Multi-kernel-based support vector machine (SVM) model structure of nonlinear systems and its specific identification method is proposed, which is composed of a SVM with linear kernel function followed in series by a...Multi-kernel-based support vector machine (SVM) model structure of nonlinear systems and its specific identification method is proposed, which is composed of a SVM with linear kernel function followed in series by a SVM with spline kernel function. With the help of this model, nonlinear model predictive control can be transformed to linear model predictive control, and consequently a unified analytical solution of optimal input of multi-step-ahead predictive control is possible to derive. This algorithm does not require online iterative optimization in order to be suitable for real-time control with less calculation. The simulation results of pH neutralization process and CSTR reactor show the effectiveness and advantages of the presented algorithm.展开更多
To decrease breakdown time and improve machine operation reliability,accurate residual useful life(RUL) prediction has been playing a critical role in condition based monitoring.A data fusion method was proposed to ac...To decrease breakdown time and improve machine operation reliability,accurate residual useful life(RUL) prediction has been playing a critical role in condition based monitoring.A data fusion method was proposed to achieve online RUL prediction of slewing bearings,which consisted of a reliability based RUL prediction model and a data driven failure rate(FR) estimation model.Firstly,an RUL prediction model was developed based on modified Weibull distribution to build the relationship between RUL and FR.Secondly,principal component analysis(PCA) was introduced to process multi-dimensional life-cycle vibration signals,and continuous squared prediction error(CSPE) and its time-domain features were employed as equipment performance degradation features.Afterwards,an FR estimation model was established on basis of the degradation features and relevant FRs using simplified fuzzy adaptive resonance theory map(SFAM) neural network.Consequently,real-time FR of equipment can be obtained through FR estimation model,and then accurate RUL can be calculated through the RUL prediction model.Results of a slewing bearing life test show that CSPE is an effective indicator of performance degradation process of slewing bearings,and that by combining actual load condition and real-time monitored data,the calculation time is reduced by 87.3%and the accuracy is increased by 0.11%,which provides a potential for online RUL prediction of slewing bearings and other various machineries.展开更多
There are often system. The high measure many inter-harmonics in power t accuracy of inter-harmonics order, amplitude and initial phase is needed. A new approach is presented for inter-harmonic modeling and parameter ...There are often system. The high measure many inter-harmonics in power t accuracy of inter-harmonics order, amplitude and initial phase is needed. A new approach is presented for inter-harmonic modeling and parameter estimation based on linear support vector machine (SVM). Firstly, parameter estimation of linear model is realized based on standard linear SVM. Then, interharmonic model is transformed to a linear model according to trigonometric functions. The approach obtains order of inter-harmonic model with windowed Blackman-Tukey (BT) spectrum analysis, and gets number and frequency of harmonics. Finally, the linear SVM is applied to estimate the inter-harmonic parameters, amplitude and initial phase. The simulation results show that the proposed approach has high precision and good antinoise. The accuracy of three parameters are all higher than 98%.展开更多
The cable-strut structural system is statically and kinematically indeterminate. The initial pre-stress is a key factor for determining the shape and load carrying capacity. A new numerical algorithm is presented here...The cable-strut structural system is statically and kinematically indeterminate. The initial pre-stress is a key factor for determining the shape and load carrying capacity. A new numerical algorithm is presented herein for the initial pre-stress finding procedure of complete cable-strut assembly. This method is based on the linear adjustment theory and does not take into account the material behavior. By using this method,the initial pre-stress of the multi self-stress modes can be found easily and the cal-culation process is simplified and efficient also. Finally,the initial pre-stress and structural performances of a particular Levy cable dome are analyzed comprehensively. The algorithm has proven to be efficient and correct,and the numerical results are valuable for practical design of Levy cable dome.展开更多
Combining mathematical morphology (MM),nonparametric and nonlinear model,a novel approach for predicting slope displacement was developed to improve the prediction accuracy.A parallel-composed morphological filter wit...Combining mathematical morphology (MM),nonparametric and nonlinear model,a novel approach for predicting slope displacement was developed to improve the prediction accuracy.A parallel-composed morphological filter with multiple structure elements was designed to process measured displacement time series with adaptive multi-scale decoupling.Whereafter,functional-coefficient auto regressive (FAR) models were established for the random subsequences.Meanwhile,the trend subsequence was processed by least squares support vector machine (LSSVM) algorithm.Finally,extrapolation results obtained were superposed to get the ultimate prediction result.Case study and comparative analysis demonstrate that the presented method can optimize training samples and show a good nonlinear predicting performance with low risk of choosing wrong algorithms.Mean absolute percentage error (MAPE) and root mean square error (RMSE) of the MM-FAR&LSSVM predicting results are as low as 1.670% and 0.172 mm,respectively,which means that the prediction accuracy are improved significantly.展开更多
