measurement window and realizes the robust measurement update of the filter.The combination of GPR and the USQUE algorithm establishes a robust mechanism framework,which enhances the robustness and stability of traditional methods.The results of the trajectory simulation experiment and SINS/GNSS car-mounted tests indicate that the strategy has strong robustness and high estimation accuracy,which demonstrates the effectiveness of the proposed method.展开更多
The computational cost of support vector regression in the training phase is O （N^3）, which is very expensive for a large scale problem. In addition, the solution of support vector regression is of parsimoniousness,...The computational cost of support vector regression in the training phase is O （N^3）, which is very expensive for a large scale problem. In addition, the solution of support vector regression is of parsimoniousness, which has relation to a part of the whole training data set. Hence, it is reasonable to reduce the training data set. Aiming at the scheme based on k-nearest neighbors to reduce the training data set with the computational complexity O （kMN^2）, an improved scheme is proposed to accelerate the reducing phase, which cuts down the computational complexity from O （kMN^2） to O （MN^2）. Finally, experimental results on benchmark data sets validate the effectiveness of the improved scheme.展开更多
To overcome the disadvantage that the standard least squares support vector regression(LS-SVR) algorithm is not suitable to multiple-input multiple-output(MIMO) system modelling directly,an improved LS-SVR algorithm w...To overcome the disadvantage that the standard least squares support vector regression(LS-SVR) algorithm is not suitable to multiple-input multiple-output(MIMO) system modelling directly,an improved LS-SVR algorithm which was defined as multi-output least squares support vector regression(MLSSVR) was put forward by adding samples' absolute errors in objective function and applied to flatness intelligent control.To solve the poor-precision problem of the control scheme based on effective matrix in flatness control,the predictive control was introduced into the control system and the effective matrix-predictive flatness control method was proposed by combining the merits of the two methods.Simulation experiment was conducted on 900HC reversible cold roll.The performance of effective matrix method and the effective matrix-predictive control method were compared,and the results demonstrate the validity of the effective matrix-predictive control method.展开更多
Modern agricultural mechanization has put forward higher requirements for the intelligent defect diagnosis.However,the fault features are usually learned and classified under all speeds without considering the effects...Modern agricultural mechanization has put forward higher requirements for the intelligent defect diagnosis.However,the fault features are usually learned and classified under all speeds without considering the effects of speed fluctuation.To overcome this deficiency,a novel intelligent defect detection framework based on time-frequency transformation is presented in this work.In the framework,the samples under one speed are employed for training sparse filtering model,and the remaining samples under different speeds are adopted for testing the effectiveness.Our proposed approach contains two stages:1)the time-frequency domain signals are acquired from the mechanical raw vibration data by the short time Fourier transform algorithm,and then the defect features are extracted from time-frequency domain signals by sparse filtering algorithm;2)different defect types are classified by the softmax regression using the defect features.The proposed approach can be employed to mine available fault characteristics adaptively and is an effective intelligent method for fault detection of agricultural equipment.The fault detection performances confirm that our approach not only owns strong ability for fault classification under different speeds,but also obtains higher identification accuracy than the other methods.展开更多
Because of the relativity among the parameters, partial least square regression(PLSR)was applied to build the model and get the regression equation. The improved algorithm simplified the calculating process greatly be...Because of the relativity among the parameters, partial least square regression(PLSR)was applied to build the model and get the regression equation. The improved algorithm simplified the calculating process greatly because of the reduction of calculation. The orthogonal design was adopted in this experiment. Every sample had strong representation, which could reduce the experimental time and obtain the overall test data. Combined with the formation problem of gas metal arc weld with big current, the auxiliary analysis technique of PLSR was discussed and the regression equation of form factors (i.e. surface width, weld penetration and weld reinforcement) to process parameters(i.e. wire feed rate, wire extension, welding speed, gas flow, welding voltage and welding current)was given. The correlativity structure among variables was analyzed and there was certain correlation between independent variables matrix X and dependent variables matrix Y. The regression analysis shows that the welding speed mainly influences the weld formation while the variation of gas flow in certain range has little influence on formation of weld. The fitting plot of regression accuracy is given. The fitting quality of regression equation is basically satisfactory.展开更多
Gaussian process(GP)has fewer parameters,simple model and output of probabilistic sense,when compared with the methods such as support vector machines.Selection of the hyper-parameters is critical to the performance o...Gaussian process(GP)has fewer parameters,simple model and output of probabilistic sense,when compared with the methods such as support vector machines.Selection of the hyper-parameters is critical to the performance of Gaussian process model.However,the common-used algorithm has the disadvantages of difficult determination of iteration steps,over-dependence of optimization effect on initial values,and easily falling into local optimum.To solve this problem,a method combining the Gaussian process with memetic algorithm was proposed.Based on this method,memetic algorithm was used to search the optimal hyper parameters of Gaussian process regression(GPR)model in the training process and form MA-GPR algorithms,and then the model was used to predict and test the results.When used in the marine long-range precision strike system(LPSS)battle effectiveness evaluation,the proposed MA-GPR model significantly improved the prediction accuracy,compared with the conjugate gradient method and the genetic algorithm optimization process.展开更多
基金supported by the National Natural Science Foundation of China(62375013).
