Detection method and device for vibration state of power equipment, equipment and medium

By training and classifying the feature parameters of vibration signals of power equipment using a decision tree classification model, the problem of lag in abnormal vibration detection of power equipment is solved, and more accurate vibration state detection is achieved.

CN118960942APending Publication Date: 2024-11-15SICHUAN POWER EHV OVERHAUL
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411031404.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Existing technologies for detecting abnormal vibrations in power equipment are not precise enough, making it difficult to distinguish between vibrations generated during normal operation and those generated during abnormal operation, leading to delays in troubleshooting.

Method used

A vibration detection method based on a decision tree classification model is adopted. The mechanical vibration signal of the power equipment is acquired by the sensor. The decision tree classification model is constructed by feature parameter training and envelope analysis to classify and regress the electrical signal and output the vibration state detection result.

Benefits of technology

It improves the accuracy of vibration status detection for power equipment, enables timely detection of abnormal vibrations, and avoids delays in fault diagnosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118960942A_ABST
    Figure CN118960942A_ABST
Patent Text Reader

Abstract

The invention discloses a power equipment vibration state detection method and device, equipment and a medium, and relates to the technical field of power equipment detection, and the technical scheme is characterized in that an electric signal generated by mechanical vibration of to-be-detected power equipment is acquired; inputting the electric signal into a vibration detection model for classification and regression so as to output a detection result of the vibration state of the to-be-detected power equipment; wherein the vibration detection model is a decision tree classification model obtained by training on the basis of characteristic parameters of a historical electric signal set of to-be-detected power equipment, and the historical electric signal set comprises historical electric signals generated by normal vibration and abnormal vibration of the to-be-detected power equipment; the characteristic parameters comprise frequencies, amplitudes and phases of historical electric signals of the to-be-detected power equipment under normal vibration and abnormal vibration. The problem of hysteresis caused by manual detection of abnormal vibration of the power equipment is solved, and the accuracy of vibration state detection of the power equipment is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of power equipment detection, and in particular to a method, device, equipment and medium for detecting the vibration state of power equipment. Background Art

[0002] Power equipment is running in the entire power system. After working for a long time, some abnormal conditions or faults are inevitable. Timely discovery of these abnormal conditions or faults is conducive to reducing the damage to the power equipment. Timely repair and troubleshooting can avoid major power accidents caused by power equipment failures and help maintain the stable operation of the power system. Abnormal vibration of power equipment is a significant feature of power equipment in an abnormal state or fault state.

[0003] At present, the detection of abnormal vibration of power equipment is not accurate enough, and it is difficult to distinguish the difference between the vibration generated by normal operation of the equipment and the vibration generated by abnormal operation. The method of regular inspection at the power equipment operation site can only play a certain preventive role. During the period from abnormal vibration to damage of the power equipment, it is difficult to detect the vibration caused by the working environment noise and abnormal or faulty operation of the power equipment through manual inspection. Therefore, the result of manual inspection is that the power equipment is not discovered until a certain period of time after the failure, resulting in a lag in the troubleshooting of the power equipment. Summary of the invention

[0004] The present application provides a method, device, equipment and medium for detecting the vibration state of electric equipment, which solves the problem of hysteresis caused by manual detection of abnormal vibration of electric equipment and improves the accuracy of vibration state detection of electric equipment.

[0005] In a first aspect of the present application, a method for detecting a vibration state of an electric device is provided, the method comprising:

[0006] Acquire the electrical signal generated by mechanical vibration of the electrical equipment to be tested;

[0007] Inputting the electrical signal into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected;

[0008] Among them, the vibration detection model is a decision tree classification model trained based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected, wherein the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

[0009] In one implementation, the acquired electrical signal is converted by a sensor having a piezoelectric ceramic material installed on the surface of the electrical equipment to be detected.

[0010] In one implementation, before the electrical signal to be detected is input into a vibration detection model for classification and regression to output the detection result of the vibration state of the electrical equipment to be detected, the method also includes: filtering out the interference signal present in the electrical signal based on the characteristic parameters of the electrical signal of the electrical equipment to be detected under normal vibration and abnormal vibration.

