Study On Modeling And Control Of Omethoate Synthesis Process Based On PSO Algorithm

Agriculture is the foundation of national economy, it plays an important role in the country's development and social stability. Pesticide is an important factor in increasing agricultural production. Omethoate is one of the pesticide which is widely used in agriculture production. The merits of its quality has great impact on agricultural production. Quality of the Omethoate has been directly impacted by the better or worse control of synthesis reactor's temperature. So establishing a better model and implementing precision control for its synthetic process will help to improve production efficiency and product quality, reduce production costs, result in significant economic benefits.Omethoate synthesis is a typical batch process, with its temperature plant having characteristics of multiple variables, nonlinear, time-varying, time-delay and so on. It is difficult to establish its model to use conventional methods. It can not be achieved satisfactory control effect to use traditional control methods. In recent years, intelligent theory methods included fuzzy logic, neural networks, intelligent evolutionary algorithms, and the integrated approaches of them, provide an effective way for modeling and controlling of such complex objects.In this paper, the neural network, particle swarm optimization, fuzzy logic are combined to model and control the temperature object of Omethoate's synthesis process. Firstly, in order to against the shortcomings of standard PSO algorithm, dynamic inertia adjustment strategy which is based on the improved algorithm of the speed item discarded is adopted. Its performance is tested with the benchmark function, results show that the improved PSO algorithm have features of fast convergence and high precision search. Secondly, characteristics of the synthetic Omethoate is analyzed. In view of the good performance of improved PSO algorithm, which will be combined with the recurrent BP network to identify the temperature object.The PSO-BP network model is established for temperature object, and compared with the static BP model.The simulation results show that the model combines the global optimization ability of particle swarm and local search advantage of BP algorithm better reflects the actual system dynamic performance. It has features of high accuracy and good performance. Thirdly, FNN controller is designed for Omethoate synthesis process based on the theory of fuzzy neural network. Fuzzy neural network transforms the parameter search optimization problem of fuzzy control rules and membership functions into the parameter optimization problem of neural network. According to the method of PSO training neural network, fuzzy neural network parameters is optimized and trained by the improved PSO algorithm using the data collected from the actual production process. Finally, the controller and the model of temperature identification is combined together to constitute temperature control system of Omethoate synthetic reaction process. Simulation results show that this control scheme can achieve satisfactory control performance.