A method and system for fault prevention of diesel generator sets with real-time perception

Abstract

The present invention relates to the technical field of state monitoring and fault diagnosis of electrical equipment, and discloses a method and system for fault prevention of diesel generator sets with real-time perception. The method includes obtaining the operation data of the diesel generator set; extracting high-frequency features according to the mechanical vibration data to obtain the mechanical fault trend; performing trend analysis on the operation data of the diesel generator set by using Fourier transform to obtain the electrical system fault trend; performing time series analysis on the control signal data to obtain the control signal fault trend; performing collaborative analysis on the load change data to obtain the load influence factor; performing correlation analysis on the operation data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend and the load influence factor, and establishing a comprehensive fault prediction model to obtain a fault prevention strategy. This method can achieve global dynamic fault prevention of diesel generator sets.

Description

Technical Field

[0001] The present invention relates to the technical field of state monitoring and fault diagnosis of electrical equipment, and particularly to a method and system for preventing faults in a diesel generator set with real-time perception. Background Art

[0002] At present, diesel generator sets are widely used in industries, commerce, and critical infrastructure. As important backup or main power supply equipment, the stability and reliability of their operation directly affect the security of power supply. However, due to the influence of load changes, environmental factors, and mechanical wear during the long-term operation of diesel generator sets, faults are likely to occur, resulting in sudden shutdowns, which seriously affect production and power supply stability. Therefore, how to achieve early prevention of faults in diesel generator sets and improve the reliability and operation efficiency of equipment has become the focus of current technical research.

[0003] In an existing technology, regular manual inspections, fixed sensor monitoring, and statistical analysis methods based on historical data are used to evaluate the health status of diesel generator sets. For example, operation data of the generator set is collected by installing vibration sensors, temperature sensors, and oil pressure monitoring devices, and fault judgments are made relying on preset thresholds or empirical rules. Some systems use fault diagnosis methods based on offline analysis to model fault modes through historical data to predict possible faults, combine signal processing methods to extract features from the vibration signals, noise signals, or emission data of diesel generator sets to identify abnormal operating conditions, and use statistical learning or mechanism-based modeling methods for fault classification and early warning. In addition, a centralized fault management system based on remote monitoring is also used to analyze the operation status of diesel generator sets through cloud data to provide operation and maintenance guidance.

[0004] However, the manual inspection cycle is long, and it is difficult to detect potential hazards in a timely manner. The layout range of fixed sensors is limited, and it is impossible to comprehensively perceive the operation status of the generator set. The modeling method based on offline analysis is difficult to dynamically adapt to complex operating condition changes. The signal processing method is prone to identification errors when facing noise interference or multi-variable coupling characteristics. The remote monitoring system may lead to inaccurate fault prediction in the case of data transmission delay or insufficient model generalization ability. Thus, there is a problem in the prior art that it is difficult to achieve global dynamic fault prevention of diesel generator sets. Summary of the Invention

[0005] The present invention provides a method and system for preventing faults in a diesel generator set with real-time perception to achieve global dynamic fault prevention of diesel generator sets.

[0006] In a first aspect, to solve the above technical problems, the present invention provides a method for preventing faults in a diesel generator set with real-time perception, including:

[0007] Obtain the operating data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data;

[0008] According to the mechanical vibration data, use wavelet transform to extract high-frequency features and obtain the mechanical fault trend;

[0009] According to the operating data of the diesel generator set, use Fourier transform for trend analysis to obtain the electrical system fault trend;

[0010] According to the control signal data, perform time series analysis, extract the time series change law, and obtain the control signal fault trend;

[0011] According to the load change data, combine the ambient temperature data and the humidity information data for multivariate collaborative analysis to obtain the load impact factor;

[0012] According to the operating data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor, perform correlation analysis and establish a comprehensive fault prediction model to obtain the fault prevention strategy.

[0013] In an alternative embodiment, the obtaining of the operating data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data, includes:

[0014] Obtain the original mechanical vibration data, original electrical parameter data, original control signal data, original load change data, original ambient temperature data, and original humidity information data;

[0015] Define the original operating data of the diesel generator set to include the original mechanical vibration data, the original electrical parameter data, the original control signal data, the original load change data, the original ambient temperature data, and the original humidity information data;

[0016] According to the original operating data of the diesel generator set, perform integrity analysis and use interpolation algorithm to fill in the missing values to obtain the complete operating data of the diesel generator set;

[0017] According to the complete operating data of the diesel generator set, perform data format standardization processing to obtain the standardized operating data of the diesel generator set;

[0018] The operating data of the diesel generator set includes mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data.

[0019] In an alternative embodiment, the method of extracting high-frequency features from the mechanical vibration data by wavelet transform to obtain the mechanical fault trend includes:

[0020] Performing feature extraction on the mechanical vibration data by wavelet transform to obtain high-frequency feature components;

[0021] Based on the high-frequency feature components, combining with a preset mechanical fault feature library, performing mechanical fault type frequency feature matching to obtain potential mechanical fault types;

[0022] Calculating the resonance frequencies of each component of the mechanical structure based on the mechanical vibration data, analyzing the corresponding relationship between mechanical wear and the vibration spectrum, and obtaining the degree of mechanical wear;

[0023] Calculating the acceleration characteristics of mechanical looseness faults based on the mechanical vibration data, performing acceleration time-domain signal analysis, and obtaining the risk of mechanical looseness;

[0024] Based on the potential mechanical fault types, the degree of mechanical wear, and the risk of mechanical looseness, combining with a preset mechanical fault feature library, performing comprehensive feature matching of mechanical fault trends to obtain the mechanical fault trend.

