Multi-dimensional data fusion monitoring method, monitoring system, electronic device and storage medium

Abstract

The application discloses a monitoring method of multi-dimensional data fusion, comprising the following steps: acquiring historical multi-element data, preprocessing historical multi-element data sources, and obtaining historical multi-dimensional data in a unified data format; extracting features of the historical multi-dimensional data in the unified data format, and obtaining a historical multi-dimensional feature vector; establishing an electrical equipment state monitoring data model by using the historical multi-dimensional feature vector; acquiring real-time multi-element data, preprocessing real-time multi-element data sources, and obtaining real-time multi-dimensional data in a unified data format; extracting features of the real-time multi-dimensional data in the unified data format, and obtaining a real-time multi-dimensional feature vector; projecting the real-time multi-dimensional feature vector onto the electrical equipment state monitoring data model, and obtaining a projection result; comparing the projection result with a preset threshold, and obtaining a data monitoring result. The application also discloses a monitoring system, an electronic device and a storage medium. The application overcomes the problems of different data focuses and structures of business systems of enterprises, information islands and repeated cross problems.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of data processing, and particularly relates to a multi-dimensional data fusion monitoring method, a monitoring system, an electronic device and a storage medium. BACKGROUND

[0002] The petrochemical industry currently has the following problems: a large number of electrical equipment types, a large quantity, and up to 100 communication protocols of data sources; various electrical monitoring systems are not connected to each other, forming multiple information islands; various system data sources are independent of each other, lack a unified data warehouse, cannot realize data sharing and sharing, cannot apply data mining technology to realize fault early warning, and cannot obtain technical and management benefits from big data. Moreover, due to the lack of a unified alarm information platform, when handling sudden defects, the operation and maintenance personnel may face problems such as information overlap, information explosion, and different time scales. In addition, the data of each platform is not open to the outside, and high interface fees need to be paid to the equipment manufacturer when accessing, and the data belonging to the user is taken as one's own. The one-time investment for equipment intelligent transformation is large; the labor intensity of inspection work is large, and the quality and efficiency need to be improved, and the like.

[0003] In the prior art, there are more studies on specific electrical equipment online state monitoring at home and abroad, and there are very few studies on overall electrical state monitoring platforms. The existing literature has proposed an online state monitoring implementation method and device structure of a transformer; some documents detail the transformer fault mechanism, online monitoring theory and method; similarly, other documents also separately study specific electrical equipment, such as the fault mechanism and online monitoring analysis method of GIS switch cabinet and power transmission line. These specific research contents and achievements belong to the basic work of the electrical equipment state monitoring platform, constitute part of the sensing layer or data layer of the entire platform system, and provide basic technical support for the multi-dimensional data fusion monitoring method.

[0004] In the prior art, there is no effective technical solution to the problem that multiple data sources cannot be fused in a unified platform for unified management and data state monitoring. SUMMARY

[0005] To solve the problems in the prior art, the present application proposes a multi-dimensional data fusion monitoring method, system, electronic device and storage medium, which applies multi-dimensional feature data fusion technology, breaks through multiple information islands, and makes data of different sources unified in the multi-dimensional data fusion monitoring system, thereby providing an accurate data basis for the next step of data state detection.

[0006] In a first aspect, the present application provides a multi-dimensional data fusion monitoring method, comprising the following steps:

[0007] Obtain historical multi-dimensional data, and pre-process the historical multi-dimensional data source to obtain historical multi-dimensional data in a unified data format;

[0008] Extract features from the historical multi-dimensional data in the unified data format to obtain a historical multi-dimensional feature vector;

[0009] Establish an electrical equipment state monitoring data model using the historical multi-dimensional feature vector;

[0010] Save the historical multi-dimensional data, the historical multi-dimensional feature vector, and the electrical equipment state monitoring data model to an electrical equipment state monitoring data warehouse according to a pre-established index;

[0011] Obtain real-time multi-dimensional data, and pre-process the real-time multi-dimensional data source to obtain real-time multi-dimensional data in a unified data format;

[0012] Extract features from the real-time multi-dimensional data in the unified data format to obtain a real-time multi-dimensional feature vector;

[0013] Project the real-time multi-dimensional feature vector onto the established electrical equipment state monitoring data model to obtain a projection result;

[0014] Compare the projection result with a preset threshold to obtain a data monitoring result.

