Fault diagnosis method based on small sample data
Through small sample data interpolation and preprocessing, fault characteristic data is extracted, and fast and accurate fault diagnosis is achieved, and complex problems of large amounts of data acquisition and processing in the existing technology are solved, thereby improving fault diagnosis efficiency.
Patent Information
- Application Number
- CN202411874683.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-18
- Publication Date
- 2025-04-11
AI Technical Summary
Existing fault diagnosis methods require a large amount of raw data, resulting in complex data acquisition and processing, inefficient and inability to meet the requirements of fast, accurate and efficient fault diagnosis.
By obtaining small sample running data, inserting data worthy of large sample data, preprocessing and feature extraction, and using fault feature data for fault diagnosis.
It realizes fast, accurate and efficient troubleshooting required for only a small amount of raw data, reducing the requirements of data acquisition time and diagnostic efficiency.
Smart Images
Figure CN119807660A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of equipment fault diagnosis, and in particular to a fault diagnosis method based on small sample data. Background Art
[0002] Fault diagnosis generally requires the collection of a large amount of raw data and a series of processing of the raw data to complete the fault diagnosis. The need for a large amount of raw data will lead to complex algorithms such as data collection, data outlier elimination, data filtering, feature extraction, and pattern recognition, large amount of calculation, and low fault diagnosis efficiency, which cannot meet the modern requirements of fast, accurate, and efficient fault diagnosis. Summary of the invention
[0003] The purpose of this application is to provide a fault diagnosis method based on small sample data, which can achieve fast, accurate and efficient fault diagnosis with only a small amount of original equipment operation data.
[0004] To achieve the above objectives, this application provides the following solutions:
[0005] In a first aspect, the present application provides a fault diagnosis method based on small sample data, comprising:
[0006] Obtain a small sample of operating data for the equipment to be diagnosed;
[0007] Performing data interpolation on the small sample operation data to obtain large sample operation data;
[0008] Preprocessing the large sample operation data to obtain preprocessed large sample operation data;
[0009] Extracting fault feature data from the preprocessed large sample operation data;
[0010] Fault diagnosis is performed according to the fault characteristic data.
[0011] According to the specific embodiments provided in this application, this application discloses the following technical effects:
[0012] The present application provides a fault diagnosis method based on small sample data, the method comprising: obtaining small sample operation data of the equipment to be diagnosed; interpolating the small sample operation data to obtain large sample operation data; preprocessing the large sample operation data to obtain preprocessed large sample operation data; extracting fault feature data from the preprocessed large sample operation data; and performing fault diagnosis based on the fault feature data. The present invention obtains large sample data from small sample data, thereby realizing fault diagnosis, and only requires a small amount of raw data to complete the fault diagnosis of the equipment, which has the advantages of short raw data acquisition time and high fault diagnosis efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0014] Figure 1 This is an application environment diagram of a fault diagnosis method based on small sample data in an embodiment of the present application;
[0015] Figure 2 A flowchart of a fault diagnosis method based on small sample data provided in one embodiment of the present application;
[0016] Figure 3 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0018] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0019] The fault diagnosis method based on small sample data provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal communicates with the server through the network. The data storage system can store the data that the server needs to process. The data storage system can be set up separately, integrated on the server, or placed on the cloud or other servers. The terminal can send the small sample operation data of the equipment to be diagnosed to the server. After the server receives the small sample operation data of the equipment to be diagnosed, the server interpolates the small sample operation data to obtain the large sample operation data; pre-processes the large sample operation data to obtain the pre-processed large sample operation data; extracts the fault feature data from the pre-processed large sample operation data; and performs fault diagnosis according to the fault feature data. The server can feed back the obtained fault diagnosis results to the terminal. In addition, in some embodiments, the fault diagnosis method based on small sample data can also be implemented by a server or a terminal alone, such as the terminal can directly perform fault diagnosis processing on the small sample operation data of the equipment to be diagnosed, or the server can obtain the small sample operation data of the equipment to be diagnosed from the data storage system and perform fault diagnosis processing.