[0025] In an alternative embodiment, the method of performing trend analysis on the diesel generator set operation data by Fourier transform to obtain the electrical system fault trend includes:

[0026] Performing analysis on the electrical parameter data by Fourier transform to obtain harmonic components;

[0027] Based on the harmonic components, combining with a preset electrical fault feature library, performing electrical fault type harmonic feature matching to obtain potential electrical fault types of the electrical system;

[0028] Performing insulation aging rate analysis based on the environmental temperature data and the humidity information data to obtain the insulation aging trend;

[0029] Based on the insulation aging trend and the potential electrical fault types of the electrical system, combining with a preset electrical fault feature library, performing comprehensive feature matching of electrical fault trends to obtain the electrical fault trend.

[0030] In an alternative embodiment, the method of performing time series analysis on the control signal data to extract time series change rules and obtain the control signal fault trend includes:

[0031] Performing feature extraction on the control signal data by time series decomposition to obtain the time series change rules of the control signal;

[0032] According to the regular pattern of the timing change of the control signal, match it with the fault waveform features in the preset control signal fault feature library to obtain the abnormal type of the control signal;

[0033] According to the abnormal type of the control signal, combine with the operating state of the control system, analyze the response time and overshoot of the control signal, and obtain the dynamic stability of the control system;

[0034] According to the regular pattern of the timing change of the control signal, the abnormal type of the control signal, and the dynamic stability of the control system, combine with the preset control signal fault feature library, conduct comprehensive feature matching of the control signal fault trend, and obtain the control signal fault trend.

[0035] In an alternative embodiment, the multivariate collaborative analysis based on the load change data, combined with the environmental temperature data and the humidity information data to obtain the load impact factor includes:

[0036] Extract features from the load change data to obtain a load change feature vector including load magnitude, change rate, and periodic characteristics;

[0037] According to the load change feature vector, use the finite element analysis method to conduct mechanical stress distribution analysis to obtain the risk of mechanical stress exceeding the limit;

[0038] According to the load change feature vector, combine with the tribology model to conduct mechanical wear rate analysis under different load conditions to obtain the trend of increased mechanical wear;

[0039] According to the load change feature vector, use the modal analysis method to conduct wavelet analysis on the mutation signal, extract transient vibration characteristics, and obtain the risk of mechanical resonance;

[0040] According to the risk of mechanical stress exceeding the limit, the trend of increased mechanical wear, the risk of mechanical resonance, the environmental temperature data, and the humidity information data, use the support vector machine algorithm to conduct multivariate collaborative analysis to obtain the load impact factor.

[0041] In an alternative embodiment, the correlation analysis and establishment of a comprehensive fault prediction model based on the diesel generator set operation data, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor to obtain the fault prevention strategy includes:

[0042] According to the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor, use the multi-source data fusion technology to conduct correlation impact analysis to obtain a multi-dimensional correlation matrix;

[0043] According to the multi-dimensional correlation matrix, a deep neural network algorithm is used for training and modeling to obtain a fault probability prediction model;

[0044] According to the fault probability prediction model, combined with a preset maintenance cost and risk consequence weight table, dynamic optimization of model parameters is carried out to obtain a comprehensive fault prediction model;

[0045] The operation data of the diesel generator set is input into the comprehensive fault prediction model to obtain a fault prevention strategy.

[0046] In a second aspect, the present invention provides a real-time perception diesel generator set fault prevention system, including:

[0047] A data acquisition module for acquiring the operation data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, environmental temperature data, and humidity information data;

[0048] A mechanical inspection module for extracting high-frequency features by wavelet transform according to the mechanical vibration data to obtain a mechanical fault trend;

[0049] An electrical analysis module for performing trend analysis by Fourier transform according to the operation data of the diesel generator set to obtain an electrical system fault trend;

[0050] A control diagnosis module for performing time series analysis according to the control signal data, extracting the time series change rule, and obtaining a control signal fault trend;

[0051] A load analysis module for performing multivariate collaborative analysis according to the load change data, combining the environmental temperature data and the humidity information data to obtain a load influence factor;

[0052] A result output module for performing correlation analysis according to the operation data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load influence factor, and establishing a comprehensive fault prediction model to obtain a fault prevention strategy. In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the real-time perception diesel generator set fault prevention method described in any one of the above.

[0053] In a fourth aspect, the present invention further provides a computer-readable storage medium. The computer-readable storage medium includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the real-time perception diesel generator set fault prevention method described in any one of the above.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The present invention discloses a method for preventing faults in a diesel generator set with real-time perception, which includes obtaining the operating data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, environmental temperature data, and humidity information data; according to the mechanical vibration data, using wavelet transform to extract high-frequency features to obtain the mechanical fault trend; according to the operating data of the diesel generator set, using Fourier transform for trend analysis to obtain the electrical system fault trend; according to the control signal data, performing time series analysis to extract the time series change law to obtain the control signal fault trend; according to the load change data, combining the environmental temperature data and the humidity information data for multivariate collaborative analysis to obtain the load impact factor; according to the operating data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor, performing correlation analysis and establishing a comprehensive fault prediction model to obtain a fault prevention strategy.

