Device information diagnosis method and device, equipment and computer readable storage medium

Abstract

This application relates to the field of equipment information diagnosis technology, and discloses an equipment information diagnosis method, apparatus, device, and computer-readable storage medium. By grouping acquired equipment data and obtaining the first data name of each piece of equipment data contained in each data group, a diagnostic model for analyzing and diagnosing the equipment data is generated. Then, the second data name corresponding to the data that the diagnostic model supports inputting and diagnosing is determined. The equipment data corresponding to the first data name that is identical to the second data name is determined from the data group and input as the input data into the diagnostic model to obtain the diagnostic result. This improves the efficiency of maintenance personnel in diagnosing faults when there are many devices and a large amount of equipment data, and allows for faster diagnostic results, enabling maintenance personnel to maintain the equipment more promptly.

Description

Technical Field

[0001] This application relates to the field of equipment diagnostic technology, specifically to a method, apparatus, device, and computer-readable storage medium for equipment information diagnosis. Background Technology

[0002] In recent years, with the development of technology, enterprises have invested more and more in the number of production equipment in factories, and the structure of the equipment has become more and more complex. With the widespread application and deployment of sensor technology, when multiple devices are operating at the same time, the equipment data generated by the sensors monitoring the equipment is vast. Diagnostic analysis models used to diagnose equipment faults often need to take data from multiple devices from the same source as input for calculation and analysis to obtain the analysis results of a certain device or module, so that maintenance personnel can maintain the faulty equipment or module based on the analysis results.

[0003] In practice, maintenance personnel need to filter equipment data from different devices and manually input it into the model to obtain analysis results for one or more devices that need to be diagnosed. The input process is time-consuming and laborious, which is not conducive to the efficiency of equipment fault diagnosis and affects the timeliness of equipment maintenance. Summary of the Invention

[0004] In view of the above problems, embodiments of this application provide a device information diagnosis method, apparatus, device, and computer-readable storage medium to solve the problem of cumbersome device data input model process.

[0005] According to one aspect of the embodiments of this application, a device information diagnosis method is provided, comprising: acquiring device data; analyzing the device data to obtain the data source of the device data; assigning device data with the same data source to the same group to obtain at least one data group, each data group including at least one device data; acquiring a first data name for each device data contained in each data group; generating a diagnostic model for analyzing and diagnosing the device data; determining a second data name, the second data name being the name corresponding to data that the diagnostic model supports input for diagnosis; determining the device data corresponding to the first data name that is the same as the second data name from the at least one data group, and using it as input data; inputting the input data into the diagnostic model to obtain a diagnostic result. By grouping the acquired equipment data and obtaining the first data name for each piece of equipment data within each data group, a diagnostic model for analyzing and diagnosing the equipment data is generated. Then, the second data name corresponding to the data that the diagnostic model supports inputting and diagnosing is determined. Equipment data with the same first data name as the second data name is selected from the data group and input into the diagnostic model as input data to obtain diagnostic results. This ensures that equipment data from the same data source can correspond to inputs with consistent names in the diagnostic model, enabling rapid diagnosis of the corresponding equipment, module, or component. When there is a large amount of equipment data, the process of inputting the equipment data into the model is faster, eliminating the need for manual input of names sequentially. This improves the efficiency of maintenance personnel in diagnosing faults when there are many devices and a large amount of equipment data, allowing the diagnostic model to analyze and diagnose the equipment data more quickly and obtain diagnostic results, thus facilitating timely maintenance by maintenance personnel.

[0006] In one optional approach, after assigning device data from the same data source to the same group to obtain at least one data group, the device information diagnosis method further includes: determining whether duplicate device data exists in each data group; if so, retaining one device data from the duplicate device data in the same data group; and compressing the device data in the data group to obtain new device data. By removing duplicate device data and compressing the device data to obtain new device data, redundant device data will not affect the accuracy of the analysis results during subsequent analysis. This also makes the device data more concise, avoiding unexpected long processing times due to data redundancy. Furthermore, the compressed new device data occupies less space, reducing the pressure on data transmission and processing.

[0007] In one optional approach, compressing the equipment data in the data set to obtain new equipment data includes: compressing the equipment data using a segmented clustered average algorithm to obtain new equipment data. Because industrial equipment generates equipment data with high dimensionality and large volume, most data analysis methods often struggle to support such a large workload. By employing a segmented clustered average algorithm to compress the equipment data, excessive loss of equipment data characteristics is minimized, and the shape and trend characteristics of the equipment data are better preserved. The resulting data sequence is smoother, making subsequent diagnostic analysis of the equipment data more accurate.

