电子科技2025，Vol.38Issue(7)：82-88,7.DOI:10.16180/j.cnki.issn1007-7820.2025.07.011
基于神经网络逆模型的制冷系统解耦控制
Research on Decoupling Control of Refrigeration System Based on Neural Network Inverse Model
摘要
Abstract
In view of the nonlinearity and multi-variable coupling of compression refrigeration system,the in-verse system control method of α-order neural network is used to decouple it into two first-order subsystems:super-heat and evaporation temperature.On this basis,the linear closed-loop controller PID(Proportional Integration Dif-ferentiation)is added to realize the high performance decoupling control of the system.The results show that the pro-posed method is simple in structure and easy to implement,and effectively avoids the shortcomings of the traditional control method which depends on the accuracy of the system model.The step response time for both superheat and evaporation temperature is reduced by 234 s and 360 s,respectively.The overshoot of the evaporation temperature and superheat under step perturbation is decreased by 9.4%and 13.3%,respectively,demonstrating that the proposed method displays better dynamic performance and stability.关键词
制冷系统/蒸发器/过热度/蒸发温度/逆系统/RBF神经网络/PID/解耦Key words
refrigeration system/evaporator/superheat/evaporation temperature/inverse system/RBF neural net-work/PID/decoupling分类
信息技术与安全科学引用本文复制引用
王俊超,丁绪东,杨远星,刘雨婷,杨玉萍..基于神经网络逆模型的制冷系统解耦控制[J].电子科技,2025,38(7):82-88,7.基金项目
山东省重大科技创新工程项目(2019JZZY020812) （2019JZZY020812）
山东省自然科学基金面(ZR2020MF070)Major Science and Technology Innovation Project of Shandong(2019JZZY020812) （ZR2020MF070）
Natural Science Foundation of Shandong(ZR2020MF070) （ZR2020MF070）
