A method, device, equipment, and storage medium for filling device status data

Abstract

An embodiment of the present invention discloses a method, device, equipment, and storage medium for filling device status data. The method includes: obtaining to-be-processed device status data of a target pumped-storage device; where the to-be-processed device status data includes at least one of operating temperature, voltage, current, and pressure pulsation; determining a missing data segment in the to-be-processed device status data and an associated data segment corresponding to the missing data segment; and inputting the missing data segment and the associated data segment into a pre-trained target data filling model to obtain a target data filling result corresponding to the missing data segment. The technical solution of the embodiment of the present invention solves the problem of large data errors in data filling in the prior art, can automatically identify the missing data segment and the associated data segment corresponding to the missing data segment, and perform data filling on the missing data segment based on the pre-trained model, improving the quality of data filling.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of pumped storage, and in particular, to a method, device, equipment, and storage medium for filling device status data. Background Art

[0002] During the operation of pumped storage units, in order to monitor the working status of key devices and components of pumped storage and ensure the safe operation of the equipment, it is often necessary to detect the time-varying curves of various status data of the equipment, such as the operating temperature of components, component voltage, current, and unit pressure pulsation, to monitor various working statuses of the equipment.

[0003] The status parameters of pumped storage components reflect the working status of the components and equipment, and show fluctuating changes over time with the operating conditions of the components. In actual operation, there are many abnormal values and missing values in the measured point data. Affected by signal delay, equipment failures, etc., the status parameter data received by the monitoring system has many missing values. Missing measured point data and abnormal measured point data often appear continuously, presenting as segmented missing in the data discrete curve. Existing technologies mostly complement missing data based on interpolation or fitting. For missing data segments, simple interpolation methods are difficult to quickly and effectively complement missing data, and it is easy to have a large error between the interpolated data and the actual data. Moreover, the data curve directly obtained by fitting has a large error from the real situation, which will affect the analysis of the monitoring system. Summary of the Invention

[0004] Embodiments of the present invention provide a method, device, equipment, and storage medium for filling device status data, which can automatically identify missing data segments and associated data segments corresponding to the missing data segments, and fill the missing data segments based on a pre-trained model, improving the quality of data filling.

[0005] In a first aspect, embodiments of the present invention provide a method for filling device status data, the method comprising:

[0006] Obtain the device status data to be processed of the target pumped storage device; wherein, the device status data to be processed includes at least one of operating temperature, voltage, current, and pressure pulsation;

[0007] Determine the missing data segments in the device status data to be processed and the associated data segments corresponding to the missing data segments;

[0008] Input the missing data segments and the associated data segments into a pre-trained target data filling model to obtain the target data filling result corresponding to the missing data segments.

[0009] In a second aspect, an embodiment of the present invention provides a device status data filling device, which includes:

[0010] A status data acquisition module, configured to acquire to-be-processed device status data of a target pumped storage device; wherein, the to-be-processed device status data includes at least one of operating temperature, voltage, current, and pressure pulsation;

[0011] A data segment determination module, configured to determine a missing data segment in the to-be-processed device status data and an associated data segment corresponding to the missing data segment;

[0012] A data filling module, configured to input the missing data segment and the associated data segment into a pre-trained target data filling model to obtain a target data filling result corresponding to the missing data segment.

[0013] In a third aspect, an embodiment of the present invention provides a computer device, which includes:

[0014] One or more processors;

[0015] A memory, configured to store one or more programs;

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the device status data filling method according to any embodiment.

[0017] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the device status data filling method according to any embodiment.

[0018] The technical solution provided by the embodiment of the present invention is to acquire to-be-processed device status data of a target pumped storage device; wherein, the to-be-processed device status data includes at least one of operating temperature, voltage, current, and pressure pulsation; determine a missing data segment in the to-be-processed device status data and an associated data segment corresponding to the missing data segment; input the missing data segment and the associated data segment into a pre-trained target data filling model to obtain a target data filling result corresponding to the missing data segment. The technical solution of the embodiment of the present invention solves the problem of large data errors in data filling in the prior art, can automatically identify a missing data segment and an associated data segment corresponding to the missing data segment, and perform data filling on the missing data segment based on a pre-trained model, improving the quality of data filling. Description of the Drawings