The performance of data-driven models relies heavily on the amount and quality of training samples, so it might deteriorate significantly in the regions where samples are scarce. The objective of this paper is to deve...The performance of data-driven models relies heavily on the amount and quality of training samples, so it might deteriorate significantly in the regions where samples are scarce. The objective of this paper is to develop an online SVR model updating strategy to track the change in the process characteristics efficiently with affordable computational burden. This is achieved by adding a new sample that violates the Karush–Kuhn–Tucker conditions of the existing SVR model and by deleting the old sample that has the maximum distance with respect to the newly added sample in feature space. The benefits offered by such an updating strategy are exploited to develop an adaptive model-based control scheme, where model updating and control task perform alternately.The effectiveness of the adaptive controller is demonstrated by simulation study on a continuous stirred tank reactor. The results reveal that the adaptive MPC scheme outperforms its non-adaptive counterpart for largemagnitude set point changes and variations in process parameters.展开更多
This paper presents a nonlinear model predictive control(NMPC) approach based on support vector machine(SVM) and genetic algorithm(GA) for multiple-input multiple-output(MIMO) nonlinear systems.Individual SVM is used ...This paper presents a nonlinear model predictive control(NMPC) approach based on support vector machine(SVM) and genetic algorithm(GA) for multiple-input multiple-output(MIMO) nonlinear systems.Individual SVM is used to approximate each output of the controlled plant Then the model is used in MPC control scheme to predict the outputs of the controlled plant.The optimal control sequence is calculated using GA with elite preserve strategy.Simulation results of a typical MIMO nonlinear system show that this method has a good ability of set points tracking and disturbance rejection.展开更多
Second lining stability, which is the last protection in tunnel engineering, is critically important. The theological properties of the surrounding rock heavily affect second lining stability. In this work, we used la...Second lining stability, which is the last protection in tunnel engineering, is critically important. The theological properties of the surrounding rock heavily affect second lining stability. In this work, we used laboratory triaxial compressive rheological limestone tests to study nonlinear creep damage characteristics of surrounding rock mass in construction projects. We established a nonlinear creep damage constitutive model for the rock mass, as well as a constitutive model numerical implementation made by programming. Second, we introduced a new foam concrete with higher compression performance and good ductility and studied its mechanical properties through uniaxial and triaxial tests. This concrete was used as the filling material for the reserved deformation layer between the primary support and second lining. Finally, we proposed a high efficiency and accuracy staged optimization method. The minimum reserved deformation layer thickness was established as the optimization goal, and the presence of plastic strain in the second lining after 100 years of surrounding rock creep was used as an evaluation index. Reserved deformation layer thickness optimization analysis reveals no plastic strain in the second lining when the reserved deformation minimum thickness layer is 28.50 cm. The results show that the new foam concrete used as a reserved deformation layer filling material can absorb creep deformation of surrounding rock mass, reduce second lining deformation that leads to plastic strain, and ensure long-term second lining stability.展开更多
Full-scale model tests were carried out on a 30 m span prestressed concrete box girder and a 20 m span prestressed concrete hollow slab. Failure models were prestressed reinforcement tensile failure and crashing of ro...Full-scale model tests were carried out on a 30 m span prestressed concrete box girder and a 20 m span prestressed concrete hollow slab. Failure models were prestressed reinforcement tensile failure and crashing of roof concrete, respectively. The ductility indexes of the box girder and hollow slab were 1.99 and 1.23, respectively, according to the energy viewpoint. Based on the horizontal section hypothesis, the nonlinear computation procedure was established using the limited banding law, and it could carry out the entire performance analysis including the unloading, mainly focusing on the ways to achieve the unloading curves computation through stress-strain, moment-curvature and load-displacement curves. Through the procedure, parameters that influence on the bearing capacity, deformation performance and ductility of the structures were analyzed. Those parameters were quantity of prestressed reinforcement and tension coefficients of prestressed reinforcement. From the analysis, some useful conclusions can be obtained.展开更多
