Novel optimal compensation control strategy for batch process based on a process migration model

Abstract

The invention relates to a novel optimal compensation control strategy for a batch process based on a process migration model. Two A and B similar process data are obtained and a process migration model is established; an optimization problem based on the process migration model is constructed and an initial optimal solution is calculated; a query data is set and similarity between the query dataand historical data is calculated; input and output deviation data of the selected data in the similar process are obtained to obtain an optimal compensation value; a predicted performance index deviation is calculated; a modeling data set of a new batch process is updated; a latest prediction error of a Kth batch and a first confidence interval and a second confidence interval are calculated; whether to carry out data rejection and process migration is determined based on a prediction error; and the radius of a trust domain is updated. With the process migration model, a problem that modelingcan not be preformed because of insufficient data in the new batch process is solved effectively; with the optimal compensation control strategy, the second-best solution of the new batch process optimization problem is compensated; and the model is updated in real time based on online data to compensate mismatching of the model object.

Description

technical field [0001] The invention belongs to the field of optimal control of batch processes in industrial production processes, and in particular relates to an optimal compensation control strategy for batch processes based on a process migration model. Background technique [0002] Batch process refers to the production process in which raw materials are processed into specific products according to the specified process flow within a specified time. It is sometimes called batch process, which has the characteristics of parameters changing with time and working status unstable. In modern industrial processing and production, batch process is widely used in the production of food, medicine, chemical and other products. It has many advantages such as simple production process, high flexibility, and low investment cost. With the rapid development of society, people's requirements for product quality, production cost, environmental protection and other aspects are constantl...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems existing in the above-mentioned prior art, the present invention provides a batch process optimal compensation control strategy based on a process migration model, which can effectively solve the problem of insufficient modeling data for the new batch process and solve the problem of optimization of energy efficiency The NCO mismatch problem of the model will help improve the effect of final product optimization and greatly improve the comprehensive economic benefits of the enterprise

Images