[0019] Figure 1 is a flowchart of a device status data filling method provided by an embodiment of the present invention;

[0020] Figure 2 It is another flowchart of the device state data filling method provided by the embodiments of the present invention;

[0021] Figure 3 It is a schematic structural diagram of a device state data filling device provided by the embodiments of the present invention;

[0022] Figure 4 It is a schematic structural diagram of a computer device provided by the embodiments of the present invention. Detailed implementation manners

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] Figure 1 It is a flowchart of a device state data filling method provided by the embodiments of the present invention. The embodiments of the present invention can be applied to the scenario of filling in missing data in the device state data of pumped-storage equipment. This method can be executed by a device state data filling device, and this device can be implemented in a software and / or hardware manner.

[0025] As Figure 1 shown, the device state data filling method includes the following steps:

[0026] S110. Obtain the to-be-processed device state data of the target pumped-storage equipment.

[0027] Among them, the target pumped-storage equipment may be the pumped-storage equipment that needs to perform state data observation. The to-be-processed device state data may be the state data of the target pumped-storage equipment that needs to be filled in with data. Specifically, the to-be-processed device state data includes at least one of operating temperature, voltage, current, and pressure pulsation. Exemplarily, the state data of the target pumped-storage equipment within a preset time period can be detected by a corresponding state data detection device, and then the to-be-processed device state data can be obtained.

[0028] S120. Determine the missing data segment in the to-be-processed device state data and the associated data segment corresponding to the missing data segment.

[0029] Among them, the missing data segment can be a data segment for which data filling is required. Exemplarily, the missing data in the device status data to be processed can be identified, and a data segment with the number of consecutive missing data greater than a preset missing threshold can be used as the missing data segment. That is, in the embodiments of the present invention, the missing data segment can be identified first, and then the missing data in the missing data segment can be filled. Further, the associated data segment can be a data segment having a time association relationship with the missing data segment. Optionally, the status data of a preset associated quantity before the missing data segment in the device status data to be processed can be used to form the associated data segment. Subsequently, based on the time association relationship between the associated data segment and the missing data segment, the missing data in the missing data segment can be predicted, so as to achieve the filling of the missing data.

[0030] S130. Input the missing data segment and the associated data segment into a pre-trained target data filling model to obtain a target data filling result corresponding to the missing data segment.

[0031] Among them, the target data filling model can be a model for filling missing data. Optionally, a preset neural network model can be trained to obtain the target data filling model. The target data filling result can be the result after filling the missing data in the missing data segment. Specifically, the target data filling model can predict the missing data in the missing data segment based on the time association relationship between the associated data segment and the missing data segment, output the predicted status data, and use the predicted status data as the target data filling result.

[0032] The technical solution provided by the embodiments of the present invention is to obtain the device status data to be processed of the target pumped storage device; among them, the device status data to be processed includes at least one of operating temperature, voltage, current, and pressure pulsation; determine the missing data segment in the device status data to be processed and the associated data segment corresponding to the missing data segment; input the missing data segment and the associated data segment into a pre-trained target data filling model to obtain a target data filling result corresponding to the missing data segment. The technical solution of the embodiments of the present invention solves the problem of large data errors in data filling in the prior art, can automatically identify the missing data segment and the associated data segment corresponding to the missing data segment, and fill the missing data segment based on the pre-trained model, improving the quality of data filling.

[0033] Figure 2It is a flowchart of another method for filling device status data provided by an embodiment of the present invention. The embodiment of the present invention is applicable to the scenario of filling missing data in the device status data of pumped-storage equipment. On the basis of the above embodiment, this embodiment further illustrates how to determine the missing data segment in the to-be-processed device status data and the associated data segment corresponding to the missing data segment; and how to train a target data filling model. This device can be implemented in a software and / or hardware manner and integrated into a computer device with application development functions.

[0034] As Figure 2 shown, the method for filling device status data includes the following steps:

[0035] S210. Obtain the to-be-processed device status data of the target pumped-storage equipment.