Reliable forecasts of the price of oil are of interest for a wide range of applications. For example, central banks and private sector forecasters view the price of oil as one of the key variables in generating macroe...Reliable forecasts of the price of oil are of interest for a wide range of applications. For example, central banks and private sector forecasters view the price of oil as one of the key variables in generating macroeconomic projections and in assessing macroeconomic risks. Of particular interest is the question of the extent to which the price of oil is helpful in predicting recessions. This paper presents a statistical learning method (SLM) based on combined fuzzy system (FS), artificial neural network (ANN), and support vector regression (SVR) to cope with optimum long-term oil price forecasting in noisy, uncertain, and complex environments. A number of quantitative factors were discovered from this model and used as the input. For verification and testing, the West Texas Intermediate (WT1) crude oil spot price is used to test the effectiveness of the proposed learning methodology. Empirical results reveal that the proposed SLM-based forecasting can model the nonlinear relationship between the input variables and price very well. Furthermore, in-sample and out-of-sample prediction performance also demonstrates that the proposed SLM model can produce more accurate prediction results than other nonlinear models.展开更多
As the running speed of high-speed trains increases, aerodynamic drag becomes the key factor which limits the further increase of the running speed and energy consumption. Aerodynamic lift of the trailing car also bec...As the running speed of high-speed trains increases, aerodynamic drag becomes the key factor which limits the further increase of the running speed and energy consumption. Aerodynamic lift of the trailing car also becomes the key force which affects the amenity and safety of the train. In the present paper, a simplified CRH380A high-speed train with three carriages is chosen as the model in order to optimize aerodynamic drag of the total train and aerodynamic lift of the trailing car. A constrained mul- ti-objective optimization design of the aerodynamic head shape of high-speed trains based on adaptive non-dominated sorting genetic algorithm is also developed combining local function three-dimensional parametric approach and central Latin hypercube sampling method with maximin criteria based on the iterative local search algorithm. The results show that local function parametric approach can be well applied to optimal design of complex three-dimensional aerodynamic shape, and the adaptive non-dominated sorting genetic algorithm can be more accurate and efficient to find the Pareto front. After optimization the aerodynamic drag of the simplified train with three carriages is reduced by 3.2%, and the lift coefficient of the trailing car by 8.24%, the volume of the streamlined head by 2.16%; the aerodynamic drag of the real prototype CRH380A is reduced by 2.26%, lift coefficient of the trailing car by 19.67%. The variation of aerodynamic performance between the simplified train and the true train is mainly concentrated in the deformation region of the nose cone and tail cone. The optimization approach proposed in the present paper is simple yet efficient, and sheds lights on the constrained multi-objective engineering optimization design of aerodynamic shape of high-speed trains.展开更多
System identification is an effective way for modeling ship manoeuvring motion and ship manoeuvrability prediction. Support vector machine is proposed to identify the manoeuvring indices in four different response mod...System identification is an effective way for modeling ship manoeuvring motion and ship manoeuvrability prediction. Support vector machine is proposed to identify the manoeuvring indices in four different response models of ship steering motion, including the first order linear, the first order nonlinear, the second order linear and the second order nonlinear models. Predictions of manoeuvres including trained samples by using the identified parameters are compared with the results of free-running model tests. It is discussed that the different four categories are consistent with each other both analytically and numerically. The generalization of the identified model is verified by predicting different untrained manoeuvres. The simulations and comparisons demonstrate the validity of the proposed method.展开更多
An infection-age structured epidemic model with a nonlinear incidence rate is investigated.We formulate the model as an abstract non-densely defined Cauchy problem and derive the condition which guarantees the existen...An infection-age structured epidemic model with a nonlinear incidence rate is investigated.We formulate the model as an abstract non-densely defined Cauchy problem and derive the condition which guarantees the existence and uniqueness for positive age-dependent equilibrium of the model.By analyzing the associated characteristic transcendental equation and applying the normal form theory presented recently for non-densely defined semilinear equations,we show that the SIR(susceptible-infected-recovered)epidemic model undergoes Zero-Hopf bifurcation at the positive equilibrium which is the main result of this paper.展开更多
基金Supported by the State Key Development Program for Basic Research of China (No.2002CB312200) and the National Natural Science Foundation of China (No.60574019).
