Data analysis method and device, equipment, storage medium and program product
The use of time series databases and analysis models in energy storage EMS systems addresses inefficiencies in data storage and analysis, enhancing efficiency and accuracy while supporting timely fault detection and risk warnings.
Patent Information
- Application Number
- CN202410983692.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-12-03
AI Technical Summary
Traditional energy storage EMS systems face inefficiencies in data querying and writing due to the use of relational databases, leading to low data analysis efficiency and lag in monitoring.
Implementing a data analysis method that utilizes a time series database for storing sampling signal time series data, enabling efficient data storage and retrieval through sub-library storage, data verification, and using time series data analysis models for comprehensive data analysis.
Improves data query and writing efficiency, reduces storage space, and enhances the accuracy and duration of data analysis by using time series databases and analysis models, facilitating timely fault detection and risk warnings.
Smart Images

Figure CN119066447A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of energy storage technology, and in particular to a data analysis method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art
[0002] As the scale and capacity of energy storage power stations continue to grow, the requirements for the time granularity, data capacity, and storage duration of data stored in the energy storage EMS (Energy Management System) are becoming increasingly higher.
[0003] In traditional technologies, energy storage EMS mostly uses relational databases for data storage.
[0004] However, in order to ensure the normal operation of the energy storage EMS, it is necessary to monitor the operation process of the energy storage EMS. The monitoring process requires continuous data collection and storage, but the efficiency of querying and writing data in relational databases is low, which will lead to low efficiency of data analysis and a certain lag in the monitoring of the energy storage EMS. Summary of the invention
[0005] Based on this, it is necessary to provide a data analysis method, device, computer equipment, computer-readable storage medium and computer program product that can improve data analysis efficiency in response to the above technical problems.
[0006] In a first aspect, the present application provides a data analysis method, which is applied to an energy storage energy management system, wherein the energy storage energy management system includes a time series database, including:
[0007] Writing the received sampling signal timing data into the timing database;
[0008] In response to the data analysis request, acquiring target sampling signal timing data corresponding to the data analysis request from each sampling signal timing data in the timing database;
[0009] The target sampling signal timing data is analyzed by using the timing data analysis model corresponding to the data analysis request.
[0010] In one embodiment, writing the received sampling signal timing data into a timing database includes:
[0011] Acquire sampled signal data and data configuration information corresponding to the sampled signal data, where the data configuration information includes data type information;
[0012] In the case where it is determined based on the data type information that the sampled signal data is sampled signal time series data, based on the data configuration information, locating a target subtable corresponding to the sampled signal time series data from a time series database;
[0013] Write the sampling signal timing data into the target subtable.
[0014] In one embodiment, the data configuration information further includes signal type information, power interval identification information, and power equipment identification information; based on the data configuration information, locating a target subtable corresponding to the sampling signal timing data from a timing database includes:
[0015] Based on the signal type information, locate the target super table corresponding to the sampling signal timing data from the timing database;
[0016] Based on the power interval identification information and the power equipment identification information, a target sub-table corresponding to the sampling signal timing data is located from a plurality of sub-tables corresponding to the target super-table.
[0017] In one embodiment, after writing the sampling signal timing data into the target sub-table, the method further includes:
[0018] Obtaining information about items to be counted, where the information about items to be counted at least includes time information to be counted and label information of time series data to be counted;
[0019] Based on the time information to be counted and the label information of the time series data to be counted, locate the time series data set to be counted from the time series database;
[0020] Using the statistical algorithm corresponding to the statistical item information, statistical analysis is performed on the statistical time series data set to obtain the statistical results corresponding to the statistical item information;
[0021] Write the statistical results into the time series database.
[0022] In one embodiment, the time series data analysis model includes an autoregressive integrated moving average model; before performing data analysis on the target sampling signal time series data using the time series data analysis model corresponding to the data analysis request, the method further includes:
[0023] Obtain historical time series data, and perform stationarity processing on the historical time series data to obtain stationary time series data and differential times;
[0024] Generate autocorrelation diagram and partial autocorrelation diagram based on stationary time series data, and determine the moving average term order and autoregressive term order of the autoregressive integrated moving average model based on the autocorrelation diagram and partial autocorrelation diagram;
[0025] Fits an autoregressive integrated moving average model based on the number of differences, the order of the moving average term, the order of the autoregressive term, and the historical time series data.
[0026] In one embodiment, the data analysis includes fault analysis, which includes fault classification and fault location; the time series data analysis model includes a clustering model and a K nearest neighbor model;
[0027] Through the timing data analysis model corresponding to the data analysis request, the timing data of the target sampling signal is analyzed, including:
[0028] Acquire target multi-source data corresponding to the target sampling signal timing data, the target multi-source data including at least one of monitoring data, electrical data, mechanical data, environmental data and historical data;
[0029] Fusing target multi-source data and target sampling signal time series data to obtain fused data;
[0030] Extract fault features from fused data;
[0031] The clustering model is used to classify faults based on fault characteristics to obtain the target fault type, and the K nearest neighbor model is used to locate faults based on fault characteristics to obtain the target fault location information.
[0032] In a second aspect, the present application further provides a data analysis device, which is applied to an energy storage energy management system, the energy storage energy management system includes a time series database, and the device includes:
[0033] A writing module, used for writing the received sampling signal timing data into the timing database;
[0034] An acquisition module, configured to acquire, in response to a data analysis request, target sampling signal timing data corresponding to the data analysis request from each sampling signal timing data in a timing database;
[0035] The data analysis module is used to perform data analysis on the target sampling signal time series data through the time series data analysis model corresponding to the data analysis request.
[0036] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:
[0037] Writing the received sampling signal timing data into the timing database;
[0038] In response to the data analysis request, acquiring target sampling signal timing data corresponding to the data analysis request from each sampling signal timing data in the timing database;
[0039] The target sampling signal timing data is analyzed by using the timing data analysis model corresponding to the data analysis request.