[0015] The historical multi-dimensional data and the real-time multi-dimensional data include device information data, power grid topology data, measurement data, state monitoring data, geographic information data, weather data, customer information data, metering data, and power outage information data.

[0016] The state monitoring data includes high-voltage electrical equipment operation state monitoring, secondary equipment state monitoring, sensor state monitoring, communication state monitoring, environment state monitoring, and auxiliary state monitoring.

[0017] The feature extraction method includes an expert system method, a correlation analysis method, a neural network method, and a clustering analysis method.

[0018] The multi-dimensional feature vector includes a static feature vector, a dynamic feature vector, a time series feature vector, a spatial feature vector, a multi-time scale vector, and a multi-space dimension vector.

[0019] The method for obtaining the historical multi-dimensional data or the real-time multi-dimensional data includes:

[0020] Corresponding interfaces are established for different historical multi-dimensional data or real-time multi-dimensional data;

[0021] Different historical multi-dimensional data or real-time multi-dimensional data is obtained through the corresponding interfaces.

[0022] The preprocessing of the historical multi-dimensional data or real-time multi-dimensional data includes:

[0023] The acquired different historical data or real-time multi-dimensional data is filtered to remove invalid data, and filtered data is obtained.

[0024] The filtered data is de-duplicated to remove repeated data, and de-duplicated data is obtained.

[0025] The de-duplicated data is data-transformed to obtain the historical multi-dimensional data or real-time multi-dimensional data in a unified data format.

[0026] The pre-established index includes a time index, a feature index, a vector index, and a data model index.

[0027] The real-time multi-dimensional data in the unified data format and the real-time multi-dimensional feature vector are saved into the electrical equipment state monitoring data warehouse according to the pre-established index.

[0028] In a second aspect, the application provides a multi-dimensional data fusion monitoring system, which includes a historical data acquisition module, a feature extraction module, a model establishment module, a real-time data acquisition module, a projection module, and a comparison module.

[0029] The historical data acquisition module and the real-time data acquisition module are connected to the feature extraction module, the feature extraction module is connected to the model establishment module and the projection module, and the projection module is connected to the comparison module.

[0030] The historical data acquisition module is configured to acquire historical multi-dimensional data and preprocess the historical multi-dimensional data source to obtain historical multi-dimensional data in a unified data format.

[0031] The feature extraction module is configured to extract features from the historical multi-dimensional data in the unified data format or real-time multi-dimensional data to obtain a historical multi-dimensional feature vector or real-time multi-dimensional data.

[0032] The model establishment module is configured to establish an electrical equipment state monitoring data model using the historical multi-dimensional feature vector.

[0033] The projection module is configured to project the real-time multi-dimensional feature vector onto the established electrical equipment state monitoring data model to obtain a projection result.

[0034] The comparison module is configured to compare the projection result with a preset threshold to obtain a data monitoring result.

[0035] In a third aspect, the application provides an electronic device, which includes:

[0036] one or more processors;

[0037] a memory;

[0038] one or more application programs stored in the memory and configured to be loaded and run by the one or more processors so as to perform the monitoring method of multi-dimensional data fusion.

[0039] In a fourth aspect, the present application provides a computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the monitoring of multi-dimensional data fusion according to the first aspect or any possible implementation manner of the first aspect.

[0040] The present application achieves the following beneficial effects:

[0041] The present application provides a monitoring method, system, electronic device and storage medium for multi-dimensional data fusion, which breaks through multiple information islands, and makes data from different sources unified in the monitoring system of multi-dimensional data fusion, thereby providing an accurate data basis for the next data state detection.

[0042] The present application solves the problems of large amount of information for electrical equipment state monitoring and information uncertainty, establishes a unified electrical equipment state monitoring data model, effectively overcomes the problems of different data emphasis and structure, information islands and repeated intersection of various business systems of enterprises, and is conducive to fully mining the value of electrical equipment big data, and provides strong computing and analysis conditions for electrical equipment operation analysis, risk identification, state early warning and fault processing. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 A monitoring method flow chart of multi-dimensional data fusion for an embodiment of the present application;

[0044] Figure 2 A data acquisition flow chart for an embodiment of the present application;

[0045] Figure 3 A preprocessing flow chart for an embodiment of the present application;

[0046] Figure 4 A principle block diagram of a monitoring system of multi-dimensional data fusion for an embodiment of the present application;

[0047] Figure 5 An electronic device example diagram for an embodiment of the present application. DETAILED DESCRIPTION

[0048] The present application will be further described below in conjunction with the drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and cannot be used to limit the protection scope of the present application.