[0036] Among them, the target pumped-storage equipment may be the pumped-storage equipment that needs to observe the status data. The to-be-processed device status data may be the status data of the target pumped-storage equipment that needs to be filled with data. Specifically, the to-be-processed device status data includes at least one of operating temperature, voltage, current, and pressure pulsation. Exemplarily, the status data of the target pumped-storage equipment within a preset time period can be detected by a corresponding status data detection device, and then the to-be-processed device status data can be obtained.

[0037] S220. Identify the missing data in the to-be-processed device status data, and use the data segment with the number of consecutive missing data greater than the preset missing threshold as the missing data segment.

[0038] Among them, the preset missing threshold may be a preset reference threshold for determining the missing segment. The missing data segment may be the data segment that needs to be filled with data. Specifically, the missing data in the to-be-processed device status data can be identified, and the data segment with the number of consecutive missing data greater than the preset missing threshold can be used as the missing data segment. That is, the embodiment of the present invention can first identify the missing data segment and then fill the missing data in the missing data segment.

[0039] S230. Determine the associated data segment corresponding to the missing data segment according to the status data of the preset associated quantity before the missing data segment in the to-be-processed device status data.

[0040] Among them, the preset associated quantity can be set as a quantity threshold preset for determining the associated segment corresponding to the missing data segment. The associated data segment can be a data segment having a time association relationship with the missing data segment. Specifically, the state data of the preset associated quantity before the missing data segment in the device state data to be processed can be constituted as the associated data segment. That is, a segment constituted by several state data before the missing data segment can be used as the associated data segment. Subsequently, based on the time association relationship between the associated data segment and the missing data segment, the missing data in the missing data segment can be predicted, so as to achieve the filling of the missing data.

[0041] S240. Input the missing data segment and the associated data segment into a pre-trained target data filling model to obtain a target data filling result corresponding to the missing data segment.

[0042] Among them, the target data filling model can be a model for filling missing data. Optionally, a preset neural network model can be trained to obtain the target data filling model. The target data filling result can be the result after filling the missing data in the missing data segment. Specifically, the target data filling model can predict the missing data in the missing data segment based on the time association relationship between the associated data segment and the missing data segment, and output the predicted state data, and use the predicted state data as the target data filling result.

[0043] Optionally, the training process of the target data filling model includes: obtaining a preliminary data filling model and device state sample data; determining the missing sample data segment and the associated sample data segment in the device state sample data, and determining at least one reference sample data segment from the device state sample data based on the associated sample data segment; inputting the missing sample data segment, the associated sample data segment and all the reference sample data segments into the preliminary data filling model to obtain a sample data filling result corresponding to the missing sample data segment; determining a filling loss function according to the sample data filling result and the preset data label of the device state sample data, and adjusting the preliminary data filling model according to the filling loss function to obtain the target data filling model.

[0044] Among them, the preliminary data filling model can be a data filling model that has not been trained. Optionally, an LSTM model can be used as the preliminary data filling model. This model is a deep learning model commonly used to process time series data and can capture the tasks of long-term dependencies in time series. The device status sample data can be the device status data used as model training samples. Optionally, a preset number of status data can be extracted from the historical status data of the target pumped-storage device manually, and then multiple missing data are formed, and the historical status data after data extraction is used as the device status sample data. Subsequently, based on the filling results of the missing data in the device status sample data by the preliminary data filling model, the parameters of the model can be adjusted to obtain the target data filling model.

[0045] Furthermore, the missing sample data segment can be a segment composed of the missing data in the device status sample data. Optionally, the missing sample data in the device status sample data can also be identified, and the data segment with the number of consecutive missing sample data greater than the preset missing threshold can be used as the missing sample data segment. Further, the associated sample data segment can be a data segment that has a time association relationship with the missing sample data segment. Optionally, the segment composed of the preset associated number of status sample data before the missing sample data segment in the device status sample data can also be used as the associated sample data segment. The reference sample data segment can be a reference data segment for data filling. Optionally, the sample data with a quantity greater than the data quantity in the associated sample data segment can be selected from the device status sample data to form the reference sample data segment. The preliminary data filling model can learn the association relationship between the value of the status data and time based on the reference sample data segment, which is convenient for the model to fill the missing data based on the learned association relationship later.