[0011] In one implementation, the characteristic parameters of the historical electrical signal set are obtained by analyzing the historical electrical signal set using an envelope analysis method.

[0012] In one implementation, the training process of the vibration detection model includes:

[0013] Acquire the historical electrical signal set of the power equipment to be detected;

[0014] Analyze the historical electrical signals in the historical electrical signal set based on the envelope analysis method to determine the characteristic parameters of the electrical signals of the power equipment to be detected under normal vibration and abnormal vibration, wherein the characteristic parameters include frequency, amplitude and phase;

[0015] A decision tree classification model is constructed based on a decision tree classification algorithm, and the decision tree classification model is iteratively trained based on characteristic parameters of the electrical signal of the power equipment to be detected under normal vibration and abnormal vibration to obtain the vibration detection model.

[0016] In one implementation, the method further includes:

[0017] Classifying the electrical signal based on the detection result of the vibration state of the power equipment to be detected, and determining the vibration state corresponding to the electrical signal, wherein the vibration state refers to normal vibration and abnormal vibration of the power equipment to be detected;

[0018] Based on the vibration state, the historical electrical signals of the historical signal set are updated to obtain an updated historical electrical signal set of the power equipment to be detected;

[0019] The decision tree model is iteratively trained based on the updated historical electrical signal set.

[0020] In one implementation, the method further includes:

[0021] Displaying on a display device a waveform diagram of the frequency, amplitude and phase of the electrical signal generated by mechanical vibration of the power device to be detected, and displaying the vibration state of the power device to be detected;

[0022] When it is determined that the vibration state of the power equipment to be detected is abnormal vibration, an early warning signal is generated, and the early warning signal is used to prompt that the power equipment to be detected has an abnormality that needs to be processed.

[0023] In a second aspect of the present application, a device for detecting the vibration state of an electric device is provided, the device comprising:

[0024] A signal acquisition module, used to acquire an electrical signal generated by mechanical vibration of the power equipment to be detected;

[0025] A vibration detection module is used to input the electrical signal into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on the characteristic parameters of a historical electrical signal set of the power equipment to be detected, the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

[0026] In a third aspect of the present application, a device for detecting a vibration state of an electric power device is provided, comprising:

[0027] The sensor is located on the power equipment to be detected and is used to obtain the electrical signal generated by the mechanical vibration of the power equipment to be detected;

[0028] The processor inputs the electrical signal into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on the characteristic parameters of a historical electrical signal set of the power equipment to be detected, the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

[0029] In a fourth aspect of the present application, a storage medium is provided, wherein the storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the method for detecting the vibration state of the electric power equipment provided in the present application.

[0030] Compared with the prior art, this application has the following beneficial effects:

[0031] In the present application, the electrical signal generated by the mechanical vibration of the power equipment to be detected is first obtained, and then the electrical signal is input into the vibration detection model for classification and regression, so that the detection result of the vibration state of the power equipment to be detected can be output. The vibration detection model is a decision tree classification model trained based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected. In the present application scheme, the decision tree classification model trained based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected classifies and regresses the electrical signal, and outputs the detection result of the vibration state. Specifically, the characteristic parameters refer to the differences and characteristics of the historical electrical signals under normal vibration and abnormal vibration in frequency, amplitude and phase, so that the accuracy of the detection result can be improved, thereby solving the problem of hysteresis caused by manual detection of abnormal vibration of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings described herein are used to provide a further understanding of the embodiments of the present application, constitute a part of the present application, and do not constitute a limitation on the embodiments of the present application. In the drawings:

[0033] Figure 1 A schematic diagram of the structure of a device for detecting the vibration state of electric power equipment provided in an embodiment of the application;

[0034] Figure 2 A schematic diagram of a flow chart of a method for detecting a vibration state of an electric power device provided in an embodiment of the present application;

