A fuzzy neural network controller with the teaching controller guidance and parameter regulations for vector-controlled induction motor is proposed. The design procedures of the fuzzy neural controller and the teachin...A fuzzy neural network controller with the teaching controller guidance and parameter regulations for vector-controlled induction motor is proposed. The design procedures of the fuzzy neural controller and the teaching controller are described. The parameters of the membership function are regulated by an on-line learning algorithm. The speed responses of the system under the condition, where the target functions are chosen as I qs and ω, are analyzed. The system responses with the variant of parameter moment of inertial J, viscous coefficients B and torque constant K tare also analyzed. Simulation results show that the control scheme and the controller have the advantages of rapid speed response and good robustness.展开更多
Ensemble learning is a wildly concerned issue.Traditional ensemble techniques are always adopted to seek better results with labeled data and base classifiers.They fail to address the ensemble task where only unlabele...Ensemble learning is a wildly concerned issue.Traditional ensemble techniques are always adopted to seek better results with labeled data and base classifiers.They fail to address the ensemble task where only unlabeled data are available.A label propagation based ensemble(LPBE) approach is proposed to further combine base classification results with unlabeled data.First,a graph is constructed by taking unlabeled data as vertexes,and the weights in the graph are calculated by correntropy function.Average prediction results are gained from base classifiers,and then propagated under a regularization framework and adaptively enhanced over the graph.The proposed approach is further enriched when small labeled data are available.The proposed algorithms are evaluated on several UCI benchmark data sets.Results of simulations show that the proposed algorithms achieve satisfactory performance compared with existing ensemble methods.展开更多
Based on FLUENT software, studied the blast fume diffusion characteristics of a driving face under 8 situations, and analyzed the influence to fume diffusion of different drift length and different ventilation pipe ou...Based on FLUENT software, studied the blast fume diffusion characteristics of a driving face under 8 situations, and analyzed the influence to fume diffusion of different drift length and different ventilation pipe outlet positions. Through regression analysis, functions of ventilation time to the drift length and ventilation pipe outlet position was obtained.展开更多
A fully coupled 6-degree-of-freedom nonlinear dynamic model is presented to analyze the dynamic response of a semi-submersible platform which is equipped with the dynamic positioning(DP) system. In the control force d...A fully coupled 6-degree-of-freedom nonlinear dynamic model is presented to analyze the dynamic response of a semi-submersible platform which is equipped with the dynamic positioning(DP) system. In the control force design, a dynamic model of reference linear drift frequency in the horizontal plane is introduced. The dynamic surface control(DSC) is used to design a control strategy for the DP. Compared with the traditional back-stepping methods, the dynamic surface control combined with radial basis function(RBF) neural networks(NNs) can avoid differentiating intermediate variables repeatedly in every design step due to the introduction of a first order filter. Low frequency motions obtained from total motions by a low pass filter are chosen to be the inputs for the RBF NNs which are used to approximate the low frequency wave force. Considering the propellers' wear and tear, the effect of filtering frequencies for the control force is discussed. Based on power consumptions and positioning requirements, the NN centers are determined. Moreover, the RBF NNs used to approximate the total wave force are built to monitor the disturbances. With the DP assistance, the results of fully coupled dynamic response simulations are given to illustrate the effectiveness of the proposed control strategy.展开更多
Based on the Navier-Stokes Equations (NSE), numerical simulation with fine grids is conducted to simulate the coastal surface wave changes, including wave generation, propagation, transformation and interactions betwe...Based on the Navier-Stokes Equations (NSE), numerical simulation with fine grids is conducted to simulate the coastal surface wave changes, including wave generation, propagation, transformation and interactions between waves and structures. This numerical model has been tested for the generation of the desired incident waves, including both regular and random waves. Some numerical results of this model are compared with available experimental data. In order to apply this model to actual cases, boundary conditions are considered in detail for different shoreline types (beach or breakwater, slope or vertical wall, etc. ). Finally, the utility of the model to a real coastal area is shown by applying it to a fishing port located in Shidao, Rongcheng, Shandong Province, P.R. China.展开更多
The location of wind turbines on a continuous hilly terrain has an influence on its power outputs.A CFDbased approach is developed to investigate the complex aerodynamic interference between two wind turbines and the ...The location of wind turbines on a continuous hilly terrain has an influence on its power outputs.A CFDbased approach is developed to investigate the complex aerodynamic interference between two wind turbines and the hilly terrain.In this approach,a new three-dimensional model of hilly terrain is established to analyze its viscous effect,and a wind shear is modelled through logarithmic function.They are coupled into the aerodynamics of wind turbine based on“FLUENT”software.Then we apply the proposed method to the NREL Phase VI wind turbines and compare with an experiment in the atmospheric boundary layer(ABL)wind tunnel to validate its accuracy.The simulation also investigates the power outputs of wind turbines on the flat ground and the continuous hilly terrain by changing the location of the wind turbine related to the hilly terrain and the shape of the 1st hill.The results show that the wind turbine located on the top of the 2nd hill has the maximum power;and that when the wind turbine is located on the downstream of the hill,the stall zone should be avoided,and the power of the wind turbine located on the side of the hill is higher than that of the wind turbine located on the front and rear of the hilly terrain.展开更多
文摘A fuzzy neural network controller with the teaching controller guidance and parameter regulations for vector-controlled induction motor is proposed. The design procedures of the fuzzy neural controller and the teaching controller are described. The parameters of the membership function are regulated by an on-line learning algorithm. The speed responses of the system under the condition, where the target functions are chosen as I qs and ω, are analyzed. The system responses with the variant of parameter moment of inertial J, viscous coefficients B and torque constant K tare also analyzed. Simulation results show that the control scheme and the controller have the advantages of rapid speed response and good robustness.