[0049] Embodiment:

[0050] The embodiment applies the monitoring method for multi-dimensional data fusion proposed in the application to Jinling Petrochemical. Currently, Jinling Petrochemical is facing the following problems: 1. A large number of electrical equipment, widely distributed, difficult to manage: Jinling Petrochemical currently has a long service life, a large number of electrical equipment, and a wide distribution. Whether the equipment is in normal operation needs to be combined with the running information at the time of inspection and needs to be combined with the running condition of the process device, which is difficult to manage. Taking a general-scale power distribution substation as an example, about 20 running equipment need to be checked and measured, and it takes one hour to complete the inspection of the substation according to the specified items. Completing the daily inspection of all running electrical equipment in the whole plant requires a large amount of work. 2. Few maintenance personnel, low efficiency of manual inspection: Corresponding to the large number of inspection equipment, the actual number of maintenance personnel is small, and the efficiency of manual communication with the process device is low. The equipment responsible by the electrical operation department has increased with the continuous increase of new devices, and the substation and switch station have also increased. At present, the distribution equipment is increasing, but the maintenance personnel is decreasing, which actually increases the maintenance difficulty. 3. The secondary equipment maintenance is not in place, and the production hidden danger cannot be eliminated as soon as possible: Non-planned shutdown of low-pressure pump equipment is inevitable in daily production operation, especially for non-important link pumps and devices that do not directly cause serious failure. Shutdown of these devices usually reduces the operation reliability of the key link or becomes an inducing factor for major accidents. Due to the above reasons, the daily maintenance work is difficult to reach the standard, so these problems cannot be found in time, and the production hidden danger cannot be eliminated as soon as possible, and the reliability of production operation is greatly reduced.

[0051] In a first aspect, the application provides a multi-dimensional data fusion monitoring method, as shown in Figure 1 The method comprises the following steps:

[0052] Step S1: Obtain historical multi-dimensional data, and pre-process the historical multi-dimensional data source to obtain historical multi-dimensional data in a unified data format;

[0053] Step S2: Extract features from the historical multi-dimensional data in the unified data format to obtain a historical multi-dimensional feature vector;

[0054] Step S3: Use the historical multi-dimensional feature vector to establish an electrical equipment state monitoring data model;

[0055] Step S4: Save the historical multi-dimensional data, the historical multi-dimensional feature vector, and the electrical equipment state monitoring data model to an electrical equipment state monitoring data warehouse according to a pre-established index;

[0056] Step S5: Obtain real-time multi-dimensional data, and pre-process the real-time multi-dimensional data source to obtain real-time multi-dimensional data in a unified data format;

[0057] Step S6: Extract features from the real-time multidimensional data in the unified data format to obtain a real-time multidimensional feature vector;

[0058] Step S7: Project the real-time multidimensional feature vector onto the established electrical equipment condition monitoring data model to obtain the projection result;

[0059] Step S8: Compare the projection results with the preset threshold to obtain the data monitoring results.

[0060] The historical and real-time multi-dimensional data include: equipment information data, power grid topology data, measurement data, status monitoring data, geographic information data, weather data, customer information data, metering data, and power outage information data.

[0061] The status monitoring data includes: high-voltage electrical equipment operation status monitoring, secondary equipment status monitoring, sensor status monitoring, communication status monitoring, environmental status monitoring, and auxiliary status monitoring.

[0062] (1) Monitoring the operating status of high-voltage electrical equipment

[0063] High-voltage electrical equipment mainly includes transformers, circuit breakers, capacitors, surge arresters, etc. The operating status of high-voltage electrical equipment is not entirely contained within SCADA. Taking transformers as an example, it is necessary to monitor their oil chromatography status, partial discharge status, core status, winding temperature status, etc. This type of information is extremely important and needs to be integrated into this system. Data already included in SCADA, such as equipment voltage, current, active power, power factor, etc., are then connected to this system via an interface.

[0064] (2) Secondary equipment status monitoring

[0065] The status of secondary equipment includes fault and abnormal status information of equipment such as digital electrical measuring instruments, digital relay protection, conventional relay protection devices, and automatic control devices.