[0035] Figure 3 A principle block diagram of a device for detecting the vibration state of electric power equipment provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of this application more clear, the present application is further described in detail below in conjunction with the embodiments and drawings. The illustrative implementation methods and descriptions of this application are only used to explain this application and are not intended to limit this application. Secondly, based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0037] The method of regular inspection at the power equipment operation site can only play a certain preventive role. During the period from abnormal vibration to damage of the power equipment, it is difficult to detect the vibration caused by the working environment noise and abnormal or faulty operation of the power equipment through manual inspection. Therefore, the result of manual inspection is that the power equipment is not discovered until a certain period of time has passed, resulting in a lag in the troubleshooting of the power equipment.

[0038] The present application embodiment first provides a device for detecting the vibration state of an electric device, such as Figure 1As shown, the detection device 100 at least includes: a sensor 101 and a processor 102;

[0039] Wherein, the sensor 101 is located on the power equipment to be detected and is used to obtain the electrical signal generated by the mechanical vibration of the power equipment to be detected;

[0040] The processor 102 inputs the electrical signal into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on feature parameters of a historical electrical signal set of the power equipment to be detected, wherein the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the feature parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

[0041] It is understandable that there can be multiple sensors 101, that is, each sensor 101 is installed on the surface of a type of power equipment, such as any type of power equipment such as transformers, insulators, transformer oil tanks, etc., so that signal collection of all power equipment of the same type / different types in a target area can be achieved, which can improve the detection efficiency of the power equipment operation status detection. Secondly, the communication method between the sensor 101 and the processor 102 can be based on a communication bus connection or on a wireless communication method. This is common knowledge, so only a brief explanation is given.

[0042] Secondly, it should be noted that the sensor is a passive micro-vibration signal sensor based on magnetostrictive material, that is, the sensor 101 is made of piezoelectric ceramic material based on the magnetostrictive principle. This part is installed on the surface of the electrical equipment to be detected. This is a relatively conventional technical means at present. The working principle of the sensor 101 is: as the electrical equipment to be detected vibrates, the sensor 101 uses the piezoelectric effect. The polar molecules in the piezoelectric material will deform when subjected to force, resulting in the rearrangement of internal charges. Therefore, a potential difference signal is generated on the surface of the piezoelectric ceramic material. The potential difference of the generated potential difference signal is proportional to the amplitude and frequency of the vibration of the electrical equipment to be detected. By converting this potential difference signal, an electrical signal generated by mechanical vibration can be obtained.

[0043] In this embodiment, the detection device first obtains the electrical signal generated by the mechanical vibration of the power equipment to be detected, and then inputs the electrical signal into the vibration detection model for classification and regression, so as to output the detection result of the vibration state of the power equipment to be detected. The vibration detection model is a decision tree classification model trained based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected. In the present application scheme, the decision tree classification model trained based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected classifies and regresses the electrical signal, and outputs the detection result of the vibration state. Specifically, the characteristic parameters refer to the differences and characteristics of the historical electrical signals under normal vibration and abnormal vibration in frequency, amplitude and phase, so that the accuracy of the detection result can be improved, thereby solving the problem of hysteresis caused by manual detection of abnormal vibration of power equipment.

[0044] Based on the above-mentioned detection device for the vibration state of the electric equipment, an embodiment of the present application also provides a detection method for the vibration state of the electric equipment. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] Please refer to Figure 2 , Figure 2 A flow chart of a method for detecting the vibration state of an electric device provided in an embodiment of the present application is shown as follows: Figure 2 As shown, the method includes:

[0046] S201, obtaining an electrical signal generated by mechanical vibration of the power equipment to be detected.

