Traditional centralized Proportional Integral(PI)controller design methods based on Equivalent Transfer Functions(ETFs)have poor decoupling effect in turboprop engines.In this paper,a centralized PI design method base...Traditional centralized Proportional Integral(PI)controller design methods based on Equivalent Transfer Functions(ETFs)have poor decoupling effect in turboprop engines.In this paper,a centralized PI design method based on dynamic imaginary matrix and equivalent transfer function is proposed.Firstly,a method for solving equivalent transfer functions based on the dynamic imaginary matrix is proposed,which adopts dynamic imaginary matrix to describe the dynamic characteristics of the system,and obtains the equivalent transfer function based on the dynamic imaginary matrix characteristics.Secondly,for the equivalent transfer function,a central-ized PI control gain is designed using the Taylor expansion method.Meanwhile,this paper further proves that the centralized PI design method proposed in this paper has integral stability.Consid-ering the impact of altitude and Mach number on turboprop engines,a linear feedforward control method based on the transfer function matrix is further proposed based on the centralized PI con-troller,and the stability of the entire comprehensive control method is proved.Finally,to ensure the safe and effective operation of the turboprop engine,a temperature and torque limiting protection controller is designed for the turboprop engine.Simulation results show that the centralized PI con-troller design method and linear feedforward control method proposed can effectively improve the control quality of turboprop engine control systems.展开更多
Multivariate time series(MTS)data are vital for various applications,particularly in machine learning tasks.However,challenges such as sensor failures can result in irregular and misaligned data with missing values,th...Multivariate time series(MTS)data are vital for various applications,particularly in machine learning tasks.However,challenges such as sensor failures can result in irregular and misaligned data with missing values,thereby complicating their analysis.While recent advancements use graph neural networks(GNNs)to manage these Irregular Multivariate Time Series(IMTS)data,they generally require a reliable graph structure,either pre-existing or inferred from adequate data to properly capture node correlations.This poses a challenge in applications where IMTS data are often streamed and waiting for future data to estimate a suitable graph structure becomes impractical.To overcome this,we introduce a dynamic GNN model suited for streaming characteristics of IMTS data,incorporating an instance-attention mechanism that dynamically learns and updates graph edge weights for real-time analysis.We also tailor strategies for high-frequency and low-frequency data to enhance prediction accuracy.Empirical results on real-world datasets demonstrate the superiority of our proposed model in both classification and imputation tasks.展开更多
In engineering systems,uncertainties in input parameters can significantly influence the output responses.This paper proposes a model distance-based approach to perform global sensitivity analysis for quantifying the ...In engineering systems,uncertainties in input parameters can significantly influence the output responses.This paper proposes a model distance-based approach to perform global sensitivity analysis for quantifying the influence of input uncertainties on multiple responses in an engineering system.The sensitivity indices are determined by comparing a reference model that incorporates all system uncertainties,with an altered model,where specific uncertainties are constrained.The proposed framework employs probability distance measures such as Hellinger distance,Kullback-Leibler divergence,and I2 norm which are based on joint probability density functions.The study also demonstrates the equivalence between the l2 norm-based approach and Sobol's analysis in multivariate sensitivity context.The proposed methodology effectively manages correlated random variables,accommodates both Gaussian and non-Gaussian distributions,and allows for the grouping of input variables.Ilustrative examples consist of static analysis of a truss system and dynamic analysis of a frame subjected to seismic excitation.The sensitivity indices are estimated using brute-force Monte Carlo simulations.The relative ranking of these sensitivity indices can be utilized to identify the most and least significant variables contributing to the response uncertainty.The numerical results show a consistent ranking of input variables across different probability measures,indicating the robustness of proposed framework.展开更多
The zero coprime system equivalence is one of important research in the theory of multidimensional system equivalence,and is closely related to zero coprime equivalence of multivariate polynomial matrices.We first dis...The zero coprime system equivalence is one of important research in the theory of multidimensional system equivalence,and is closely related to zero coprime equivalence of multivariate polynomial matrices.We first discuss the relation between zero coprime equivalence and unimodular equivalence for polynomial matrices.Then,we investigate the zero coprime equivalence problem for several classes of polynomial matrices,some novel findings and criteria on reducing these matrices to their Smith normal forms are obtained.Finally,an example is provided to illustrate the main results.展开更多