[0008] In one optional approach, before generating the diagnostic model for analyzing and diagnosing equipment data, the equipment information diagnostic method further includes: obtaining a fault correlation table of the equipment data, wherein the fault correlation table characterizes the probability of failure of the equipment corresponding to the data source when the difference between the parameter size of the equipment data and the standard value is equal to a preset percentage of the standard value, and the standard value is the value of the factory parameter calibration of the equipment corresponding to the data source; according to the fault correlation table, equipment data in the data group whose failure probability is greater than a first preset threshold is used as first feature data; generating the diagnostic model for analyzing and diagnosing equipment data includes: sequentially inputting the first feature data in each data group into a neural network model for training to obtain the diagnostic model. By looking up a table and setting a first preset threshold, equipment data with a high correlation to equipment failure is identified as first feature data, and the first feature data is input into the neural network model for training, which makes the obtained diagnostic model more sensitive to diagnosing equipment failures and improves the accuracy of the diagnostic model in diagnosing equipment failures.

[0009] In one optional approach, obtaining the first data name of each device data contained in each data group includes: obtaining the data type of the device data contained in the data group; determining whether the data type is a database type, where the database type contains multiple sub-data items; if so, merging the names of each sub-data item in the device data corresponding to the current data type as the first data name; if not, using the file name of the device data corresponding to the current data type as the first data name. By determining the data type of the device data and adopting different methods for obtaining the first data name based on whether it is a database type, the method ensures that even when the device data is contained in a database file, the acquisition of the first data name will not be incorrect, and the probability of the obtained first data name matching the second data name of the diagnostic model is higher. Since device data in some fields may be represented by a single file reflecting a single data value, and when the device data is not a database type, its file name is used as the first data name, the device information diagnostic method provided in this application can exhibit good adaptability to more types of device data.

[0010] In one optional approach, after inputting the data to be input into the diagnostic model and obtaining the diagnostic result, the equipment information diagnostic method further includes: determining whether the diagnostic result is a fault; if so, using the data to be input as the second feature data. By determining whether the diagnostic result is a fault, and then using the data to be input from the equipment diagnosed as faulty as the second feature data, maintenance personnel can easily retrieve the data to be input when the equipment is faulty for verification and analysis.

[0011] In one optional approach, if the input data is used as the second feature data, the equipment information diagnosis method further includes: obtaining the quantity of the second feature data; determining whether the quantity of the second feature data is greater than or equal to a second preset threshold; if not, adding a piece of equipment data with the same data source as the second feature data to the second feature data, until the quantity of the second feature data equals the second preset threshold or all equipment data with the same data source as the second feature data have been added to the second feature data. The second feature data is used as input to a neural network model for training to obtain an optimized model. The optimized model is used as a new diagnostic model, which is used to analyze and diagnose equipment data in subsequent steps. In this way, the diagnostic model trained by the neural network model can better adapt to the actual operating state of the equipment. The model trained based on equipment data when the equipment itself malfunctions will be more sensitive to the analysis and diagnosis of possible malfunctions of the same equipment, helping to improve the accuracy and efficiency of the new diagnostic model in diagnosing equipment malfunctions.

[0012] According to another aspect of the embodiments of this application, a device information diagnostic apparatus is provided, comprising: a first acquisition module, a first processing module, a second processing module, a second acquisition module, a third processing module, a first determination module, a second determination module, and a fourth processing module. The first acquisition module is used to acquire device data; the first processing module is used to analyze the device data to obtain the data source of the device data; the second processing module is used to allocate device data with the same data source to the same group to obtain at least one data group; the second acquisition module is used to acquire a first data name for each device data contained in each data group; the third processing module is used to generate a diagnostic model for analyzing and diagnosing the device data; the first determination module is used to obtain a second data name; the second determination module is used to determine the device data corresponding to the first data name that is the same as the second data name from the at least one data group and use it as input data; and the fourth processing module is used to input the input data into the diagnostic model to obtain a diagnostic result.

[0013] According to another aspect of the embodiments of this application, a device information diagnostic device is provided, including: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus; the memory is used to store at least one program, which causes the processor to perform the operation of any of the above device information diagnostic methods.

[0014] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein executable instructions are stored in the storage medium, and the executable instructions cause the device information diagnostic device to perform the operation of any of the above-described device information diagnostic methods.

[0015] The above description is merely an overview of the technical solutions of the embodiments of this application. In order to better understand the technical means of the embodiments of this application and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of this application more obvious and understandable, specific implementation methods of this application are described below. Attached Figure Description

[0016] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0017] Figure 1 A schematic flowchart illustrating the device information diagnosis method provided in the embodiments of this application;

[0018] Figure 2 This is a flowchart illustrating the steps following step 130 of this application;

[0019] Figure 3 This is a schematic diagram illustrating one embodiment of step 1303 of this application;

[0020] Figure 4 This is a flowchart illustrating the steps preceding step 150 of this application;

[0021] Figure 5 This is a flowchart illustrating a sub-step of step 140 in this application;

[0022] Figure 6 This is a flowchart illustrating the steps following step 180 of this application;

[0023] Figure 7 This is a flowchart illustrating the steps following step 1802 in this application;

[0024] Figure 8 This is a functional block diagram of the device information diagnostic apparatus provided in the embodiments of this application;

[0025] Figure 9 This is a schematic diagram of the device information diagnostic device provided in an embodiment of this application. Detailed Implementation

[0026] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein.