[0046] The sample data filling result can be the filling result for the missing sample data segment. Specifically, the status data corresponding to the missing sample data segment output by the model can be used as the sample data filling result. The filling loss function can be the error function when the preliminary data filling model performs data filling. Specifically, the data in the sample data filling result and the preset data label can be substituted into the preset loss function to obtain the filling loss function. Further, the model parameters in the preliminary data filling model can be adjusted based on the filling loss function to obtain the target data filling model.

[0047] Optionally, determining at least one reference sample data segment from the device status sample data based on the associated sample data segment includes: traversing each sample data in the device status sample data, selecting a preset number of data from the device status sample data starting from the sample data, obtaining an initial sample data segment corresponding to each sample data, and determining the number of missing data in each initial sample data segment; wherein the preset number of samples is greater than or equal to the number of missing sample data in the missing sample data segment; screening the initial sample data segments according to the number of missing data to determine candidate sample data segments; and screening out reference sample data segments from the candidate sample data segments based on the similarity between the candidate sample data segments and the associated sample data segment.

[0048] Wherein, the preset number of samples can be the number of data in the reference sample data segment. The initial sample data segment can be the original sample data segment selected from the device status sample data. Specifically, starting from each sample data in the device status sample data, the data segment composed of the preset number of data after the sample data can be used as the initial sample data segment. The number of missing sample data can be the number of missing data in the initial sample data segment. Specifically, after determining the initial sample data segment, the missing data in the initial sample data segment can be identified, and the number of missing sample data can be determined according to the number of identified missing data. The candidate sample data segment can be the data segment screened out from the initial sample data segment. Specifically, the candidate sample data segment can be determined by screening the initial sample data segment according to the number of missing data. Further, based on the similarity between the candidate sample data segment and the associated sample data segment, the candidate sample data segment can be screened again to obtain the reference sample data segment.

[0049] Optionally, screening the initial sample data segments according to the number of missing data to determine candidate sample data segments includes: for each candidate sample data segment, determining the absolute difference between the first data value in the initial sample data segment and the first data value in the associated sample data segment; and when the absolute difference is less than the preset difference threshold and the number of missing data in the candidate sample data segment is less than the preset sample missing threshold, taking the candidate sample data segment as the candidate sample data segment.

[0050] Optionally, based on the similarity between the candidate sample data segment and the associated sample data segment, the reference sample data segment is screened out from the candidate sample data segment, including: for each candidate sample data segment, using the sample data of the first associated sample data quantity in the candidate sample data segment as the sample data to be analyzed; for each sample data to be analyzed, determining the corresponding control data from the associated sample data segment and determining the data similarity value between the sample data to be analyzed and the control data; when the sum value of the data similarity values of all the sample data to be analyzed in the candidate sample data segment is less than or equal to the preset similarity threshold, taking the candidate sample data segment as the reference sample data segment.

[0051] Among them, the sample data to be analyzed can be the data used for similarity degree analysis in the candidate sample data segment. The quantity of associated sample data can be the quantity of data in the associated sample data segment. Exemplarily, if the quantity of associated sample data is 50, the first 50 sample data in the candidate sample data segment can be used as the sample data to be analyzed. By determining the sample data to be analyzed, it is to make the quantity of data in the candidate sample data segment consistent with that in the associated sample data segment, facilitating subsequent similarity degree analysis.

[0052] Further, the control data can be the data used for similarity degree analysis with the sample data to be analyzed. Specifically, the data with the same data sorting as the sample data to be analyzed in the associated sample data segment can be used as the control data. Exemplarily, when analyzing the tenth sample data to be analyzed, the tenth data in the associated sample data segment can be used as the control data corresponding to the sample data to be analyzed. Further, the data similarity value can be a value reflecting the similarity degree between two data. The data similarity value can be determined according to the data difference between the sample data to be analyzed and the control data. Further, after determining the data similarity value corresponding to each sample data to be analyzed, the sum of the data similarity values of all the sample data to be analyzed in the candidate sample data segment can be added up, and then the obtained sum value is compared with the preset similarity threshold. When the sum value of the data similarity values of all the sample data to be analyzed in the candidate sample data segment is less than or equal to the preset similarity threshold, taking the candidate sample data segment as the reference sample data segment.