A new algorithm for constructing an inverse of a multivariable linear system is presented. This algorithm makes the constructing an inverse of the higher order matrices into searching for the equivalent normal form o...A new algorithm for constructing an inverse of a multivariable linear system is presented. This algorithm makes the constructing an inverse of the higher order matrices into searching for the equivalent normal form of the lower order matrices. Consequently, the calculation is more simple efficient and programmed than previous methods. Another result of the paper is that the lower reduced inverse system is obtained, by selecting special bases of the observable space of the original systems, it reveals the effect of the observability of the original systems on the order of the inverse systems.展开更多
Through modifying the CPN model, a kind of multivariable fuzzy model is put forward, and the matching fuzzy multistep predictive control algorithm is deduced based on the model. The modified model works in a competiti...Through modifying the CPN model, a kind of multivariable fuzzy model is put forward, and the matching fuzzy multistep predictive control algorithm is deduced based on the model. The modified model works in a competitive output manner which results in its local representation property. While studying on line, only a few parameters need to be regulated. So the model has the merits of fast learning and on line self organizing modeling. The control algorithm is simple, adaptive and useful in multivariable and time delay systems. Applying the algorithm in a paper making system, simulation shows its good effect.展开更多
A decentralized model reference adaptive control (MRAC) scheme is proposed and applied to design a multivariable control system of a dual-spool turbofan engine.Simulation studies show good static and dynamic performan...A decentralized model reference adaptive control (MRAC) scheme is proposed and applied to design a multivariable control system of a dual-spool turbofan engine.Simulation studies show good static and dynamic performance of the system over the fullflight envelope. Simulation results also show the good effectiveness of reducing interactionin the multivariable system with significant coupling. The control system developed has awide frequency band to satisfy the strict engineering requirement and is practical for engineering applications.展开更多
This paper presents a multivariable generalized predictive controller with proportion and integration structure by modifying the quadratic criterion of the usual MGPC. The control performance has been improved greatl...This paper presents a multivariable generalized predictive controller with proportion and integration structure by modifying the quadratic criterion of the usual MGPC. The control performance has been improved greatly. The effectiveness of the controller is demonstrated by the simulation result.展开更多
A constrained decoupling (generalized predictive control) GPC algorithm is proposed for MIMO (malti-input multi-output) system. This algorithm takes account of all constraints of inputs and their increments. By solvin...A constrained decoupling (generalized predictive control) GPC algorithm is proposed for MIMO (malti-input multi-output) system. This algorithm takes account of all constraints of inputs and their increments. By solving matrix equations, the multi-step predictive decoupling controllers are realized. This algorithm need not solve Diophantine functions, and weakens the cross-coupling of the variables. At last the simulation results demon- strate the effectiveness of this proposed strategy.展开更多
Standard genetic algorithms （SGAs） are investigated to optimise discrete-time proportional-integral-derivative （PID） con- troller parameters, by three tuning approaches, for a multivariable glass furnace process w...Standard genetic algorithms （SGAs） are investigated to optimise discrete-time proportional-integral-derivative （PID） con- troller parameters, by three tuning approaches, for a multivariable glass furnace process with loop interaction. Initially, standard genetic algorithms （SGAs） are used to identify control oriented models of the plant which are subsequently used for controller optimisa- tion. An individual tuning approach without loop interaction is considered first to categorise the genetic operators, cost functions and improve searching boundaries to attain the desired performance criteria. The second tuning approach considers controller parameters optimisation with loop interaction and individual cost functions. While, the third tuning approach utilises a modified cost function which includes the total effect of both controlled variables, glass temperature and excess oxygen. This modified cost function is shown to exhibit improved control robustness and disturbance rejection under loop interaction.展开更多