[0027] The inventors of this application understand that, with the development of modern technology, in many fields, people often need to use a large number of devices to work together to achieve one or more goals, such as in the Internet of Things (IoT) field and automated processing plants. When a large number of devices are operating simultaneously, if a device malfunctions, it will seriously affect the efficiency of the operation and may even cause several other related devices to malfunction. Therefore, people often analyze and diagnose equipment data in real time to accurately locate equipment faults so that maintenance personnel can repair the equipment in a timely manner.

[0028] The inventors of this application have noted that, in order to improve efficiency and save manpower, in many fields, maintenance personnel generally use machine learning-generated models to analyze and diagnose equipment data. Maintenance personnel input the real-time data generated by the equipment into the model, and the model calculates the analysis results of the equipment. Based on the analysis results, it is determined whether the equipment has a fault, so that maintenance personnel can perform timely maintenance on faulty equipment.

[0029] However, the inventors of this application have found that in actual operation, maintenance personnel need to filter equipment data from different devices and manually input it into the model to obtain analysis results for one or more devices that need to be diagnosed. The input process is time-consuming and laborious, which is not conducive to the efficiency of equipment fault diagnosis and affects the timeliness of equipment maintenance.

[0030] Based on this, the inventors, through research, designed an equipment information diagnosis method. This method involves grouping the acquired equipment data and obtaining the first data name for each piece of equipment data within each data group. After generating a diagnostic model for analyzing and diagnosing the equipment data, the second data name corresponding to the data supported for input and diagnosis by the diagnostic model is determined. The equipment data corresponding to the first data name with the same second data name is then selected from the data group and input into the diagnostic model as the input data to obtain the diagnostic result. This method eliminates the need for maintenance personnel to manually search for and input equipment data from the equipment data into the model, improving the efficiency of fault diagnosis when there are many devices and a large amount of equipment data. It allows for faster diagnostic results and more timely equipment maintenance.

[0031] The device information diagnosis method disclosed in this application can be used, but is not limited to, for fault diagnosis of industrial Internet devices. It can also be applied to any other components or devices that need to be diagnosed. In this application embodiment, the method for fault diagnosis of industrial Internet devices is used as an example for illustration.

[0032] Figure 1 This illustration shows a flowchart of a device information diagnosis method provided in an embodiment of this application. The method is executed by a device information diagnosis device, which may be, for example, a computer, a server, etc. Figure 1 As shown, the method includes the following steps:

[0033] Step 110: Obtain device data.

[0034] In this step, device data can be acquired either by establishing a communication protocol with external devices or by directly retrieving it from local storage according to file paths. Different methods can be used depending on the source and type of the device data. The purpose of acquiring device data is to obtain data that reflects the current or a specific time period of the device's operating status for analysis in subsequent steps. In some cases, the acquired device data may have been forwarded or processed by other intermediary devices; as long as the data still reflects the operating status of its source device to some extent after forwarding or processing, it is acceptable.

[0035] Step 120: Analyze the equipment data to obtain the data source of the equipment data.

[0036] In this step, the device data is analyzed to determine the data source. This can be done by directly reading the file path of the data or by locating the data source through the source information attached to the device data's communication protocol.

[0037] The data source refers to the source of the equipment data, which can be external equipment, the equipment itself, or a module or component within the same equipment. The purpose of obtaining the data source is to ensure that the acquired equipment data corresponds to the operating status of the equipment, module, or component it reflects, so as to better diagnose and analyze the operating status of each equipment, module, or component in subsequent steps.

[0038] Step 130: Assign device data from the same data source to the same group to obtain at least one data group, and each data group includes at least one device data.

[0039] In this step, the device data is classified according to its source. Specifically, if all device data comes from the same source, then only one data group containing all device data is obtained after grouping. If some device data comes from device A and some comes from device B, then two data groups are obtained after grouping. One data group contains all device data from device A, and the other data group contains all device data from device B. If some device data comes from device A, some comes from module a1 in device A, and the rest comes from device B, then two classification results may be obtained according to the data classification method: 1) Three data groups are obtained, with the first data group containing device data from device A, the second data group containing device data from module a1, and the third data group containing device data from device B; 2) Two data groups are obtained, with the first data group containing device data from device A and module a1, and the second data group containing device data from device B.

[0040] The above embodiments describe multiple possible classification results for the same device data, depending on the different methods the software uses to classify data according to similar items. It is understood that the number of data groups obtained in step 130 will vary depending on the classification method used. Regardless of the classification method used, the essence is to group device data from multiple devices, modules, or components according to different sources, so as to include the data of the same device, module, or component in one group to facilitate subsequent diagnostic analysis. This allows maintenance personnel to determine which specific device, module, or component has a fault based on the diagnostic results of the device data in the same data group. The different classification methods only affect the accuracy of the maintenance personnel in locating the source of the fault in the device, module, or component. Those skilled in the art can flexibly choose the method of classifying device data according to the data source according to actual operational needs, so that the speed of grouping device data and the accuracy of subsequent device location by maintenance personnel meet the actual operational requirements.

[0041] Step 140: Obtain the first data name of each device data contained in each data group.

