Research On Metering Control System Of 800 Type Cement Stabilized Macadam Mixing Equipment

The road structure system of "strong base and thin surface" is widely used in the construction of Chinese highway.As the main material of the highway base,the cement stabilized macadam mixture has a great influence on the road performance.Cement stabilized gravel mixing equipment is widely used in highway construction as special equipment for production of cement stabilized macadam mixture.But at present,the metering accuracy of domestic Cement stabilized gravel mixing equipment is low,which is easy to cause the poor performance of cement stabilized macadam mixture.Aiming at this problem,the thesis takes a domestic 800 type cement stabilized gravel mixing equipment as the research object,and studies the structure and algorithm precision of the metering control system of 800 type cement stabilized gravel mixing equipment.First of all,the scale structure of the metering system is studied.Based on the mechanical analysis of common scale frame structure,the direct load-bearing scale frame structure is proposed,the structure design and feasibility analysis are carried out,and the load cell is analyzed and selected.The network structure of metering control system is improved.Based on PROFINET industrial ethernet technology,PROFINET IO distributed metering control system and control network of cement stabilized gravel mixing equipment are established.According to the performance requirements of metering control system,the hardware of control system is selected.Secondly,the metering control algorithm of cement stabilized gravel mixing equipment is analyzed.Since the PID algorithm of metering control system can not adjust three main control parameters online,therefore,DDPG deep reinforcement learning algorithm is used for online self-tuning of PID control parameters.Based on the mathematical model of metering system,the DDPG-PID controller of cement stabilized gravel mixing equipment is designed,and then the dynamic response and follow-up performance of the controller are verified by MATLAB.The DDPG-PID controller and the conventional PID controller are simulated with or without external disturbance respectively.The simulation results show that: When there is no external disturbance,the time to reach steady state of DDPG-PID controller is shortened by 58.41%,the maximum overshoot is reduced by 16.97%,and the maximum amplitude error is reduced by 12.82%;When there is external disturbance,the cumulative output error of DDPG-PID controller is reduced by 4.91%,the maximum overshoot is reduced by 19.28%,the maximum amplitude error is reduced by 15.2%,and the result is no overshoot and oscillation.The dynamic response and follow performance of DDPG-PID controller are better than PID controller.Finally,the simulation results show that DDPG-PID control performance can be further improved in the face of complex external environment.Therefore,the DDPG algorithm is improved to Twin Delay DDPG algorithm,and the neural network structure and reward function of the original model are optimized,and then,twin-delay DDPG-PID controller and DDPG-PID controller are compared and simulated in the same environment.The results show that: When there is no external disturbance,the time for the Twin Delay DDPG-PID controller to reach the steady state is shortened by 48.97%,the output cumulative error is reduced by1.95%,and the maximum amplitude error is reduced by 1.18%;When there are external disturbances,the cumulative output error of the improved twin-delay DDPG-PID controller is reduced by 1.05%,the maximum delay time is reduced by 0.15 seconds,and the maximum amplitude error is reduced by 3.64%.In summary,the Twin Delay DDPG-PID controller has faster learning speed and control accuracy in complex environments,and the Twin Delay DDPG-PID controller is more suitable for the needs of cement stabilized gravel mixing equipment than the DDPG-PID controller.