Blast furnace （BF） ironmaking process has complex and nonlinear dynamic characteristics. The molten iron temperature （MIT） as well as Si, P and S contents of molten iron is difficult to be directly measured online...Blast furnace （BF） ironmaking process has complex and nonlinear dynamic characteristics. The molten iron temperature （MIT） as well as Si, P and S contents of molten iron is difficult to be directly measured online, and large-time delay exists in offline analysis through laboratory sampling. A nonlinear multivariate intelligent modeling method was proposed for molten iron quality （MIQ） based on principal component analysis （PCA） and dynamic ge- netic neural network. The modeling method used the practical data processed by PCA dimension reduction as inputs of the dynamic artificial neural network （ANN）. A dynamic feedback link was introduced to produce a dynamic neu- ral network on the basis of traditional back propagation ANN. The proposed model improved the dynamic adaptabili- ty of networks and solved the strong fluctuation and resistance problem in a nonlinear dynamic system. Moreover, a new hybrid training method was presented where adaptive genetic algorithms （AGA） and ANN were integrated, which could improve network convergence speed and avoid network into local minima. The proposed method made it easier for operators to understand the inside status of blast furnace and offered real-time and reliable feedback infor- mation for realizing close-loop control for MIQ. Industrial experiments were made through the proposed model based on data collected from a practical steel company. The accuracy could meet the requirements of actual operation.展开更多
The control of gas fractionation unit(GFU) in petroleum industry is very difficult due to multivariable characteristics and a large time delay.PID controllers are still applied in most industry processes.However,the t...The control of gas fractionation unit(GFU) in petroleum industry is very difficult due to multivariable characteristics and a large time delay.PID controllers are still applied in most industry processes.However,the traditional PID control has been proven not sufficient and capable for this particular petro-chemical process.In this work,an incremental multivariable predictive functional control(IMPFC) algorithm was proposed with less online computation,great precision and fast response.An incremental transfer function matrix model was set up through the step-response data,and predictive outputs were deduced with the theory of single-value optimization.The results show that the method can optimize the incremental control variable and reject the constraint of the incremental control variable with the positional predictive functional control algorithm,and thereby making the control variable smoother.The predictive output error and future set-point were approximated by a polynomial,which can overcome the problem under the model mismatch and make the predictive outputs track the reference trajectory.Then,the design of incremental multivariable predictive functional control was studied.Simulation and application results show that the proposed control strategy is effective and feasible to improve control performance and robustness of process.展开更多
Accurate sales prediction in filling stations is the basis to fill in the refined oil in time and avoid the outof-stock as much as possible.Considering the defect of great“lag”in the general time series model,this p...Accurate sales prediction in filling stations is the basis to fill in the refined oil in time and avoid the outof-stock as much as possible.Considering the defect of great“lag”in the general time series model,this paper summarizes the multiple factors that influence the oil sales and develops a multivariable long short-term memory(LSTM)neural network,with the hyper-parameters being improved by the genetic algorithm(GA).To further improve the prediction accuracy,the proposed LSTM neural network is generalized to bidirectional LSTM(Bi LSTM),in which the impact of future factors on present sales can be taken into account by backward training.Finally,different LSTM structures and genetic algorithm parameters are tested to discuss their impact on prediction accuracy.Results demonstrated that genetic algorithm improved Bi LSTM model is superior to extreme gradient boosting,ARIMA,and artificial neural network,having the highest accuracy of 89%.展开更多
Molten iron temperature as well as Si, P, and S contents is the most essential molten iron quality （MIQ） indices in the blast furnace （BF） ironmaking, which requires strict monitoring during the whole ironmaking p...Molten iron temperature as well as Si, P, and S contents is the most essential molten iron quality （MIQ） indices in the blast furnace （BF） ironmaking, which requires strict monitoring during the whole ironmaking production. However, these MIQ parameters are difficult to be directly measured online, and large-time delay exists in off-line analysis through laboratory sampling. Focusing on the practical challenge, a data-driven modeling method was presented for the prediction of MIQ using the improved muhivariable incremental random vector functional-link net- works （M-I-RVFLNs）. Compared with the conventional random vector functional-link networks （RVFLNs） and the online sequential RVFLNs, the M-I-RVFLNs have solved the problem of deciding the optimal number of hidden nodes and overcome the overfitting problems. Moreover, the proposed M I RVFLNs model has exhibited the potential for multivariable prediction of the MIQ and improved the terminal condition for the multiple-input multiple-out- put （MIMO） dynamic system, which is suitable for the BF ironmaking process in practice. Ultimately, industrial experiments and contrastive researches have been conducted on the BF No. 2 in Liuzhou Iron and Steel Group Co. Ltd. of China using the proposed method, and the results demonstrate that the established model produces better estima ting accuracy than other MIQ modeling methods.展开更多