[0053] Optionally, determining the data similarity value between the sample data to be analyzed and the control data includes: when the control data is missing data, using the preset similarity value as the data similarity value between the sample data to be analyzed and the control data; when the control data is non-missing data, using the square of the numerical difference between the sample data to be analyzed and the control data as the data similarity value.

[0054] Exemplarily, the following provides a specific implementation manner for filling device status data. The specific method is as follows:

[0055] Step1 Select associated data

[0056] Suppose a certain state parameter has a total of n normal measured point data, denoted as [A1, A2, A3... An]. n . For a missing segment with a length of m, that is, this segment has a total of m data acquisition points (including missing values, abnormal values, and a small amount of normal data) in it, and most of the data are abnormal data. If we want to fill this missing segment, first select the 50 measured point data before the missing segment as the associated data of the missing segment, denoted as: [i1, i2, i3... i 49 , i 50 .

[0057] Step2 Select the "window"

[0058] In order to find the law of the change of time series data, it is necessary to find a data segment similar to the associated data among all measured point data, which is called the "window". The method is as follows.

[0059] First, based on i1, traverse the measured point data. Find the measured point data similar to i1, that is, for a measured point data A n , record the sum of the number of missing values and abnormal values within the subsequent (m + 50) sampling points as t. Denote the theoretical maximum value and theoretical minimum value of the state parameter i as i max , i min . If it satisfies

[0060]

[0061] Then A n is the window reference, and select the data segment composed of A n and the subsequent (m + 50) data in total, denoted as [a1, a2, a3... a 50 ... a 50+m . Analyze the similarity between [a1, a2, a3... a 50 and the associated data. For continuous data, consider using variance analysis to analyze its similarity with the associated data. Set the standard as S0, and its calculation formula is

[0062] S0 = 0.4(i max - i min ) (2)

[0063] If it satisfies

[0064]

[0065] Then judge that [a1, a2, a3... a 50 is similar to the associated data. Among them, s tis the square of the difference between two data segments. If there are missing values in the data segment, the s value at this point is replaced with a fixed value, specifically t as

[0066]

[0067] [a1, a2, a3... a 50 is similar to the associated data, then [a1, a2, a3... a 50 ... a 50+m segment is identified as a "window". The "window" has similarity with the missing segment and can be used for machine learning of LSTM, thereby helping the LSTM model output data to fill the missing segment.

[0068] Step3 Fill the missing segment

[0069] Traverse all the measured point data, and input the obtained multiple windows into the LSTM model for training. After learning, the LSTM model extends based on the associated data [i1, i2, i3... i 49 , i 50 and fills the subsequent missing segments.

[0070] By filling the missing segments one by one in this way, the processed state parameter data curve can be obtained. The technical solution provided by the embodiment of the present invention selects the "window" by calculating similarity, and uses the method of replacing missing data with fixed values, which can select high-quality input data and avoid the interference caused by too many missing values in the input data. In addition, by introducing the LSTM model, it can capture the development law of various device state parameters over time, thereby generating output segments to fill the missing segments and achieving the purpose of data governance.

[0071] The technical solution provided by the embodiment of the present invention includes obtaining the to-be-processed device state data of the target pumped storage device; identifying the missing data in the to-be-processed device state data, and taking the data segment with the number of consecutive missing data greater than the preset missing threshold as the missing data segment; determining the associated data segment corresponding to the missing data segment according to the preset number of associated state data before the missing data segment in the to-be-processed device state data; inputting the missing data segment and the associated data segment into the pre-trained target data filling model to obtain the target data filling result corresponding to the missing data segment. The technical solution of the embodiment of the present invention solves the problem of large data error in data filling in the prior art, can automatically identify the missing data segment and the associated data segment corresponding to the missing data segment, and perform data filling on the missing data segment based on the pre-trained model, improving the quality of data filling.

[0072] Figure 3It is a schematic structural diagram of a device state data filling device provided by an embodiment of the present invention. The embodiment of the present invention is applicable to the scenario of filling in missing data in the device state data of pumped-storage equipment. The device can be implemented in the form of software and / or hardware and integrated in a computer device with application development functions.

[0073] As Figure 3 shown, the device state data filling device includes: a state data acquisition module 310, a data segment determination module 320, and a data filling module 330.

