Adaptive prediction control method based on Hammerstein model

Abstract

The invention discloses an adaptive prediction control method based on the Hammerstein model. The adaptive prediction control method based on the Hammerstein model specifically comprises a first step that a data collection channel collects technological parameter values in the process of intermittent chemical production in real time and conducts data pre-processing, a second step that the processed data are transmitted to a neural network identifier, modeling is conducted by the neural network identifier, and the established model is revised in a simulating mode, a third step that a mixed adaptive prediction controller reads model parameters, and generates a control parameter to control motions of an executing mechanism, and a fourth step that a control algorithm is achieved. According to the adaptive prediction control method based on the Hammerstein model, a control system is designed through the control method based on the Hammerstein model, a nonlinear inversion element is introduced, prediction strategy is designed aiming at a linear element, and then the actual controlled quantity is determined through solving of a nonlinear algebraic equation set. Therefore, the calculated quantity of on-line optimization is reduced, and precise effective control over the intermittent chemical process is achieved.

Description

technical field [0001] The invention relates to an advanced control method for a chemical production process, in particular to an intelligent self-adaptive predictive control method for a batch chemical production process based on a Hammerstein model, belonging to the field of industrial control. Background technique [0002] Batch process refers to the process of processing a limited amount of materials in one or more equipment according to the prescribed processing sequence to obtain a limited amount of products. Due to the advantages of short process, simple equipment, low investment, quick results, and easy replacement of varieties, intermittent production is widely used in fine chemical, pesticide chemical and biomedical production, and its application prospects are very promising. In recent years, the scale of fine chemical production process has continued to expand, the complexity of the process has increased, the requirements for product quality have increased, and t...

AI Technical Summary

Problems solved by technology

[0004] Due to the time-varying and non-linear characteristics of the intermittent process, designing a controller based on the linear system theory cannot meet the control requirements, because this method is simple in calculation and strong in real-time, but its design structure is complex, the amount of calculation is large, and its stability is difficult to obtain. ensure

Due to the influence of various uncertain factors in the actual batch chemical process system, it is difficult for the designer to obtain an accurate model description of the thermal system object in a large working range

The traditional mechanism model method based on the material-energy balance equation is difficult to meet the real-time requirements of control optimization due to the complexity of the modeling process, model structure and calculation method. Therefore, the calculation results inevitably have considerable deviations from the actual operating data, which reduces its practicability; Additional costs may even interfere with normal safe production

Many process characteristic parameters are difficult to measure, there are many operating constraints, there are many disturbances, the process is irreversible, and it is difficult to take remedial measures, which makes the control of batch process very difficult. Therefore, it is researched and promoted in the batch production process. It is necessary and urgent to apply various effective advanced control strategies

Images