Based on the generalized variational principle and B-spline wavelet on the interval （BSWI）, the multivariable BSWI elements with two kinds of variables （TBSWI） for hyperboloidal shell and open cylindrical shell ar...Based on the generalized variational principle and B-spline wavelet on the interval （BSWI）, the multivariable BSWI elements with two kinds of variables （TBSWI） for hyperboloidal shell and open cylindrical shell are constructed in this paper. Different from the traditional method, the present one treats the generalized displacement and stress as independent variables. So differentiation and integration are avoided in calculating generalized stress and thus the precision is improved. Furthermore, compared with commonly used Daubechies wavelet, BSWI has explicit expression and excellent approximation property and thus further guarantee satisfactory results. Finally, the efficiency of the constructed multivariable shell elements is validated through several numerical examples.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
This paper proposes a multivariable fixed-time leaderfollower formation control method for a group of nonholonomic mobile robots, which has the ability to estimate multiple uncertainties. Firstly, based on the state s...This paper proposes a multivariable fixed-time leaderfollower formation control method for a group of nonholonomic mobile robots, which has the ability to estimate multiple uncertainties. Firstly, based on the state space model of the leader-follower formation, a multivariable fixed-time formation kinematics controller is designed. Secondly, to overcome uncertainties existing in the nonholonomic mobile robot system, such as load change,friction, external disturbance, a multivariable fixed-time torque controller based on the fixed-time disturbance observer at the dynamic level is designed. The designed torque controller is cascaded with the formation controller and finally realizes accurate estimation of the uncertain part of the system, the follower tracking of reference velocity and the desired formation of the leader and the follower in a fixed-time. The fixed-time upper bound is completely determined by the controller parameters, which is independent of the initial state of the system. The multivariable fixed-time control theory and the Lyapunov method are adopted to ensure the system stability.Finally, the effectiveness of the proposed algorithm is verified by the experimental simulation.展开更多
This paper describes empirical research on the model, optimization and supervisory control of beer fermentation.Conditions in the laboratory were made as similar as possible to brewery industry conditions. Since mathe...This paper describes empirical research on the model, optimization and supervisory control of beer fermentation.Conditions in the laboratory were made as similar as possible to brewery industry conditions. Since mathematical models that consider realistic industrial conditions were not available, a new mathematical model design involving industrial conditions was first developed. Batch fermentations are multiobjective dynamic processes that must be guided along optimal paths to obtain good results.The paper describes a direct way to apply a Pareto set approach with multiobjective evolutionary algorithms (MOEAs).Successful finding of optimal ways to drive these processes were reported.Once obtained, the mathematical fermentation model was used to optimize the fermentation process by using an intelligent control based on certain rules.展开更多
A novel method of incorporating generalized predictive control (GPC) algorithms based on quantitative feedback theory (QFT) principles is proposed for solving the feedback control problem of the highly uncertain and c...A novel method of incorporating generalized predictive control (GPC) algorithms based on quantitative feedback theory (QFT) principles is proposed for solving the feedback control problem of the highly uncertain and cross-coupling plants. The quantitative feedback theory decouples the multi-input and multi-output (MIMO) plant and is also used to reduce the uncertainties of the system, stabilize the system, and achieve tracking performance of the system to a certain extent. Single-input and single-output (SISO) generalized predictive control is used to achieve performance with higher performance. In GPC, the model is identified on-line, which is based on the QFT input and the plant output signals. The simulation results show that the performance of the system is superior to the performance when only QFT is used for highly uncertain MIMO plants.展开更多
Pseudo-division algorithm for matrix multivariable polynomial are given, thereby with the view of differential algebra, the sufficient and necessary conditions for transforming a class of partial differential equation...Pseudo-division algorithm for matrix multivariable polynomial are given, thereby with the view of differential algebra, the sufficient and necessary conditions for transforming a class of partial differential equations into infinite dimensional Hamiltonianian system and its concrete form are obtained. Then by combining this method with Wu's method, a new method of constructing general solution of a class of mechanical equations is got, which several examples show very effective.展开更多
基金support by the National Natural Science Foundation of China (No.52202474)China Postdoctoral Science Foundation (No.2023M731655)+3 种基金Major Projects of National Science and Technology,China (No.J2019-I-0020-0019)Advanced Aviation Power Innovation Workstation Project,China (No.HKCX2022-01-026-03)Basic Research Business Fees for Central Universities,China (No.NT2023004)Nanjing University of Aeronautics and Astronautics Forward-looking Layout Research Project,China (No.1002-ILA22037-1A22).
