Research On Pressure Control Technology Of Low Pressure Casting Liquid Surface Based On Fuzzy Neural Network PID

Low pressure casting is a commonly used technique for producing aluminum alloy castings.Compared with the traditional gravity casting technique,it has the characteristics of high metal liquid utilization rate and dense casting structure.Liquid surface pressurization system is the core of low pressure casting.The response speed and control precision of the liquid surface pressure directly affect the quality of the casting.At present,the liquid pressure control precision of domestic low-pressure casting equipment is not accurate enough,which limits the quality of casting products and casting types.In this paper,the control algorithm of low pressure casting liquid surface pressurization system is studied.The main work and conclusions of this paper are as follows:(1)Starting from the principle of low pressure casting,the characteristics of low pressure casting and the process requirements of low pressure casting are analyzed.The composition of low pressure casting equipment is introduced and the relationship between liquid surface pressurization system and low pressure casting equipment is analyzed.The gas path of the liquid surface pressurization system is designed,and the mathematical model of the main links of the liquid surface pressurization system is established.(2)The control requirements of liquid surface pressurization system are studied.The influence of liquid surface pressure on casting forming and the characteristics of common control algorithms are analyzed.PID,fuzzy PID and fuzzy neural network PID are designed and MATLAB is used for simulation.The performance indexes of three kinds of controllers are compared and analyzed,and the pressure control algorithm of low pressure casting liquid level based on fuzzy neural network PID is determined.(3)The necessity of parameter optimization of fuzzy neural network is analyzed from two aspects: the limitation of BP algorithm and the influence of initial parameters of fuzzy neural network on the control effect.The basic theory of fruit fly optimization algorithm is expounded.The generation mechanism of candidate solution,search radius and search strategy of standard fruit fly optimization algorithm are improved.A hybrid algorithm based on improved fruit fly optimization algorithm and BP algorithm is designed,and the initial parameters of the fuzzy neural network are optimized.In the tracking test of low pressure casting process pressure,the optimized fuzzy neural network PID controller is superior to the conventional PID controller and fuzzy PID controller in the indexes of maximum error and average error.(4)The hardware of the control system is designed according to the control requirement of the low pressure casting equipment.The control system with PLC as the lower computer and touch screen as the upper computer is determined,and the hardware is selected.PLC and HMI are configured,and the PLC program of liquid pressure system and the visual interface of HMI are designed.The fuzzy neural network PID control is realized in s7-1500 by designing and calling function block.The feasibility of the fuzzy neural network PID controller in the industrial field control of low pressure casting liquid surface pressurization is verified.In summary,this paper studied the pressure control technology of low pressure casting liquid surface pressurization and discussed the pressure control algorithm of low pressure casting liquid surface.A fuzzy neural network PID controller is designed,and an improved fruit fly optimization algorithm is proposed to optimize the initial parameters of the controller,which improves the control precision of low pressure casting liquid surface pressure.Finally,the feasibility of fuzzy neural network PID control in low pressure casting liquid surface pressurized industrial field is verified.