CN-122020164-A - Power generation equipment fault analysis method and system

CN122020164ACN 122020164 ACN122020164 ACN 122020164ACN-122020164-A

Abstract

The invention discloses a power generation equipment fault analysis method and system. The method comprises the steps of collecting current operation data of power generation equipment in real time, preprocessing the current operation data to obtain a current input vector, obtaining a predicted value of a target measurement point by utilizing an intelligent prediction model obtained by training based on a historical normal operation sample data set based on the current input vector, comparing the predicted value with a corresponding actual measurement value to obtain a residual value, comparing the residual value with a residual threshold value obtained by statistics based on the historical normal operation sample data set, and judging that the operation state of the power generation equipment is abnormal and triggering early warning when the residual value exceeds the residual threshold value. The invention solves the technical problem of inaccurate monitoring in the operation monitoring of the power generation equipment in the prior art.

Inventors

LI ENPENG
AI XIN
JI HAILONG
HE HAOMIN
YAO ZHILIN
LI RUYI
ZHANG JUN

Assignees

深能智慧能源科技有限公司
深能(河源)电力有限公司

Dates

Publication Date: 20260512
Application Date: 20260122

Claims (10)

1. The power generation equipment fault early warning analysis method is characterized by comprising the following steps of: collecting current operation data of power generation equipment in real time, and preprocessing the current operation data to obtain a current input vector; Based on the current input vector, an intelligent prediction model obtained by training based on a historical normal operation sample data set is utilized to obtain a predicted value of a target measuring point, and the predicted value is compared with a corresponding actual measured value to obtain a residual value; and comparing the residual value with a residual threshold value obtained based on statistics of a historical normal operation sample data set, and judging that the operation state of the power generation equipment is abnormal and triggering early warning when the residual value exceeds the residual threshold value.
2. The method of claim 1, wherein the intelligent predictive model is derived by: collecting historical operation data of the power generation equipment, and performing data cleaning, normal working condition screening and normalization on the historical operation data to obtain a historical normal operation sample data set, wherein the historical operation data comprises measuring point parameters of at least one of deaerator water level, pressure, temperature, condensate flow, heating steam flow and valve opening; And determining a mapping relation between an input vector formed by a plurality of associated measuring point parameters and an output formed by target measuring point parameters by adopting a support vector machine regression algorithm based on the historical normal operation sample data set so as to obtain the intelligent prediction model describing the internal association relation of the measuring point parameters.
3. The method of claim 2, wherein determining a mapping relationship between an input vector of a plurality of associated site parameters and an output of a target site parameter using a support vector machine regression algorithm comprises: selecting a plurality of associated measuring point parameters with physical association relation with a target measuring point parameter from the plurality of measuring point parameters, constructing an input vector composed of the plurality of associated measuring point parameters, and taking the target measuring point parameter as output; and constructing a kernel function by using a Gaussian radial basis kernel function, and modeling the relation between the input vector and the target measuring point parameter by using a support vector machine regression algorithm based on the kernel function, the Lagrange multiplier and the bias term to obtain the mapping relation.
4. The method of claim 3, wherein selecting the plurality of associated site parameters from the plurality of site parameters that have a physical association with the target site parameter comprises selecting the plurality of associated site parameters from the plurality of site parameters that have a physical association with the target site parameter based on at least one of a mass balance relationship between a deaerator water level and a condensate flow rate, a feedwater pump outlet flow rate, a thermodynamic relationship between a deaerator pressure and a heating steam flow rate, and a valve opening.
5. The method according to claim 2, wherein after determining that the operation state of the power generation equipment is abnormal and triggering the early warning, the method further comprises comprehensively analyzing residual variation conditions of a plurality of input measuring points with process association relation with the target measuring point, and determining an abnormal source measuring point and corresponding equipment or process links thereof to obtain a fault auxiliary diagnosis result.
6. The method of claim 2, wherein performing data cleaning, normal condition screening, and normalization on the historical operating data to obtain a historical normal operating sample dataset comprises: removing data exceeding a physical range, data with continuous identical values exceeding a preset time threshold value and data with change rate exceeding a physical limit from the historical operation data to obtain cleaned data; Screening data in a rated load stable operation interval from the cleaned data based on unit load conditions to obtain normal working condition data; And processing the normal working condition data by adopting a normalization method to obtain the historical normal operation sample data set.
7. A power generation equipment fault early warning analysis system, characterized by comprising: the preprocessing module is configured to acquire current operation data of the power generation equipment in real time, and preprocess the current operation data to obtain a current input vector; The residual determination module is configured to obtain a predicted value of a target measuring point by utilizing an intelligent prediction model obtained by training based on a historical normal operation sample data set based on the current input vector, and compare the predicted value with a corresponding actual measured value to obtain a residual value; And the early warning analysis module is configured to compare the residual value with a residual threshold value obtained based on statistics of a historical normal operation sample data set, and when the residual value exceeds the residual threshold value, the early warning analysis module judges that the operation state of the power generation equipment is abnormal and triggers early warning.
8. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored program, wherein the program when run controls a device in which the computer readable storage medium is located to perform the method according to any one of claims 1 to 6.
9. A computer device is characterized by comprising a memory and a processor, The memory stores a computer program; the processor being operative to execute a computer program stored in the memory, the computer program when run causes the processor to perform the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