[0056] The present invention realizes the comprehensive perception of the operating state of the diesel generator set by constructing a multi-source data fusion analysis system, and constructs an accurate fault prediction model based on the feature extraction and trend analysis of different types of data. First, by collecting the mechanical vibration data, electrical parameter data, control signal data, load change data, environmental temperature data, and humidity information data of the diesel generator set, the comprehensiveness of the monitoring data is ensured. Then, wavelet transform is used to extract the high-frequency features of the mechanical vibration data to identify potential mechanical fault trends; Fourier transform is used to perform frequency domain analysis on the electrical parameter data to reveal potential electrical system fault trends in the electrical system; the control signal data is processed by the time series analysis method to extract the time series change law, thereby identifying abnormal trends in the control system; combining the load change data, environmental temperature, and humidity information, multivariate collaborative analysis is carried out to obtain the load impact factor. Based on the above analysis results, correlation analysis is performed to establish a comprehensive fault prediction model to fuse multi-source data, improve the accuracy of fault prevention, and obtain a fault prevention strategy. The present invention can adaptively adjust the warning mechanism and maintenance measures according to different operating states of the diesel generator set, making up for the limitations of the prior art in fault prevention, and thus realizing the global dynamic fault prevention of the diesel generator set. Description of the Drawings

[0057] Figure 1 is a schematic flowchart of a method for preventing faults in a diesel generator set with real-time perception provided by the first embodiment of the present invention;

[0058] Figure 2It is a schematic structural diagram of a real-time perception diesel generator set fault prevention system provided by the second embodiment of the present invention. Specific implementation manners

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0060] Refer to Figure 1 , the first embodiment of the present invention provides a real-time perception diesel generator set fault prevention method, including the following steps:

[0061] S11, obtain the operation data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data and humidity information data;

[0062] S12, according to the mechanical vibration data, adopt wavelet transform to extract high-frequency features to obtain the mechanical fault trend;

[0063] S13, according to the operation data of the diesel generator set, adopt Fourier transform for trend analysis to obtain the electrical system fault trend;

[0064] S14, according to the control signal data, perform time series analysis, extract the time series change law to obtain the control signal fault trend;

[0065] S15, according to the load change data, combine the ambient temperature data and the humidity information data for multivariate collaborative analysis to obtain the load influence factor;

[0066] S16, according to the operation data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend and the load influence factor, perform correlation analysis and establish a comprehensive fault prediction model to obtain the fault prevention strategy.

[0067] In step S11, it is necessary to obtain the operation data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data and humidity information data.

[0068] In one implementation manner, the obtaining of the operation data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data and humidity information data, includes:

[0069] Obtain the original mechanical vibration data, original electrical parameter data, original control signal data, original load change data, original ambient temperature data, and original humidity information data; define the original operating data of the diesel generator set to include the original mechanical vibration data, the original electrical parameter data, the original control signal data, the original load change data, the original ambient temperature data, and the original humidity information data; perform integrity analysis based on the original operating data of the diesel generator set, and use an interpolation algorithm to fill in the missing values to obtain the complete operating data of the diesel generator set; perform data format standardization processing based on the complete operating data of the diesel generator set to obtain the standardized operating data of the diesel generator set; the operating data of the diesel generator set includes mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data.

[0070] It should be noted that the original mechanical vibration data can be used to measure the vibration of the mechanical system through devices such as accelerometers and displacement sensors, and record parameters such as the amplitude and frequency of the vibration; the original electrical parameter data can be obtained through electrical parameter measuring instruments, such as voltmeters, ammeters, power analyzers, etc., to obtain electrical parameter data such as voltage, current, and power in the circuit; the original control signal data can be recorded by a data acquisition system for signals issued by the control system, such as PWM signals, analog signals, etc.; the original load change data can be collected in real time through sensors; the original ambient temperature data can be used to measure the change of the ambient temperature through temperature sensors, such as thermocouples and thermistors; the original humidity information data can be obtained through humidity sensors. Integrity analysis can identify missing values in the data, including checking whether the data records are continuous and whether there is data loss; after identifying the missing values, an interpolation algorithm is used to fill in these missing values. There are various interpolation algorithms, such as linear interpolation, Lagrange interpolation, KNN interpolation, etc., which are not required in the embodiments of the present invention. Data standardization processing is to standardize and unify the data, so that data from different sources, different formats, and different precisions have consistency and comparability when performing analysis and applications, and improve the quality and credibility of the data.

[0071] In step S12, it is necessary to extract high-frequency features using wavelet transform based on the mechanical vibration data to obtain the mechanical fault trend.