[0047] In this embodiment, the power equipment to be detected refers to power equipment of the same type and specification, such as a power transformer used in a 500 kV transmission line, and another example is a power transformer used in a 110 kV transmission line. When acquiring its electrical signal, only electrical signals of the same type and specification are acquired, and then the electrical signal is input into the vibration detection model given in the following steps. As the power equipment to be detected vibrates, the sensor with piezoelectric ceramic material will generate a potential difference signal on the surface of the piezoelectric ceramic material according to the piezoelectric effect. The polar molecules in the piezoelectric material will deform when subjected to force, resulting in the rearrangement of the internal charge. The potential difference of the generated potential difference signal is proportional to the amplitude and frequency of the vibration of the power equipment to be detected. By converting this potential difference signal, the electrical signal generated by the mechanical vibration can be obtained.

[0048] S202, inputting the electrical signal into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on feature parameters of a historical electrical signal set of the power equipment to be detected, the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the feature parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

[0049] In this embodiment, in combination with the contents described in the above embodiments, the acquired electrical signal is input into the corresponding vibration detection model according to the type and specifications of the electrical equipment to be detected, and the detection results of the vibration state of the electrical equipment to be detected are output, and the detection results are divided into normal vibration and abnormal vibration.

[0050] Specifically, the vibration detection model is a decision tree classification model trained based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected. Before training and learning, the characteristic parameters of the electrical signals of the power equipment of the same type and specification as the power equipment to be detected under normal vibration and abnormal vibration are collected to construct a historical electrical signal set, which includes two groups of parameters, normal and abnormal, such as 450 groups of historical electrical signals under normal vibration and 50 groups of historical electrical signals under abnormal vibration.

[0051] More specifically, the converted historical electrical signals are processed by the envelope analysis method to obtain the differences and characteristics of the historical electrical signals in frequency, amplitude and phase under normal vibration state and abnormal vibration state, thereby obtaining the characteristic parameters under the two vibration states.

[0052] In some embodiments, the characteristic parameters of the historical electrical signal set are obtained by analyzing the historical electrical signal set using an envelope analysis method.

[0053] Specifically, the envelope analysis method includes more vibration signal feature analysis than the traditional analysis method. This method requires the analytical signal of the original signal to extract the vibration characteristics of the original system. The analytical signal is a complex-valued function without negative frequency components, so the Fourier transform S(f) of the time domain signal S(t) has Hermitian symmetry at the zero frequency axis, that is, S(-f) = S(f) * ,S(f) * is the complex conjugate of S(f).

[0054] Define the parsing function as: In formula (1), u(f) is the unit step function, sgn(f) is the sign function, the analytical function only contains the non-negative frequency components of S(f), and due to the Hermitian symmetry of S(f), this operation is reversible.

[0055] S a The analytical signal of (t) can be expressed as S a (f) is driven by the inverse Fourier transform. Specifically, S a The expression of (t) is: In formula (3), is the Hilbert transform, * is the convolution factor, and j is the imaginary unit operator. m (t)=|S a (t)| is the instantaneous amplitude or envelope, θ(t) = arg[S a (t)] is the instantaneous phase. After the original signal undergoes the above analytical signal modulation process, a more obviously differentiated characteristic parameter can be obtained, so the characteristic parameter can be used as a basis for judging abnormal vibration.

[0056] On the basis of the envelope analysis method described in the above embodiment, this embodiment constructs a decision tree classification algorithm, and iteratively trains the decision tree classification algorithm based on the above feature parameters. When the training round is reached, the vibration detection model can be obtained. Specifically, the decision tree classification algorithm is a commonly used supervised learning algorithm, which represents the mapping of the feature / input space based on the training data to various classes or continuous values. This learning method uses a decision tree as a prediction model. There are two basic types of decision trees: classification and regression. In a classification tree, the output is a finite set of classes, while in a regression tree, the output is a continuous value. There are different algorithms for making decision trees. In a decision tree, each node defines a threshold or some binary yes / no criteria to classify a specific feature into a class. A node can be considered as a series of questions. When moving from the root of the tree to its branches, it can be considered as a set of questions and narrows the feature space to a specific class representation. A major advantage of the decision tree algorithm is that it implicitly performs feature selection. The most important feature with the highest predictive power goes to the root, the next highest predictive feature goes to its child feature, and so on. Furthermore, if not managed properly, decision tree learners may create overly complex trees and overfitting of the data may occur. The simplified steps of a decision tree classification algorithm are as follows:

[0057] In the loop:

[0058] 1. Select the best attribute / feature (e.g., Y).