Description

Power generation equipment fault analysis method and system Technical Field The invention relates to the field of artificial intelligence, in particular to a method and a system for analyzing faults of power generation equipment. Background At present, an automatic control system (DCS, PLC and the like) of a thermal power plant has a constant value alarm function and can generate an alarm when a monitored parameter reaches a set alarm value. The partial power generation equipment fault prediction software system can perform a conventional basic early warning function to fight a time window for the early intervention of power plant operators. However, these systems have the common limitations of adopting a fixed threshold alarm mechanism, being unable to adapt to different operation conditions, lacking in correlation analysis among parameters, being unable to identify gradual failure trend, having too short alarm delay or advance time, usually being triggered when failure has occurred or is approaching a critical state, and being unable to provide intelligent diagnosis analysis of failure cause. Therefore, the power enterprises are urgent to establish a more efficient intelligent monitoring and early warning technology, and realize prospective early warning on faults of the deaerators by deeply mining the existing data resources, so that safe and reliable operation of the equipment is ensured. Disclosure of Invention The embodiment of the invention provides a power generation equipment fault analysis method and system, which at least solve the technical problem of inaccurate monitoring in the operation monitoring of power generation equipment in the prior art. According to one aspect of the embodiment of the invention, the power generation equipment fault early warning analysis method comprises the steps of collecting current operation data of the power generation equipment in real time, preprocessing the current operation data to obtain a current input vector, obtaining a predicted value of a target measuring point by utilizing an intelligent prediction model obtained by training based on a historical normal operation sample data set based on the current input vector, comparing the predicted value with a corresponding practical measured value to obtain a residual value, comparing the residual value with a residual threshold value obtained by statistics based on the historical normal operation sample data set, and judging that the operation state of the power generation equipment is abnormal and triggering early warning when the residual value exceeds the residual threshold value. According to another aspect of the embodiment of the invention, a power generation equipment fault early warning analysis system is provided, which comprises a preprocessing module, a residual determination module and an early warning analysis module, wherein the preprocessing module is used for acquiring current operation data of power generation equipment in real time, preprocessing the current operation data to obtain a current input vector, the residual determination module is used for obtaining a predicted value of a target measurement point by utilizing an intelligent prediction model obtained by training based on a historical normal operation sample data set based on the current input vector and comparing the predicted value with a corresponding actual measurement value to obtain a residual value, and the early warning analysis module is used for comparing the residual value with a residual threshold value obtained by statistics based on the historical normal operation sample data set and judging that the operation state of the power generation equipment is abnormal and triggering early warning when the residual value exceeds the residual threshold value. According to the method, the device and the system, current operation data of the power generation equipment are collected in real time, the current operation data are preprocessed to obtain a current input vector, based on the current input vector, an intelligent prediction model obtained through training based on a historical normal operation sample data set is utilized to obtain a predicted value of a target measurement point, the predicted value is compared with a corresponding actual measurement value to obtain a residual value, the residual value is compared with a residual threshold value obtained through statistics based on the historical normal operation sample data set, and when the residual value exceeds the residual threshold value, the abnormal operation state of the power generation equipment is judged and early warning is triggered. By the aid of the scheme, the technical problem that monitoring is inaccurate in operation monitoring of power generation equipment in the prior art is solved. Drawings The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specifica