CN-122014409-A - Method and system for monitoring running state of diesel generator

CN122014409ACN 122014409 ACN122014409 ACN 122014409ACN-122014409-A

Abstract

The application provides a diesel generator running state monitoring method and system, which are characterized in that time-frequency domain characteristic information for representing the running state stability of a diesel generator is extracted from multisource monitoring parameters of the diesel generator under the running condition, an abnormality identification layer and a fault mechanism-data hybrid driving reasoning layer for the diesel generator are constructed, a suspected abnormality region of the diesel generator under the running condition is determined through the abnormality identification layer and the time-frequency domain characteristic information, fault reasoning results of abnormality type, fault cause and fault severity of the diesel generator are determined through the fault mechanism-data hybrid driving reasoning layer and the suspected abnormality region, running health degree score of the diesel generator is determined, and the running state of the diesel generator is subjected to grading diagnosis and early warning through the running health degree score and the fault reasoning results. By adopting the scheme of the application, the abnormal association area can be accurately identified based on the complex operation data of the diesel generator so as to monitor the operation state of the diesel generator.

Inventors

Tu Yongmao
WU XIAOQING

Assignees

广州保和电力技术服务有限公司

Dates

Publication Date: 20260512
Application Date: 20260403

Claims (10)

1. The method for monitoring the running state of the diesel generator is characterized by comprising the following steps of: Acquiring multisource monitoring parameters of a diesel generator under an operation condition, wherein the multisource monitoring parameters comprise electrical parameters and mechanical operation parameters of the diesel generator; Performing time-frequency domain analysis on the multisource monitoring parameters to obtain time-frequency domain characteristic information for representing the running state stability of the diesel generator; Constructing an anomaly identification layer for performing anomaly region pre-diagnosis on the diesel generator and a fault mechanism-data hybrid driving reasoning layer for performing anomaly type and fault cause reasoning based on a mapping relation among a fault mechanism knowledge base, historical multisource monitoring parameters and historical anomaly data corresponding to the diesel generator; Performing abnormality positioning pre-diagnosis on the running state of the diesel generator through the abnormality identification layer and the time-frequency domain characteristic information to obtain a suspected abnormality region of the diesel generator under the running condition, and performing fault reasoning on a distortion characteristic set corresponding to the suspected abnormality region through the fault mechanism-data hybrid driving reasoning layer to obtain a fault reasoning result of the abnormality type, the fault cause and the fault severity of the diesel generator; Generating an operation health degree score of the diesel generator according to a time sequence degradation rule of the multi-source monitoring parameter in a time dimension and an abnormal diagnosis result, and carrying out hierarchical diagnosis early warning on the operation state of the diesel generator according to the operation health degree score and the fault reasoning result.
2. The method of claim 1, wherein performing time-frequency domain analysis on the multisource monitoring parameters to obtain time-frequency domain feature information for characterizing the stability of the operating state of the diesel generator specifically comprises: preprocessing the multisource monitoring parameters to obtain preprocessed multisource monitoring parameters; and extracting time-frequency domain characteristics of the preprocessed multisource monitoring parameters to obtain time-frequency domain characteristic information representing the running state of the diesel generator.
3. The method of claim 1, wherein constructing an anomaly identification layer for performing anomaly region pre-diagnosis on the diesel generator and a failure mechanism-data hybrid driving reasoning layer for performing anomaly type and failure cause reasoning based on a mapping relation among a failure mechanism knowledge base, a historical multi-source monitoring parameter and historical anomaly data corresponding to the diesel generator specifically comprises: constructing a fault mechanism knowledge base corresponding to the diesel generator; acquiring historical multisource monitoring parameters and historical abnormal data corresponding to the diesel generator; determining a mapping relation among the fault mechanism knowledge base, the historical multi-source monitoring parameters and the historical abnormal data; Constructing an abnormality identification layer for performing abnormality region pre-diagnosis on the diesel generator according to the mapping relation; acquiring historical abnormal type data of the diesel generator; And constructing a fault mechanism-data hybrid driving reasoning layer for carrying out abnormality type and fault cause reasoning on the diesel generator through the mapping relation and the historical abnormality type data.