[0040] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:
[0041] Writing the received sampling signal timing data into the timing database;
[0042] In response to the data analysis request, acquiring target sampling signal timing data corresponding to the data analysis request from each sampling signal timing data in the timing database;
[0043] The target sampling signal timing data is analyzed by using the timing data analysis model corresponding to the data analysis request.
[0044] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:
[0045] Writing the received sampling signal timing data into the timing database;
[0046] In response to the data analysis request, acquiring target sampling signal timing data corresponding to the data analysis request from each sampling signal timing data in the timing database;
[0047] The target sampling signal timing data is analyzed by using the timing data analysis model corresponding to the data analysis request.
[0048] The above-mentioned data analysis method, device, computer equipment, computer-readable storage medium and computer program product, firstly, realize the sub-library storage of the sampled signal data by writing the received sampling signal time series data into the time series database, and use the time series database to store the sampling signal time series data, which can effectively improve the query and writing efficiency of the time series data; then, by responding to the data analysis request, the target sampling signal time series data corresponding to the data analysis request is obtained from each sampling signal time series data in the time series database, and the target sampling signal time series data is efficiently obtained after receiving the data analysis request; then, the target sampling signal time series data is analyzed by the time series data analysis model corresponding to the data analysis request, and the target sampling signal time series data is efficiently analyzed. Many sampling signals in the monitoring process of energy storage equipment have time series properties. The time series database can not only ensure the orderly storage of these time series data, but also improve the efficiency of writing and querying the time series data into the database. The improvement of data query efficiency can effectively shorten the overall time of data analysis and improve the efficiency of data analysis. On the other hand, there may be repeated or slightly changing data in the sampled signal. The time series database can use compression algorithms to effectively reduce the storage space, thereby increasing the storage period, capacity and duration of the sampled signal time series data, providing more sufficient data support for subsequent data analysis, thereby improving the accuracy of subsequent data analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0050] Figure 1 A schematic diagram of a data analysis method in one embodiment;
[0051] Figure 2 is a flow chart of a data analysis method in another embodiment;
[0052] Figure 3 is a flow chart of a data analysis method in yet another embodiment;
[0053] Figure 4 is a schematic diagram of a super table in one embodiment;
[0054] Figure 5 is a schematic diagram of a sub-table in one embodiment;
[0055] Figure 6 is a structural block diagram of a data analysis device in one embodiment;
[0056] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0057] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0058] In an exemplary embodiment, Figure 1 As shown, a data analysis method is provided, which is applied to an energy storage energy management system. This embodiment uses the method applied to a terminal corresponding to the energy storage energy management system as an example, wherein the terminal may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices, and the Internet of Things devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. It is understandable that the method may also be applied to a server corresponding to the energy storage energy management system, and may also be applied to a system including a terminal and a server, and may be implemented through the interaction between the terminal and the server.
[0059] As an example, the energy storage energy management system includes at least a message monitoring module, a sampling statistics module, a data transmission module and a real-time library, and may also include a time series database, a non-time series database, a data analysis module, etc. It is understandable that the database and the data analysis module may be deployed on the same terminal or server, or may be deployed on different terminals or servers respectively, which may be determined according to actual conditions, and this embodiment does not limit this.
[0060] As an example, the time series database can run in multiple threads, and illustratively, the multiple threads can include at least one of a data storage thread, a power outage statistics thread, a daily statistics thread, a sampling statistics thread, and an electric quantity statistics thread, etc. Among them, the data storage thread is used to perform tasks related to data storage; the power outage statistics thread is used to perform power outage statistics or predictions based on sampling information; the daily statistics thread is used to provide daily statistics of data to facilitate monitoring and prediction of data; the sampling statistics thread is used to classify or personalize the storage of sampling information to facilitate data monitoring and calculation; and the electric quantity statistics thread is used to perform tasks related to electric energy statistics.
[0061] In this embodiment, the method includes the following steps S10-S30. Among them:
[0062] Step S10, writing the received sampling signal timing data into a timing database.
[0063] Among them, the energy storage energy management system is a core component of the energy storage system and a bridge connecting the energy storage device and the power grid. It is mainly used to monitor, control and optimize the operation of the energy storage device. A database can be deployed in the energy storage energy management system, and the database includes a time series database and a non-time series database. A non-time series database refers to a database used to process, store and retrieve data other than time series data, and may include at least one of a relational database, a non-relational database, a graph database, etc. A time series database refers to a database used to process, store and retrieve time series data. Time series data refers to a set of ordered data points, each of which is associated with a specific timestamp, indicating that the data is captured or generated at a certain exact moment. Sampled signal time series data refers to sampled signal data with time series properties obtained by sampling the energy storage device by the signal measurement point. A signal measurement point refers to a device used to monitor and collect key data of each energy storage device in the energy storage system, and may include various sensors or other data acquisition devices. The signal measurement point can monitor the working status and performance of each energy storage device in the energy storage system in real time.
[0064] Exemplarily, during the operation of each energy storage device in the energy storage system, each energy storage device can be continuously sampled by at least one signal measuring point, and at least one of the operating data generated by each energy storage device during operation and the environmental data of the environment in which the energy storage device is located can be collected. The signal measuring point can be connected to the real-time library by communication, and the collected operating data and environmental data, etc., can be transmitted and stored in the real-time library as sampled signal data. Furthermore, according to the pre-set database configuration rules, different databases can be allocated for storage for these sampled signal data entering the real-time library, thereby realizing the sub-library storage of the sampled signal data. Exemplarily, the database can include a time series database and a relational database; the sampled signal time series data in the sampled signal data can be written into the time series database, and the sampled signal non-time series data in the sampled signal data other than the sampled signal time series data can be written into the relational database. In this way, since each database can be configured with different storage strategies, the most suitable database can be matched for different data, and the data storage effect of various types of data can be comprehensively improved.