文摘Multi-kernel-based support vector machine (SVM) model structure of nonlinear systems and its specific identification method is proposed, which is composed of a SVM with linear kernel function followed in series by a SVM with spline kernel function. With the help of this model, nonlinear model predictive control can be transformed to linear model predictive control, and consequently a unified analytical solution of optimal input of multi-step-ahead predictive control is possible to derive. This algorithm does not require online iterative optimization in order to be suitable for real-time control with less calculation. The simulation results of pH neutralization process and CSTR reactor show the effectiveness and advantages of the presented algorithm.
基金Projects(51375222,51175242)supported by the National Natural Science Foundation of China
文摘To decrease breakdown time and improve machine operation reliability,accurate residual useful life(RUL) prediction has been playing a critical role in condition based monitoring.A data fusion method was proposed to achieve online RUL prediction of slewing bearings,which consisted of a reliability based RUL prediction model and a data driven failure rate(FR) estimation model.Firstly,an RUL prediction model was developed based on modified Weibull distribution to build the relationship between RUL and FR.Secondly,principal component analysis(PCA) was introduced to process multi-dimensional life-cycle vibration signals,and continuous squared prediction error(CSPE) and its time-domain features were employed as equipment performance degradation features.Afterwards,an FR estimation model was established on basis of the degradation features and relevant FRs using simplified fuzzy adaptive resonance theory map(SFAM) neural network.Consequently,real-time FR of equipment can be obtained through FR estimation model,and then accurate RUL can be calculated through the RUL prediction model.Results of a slewing bearing life test show that CSPE is an effective indicator of performance degradation process of slewing bearings,and that by combining actual load condition and real-time monitored data,the calculation time is reduced by 87.3%and the accuracy is increased by 0.11%,which provides a potential for online RUL prediction of slewing bearings and other various machineries.
基金National Natural Science Foundation of China(No.60774011)Natural Science Foundation of zhejiang Province,China(No.Y1090182)
文摘There are often system. The high measure many inter-harmonics in power t accuracy of inter-harmonics order, amplitude and initial phase is needed. A new approach is presented for inter-harmonic modeling and parameter estimation based on linear support vector machine (SVM). Firstly, parameter estimation of linear model is realized based on standard linear SVM. Then, interharmonic model is transformed to a linear model according to trigonometric functions. The approach obtains order of inter-harmonic model with windowed Blackman-Tukey (BT) spectrum analysis, and gets number and frequency of harmonics. Finally, the linear SVM is applied to estimate the inter-harmonic parameters, amplitude and initial phase. The simulation results show that the proposed approach has high precision and good antinoise. The accuracy of three parameters are all higher than 98%.
基金Project (No.863-705-210) supported by the Hi-Tech Research and Development Program (863) of China
文摘The cable-strut structural system is statically and kinematically indeterminate. The initial pre-stress is a key factor for determining the shape and load carrying capacity. A new numerical algorithm is presented herein for the initial pre-stress finding procedure of complete cable-strut assembly. This method is based on the linear adjustment theory and does not take into account the material behavior. By using this method,the initial pre-stress of the multi self-stress modes can be found easily and the cal-culation process is simplified and efficient also. Finally,the initial pre-stress and structural performances of a particular Levy cable dome are analyzed comprehensively. The algorithm has proven to be efficient and correct,and the numerical results are valuable for practical design of Levy cable dome.