文摘Traditional centralized Proportional Integral(PI)controller design methods based on Equivalent Transfer Functions(ETFs)have poor decoupling effect in turboprop engines.In this paper,a centralized PI design method based on dynamic imaginary matrix and equivalent transfer function is proposed.Firstly,a method for solving equivalent transfer functions based on the dynamic imaginary matrix is proposed,which adopts dynamic imaginary matrix to describe the dynamic characteristics of the system,and obtains the equivalent transfer function based on the dynamic imaginary matrix characteristics.Secondly,for the equivalent transfer function,a central-ized PI control gain is designed using the Taylor expansion method.Meanwhile,this paper further proves that the centralized PI design method proposed in this paper has integral stability.Consid-ering the impact of altitude and Mach number on turboprop engines,a linear feedforward control method based on the transfer function matrix is further proposed based on the centralized PI con-troller,and the stability of the entire comprehensive control method is proved.Finally,to ensure the safe and effective operation of the turboprop engine,a temperature and torque limiting protection controller is designed for the turboprop engine.Simulation results show that the centralized PI con-troller design method and linear feedforward control method proposed can effectively improve the control quality of turboprop engine control systems.
基金supported by the UoA Start-up Grant,UQ Cyber Security Seed Funding,the Australian Research Council Linkage Project(LP230200821)the Australian Research Council Early Career Industry Fellowship(IE240100275)+1 种基金the Australian Research Council Discovery Project(DP240103070)the Australian Research Council Discovery Early Career Researcher Award(DE230101116).
文摘Multivariate time series(MTS)data are vital for various applications,particularly in machine learning tasks.However,challenges such as sensor failures can result in irregular and misaligned data with missing values,thereby complicating their analysis.While recent advancements use graph neural networks(GNNs)to manage these Irregular Multivariate Time Series(IMTS)data,they generally require a reliable graph structure,either pre-existing or inferred from adequate data to properly capture node correlations.This poses a challenge in applications where IMTS data are often streamed and waiting for future data to estimate a suitable graph structure becomes impractical.To overcome this,we introduce a dynamic GNN model suited for streaming characteristics of IMTS data,incorporating an instance-attention mechanism that dynamically learns and updates graph edge weights for real-time analysis.We also tailor strategies for high-frequency and low-frequency data to enhance prediction accuracy.Empirical results on real-world datasets demonstrate the superiority of our proposed model in both classification and imputation tasks.
文摘In engineering systems,uncertainties in input parameters can significantly influence the output responses.This paper proposes a model distance-based approach to perform global sensitivity analysis for quantifying the influence of input uncertainties on multiple responses in an engineering system.The sensitivity indices are determined by comparing a reference model that incorporates all system uncertainties,with an altered model,where specific uncertainties are constrained.The proposed framework employs probability distance measures such as Hellinger distance,Kullback-Leibler divergence,and I2 norm which are based on joint probability density functions.The study also demonstrates the equivalence between the l2 norm-based approach and Sobol's analysis in multivariate sensitivity context.The proposed methodology effectively manages correlated random variables,accommodates both Gaussian and non-Gaussian distributions,and allows for the grouping of input variables.Ilustrative examples consist of static analysis of a truss system and dynamic analysis of a frame subjected to seismic excitation.The sensitivity indices are estimated using brute-force Monte Carlo simulations.The relative ranking of these sensitivity indices can be utilized to identify the most and least significant variables contributing to the response uncertainty.The numerical results show a consistent ranking of input variables across different probability measures,indicating the robustness of proposed framework.