[0065] As an example, after receiving the sampling signal timing data, the sampling signal timing data may be deduplicated to avoid affecting the writing efficiency. The deduplication method includes hash table deduplication technology and the like.
[0066] As an example, in the process of writing the sampling signal time series data into the time series database, data verification technology and data sampling can also be used to ensure the quality of the data and prevent data errors and abnormal data.
[0067] As an example, when creating each database, in addition to standard options such as database name, character encoding, and proofreading rules, you can also specify multiple parameters such as retention period, number of copies, cache size, time accuracy, maximum and minimum number of records in a file block, whether to compress, and the number of days covered by a data file, to increase the freedom of data storage and adapt to different actual needs.
[0068] Step S20, in response to the data analysis request, obtaining the target sampling signal timing data corresponding to the data analysis request from the timing data of each sampling signal in the timing database.
[0069] Among them, data analysis refers to the comprehensive health diagnosis and analysis of the energy storage system from one or more aspects such as data quality, behavior analysis, abnormal aging and risk warning based on the long-term and massive data accumulated by the energy storage energy management system. Data quality analysis refers to the process of evaluating the accuracy, completeness, consistency, timeliness and effectiveness of data. The quality of data directly affects the reliability of decisions and analysis based on these data. The main goal of data quality analysis is to identify and quantify the problems in the data so as to clean and correct the data and improve the overall quality of the data. Behavior analysis refers to the process of analyzing behavioral data that characterizes user behavior. Behavior analysis can include power consumption behavior analysis and management behavior analysis according to different users. Through power consumption behavior analysis, it can help the management of the energy storage system to deeply understand the needs, preferences, pain points and habits of using products or services of users, provide personalized services for different users, optimize user experience, and formulate different transmission strategies, pricing strategies, etc., to optimize product functions. Abnormal aging diagnosis refers to the monitoring process of whether there are abnormal aging faults in various energy storage devices, modules, components, etc. in the energy storage system, so that when abnormal aging faults are detected, early warnings and maintenance can be carried out in time to reduce losses. Risk warning refers to the process of early identification and warning of potential risks in energy storage systems, which can help guide the operation and maintenance of energy storage systems, so as to deal with them before failures occur, thereby effectively preventing the occurrence of failures.
[0070] For example, the user can send a data analysis request to the energy storage management system through the user terminal when data analysis is required according to actual needs. After receiving the data analysis request, the terminal of the energy storage management system extracts data information from the data analysis request, and then queries the sampling signal timing data matching the data information from the timing database based on the data information, and uses the sampling signal timing data matching the data information as the target sampling signal timing data.
[0071] As an example, when a non-time series database is also provided in the energy storage energy management system, after extracting data information from the data request, the terminal of the energy storage energy management system can also query the sampling signal non-time series data matching the data information from the non-time series database based on the data information, and then, the sampling signal non-time series data and the sampling signal time-demand data can be used together for subsequent data analysis.
[0072] Step S30: performing data analysis on the target sampling signal time series data by using the time series data analysis model corresponding to the data analysis request.
[0073] Among them, the time series data analysis model refers to a model that can analyze time series data, such as an elementary model, a clustering model, a recurrent neural network, a convolutional neural network, a large model, etc. For example, a long short-term memory network can be used to predict the electricity demand in the future; a clustering model can be used to classify users and provide similar services to users with similar electricity usage behaviors; an elementary model can also be fitted based on the sampling signal time series data to infer the development trend of the sampling signal time series data in the next period of time. Specifically, different models can be selected and used according to the actual data analysis needs, and this embodiment does not limit this.
[0074] Exemplarily, after acquiring the target sampling signal time series data, the time series data analysis model matching the data analysis request can be queried from at least one preset data analysis model according to the data analysis request, the target sampling signal time series data can be input into the time series data analysis model, and the input target sampling signal time series data can be analyzed by the time series data analysis model to obtain the data analysis result. The data analysis result can be in the form of classification labels or numerical values, for example, it can be the existence or non-existence of data quality problems, or the probability that energy storage device A has fault B, etc.
[0075] As an example, the target sampling signal time series data may be preprocessed before being input into the time series data analysis model. The preprocessing includes at least one of data cleaning, deletion of missing values, deduplication processing, outlier detection and correction, and normalization processing.
[0076] As an example, data analysis includes power consumption behavior analysis; the target sampling signal time series data can be power consumption behavior data, and the time series data analysis model can use a clustering algorithm, such as the K-means clustering algorithm, the K-nearest neighbor algorithm, etc. The K-means clustering algorithm is conducive to the development and processing of large-scale data, and can also significantly improve the processing capacity and make data processing more efficient. Aggregating users with similar power consumption behaviors can facilitate the energy storage system to provide more reasonable services, charge different taxes according to different types of power consumption behaviors, adjust the power transmission efficiency of the power grid according to different types of power consumption behaviors, and improve the utilization rate of electric energy.
[0077] As an example, the electricity consumption behavior data includes load data. A typical load curve can be constructed in advance based on historical load data. After obtaining the load data to be analyzed that needs to be analyzed, a load curve to be analyzed is constructed based on the load data to be analyzed, and the load curve to be analyzed is compared with the typical load curve for similarity. According to the similarity comparison result, it can be analyzed whether the load data to be analyzed contains uncertain loads. Uncertain loads include the impact of unforeseen factors such as prediction errors, weather changes, and emergencies on the load. Furthermore, short-term load prediction and detection of abnormal electricity consumption behavior can be achieved based on uncertain loads through clustering models or neural network models. It is also possible to achieve long-term load prediction by integrating the load data to be analyzed and other related factor data through clustering models or neural network models. Other related factor data include temperature, time, holiday information, etc. Machine learning algorithms can also be used to perceive electricity usage behavior and evaluate load composition based on the load data to be analyzed. Power usage behavior perception refers to identifying and parsing different power-consuming devices or power-consuming activities from the load data to be analyzed, and load composition evaluation refers to analyzing the composition of different types of loads in the power system. Through power usage behavior perception and load composition evaluation, power consumption patterns can be analyzed to improve energy efficiency, optimize power distribution, and provide users with more intelligent services.