[0042] In this step, the first data name can be obtained by searching the characters in the device data to find the name information of the data contained in the code or character information of the device data. For example, when the device data contains some kind of programming code, the first data name can be determined by searching the characters before the "=" sign to obtain the names of each parameter in the device data. Alternatively, when the device data is text data, the first data name can be determined by searching the characters before the ":". Or, when the device data is data in a database file, the name of each parameter in the database file can be directly read as the first data name.

[0043] Step 150: Generate a diagnostic model for analyzing and diagnosing equipment data.

[0044] In this step, the diagnostic model can be a common training model used for data analysis, fault prediction and fault diagnosis of equipment data. The model can be generated by a machine learning model, such as a hybrid model generated by feature selection based on random forest algorithm and temporal convolutional network as the diagnostic model. This application embodiment does not make any special limitation on this.

[0045] Step 160: Determine the second data name, which is the name corresponding to the data that the diagnostic model supports input for diagnosis.

[0046] In this step, the second data name refers to the name of the data that can be input into the diagnostic model as the basis for diagnosis. Existing training models often have multiple input parameters, all of which are predefined in the model. For example, in an LSTM model within a neural network, there are three input parameters: time_step, input_size, and batch_size. The determined second data names are time_step, input_size, and batch_size. The second data name can be determined by searching the model's program code or directly reading the pre-set input interface names of the model. The determination of the second data name is used in subsequent steps to complement the first data name and assist in inputting device data into the diagnostic model.

[0047] Step 170: Determine the device data corresponding to the first data name that is the same as the second data name from at least one data group, and use it as the input data.

[0048] In this step, determining the device data corresponding to the first data name that is the same as the second data name from at least one data group and using it as input data means that if the first data name of the device data in the data group is the same as the determined second data name, then the device data is used as input data. The input data is used to input into the diagnostic model to obtain the diagnostic result.

[0049] Step 180: Input the data to be input into the diagnostic model to obtain the diagnostic results.

[0050] In this step, the device data with the same first data name and second data name, i.e., the input data, is input into the diagnostic model. Specifically, for example, if the data name of the input data is batch_size, and the second data names of the diagnostic model are time_step, input_size, and batch_size, then the input data is input into batch_size in the diagnostic model as input.

[0051] Steps 110 to 180 ensure that equipment data from the same data source corresponds to inputs with identical names in the diagnostic model, enabling rapid diagnosis of the corresponding equipment, module, or component. When a large amount of equipment data exists, the process of inputting the data into the model is faster, eliminating the need for manual input by name. This improves the efficiency of maintenance personnel in diagnosing faults when dealing with numerous devices and their data, allowing the diagnostic model to analyze and diagnose the data more quickly and obtain diagnostic results, thus facilitating timely maintenance.

[0052] Please see Figure 2 , Figure 2 This is a flowchart illustrating the steps following step 130 of this application. Figure 2 As shown, in some embodiments of this application, after step 130, the following steps are also included:

[0053] Step 1301: Determine whether there is duplicate device data in each data group;

[0054] If the determination is yes, then perform the following steps:

[0055] Step 1302: Retain one device data from the duplicate device data in the same data group;

[0056] Step 1303: Compress the device data in the data group to obtain new device data.

[0057] In steps 1301 to 1302, since the data sources of the device data in the same data group are consistent, the system determines whether there is duplicate device data in each data group. If the determination is yes, only one device data in the duplicate device data is retained in the same data group. This ensures that when the device data from the same data source is duplicated due to unexpected events such as device data transmission failure, the extra data can be removed.

[0058] In step 1303, compressing the device data in the data group to obtain new device data means reducing the amount of data to reduce the data storage space requirement without losing useful information. For example, run-length encoding in lossless compression can be used to compress the device data in the data group so that the new device data occupies less space.

[0059] By removing duplicate device data and compressing it to obtain new device data, redundant device data will not affect the accuracy of the analysis results in subsequent analysis. This also makes the device data more concise, avoiding unexpected long processing times due to data duplication and redundancy. The new device data obtained after compression occupies less space, reducing the pressure on data transmission and processing.

[0060] Please see Figure 3 , Figure 3 This is a schematic diagram illustrating one embodiment of step 1303 of this application. Figure 3 As shown, in some embodiments of this application, step 1303 includes:

[0061] Step 13031: Compress the device data using a segmented clustered average algorithm to obtain new device data.

[0062] In step 13031, the Partition-based Approximate Aggregation algorithm divides the time-series data in the device data into multiple equal-sized intervals along the horizontal direction, sorts them according to their original order, and uses the average value of each interval to form a new sequence representing the original data. The new sequence obtained by this method is the partition-based approximate average representation of the original time-series data. For example, suppose the time-series data sequence in the device data is X = {x1, x2, ..., x...} n ,}, where n∈N + The segmented clustered average algorithm divides the time series data into k intervals sequentially, resulting in a sequence X. PAA ={x1,x2,…,x kTechnicians can set appropriate interval lengths according to actual needs. When the interval length is longer, the number of intervals obtained is smaller, the compression degree of equipment data is higher, and the feature loss of equipment data is higher. Conversely, when the interval length is shorter, the number of intervals obtained is larger, the compression degree of time series data is lower, and the feature loss of equipment data is lower.