[0072] In one implementation, the extracting high-frequency features using wavelet transform based on the mechanical vibration data to obtain the mechanical fault trend includes:

[0073] Based on the mechanical vibration data, wavelet transform is used for feature extraction to obtain high-frequency feature components; based on the high-frequency feature components, combined with a preset mechanical fault feature library, mechanical fault type frequency feature matching is performed to obtain potential mechanical fault types; based on the mechanical vibration data, the resonance frequencies of each component of the mechanical structure are calculated, and the corresponding relationship between mechanical wear and vibration spectrum is analyzed to obtain the degree of mechanical wear; based on the mechanical vibration data, the acceleration characteristics of mechanical looseness faults are calculated, and acceleration time-domain signal analysis is performed to obtain the risk of mechanical looseness; based on the potential mechanical fault types, the degree of mechanical wear, and the risk of mechanical looseness, combined with a preset mechanical fault feature library, mechanical fault trend comprehensive feature matching is performed to obtain the mechanical fault trend.

[0074] It should be noted that wavelet transform can generate wavelet coefficients at different scales, and these coefficients reflect the characteristics of the signal at different scales and positions, facilitating the screening and extraction of different signal components. The mechanical fault trend refers to the possible fault types and their development trends of mechanical components during the operation of a diesel generator set, including but not limited to mechanical wear, looseness, fracture, etc. The acquisition of the mechanical fault trend depends on the in-depth analysis of mechanical vibration data. Specifically, high-frequency features are extracted through wavelet transform, and the vibration signal is decomposed to identify different frequency components, thereby revealing abnormal vibration patterns. By matching the extracted high-frequency feature components with a preset mechanical fault feature library, potential mechanical fault types can be identified, and the degree of mechanical wear can be calculated and analyzed in combination with the resonance frequency. At the same time, the risk of mechanical looseness is judged through acceleration time-domain signal analysis. Finally, through comprehensive fault feature matching, the mechanical fault trend is obtained. The mechanical fault trend is used for the real-time health monitoring of diesel generator sets, providing data support for subsequent fault prediction and prevention strategies to ensure the stable operation of the equipment.

[0075] In step S13, it is necessary to perform trend analysis on the diesel generator set operation data using Fourier transform to obtain the electrical system fault trend.

[0076] In one implementation, the performing trend analysis on the diesel generator set operation data using Fourier transform to obtain the electrical system fault trend includes:

[0077] Based on the electrical parameter data, Fourier transform is used for analysis to obtain harmonic components; based on the harmonic components, combined with a preset electrical fault feature library, electrical fault type harmonic feature matching is performed to obtain potential electrical fault types of the electrical system; based on the ambient temperature data and the humidity information data, insulation aging rate analysis is performed to obtain the insulation aging trend; based on the insulation aging trend and the potential electrical fault types of the electrical system, combined with a preset electrical fault feature library, electrical fault trend comprehensive feature matching is performed to obtain the electrical fault trend.

[0078] It should be noted that the electrical system fault trend refers to the types of faults that may occur in the electrical system during the operation of a diesel generator set and their evolution trends, including but not limited to problems such as harmonic distortion, insulation aging, short circuit, and overload. Obtaining the electrical system fault trend depends on the spectral analysis of electrical parameter data. Specifically, the electrical signal is decomposed through Fourier transform to identify harmonic components. By matching the extracted harmonic components with a preset electrical fault feature library, potential fault types of the electrical system are obtained, and the insulation aging rate is analyzed by combining environmental temperature and humidity information to evaluate the long-term health status of the electrical system. Finally, through comprehensive feature matching, the electrical fault trend is obtained. The electrical fault trend can be used for the electrical health monitoring of diesel generator sets, providing support for fault prediction and prevention strategies to ensure the stability and reliability of system operation.

[0079] In step S14, time series analysis needs to be performed based on the control signal data to extract the time series change law and obtain the control signal fault trend.

[0080] In one implementation, the performing time series analysis based on the control signal data to extract the time series change law and obtain the control signal fault trend includes:

[0081] Based on the control signal data, time series decomposition is used for feature extraction to obtain the control signal time series change law; according to the control signal time series change law, it is combined with the fault waveform features in a preset control signal fault feature library for matching to obtain the control signal anomaly type; according to the control signal anomaly type, combined with the operating state of the control system, the response time and overshoot of the control signal are analyzed to obtain the dynamic stability of the control system; according to the control signal time series change law, the control signal anomaly type, and the dynamic stability of the control system, combined with a preset control signal fault feature library, comprehensive feature matching of the control signal fault trend is performed to obtain the control signal fault trend.

[0082] It should be noted that the control signal fault trend refers to the signal anomalies and their trends over time in the control system of a diesel generator set, including but not limited to signal delay, loss, abnormal fluctuations, overshoot, and undershoot. Obtaining the control signal fault trend depends on the time series analysis of control signal data. Specifically, the time series variation law of the signal is extracted through time series decomposition to identify the long-term trend, periodic fluctuations, and sudden anomalies of signal characteristics. By matching the extracted time series characteristics with a preset control signal fault feature library, the type of control signal anomaly is obtained. Combining with the operating state of the diesel generator set, the dynamic response characteristics of the control system, such as signal response time and overshoot, are analyzed to obtain the dynamic stability of the control system. Finally, the control signal fault trend is obtained through comprehensive feature matching. The control signal fault trend can be used for the health monitoring of the control system of the diesel generator set, providing data support for abnormal signal detection and fault prevention, ensuring the stability and response accuracy of the control system, and improving the overall reliability of the diesel generator set.

[0083] In step S15, it is necessary to perform multivariate collaborative analysis based on the load change data, combined with the environmental temperature data and the humidity information data, to obtain the load impact factor.

