An improved elman neural network controller based on quasi‐ARX neural network for nonlinear systems

Abstract

An improved Elman neural network (IENN) controller with particle swarm optimization (PSO) is presented for nonlinear systems. The proposed controller is composed of a quasi-ARX neural network (QARXNN) prediction model and a switching mechanism. The switching mechanism is used to guarantee that the prediction model works well. The primary controller is designed based on IENN using the backpropagation (BP) learning algorithm with PSO. PSO is used to adjust the learning rates in the BP process for improving the learning capability. The adaptive learning rates of the controller are investigated via the Lyapunov stability theorem. The proposed controller performance is verified through numerical simulation. The method is compared with the fuzzy switching and 0/1 switching methods to show its effectiveness in terms of stability, accuracy, and robustness.

A Prediction Model of Forest Preliminary Precision Fertilization Based on Improved GRA-PSO-BP Neural Network

“…e structure of the basic BPNN consists of the input layer, the hidden layer, and the output layer, as shown in Figure 1. e BPNN algorithm consists of two parts: the forward transfer of the input data and the back propagation of the error between the output data and the expected data [26,27]. One part is the input data through the input layer to the hidden layer, the hidden layer to the output layer.…”

A Self-Organizing Quasi-Linear ARX RBFN Model for Nonlinear Dynamical Systems Identification

“…• Method 1: Adaptive fuzzy switching controller based on SOQARX-RBFN model [8] • Method 2: An improved Elman neural network (IENN) controller based on SOQARX-RBFN [18] Since the quasi-linear ARX RBFN prediction model described by Eq. (17) is linear in the input variable u(t), it is easy to design the control system.…”

“…Please refer Refs. [8] and [18] for the details of control systems based on the quasi-linear ARX RBFN models. In Fig.…”

Livestock Disease Prevention and Control Management System in Financial Venture Capital

“…The information of the hidden layer is stored in the continuous layer with a certain delay, which improves the memory of the information of the last time. In this way, the internal feedback is achieved and the intermediate process is reduced, making the network system better than the usual static neural network.Scholar Chen X based on fog is sensitive to temperature change the fact that the temperature through a set of experimental results and error analysis, this paper proposes a new error processing technology, the method includes both denoising and modeling and compensation, after denoising part, this paper proposes a dynamic modified Elman neural network (ENN) modeling method, modeling and compensation results show that this model can effectively reduce the temperature changes and compensation caused by drift [8].Scholar W Xiang by Elman neural network model is established to predict the change of aquaculture water body NH3 -N, record into the water day food, dissolved oxygen in water, water temperature, water temperature, turbidity and precipitation index as input variables, select the pond NH3 -N content as the output variable, the results show that the model can well simulate the change trend of aquaculture water body NH3 -N, based on Elman neural network prediction model of aquaculture water body nonlinear dynamic changes of NH3 -N content has a strong ability of description,It has good adaptability and accuracy in practical application [9].In order to achieve the optimal operation of power system, power generation must meet the demand of power load [10], the scholar L Yang proposed an improved prediction engine based on Elman neural network (IENN) to predict the load signal, and used the intelligent algorithm to optimize the weights of the prediction engine, and obtained a better prediction result [11].Scholars I Sutrisno in view of the nonlinear system, puts forward an improved particle swarm optimization (IENN) Elman neural network controller, the controller by the ARX (QARXNN) neural network prediction model and a switching mechanism, the switching mechanism, ensure the good running of forecasting model, design the main controller based on IENN, adopt the BP learning algorithm with PSO, the algorithm by adjusting the learning rate to improve learning ability in the process of BP, the performance of the controller is verified by numerical simulation.By comparing with fuzzy switching method and 0/1 switching method, the effectiveness of this method in terms of stability, accuracy and robustness is proved [12].Scholar CH Tsai proposed an Elman neural network controller based on the improved differential evolution algorithm to control the squirrel cage induction power generation system for grid-connected wind power generation, and the experimental results verified the feasibility and effectiveness of the proposed system in grid-connected wind power applications [13].Scholars XL Zhu, a feed-forward neural network based adaptive vector quantization (AVQ) network clustering algorithm, is used for machine tool thermal error compensation in the choice of temperature measurement point, the method used in machining center, measurement point from 18 reduced to a 3, based on the outputinput feedback Elman neural network model, established the thermal error and the relationship between the critical temperature measuring points, the results show that this method can effectively eliminate the coupling between the temperature measuring point, improve the accuracy of thermal error model and robustness [14]…”