基金Project(20090162120084)supported by Research Fund for the Doctoral Program of Higher Education of ChinaProject(08JJ4014)supported by the Natural Science Foundation of Hunan Province,China
文摘Combining mathematical morphology (MM),nonparametric and nonlinear model,a novel approach for predicting slope displacement was developed to improve the prediction accuracy.A parallel-composed morphological filter with multiple structure elements was designed to process measured displacement time series with adaptive multi-scale decoupling.Whereafter,functional-coefficient auto regressive (FAR) models were established for the random subsequences.Meanwhile,the trend subsequence was processed by least squares support vector machine (LSSVM) algorithm.Finally,extrapolation results obtained were superposed to get the ultimate prediction result.Case study and comparative analysis demonstrate that the presented method can optimize training samples and show a good nonlinear predicting performance with low risk of choosing wrong algorithms.Mean absolute percentage error (MAPE) and root mean square error (RMSE) of the MM-FAR&LSSVM predicting results are as low as 1.670% and 0.172 mm,respectively,which means that the prediction accuracy are improved significantly.
基金Supported by the National Basic Research Program of China(2012CB720500)Postdoctoral Science Foundation of China(2013M541964)Fundamental Research Funds for the Central Universities(13CX05021A)
文摘The performance of data-driven models relies heavily on the amount and quality of training samples, so it might deteriorate significantly in the regions where samples are scarce. The objective of this paper is to develop an online SVR model updating strategy to track the change in the process characteristics efficiently with affordable computational burden. This is achieved by adding a new sample that violates the Karush–Kuhn–Tucker conditions of the existing SVR model and by deleting the old sample that has the maximum distance with respect to the newly added sample in feature space. The benefits offered by such an updating strategy are exploited to develop an adaptive model-based control scheme, where model updating and control task perform alternately.The effectiveness of the adaptive controller is demonstrated by simulation study on a continuous stirred tank reactor. The results reveal that the adaptive MPC scheme outperforms its non-adaptive counterpart for largemagnitude set point changes and variations in process parameters.
基金Supported by the National Natural Science Foundation of China(21076179)the National Basic Research Program of China(2012CB720500)
文摘This paper presents a nonlinear model predictive control(NMPC) approach based on support vector machine(SVM) and genetic algorithm(GA) for multiple-input multiple-output(MIMO) nonlinear systems.Individual SVM is used to approximate each output of the controlled plant Then the model is used in MPC control scheme to predict the outputs of the controlled plant.The optimal control sequence is calculated using GA with elite preserve strategy.Simulation results of a typical MIMO nonlinear system show that this method has a good ability of set points tracking and disturbance rejection.
基金Projects(51409154,41372289)supported by the National Natural Science Foundation of ChinaProjects(2015JQJH106,2014TDJH103)supported by Research Fund of Shandong University of Science and Technology,China
文摘Second lining stability, which is the last protection in tunnel engineering, is critically important. The theological properties of the surrounding rock heavily affect second lining stability. In this work, we used laboratory triaxial compressive rheological limestone tests to study nonlinear creep damage characteristics of surrounding rock mass in construction projects. We established a nonlinear creep damage constitutive model for the rock mass, as well as a constitutive model numerical implementation made by programming. Second, we introduced a new foam concrete with higher compression performance and good ductility and studied its mechanical properties through uniaxial and triaxial tests. This concrete was used as the filling material for the reserved deformation layer between the primary support and second lining. Finally, we proposed a high efficiency and accuracy staged optimization method. The minimum reserved deformation layer thickness was established as the optimization goal, and the presence of plastic strain in the second lining after 100 years of surrounding rock creep was used as an evaluation index. Reserved deformation layer thickness optimization analysis reveals no plastic strain in the second lining when the reserved deformation minimum thickness layer is 28.50 cm. The results show that the new foam concrete used as a reserved deformation layer filling material can absorb creep deformation of surrounding rock mass, reduce second lining deformation that leads to plastic strain, and ensure long-term second lining stability.
基金National Natural Science Foundation of China(No.50678063)
文摘Full-scale model tests were carried out on a 30 m span prestressed concrete box girder and a 20 m span prestressed concrete hollow slab. Failure models were prestressed reinforcement tensile failure and crashing of roof concrete, respectively. The ductility indexes of the box girder and hollow slab were 1.99 and 1.23, respectively, according to the energy viewpoint. Based on the horizontal section hypothesis, the nonlinear computation procedure was established using the limited banding law, and it could carry out the entire performance analysis including the unloading, mainly focusing on the ways to achieve the unloading curves computation through stress-strain, moment-curvature and load-displacement curves. Through the procedure, parameters that influence on the bearing capacity, deformation performance and ductility of the structures were analyzed. Those parameters were quantity of prestressed reinforcement and tension coefficients of prestressed reinforcement. From the analysis, some useful conclusions can be obtained.