[0078] As an example, in addition to the target sampling signal timing data, big data can also be used to obtain target multi-source data related to the target sampling signal timing data, such as monitoring data, electrical data, mechanical data, environmental data, and historical data; further, the large model sampling attention mechanism can be used to perform reasoning analysis on the target sampling signal timing data and the target multi-source data, capture the temporal correlation of the target sampling signal timing data, and the correlation between the target sampling signal timing data and each data in the target multi-source data, so as to predict the failure risk of the energy storage equipment in the future, and issue a risk warning in time when the failure risk is detected to be higher than the preset risk threshold. In this way, a risk warning can be issued before the failure occurs, thereby preventing the failure from occurring and minimizing the loss and safety risks.
[0079] As an example, after performing data analysis on the target sampling signal time series data through the time series data analysis model corresponding to the data analysis request, it may also include: displaying the data analysis result of the data analysis through a user terminal or a management terminal.
[0080] As an example, after performing data analysis on the target sampling signal time series data through the time series data analysis model corresponding to the data analysis request, it can also include: when any fault is detected in the energy storage system, a fault warning is issued. Among them, the method of fault warning can include outputting fault warning information through the target energy storage device with fault or an alarm device set within a certain distance range of the target energy storage device, such as emitting an alarm sound, turning on an alarm light, etc.; the method of fault warning can also include sending fault warning information to the management terminal, such as pop-up window, calling and broadcasting fault warning voice information after the call is connected, etc.
[0081] In the above data analysis method, firstly, by writing the received sampling signal time series data into the time series database, the sub-library storage of the sampling signal data is realized. Using the time series database to store the sampling signal time series data can effectively improve the query and writing efficiency of the time series data; then, by responding to the data analysis request, the target sampling signal time series data corresponding to the data analysis request is obtained from each sampling signal time series data in the time series database, and the efficient acquisition of the target sampling signal time series data is realized after receiving the data analysis request; then, through the time series data analysis model corresponding to the data analysis request, the target sampling signal time series data is analyzed, and the efficient data analysis of the target sampling signal time series data is realized. Many sampling signals in the monitoring process of energy storage equipment have time series properties. The time series database can not only ensure the orderly storage of these time series data, but also improve the efficiency of writing and querying the time series data into the database. The improvement of data query efficiency can effectively shorten the overall time of data analysis and improve the efficiency of data analysis. On the other hand, there may be repeated or slightly changing data in the sampled signal. The time series database can use compression algorithms to effectively reduce the storage space, thereby increasing the storage period, capacity and duration of the sampled signal time series data, providing more sufficient data support for subsequent data analysis, thereby improving the accuracy of subsequent data analysis.
[0082] In an exemplary embodiment, Figure 2 As shown, writing the received sampling signal timing data into the timing database includes steps S11 to S13. Among them:
[0083] Step S11, acquiring sampled signal data and data configuration information corresponding to the sampled signal data, wherein the data configuration information includes data type information.
[0084] It should be noted that the energy storage system involves a large number of sampled signal data with many types. If all of them are mixed and stored together, the accuracy and efficiency of data query and acquisition will be low.
[0085] The data configuration information refers to information related to the database configuration rules of the database, including at least one of data type information, signal type information, power interval identification information, and power equipment identification information. The database configuration rules of the database can be determined in advance according to actual conditions and test results, for example, time series data is stored in a time series database, and non-time series data is stored in a non-time series database; divided tables are stored according to the power equipment identification information, etc.
[0086] Exemplarily, during the operation of each energy storage device in the energy storage system, each energy storage device can be continuously sampled by at least one signal measuring point, and at least one of the operating data generated by each energy storage device during operation and the environmental data of the environment in which the energy storage device is located can be collected. The signal measuring point can transmit and store the collected operating data and environmental data, etc., as sampled signal data to the real-time library through communication connection with the real-time library. Furthermore, these sampled signal data and data configuration information corresponding to the sampled signal data can be obtained from the real-time library.
[0087] Step S12: when it is determined based on the data type information that the sampled signal data is sampled signal time series data, based on the data configuration information, locate the target sub-table corresponding to the sampled signal time series data from the time series database.
[0088] The target sub-table refers to a data storage unit in a time series database for storing time series data of each sampling signal.
[0089] Exemplarily, after determining the data type information corresponding to the sampled signal data, it is possible to first determine whether the sampled signal data is sampled signal time series data based on the data type information. If the sampled signal data is determined to be sampled signal time series data, it can be written into a time series database. If the sampled signal data is determined not to be sampled signal time series data, it can be written into a non-time series database, thereby realizing the sub-library storage of the sampled signal data. If the sampled signal data is determined to be sampled signal time series data, the target sub-table matching the data configuration information can be determined by querying a preset time series database configuration rule.
[0090] As an example, the data type information may be an enablement flag of a time series database. Based on the actual conditions of each signal measurement point, an enablement flag of a time series database may be added to the signal measurement point that can collect time series data in advance; after acquiring the sampled signal data, it may be queried whether there is an enablement flag of a time series database in the signal measurement point information that collects the sampled signal data; if there is an enablement flag of a time series database, it is determined that the sampled signal data is sampled signal time series data and can be written into the time series database; otherwise, it is determined that the sampled signal data is not sampled signal time series data and cannot be written into the time series database.
[0091] Determine that the sampled signal data is sampled signal timing data based on the data type information
[0092] Step S13, writing the sampling signal timing data into the target sub-table.
[0093] Exemplarily, after the target sub-table is determined, a new column may be added to the target sub-table, and the sampling signal timing data may be written into the new column in the target sub-table.
[0094] In this embodiment, by means of sub-library and sub-table, it is possible to achieve classified and orderly storage of a large number of sampling signal data of various types, thereby improving the accuracy and efficiency of data query and acquisition.