[0066] (3) Sensor status monitoring

[0067] Sensor status includes the effective / failure status information of devices such as current transformers, voltage transformers, temperature sensors, vibration sensors, and arc sensors.

[0068] (4) Communication status monitoring

[0069] Communication status includes information such as the connectivity of secondary devices and communication equipment, and the bit error rate.

[0070] (5) Environmental status monitoring

[0071] The environmental state includes temperature, humidity, water immersion, lightning strike and other information in the station, indoor, cabinet and loop.

[0072] (6) Auxiliary state monitoring

[0073] The auxiliary state monitoring is the state monitoring of other auxiliary systems, for example, independent arc protection system, wireless temperature measurement system, temperature and humidity control system, fire alarm system, online insulation monitoring system and video monitoring system, which can be connected to the platform and the data can be fused into the corresponding equipment.

[0074] The feature extraction method includes expert system method, correlation analysis method, neural network method and cluster analysis method.

[0075] The feature extraction by the expert system method, correlation analysis method, neural network method and cluster analysis method can be realized by the person skilled in the art, and the present application does not expand the description.

[0076] The multi-dimensional feature vector includes static feature vector, dynamic feature vector, time sequence feature vector, spatial feature vector, multi-time scale vector and multi-space dimension vector.

[0077] The historical multi-element data or real-time multi-element data is obtained, as shown in Figure 2 The method comprises the following steps:

[0078] Step S100: establishing a corresponding interface for different historical multi-element data or real-time multi-element data;

[0079] Step S101: obtaining different historical multi-element data or real-time multi-element data through the corresponding interface.

[0080] The historical multi-element data or real-time multi-element data is preprocessed to obtain historical multi-dimensional data or real-time multi-element data in a unified data format, as shown in Figure 3 The method comprises the following steps:

[0081] Step S200: filtering the obtained different historical data or real-time multi-element data to remove invalid data and obtain filtered data;

[0082] Step S201: removing repeated data from the filtered data to obtain de-duplicated data;

[0083] Step S202: data transformation is performed on the de-duplicated data to obtain historical multi-dimensional data or real-time multi-element data in a unified data format.

[0084] The pre-established index includes time index, feature index, vector index and data model index.

[0085] The real-time multidimensional data in the unified data format and the real-time multidimensional feature vector are saved into the electrical equipment state monitoring data warehouse according to pre-established indexes.

[0086] In a second aspect, the application provides a multidimensional data fusion monitoring system, as shown in the accompanying drawings, comprising a historical data acquisition module, a feature extraction module, a model establishment module, a real-time data acquisition module, a projection module, and a comparison module. Figure 4

[0087] The historical data acquisition module and the real-time data acquisition module are connected to the feature extraction module, the feature extraction module is connected to the model establishment module and the projection module, and the projection module is connected to the comparison module.

[0088] The historical data acquisition module is configured to acquire historical multidimensional data, and pre-process the historical multidimensional data source to obtain historical multidimensional data in a unified data format.

[0089] The feature extraction module is configured to extract features from the historical multidimensional data in the unified data format or real-time multidimensional data to obtain a historical multidimensional feature vector or real-time multidimensional data.

[0090] The model establishment module is configured to establish an electrical equipment state monitoring data model using the historical multidimensional feature vector.

[0091] The projection module is configured to project the real-time multidimensional feature vector onto the established electrical equipment state monitoring data model to obtain a projection result.

[0092] The comparison module is configured to compare the projection result with a preset threshold to obtain a data monitoring result.

[0093] In a third aspect, the application provides an electronic device, comprising:

[0094] One or more processors;

[0095] A memory;

[0096] One or more application programs stored in the memory and configured to be loaded and run by the one or more processors to perform the multidimensional data fusion monitoring method.

[0097] As shown in the accompanying drawings, the electronic device 100 comprises a processor 101 and a memory 103. Figure 5 The structure of the electronic device 100 does not constitute a limitation on the embodiments of the application.

[0098] The structure of the electronic device 100 does not constitute a limitation on the embodiments of the application.​

[0099] The processor 101 can be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure. The processor 101 can also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0100] The bus 102 can include a path for transmitting information between the above-mentioned components. The bus 102 can be a PCI bus or an EISA bus, etc. The bus 102 can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 5 In the figure, only one thick line is used, but it does not mean that there is only one bus or only one type of bus.

