Generator fault automatic detection and report method and related system

Abstract

The invention belongs to the field of fault inspection, and discloses a generator fault automatic detection and report method and a related system.According to the generator fault automatic detection and report method and system, most valuable fault features can be extracted from high-dimensional original data through a fast Fourier transform (FFT) processing technology, the data size of subsequent processing and analysis is reduced, and the fault detection efficiency is improved. And the fault diagnosis is more focused and efficient. By preprocessing the signals, the influence of interference signals and noise can be effectively reduced, so that the accuracy of fault diagnosis is improved. The method is particularly important for identifying subtle fault signs, and potential fault problems can be found as soon as possible. According to the method, whether the generator is in a normal state or a specific fault state can be quickly and accurately identified by utilizing a machine learning algorithm of the SVM, which is crucial for timely maintenance and reduction of downtime.

Description

Technical Field

[0001] The invention belongs to the field of fault inspection, and in particular relates to a generator fault automatic detection and reporting method and a related system. Background Art

[0002] With the continuous development of industrial automation and intelligence, power generation equipment, as an important power source for industrial production, has a great significance for ensuring production continuity and reducing economic losses. However, in the long-term operation of generators, various faults are prone to occur due to various factors such as wear, overload, environmental factors, etc. If these faults are not detected and diagnosed in time, they may cause equipment performance degradation and even lead to serious accidents, resulting in huge economic losses and safety hazards.

[0003] Traditional generator fault detection methods mainly rely on manual inspections and empirical judgments. This method is not only inefficient, but also limited by personnel skills and experience, making it difficult to accurately and timely detect faults. In addition, manual detection methods cannot achieve real-time analysis and processing of large amounts of data, making it difficult to meet the requirements of modern industry for high efficiency and high reliability. In order to solve the above problems, researchers and engineers have developed various automatic fault detection systems. These systems usually include modules such as data acquisition, signal processing, and fault diagnosis, which can monitor and analyze the operating status of the generator in real time, thereby achieving automatic detection and reporting of faults.

[0004] However, the existing automatic fault detection systems still have some shortcomings, such as low fault diagnosis accuracy, slow system response speed, and low data management efficiency. Summary of the invention

[0005] The purpose of the present invention is to overcome the above-mentioned shortcomings of low fault diagnosis accuracy, slow system response speed and low data management efficiency, and to provide a generator fault automatic detection and reporting method and related system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for automatically detecting and reporting a generator fault, comprising the following steps: Collect the continuous time signals of each sensor of the generator, and convert the continuous time signals into discrete time signals through sampling conversion; Preprocess discrete time signals to remove high frequency noise and low frequency drift; Performing fast Fourier transform on the preprocessed discrete time signal to convert the preprocessed discrete time signal from the time domain to the frequency domain; Calculate the required eigenvalues ​​in the frequency domain and generate the required report based on the eigenvalues; Extracting a fault feature vector from the required feature values, converting the fault feature vector into a data set, and dividing the data set into training samples of a normal state and different fault states; Construct an SVM model, use different training samples to train the SVM model, and after verification, obtain a trained SVM model; All feature values ​​are input into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

[0007] A further improvement of the present invention is that, when preprocessing the discrete-time signal, the discrete-time signal is first subjected to denoising, and then the denoised discrete-time signal is subjected to filtering.

[0008] A further improvement of the present invention is that the method for performing fast Fourier transform on the preprocessed discrete-time signal is as follows:

[0009] in, is the discrete-time signal in the frequency domain complex frequency components, In the time domain Sample values, is the total number of sampling points, is an imaginary unit, is the frequency index, ranging from 0 to .

[0010] A further improvement of the present invention is that the required characteristic values ​​in the frequency domain include frequency peak value, spectrum energy and frequency bandwidth.

[0011] A further improvement of the present invention is that the spectrum energy is calculated as follows:

[0012]

[0013] in, is the spectrum energy, For the The magnitude of the complex frequency components, and Respectively Complex frequency components The real and imaginary parts of .