基金Supported by the National Natural Science Foundation of China(12271154)the Natural Science Foundation of Hunan Province(2022JJ30234)the Postgraduate Scientific Research Innovation Project of Hunan Province(CX20231032)。
文摘The zero coprime system equivalence is one of important research in the theory of multidimensional system equivalence,and is closely related to zero coprime equivalence of multivariate polynomial matrices.We first discuss the relation between zero coprime equivalence and unimodular equivalence for polynomial matrices.Then,we investigate the zero coprime equivalence problem for several classes of polynomial matrices,some novel findings and criteria on reducing these matrices to their Smith normal forms are obtained.Finally,an example is provided to illustrate the main results.
文摘A new algorithm for constructing an inverse of a multivariable linear system is presented. This algorithm makes the constructing an inverse of the higher order matrices into searching for the equivalent normal form of the lower order matrices. Consequently, the calculation is more simple efficient and programmed than previous methods. Another result of the paper is that the lower reduced inverse system is obtained, by selecting special bases of the observable space of the original systems, it reveals the effect of the observability of the original systems on the order of the inverse systems.
文摘Through modifying the CPN model, a kind of multivariable fuzzy model is put forward, and the matching fuzzy multistep predictive control algorithm is deduced based on the model. The modified model works in a competitive output manner which results in its local representation property. While studying on line, only a few parameters need to be regulated. So the model has the merits of fast learning and on line self organizing modeling. The control algorithm is simple, adaptive and useful in multivariable and time delay systems. Applying the algorithm in a paper making system, simulation shows its good effect.
文摘A decentralized model reference adaptive control (MRAC) scheme is proposed and applied to design a multivariable control system of a dual-spool turbofan engine.Simulation studies show good static and dynamic performance of the system over the fullflight envelope. Simulation results also show the good effectiveness of reducing interactionin the multivariable system with significant coupling. The control system developed has awide frequency band to satisfy the strict engineering requirement and is practical for engineering applications.
文摘This paper presents a multivariable generalized predictive controller with proportion and integration structure by modifying the quadratic criterion of the usual MGPC. The control performance has been improved greatly. The effectiveness of the controller is demonstrated by the simulation result.
基金Supported by the National Natural Science Foundation of China (No.60374037, No.60574036), the Program for New Century Excellent Talents in University of China (NCET), and the Specialized Research Fund for the Doctoral Program of Higher Edu-cation of China (No.20050055013).
文摘A constrained decoupling (generalized predictive control) GPC algorithm is proposed for MIMO (malti-input multi-output) system. This algorithm takes account of all constraints of inputs and their increments. By solving matrix equations, the multi-step predictive decoupling controllers are realized. This algorithm need not solve Diophantine functions, and weakens the cross-coupling of the variables. At last the simulation results demon- strate the effectiveness of this proposed strategy.
文摘Standard genetic algorithms （SGAs） are investigated to optimise discrete-time proportional-integral-derivative （PID） con- troller parameters, by three tuning approaches, for a multivariable glass furnace process with loop interaction. Initially, standard genetic algorithms （SGAs） are used to identify control oriented models of the plant which are subsequently used for controller optimisa- tion. An individual tuning approach without loop interaction is considered first to categorise the genetic operators, cost functions and improve searching boundaries to attain the desired performance criteria. The second tuning approach considers controller parameters optimisation with loop interaction and individual cost functions. While, the third tuning approach utilises a modified cost function which includes the total effect of both controlled variables, glass temperature and excess oxygen. This modified cost function is shown to exhibit improved control robustness and disturbance rejection under loop interaction.