[0095] In an exemplary embodiment, the data configuration information also includes signal type information, power interval identification information and power equipment identification information; Figure 3 As shown, based on the data configuration information, locating the target sub-table corresponding to the sampling signal timing data from the timing database includes steps S121 to S122. Among them:
[0096] Step S121: based on the signal type information, locate the target super table corresponding to the sampled signal time series data from the time series database.
[0097] It should be noted that the energy storage system involves a large number of sampling signal time series data with many types. If all of them are mixed and stored together, the accuracy and efficiency of time series data query and acquisition will be low.
[0098] Among them, the super table exists as a template or pattern. The super table defines one or more column families, as well as the attributes and data types of each column. The super table itself does not store any data. The data is stored in its corresponding sub-table. The super table is used to batch create sub-tables with the same structure, which simplifies database management and can uniformly define storage policies, data retention policies, etc. An example of a super table is Figure 4 As shown, Figure 4From left to right, each column is the name of the subtable (name), the creation time of the subtable (created_time), the column family of the subtable (columns), the tag of the subtable (tags), and the number of subtables (tables); yx in the table name means telesignaling, yc means telemetry, and ym means telepulse. The number after the abbreviation indicates the subtable code; the number 4001 in the column family of the subtable indicates that the subtable has a total of 4001 columns. It can also be specified that the first column is a timestamp, and the 2nd to 4001st columns are the collected physical quantities; the data 2 in the tag of the subtable indicates that there are two tags in total, and it can also be specified that these two tags are power interval identification information and power equipment identification information. Secondly, the zeroth column is specified as "TS" and the data type is timestamp. The following four thousand columns are specified as the collected physical quantities, and the data type is floating point. Finally, "BAYID" is specified as interval, and "UNITID" is specified as equipment as static tags. The subtable is a data storage unit created based on the supertable. Each subtable can inherit the structure and attributes of the supertable, but can write different data. Subtables are used to store specific data related to a specific instance or object. For example, different subtables can correspond to different devices, sensors, or data sources, and each subtable stores the time series data of its specific entity. An example of a subtable is Figure 5 As shown, Figure 5 The first column is the timestamp, and each subsequent column is the specific value of the collected sampling signal timing data.
[0099] The sampling signal time series data collected in the energy storage system are usually divided into three signal types: telemetry signal, telesignaling signal and telepulse signal. That is, the signal type of the sampling signal time series data can include one of telemetry signal, telesignaling signal and telepulse signal. Telemetry signal usually refers to continuously changing analog quantity, such as voltage, current, temperature, etc., and the data has continuity and high sampling rate; telesignaling signal usually refers to discrete switch quantity or status information, such as circuit breaker status, alarm signal, etc., the data is discrete and usually updated at a low frequency; telepulse signal usually refers to pulse count or cumulative quantity, such as meter reading, flow count, etc., the data is cumulative and the update may be irregular. The characteristics of the signal time series data used by these three signal types and the requirements for writing and query are different. Therefore, different types of sampling signal time series data can be stored separately, that is, different super tables can be created in advance for telemetry, telesignaling and telepulse.
[0100] Exemplarily, after determining that the sampled signal data is sampled signal time series data, a target supertable matching the signal type information may be determined from various supertables. For example, if the signal type information is a telemetry signal, the supertable corresponding to the telemetry signal is determined as the target supertable.
[0101] Step S122: based on the power interval identification information and the power equipment identification information, locate the target sub-table corresponding to the sampling signal timing data from the multiple sub-tables corresponding to the target super-table.
[0102] Each subtable can inherit the structure and attributes of the supertable and have its own label, which can be used to distinguish different subtables. In the monitoring process of energy storage equipment, different devices usually need to be monitored separately. Therefore, when creating a subtable, the power interval identification information and the power equipment identification information can be used as the label of the subtable, and then the sampling signal timing data can be written and stored based on the power interval identification information and the power equipment identification information.
[0103] Exemplarily, after determining the target supertable, the target subtable matching both the power interval identification information and the power equipment identification information can be determined from multiple subtables corresponding to the target supertable by querying the label information of each subtable in the target supertable.
[0104] As an example, a time series database can be created through TDengine (a time series database). First, different super tables are created for the three different types of sampled signal time series data: telemetry, telesignaling, and telepulse. When creating a super table, it is necessary to provide the table name, data column definition, label, etc. In the definition of the data column, the first column can be determined as a timestamp, and the other columns can be determined as the collected physical quantity. The data type can be at least one of integer, floating point, string, etc. The data type of the label can be at least one of integer, floating point, string, etc., and the static attribute of the signal measurement point can be used as a label. The subtable can be created dynamically, that is, after receiving the sampled signal time series data, if a matching target subtable already exists, the sampled signal time series data is written into the target subtable. If there is no matching target subtable, a new target subtable is created first, and then the sampled signal time series data is written into the newly created target subtable. The table name of the subtable can be "super table name + power interval identification information + power equipment identification information". To write to a time series database, you can use the structured query language, or execute INSERT statements through a connector to insert data, or manually enter INSERT statements through the TDengine CLI (Command-LineInterface, command line) to insert data. To query a time series database, you can use the structured query language to query, or manually execute structured query language ad hoc queries through the TDengine command line tool taos. A time series database can also use structured query language to define real-time stream transformations. When data is written to the source table of the stream, the data will be automatically processed in the defined manner, and the results will be pushed to the destination table according to the defined trigger mode. It provides a lightweight solution to replace complex stream processing systems, and can provide millisecond-level calculation result delays in the case of high-throughput data writing.
[0105] In this embodiment, by means of sub-tables, it is possible to achieve classified and orderly storage of a large number of sampling signal time series data of various types, thereby improving the accuracy and efficiency of time series data query and acquisition.