[0059] 2. Assign Y as the decision attribute of the node.

[0060] 3. For each value of Y, create a descendant of node.

[0061] 4. Sort the training examples into leaves.

[0062] 5. If the sample is classified correctly, then stop.

[0063] 6.Else traverse the leaves.

[0064] After the cycle of steps 1-6 above, the decision tree analysis algorithm can effectively select important features and non-important features for detecting abnormal vibration. Initially, some data that have been detected and collected are input into the decision tree analysis algorithm for preliminary learning. As the detection proceeds, the detection parameters are continuously learned and deepened, and the detection accuracy will gradually improve.

[0065] In some embodiments, the training process of the vibration detection model includes: obtaining the historical electrical signal set of the electrical equipment to be detected; analyzing the historical electrical signals in the historical electrical signal set based on the envelope analysis method, and determining the characteristic parameters of the electrical signals of the electrical equipment to be detected under normal vibration and abnormal vibration, wherein the characteristic parameters include frequency, amplitude and phase; constructing a decision tree classification model based on a decision tree classification algorithm, and iteratively training the decision tree classification model based on the characteristic parameters of the electrical signals of the electrical equipment to be detected under normal vibration and abnormal vibration to obtain the vibration detection model.

[0066] In this embodiment, the iterative training method of the decision tree classification algorithm based on characteristic parameters is a conventional technical means of those skilled in the art. This embodiment only provides the following overview. Specifically, the characteristic parameters of the historical electrical signal set of the power equipment of the same type and specification as the power equipment to be detected are trained and learned to obtain two sets of unused parameters to determine the state of the electrical equipment. At the same time, the final judgment result will be fed back to this part, and the corresponding data records will be saved for learning and further optimized for classification. Secondly, the iteration and stop of the algorithm training to obtain the final vibration detection model, the loss function used is a conventional technical means as a decision tree classification algorithm, so this embodiment will not explain it.

[0067] In some embodiments, before executing step S202, the method provided in this embodiment further includes: filtering out interference signals existing in the electrical signals according to characteristic parameters of the electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

[0068] Specifically, according to the signal characteristics of the historical electrical signal in the normal vibration state and the abnormal vibration state, a digital passband filter is designed to remove the signals of other frequencies from environmental interference and the like except the two, so as to improve the accuracy of subsequent judgments. Furthermore, the digital filter has the characteristics of precise filter design, easy implementation in a microprocessor, and modifiable filter characteristics. In the proposed sensor based on magnetostrictive materials, monitoring needs to be carried out in real time, and thus a high-performance microprocessor is required. Sufficient time is reserved between the analysis steps, and the analysis is usually performed every hour. This embodiment provides a wireless pulse response filter, which has the advantage of requiring fewer orders than a finite pulse filter. This filter has a nonlinear phase characteristic and is suitable for online filtering applications with sufficient time intervals. The analytical signal passes through this filter, and other signals irrelevant to the vibration state are greatly attenuated.

[0069] In other words, this embodiment combines the signal characteristics of the normal vibration state and the abnormal vibration state obtained by analyzing the historical electrical signals, designs a digital passband filter, and removes signals of other frequencies other than the normal vibration state and the abnormal vibration state, so as to improve the accuracy of subsequent vibration state detection.

[0070] In some embodiments, the detection method provided in this embodiment also includes: classifying the electrical signal based on the detection result of the vibration state of the power equipment to be detected, and determining the vibration state corresponding to the electrical signal, wherein the vibration state refers to the normal vibration and abnormal vibration of the power equipment to be detected; updating the historical electrical signals of the historical signal set based on the vibration state to obtain an updated historical electrical signal set of the power equipment to be detected; and iteratively training the decision tree model based on the updated historical electrical signal set.

