Research On Control System Of Batching Scale And Its High Precision Dynamic Weighing Technology

The loss-in-weight dosage control system is a kind of dosage system that uses dynamic weighing for metering,which has the advantages of high accuracy and easy maintenance,and is especially suitable for modern industrial dosage production.However,most of the dosage control systems do not have a good method to deal with the non-linear errors generated in the dynamic weighing process,but they are processed by a single filtering method,which is not effective when the errors are large and the environment is variable.In order to realize the high accuracy and fast response of the dosage control system,this topic proposes a Kalman filterPSO-GRU neural network combined model adaptive filtering algorithm,which can effectively eliminate the nonlinear error of dynamic weighing process and improve the accuracy of dynamic weighing process.And a two-stage fuzzy PID control technique is proposed for the control process of the dosage system,and it can realize the fast response and stability of the dosage control process.The main work of this paper is as follows.(1)Analyze the principle of batching scale and its realization goal,and design the lossin-weight weighing structure.Design the overall control framework of the batching control system according to the connection between the control system and the field equipment and instruments.According to the functional requirements of the control system,the corresponding dosage control strategy is designed.(2)In order to eliminate the nonlinearity error generated by mechanical vibration and motor rotation in the dynamic weighing process,to get the real weighing data sequence and to improve the dosage accuracy,an adaptive filter combining Kalman filter and GRU neural network algorithm is designed to eliminate the multivariable nonlinear errors.In this study,the error at no-load is analyzed,and the application of Kalman filter algorithm and GRU neural network algorithm in error filtering is studied separately,and the optimization of GRU neural network is also analyzed,and a combined Kalman filter-PSO-GRU neural network filtering model is proposed on its basis.The combined filtering model is applied to the dynamic weighing data processing,and it achieves better results than the single model,which provides a new method to achieve high accuracy in dynamic weighing.(3)In the dynamic weighing environment,the conventional control algorithm cannot meet the requirements of the changing environment due to the variable environment.In order to take into account the stability of the dosage speed and the accuracy of the dosage,the dosage control process is analyzed and studied,and a two-stage dosage control system based on the fuzzy control method optimized PID control is proposed,and the mathematical model of the dosage process is studied to determine the initialization curve of the dosage control speed in one stage and the design of the fuzzy PID controller in two stages.Finally,the experiments show that the control method is able to achieve stable and fast response of batching speed,and has good adaptability and anti-interference performance.(4)Finally,according to the above key technical points,the hardware is selected and designed,the batching control platform is built,and the designed control scheme is realized,and the proposed Kalman filter-PSO-GRU neural network combination model and two-stage fuzzy PID control algorithm are experimented and simulated,and finally the experimental verification results show that the proposed scheme can improve the accuracy in the dynamic weighing process,and at the same time realize the rapid response and stability of speed control,and the error accuracy of the overall batching does not exceed 0.5%,in line with technical specifications.