[0084] In one implementation, the performing multivariate collaborative analysis based on the load change data, combined with the environmental temperature data and the humidity information data, to obtain the load impact factor includes:

[0085] Based on the load change data, feature extraction is performed to obtain a load change feature vector including load magnitude, change rate, and periodic characteristics; based on the load change feature vector, a finite element analysis method is used to perform mechanical stress distribution analysis to obtain the risk of mechanical stress exceeding the limit; based on the load change feature vector, combined with a tribology model, the mechanical wear rate analysis under different load conditions is performed to obtain the trend of mechanical wear aggravation; based on the load change feature vector, a modal analysis method is used to perform wavelet analysis on the mutation signal to extract transient vibration characteristics to obtain the risk of mechanical resonance; based on the risk of mechanical stress exceeding the limit, the trend of mechanical wear aggravation, the risk of mechanical resonance, the environmental temperature data, and the humidity information data, a support vector machine algorithm is used to perform multivariate collaborative analysis to obtain the load impact factor.

[0086] It should be noted that the load impact factor represents the degree to which the mechanical and electrical performance of a diesel generator set is affected by environmental factors under different load conditions. The acquisition of the load impact factor is based on multivariate collaborative analysis, comprehensively considering the load change characteristic vector, the risk of mechanical stress exceeding the limit, the trend of increased mechanical wear, the risk of mechanical resonance, the dynamic impact of the environmental temperature data and the humidity information data. The load change characteristic vector is obtained through a feature extraction method, and in combination with finite element analysis, tribology modeling and modal analysis, the risk of mechanical stress exceeding the limit, the trend of increased mechanical wear and the risk of mechanical resonance are respectively evaluated. At the same time, using the environmental temperature data and the humidity information data, combined with machine learning algorithms for multivariate analysis, a comprehensive load impact factor is calculated. The load impact factor is used to evaluate the reliability of the diesel generator set during actual operation, identify potential mechanical and electrical fault risks in advance, so as to optimize the load management strategy and improve the stability and service life of the unit.

[0087] In step S16, it is necessary to perform correlation analysis based on the diesel generator set operation data, the mechanical fault trend, the electrical system fault trend, the control signal fault trend and the load impact factor, and establish a comprehensive fault prediction model to obtain a fault prevention strategy.

[0088] In one implementation, the performing correlation analysis based on the diesel generator set operation data, the mechanical fault trend, the electrical system fault trend, the control signal fault trend and the load impact factor, and establishing a comprehensive fault prediction model to obtain a fault prevention strategy includes:

[0089] Based on the mechanical fault trend, the electrical system fault trend, the control signal fault trend and the load impact factor, using multi-source data fusion technology to perform correlation impact analysis to obtain a multi-dimensional correlation matrix; according to the multi-dimensional correlation matrix, using a deep neural network algorithm for training and modeling to obtain a fault probability prediction model; according to the fault probability prediction model, combined with a preset maintenance cost and risk consequence weight table, perform dynamic optimization of the model parameters to obtain a comprehensive fault prediction model; input the diesel generator set operation data into the comprehensive fault prediction model to obtain a fault prevention strategy.

[0090] It should be noted that the acquisition of the multi-dimensional correlation matrix involves data preprocessing, feature extraction, multi-source data fusion technology, and correlation impact analysis; generating a multi-dimensional correlation matrix to analyze the mutual influence relationship among various fault trends. The fault prevention strategy is a decision data set including operation parameter adjustment suggestions, maintenance priority ranking, and resource allocation plans, which is generated through multi-dimensional fault correlation analysis and cost-risk optimization modeling. Specifically, the fault prevention strategy includes dynamic adjustment instructions for the operation parameters of the diesel generator set, such as adjusting the generator output power range, optimizing the working intensity of the cooling system, regulating the fuel supply rate, etc., and also includes maintenance measure suggestions, such as determining whether the maintenance priority is immediate shutdown for repair or observation of operation, listing spare part replacement lists such as bearing models, adjusting the lubrication cycle, etc., and also includes resource allocation plans, such as formulating a manpower scheduling plan like the division of the maintenance team, an equipment deployment plan like the startup time of the standby unit, and the budget allocation ratio. The generation of these strategies depends on the comprehensive analysis of multi-source data, such as input mechanical fault trends like the bearing wear degree reaching 80%, electrical system fault trends like the insulation aging rate being 0.5% per month, control signal fault trends like the response overshoot being 15%, load impact factors like the mechanical stress overrun risk level being grade three, and real-time operation data like the current load rate being 85%. By mapping these data into feature vectors and using the fault evolution knowledge graph to establish association rules, such as increased bearing wear will lead to increased vibration and then increase the risk of electrical connection loosening, and then combining the maintenance cost like the cost of replacing the bearing being five thousand yuan and the risk consequence weight like the shutdown loss being fifty thousand yuan per hour, using a deep neural network to calculate the optimal strategy. For example, when the comprehensive fault prediction model outputs a bearing fault probability of 92% and an electrical joint loosening probability of 65%, the generated fault prevention strategy may include immediately reducing the load to 70% to slow down the bearing wear rate, arranging to replace the bearing during the low-load period at night within three days to reduce the maintenance cost, synchronously checking and strengthening the electrical joints to prevent chain faults, and at the same time allocating two maintenance personnel and the spare bearing inventory number B203. These strategies are pushed to the operation and maintenance terminal in the form of an instruction list, and a work order number and an execution schedule are automatically generated, so as to achieve a closed-loop control from fault prediction to prevention execution by quantifying the fault evolution path and economic constraints, and finally achieve a balance between minimizing risks and optimizing operation and maintenance costs, realizing the global dynamic fault prevention of the diesel generator set.