[0074] Among them, the state data acquisition module 310 is used to acquire the to-be-processed device state data of the target pumped-storage equipment; wherein, the to-be-processed device state data includes at least one of operating temperature, voltage, current, and pressure pulsation; the data segment determination module 320 is used to determine the missing data segment in the to-be-processed device state data and the associated data segment corresponding to the missing data segment; the data filling module 330 is used to input the missing data segment and the associated data segment into a pre-trained target data filling model to obtain the target data filling result corresponding to the missing data segment.

[0075] The technical solution provided by the embodiment of the present invention is to acquire the to-be-processed device state data of the target pumped-storage equipment; wherein, the to-be-processed device state data includes at least one of operating temperature, voltage, current, and pressure pulsation; determine the missing data segment in the to-be-processed device state data and the associated data segment corresponding to the missing data segment; input the missing data segment and the associated data segment into a pre-trained target data filling model to obtain the target data filling result corresponding to the missing data segment. The technical solution of the embodiment of the present invention solves the problem of large data errors in data filling in the prior art, can automatically identify the missing data segment and the associated data segment corresponding to the missing data segment, and perform data filling on the missing data segment based on the pre-trained model, improving the quality of data filling.

[0076] In an optional implementation manner, the data segment determination module 320 is specifically used to: identify the missing data in the to-be-processed device state data, and use the data segment with the number of consecutive missing data greater than a preset missing threshold as the missing data segment; determine the associated data segment corresponding to the missing data segment according to the state data of a preset associated number before the missing data segment in the to-be-processed device state data.

[0077] In an alternative embodiment, the device status data filling device further includes: a data filling model training module, configured to: obtain a preliminary data filling model and device status sample data; determine the missing sample data segments and associated sample data segments in the device status sample data, and determine at least one reference sample data segment from the device status sample data based on the associated sample data segments; input the missing sample data segments, associated sample data segments, and all reference sample data segments into the preliminary data filling model to obtain a sample data filling result corresponding to the missing sample data segments; determine a filling loss function according to the sample data filling result and the preset data tags of the device status sample data, and adjust the preliminary data filling model according to the filling loss function to obtain the target data filling model.

[0078] In an alternative embodiment, the data filling model training module includes: a reference sample data segment determining sub-module, configured to: traverse each sample data in the device status sample data, select data with a preset sample quantity from the device status sample data starting from the sample data, obtain an initial sample data segment corresponding to each sample data, and determine the number of missing data in each initial sample data segment; wherein the preset sample quantity is greater than or equal to the number of missing sample data in the missing sample data segment; screen the initial sample data segments according to the number of missing data to determine candidate sample data segments; and screen out the reference sample data segments from the candidate sample data segments based on the similarity degree between the candidate sample data segments and the associated sample data segments.

[0079] In an alternative embodiment, the reference sample data segment determining sub-module includes: a candidate sample data segment determining unit, configured to: for each candidate sample data segment, determine the absolute difference between the first data value in the initial sample data segment and the first data value in the associated sample data segment; and when the absolute difference is less than a preset difference threshold and the number of missing data in the candidate sample data segment is less than a preset sample missing threshold, use the candidate sample data segment as the candidate sample data segment.

[0080] In an alternative embodiment, the reference sample data segment determination sub-module further includes: a reference sample data segment determination unit, configured to: for each candidate sample data segment, use the first associated sample data quantity of sample data in the candidate sample data segment as the sample data to be analyzed; for each sample data to be analyzed, determine the corresponding control data from the associated sample data segment for the sample data to be analyzed, and determine the data similarity value between the sample data to be analyzed and the control data; when the sum value of the data similarity values of all the sample data to be analyzed in the candidate sample data segment is less than or equal to a preset similarity threshold, use the candidate sample data segment as the reference sample data segment.

[0081] In an alternative embodiment, the reference sample data segment determination unit includes: a data similarity value determination sub-unit, configured to: when the control data is vacant data, use a preset similarity value as the data similarity value between the sample data to be analyzed and the control data; when the control data is non-vacant data, use the square of the numerical difference between the sample data to be analyzed and the control data as the data similarity value.