[0106] In an exemplary embodiment, after writing the sampling signal timing data into the target sub-table, steps S14 to S17 are also included. Among them:
[0107] Step S14, obtaining information of items to be counted, where the information of items to be counted at least includes time information to be counted and time series data label information to be counted.
[0108] Among them, the items to be counted refer to statistical items that can perform statistical analysis on the sampling signal time series data in advance. The items to be counted can correspond to statistical algorithms one by one. The statistical algorithms include summing, finding the maximum value, finding the minimum value, finding the average value, and counting the proportion of data above a preset threshold, etc. The specific method can be determined according to the actual situation, and this embodiment does not limit this. The time to be counted refers to the time range to which the time series data to be counted belongs. The label information of the time series data to be counted includes power interval identification information and power equipment identification information, which is used to determine the subtable to which the time series data to be counted belongs.
[0109] Exemplarily, the items to be counted can be preset, and in the process of writing sampling signal timing data into the timing database, data statistics can be performed regularly or in a triggered manner. When performing data statistics, the information of the items to be counted that are preset can be obtained first.
[0110] Step S15: based on the time information to be counted and the label information of the time series data to be counted, locate the time series data set to be counted from the time series database.
[0111] Exemplarily, based on the label information of the time series data to be counted, the sub-table to which the time series data to be counted belongs is located from the time series database, and then the time series data to be counted that matches the time information to be counted is filtered out from the sampling signal time series data stored in the sub-table to which the time series data to be counted belongs, and these time series data to be counted constitute the time series data set to be counted.
[0112] Step S16, using a statistical algorithm corresponding to the statistical item information, statistically analyzing the statistical time series data set to be statistically analyzed, and obtaining statistical results corresponding to the statistical item information.
[0113] Exemplarily, after determining the time series data set to be counted, the corresponding statistical algorithm is queried according to the statistical item information, and the statistical algorithm corresponding to the sampled statistical item information is used to perform statistical analysis on the time series data set to be counted to obtain the statistical result corresponding to the statistical item information.
[0114] Step S17, writing the statistical results into the time series database.
[0115] For example, after obtaining the statistical results, the statistical results can be written into a statistical table in the time series database, so that if these statistical results are needed later, they can be directly queried from the statistical table without taking up time for data analysis.
[0116] In this embodiment, after the sampling signal timing data is written into the timing database and before data analysis, data statistics are performed on the sampling signal timing data written into the timing database. In this way, when these statistical results are needed during data analysis, they can be directly obtained without taking up data analysis time, thereby improving the efficiency of data analysis.
[0117] In an exemplary embodiment, the time series data analysis model includes an autoregressive integrated moving average model; before the time series data analysis model corresponding to the data analysis request is used to analyze the time series data of the target sampling signal, steps S01 to S03 are also included. Among them:
[0118] Step S01, obtain historical time series data, and perform stationarity processing on the historical time series data to obtain stationary time series data and differential times.
[0119] Among them, the Autoregressive Integrated Moving Average Model (ARIMA model for short) is a statistical model used to analyze and predict time series data, which can predict future time series values. Before using the ARIMA model for prediction, it is necessary to first build the ARIMA model based on historical time series data.
[0120] For example, the historical time series data collected within a period of time can be obtained from the time series database, and the ADF (Augmented Dickey-Fuller, an augmented form of the Dickey-Fuller test) test can be used to test the stationarity of the historical time series data. Stationarity means that the mean and variance of the historical time series data remain unchanged at different time points. The null hypothesis of the ADF test is that the historical time series data has a unit root, that is, the historical time series data is not stationary. If the test statistic is less than the critical value and the p-value is less than the significance level, for example, 0.05, the null hypothesis can be rejected, that is, the historical time series data does not have a unit root, and the historical time series data can be considered to be stationary; if the test statistic is greater than the critical value, or the p-value is greater than the significance level, the null hypothesis cannot be rejected, that is, the historical time series data has a unit root, and the historical time series data is non-stationary. After the historical time series data is determined to be non-stationary, the historical time series data can be differentiated, and then the ADF test can be used to test the stationarity of the differentiated historical time series data. This cycle can be repeated until the differentiated historical time series data is stable, and the stable time series data and the number of differences can be obtained.
[0121] Step S02, generating an autocorrelation diagram and a partial autocorrelation diagram based on the stationary time series data, and determining the moving average term order and the autoregressive term order of the autoregressive integrated moving average model based on the autocorrelation diagram and the partial autocorrelation diagram.
[0122] Exemplarily, an autocorrelation graph and a partial autocorrelation graph are generated based on stationary time series data. If the ACF graph decays slowly and the PACF graph is truncated after a certain point, the lag order of the truncation point can be used as a candidate value of the autoregressive term order; if the PACF graph shows a clear peak, the autoregressive term order is greater than 0; if the PACF graph decays slowly and the ACF graph is truncated after a certain point, the lag order of the truncation point can be used as a candidate value of the moving average term order; if the ACF graph shows a clear peak, the moving average term order is greater than 0; in this way, the candidate value of the moving average term order and the candidate value of the autoregressive term order can be determined. Furthermore, the Akaike information criterion, the modified Akaike information criterion or the Bayesian information criterion can be used to compare models under different parameter combinations, so as to determine the optimal one from the candidate values of the moving average term order and the candidate values of the autoregressive term order as the moving average term order and the autoregressive term order.
[0123] Step S03, fitting an autoregressive integrated moving average model based on the number of differences, the order of the moving average term, the order of the autoregressive term and the historical time series data.
[0124] Exemplarily, after determining the number of differences, the order of moving average terms and the order of autoregressive terms, the basic form of the autoregressive integrated moving average model can be determined, and then the autoregressive integrated moving average model can be fitted based on the historical time series data according to the determined basic form to obtain the autoregressive integrated moving average model.
[0125] As an example, after determining the number of differences, the order of the moving average term, and the order of the autoregressive term, and before fitting the autoregressive integrated moving average model, you can also perform model diagnostics to check the residuals of the model. The ideal residuals should have zero mean, constant variance, and no autocorrelation. If the residuals show certain patterns, the model parameters may need to be adjusted.