[0014] A further improvement of the present invention is that the calculation method of the trained SVM model is as follows:

[0015] The constraints are:

[0016] in, For the The Lagrange multiplier of samples, , is the penalty parameter, For the The Lagrange multiplier of samples, For the The class labels of samples, For the The class labels of samples, For the The feature vector of the sample For the The feature vector of the samples, is the sample size, is the kernel function.

[0017] A further improvement of the present invention is that the decision function of the trained SVM model is:

[0018] in, is the decision function, is the bias term.

[0019] In a second aspect, the present invention provides a generator fault automatic detection and reporting system, comprising: The sampling conversion module is used to collect the continuous time signals of each sensor of the generator and convert the continuous time signals into discrete time signals through sampling conversion; A preprocessing module is used to preprocess discrete time signals to eliminate high-frequency noise and low-frequency drift; A fast Fourier transform module is used to perform fast Fourier transform on the preprocessed discrete time signal, so as to convert the preprocessed discrete time signal from the time domain to the frequency domain; A report generation module is used to calculate the required eigenvalues ​​in the frequency domain and generate the required report based on the eigenvalues; A feature vector extraction module is used to extract a fault feature vector from the required feature values, convert the fault feature vector into a data set, and divide the data set into training samples of a normal state and different fault states; The model building module is used to build the SVM model. Different training samples are used to train the SVM model. After verification, the trained SVM model is obtained. The model output module is used to input all feature values ​​into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

[0020] In a third aspect, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of a method for automatically detecting and reporting generator faults when executing the computer program.

[0021] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of a method for automatic detection and reporting of generator faults.

[0022] Compared with the prior art, the present invention has the following beneficial effects: The present invention uses the fast Fourier transform (FFT) processing technology to extract the most valuable fault features from high-dimensional raw data, which not only reduces the amount of data for subsequent processing and analysis, but also makes fault diagnosis more focused and efficient. By preprocessing the signal, the influence of interference signals and noise can be effectively reduced, thereby improving the accuracy of fault diagnosis. This is particularly important for identifying subtle signs of faults and helps to detect potential fault problems early. The present invention uses the machine learning algorithm of the support vector machine (SVM) to quickly and accurately identify whether the generator is in a normal state or a specific fault state, which is crucial for timely maintenance and reducing downtime. Through continuous training and learning, the fault diagnosis algorithm module can be adjusted for the specific operating mode and fault mode of the generator, thereby improving the accuracy of diagnosis. It can not only diagnose faults that have occurred, but also predict potential faults by analyzing the development trend of fault characteristics, providing support for preventive maintenance.

[0023] Furthermore, by optimizing algorithm parameters and selecting appropriate kernel functions, the fault diagnosis algorithm module of the present invention can effectively reduce misdiagnosis and missed diagnosis and improve the reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flow chart of the present invention; Figure 2 is a system diagram of the present invention; Figure 3 This is a system diagram of Example 4. DETAILED DESCRIPTION

[0025] In order to further understand the content of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0026] Embodiment 1: See also Figure 1 , a method for automatic detection and reporting of generator faults, comprising the following steps: S1, collects the continuous time signals of each sensor of the generator, and converts the continuous time signals into discrete time signals through sampling conversion.

[0027] S2, preprocesses the discrete time signal to eliminate high-frequency noise and low-frequency drift.

[0028] S3, performing fast Fourier transform on the preprocessed discrete-time signal to convert the preprocessed discrete-time signal from the time domain to the frequency domain.

[0029] S4, calculating the required eigenvalues ​​in the frequency domain, and generating the required report based on the eigenvalues.

[0030] S5, extracting a fault feature vector from the required feature values, converting the fault feature vector into a data set, and dividing the data set into training samples of a normal state and different fault states.

[0031] S6, constructing an SVM model, using different training samples to train the SVM model, and obtaining a trained SVM model after verification.