文摘As the core component of inertial navigation systems, fiber optic gyroscope (FOG), with technical advantages such as low power consumption, long lifespan, fast startup speed, and flexible structural design, are widely used in aerospace, unmanned driving, and other fields. However, due to the temper-ature sensitivity of optical devices, the influence of environmen-tal temperature causes errors in FOG, thereby greatly limiting their output accuracy. This work researches on machine-learn-ing based temperature error compensation techniques for FOG. Specifically, it focuses on compensating for the bias errors gen-erated in the fiber ring due to the Shupe effect. This work pro-poses a composite model based on k-means clustering, sup-port vector regression, and particle swarm optimization algo-rithms. And it significantly reduced redundancy within the sam-ples by adopting the interval sequence sample. Moreover, met-rics such as root mean square error (RMSE), mean absolute error (MAE), bias stability, and Allan variance, are selected to evaluate the model’s performance and compensation effective-ness. This work effectively enhances the consistency between data and models across different temperature ranges and tem-perature gradients, improving the bias stability of the FOG from 0.022 °/h to 0.006 °/h. Compared to the existing methods utiliz-ing a single machine learning model, the proposed method increases the bias stability of the compensated FOG from 57.11% to 71.98%, and enhances the suppression of rate ramp noise coefficient from 2.29% to 14.83%. This work improves the accuracy of FOG after compensation, providing theoretical guid-ance and technical references for sensors error compensation work in other fields.
基金National Natural Science Foundation of China(62203118)。
文摘Piezo actuators are widely used in ultra-precision fields because of their high response and nano-scale step length.However,their hysteresis characteristics seriously affect the accuracy and stability of piezo actuators.Existing methods for fitting hysteresis loops include operator class,differential equation class,and machine learning class.The modeling cost of operator class and differential equation class methods is high,the model complexity is high,and the process of machine learning,such as neural network calculation,is opaque.The physical model framework cannot be directly extracted.Therefore,the sparse identification of nonlinear dynamics(SINDy)algorithm is proposed to fit hysteresis loops.Furthermore,the SINDy algorithm is improved.While the SINDy algorithm builds an orthogonal candidate database for modeling,the sparse regression model is simplified,and the Relay operator is introduced for piecewise fitting to solve the distortion problem of the SINDy algorithm fitting singularities.The Relay-SINDy algorithm proposed in this paper is applied to fitting hysteresis loops.Good performance is obtained with the experimental results of open and closed loops.Compared with the existing methods,the modeling cost and model complexity are reduced,and the modeling accuracy of the hysteresis loop is improved.
基金The project was supported by the National Science Foundation of China (70572045)
文摘A novel algorithm for constructing support vector machine regression ensemble is proposed. As to regression prediction, support vector machine regression （SVMR） ensemble is proposed by resampling from given training data sets repeatedly and aggregating several independent SVMRs, each of which is trained to use a replicated training set. After training, several independently trained SVMRs need to be aggregated in an appropriate combination manner. Generally, the linear weighting is usually used like expert weighting score in Boosting Regression and it is without optimization capacity. Three combination techniques are proposed, including simple arithmetic mean, linear least square error weighting and nonlinear hierarchical combining that uses another upper-layer SVMR to combine several lower-layer SVMRs. Finally, simulation experiments demonstrate the accuracy and validity of the presented algorithm.