[0126] In this embodiment, many sampling signal time series data collected in the energy storage system often change over time, showing seasonal and periodic non-stationary characteristics. The autoregressive integrated moving average model can convert the non-stationary time series into a stationary series and can capture the autocorrelation between data, thereby realizing future predictions based on past observations and improving the timeliness of data analysis.
[0127] In an exemplary embodiment, the data analysis result includes the fault analysis result, the data analysis includes the fault analysis, the fault analysis includes the fault classification and the fault location; the time series data analysis model includes the clustering model and the K nearest neighbor model; the target sampling signal time series data is analyzed by the time series data analysis model corresponding to the data analysis request, and the step S31 to step S34 are also included. Among them:
[0128] Step S31, obtaining target multi-source data corresponding to the target sampling signal timing data, where the target multi-source data includes at least one of monitoring data, electrical data, mechanical data, environmental data and historical data.
[0129] It should be noted that the autoregressive integrated moving average model can better capture the linear relationship between data, however, it is difficult to capture the nonlinear relationship between data. Therefore, if there is a significant nonlinear relationship between the detection target and the input data, the detection accuracy will be low. In addition to the sampling signal time series data, the data related to the detection target can also include multi-source data, which includes sampling signal non-time series data and data provided by other terminals, such as monitoring data, electrical data, mechanical data, environmental data, historical data, etc.
[0130] Exemplarily, after receiving a data analysis request, data analysis type information can be first extracted from the data analysis request, and the data analysis type corresponding to the data analysis request can be determined based on the data analysis type information. When the data analysis type is determined to be fault analysis, based on the correspondence between the preset data analysis type and the data, the target sampling signal timing data required for fault analysis and the target multi-source data corresponding to these target sampling signal timing data can be determined, and the target sampling signal timing data and the target multi-source data corresponding to these target sampling signal timing data can be obtained.
[0131] Step S32, fusing the target multi-source data and the target sampling signal time series data to obtain fused data.
[0132] Exemplarily, the target multi-source data and the target sampling signal time series data can be fused by means of data association, cleaning, feature extraction, etc. to obtain fused data.
[0133] Step S33: extracting fault features from the fused data.
[0134] Exemplarily, feature extraction is performed on the fused data through a feature extraction model to obtain fault features.
[0135] Step S34, classify the faults based on the fault features using a clustering model to obtain a target fault type, and locate the faults based on the fault features using a K nearest neighbor model to obtain target fault location information of the fault.
[0136] Exemplarily, after receiving a data analysis request, the data analysis type information can be first extracted from the data analysis request. Based on the correspondence between the preset data analysis type and the time series data analysis model, it can be determined that the time series data analysis model for fault analysis includes a clustering model and a K nearest neighbor model, wherein the clustering model is used for fault classification and the K nearest neighbor model is used for fault location. After extracting the fault feature, the fault feature can be input into the clustering model, and the similarity between the fault feature and each fault type is detected by the clustering model. The fault type with the highest similarity to the fault feature and the maximum similarity greater than the preset similarity threshold is output as the target fault type. If the maximum similarity is not greater than the preset similarity threshold, it can be considered that the energy storage device currently has no fault. Synchronously, or after the energy storage device is detected to have a fault through the clustering model, the fault feature is input into the K nearest neighbor model, and the K nearest neighbor model is used to find the K training samples closest to the fault feature based on the fault feature, that is, the K nearest neighbors, and analyze the common features of the K nearest neighbors, so as to determine the position with the highest probability of the fault occurring, and obtain the target fault location information.
[0137] In this embodiment, by combining the target sampling signal timing data and the target multi-source data, more accurate fault classification and fault location are achieved. The K nearest neighbor model can be used to isolate faults, effectively reduce the scope of faults, facilitate rapid location of fault points, and contribute to timely resolution of faults, thereby improving the safety of the energy storage system.
[0138] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0139] Based on the same inventive concept, the embodiment of the present application also provides a data analysis device for implementing the data analysis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more data analysis device embodiments provided below can refer to the limitations on the data analysis method above, and will not be repeated here.
[0140] In an exemplary embodiment, Figure 6As shown, a data analysis device is provided, which is applied to an energy storage energy management system. The energy storage energy management system includes a time series database. The device includes: a writing module 602, an acquisition module 604 and a data analysis module 606, wherein:
[0141] A writing module 602 is used to write the received sampling signal time series data into a time series database;
[0142] The acquisition module 604 is used to acquire the target sampling signal timing data corresponding to the data analysis request from the timing data of each sampling signal in the timing database in response to the data analysis request;
[0143] The data analysis module 606 is used to perform data analysis on the target sampling signal time series data through the time series data analysis model corresponding to the data analysis request.
[0144] In an exemplary embodiment, the writing module 602 is further configured to:
[0145] Acquire sampled signal data and data configuration information corresponding to the sampled signal data, where the data configuration information includes data type information;
[0146] In the case where it is determined based on the data type information that the sampled signal data is sampled signal time series data, based on the data configuration information, locating a target subtable corresponding to the sampled signal time series data from a time series database;
[0147] Write the sampling signal timing data into the target subtable.
[0148] In an exemplary embodiment, the data configuration information further includes signal type information, power interval identification information and power equipment identification information; the writing module 602 is also used to:
[0149] Based on the signal type information, locate the target super table corresponding to the sampling signal timing data from the timing database;
[0150] Based on the power interval identification information and the power equipment identification information, a target sub-table corresponding to the sampling signal timing data is located from a plurality of sub-tables corresponding to the target super-table.
