Aimed at plane nonlinear two-degree-of-freedom (2-dof) manipulator, which is a nonlinear multi-input and multi-output(MIMO) system, its joint angles are controlled by model-free adaptive control (MFAC) theory to reali...
详细信息
ISBN:
(纸本)9781538626184
Aimed at plane nonlinear two-degree-of-freedom (2-dof) manipulator, which is a nonlinear multi-input and multi-output(MIMO) system, its joint angles are controlled by model-free adaptive control (MFAC) theory to realize trajectory tracking. the nonlinear system model is replaced by the compact form dynamic linearization time-varying model, and the pseudo-Jacobian matrix of the system is estimated on the basis of the input and output data of the manipulator model. the simulation results show that the compact form dynamic linearized model-free adaptive control (CFDL-MFAC) algorithm can effectively ensure the tracking performance of the system output, and the error remains within a certain range.
this paper investigates a PD-type spatial iterative learningcontrol (SILC) method for the wind turbine pitch control system in order to maintain the stationary output power constant withthe wind speed increase in re...
详细信息
ISBN:
(纸本)9781538626184
this paper investigates a PD-type spatial iterative learningcontrol (SILC) method for the wind turbine pitch control system in order to maintain the stationary output power constant withthe wind speed increase in region 3. the pitch control system is considered as the repetitive operation system, then the temporal domain linear time-invariant pitch control system transforms to a spatial domain linear spatial-variant pitch control system, the PD-type SILC algorithmic generates the upgraded pitch angle control inputs by compensating for the initial input with proportional and derivative actions based on the tracking error between the desired output rotor speed and the measured rotor speed in real time. By adopting the Lebesgue-p norm and the generalized Young inequality of convolution integral, the convergence of the PD-type SILC for pitch control system is derived. Finally, some numerical simulations are presented to verify the effectiveness and validity of the SILC in wind turbine pitch control system.
this paper studies a discrete control algorithm of through sampling feedback. On the digital computer control platform, a sampled-datacontroller is designed to control a quadrotor UAV system based on backstepping con...
详细信息
ISBN:
(纸本)9781538626184
this paper studies a discrete control algorithm of through sampling feedback. On the digital computer control platform, a sampled-datacontroller is designed to control a quadrotor UAV system based on backstepping control method and the proposed continuous controller. the results show that, compared withtheoretically continuous time control scheme, our discrete controller can realize the purpose of stable flight. Finally, a simulation example is given to show the effectiveness of the proposed control scheme.
In recent years, the policy gradient method in intensive learning has attracted wide attention with its good convergence performance. At the same time, regulation of hyper parameters is also a matter of concern. Based...
详细信息
ISBN:
(纸本)9781538626184
In recent years, the policy gradient method in intensive learning has attracted wide attention with its good convergence performance. At the same time, regulation of hyper parameters is also a matter of concern. Based on the advantages of Actor-Critic structure (AC), the Natural-Gradient Actor-Critic algorithm (NAC) in the discount model is studied in this article. then the Natural-Gradient Actor-Critic with ADADELTA (A-NAC) algorithm is proposed. the use of ADADELTA is adapted to adjust the learning rate in the actor network, and further improves the convergence speed of the NAC algorithm. Simulation results show that NAC/A-NAC have better learning efficiency and faster convergence rate than regular gradient AC methods.
In this article, direct at the batch process controlled by iterative learningcontrol (ILC), it proposes a more reasonable benchmark for the control performance assessment (CPA): the model predictive control (MPC) ben...
详细信息
ISBN:
(纸本)9781538626184
In this article, direct at the batch process controlled by iterative learningcontrol (ILC), it proposes a more reasonable benchmark for the control performance assessment (CPA): the model predictive control (MPC) benchmark. First, ILC-controlled batch process is converted to a 2-D Fornasini-Marchesini (FM) model. On this basis, the cost function is constructed and the optimal learning law can be found. then the 2D MPC performance tradeoff surface for assessment is obtained. Finally, a set of simulation experiments prove the effectiveness and feasibility of the proposed method.
this paper presents the development of an adaptive neural network (NN) control method for non-linear quarter-vehicle model which has the characteristics of road disturbance, parameter uncertainties and unknown dead-zo...