基金supported by the National Natural Science Foundation of China(71071077)the Ministry of Education Key Project of National Educational Science Planning(DFA090215)+1 种基金China Postdoctoral Science Foundation(20100481137)Funding of Jiangsu Innovation Program for Graduate Education(CXZZ11-0226)
文摘The construction method of background value is improved in the original multi-variable grey model （MGM（1,m）） from its source of construction errors. The MGM（1,m） with optimized background value is used to eliminate the random fluctuations or errors of the observational data of all variables, and the combined prediction model together with the multiple linear regression is established in order to improve the simulation and prediction accuracy of the combined model. Finally, a combined model of the MGM（1,2） with optimized background value and the binary linear regression is constructed by an example. The results show that the model has good effects for simulation and prediction.
基金supported by the National Natural Science Foundation of China(61172127)the Natural Science Foundation of Anhui Province(1408085MF121)
文摘Removal of cloud cover on the satellite remote sensing image can effectively improve the availability of remote sensing images. For thin cloud cover, support vector value contourlet transform is used to achieve multi-scale decomposition of the area of thin cloud cover on remote sensing images. Through enhancing coefficients of high frequency and suppressing coefficients of low frequency, the thin cloud is removed. For thick cloud cover, if the areas of thick cloud cover on multi-source or multi-temporal remote sensing images do not overlap, the multi-output support vector regression learning method is used to remove this kind of thick clouds. If the thick cloud cover areas overlap, by using the multi-output learning of the surrounding areas to predict the surface features of the overlapped thick cloud cover areas, this kind of thick cloud is removed. Experimental results show that the proposed cloud removal method can effectively solve the problems of the cloud overlapping and radiation difference among multi-source images. The cloud removal image is clear and smooth.
基金Project(50304010) supported by the National Natural Science Foundation of China
文摘In order to optimize plastic viscosity of 18 mPa·s circulating micro-bubble drilling fluid formula,orthogonal and uniform experimental design methods were applied,and the plastic viscosities of 36 and 24 groups of agent were tested,respectively.It is found that these two experimental design methods show drawbacks,that is,the amount of agent is difficult to determine,and the results are not fully optimized.Therefore,multiple regression experimental method was used to design experimental formula.By randomly selecting arbitrary agent with the amount within the recommended range,17 groups of drilling fluid formula were designed,and the plastic viscosity of each experiment formula was measured.Set plastic viscosity as the objective function,through multiple regressions,then quadratic regression model is obtained,whose correlation coefficient meets the requirement.Set target values of plastic viscosity to be 18,20 and 22 mPa·s,respectively,with the trial method,5 drilling fluid formulas are obtained with accuracy of 0.000 3,0.000 1 and 0.000 3.Arbitrarily select target value of each of the two groups under the formula for experimental verification of drilling fluid,then the measurement errors between theoretical and tested plastic viscosity are less than 5%,confirming that regression model can be applied to optimizing the circulating of plastic-foam drilling fluid viscosity.In accordance with the precision of different formulations of drilling fluid for other constraints,the methods result in the optimization of the circulating micro-bubble drilling fluid parameters.
基金supported by the National Key Research and Development Program of China(2018YFB1003700)the Scientific and Technological Support Project(Society)of Jiangsu Province(BE2016776)+2 种基金the“333” project of Jiangsu Province(BRA2017228 BRA2017401)the Talent Project in Six Fields of Jiangsu Province(2015-JNHB-012)
文摘For imbalanced datasets, the focus of classification is to identify samples of the minority class. The performance of current data mining algorithms is not good enough for processing imbalanced datasets. The synthetic minority over-sampling technique(SMOTE) is specifically designed for learning from imbalanced datasets, generating synthetic minority class examples by interpolating between minority class examples nearby. However, the SMOTE encounters the overgeneralization problem. The densitybased spatial clustering of applications with noise(DBSCAN) is not rigorous when dealing with the samples near the borderline.We optimize the DBSCAN algorithm for this problem to make clustering more reasonable. This paper integrates the optimized DBSCAN and SMOTE, and proposes a density-based synthetic minority over-sampling technique(DSMOTE). First, the optimized DBSCAN is used to divide the samples of the minority class into three groups, including core samples, borderline samples and noise samples, and then the noise samples of minority class is removed to synthesize more effective samples. In order to make full use of the information of core samples and borderline samples,different strategies are used to over-sample core samples and borderline samples. Experiments show that DSMOTE can achieve better results compared with SMOTE and Borderline-SMOTE in terms of precision, recall and F-value.