基金Sponsored by National Natural Science Foundation of China(61290323,61333007,614730646)IAPI Fundamental Research Funds(2013ZCX02-09)+1 种基金Fundamental Research Funds for the Central Universities of China(N130508002,N130108001)National High-tech Research and Development Program of China(2015AA043802)
文摘Blast furnace （BF） ironmaking process has complex and nonlinear dynamic characteristics. The molten iron temperature （MIT） as well as Si, P and S contents of molten iron is difficult to be directly measured online, and large-time delay exists in offline analysis through laboratory sampling. A nonlinear multivariate intelligent modeling method was proposed for molten iron quality （MIQ） based on principal component analysis （PCA） and dynamic ge- netic neural network. The modeling method used the practical data processed by PCA dimension reduction as inputs of the dynamic artificial neural network （ANN）. A dynamic feedback link was introduced to produce a dynamic neu- ral network on the basis of traditional back propagation ANN. The proposed model improved the dynamic adaptabili- ty of networks and solved the strong fluctuation and resistance problem in a nonlinear dynamic system. Moreover, a new hybrid training method was presented where adaptive genetic algorithms （AGA） and ANN were integrated, which could improve network convergence speed and avoid network into local minima. The proposed method made it easier for operators to understand the inside status of blast furnace and offered real-time and reliable feedback infor- mation for realizing close-loop control for MIQ. Industrial experiments were made through the proposed model based on data collected from a practical steel company. The accuracy could meet the requirements of actual operation.
基金Project(61203021)supported by the National Natural Science Foundation of ChinaProject(2011216011)supported by the Scientific and Technological Program of Liaoning Province,China+2 种基金Project(2013020024)supported by the Natural Science Foundation of Liaoning Province,ChinaProject(2012BAF05B00)supported by the National Science and Technology Support Program,ChinaProject(LJQ2015061)supported by the Program for Liaoning Excellent Talents in Universities,China
文摘The control of gas fractionation unit(GFU) in petroleum industry is very difficult due to multivariable characteristics and a large time delay.PID controllers are still applied in most industry processes.However,the traditional PID control has been proven not sufficient and capable for this particular petro-chemical process.In this work,an incremental multivariable predictive functional control(IMPFC) algorithm was proposed with less online computation,great precision and fast response.An incremental transfer function matrix model was set up through the step-response data,and predictive outputs were deduced with the theory of single-value optimization.The results show that the method can optimize the incremental control variable and reject the constraint of the incremental control variable with the positional predictive functional control algorithm,and thereby making the control variable smoother.The predictive output error and future set-point were approximated by a polynomial,which can overcome the problem under the model mismatch and make the predictive outputs track the reference trajectory.Then,the design of incremental multivariable predictive functional control was studied.Simulation and application results show that the proposed control strategy is effective and feasible to improve control performance and robustness of process.
基金partially supported by the National Natural Science Foundation of China(51874325)Science Foundation of China University of Petroleum,Beijing(2462021BJRC009)。
文摘Accurate sales prediction in filling stations is the basis to fill in the refined oil in time and avoid the outof-stock as much as possible.Considering the defect of great“lag”in the general time series model,this paper summarizes the multiple factors that influence the oil sales and develops a multivariable long short-term memory(LSTM)neural network,with the hyper-parameters being improved by the genetic algorithm(GA).To further improve the prediction accuracy,the proposed LSTM neural network is generalized to bidirectional LSTM(Bi LSTM),in which the impact of future factors on present sales can be taken into account by backward training.Finally,different LSTM structures and genetic algorithm parameters are tested to discuss their impact on prediction accuracy.Results demonstrated that genetic algorithm improved Bi LSTM model is superior to extreme gradient boosting,ARIMA,and artificial neural network,having the highest accuracy of 89%.
基金Item Sponsored by National Natural Science Foundation of China(61290323,61333007,61473064)Fundamental Research Funds for Central Universities of China(N130108001)+1 种基金National High Technology Research and Development Program of China(2015AA043802)General Project on Scientific Research for Education Department of Liaoning Province of China(L20150186)
文摘Molten iron temperature as well as Si, P, and S contents is the most essential molten iron quality （MIQ） indices in the blast furnace （BF） ironmaking, which requires strict monitoring during the whole ironmaking production. However, these MIQ parameters are difficult to be directly measured online, and large-time delay exists in off-line analysis through laboratory sampling. Focusing on the practical challenge, a data-driven modeling method was presented for the prediction of MIQ using the improved muhivariable incremental random vector functional-link net- works （M-I-RVFLNs）. Compared with the conventional random vector functional-link networks （RVFLNs） and the online sequential RVFLNs, the M-I-RVFLNs have solved the problem of deciding the optimal number of hidden nodes and overcome the overfitting problems. Moreover, the proposed M I RVFLNs model has exhibited the potential for multivariable prediction of the MIQ and improved the terminal condition for the multiple-input multiple-out- put （MIMO） dynamic system, which is suitable for the BF ironmaking process in practice. Ultimately, industrial experiments and contrastive researches have been conducted on the BF No. 2 in Liuzhou Iron and Steel Group Co. Ltd. of China using the proposed method, and the results demonstrate that the established model produces better estima ting accuracy than other MIQ modeling methods.