[0091] For the convenience of understanding the present invention, some preferred embodiments of the present invention will be further described below.

[0092] The working process of the present invention will be described below by taking a relatively common scenario as an example. Please also refer to Figure 2 , which is Figure 1 a schematic diagram of the working scenario of the method of

[0093] A large data center is equipped with a set of diesel generator sets as the backup power supply of the data center to ensure that the servers can operate continuously and stably in case of a main grid failure. The diesel generator sets need to have high reliability and the ability to operate stably for a long time. Therefore, during their operation, real-time monitoring and prevention of potential faults are the key to ensuring the reliability of the system.

[0094] During the application of the present invention, the operation data of the diesel generator sets are first obtained, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data. These data are collected by multiple sensors and monitoring devices and are recorded and analyzed in real time by a computer system.

[0095] After obtaining the data, the mechanical vibration data is first processed. Wavelet transform is used to extract high-frequency features and analyze the spectral changes of the vibration signals. By matching with a preset mechanical fault feature library, mechanical problems such as possible bearing wear, gear faults, or looseness are identified. At the same time, the resonance frequency of the mechanical structure is calculated, and the relationship between the vibration characteristics and the wear state of the mechanical components is analyzed to further evaluate the risk of mechanical looseness. Through the comprehensive feature matching of the above analysis results, the mechanical fault trend of the diesel generator sets is obtained.

[0096] Meanwhile, Fourier transform analysis is performed on the electrical parameter data of the diesel generator sets to extract harmonic components, and combined with a preset electrical fault feature library, potential fault types of the electrical system are identified. For example, if it is analyzed that the content of a specific harmonic increases abnormally, it may indicate problems such as local short circuit or insulation aging of the generator winding. In addition, combined with the ambient temperature data and humidity information data, the aging rate of the insulating material is analyzed, and based on this, potential electrical faults that may occur in the future are predicted. According to the potential fault types of the electrical system and the insulation aging trend, combined with a preset electrical fault feature library for comprehensive feature matching, the electrical system fault trend is obtained.

[0097] In terms of control signals, the system uses time series analysis methods to decompose the timing changes of the control signals and extract the timing change rules of the control signals. By matching with the fault waveform features in a preset control signal fault feature library, the abnormal types of the control signals are obtained, such as signal delay, pulse loss, or noise interference. At the same time, combined with the operation state of the diesel generator sets, the response time and overshoot of the control signals are analyzed to obtain the dynamic stability of the control system. Finally, based on the comprehensive analysis of the control signal timing change rules, the control signal abnormal types, and the dynamic stability of the control system, the control signal fault trend is obtained.

[0098] In addition, the system processes the load change data, and performs multivariate collaborative analysis in combination with the ambient temperature data and the humidity information data. First, the load magnitude, change rate, and periodic characteristics are extracted from the load change data to obtain a load change feature vector, and the mechanical stress distribution is calculated using the finite element analysis method based on the load change feature vector to identify the risk of mechanical stress exceeding the limit. At the same time, in combination with the tribology model, the mechanical wear rate under different load conditions is evaluated to obtain the trend of increasing mechanical wear. In addition, modal analysis is performed on the mutation signal to extract transient vibration characteristics to identify the risk of mechanical resonance. Finally, in combination with the risk of mechanical stress exceeding the limit, the trend of increasing mechanical wear, the risk of mechanical resonance, the ambient temperature data, and the humidity information data, the system uses the support vector machine algorithm for multivariate collaborative analysis to obtain the load impact factor.

[0099] After obtaining the mechanical fault trend, electrical system fault trend, control signal fault trend, and load impact factor, correlation analysis is performed on all the data, and a comprehensive fault prediction model is established. Using the multi-source data fusion technology, a multi-dimensional correlation matrix is generated to analyze the mutual influence relationship between various fault trends. Based on this matrix, the system uses the deep neural network algorithm for training and modeling to establish a fault probability prediction model, and in combination with the preset maintenance cost and risk consequence weight table, the model parameters are dynamically optimized to generate a comprehensive fault prediction model. The operation data of the diesel generator set is input into the comprehensive fault prediction model to obtain a fault prevention strategy.

[0100] In specific applications, the operation data of the diesel generator set is input in real time, and the fault occurrence probability in the current state is calculated through the comprehensive fault prediction model. When the fault probability exceeds the preset threshold, the system automatically generates a fault prevention strategy, such as adjusting the load distribution, optimizing the control signal, replacing the worn parts in advance, or increasing the insulation protection measures, etc., so as to minimize the risk of the diesel generator set failing.

[0101] Through the method of the present invention, the diesel generator set can monitor its own state in real time during operation and predict possible faults in advance, thereby improving the reliability and service life of the equipment. In scenarios with high reliability requirements for power supply such as data centers, hospitals, and industrial parks, the application of this method can effectively reduce the risk of service interruption caused by generator failures and ensure the stable operation of the power system.