[0032] S7, input all the characteristic values ​​into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

[0033] Embodiment 2: See also Figure 2 , generator fault automatic detection and reporting system, including: The sampling conversion module is used to collect the continuous time signals of each sensor of the generator and convert the continuous time signals into discrete time signals through sampling conversion; A preprocessing module is used to preprocess discrete time signals to eliminate high-frequency noise and low-frequency drift; A fast Fourier transform module is used to perform fast Fourier transform on the preprocessed discrete time signal, so as to convert the preprocessed discrete time signal from the time domain to the frequency domain; A report generation module is used to calculate the required eigenvalues ​​in the frequency domain and generate the required report based on the eigenvalues; A feature vector extraction module is used to extract a fault feature vector from the required feature values, convert the fault feature vector into a data set, and divide the data set into training samples of a normal state and different fault states; The model building module is used to build the SVM model. Different training samples are used to train the SVM model. After verification, the trained SVM model is obtained. The model output module is used to input all feature values ​​into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

[0034] Embodiment 3: The method for automatically detecting and reporting generator faults comprises the following steps: S1, collects the continuous time signals of each sensor of the generator, and converts the continuous time signals into discrete time signals through sampling conversion.

[0035] S2, preprocesses the discrete time signal to eliminate high-frequency noise and low-frequency drift.

[0036] S3, performing fast Fourier transform on the preprocessed discrete-time signal to convert the preprocessed discrete-time signal from the time domain to the frequency domain.

[0037] The method of fast Fourier transform is as follows:

[0038] in, is the discrete-time signal in the frequency domain complex frequency components, In the time domain Sample values, is the total number of sampling points, is an imaginary unit, is the frequency index, ranging from 0 to .

[0039] S4, calculate the required eigenvalues ​​in the frequency domain, and generate the required report based on the eigenvalues. The required eigenvalues ​​in the frequency domain include frequency peak, spectrum energy, and frequency bandwidth. The spectrum energy is calculated as follows:

[0040]

[0041] in, is the spectrum energy, For the The magnitude of the complex frequency components, and Respectively Complex frequency components The real and imaginary parts of .

[0042] S5, extracting a fault feature vector from the required feature values, converting the fault feature vector into a data set, and dividing the data set into training samples of a normal state and different fault states.

[0043] S6, construct the SVM model, use different training samples to train the SVM model, and after verification, obtain the trained SVM model. The calculation method of the trained SVM model is as follows:

[0044] The constraints are:

[0045] in, For the The Lagrange multiplier of samples, , is the penalty parameter, For the The Lagrange multiplier of samples, For the The class label of each sample (+1 or -1), For the The class labels of samples, For the The feature vector of the sample For the The feature vector of the samples, is the sample size, is the kernel function.

[0046] The decision function is:

[0047] in, is the decision function, is the bias term.

[0048] S7, input all the characteristic values ​​into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

[0049] Through the above method, the SVM algorithm can learn a decision boundary from the feature vector to distinguish between normal and fault states. In practical applications, the performance of SVM depends largely on the selection of kernel functions and the optimization of model parameters.

[0050] During the operation of the generator, an abnormal vibration pattern is detected, indicating that the bearing may be faulty. Subsequently, the fault database is triggered, and the detailed information of the fault event is first recorded in the database through an SQL insert statement. The specific contents include: inserting a new record in the FaultRecords table, in which the Timestamp field records the exact time when the fault occurred, the MachineID field identifies the number of the generator with the fault, the FaultType field is set to "bearing fault", the Description field provides a detailed description of the fault, such as "Vibration analysis shows signs of wear on the bearing", and the Severity field is set to 2 according to the abnormality of the vibration amplitude (assuming that the severity is divided into levels 1 to 5). At the same time, the fault database also supports query functions. For example, a SQL query statement can be executed to retrieve all records marked as "bearing fault" in the past month for trend analysis and maintenance planning. In addition, the database can also update fault records. For example, after the bearing is replaced, the operator can use a SQL update statement to change the Severity field of the relevant record to 0 and update the Description field to "The bearing has been replaced and the fault has been resolved." In this way, the fault database effectively manages the storage, retrieval and update of fault data, providing valuable information for generator maintenance and reliability analysis.

