Method and system for diagnosing leakage fault of high-pressure heater based on neural network and thermodynamic modeling

Abstract

The invention provides a method and system for diagnosing leakage faults of high-pressure heaters based on neural network and thermodynamic modeling. The present invention comprises the following steps: constructing a mathematical model about the opening of each drain regulating valve of each high pressure heater and the fluid flow through the drain regulating valve based on collected historical data; extracting the characteristic vector parameters of the high pressure heater system during normal operation, constructing Training set and test set; model training is carried out on the training set data through neural network algorithm, and the maximum error range of the comparison chart between the opening degree of the heater drain regulating valve and the actual drain regulating valve opening is recorded; the real-time calculation of each high-pressure steam extraction flow rate and The first safety diagnosis is completed by the difference between the two drain flows; the second safety diagnosis is completed by comparing the difference between the opening degree of the high pressure heater drain valve calculated by the neural network model and the actual opening degree, and the first safety diagnosis is automatically switched and a second safety diagnostic. The invention has higher accuracy and can greatly improve the real-time performance of leakage diagnosis of the high pressure heater.

Description

technical field [0001] The invention relates to the field of power system fault diagnosis, in particular to a method and system for diagnosing leakage faults of high-pressure heaters based on neural network and thermodynamic modeling. Background technique [0002] The high-pressure heater is a device that uses the partial extraction of the steam turbine to heat the feed water. As a heat conversion device, it is mainly used in the heat recovery system of large thermal power units. The high-pressure heater is composed of a shell and a piping system. A steam condensing section is set in the upper part of the inner cavity of the shell, a drain cooling section is set in the lower part, and a water supply inlet and a water supply outlet are set at the top of the inlet and outlet pipes. When the superheated steam enters the casing from the inlet, it can heat the feed water in the upper main coil, and after the steam condenses into water, the condensed hot water can heat part of th...

AI Technical Summary

Problems solved by technology

[0003] The traditional high-pressure heater leakage diagnosis needs to be judged by the operation experience of the centralized control personnel, and the unit has no high-pressure heater leakage alarm, which makes the high-pressure heater leakage diagnosis accuracy and real-time performance poor

In the existing technology, some flow measurement points are added to assist judgment. The cost of installing measurement points is relatively high, and the maintenance cost of the measurement points in the later period also increases accordingly. After the pipeline is drilled and the flow measurement points are installed, the original structure of the pipeline is destroyed and Increased throttling loss, poor economy and safety

[0004] Publication No. 109459195 A "Method and System for Judging Leakage of High-Pressure Heater System" discloses a method based on heat balance principle, material conservation principle and multiple historical measuring point data during normal operation of the high-pressure heater system to calculate the The method and system for determining the leakage criterion based on the deviation values ​​of multiple historical drainage volumes at the outlet of the high-pressure heater system, and then according to the preset algorithm, compared with the existing method that only judges by disease control personnel, has greatly improved progress, but the accuracy of this method depends entirely on the accuracy of the hydrophobic flow measurement points, and there is only one criterion. During the load lifting period in real projects, the water flow fluctuates greatly. This method needs to widen the time domain, and the longer the time domain wide, the more laggy the diagnostic response

Facing the fault with high leakage and large leakage, the diagnosis information cannot be given in time, and the fault diagnosis response speed of this method is slow

Images