[0102] In summary, the present invention discloses a method for preventing faults in a diesel generator set with real-time perception, including obtaining the operating data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data; extracting high-frequency features by using wavelet transform according to the mechanical vibration data to obtain the mechanical fault trend; performing trend analysis by using Fourier transform according to the operating data of the diesel generator set to obtain the electrical system fault trend; performing time series analysis according to the control signal data to extract the time series change rule and obtain the control signal fault trend; performing multivariate collaborative analysis by combining the load change data, the ambient temperature data, and the humidity information data to obtain the load impact factor; performing correlation analysis according to the operating data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor, and establishing a comprehensive fault prediction model to obtain a fault prevention strategy.

[0103] The present invention realizes the comprehensive perception of the operating state of the diesel generator set by constructing a multi-source data fusion analysis system, and constructs an accurate fault prediction model based on the feature extraction and trend analysis of different types of data. First, the comprehensiveness of the monitoring data is ensured by collecting the mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data of the diesel generator set. Then, wavelet transform is used to extract the high-frequency features of the mechanical vibration data to identify potential mechanical fault trends; Fourier transform is used to perform frequency domain analysis on the electrical parameter data to reveal possible electrical system fault trends in the electrical system; the control signal data is processed by the time series analysis method to extract the time series change rule, thereby identifying the abnormal trend of the control system; multivariate collaborative analysis is carried out by combining the load change data, ambient temperature, and humidity information to obtain the load impact factor. Based on the above analysis results, correlation analysis is performed to establish a comprehensive fault prediction model to fuse multi-source data, improve the accuracy of fault prevention, and obtain a fault prevention strategy. The present invention can adaptively adjust the early warning mechanism and maintenance measures according to different operating states of the diesel generator set, making up for the limitations of the prior art in fault prevention, and thus realizing the global dynamic fault prevention of the diesel generator set.

[0104] Referring to Figure 2 , the second embodiment of the present invention provides a system for preventing faults in a diesel generator set with real-time perception, including:

[0105] A data acquisition module, configured to acquire the operating data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data;

[0106] A mechanical inspection module, configured to extract high-frequency features by using wavelet transform according to the mechanical vibration data, and obtain a mechanical fault trend;

[0107] An electrical analysis module, configured to perform trend analysis by using Fourier transform according to the operation data of the diesel generator set, and obtain an electrical system fault trend;

[0108] A control diagnosis module, configured to perform time series analysis according to the control signal data, extract time series change rules, and obtain a control signal fault trend;

[0109] A load analysis module, configured to perform multivariate collaborative analysis according to the load change data, in combination with the ambient temperature data and the humidity information data, and obtain a load impact factor;

[0110] A result output module, configured to perform correlation analysis according to the operation data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor, and establish a comprehensive fault prediction model, and obtain a fault prevention strategy.

[0111] It should be noted that a real-time perception diesel generator set fault prevention device provided in an embodiment of the present invention is used to execute all process steps of a real-time perception diesel generator set fault prevention method in the above embodiment. The working principles and beneficial effects of the two correspond one by one, and thus will not be elaborated herein.

[0112] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a real-time perception diesel generator set fault prevention program. When the processor executes the computer program, the steps in the above-mentioned embodiments of the real-time perception diesel generator set fault prevention method are implemented, such as Figure 1 the step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented, such as the load analysis module.

[0113] Exemplarily, the computer program may be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0114] The electronic device can be a computing device such as a desktop computer, notebook, handheld computer, and smart tablet, etc. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0115] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device, and connects various parts of the entire electronic device through various interfaces and lines.

[0116] The memory can be used to store the computer programs and / or modules. The processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.), etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one magnetic disk storage device, flash device, or other volatile solid-state storage devices.

[0117] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0118] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0119] The specific embodiments described above have further elaborated on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for preventing faults in a diesel generator set with real-time perception, characterized in that, Executed by a computer, including: Obtain the operating data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, environmental temperature data, and humidity information data; According to the mechanical vibration data, use wavelet transform to extract high-frequency features and obtain the mechanical fault trend, including: According to the mechanical vibration data, use wavelet transform for feature extraction to obtain high-frequency feature components; According to the high-frequency feature components, combined with a preset mechanical fault feature library, perform mechanical fault type frequency feature matching to obtain potential mechanical fault types; According to the mechanical vibration data, calculate the resonance frequencies of each component of the mechanical structure, analyze the corresponding relationship between mechanical wear and vibration spectrum, and obtain the degree of mechanical wear; According to the mechanical vibration data, calculate the acceleration characteristics of mechanical looseness faults, perform acceleration time-domain signal analysis, and obtain the risk of mechanical looseness; According to the potential mechanical fault types, the degree of mechanical wear, and the risk of mechanical looseness, combined with a preset mechanical fault feature library, perform comprehensive feature matching of mechanical fault trends to obtain the mechanical fault trend; According to the operating data of the diesel generator set, use Fourier transform for trend analysis to obtain the electrical system fault trend; According to the control signal data, perform time series analysis, extract the time series change law, and obtain the control signal fault trend; According to the load change data, combined with the environmental temperature data and the humidity information data, perform multivariate collaborative analysis to obtain the load impact factor; According to the operating data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor, perform correlation analysis and establish a comprehensive fault prediction model to obtain a fault prevention strategy, including: According to the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load impact factor, use multi-source data fusion technology to perform correlation impact analysis to obtain a multi-dimensional correlation matrix; According to the multi-dimensional correlation matrix, use a deep neural network algorithm for training and modeling to obtain a fault probability prediction model; According to the fault probability prediction model, combined with a preset maintenance cost and risk consequence weight table, perform dynamic optimization of model parameters to obtain a comprehensive fault prediction model; Input the operating data of the diesel generator set into the comprehensive fault prediction model to obtain a fault prevention strategy.