文摘The thermal induced errors can account for as much as 70% of the dimensional errors on a workpiece. Accurate modeling of errors is an essential part of error compensation. Base on analyzing the existing approaches of the thermal error modeling for machine tools, a new approach of regression orthogonal design is proposed, which combines the statistic theory with machine structures, surrounding condition, engineering judgements, and experience in modeling. A whole computation and analysis procedure is given. Therefore, the model got from this method are more robust and practical than those got from the present method that depends on the modeling data completely. At last more than 100 applications of CNC turning center with only one thermal error model are given. The cutting diameter variation reduces from more than 35 μm to about 12 μm with the orthogonal regression modeling and compensation of thermal error.
文摘A multivariable regression(MVR) approach is proposed to identify the real power transfer between generators and loads.Based on solved load flow results,it first uses modified nodal equation method(MNE) to determine real power contribution from each generator to loads.Then,the results of MNE method and load flow information are utilized to determine suitable regression coefficients using MVR model to estimate the power transfer.The 25-bus equivalent system of south Malaysia is utilized as a test system to illustrate the effectiveness of the MVR output compared to that of the MNE method.The error of the estimate of MVR method ranges from 0.001 4 to 0.007 9.Furthermore,when compared to MNE method,MVR method computes generator contribution to loads within 26.40 ms whereas the MNE method takes 360 ms for the calculation of same real power transfer allocation.Therefore,MVR method is more suitable for real time power transfer allocation.
基金supported by the National Natural Science Foundation of China (61074127)
文摘As the solutions of the least squares support vector regression machine （LS-SVRM） are not sparse, it leads to slow prediction speed and limits its applications. The defects of the ex- isting adaptive pruning algorithm for LS-SVRM are that the training speed is slow, and the generalization performance is not satis- factory, especially for large scale problems. Hence an improved algorithm is proposed. In order to accelerate the training speed, the pruned data point and fast leave-one-out error are employed to validate the temporary model obtained after decremental learning. The novel objective function in the termination condition which in- volves the whole constraints generated by all training data points and three pruning strategies are employed to improve the generali- zation performance. The effectiveness of the proposed algorithm is tested on six benchmark datasets. The sparse LS-SVRM model has a faster training speed and better generalization performance.
基金Project(70671039) supported by the National Natural Science Foundation of China
文摘A combined model based on principal components analysis (PCA) and generalized regression neural network (GRNN) was adopted to forecast electricity price in day-ahead electricity market. PCA was applied to mine the main influence on day-ahead price, avoiding the strong correlation between the input factors that might influence electricity price, such as the load of the forecasting hour, other history loads and prices, weather and temperature; then GRNN was employed to forecast electricity price according to the main information extracted by PCA. To prove the efficiency of the combined model, a case from PJM (Pennsylvania-New Jersey-Maryland) day-ahead electricity market was evaluated. Compared to back-propagation (BP) neural network and standard GRNN, the combined method reduces the mean absolute percentage error about 3%.
文摘In this article, authors introduce a method to assess local influence of obser- vations on the parameter estimates and prediction in multivariate regression model. The diagnostics under the perturbations of error variance, response variables and explanatory variables are derived, and the results are compared with those of case- deletion. Two examples are analyzed for illustration.
基金supported by the National Natural Science Foundation of China(61873275,61703419,425317829).
文摘High-precision filtering estimation is one of the key techniques for strapdown inertial navigation system/global navigation satellite system(SINS/GNSS)integrated navigation system,and its estimation plays an important role in the performance evaluation of the navigation system.Traditional filter estimation methods usually assume that the measurement noise conforms to the Gaussian distribution,without considering the influence of the pollution introduced by the GNSS signal,which is susceptible to external interference.To address this problem,a high-precision filter estimation method using Gaussian process regression(GPR)is proposed to enhance the prediction and estimation capability of the unscented quaternion estimator(USQUE)to improve the navigation accuracy.Based on the advantage of the GPR machine learning function,the estimation performance of the sliding window for model training is measured.This method estimates the output of the observation information source through the measurement window and realizes the robust measurement update of the filter.The combination of GPR and the USQUE algorithm establishes a robust mechanism framework,which enhances the robustness and stability of traditional methods.The results of the trajectory simulation experiment and SINS/GNSS car-mounted tests indicate that the strategy has strong robustness and high estimation accuracy,which demonstrates the effectiveness of the proposed method.
基金supported by the National Natural Science Foundation of China(50576033).