[0082] The device status data filling device provided by the embodiments of the present invention can execute the device status data filling method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0083] Figure 4 It is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Figure 4 It shows a block diagram of an exemplary computer device 12 suitable for implementing the embodiments of the present invention. Figure 4 The shown computer device 12 is only an example, and should not bring any limitation to the functions and usage scope of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capabilities and can be configured in the device status data filling device.

[0084] As Figure 4 shown, the computer device 12 is presented in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0085] The bus 18 can be one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an Accelerated Graphics Port, a processor bus, or a local bus using any of the various bus structures. By way of example, these architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

[0086] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including both volatile and nonvolatile media, removable and non-removable media.

[0087] The system memory 28 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. The computer device 12 can further include other removable / non-removable, volatile / nonvolatile computer system storage media. By way of example only, a storage system 34 can be used for reading from and writing to non-removable, nonvolatile magnetic media ( Figure 4 not shown and typically called a "hard disk drive"). Although Figure 4 not shown in the figure, a disk drive for reading from and writing to a removable nonvolatile disk (such as a "floppy disk"), and an optical disk drive for reading from and writing to a removable nonvolatile optical disk (such as a CD-ROM, DVD-ROM, or other optical media) can be provided. In these cases, each drive can be connected to the bus 18 by one or more data media interfaces. The system memory 28 can include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of the embodiments of the present invention.

[0088] A program / utility 40 having a set (at least one) of program modules 42 can be stored, for example, in the system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, and an implementation of a network environment may be included in each or some combination of these examples. The program modules 42 generally carry out the functions and / or methods in the embodiments described in the present invention.

[0089] The computer device 12 can also communicate with one or more external devices 14 (such as a keyboard, a pointing device, a display 24, etc.), and can also communicate with one or more devices that enable a user to interact with the computer device 12, and / or communicate with any device that enables the computer device 12 to communicate with one or more other computing devices (such as a network card, a modem, etc.). Such communication can be carried out through the input / output (I / O) interface 22. Moreover, the computer device 12 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 20. As Figure 4 shown, the network adapter 20 communicates with other modules of the computer device 12 through the bus 18. It should be understood that although Figure 4 not shown in the figure, other hardware and / or software modules can be used in combination with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0090] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28. For example, it implements the device status data filling method provided by the embodiments of the present invention. The method includes:

[0091] Obtaining the to-be-processed device status data of the target pumped-storage device; wherein, the to-be-processed device status data includes at least one of operating temperature, voltage, current, and pressure pulsation;

[0092] Determining the missing data segments in the to-be-processed device status data and the associated data segments corresponding to the missing data segments;

[0093] Inputting the missing data segments and the associated data segments into a pre-trained target data filling model to obtain the target data filling result corresponding to the missing data segments.

[0094] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the device status data filling method provided by any embodiment of the present invention, including:

[0095] Obtaining the to-be-processed device status data of the target pumped-storage device; wherein, the to-be-processed device status data includes at least one of operating temperature, voltage, current, and pressure pulsation;

[0096] Determining the missing data segments in the to-be-processed device status data and the associated data segments corresponding to the missing data segments;

[0097] Input the vacant data segment and the associated data segment into a pre-trained target data filling model to obtain the target data filling result corresponding to the vacant data segment.

[0098] The computer storage medium of the embodiments of the present invention may adopt any combination of one or more computer-readable media. The computer-readable media may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0099] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which the computer-readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0100] The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0101] Computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0102] Those of ordinary skill in the art should understand that the above-described modules or steps of the present invention may be implemented using a general-purpose computing device. They may be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Optionally, they may be implemented using program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps of them may be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0103] Note that the above is only the preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments here. Various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments may be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for filling device status data, characterized in that, Including: Obtain the device status data to be processed of the target pumped storage device; wherein, the device status data to be processed includes at least one of operating temperature, voltage, current, and pressure pulsation; Determine the missing data segments in the device status data to be processed and the associated data segments corresponding to the missing data segments; Input the missing data segments and the associated data segments into a pre-trained target data filling model to obtain the target data filling result corresponding to the missing data segments; The training process of the target data filling model includes: Obtain a preliminary data filling model and device status sample data; Determine the missing sample data segments and associated sample data segments in the device status sample data; Traverse each sample data in the device status sample data, select data with a preset sample quantity from the device status sample data starting from the sample data, obtain the initial sample data segments corresponding to each sample data, and determine the number of missing data in each initial sample data segment; wherein, the preset sample quantity is greater than or equal to the number of missing sample data in the missing sample data segments; For each initial sample data segment, determine the absolute difference between the first data value in the initial sample data segment and the first data value in the associated sample data segment; When the absolute difference is less than a preset difference threshold and the number of missing data in the initial sample data segment is less than a preset sample missing threshold, use the initial sample data segment as a candidate sample data segment; Based on the similarity degree between the candidate sample data segments and the associated sample data segments, screen out the reference sample data segments from the candidate sample data segments; Input the missing sample data segments, the associated sample data segments, and all the reference sample data segments into the preliminary data filling model to obtain the sample data filling result corresponding to the missing sample data segments; Determine a filling loss function according to the sample data filling result and the preset data label of the device status sample data, and adjust the preliminary data filling model according to the filling loss function to obtain the target data filling model.

