Hybrid identification method and system for thermal process in thermal power plant

Abstract

The invention relates to a hybrid identification method and system for a thermal process of a thermal power plant, which establishes an identification sequence corresponding to preset load operating point points, and correspondingly classifies collected real-time data into the identification sequence. According to the identification sequence, use the least square method to identify the initial values ​​of the time constant, delay time and gain of the thermal process object, and then use the linear method to fit the time constant, delay time and gain of the thermal process object to obtain the corresponding function expression . Finally, the Elman neural network is used to correct the gain of the thermal process object, and a first-order inertia plus pure delay model is obtained. When identifying the parameters of the thermal controlled object model, the neural network can be used to simulate the input and output relationship of the actual system by using the arbitrary approximation ability of the neural network to any nonlinear mapping, and continuously adjust the identification accuracy according to the input and output data to ensure the accuracy of the gain identification. Accuracy improves the thermal process safety of thermal power plants.

Description

technical field [0001] The invention relates to the technical field of thermal process in thermal power plants, in particular to a method and system for identifying a mixture of thermal process in thermal power plants. Background technique [0002] The dynamic characteristics of the controlled object in the thermal process of a thermal power plant refer to the dynamic relationship between the output and input of the control object. The identification of the transfer function model of the thermal object has always been an important research content in the field of system identification. The dynamic characteristics of the thermal system, the control scheme design of the control system and the parameter setting of the regulator are of great significance. [0003] Model identification includes two parts: model structure assumption and model parameter identification. Because there are many factors affecting the dynamic characteristics of the controlled object, it is difficult to...

AI Technical Summary

Problems solved by technology

Since there are many factors affecting the dynamic characteristics of the controlled object, it is difficult to describe them accurately

Due to the nonlinear characteristics of the thermal system, the gain of many controlled objects is not constant. Compared with the time constant T and delay time τ, the change of the gain will have a more important impact on the control quality of the system, so the gain Identification is of great significance in the whole process of model determination, if it cannot be accurately identified, it will lead to low safety of thermal process in thermal power plants

Images