[0101] The memory 103 can be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, an EEPROM, a CD-ROM or other optical disk storage, an optical disk storage (including a compact disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but not limited thereto.

[0102] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the multi-dimensional data fusion monitoring method of the first aspect or any possible implementation manner of the first aspect.

[0103] The applicant of the present application has made a detailed description and explanation of the embodiments of the present application in combination with the drawings of the specification, but those skilled in the art should understand that the above embodiments are only preferred embodiments of the present application, and the detailed description is only to help the reader better understand the spirit of the present application, and is not a limitation on the protection scope of the present application. On the contrary, any improvement or modification made on the basis of the spirit of the present application should fall within the protection scope of the present application.

Claims

1. A monitoring method for multidimensional data fusion, characterized in that, Includes the following steps: Historical multidimensional data is acquired and preprocessed to obtain historical multidimensional data in a unified data format; Features are extracted from historical multidimensional data in the unified data format to obtain historical multidimensional feature vectors; The historical multidimensional feature vectors are used to establish an electrical equipment condition monitoring data model; The historical multidimensional data, historical multidimensional feature vectors, and electrical equipment condition monitoring data model are saved into the electrical equipment condition monitoring data warehouse according to the pre-established index. Acquire real-time multidimensional data and preprocess the real-time multidimensional data to obtain real-time multidimensional data in a unified data format; Features are extracted from real-time multidimensional data in the unified data format to obtain a real-time multidimensional feature vector; The real-time multidimensional data in the unified data format and the real-time multidimensional feature vector are saved into the electrical equipment condition monitoring data warehouse according to the pre-established index; The real-time multidimensional feature vector is projected onto the established electrical equipment condition monitoring data model to obtain the projection result; The projection results are compared with the preset threshold to obtain the data monitoring results; The historical multi-dimensional data and real-time multi-dimensional data include: equipment information data, power grid topology data, measurement data, status monitoring data, geographic information data, weather data, customer information data, metering data, and power outage information data; The methods for extracting features include: expert system methods, association analysis methods, neural network methods, and cluster analysis methods; The acquisition of historical multivariate data or real-time multivariate data includes: Establish corresponding interfaces for different historical or real-time multivariate data. Different historical multivariate data or real-time multivariate data can be obtained through the corresponding interfaces; The historical or real-time multidimensional data is preprocessed to obtain historical or real-time multidimensional data in a unified data format, including: The different historical data or real-time multi-dimensional data are filtered to remove invalid data and obtain the filtered data. The filtered data is deduplicated to remove duplicate data, resulting in deduplicated data. The deduplicated data is transformed to obtain historical or real-time multidimensional data in a unified data format.

2. The monitoring method for multidimensional data fusion according to claim 1, characterized in that: The status monitoring data includes: high-voltage electrical equipment operation status monitoring, secondary equipment status monitoring, sensor status monitoring, communication status monitoring, environmental status monitoring, and auxiliary status monitoring.

3. A monitoring system for multi-dimensional data fusion, characterized in that: The monitoring method for implementing the multidimensional data fusion according to any one of claims 1-2 includes: a historical data acquisition module, a feature extraction module, a model building module, a real-time data acquisition module, a projection module, and a comparison module; The historical data acquisition module and the real-time data acquisition module are respectively connected to the feature extraction module. The feature extraction module is respectively connected to the model building module and the projection module. The projection module is connected to the comparison module. The historical data acquisition module is used to acquire historical multidimensional data and preprocess the historical multidimensional data to obtain historical multidimensional data in a unified data format. The feature extraction module is used to extract features from historical multidimensional data or real-time multidimensional data in the unified data format to obtain historical multidimensional feature vectors or real-time multidimensional data. The model building module is used to build an electrical equipment condition monitoring data model using the historical multidimensional feature vectors; the projection module is used to project the real-time multidimensional feature vectors onto the built electrical equipment condition monitoring data model to obtain the projection result. The comparison module is used to compare the projection result with a preset threshold to obtain the data monitoring result.

4. An electronic device, characterized in that, The electronic device includes: One or more processors; Memory; One or more applications, which are stored in the memory and configured to be loaded and run by the one or more processors to perform the monitoring method of multidimensional data fusion as described in any one of claims 1 to 2.

5. A computer-readable storage medium, characterized in that, It stores a computer program that can be loaded and run by a processor to perform the monitoring method of multidimensional data fusion as described in any one of claims 1 to 2.

Images