基金supported by the National Natural Science Foundation of China (No. 50875195)the Foundation for the Author of National Excellent Doctoral Dissertation of China (No. 2007B33)the Key Project of the National Natural Science Foundation of China (No. 51035007)
文摘Based on the generalized variational principle and B-spline wavelet on the interval （BSWI）, the multivariable BSWI elements with two kinds of variables （TBSWI） for hyperboloidal shell and open cylindrical shell are constructed in this paper. Different from the traditional method, the present one treats the generalized displacement and stress as independent variables. So differentiation and integration are avoided in calculating generalized stress and thus the precision is improved. Furthermore, compared with commonly used Daubechies wavelet, BSWI has explicit expression and excellent approximation property and thus further guarantee satisfactory results. Finally, the efficiency of the constructed multivariable shell elements is validated through several numerical examples.
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(61872204)the Natural Science Foundation of Heilongjiang Province of China(F2015025)。
文摘This paper proposes a multivariable fixed-time leaderfollower formation control method for a group of nonholonomic mobile robots, which has the ability to estimate multiple uncertainties. Firstly, based on the state space model of the leader-follower formation, a multivariable fixed-time formation kinematics controller is designed. Secondly, to overcome uncertainties existing in the nonholonomic mobile robot system, such as load change,friction, external disturbance, a multivariable fixed-time torque controller based on the fixed-time disturbance observer at the dynamic level is designed. The designed torque controller is cascaded with the formation controller and finally realizes accurate estimation of the uncertain part of the system, the follower tracking of reference velocity and the desired formation of the leader and the follower in a fixed-time. The fixed-time upper bound is completely determined by the controller parameters, which is independent of the initial state of the system. The multivariable fixed-time control theory and the Lyapunov method are adopted to ensure the system stability.Finally, the effectiveness of the proposed algorithm is verified by the experimental simulation.
文摘This paper describes empirical research on the model, optimization and supervisory control of beer fermentation.Conditions in the laboratory were made as similar as possible to brewery industry conditions. Since mathematical models that consider realistic industrial conditions were not available, a new mathematical model design involving industrial conditions was first developed. Batch fermentations are multiobjective dynamic processes that must be guided along optimal paths to obtain good results.The paper describes a direct way to apply a Pareto set approach with multiobjective evolutionary algorithms (MOEAs).Successful finding of optimal ways to drive these processes were reported.Once obtained, the mathematical fermentation model was used to optimize the fermentation process by using an intelligent control based on certain rules.
基金Supported by the National Natural Science Foundation of China (No.60374037, No.60574036), the Program for New Century Excellent Talents in Education Ministry (NCET), and the Specialized Research Fund for the Doctoral Program of Higher Education of China (No.20050055013).
文摘A novel method of incorporating generalized predictive control (GPC) algorithms based on quantitative feedback theory (QFT) principles is proposed for solving the feedback control problem of the highly uncertain and cross-coupling plants. The quantitative feedback theory decouples the multi-input and multi-output (MIMO) plant and is also used to reduce the uncertainties of the system, stabilize the system, and achieve tracking performance of the system to a certain extent. Single-input and single-output (SISO) generalized predictive control is used to achieve performance with higher performance. In GPC, the model is identified on-line, which is based on the QFT input and the plant output signals. The simulation results show that the performance of the system is superior to the performance when only QFT is used for highly uncertain MIMO plants.
文摘Pseudo-division algorithm for matrix multivariable polynomial are given, thereby with the view of differential algebra, the sufficient and necessary conditions for transforming a class of partial differential equations into infinite dimensional Hamiltonianian system and its concrete form are obtained. Then by combining this method with Wu's method, a new method of constructing general solution of a class of mechanical equations is got, which several examples show very effective.