文摘Reliable forecasts of the price of oil are of interest for a wide range of applications. For example, central banks and private sector forecasters view the price of oil as one of the key variables in generating macroeconomic projections and in assessing macroeconomic risks. Of particular interest is the question of the extent to which the price of oil is helpful in predicting recessions. This paper presents a statistical learning method (SLM) based on combined fuzzy system (FS), artificial neural network (ANN), and support vector regression (SVR) to cope with optimum long-term oil price forecasting in noisy, uncertain, and complex environments. A number of quantitative factors were discovered from this model and used as the input. For verification and testing, the West Texas Intermediate (WT1) crude oil spot price is used to test the effectiveness of the proposed learning methodology. Empirical results reveal that the proposed SLM-based forecasting can model the nonlinear relationship between the input variables and price very well. Furthermore, in-sample and out-of-sample prediction performance also demonstrates that the proposed SLM model can produce more accurate prediction results than other nonlinear models.
基金supported by the Major State Basic Research Development Program of China ("973" Program) (Grant No. 2011CB711100) National Key Technology R&D Program (Grant No. 2009BAQG12A03)
文摘As the running speed of high-speed trains increases, aerodynamic drag becomes the key factor which limits the further increase of the running speed and energy consumption. Aerodynamic lift of the trailing car also becomes the key force which affects the amenity and safety of the train. In the present paper, a simplified CRH380A high-speed train with three carriages is chosen as the model in order to optimize aerodynamic drag of the total train and aerodynamic lift of the trailing car. A constrained mul- ti-objective optimization design of the aerodynamic head shape of high-speed trains based on adaptive non-dominated sorting genetic algorithm is also developed combining local function three-dimensional parametric approach and central Latin hypercube sampling method with maximin criteria based on the iterative local search algorithm. The results show that local function parametric approach can be well applied to optimal design of complex three-dimensional aerodynamic shape, and the adaptive non-dominated sorting genetic algorithm can be more accurate and efficient to find the Pareto front. After optimization the aerodynamic drag of the simplified train with three carriages is reduced by 3.2%, and the lift coefficient of the trailing car by 8.24%, the volume of the streamlined head by 2.16%; the aerodynamic drag of the real prototype CRH380A is reduced by 2.26%, lift coefficient of the trailing car by 19.67%. The variation of aerodynamic performance between the simplified train and the true train is mainly concentrated in the deformation region of the nose cone and tail cone. The optimization approach proposed in the present paper is simple yet efficient, and sheds lights on the constrained multi-objective engineering optimization design of aerodynamic shape of high-speed trains.
基金the Special Research Fund for the Doctoral Program of Higher Education (No. 20050248037)the National Natural Science Foundation of China(No. 50779033)
文摘System identification is an effective way for modeling ship manoeuvring motion and ship manoeuvrability prediction. Support vector machine is proposed to identify the manoeuvring indices in four different response models of ship steering motion, including the first order linear, the first order nonlinear, the second order linear and the second order nonlinear models. Predictions of manoeuvres including trained samples by using the identified parameters are compared with the results of free-running model tests. It is discussed that the different four categories are consistent with each other both analytically and numerically. The generalization of the identified model is verified by predicting different untrained manoeuvres. The simulations and comparisons demonstrate the validity of the proposed method.
基金supported by National Natural Science Foundation of China (Grant Nos. 11471044 and 11371058)the Fundamental Research Funds for the Central Universities
文摘An infection-age structured epidemic model with a nonlinear incidence rate is investigated.We formulate the model as an abstract non-densely defined Cauchy problem and derive the condition which guarantees the existence and uniqueness for positive age-dependent equilibrium of the model.By analyzing the associated characteristic transcendental equation and applying the normal form theory presented recently for non-densely defined semilinear equations,we show that the SIR(susceptible-infected-recovered)epidemic model undergoes Zero-Hopf bifurcation at the positive equilibrium which is the main result of this paper.