[0063] Because industrial equipment generates high-dimensional and large-volume data, most data analysis methods often struggle to support such a large workload. By using a segmented clustered average algorithm to compress the equipment data, compression of high-dimensional and large-volume equipment data does not easily lose too many of the equipment data's features. It can better preserve the shape and trend characteristics of the equipment data, resulting in a smoother data sequence, which makes the diagnostic analysis of the equipment data in subsequent steps more accurate.

[0064] Please see Figure 4 , Figure 4 This is a flowchart illustrating the steps preceding step 150 of this application. Figure 4 As shown, in one embodiment of this application, before step 150, the following steps are further included:

[0065] Step 1510: Obtain the fault correlation table of the equipment data. The fault correlation table is used to characterize the fault probability of the equipment corresponding to the data source when the difference between the parameter size of the equipment data and the standard value is equal to a preset percentage of the standard value. The standard value is the value of the factory parameter calibration of the equipment corresponding to the data source.

[0066] Specifically, the preset percentage can be 10%, 20%, 30%, etc., and its specific value is not limited here.

[0067] Step 1520: Based on the fault correlation table, select the equipment data in the data group whose fault probability is greater than the first preset threshold as the first feature data;

[0068] Step 1530: Input the first feature data from each data group into the neural network model for training to obtain the diagnostic model.

[0069] In step 1510, the fault correlation table can be obtained through multiple experimental tests or by analyzing existing equipment data and results. For example, taking a preset percentage of 20%, there are 20 identical pieces of equipment in a certain operation. The equipment data obtained for each piece of equipment includes temperature data and speed data. The rated temperature and rated speed specified in the factory parameters of these pieces of equipment are 80℃ and 6000 rpm, respectively. Suppose that the temperature data of 10 pieces of equipment are 100℃, 75℃, 60℃, 79℃, 100℃, 82℃, 98℃, 100℃, 87℃, and 100℃, and the speed data of the remaining 10 pieces of equipment are 4800 rpm and 6200 rpm, respectively. Based on the above equipment data, the following speeds are listed: 5800 rpm, 7200 rpm, 5500 rpm, 7200 rpm, 6500 rpm, 6000 rpm, 4800 rpm, and 7200 rpm. The following five temperature data points show a difference of 20% between their parameters and the standard values: 100℃, 60℃, 100℃, 100℃, and 100℃. These are referred to as abnormal temperature data. Similarly, the following five speed data points show a difference of 20% between their parameters and the standard values: 4800 rpm, 7200 rpm, 7200 rpm, 4800 rpm, and 7200 rpm. These are referred to as abnormal speed data. If 4 out of 5 devices corresponding to the source of abnormal temperature data malfunction, the probability of failure for the temperature data in the device data can be calculated as 80%. If 2 out of 5 devices corresponding to the source of abnormal speed data malfunction, the probability of failure for the speed data in the device data can be calculated as 40%. Recording the failure probabilities of the temperature and speed data in the device data into a fault correlation table reveals that device data with higher failure probabilities in the fault correlation table has a higher correlation with device failure.

[0070] In step 1520, by looking up a table, the equipment data with a failure probability greater than a first preset threshold is used as the first feature data. The first feature data is the equipment data with a high correlation to equipment failure, which can be used as the data basis for equipment diagnosis and analysis in subsequent steps.

[0071] In step 1530, the first feature data is input into the neural network model in the following way: the first feature data belonging to the same data group is input into the neural network model for training, and the resulting diagnostic model mainly diagnoses the device data from the same data source. Then, the first feature data belonging to another data group is input into the neural network model for training, and a diagnostic model is mainly diagnosing the device data from another data source. This process is repeated until the first feature data of each data group has been input into the neural network model and training is completed.

[0072] Furthermore, in order to reduce the number of diagnostic models and make them easier to use, the above steps of sequentially inputting the first feature data from each data group into the neural network model for training to obtain the diagnostic model can also be performed by sequentially inputting multiple data groups into the neural network model to train the same diagnostic model. The resulting diagnostic model can perform diagnostic analysis on device data from multiple data sources.

[0073] In steps 1510 to 1530, by looking up a table and setting a first preset threshold, the equipment data with a high correlation to the equipment fault is identified as the first feature data. The first feature data is then input into the neural network model for training, which makes the obtained diagnostic model more sensitive to the diagnosis of equipment faults and improves the accuracy of the diagnostic model in diagnosing equipment faults.

[0074] Please see Figure 5 , Figure 5 This is a flowchart illustrating a sub-step of step 140 in this application. Figure 5 As shown, in some embodiments of this application, step 140 further includes:

[0075] Step 141: Obtain the data type of the device data contained in the data group;

[0076] Step 142: Determine if the data type is a database type. A database type is a data type that contains multiple sub-data items.

[0077] If the determination is yes, then perform the following steps:

[0078] Step 143: Merge the names of each sub-data item in the device data corresponding to the current data type as the first data name;

[0079] If the result is negative, proceed with the following steps:

[0080] Step 144: Use the file name of the device data corresponding to the current data type as the first data name.

