Data pretreatment method for steam water leakage diagnosis of heat recovery boiler

Abstract

The invention discloses a data pretreatment method for a steam water leakage diagnosis of a heat recovery boiler. The data pretreatment method comprises the following steps that: (1) the operation data are collected and recorded; (2) the abnormal data among the operating data are removed; (3) the operating data in steady operating condition are screened; (4) steady-state segments are conducted with the mean value polymerization on the basis of the abnormal data removal and the steady-state screening. The data pretreatment method for the steam water leakage diagnosis of the heat recovery boiler is put forward according to the characteristics of the operation data of the gas-steam combined cycle unit heat recovery boiler. The data pretreatment method has the advantages of high reliability, improved operating data quality and improved fault diagnosis accuracy rate.

Description

technical field [0001] The invention belongs to the field of steam-water leakage diagnosis of power station units, and more specifically relates to a data preprocessing method for steam-water leakage diagnosis of waste heat boilers. [0002] technical background [0003] Coal-fired units with a large proportion in my country's power structure have poor peak-shaving capabilities. In contrast, gas-steam combined cycle units have the characteristics of fast start-up and stop, flexible operation, good peak-shaving capabilities, and are suitable for grid load peak-shaving. This situation makes my country's gas-steam combined cycle units participate in deep peak shaving of the power grid for a long time, and the load change rate is large during the operation of the units, making it difficult to maintain stable operating conditions for a long time. The long-term operation of the unit in this mode is prone to leakage of the waste heat boiler tube, which has a serious impact on the sa...

AI Technical Summary

Problems solved by technology

[0004] Since the gas-steam combined cycle unit participates in the deep peak shaving of the power grid for a long time during operation, its operating data changes rapidly, including a large amount of abnormal, unstable, and disturbed data

These data cannot accurately reflect the current operating state of the unit. When using these data for fault diagnosis, the accuracy of fault diagnosis cannot be guaranteed, which will interfere with the operator's judgment on the actual operating state of the unit.

However, the existing fault diagnosis methods, especially for gas-steam combined cycle unit waste heat boiler steam-water leakage fault diagnosis methods, lack of attention to the preprocessing of operating data, and the accuracy of leakage diagnosis is low

Images