4. The method of claim 1, wherein performing abnormality localization pre-diagnosis on the operation state of the diesel generator through the abnormality identification layer and the time-frequency domain feature information, and obtaining a suspected abnormality region of the diesel generator under the operation condition specifically comprises: performing abnormality diagnosis on the time-frequency domain characteristic information according to the abnormality recognition layer to obtain abnormality diagnosis information in the time-frequency domain characteristic information; Determining an abnormal coupling judgment rule of time domain stability characteristics and frequency domain energy distribution characteristics of the diesel generator under an operation condition; and carrying out abnormal positioning on the running state of the diesel generator through the abnormal diagnosis information and the abnormal coupling judgment rule to obtain a suspected abnormal region of the diesel generator under the running condition.
5. The method of claim 1, wherein the fault mechanism-data hybrid drive reasoning layer performs fault reasoning on the distortion feature set corresponding to the suspected abnormal region, and the obtaining the fault reasoning result of the diesel generator abnormal type, the fault cause and the fault severity specifically includes: determining a distortion feature set corresponding to the suspected abnormal region in the time-frequency domain feature information; Inputting the distortion feature set into the fault mechanism-data hybrid driving reasoning layer, and carrying out fault reasoning on the abnormality type or abnormality reason of the distortion feature set to obtain a fault reasoning result of the abnormality type, the fault cause and the fault severity of the diesel generator.
6. The method of claim 1, wherein generating the running health score of the diesel generator based on the time-series degradation law and the abnormality diagnosis result of the multi-source monitoring parameter in the time dimension specifically comprises: determining a time sequence degradation rule and an abnormality diagnosis result of the multi-source monitoring parameter in a time dimension; determining a reference time sequence degradation rule of the diesel generator; and determining the running health degree score of the diesel generator according to the time sequence degradation rule, the abnormality diagnosis result and the reference time sequence degradation rule.
7. The method of claim 1, wherein performing a hierarchical diagnostic pre-warning of the operational status of the diesel generator from the operational health score and the fault inference result comprises: Deep fusion is carried out on the operation health degree score and the fault reasoning result to obtain diagnosis fusion information of the operation state of the diesel generator; Determining a hierarchical diagnosis early warning decision of the running state of the diesel generator; and diagnosing the diagnosis fusion information according to the hierarchical diagnosis early warning decision, and performing hierarchical diagnosis early warning on the running state of the diesel generator according to a diagnosis result.
8. A diesel generator operating condition monitoring system, comprising: The acquisition module is used for acquiring multisource monitoring parameters of the diesel generator under the operating condition, wherein the multisource monitoring parameters comprise electrical parameters and mechanical operating parameters of the diesel generator; the processing module is used for carrying out time-frequency domain analysis on the multisource monitoring parameters to obtain time-frequency domain characteristic information for representing the running state stability of the diesel generator; The processing module is further used for constructing an anomaly identification layer for performing anomaly region pre-diagnosis on the diesel generator and a fault mechanism-data hybrid driving reasoning layer for performing anomaly type and fault cause reasoning based on a mapping relation among a fault mechanism knowledge base, historical multisource monitoring parameters and historical anomaly data corresponding to the diesel generator; The processing module is further used for carrying out abnormality positioning pre-diagnosis on the running state of the diesel generator through the abnormality identification layer and the time-frequency domain characteristic information to obtain a suspected abnormality region of the diesel generator under the running working condition, and carrying out fault reasoning on a distortion characteristic set corresponding to the suspected abnormality region through the fault mechanism-data hybrid driving reasoning layer to obtain a fault reasoning result of the abnormality type, the fault cause and the fault severity of the diesel generator; and the execution module is used for generating an operation health degree score of the diesel generator according to a time sequence degradation rule and an abnormality diagnosis result of the multi-source monitoring parameter in a time dimension, and carrying out hierarchical diagnosis early warning on the operation state of the diesel generator according to the operation health degree score and the fault reasoning result.
9. A computer device comprising a memory storing code and a processor configured to obtain the code and to perform the diesel generator operating condition monitoring method of any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the diesel generator operation state monitoring method according to any one of claims 1 to 7.