[0081] In step 141, the data type of the device data refers to the file extension of the device data. For example, when the device data is log data.log, the obtained data type is log; when the device data is text data output.txt, the obtained data type is txt.

[0082] The data types obtained from the device data contained in the data group may also be of various types. This usually occurs when there are multiple different devices in a single operation. Since the device data obtained from different data sources comes from multiple different devices, their data formats may differ. For example, the obtained data types may be: txt, xml, log, and mdf.

[0083] In steps 142 to 144, the data type of the device data obtained in step 141 is determined. If it is a database type, the names of each sub-data item in the database are merged to form the first data name. The database type refers to the file type of the file used to contain the data in the database. For example, common database types include SQL databases, Access databases, Oracle databases, etc., with file extensions such as .mdf, .mdb, and .dbf. It can be understood that if the file extension of a device data item matches any existing database file extension, then the device data can be considered a database type. In this case, to obtain the first data name, the names of the sub-data items contained in the database should be merged to form the first data name. If the data type of the device data does not match any existing database file extension, then the file name of the device data is used as the first data name. For example, if the device data is crash-report.txt, then the file name crash-report is used as the first data name.

[0084] By determining the data type of the device data and adopting different methods for obtaining the first data name based on whether it is a database type, the method ensures that the first data name is obtained correctly even when the device data is contained in a database file. The probability of the obtained first data name matching the second data name of the diagnostic model is also higher. Since device data in some fields may be represented by a single file reflecting a single data value, the device information diagnostic method provided in this application is obtained as the first data name when the device data is not a database type. This allows the device information diagnostic method to demonstrate good adaptability to a wider range of device data types.

[0085] like Figure 6 As shown, Figure 6 This is a flowchart illustrating the steps following step 180 of this application. Figure 6As shown, in some embodiments of this application, after step 180, the method further includes:

[0086] Step 1801: Determine whether the diagnostic result indicates a fault;

[0087] If the determination is yes, then perform the following steps:

[0088] Step 1802: Use the data to be input as the second feature data.

[0089] In steps 1801 to 1802, by determining whether the diagnostic result is a fault, the data to be input for the equipment diagnosed as faulty is used as the second feature data, which makes it easier for maintenance personnel to retrieve the data to be input when the equipment is faulty for verification and analysis.

[0090] like Figure 7 As shown, Figure 7 This is a flowchart illustrating the steps following step 1802 of this application. Figure 7 As shown, in some embodiments of this application, after step 1802, the method further includes:

[0091] Step 18021: Obtain the number of second feature data;

[0092] Step 18022: Determine whether the quantity of the second feature data is greater than or equal to the second preset threshold;

[0093] If the result is negative, proceed with the following steps:

[0094] Step 18023: Add a device data source consistent with the second feature data to the second feature data until the number of second feature data is equal to the second preset threshold or all device data sources consistent with the second feature data have been added to the second feature data. The second feature data is used to input the neural network model for training to obtain an optimized model. The optimized model is used as a new diagnostic model. The diagnostic model is used to analyze and diagnose the device data in subsequent steps.

[0095] In step 18021, the quantity of the second feature data can be obtained through conventional data statistics. It should be noted that if the second feature data is a data type that contains multiple sub-data in a single file, such as a database type, the quantity of the second feature data should be obtained by obtaining the quantity of the sub-data it contains, rather than obtaining the number of files. If the second feature data is a single file reflecting a single piece of data, the quantity of the second feature data can be obtained by counting the number of files.

[0096] In steps 18022 to 18023, the quantity of the second feature data is judged. If the quantity is not greater than or equal to the second preset threshold, a device data with the same data source as the second feature data is added to the second feature data. This step is repeated until the quantity of the second feature data meets the condition of the second preset threshold or all device data has been added to the second feature data. The purpose is to use the second feature data obtained in steps 1801 to 1802, which can reflect the fault status of the device to a large extent, to input into the neural network model for model training. However, if the quantity of the second feature data is too small, it may lead to poor model training effect or failure to complete training. Therefore, device data from the same data source, that is, from the same device, module or component, is added to the second feature data until the quantity of the second feature data reaches the standard.

[0097] In this way, the diagnostic model trained by the neural network model can better adapt to the actual operating state of the equipment. The model trained based on the equipment data when the equipment itself fails will be more sensitive to the analysis and diagnosis of possible failures of the same equipment, which will help improve the accuracy and efficiency of the new diagnostic model in diagnosing equipment failures.

[0098] Figure 8 A functional block diagram of a device information diagnostic apparatus 800 according to an embodiment of this application is shown. Figure 8 As shown, the device information diagnostic device 800 includes: a first acquisition module 801, a first processing module 802, a second processing module 803, a second acquisition module 804, a third processing module 805, a first determination module 806, a second determination module 807, and a fourth processing module 808. The first acquisition module 801 acquires device data; the first processing module 802 analyzes the device data to determine its data source; the second processing module 803 assigns device data with the same data source to the same group to obtain at least one data group; the second acquisition module 804 acquires the first data name of each device data contained in each data group; the third processing module 805 generates a diagnostic model for analyzing and diagnosing the device data; the first determination module 806 determines the second data name; the second determination module 807 determines the device data corresponding to the first data name with the same second data name from at least one data group and uses it as input data; and the fourth processing module 808 inputs the input data into the diagnostic model to obtain a diagnostic result.

