Predictive control of microwave hot-air coupled drying model based on GWO-BP neural network

被引：16

作者：

Bai, Haoran ^{[1
]}

Chu, Ziyi ^{[1
]}

Wang, Dongwei ^{[1
]}

Bao, Yan ^{[2
]}

Qin, Liyang ^{[2
]}

Zheng, Yuhui ^{[1
]}

Li, Fengmei ^{[1
]}

机构：

[1] Qingdao Agr Univ, Coll Mech & Elect Engn, 700 Great Wall Rd, Qingdao 266109, Peoples R China

[2] China Construct Zhongxin Construct Engn Co Ltd, Chengdu, Peoples R China

来源：

DRYING TECHNOLOGY | 2023年 / 41卷 / 07期

关键词：

Grey wolf optimization algorithm; BP neural network; predictive Control; projection conjugate gradient method; fruit and vegetable drying;

D O I：

10.1080/07373937.2022.2124262

中图分类号：

TQ [化学工业];

学科分类号：

0817 ;

摘要：

A GB-MPC control algorithm (GWO-BP-MPC) was proposed to solve the problem of precise temperature control of fruit and vegetable coupling drying devices. Firstly, the BP (Back Propagation) neural network was improved using the Grey Wolf Optimizer (GWO) algorithm to increase the relevance and accuracy of the prediction model. By means of an improved neural network, we developed a high-accuracy predictive model for temperature control of drying units. Secondly, the projection conjugate gradient method was proposed for nonlinear optimization of the control system to improve the solving speed and accuracy of the optimal solution. The GB-MPC control algorithm was compared with the PID controller. The experimental results shown that the convergence speed of GB-MPC control was faster, the time took to reach a steady state in a single stage was shortened by 47 seconds compared with PID control. In the control process, the temperature change range of the GB-MPC control algorithm was smaller and there was no overshoot problem, which gave a better control effect than PID.

引用

页码：1148 / 1158

页数：11

共 21 条

[11] How effective is the Grey Wolf optimizer in training multi-layer perceptrons [J].

Mirjalili, Seyedali .

APPLIED INTELLIGENCE, 2015, 43 (01) :150-161

[12] Grey Wolf Optimizer [J].

Mirjalili, Seyedali ;

Mirjalili, Seyed Mohammad ;

Lewis, Andrew .

ADVANCES IN ENGINEERING SOFTWARE, 2014, 69 :46-61

[13]

More J J., 1982, Newton's method (No. ANL-82-8), DOI DOI 10.2172/5326201

[14] Conjugate gradient method [J].

Nazareth, J. L. .

WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL STATISTICS, 2009, 1 (03) :348-353

[15] Adaptive lasso echo state network based on modified Bayesian information criterion for nonlinear system modeling [J].

Qiao, Junfei ;

Wang, Lei ;

Yang, Cuili .

NEURAL COMPUTING & APPLICATIONS, 2019, 31 (10) :6163-6177

[16]

Raeihagh H, 2020, J LOSS PREVENT PROC, V68, DOI [10.1016/j.jip.2020.140438, 10.1016/j.jlp.2020.104238]

[17] Real-time evaluation of artificial neural network-developed model of banana slice kinetics in microwave-hot air dryer [J].

Sabzevari, Masumeh ;

Behroozi-Khazaei, Nasser ;

Darvishi, Hosain .

JOURNAL OF FOOD PROCESS ENGINEERING, 2021, 44 (09)

[18] Design and Implementation of Neural Networks Neurons with RadBas, LogSig, and TanSig Activation Functions on FPGA [J].

Sahin, I. ;

Koyuncu, I. .

ELEKTRONIKA IR ELEKTROTECHNIKA, 2012, 120 (04) :51-54

[19] Model predictive control and neural network predictive control of TAME reactive distillation column [J].

Sharma, Neha ;

Singh, Kailash .

CHEMICAL ENGINEERING AND PROCESSING-PROCESS INTENSIFICATION, 2012, 59 :9-21

[20]

Su M., 2010, Journal of Nonlinear Analysis and Optimization, V1, P35

← 1 2 3 →