Research On The Intelligent Control Method Of The Airborne Evaporative Cycle Refrigeration System

There are significant differences between the system control requirements in the ground environment and that in the airborne environment.The research on the control method of the airborne evaporative cycle refrigeration system is helpful to improve its application effect in the airborne environment.In this paper,based on the analysis of the characteristics of the airborne evaporative cycle refrigeration system,the intelligent control method of the system is studied.(1)The simulation model of the air-borne evaporative cycle refrigeration system is established.The coupling response of the parameters of the internal components and the external disturbance to the system parameters and the law of the variation of the refrigeration coefficient are studied numerically,which provides the theoretical basis for the study of the control method under the multi disturbance of the system.(2)Taking the thermal load of electronic equipment and the state change of stamping air as disturbance,the control strategies of compressor controlling the temperature of cold plate outlet and expansion valve controlling the superheat of evaporator outlet are adopted.The control effects of PID and single neuron PID control methods are compared and analyzed.The results show that the two control methods have the same effect when the stamping air state changes.However,when the thermal load changes suddenly,the superheat overshoot of the evaporator outlet under the single neuron PID control method is more than twice as low as that under the PID control method.The regulating time of the two control methods is 600s～700s.The control effect is still difficult to meet the requirements of airborne environmental control fast and stable control.(3)A decoupling control method based on BP neural network is proposed considering the coupling effect between the two objective control and the nonlinearity of the system.The simulation results show that the overshoot of the control target under the BP neural network decoupling control method is 20% lower than that of the single neuron PID control method,and the regulating time is reduced by more than 200 s.The control effect is improved effectively.The intelligent control method of evaporative cycle refrigeration system studied in this paper can provide technical reference for the optimal design of airborne environmental control system.