[0099] In some embodiments, the device information diagnostic apparatus 800 further includes a first judgment module, a fifth processing module, and a sixth processing module. The first judgment module is used to determine whether there is duplicate device data in each data group; the fifth processing module is used to retain one device data in the same data group when there is duplicate device data; and the sixth processing module is used to compress the device data in the data group to obtain new device data.

[0100] In some embodiments, the sixth processing module further includes a compression unit. The compression unit is used to compress the device data using a segmented clustered average algorithm to obtain new device data.

[0101] In some embodiments, the device information diagnostic apparatus 800 further includes a third acquisition module and a seventh processing module. The third acquisition module is used to acquire a fault correlation table of device data. The fault correlation table is used to characterize the fault probability of the device corresponding to the data source when the difference between the parameter size of the device data and the standard value is equal to 20% of the standard value. The standard value is the value of the factory parameter calibration of the device corresponding to the data source. The seventh processing module is used to select device data in the data group whose fault probability is greater than a first preset threshold as first feature data according to the fault correlation table.

[0102] In some embodiments, the third processing module 805 further includes an input unit. The input unit is used to sequentially input the first feature data from each data group into the neural network model for training to obtain a diagnostic model.

[0103] In some embodiments, the second acquisition module 804 further includes an acquisition unit, a judgment unit, a first processing unit, and a second processing unit. The acquisition unit is used to acquire the data type of the device data contained in the data group; the judgment unit is used to determine whether the data type is a database type, where the database type is a data type containing multiple sub-data items; the first processing unit is used to merge the names of each sub-data item in the device data corresponding to the current data type as a first data name when the data type is a database type; and the second processing unit is used to use the file name of the device data corresponding to the current data type as the first data name when the data type is not a database type.

[0104] In some embodiments, the device information diagnostic apparatus 800 further includes a second judgment module and an eighth processing module. The second judgment module is used to determine whether the diagnostic result is a fault, and the eighth processing module is used to use the data to be input as second feature data when the diagnostic result is a fault.

[0105] In some embodiments, the device information diagnostic apparatus 800 further includes a fourth acquisition module, a third judgment module, and a ninth processing module. The fourth acquisition module acquires the quantity of second feature data; the third judgment module determines whether the quantity of second feature data is greater than or equal to a second preset threshold; and the ninth processing module adds a device data item with the same data source as the second feature data to the second feature data when the quantity of the second feature data is less than the second preset threshold, until the quantity of the second feature data equals the second preset threshold or all device data items with the same data source as the second feature data have been added to the second feature data. The second feature data is used as input to a neural network model for training to obtain an optimized model. The optimized model is used as a new diagnostic model, which is used to analyze and diagnose the device data in subsequent steps.

[0106] According to another aspect of the embodiments of this application, a device information diagnostic device is provided. Please refer to [link / reference needed]. Figure 9 , Figure 9 The diagram shows a structural schematic of a device information diagnostic device according to an embodiment of this application. The specific embodiments of this application do not limit the specific implementation of the device information diagnostic device.

[0107] like Figure 9 As shown, the device information diagnostic device may include: processor 902, memory 906, communication interface 904 and communication bus 908.

[0108] The processor 902, memory 906, and communication interface 904 communicate with each other via communication bus 908. The memory 906 stores at least one program 910, which causes the processor 902 to execute the relevant steps as described in the above-described embodiment of the device information diagnosis method.

[0109] Specifically, program 910 may include program code, which includes computer-executable instructions.

[0110] The processor 902 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The device information diagnostic device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0111] Memory 906 is used to store program 910. Memory 906 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0112] Specifically, program 910 can be called by processor 902 to enable the device information diagnostic device to perform the following operations: acquire device data; analyze the device data to obtain the data source of the device data; assign device data with the same data source to the same group to obtain at least one data group; each data group includes at least one piece of device data; acquire a first data name for each piece of device data contained in each data group; generate a diagnostic model for analyzing and diagnosing the device data; determine a second data name; the second data name is the name corresponding to the data that the diagnostic model supports input for diagnosis; determine the device data corresponding to the first data name that is the same as the second data name from at least one data group, and use it as the input data; input the input data into the diagnostic model to obtain the diagnostic result.

[0113] This application also provides a computer-readable storage medium storing executable instructions. When the executable instructions are executed on a device information diagnostic device, the device information diagnostic device performs the device information diagnostic method in any of the above embodiments.

[0114] Specifically, the executable instructions can be used to enable the device information diagnostic device to perform the following operations: analyze device data to obtain the data source of the device data; assign device data with the same data source to the same group to obtain at least one data group; each data group includes at least one piece of device data; obtain a first data name for each piece of device data contained in each data group; generate a diagnostic model for analyzing and diagnosing the device data; determine a second data name; the second data name is the name corresponding to the data that the diagnostic model supports inputting and diagnosing; determine the device data corresponding to the first data name that is the same as the second data name from at least one data group, and use it as the input data; input the input data into the diagnostic model to obtain the diagnostic result.