[0020] The terminals may be, but are not limited to, various desktop computers, laptops, smart phones, tablet computers, IoT devices and portable wearable devices. IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server may be implemented as an independent server or a server cluster consisting of multiple servers, or a cloud server.
[0021] In an exemplary embodiment, Figure 2 As shown, a fault diagnosis method based on small sample data is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or can be executed by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server in the example is used to illustrate, including the following steps 101 to 105. Among them:
[0022] Step 101: Obtain a small sample of operating data of the equipment to be diagnosed.
[0023] Step 102: interpolate the small sample operation data to obtain large sample operation data.
[0024] Step 103: preprocess the large sample operation data to obtain preprocessed large sample operation data.
[0025] Step 104: extracting fault feature data from the preprocessed large sample operation data.
[0026] Step 105: performing fault diagnosis according to the fault characteristic data.
[0027] Implement the above steps 101 to 105, and use small sample data to complete fault diagnosis. The transformation of small sample data to a large amount of original data is mainly completed by generating data, eliminating data errors, and injecting data features on small sample data, and then the fault diagnosis is completed by processing the original data. This method only requires a small amount of original data to complete equipment fault diagnosis, has the advantages of short original data collection time and high fault diagnosis efficiency, and is suitable for completing rapid fault diagnosis in wartime.
[0028] In another exemplary embodiment of the present application, small sample data is converted into a large sample: Assume that the collected small sample data can be represented by a data vector X, that is, X = [x1, x2, ... x N ], where N is the total number of sampling points of the collected small sample data. Step 1: Generate data error information. Based on prior data information, the collected data error generally follows a Gaussian distribution with a mean of 1 and a variance of 5. Based on this distribution information, 2N error information is generated using software programming. Assuming that the generated error information is E, then E = [e1, e2, ...e M ], where M = 2N. The prior data is the data of the same type of samples accumulated at different times. Through classification statistics and frequency histogram methods, the data error is statistically found to obey this distribution; the error generation method is to use matlab programming to randomly generate 2N error data that conform to this distribution. Gaussian distribution is a common form of random distribution. Matlab has a random number generation module that conforms to this distribution. This module can be used to generate 2N error data. Step 2: Interpolate the original small sample data. First, let Y = [y1, y2, …y N ], where y i =(x i +x i+1 ) / 2+x N / M, then the data vector X after interpolation transformation can be expressed as X_inter = [x1, y1, x2, y2, ...x N ,y N ], for the convenience of subsequent writing, let Z = X_inter = [z1, z2, ... z M ], where M = 2N. The third step is to generate the transformed large sample data Z_data. Z _ datai=[(z i +e i )+(z i -e i )] / 2+abs(z i) / N, where abs() represents the modulus of the data. In step 102, the small sample operation data is interpolated to obtain the large sample operation data, which specifically includes:
[0029] (a1) Determine data error based on historical equipment operation data.
[0030] (a2) Determine interpolation data of the small sample running data.
[0031] The formula for determining the interpolation data of the small sample running data is:
[0032] y i＝ (x i +x i+1 ) / 2+x N / M;
[0033] Among them, y i represents the i-th interpolation data; x i and x i+1 represents the data at sampling point i and sampling point i+1 in the small sample running data; N represents the total number of sampling points of the small sample running data; M=2N.
[0034] (a3) constructing interpolation transformed data according to the interpolation data and the small sample operation data.
[0035] (a4) constructing the large sample operation data according to the interpolated data and the data error. The calculation formula for each sampling point data in the large sample operation data is:
[0036] Z_datai=[(z i +e i )+(z i -e i )] / 2+abs(z i ) / N;
[0037] Among them, Z datai represents the data of sampling point i in the large sample running data; z i Represents the data of sampling point i in the interpolated data; e i It represents the data error corresponding to sampling point i; N represents the total number of sampling points of small sample running data; abs() represents the modulus of the data.