[0071] In this embodiment, since the historical electrical signal set is trained with characteristic parameters corresponding to defined normal vibrations and abnormal vibrations, it cannot well characterize the characteristic parameters in the electrical signals collected by the power equipment to be detected working in an actual environment. Therefore, this embodiment adds the corresponding electrical signals to the historical electrical signal set based on the vibration state corresponding to the electrical signals, thereby updating the historical electrical signal set, thereby enriching the characteristic parameters of the historical electrical signal set. The decision tree model is iteratively trained again based on the updated historical electrical signal set, which can continuously learn and deepen the detection parameters, and the detection accuracy will gradually improve.

[0072] In some embodiments, the detection method provided in this embodiment also includes: displaying a waveform diagram of the frequency, amplitude and phase of the electrical signal generated by mechanical vibration of the electrical equipment to be detected in a display device, and displaying the vibration state of the electrical equipment to be detected; when it is determined that the vibration state of the electrical equipment to be detected is abnormal vibration, generating early warning information, and the early warning signal is used to prompt that the electrical equipment to be detected has an abnormality that needs to be processed.

[0073] Specifically, in the above Figure 1 On the basis of what is shown, the detection device can also be connected to a display device, such as a monitor, oscilloscope, etc., to display a waveform diagram of the frequency, amplitude and phase of the electrical signal generated by the mechanical vibration of the electrical equipment to be detected in the display device, so as to facilitate the detection personnel to better observe the changes in characteristic parameters during the mechanical vibration process. In addition, early warning information can be generated based on the detection results to prompt the detection personnel to repair or replace the electrical equipment to be detected.

[0074] Please refer to Figure 3 , Figure 3 A schematic diagram of a device for detecting the vibration state of an electric device provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the device comprises:

[0075] The signal acquisition module 301 is used to acquire the electrical signal generated by the mechanical vibration of the power equipment to be detected;

[0076] The vibration detection module 302 is used to input the electrical signal into a vibration detection model for classification and regression to output the detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected, the historical electrical signal set includes the historical electrical signals generated by the normal vibration and abnormal vibration of the power equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

[0077] In this embodiment, the detection device first obtains the electrical signal generated by the mechanical vibration of the power equipment to be detected, and then inputs the electrical signal into the vibration detection model for classification and regression, so as to output the detection result of the vibration state of the power equipment to be detected. The vibration detection model is a decision tree classification model obtained by training based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected. In the present application scheme, the decision tree classification model obtained by training based on the characteristic parameters of the historical electrical signal set of the power equipment to be detected classifies and regresses the electrical signal, and outputs the detection result of the vibration state. Specifically, the characteristic parameters refer to the differences and characteristics of the historical electrical signals under normal vibration and abnormal vibration in frequency, amplitude and phase, so that the accuracy of the detection result can be improved, thereby solving the problem of hysteresis caused by manual detection of abnormal vibration of power equipment.

[0078] It should be noted that the modules in the device for detecting the vibration state of the electric equipment in the embodiment of the present application correspond one by one to the various implementation steps of the method for detecting the vibration state of the electric equipment provided in the above embodiment. Since the above description has been carried out in detail, some details not reflected in this embodiment can be referred to the above part, and will not be elaborated here.

[0079] An embodiment of the present application further provides a storage medium, which includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the method for detecting the vibration state of the power equipment provided in the present application.

[0080] According to an embodiment of the present application, a processor is also provided, which is used to run a program, wherein when the program is running, Figure 2 The detection method of the vibration state of the electric equipment shown. Optionally, the following steps are executed when the program is running: obtaining the electric signal generated by the mechanical vibration of the electric equipment to be detected; inputting the electric signal into the vibration detection model for classification and regression to output the detection result of the vibration state of the electric equipment to be detected; wherein the vibration detection model is a decision tree classification model obtained by training based on the characteristic parameters of the historical electric signal set of the electric equipment to be detected, the historical electric signal set includes the historical electric signals generated by the normal vibration and abnormal vibration of the electric equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electric signals of the electric equipment to be detected under normal vibration and abnormal vibration.