[0115] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of this application are not directed to any particular programming language. It should be understood that the content of this application described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of this application.

[0116] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0117] Similarly, it should be understood that, in order to simplify this application and aid in understanding one or more of the various aspects of the invention, in the above description of exemplary embodiments of this application, various features of the embodiments of this application are sometimes grouped together into a single embodiment, drawing, or description thereof.

[0118] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying abstract and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying abstract and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0119] It should be noted that the above embodiments are illustrative of this application and not restrictive, and those skilled in the art can design alternative embodiments without departing from the scope. Unless otherwise specified, the steps in the above embodiments should not be construed as limiting the order of execution.

Claims

1. A method for diagnosing equipment information, characterized in that, include: Acquire device data; Analyze the device data to determine its data source; The device data from the same data source are assigned to the same group to obtain at least one data group; each data group includes at least one piece of device data. Obtain the first data name of each device data contained in each of the data groups; Generate a diagnostic model for analyzing and diagnosing the data from the device; Determine the second data name; the second data name is the name corresponding to the data that the diagnostic model supports inputting and diagnosing. From at least one of the data groups, determine the device data corresponding to the first data name that is the same as the second data name, and use it as the input data; The data to be input is input into the diagnostic model to obtain the diagnostic result.

2. The equipment information diagnosis method according to claim 1, characterized in that, After assigning device data from the same data source to the same group to obtain at least one data group, the device information diagnosis method further includes: Determine whether there is duplicate device data in each of the data groups; If so, then one of the duplicate device data is retained in the same data group; The device data in the data group is compressed to obtain new device data.

3. The equipment information diagnosis method according to claim 2, characterized in that, The step of compressing the device data in the data group to obtain new device data includes: The device data is compressed using a segmented clustered average algorithm to obtain new device data.

4. The equipment information diagnosis method according to claim 1, characterized in that, Before generating the diagnostic model for analyzing and diagnosing the device data, the device information diagnostic method further includes: Obtain the fault correlation table of the equipment data; the fault correlation table is used to characterize the failure probability of the equipment corresponding to the data source when the difference between the parameter size of the equipment data and the standard value is equal to a preset percentage of the standard value, and the standard value is the value of the factory parameter calibration of the equipment corresponding to the data source. According to the fault correlation table, the device data in the data group whose fault probability is greater than a first preset threshold is taken as the first feature data; The generation of the diagnostic model for analyzing and diagnosing the device data includes: The first feature data from each of the data groups is sequentially input into the neural network model for training to obtain the diagnostic model.

5. The equipment information diagnosis method according to claim 1, characterized in that, The step of obtaining the first data name of each device data contained in each data group includes: Obtain the data type of the device data contained in the data group; Determine whether the data type is a database type; the database type is a data type containing multiple sub-data items. If so, the names of each sub-data item in the device data corresponding to the current data type are merged to form the first data name; If not, then the file name of the device data corresponding to the current data type shall be used as the first data name.

6. The equipment information diagnosis method according to claim 1, characterized in that, After inputting the data to be input into the diagnostic model and obtaining the diagnostic result, the device information diagnostic method further includes: Determine whether the diagnostic result is a fault; If so, the input data will be used as the second feature data.

7. The equipment information diagnosis method according to claim 6, characterized in that, If so, then after using the input data as the second feature data, the device information diagnosis method further includes: Obtain the quantity of the second feature data; Determine whether the quantity of the second feature data is greater than or equal to the second preset threshold; If not, add a device data source consistent with the data source of the second feature data to the second feature data until the number of the second feature data is equal to the second preset threshold or all device data sources consistent with the data source of the second feature data have been added to the second feature data; the second feature data is used to input the neural network model for training to obtain an optimized model, and the optimized model is used as the new diagnostic model, which is used to analyze and diagnose the device data in subsequent steps.

8. A device for diagnosing equipment information, characterized in that, include: The first acquisition module is used to acquire device data; The first processing module is used to analyze the device data and obtain the data source of the device data; The second processing module is used to allocate the device data with the same data source to the same group to obtain at least one data group; The second acquisition module is used to acquire the first data name of each device data contained in each data group; The third processing module is used to generate a diagnostic model for analyzing and diagnosing the device data; The first determining module is used to determine the second data name; the second data name is the name corresponding to the data that the diagnostic model supports inputting and diagnosing. The second determining module is used to determine the device data corresponding to the first data name that is the same as the second data name from at least one of the data groups, and use it as the input data; The fourth processing module is used to input the data to be input into the diagnostic model to obtain the diagnostic results.

9. A device information diagnostic device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one program that causes the processor to perform the operation of the device information diagnostic method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores at least one program, which, when run on the device information diagnostic device, causes the device information diagnostic device to perform the operation of the device information diagnostic method as described in any one of claims 1-7.

Images