[0151] In an exemplary embodiment, the writing module 602 is further configured to:
[0152] Obtaining information about items to be counted, where the information about items to be counted at least includes time information to be counted and label information of time series data to be counted;
[0153] Based on the time information to be counted and the label information of the time series data to be counted, locate the time series data set to be counted from the time series database;
[0154] Using the statistical algorithm corresponding to the statistical item information, statistical analysis is performed on the statistical time series data set to obtain the statistical results corresponding to the statistical item information;
[0155] Write the statistical results into the time series database.
[0156] In an exemplary embodiment, the time series data analysis model includes an autoregressive integrated moving average model; the data analysis device also includes a model building module; before performing data analysis on the target sampling signal time series data through the time series data analysis model corresponding to the data analysis request, the model building module is used to:
[0157] Obtain historical time series data, and perform stationarity processing on the historical time series data to obtain stationary time series data and differential times;
[0158] Generate autocorrelation diagram and partial autocorrelation diagram based on stationary time series data, and determine the moving average term order and autoregressive term order of the autoregressive integrated moving average model based on the autocorrelation diagram and partial autocorrelation diagram;
[0159] Fits an autoregressive integrated moving average model based on the number of differences, the order of the moving average term, the order of the autoregressive term, and the historical time series data.
[0160] In an exemplary embodiment, data analysis includes fault analysis, which includes fault classification and fault location; the time series data analysis model includes a clustering model and a K nearest neighbor model; the data analysis module 606 is also used to:
[0161] Acquire target multi-source data corresponding to the target sampling signal timing data, the target multi-source data including at least one of monitoring data, electrical data, mechanical data, environmental data and historical data;
[0162] Fusing target multi-source data and target sampling signal time series data to obtain fused data;
[0163] Extract fault features from fused data;
[0164] The clustering model is used to classify faults based on fault characteristics to obtain the target fault type, and the K nearest neighbor model is used to locate faults based on fault characteristics to obtain the target fault location information.
[0165] Each module in the above data analysis device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.
[0166] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a data analysis method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.
[0167] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0168] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.
[0169] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.
[0170] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.
[0171] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0172] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.
[0173] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0174] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A data analysis method, characterized in that: The method is applied to an energy storage energy management system, which includes a time series database, including: Writing the received sampling signal timing data into the timing database; In response to a data analysis request, acquiring target sampling signal timing data corresponding to the data analysis request from each sampling signal timing data in the timing database; The target sampling signal timing data is analyzed using a timing data analysis model corresponding to the data analysis request.
2. The method according to claim 1, characterized in that The step of writing the received sampling signal timing data into the timing database comprises: Acquire sampled signal data and data configuration information corresponding to the sampled signal data, wherein the data configuration information includes data type information; In a case where it is determined based on the data type information that the sampled signal data is sampled signal time series data, based on the data configuration information, locating a target subtable corresponding to the sampled signal time series data from a time series database; The sampling signal timing data is written into the target sub-table.
3. The method according to claim 2, characterized in that The data configuration information also includes signal type information, power interval identification information, and power equipment identification information; and locating a target subtable corresponding to the sampling signal timing data from a timing database based on the data configuration information includes: Based on the signal type information, locate a target super table corresponding to the sampled signal timing data from a timing database; Based on the power interval identification information and the power equipment identification information, a target sub-table corresponding to the sampling signal timing data is located from a plurality of sub-tables corresponding to the target super-table.
4. The method according to claim 2, characterized in that: After writing the sampling signal timing data into the target sub-table, the method further includes: Acquire information of items to be counted, wherein the information of items to be counted at least includes time information to be counted and time series data label information to be counted; Based on the time information to be counted and the label information of the time series data to be counted, locating the time series data set to be counted from the time series database; Using the statistical algorithm corresponding to the statistical item information, statistical analysis is performed on the time series data set to be counted, to obtain the statistical results corresponding to the statistical item information; The statistical results are written into the time series database.
5. The method according to any one of claims 1 to 4, characterized in that: The time series data analysis model includes an autoregressive integrated moving average model; before performing data analysis on the target sampling signal time series data using the time series data analysis model corresponding to the data analysis request, the method further includes: Acquire historical time series data, and perform stationarity processing on the historical time series data to obtain stationary time series data and differential times; Generate an autocorrelation graph and a partial autocorrelation graph based on the stationary time series data, and determine the moving average term order and the autoregressive term order of the autoregressive integrated moving average model based on the autocorrelation graph and the partial autocorrelation graph; An autoregressive integrated moving average model is fitted based on the number of differences, the order of the moving average term, the order of the autoregressive term and the historical time series data.
6. The method according to any one of claims 1 to 4, characterized in that: The data analysis includes fault analysis, which includes fault classification and fault location; the time series data analysis model includes a clustering model and a K nearest neighbor model; The step of performing data analysis on the target sampling signal time series data by using the time series data analysis model corresponding to the data analysis request includes: Acquire target multi-source data corresponding to the target sampling signal timing data, wherein the target multi-source data includes at least one of monitoring data, electrical data, mechanical data, environmental data and historical data; Fusing the target multi-source data and the target sampling signal time series data to obtain fused data; extracting fault features from the fused data; The clustering model is used to classify faults based on the fault features to obtain a target fault type, and the K nearest neighbor model is used to locate faults based on the fault features to obtain target fault location information of the fault.
7. A data analysis device, characterized in that: The device is applied to an energy storage management system, the energy storage management system includes a time series database, and the device includes: A writing module, used for writing the received sampling signal timing data into the timing database; An acquisition module, configured to acquire, in response to a data analysis request, target sampling signal timing data corresponding to the data analysis request from each sampling signal timing data in the timing database; The data analysis module is used to perform data analysis on the target sampling signal time series data through the time series data analysis model corresponding to the data analysis request.
8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
Citation Information
Patent Citations
Time sequence analysis method based on time sequence database
CN114254012A
Intelligent substation data storage method based on object model
CN115525654A
Production equipment visualization method based on operation data analysis
CN116955468A
Time series data analysis method and device, computing equipment and storage medium
CN117688054A
Distribution transformer operation fault telemetering method and system
CN118130940A