Description

Method and system for monitoring running state of diesel generator Technical Field The application relates to the technical field of operation state monitoring, in particular to a method and a system for monitoring the operation state of a diesel generator. Background The operation state monitoring is a technical means for accurately judging the current working condition, evaluating the performance degradation trend and identifying potential abnormality by collecting, processing and analyzing various key parameters generated in the operation process of the equipment in real time or periodically, and provides objective basis for early warning of faults, maintenance decision optimization and full life cycle management of the equipment. The diesel generator is used as important standby power equipment, the running state of the diesel generator is directly related to the stable running of a matched system, various running data related to electricity and machinery can present complex change characteristics in the long-term running process of the equipment, meanwhile, the running of the equipment is easily influenced by various factors such as load fluctuation, environmental conditions, component aging and the like, the traditional monitoring mode is mostly dependent on single parameter judgment or manual experience, potential abnormality in the running of the equipment is difficult to comprehensively capture, the position of abnormality occurrence cannot be precisely positioned, the reason of abnormality occurrence cannot be precisely resolved, the quantitative evaluation is difficult to the whole health state of the equipment, abnormal omission judgment and misjudgment are easy to occur, the fault risk cannot be prejudged in advance, and the reliable running and operation and maintenance efficiency of the equipment are influenced. Disclosure of Invention The application provides a diesel generator running state monitoring method and system, which can accurately identify an abnormal correlation area based on complex running data of a diesel generator so as to monitor the running state of the diesel generator. In a first aspect, the present application provides a method for monitoring an operating state of a diesel generator, comprising the steps of: Acquiring multisource monitoring parameters of a diesel generator under an operation condition, wherein the multisource monitoring parameters comprise electrical parameters and mechanical operation parameters of the diesel generator; Performing time-frequency domain analysis on the multisource monitoring parameters to obtain time-frequency domain characteristic information for representing the running state stability of the diesel generator; Constructing an anomaly identification layer for performing anomaly region pre-diagnosis on the diesel generator and a fault mechanism-data hybrid driving reasoning layer for performing anomaly type and fault cause reasoning based on a mapping relation among a fault mechanism knowledge base, historical multisource monitoring parameters and historical anomaly data corresponding to the diesel generator; Performing abnormality positioning pre-diagnosis on the running state of the diesel generator through the abnormality identification layer and the time-frequency domain characteristic information to obtain a suspected abnormality region of the diesel generator under the running condition, and performing fault reasoning on a distortion characteristic set corresponding to the suspected abnormality region through the fault mechanism-data hybrid driving reasoning layer to obtain a fault reasoning result of the abnormality type, the fault cause and the fault severity of the diesel generator; Generating an operation health degree score of the diesel generator according to a time sequence degradation rule of the multi-source monitoring parameter in a time dimension and an abnormal diagnosis result, and carrying out hierarchical diagnosis early warning on the operation state of the diesel generator according to the operation health degree score and the fault reasoning result. In some embodiments, performing time-frequency domain analysis on the multisource monitoring parameter to obtain time-frequency domain feature information for characterizing the running state stability of the diesel generator specifically includes: preprocessing the multisource monitoring parameters to obtain preprocessed multisource monitoring parameters; and extracting time-frequency domain characteristics of the preprocessed multisource monitoring parameters to obtain time-frequency domain characteristic information representing the running state of the diesel generator. In some embodiments, based on a mapping relationship among a fault mechanism knowledge base, a historical multisource monitoring parameter and historical abnormal data corresponding to the diesel generator, constructing an abnormal recognition layer for performing abnormal region pre-diagnosis on th