2. The method according to claim 1, wherein The determining the missing data segments in the device status data to be processed and the associated data segments corresponding to the missing data segments includes: Identify the missing data in the device status data to be processed, and use the data segments with the number of consecutive missing data greater than a preset missing threshold as the missing data segments; Determine the associated data segments corresponding to the missing data segments according to the status data of a preset associated quantity before the missing data segments in the device status data to be processed.

3. The method according to claim 1, characterized in that, The screening out the reference sample data segments from the candidate sample data segments based on the similarity degree between the candidate sample data segments and the associated sample data segments includes: For each candidate sample data segment, use the sample data of the associated sample data quantity in the candidate sample data segment as the sample data to be analyzed; For each sample data to be analyzed, determine the control data corresponding to the sample data to be analyzed from the associated sample data segments, and determine the data similarity value between the sample data to be analyzed and the control data; In the case where the sum value of the data similarity values of all the sample data to be analyzed in the candidate sample data segment is less than or equal to a preset similarity threshold, use the candidate sample data segment as the reference sample data segment.

4. The method according to claim 3, wherein The determination of the data similarity value between the sample data to be analyzed and the control data includes: In the case where the control data is missing data, use a preset similarity value as the data similarity value between the sample data to be analyzed and the control data; In the case where the control data is non-missing data, use the square of the numerical difference between the sample data to be analyzed and the control data as the data similarity value.

5. A device status data filling device, characterized in that, The device includes: A status data acquisition module for acquiring the device status data to be processed of the target pumped-storage equipment; wherein, the device status data to be processed includes at least one of operating temperature, voltage, current, and pressure pulsation; A data segment determination module for determining the missing data segments in the device status data to be processed and the associated data segments corresponding to the missing data segments; A data filling module for inputting the missing data segments and the associated data segments into a pre-trained target data filling model to obtain the target data filling result corresponding to the missing data segments; A data filling model training module for: acquiring a preliminary data filling model and device status sample data; Determining the missing sample data segments and associated sample data segments in the device status sample data; Traverse each sample data in the device status sample data, select a preset number of data from the device status sample data starting from the sample data to obtain an initial sample data segment corresponding to each sample data, and determine the number of missing data in each initial sample data segment; wherein, the preset number of samples is greater than or equal to the number of missing sample data in the missing sample data segment; For each initial sample data segment, determine the absolute difference between the first data value in the initial sample data segment and the first data value in the associated sample data segment; In the case where the absolute difference is less than a preset difference threshold and the number of missing data in the initial sample data segment is less than a preset sample missing threshold, use the initial sample data segment as a candidate sample data segment; Based on the similarity degree between the candidate sample data segment and the associated sample data segment, screen out the reference sample data segment from the candidate sample data segment; Input the missing sample data segment, the associated sample data segment, and all the reference sample data segments into the preliminary data filling model to obtain the sample data filling result corresponding to the missing sample data segment; Determine a filling loss function according to the sample data filling result and the preset data label of the device status sample data, and adjust the preliminary data filling model according to the filling loss function to obtain the target data filling model.

6. A computer device, characterized in that, The computer device includes: One or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the device status data filling method according to any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the device status data filling method according to any one of claims 1-4.

Images