详细信息
ISBN:
(纸本)9781538626184
this paper presents the development of an adaptive neural network (NN) control method for non-linear quarter-vehicle model which has the characteristics of road disturbance, parameter uncertainties and unknown dead-zone. Considering the dead-zone slopes as a model uncertainty, an adaptive NN control scheme is developed depending on backstepping technique. In this paper, uncertain non-linear functions in suspension systems are estimated by NNs. then again, the minimal learning parameters can ensure that the computation and the complexity of system are exceedingly reduced. the stability and the signals boundedness of vehicle suspension system are proved. Finally, a given simulation example shows the feasibility of the designed approach.
In the field of microarray data research, it is quite challenging to make classification due to small sample size and the high dimension of data. Moreover, the feature selection is crucial. In this paper, we propose m...
详细信息
ISBN:
(纸本)9781538626184
In the field of microarray data research, it is quite challenging to make classification due to small sample size and the high dimension of data. Moreover, the feature selection is crucial. In this paper, we propose multidimensional mutual information (MMI) feature selection method to select the most informative features for classification. After feature selection using the proposed MMI, Extreme learning Machine (ELM) is used as an efficient classifier. So as to evaluate the performance of the proposed methodology, a typical dataset called Leukemia is selected to carry out a case study. Simulation results demonstrate the effectiveness of the proposed method.
A novel adaptive control strategy based on input/output (I/O) data is proposed in this paper to solve the problem of power management of battery energy storage system (BESS). In the proposed control strategy, a time-v...
详细信息
ISBN:
(纸本)9781538626184
A novel adaptive control strategy based on input/output (I/O) data is proposed in this paper to solve the problem of power management of battery energy storage system (BESS). In the proposed control strategy, a time-varying parameter named pseudo-partial derivative (PPD) parameter utilized in dynamic linearization is estimated by an adaptive observer. Besides, the input saturation problem is considered and a compensation signal is added to consummate the anti-windup control algorithm. Finally, simulation results are presented to validate the effectiveness and performance of the proposed control strategy.
A simple iterative learningcontrol approach is proposed to track specific target points in this work. For a general linear system, a P-type point-to-point ILC and a PD-type point-to-point ILC laws are designed, respe...
详细信息
ISBN:
(纸本)9781538626184
A simple iterative learningcontrol approach is proposed to track specific target points in this work. For a general linear system, a P-type point-to-point ILC and a PD-type point-to-point ILC laws are designed, respectively. the two control laws only use the tracking error at the specified point to update the input signal at the corresponding specified point. the input signal between two consecutive specified points remains the same as the input signal at the previous specified point. the proposed method has the advantages of simple structure and easy application. the convergence analysis and simulation results further confirmed the availability of the method.
In this paper, the event-triggered adaptive neural network-based tracking control problem is investigated for a class of single-input single-output (SISO) nonlinear systems in strict-feedback form. In the considered s...
详细信息
ISBN:
(纸本)9781538626184
In this paper, the event-triggered adaptive neural network-based tracking control problem is investigated for a class of single-input single-output (SISO) nonlinear systems in strict-feedback form. In the considered systems, there exist unknown functions which are approximated by radial basis function neural networks (RBFNNs). Moreover, the output constraint problem is also taken into account, which is solved by exploiting a barrier Lyapunov function. In order to save resources, the event-triggered control method is developed by using the backstepping technique. then, the boundedness of all variables appearing in the systems is obtained, as well as the tracking error stays in a small neighborhood of the origin. In the end, a simulation example is employed to show the effective of the proposed scheme.
暂无评论