[0081] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0082] In the above-mentioned embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided in the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0083] 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 over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0084] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0085] If the integrated unit 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 technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc. Various media that can store program codes.

[0086] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for detecting the vibration state of electric power equipment, characterized in that: Methods include: Acquire the electrical signal generated by mechanical vibration of the electrical equipment to be tested; The electrical signal is input into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on the characteristic parameters of a historical electrical signal set of the power equipment to be detected, the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

2. The method according to claim 1, characterized in that The acquired electrical signal is converted by a sensor made of piezoelectric ceramic material installed on the surface of the electrical equipment to be detected.

3. The method according to claim 2, characterized in that Before inputting the electrical signal to be detected into the vibration detection model for classification and regression to output the detection result of the vibration state of the electrical equipment to be detected, the method also includes: filtering out the interference signal present in the electrical signal according to the characteristic parameters of the electrical signal of the electrical equipment to be detected under normal vibration and abnormal vibration.

4. The method according to claim 1, characterized in that The characteristic parameters of the historical electrical signal set are obtained by analyzing the historical electrical signal set using an envelope analysis method.

5. The method according to claim 1, characterized in that The training process of the vibration detection model includes: Acquire the historical electrical signal set of the power equipment to be detected; Analyze the historical electrical signals in the historical electrical signal set based on the envelope analysis method to determine the characteristic parameters of the electrical signals of the power equipment to be detected under normal vibration and abnormal vibration, wherein the characteristic parameters include frequency, amplitude and phase; A decision tree classification model is constructed based on a decision tree classification algorithm, and the decision tree classification model is iteratively trained based on characteristic parameters of electrical signals of the power equipment to be detected under normal vibration and abnormal vibration to obtain the vibration detection model.

6. The method according to claim 1, characterized in that The method further comprises: Classifying the electrical signal based on the detection result of the vibration state of the power equipment to be detected, and determining the vibration state corresponding to the electrical signal, wherein the vibration state refers to normal vibration and abnormal vibration of the power equipment to be detected; Based on the vibration state, the historical electrical signals of the historical signal set are updated to obtain an updated historical electrical signal set of the power equipment to be detected; The decision tree model is iteratively trained based on the updated historical electrical signal set.

7. The method according to claim 1, characterized in that The method further comprises: Displaying on a display device a waveform diagram of the frequency, amplitude and phase of the electrical signal generated by mechanical vibration of the power device to be detected, and displaying the vibration state of the power device to be detected; When it is determined that the vibration state of the power equipment to be detected is abnormal vibration, an early warning signal is generated, and the early warning signal is used to prompt that the power equipment to be detected has an abnormality that needs to be processed.

8. A device for detecting the vibration state of electric power equipment, characterized in that: The device includes: A signal acquisition module, used to acquire an electrical signal generated by mechanical vibration of the power equipment to be detected; A vibration detection module is used to input the electrical signal into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on the characteristic parameters of a historical electrical signal set of the power equipment to be detected, the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

9. A device for detecting the vibration state of electric power equipment, characterized in that: include: The sensor is located on the power equipment to be detected and is used to obtain the electrical signal generated by the mechanical vibration of the power equipment to be detected; The processor inputs the electrical signal into a vibration detection model for classification and regression to output a detection result of the vibration state of the power equipment to be detected; wherein the vibration detection model is a decision tree classification model trained based on the characteristic parameters of a historical electrical signal set of the power equipment to be detected, the historical electrical signal set includes historical electrical signals generated by normal vibration and abnormal vibration of the power equipment to be detected, and the characteristic parameters include the frequency, amplitude and phase of the historical electrical signals of the power equipment to be detected under normal vibration and abnormal vibration.

10. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the method for detecting the vibration state of the electric power equipment according to any one of claims 1 to 7.

Citation Information

Cited By

  • Abnormality detection method for vibration signal of transmission system and related equipment

    CN121117813A

  • An abnormality detection method of a transmission system vibration signal and related device

    CN121117813B