Optimal Temperature Control Of Batch Process

With the fierce competition in recent years and the demand for products of many varieties and many specifications and even high value-added, batch and semi-batch mode of production has become crucial for standing on in the competition. However, the majority of domestic automation of batch process is still low, and it is also prevalent with high energy consumption, high material consumption and other issues. Therefore, to enhance competitiveness, reduce energy consumption and material consumption, using optimization and advanced control strategies in batch processes has quickly become the research hotspot. The optimal operation trajectory of batch process is always determined by experienced workers according to rich experiences. Obviously it is time-consuming and laborious. In order to overcome this, a simple and fast mathematical method is given to obtain the optimal trajectory based on process neural network (PNN). And it studied the single-batch tracking and multi-batch repetitive learning to achieve the temperature curve of batch process. The target of both the calculation of the temperature curve and tracking control of the temperature is cutting in computing and control time and improving control effect. At the same time this method provides a new reference for optimal control in the field of actual production.The main tasks include:1,It studied batch processes modeling problem by the perspective of continuous time-varying input. An automatically calculate method of optimal operation curve which used by time function model was developed based on Process Neural Network: using Fourier orthogonal basis function to deal with continuous time-varying input and the unique time aggregation operator of Process Neural Network to train the neural network. When modeling process of producing accelerant for sulfuring rubber with Process Neural Network, it got a better result. Based on this time function model, a quadratic form performance index function is presented to obtain the optimal operating trajectory that is set as set point in actual production.2,In the tracking control system, which set the optimal temperature curve acquired by numerical solution as set point:using neural networks to develop object model and controller model of batch processes and making the reactor temperature tracking the optimal temperature curve based on internal model control theory.3,For the serious non-linear of the temperature of batch process, it proposed a integrated control program which combined iterative learning control and feedback control:using feedback control to regulate tracking deviation based on a single batch tracking and using iterative learning control to finish feed-forward compensation based on the repeatability between batch to batch. When using this method to control temperature of batch process, it got a prefect tracking curve and provided an effective solution for temperature control of batch processes.