[0051] The fault database provides the generator fault automatic detection and reporting system with data storage, management and analysis capabilities. The fault database can record in detail all relevant information of each fault event, including timestamp, machine ID, fault type, description and severity, etc., to facilitate historical tracking and trend analysis. By storing a large amount of fault data, the fault database provides a valuable information resource for technical analysis and research, which helps to deeply understand the failure modes and causes of the generator. The data in the fault database can be used to formulate and optimize maintenance plans. By analyzing the frequency and type of faults, preventive maintenance can be carried out in a more targeted manner. The fault database supports fast data retrieval and update operations, which means that when a fault occurs, the present invention can quickly query historical data and update the current fault status, thereby speeding up the response speed of the entire system: the fault database can be integrated with other systems to realize data sharing and exchange, and enhance the synergy of the entire enterprise information system.

[0052] The front end of the present invention includes multiple vibration sensors and temperature sensors, which are respectively installed on the bearing part and stator winding of the generator. When the generator is running, the vibration sensor captures the vibration signal of the bearing, while the temperature sensor monitors the temperature change of the bearing and winding. These signals are further processed after being amplified, filtered and converted by the signal conditioning circuit. For example, in one test, the vibration sensor detected an abnormal vibration peak in the bearing part, and the temperature sensor showed that the temperature in the area had increased. These data were then recorded and used for subsequent fault analysis.

[0053] When it is determined that the vibration and temperature data are outside the preset normal range and a possible bearing fault is identified, an alarm is sounded through the sound and light alarm, and a red warning message is displayed on the operation interface in the control room, such as "Bearing fault alarm: machine number 1234, abnormal bearing vibration, please check." In addition, notifications can be sent to the maintenance team via SMS or email to ensure that maintenance measures are taken in a timely manner, thereby avoiding downtime losses caused by bearing failure.

[0054] In the case of a bearing failure, once the fault is confirmed, a report is automatically created containing the following content: "Bearing Fault Report: Machine No. 1234, bearing fault detected on a certain date and time. Fault Description: Vibration analysis showed signs of bearing wear and increased temperature. Severity: Level 2 (Moderate). Recommended Action: Check immediately, bearing replacement may be required." Once the report is generated, it can be emailed, printed, or stored in a database for historical records and subsequent analysis.

[0055] Once a fault is diagnosed, the diagnostic results and alarm information will be sent. In addition, it is also responsible for transmitting the final fault report and real-time data to the remote monitoring center through a secure network connection, for example, using a VPN or a dedicated network to encrypt the data and send it to the enterprise resource planning (ERP) system or cloud server, so that managers can access and analyze the data at any location.

[0056] In the case of a bearing failure, a touchscreen control panel and a graphical user interface (GUI) are also included. Operators can see real-time data displays on the GUI, such as vibration waveforms, temperature trend graphs, and system health status indicators. When a bearing failure occurs, the GUI will pop up an alarm window with detailed fault information and recommended operating procedures. Operators can use the interface to confirm receipt of the alarm, initiate an emergency shutdown procedure, or manually generate a fault report. They can also use the interface to adjust sensor parameters, view historical data, and perform system self-test procedures to ensure system reliability and continuous operation.

[0057] Embodiment 4: See also Figure 3As shown, the present invention also provides an electronic device 100 for a method for automatically detecting and reporting generator faults; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0058] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the method for automatic detection and reporting of generator faults described in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application 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 (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 can include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0059] The at least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.