[0038] In another exemplary embodiment of the present application, the large sample data is preprocessed. Step 1: Remove wild values from the data based on the hard threshold. The hard threshold for each column in Z_data is mainly obtained based on historical experience data. Step 2: Process the data to reduce errors based on the smoothing threshold. The smoothing threshold is mainly 3 / 4 times the average value of two adjacent data. When the data of a certain sampling point is greater than the smoothing threshold, the data of the sampling point is added with the smoothing threshold. If the data of the sampling point is lower than the smoothing threshold, the data of the sampling point is replaced by the smoothing threshold to complete the smoothing. Therefore, in step 103, the large sample operation data is preprocessed to obtain the preprocessed large sample operation data, which specifically includes:
[0039] The large sample operation data is preprocessed according to the hard threshold and the smoothing threshold to obtain the preprocessed large sample operation data. The more specific process is:
[0040] (b1) Comparing the data of each sampling point in the large sample operation data with the corresponding hard threshold, and screening out the large sample operation data after preliminary screening.
[0041] (b2) calculating a smoothing threshold for each sampling point data in the large sample running data after the initial screening.
[0042] For each sampling point data in the large sample running data after the initial screening, a smoothing threshold is calculated based on two adjacent data of each sampling point data. As an example, the smoothing threshold is 3\4 times the average value of the two adjacent data.
[0043] (b3) If the data of each sampling point is greater than the corresponding smoothing threshold, the data of the current sampling point plus the smoothing threshold are added as the updated data of the current sampling point.
[0044] (b4) If the data of each sampling point is less than or equal to the corresponding smoothing threshold, the smoothing threshold is used to replace the data of the current sampling point.
[0045] In another exemplary embodiment of the present application, the fault characteristics of the large sample data are taken: the fault characteristics exist in multiple data arranged at the front of the large sample data, such as 5-bit data. The large sample data Z_data is sorted from large to small, and the first 5 data are selected, Z_data1, Z_data2, Z_data3, Z_data4, and Z_data5. Therefore, in step 104, the fault characteristic data in the preprocessed large sample operation data is extracted, which specifically includes:
[0046] (c1) sorting the data of each sampling point in the preprocessed large sample running data according to size.
[0047] (c2) Selecting the largest preset number of data as fault characteristic data.
[0048] In another exemplary embodiment of the present application, fault characteristic data is used to complete fault diagnosis: when Z_data1, Z_data2, Z_data3, Z_data4, and Z_data5 are respectively less than the corresponding fault thresholds (as an example 15, 13, 12, 11, 5), it is determined that the equipment to be diagnosed does not have a fault. When any of the data is not within the above range, it is determined that the equipment to be diagnosed has a fault. At this time, it is necessary to combine the parameter information of the specific equipment to complete accurate fault diagnosis. The present application is mainly based on the general small sample data to complete the transformation of large sample data, and at the same time complete the determination of whether the equipment has a fault. Therefore, in step 105, fault diagnosis is performed according to the fault characteristic data, specifically including:
[0049] (d1) Comparing each of the fault characteristic data with the corresponding fault threshold.
[0050] (d2) If each of the fault characteristic data is smaller than the corresponding fault threshold, it is determined that the equipment to be diagnosed does not have a fault.
[0051] (d3) If at least one of the fault characteristic data is smaller than the corresponding fault threshold, it is determined that the equipment to be diagnosed has a fault.
[0052] The present application also provides an application scenario, which applies the above-mentioned fault diagnosis method based on small sample data. Specifically: The fault diagnosis method based on small sample data provided in this embodiment can be applied in the equipment fault diagnosis scenario. The equipment fault diagnosis scenario includes a data collection link, a fault diagnosis link and a fault processing link; the data collection link is used to collect small sample operation data of the equipment to be diagnosed; the fault diagnosis link is used to obtain the small sample operation data of the equipment to be diagnosed, interpolate the small sample operation data to obtain large sample operation data, pre-process the large sample operation data to obtain the pre-processed large sample operation data, extract the fault feature data in the pre-processed large sample operation data, and perform fault diagnosis according to the fault feature data; the fault processing link is used to perform fault processing when a fault is diagnosed. The fault diagnosis method based on small sample data provided in this embodiment belongs to the fault diagnosis link.