基金Project (20121101004) supported by the Major Science and Technology Program of Shanxi Province,ChinaProject (20130321004-01) supported by the Key Technologies R&D Program of Shanxi Province,China+2 种基金Project (2013M530896) supported by the Postdoctoral Science Foundation of ChinaProject (2014021022-6) supported by the Shanxi Provincial Science Foundation for Youths,ChinaProject (80010302010053) supported by the Shanxi Characteristic Discipline Fund,China
文摘Ensemble learning is a wildly concerned issue.Traditional ensemble techniques are always adopted to seek better results with labeled data and base classifiers.They fail to address the ensemble task where only unlabeled data are available.A label propagation based ensemble(LPBE) approach is proposed to further combine base classification results with unlabeled data.First,a graph is constructed by taking unlabeled data as vertexes,and the weights in the graph are calculated by correntropy function.Average prediction results are gained from base classifiers,and then propagated under a regularization framework and adaptively enhanced over the graph.The proposed approach is further enriched when small labeled data are available.The proposed algorithms are evaluated on several UCI benchmark data sets.Results of simulations show that the proposed algorithms achieve satisfactory performance compared with existing ensemble methods.
文摘Based on FLUENT software, studied the blast fume diffusion characteristics of a driving face under 8 situations, and analyzed the influence to fume diffusion of different drift length and different ventilation pipe outlet positions. Through regression analysis, functions of ventilation time to the drift length and ventilation pipe outlet position was obtained.
基金funded by the National Basic Research Program of China (Grant Nos. 2011CB013702 and 2011CB013703)the Science Fund for Creative Research Groups of the National Natural Science Foundation of China (Grant No. 50921001)
文摘A fully coupled 6-degree-of-freedom nonlinear dynamic model is presented to analyze the dynamic response of a semi-submersible platform which is equipped with the dynamic positioning(DP) system. In the control force design, a dynamic model of reference linear drift frequency in the horizontal plane is introduced. The dynamic surface control(DSC) is used to design a control strategy for the DP. Compared with the traditional back-stepping methods, the dynamic surface control combined with radial basis function(RBF) neural networks(NNs) can avoid differentiating intermediate variables repeatedly in every design step due to the introduction of a first order filter. Low frequency motions obtained from total motions by a low pass filter are chosen to be the inputs for the RBF NNs which are used to approximate the low frequency wave force. Considering the propellers' wear and tear, the effect of filtering frequencies for the control force is discussed. Based on power consumptions and positioning requirements, the NN centers are determined. Moreover, the RBF NNs used to approximate the total wave force are built to monitor the disturbances. With the DP assistance, the results of fully coupled dynamic response simulations are given to illustrate the effectiveness of the proposed control strategy.
基金supported by the National Natural Foundation of China(No.50479027)the Natural Science Foundation of Qingdao(Grant No.03-jr-15).
文摘Based on the Navier-Stokes Equations (NSE), numerical simulation with fine grids is conducted to simulate the coastal surface wave changes, including wave generation, propagation, transformation and interactions between waves and structures. This numerical model has been tested for the generation of the desired incident waves, including both regular and random waves. Some numerical results of this model are compared with available experimental data. In order to apply this model to actual cases, boundary conditions are considered in detail for different shoreline types (beach or breakwater, slope or vertical wall, etc. ). Finally, the utility of the model to a real coastal area is shown by applying it to a fishing port located in Shidao, Rongcheng, Shandong Province, P.R. China.
基金supported by the Natural Science Foundation of Jiangsu Province (No. BK20161537)National Science Key Laboratory Foundation(No.6142220180202)+1 种基金Rotor Aerodynamics Key Laboratory Foundation (No.RAL20180303-1)National Natural Science Foundation of China(No.11502105).
文摘The location of wind turbines on a continuous hilly terrain has an influence on its power outputs.A CFDbased approach is developed to investigate the complex aerodynamic interference between two wind turbines and the hilly terrain.In this approach,a new three-dimensional model of hilly terrain is established to analyze its viscous effect,and a wind shear is modelled through logarithmic function.They are coupled into the aerodynamics of wind turbine based on“FLUENT”software.Then we apply the proposed method to the NREL Phase VI wind turbines and compare with an experiment in the atmospheric boundary layer(ABL)wind tunnel to validate its accuracy.The simulation also investigates the power outputs of wind turbines on the flat ground and the continuous hilly terrain by changing the location of the wind turbine related to the hilly terrain and the shape of the 1st hill.The results show that the wind turbine located on the top of the 2nd hill has the maximum power;and that when the wind turbine is located on the downstream of the hill,the stall zone should be avoided,and the power of the wind turbine located on the side of the hill is higher than that of the wind turbine located on the front and rear of the hilly terrain.