[0060] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a method for automatically detecting and reporting a generator fault, and the processor 102 can execute the plurality of instructions to implement: Collect the continuous time signals of each sensor of the generator, and convert the continuous time signals into discrete time signals through sampling conversion; Preprocess discrete time signals to remove high frequency noise and low frequency drift; Performing fast Fourier transform on the preprocessed discrete time signal to convert the preprocessed discrete time signal from the time domain to the frequency domain; Calculate the required eigenvalues ​​in the frequency domain and generate the required report based on the eigenvalues; Extracting a fault feature vector from the required feature values, converting the fault feature vector into a data set, and dividing the data set into training samples of a normal state and different fault states; Construct an SVM model, use different training samples to train the SVM model, and after verification, obtain a trained SVM model; All feature values ​​are input into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

[0061] Example 5 If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0062] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0063] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0064] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0065] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for automatically detecting and reporting generator faults, characterized in that: The following steps are involved: Collect the continuous time signals of each sensor of the generator, and convert the continuous time signals into discrete time signals through sampling conversion; Preprocess discrete time signals to remove high frequency noise and low frequency drift; Performing fast Fourier transform on the preprocessed discrete time signal to convert the preprocessed discrete time signal from the time domain to the frequency domain; Calculate the required eigenvalues ​​in the frequency domain and generate the required report based on the eigenvalues; Extracting a fault feature vector from the required feature values, converting the fault feature vector into a data set, and dividing the data set into training samples of a normal state and different fault states; Construct an SVM model, use different training samples to train the SVM model, and after verification, obtain a trained SVM model; All feature values ​​are input into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

2. The method for automatic detection and reporting of generator faults according to claim 1, characterized in that: When preprocessing the discrete-time signal, the discrete-time signal is first denoised, and then the denoised discrete-time signal is filtered.

3. The method for automatic detection and reporting of generator faults according to claim 1, characterized in that: The method for performing fast Fourier transform on the preprocessed discrete-time signal is as follows: in, is the discrete-time signal in the frequency domain complex frequency components, In the time domain Sample values, is the total number of sampling points, is an imaginary unit, is the frequency index, ranging from 0 to .

4. The method for automatic detection and reporting of generator faults according to claim 1, characterized in that: The desired characteristic values ​​in the frequency domain include frequency peak, spectrum energy, and frequency bandwidth.

5. The method for automatic detection and reporting of generator faults according to claim 4, characterized in that: The spectrum energy is calculated as follows: in, is the spectrum energy, For the The magnitude of the complex frequency components, and Respectively Complex frequency components The real and imaginary parts of .

6. The method for automatic detection and reporting of generator faults according to claim 1, characterized in that: The calculation method of the trained SVM model is as follows: The constraints are: in, For the The Lagrange multiplier of samples, , is the penalty parameter, For the The Lagrange multiplier of samples, For the The class labels of samples, For the The class labels of samples, For the The feature vector of the sample For the The feature vector of the samples, is the sample size, is the kernel function.

7. The method for automatic detection and reporting of generator faults according to claim 6, characterized in that: The decision function of the trained SVM model is: in, is the decision function, is the bias term.

8. Generator fault automatic detection and reporting system, characterized by: include: The sampling conversion module is used to collect the continuous time signals of each sensor of the generator and convert the continuous time signals into discrete time signals through sampling conversion; A preprocessing module is used to preprocess discrete time signals to eliminate high-frequency noise and low-frequency drift; A fast Fourier transform module is used to perform fast Fourier transform on the preprocessed discrete time signal, so as to convert the preprocessed discrete time signal from the time domain to the frequency domain; A report generation module is used to calculate the required eigenvalues ​​in the frequency domain and generate the required report based on the eigenvalues; A feature vector extraction module is used to extract a fault feature vector from the required feature values, convert the fault feature vector into a data set, and divide the data set into training samples of a normal state and different fault states; The model building module is used to build the SVM model. Different training samples are used to train the SVM model. After verification, the trained SVM model is obtained. The model output module is used to input all feature values ​​into the trained SVM model to obtain whether the generator is in a normal working state or a specific fault state.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for automatic detection and reporting of generator faults according to any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for automatic detection and reporting of generator faults according to any one of claims 1 to 7 are implemented.

Images