文摘The computational cost of support vector regression in the training phase is O （N^3）, which is very expensive for a large scale problem. In addition, the solution of support vector regression is of parsimoniousness, which has relation to a part of the whole training data set. Hence, it is reasonable to reduce the training data set. Aiming at the scheme based on k-nearest neighbors to reduce the training data set with the computational complexity O （kMN^2）, an improved scheme is proposed to accelerate the reducing phase, which cuts down the computational complexity from O （kMN^2） to O （MN^2）. Finally, experimental results on benchmark data sets validate the effectiveness of the improved scheme.
基金Project(50675186) supported by the National Natural Science Foundation of China
文摘To overcome the disadvantage that the standard least squares support vector regression(LS-SVR) algorithm is not suitable to multiple-input multiple-output(MIMO) system modelling directly,an improved LS-SVR algorithm which was defined as multi-output least squares support vector regression(MLSSVR) was put forward by adding samples' absolute errors in objective function and applied to flatness intelligent control.To solve the poor-precision problem of the control scheme based on effective matrix in flatness control,the predictive control was introduced into the control system and the effective matrix-predictive flatness control method was proposed by combining the merits of the two methods.Simulation experiment was conducted on 900HC reversible cold roll.The performance of effective matrix method and the effective matrix-predictive control method were compared,and the results demonstrate the validity of the effective matrix-predictive control method.
基金Project(51675262)supported by the National Natural Science Foundation of ChinaProject(2016YFD0700800)supported by the National Key Research and Development Program of China+2 种基金Project(6140210020102)supported by the Advance Research Field Fund Project of ChinaProject(NP2018304)supported by the Fundamental Research Funds for the Central Universities,ChinaProject(2017-IV-0008-0045)supported by the National Science and Technology Major Project
文摘Modern agricultural mechanization has put forward higher requirements for the intelligent defect diagnosis.However,the fault features are usually learned and classified under all speeds without considering the effects of speed fluctuation.To overcome this deficiency,a novel intelligent defect detection framework based on time-frequency transformation is presented in this work.In the framework,the samples under one speed are employed for training sparse filtering model,and the remaining samples under different speeds are adopted for testing the effectiveness.Our proposed approach contains two stages:1)the time-frequency domain signals are acquired from the mechanical raw vibration data by the short time Fourier transform algorithm,and then the defect features are extracted from time-frequency domain signals by sparse filtering algorithm;2)different defect types are classified by the softmax regression using the defect features.The proposed approach can be employed to mine available fault characteristics adaptively and is an effective intelligent method for fault detection of agricultural equipment.The fault detection performances confirm that our approach not only owns strong ability for fault classification under different speeds,but also obtains higher identification accuracy than the other methods.
文摘Because of the relativity among the parameters, partial least square regression(PLSR)was applied to build the model and get the regression equation. The improved algorithm simplified the calculating process greatly because of the reduction of calculation. The orthogonal design was adopted in this experiment. Every sample had strong representation, which could reduce the experimental time and obtain the overall test data. Combined with the formation problem of gas metal arc weld with big current, the auxiliary analysis technique of PLSR was discussed and the regression equation of form factors (i.e. surface width, weld penetration and weld reinforcement) to process parameters(i.e. wire feed rate, wire extension, welding speed, gas flow, welding voltage and welding current)was given. The correlativity structure among variables was analyzed and there was certain correlation between independent variables matrix X and dependent variables matrix Y. The regression analysis shows that the welding speed mainly influences the weld formation while the variation of gas flow in certain range has little influence on formation of weld. The fitting plot of regression accuracy is given. The fitting quality of regression equation is basically satisfactory.
基金Project(513300303)supported by the General Armament Department,China
文摘Gaussian process(GP)has fewer parameters,simple model and output of probabilistic sense,when compared with the methods such as support vector machines.Selection of the hyper-parameters is critical to the performance of Gaussian process model.However,the common-used algorithm has the disadvantages of difficult determination of iteration steps,over-dependence of optimization effect on initial values,and easily falling into local optimum.To solve this problem,a method combining the Gaussian process with memetic algorithm was proposed.Based on this method,memetic algorithm was used to search the optimal hyper parameters of Gaussian process regression(GPR)model in the training process and form MA-GPR algorithms,and then the model was used to predict and test the results.When used in the marine long-range precision strike system(LPSS)battle effectiveness evaluation,the proposed MA-GPR model significantly improved the prediction accuracy,compared with the conjugate gradient method and the genetic algorithm optimization process.