2. The real-time perception-based diesel generator set fault prevention method according to claim 1, characterized in that, The obtaining of the operating data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, environmental temperature data, and humidity information data, includes: Obtain the original mechanical vibration data, original electrical parameter data, original control signal data, original load change data, original environmental temperature data, and original humidity information data; Define the original operating data of the diesel generator set as including the original mechanical vibration data, the original electrical parameter data, the original control signal data, the original load change data, the original environmental temperature data, and the original humidity information data; Based on the original operating data of the diesel generator set, perform integrity analysis, and use interpolation algorithm to fill in the missing values to obtain the complete operating data of the diesel generator set; Based on the complete operating data of the diesel generator set, perform standardization processing on the data format to obtain the standardized operating data of the diesel generator set; The operating data of the diesel generator set includes mechanical vibration data, electrical parameter data, control signal data, load change data, environmental temperature data, and humidity information data.

3. The real-time perception-based diesel generator set fault prevention method according to claim 1, wherein, Based on the operating data of the diesel generator set, use Fourier transform for trend analysis to obtain the electrical system fault trend, including: Based on the electrical parameter data, use Fourier transform for analysis to obtain the harmonic components; Based on the harmonic components, combine with the preset electrical fault feature library to perform harmonic feature matching of electrical fault types to obtain the potential electrical fault types of the electrical system; Based on the environmental temperature data and the humidity information data, perform insulation aging rate analysis to obtain the insulation aging trend; Based on the insulation aging trend and the potential electrical fault types of the electrical system, combine with the preset electrical fault feature library to perform comprehensive feature matching of electrical fault trends to obtain the electrical fault trend.

4. The real-time perception-based diesel generator set fault prevention method according to claim 1, characterized in that Based on the control signal data, perform time series analysis to extract the time series change law to obtain the control signal fault trend, including: Based on the control signal data, use time series decomposition for feature extraction to obtain the time series change law of the control signal; Based on the time series change law of the control signal, combine with the fault waveform features in the preset control signal fault feature library for matching to obtain the abnormal type of the control signal; Based on the abnormal type of the control signal, combine with the operating state of the control system to analyze the response time and overshoot of the control signal to obtain the dynamic stability of the control system; Based on the time series change law of the control signal, the abnormal type of the control signal, and the dynamic stability of the control system, combine with the preset control signal fault feature library to perform comprehensive feature matching of the control signal fault trend to obtain the control signal fault trend.

5. The real-time perception-based diesel generator set fault prevention method according to claim 1, wherein Based on the load change data, combine with the environmental temperature data and the humidity information data to perform multivariate collaborative analysis to obtain the load impact factor, including: Based on the load change data, perform feature extraction to obtain a load change feature vector including load magnitude, change rate, and periodic characteristics; Based on the load change feature vector, use the finite element analysis method to perform mechanical stress distribution analysis to obtain the risk of mechanical stress exceeding the limit; Based on the load change feature vector, combine with the tribology model to perform mechanical wear rate analysis under different load conditions to obtain the trend of increased mechanical wear; Based on the load change feature vector, use the modal analysis method to perform wavelet analysis on the mutation signal to extract the transient vibration characteristics to obtain the mechanical resonance risk; Based on the risk of mechanical stress exceeding the limit, the trend of increased mechanical wear, the mechanical resonance risk, the environmental temperature data, and the humidity information data, use the support vector machine algorithm to perform multivariate collaborative analysis to obtain the load impact factor.

6. A real-time perception diesel generator set fault prevention system, characterized in that, A method for preventing faults in a diesel generator set for real-time perception as described in any one of claims 1 to 5, comprising: A data acquisition module for acquiring operation data of the diesel generator set, including mechanical vibration data, electrical parameter data, control signal data, load change data, ambient temperature data, and humidity information data; A mechanical inspection module for extracting high-frequency features using wavelet transform based on the mechanical vibration data to obtain a mechanical fault trend; An electrical analysis module for performing trend analysis using Fourier transform based on the operation data of the diesel generator set to obtain an electrical system fault trend; A control diagnosis module for performing time series analysis based on the control signal data, extracting time series change rules, and obtaining a control signal fault trend; A load analysis module for performing multivariate collaborative analysis based on the load change data, combined with the ambient temperature data and the humidity information data, to obtain a load influence factor; A result output module for performing correlation analysis and establishing a comprehensive fault prediction model based on the operation data of the diesel generator set, the mechanical fault trend, the electrical system fault trend, the control signal fault trend, and the load influence factor, to obtain a fault prevention strategy.

7. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for preventing faults in a diesel generator set for real-time perception as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method for preventing faults in a diesel generator set for real-time perception as described in any one of claims 1 to 5.

Images