[0053] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 3As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store fault diagnosis results. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a fault diagnosis method based on small sample data is implemented.
[0054] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.
[0055] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0056] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0057] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0058] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0059] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.
[0060] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0061] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A fault diagnosis method based on small sample data, characterized in that: The fault diagnosis method based on small sample data includes: Obtain a small sample of operating data for the equipment to be diagnosed; Performing data interpolation on the small sample operation data to obtain large sample operation data; Preprocessing the large sample operation data to obtain preprocessed large sample operation data; Extracting fault feature data from the preprocessed large sample operation data; Fault diagnosis is performed according to the fault characteristic data.
2. The fault diagnosis method based on small sample data according to claim 1, characterized in that: The small sample operation data is interpolated to obtain large sample operation data, specifically including: Determine data errors based on historical equipment operation data; Determining interpolation data of the small sample running data; Constructing interpolation transformed data according to the interpolation data and the small sample operation data; The large sample operation data is constructed according to the interpolated transformed data and the data error.
3. The fault diagnosis method based on small sample data according to claim 1, characterized in that: The formula for determining the interpolation data of the small sample running data is: y i＝ (x i +x i+1 ) / 2+x N / M; Among them, y i represents the i-th interpolation data; x i and x i+1 represents the data at sampling point i and sampling point i+1 in the small sample running data; N represents the total number of sampling points of the small sample running data; M=2N.
4. The fault diagnosis method based on small sample data according to claim 1, characterized in that: The calculation formula for each sampling point data in the large sample operation data is: Z_datai=[(z i +e i )+(with i -e i )] / 2+abs(z i ) / N; Where Z_datai represents the data of sampling point i in the large sample operation data; z i Represents the data of sampling point i in the interpolated data; e i It represents the data error corresponding to sampling point i; N represents the total number of sampling points of small sample running data; abs() represents the modulus of the data.
5. The fault diagnosis method based on small sample data according to claim 1, characterized in that: Preprocessing the large sample operation data to obtain the preprocessed large sample operation data specifically includes: The large sample operation data is preprocessed according to the hard threshold and the smoothing threshold to obtain the preprocessed large sample operation data.
6. The fault diagnosis method based on small sample data according to claim 5, characterized in that: Preprocessing the large sample operation data according to the hard threshold and the smoothing threshold to obtain the preprocessed large sample operation data specifically includes: Compare the data of each sampling point in the large sample operation data with the corresponding hard threshold, and screen out the large sample operation data after preliminary screening; Calculating a smoothing threshold for each sampling point data in the large sample running data after the initial screening; If the data of each sampling point is greater than the corresponding smoothing threshold, the data of the current sampling point plus the sum of the smoothing threshold is used as the updated data of the current sampling point; If the data of each sampling point is less than or equal to the corresponding smoothing threshold, the smoothing threshold is used to replace the data of the current sampling point.
7. The fault diagnosis method based on small sample data according to claim 6 is characterized in that: For each sampling point data in the large sample running data after the initial screening, a smoothing threshold is calculated, specifically including: For each sampling point data in the large sample running data after the initial screening, a smoothing threshold is calculated based on two adjacent data of each sampling point data.
8. The fault diagnosis method based on small sample data according to claim 7 is characterized in that: The smoothing threshold is 3\4 times the average value of two adjacent data.
9. The fault diagnosis method based on small sample data according to claim 1, characterized in that: Extracting fault feature data from the preprocessed large sample operation data specifically includes: Sorting the data of each sampling point in the preprocessed large sample running data according to size; The largest preset number of data is selected as fault characteristic data.
10. The fault diagnosis method based on small sample data according to claim 9, characterized in that: Performing fault diagnosis according to the fault characteristic data specifically includes: Comparing each of the fault characteristic data with the corresponding fault threshold; If each of the fault characteristic data is smaller than the corresponding fault threshold, it is determined that the equipment to be diagnosed does not have a fault; If at least one of the fault characteristic data is smaller than the corresponding fault threshold, it is determined that the equipment to be diagnosed has a fault.