Substation hidden danger monitoring and early warning method, system, equipment and medium

Abstract

The invention discloses a substation hidden danger monitoring and early warning method, system and device and a medium, and the method comprises the steps: obtaining substation hidden danger monitoring data, and carrying out the data preprocessing; performing time synchronization and space alignment on the preprocessed data, and dynamically allocating weights in combination with data credibility to realize multi-source data weighted fusion; and carrying out hidden danger analysis on the fused data through a neural network, and issuing early warning information based on a detected hidden danger type. According to the invention, through cooperative monitoring of a satellite remote sensing technology and a ground sensor network, omnibearing monitoring of an internal equipment operation state and a surrounding macroscopic environment of the transformer substation is covered; by adopting a space-time alignment technology and a dynamic weight fusion algorithm, the monitoring accuracy is improved; the anti-interference capability of satellite data and the real-time performance of ground data are complemented, and a data correction technology in preprocessing is combined, so that the robustness of the system is improved; through the convolutional neural network, automatic hidden danger identification is realized, and response time and labor cost are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of substation hidden danger monitoring, and in particular to a substation hidden danger monitoring and early warning method, system, device and medium. BACKGROUND

[0002] In recent years, with the rapid development of satellite remote sensing technology and Internet of Things technology, substation hidden danger monitoring gradually evolves from single ground monitoring to space-ground integration collaborative monitoring. The application of high-resolution optical satellites and synthetic aperture radar satellites makes it possible to obtain substation and surrounding geographical environment information in a wide range and all-weather. Ground monitoring systems rely on sensor networks, such as temperature, vibration, partial discharge sensors and image acquisition equipment, to realize real-time monitoring of device operating status. In terms of data processing, multi-source data fusion technology and deep learning algorithms are introduced into hidden danger analysis to improve the intelligent level of the monitoring system. However, existing technologies mainly focus on the optimization of single data source or the monitoring of local scenes. The deep fusion and collaborative analysis of satellite and ground data are still in the exploratory stage, especially in terms of reliability, data complementary mechanism and refinement of hidden danger determination logic in complex environments.

[0003] Traditional substation monitoring technology mainly relies on ground sensor networks, which has the following limitations: limited monitoring range, ground equipment is constrained by geographical conditions and installation location, and can only cover the internal and near-distance surrounding areas of the substation, and cannot obtain macro-environment information; single data leads to one-sided analysis, and single ground sensor data cannot comprehensively evaluate the correlation between device operating status and environmental factors; insufficient anti-interference capability, ground sensors are prone to data drift or interruption in bad weather, while satellite data is stable but lacks real-time collaborative correction mechanism with ground data, resulting in reduced monitoring reliability. In addition, existing data fusion methods mainly use fixed weight allocation and do not adjust the fusion strategy based on dynamic data quality, which is difficult to meet the needs of complex scenarios. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a substation hidden danger monitoring and early warning method and system to solve the problems of limited monitoring range leading to missed detection of macro-environment hidden dangers, single data reducing analysis accuracy, insufficient anti-interference capability affecting reliability, and how to realize dynamic collaborative fusion of satellite and ground data to improve early warning accuracy.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a substation hidden danger monitoring and early warning method, comprising: obtaining substation hidden danger monitoring data and performing data preprocessing; time synchronization and space alignment are performed on the preprocessed data, and multi-source data weighted fusion is realized by combining data credibility dynamic weight distribution; hidden danger analysis is performed on the fused data through a neural network, and early warning information is issued based on the detected hidden danger type.

[0008] As a preferred scheme of the substation hidden danger monitoring and early warning method of the present application, the method comprises: the step of obtaining substation hidden danger monitoring data comprises the steps of obtaining satellite data and obtaining ground data.

[0009] As a preferred scheme of the substation hidden danger monitoring and early warning method of the present application, the method comprises: the data preprocessing comprises the steps of preprocessing the obtained satellite data; and preprocessing the obtained ground data.

[0010] As a preferred scheme of the substation hidden danger monitoring and early warning method of the present application, the method comprises: the time synchronization comprises time alignment of the satellite data and the ground data according to timestamp information of the satellite data and the ground data; and the space alignment comprises conversion of the satellite data and the ground data to a unified geographic coordinate system.

[0011] As a preferred scheme of the substation hidden danger monitoring and early warning method of the present application, the method comprises: the multi-source data weighted fusion comprises: a weighted fusion algorithm is used to calculate the weight distribution according to the credibility of the satellite data and the ground data, different weights are assigned to different types of data, and weighted fusion is performed based on the assigned weights to obtain multi-source data fusion.

[0012] As a preferred scheme of the substation hidden danger monitoring and early warning method of the present application, the method comprises: the hidden danger analysis comprises extracting spatiotemporal features in the fused data using a convolutional neural network, and determining a hidden danger type based on a preset logic; and the hidden danger type comprises foreign object intrusion hidden danger, equipment overheating hidden danger and structure deformation hidden danger.

[0013] As a preferred scheme of the substation hidden danger monitoring and early warning method of the present application, the method comprises: the early warning information based on the detected hidden danger type comprises: when a hidden danger is detected, corresponding early warning information is generated according to the hidden danger type and the severity.

[0014] In a second aspect, the present application provides a substation hidden danger monitoring and early warning system, comprising: a data processing module, a data fusion module, and a hidden danger early warning module; the data processing module is configured to acquire substation hidden danger monitoring data and perform data preprocessing; the data fusion module is configured to synchronize the preprocessed data in time and align the preprocessed data in space, and combine data credibility to dynamically assign weights to realize multi-source data weighted fusion; the hidden danger early warning module is configured to analyze hidden dangers through a neural network based on the fused data, and issue early warning information based on the detected hidden danger types.

[0015] In a third aspect, the present application provides an electronic device, comprising:

[0016] a memory and a processor;

[0017] The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions, which realize the steps of the substation hidden danger monitoring and early warning method when executed by the processor.

[0018] In a fourth aspect, the present application provides a computer readable storage medium storing computer executable instructions, which realize the steps of the substation hidden danger monitoring and early warning method when executed by the processor.

[0019] Compared with the prior art, the present application has the following beneficial effects: the present application breaks through the physical limitations of traditional ground monitoring through the cooperative monitoring of satellite remote sensing technology and ground sensor networks, covering all-around monitoring of the internal equipment operating state of the substation and the surrounding macro environment; through the use of time-space alignment technology and dynamic weight fusion algorithm, the dynamic assignment of weights according to data credibility not only more accurately associates the equipment state and external environmental factors, but also improves the accuracy of monitoring; through the complementation of the anti-interference capability of satellite data and the real-time performance of ground data, combined with the data correction technology in preprocessing, the robustness of the system is improved; through the spatio-temporal feature extraction and classification decision mechanism based on the convolutional neural network, the automatic identification of hidden dangers is realized, and the response time and labor cost are reduced. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0021] Figure 1 The overall flowchart of the substation hidden danger monitoring and early warning method according to an embodiment of the present application is shown in the figure.

[0022] Figure 2The overall flowchart of the substation hidden danger monitoring and early warning system is shown in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0024] Embodiment 1, refer to Figure 1 In an embodiment of the present application, a substation hidden danger monitoring and early warning method is provided, comprising:

[0025] S1: obtaining substation hidden danger monitoring data and performing data preprocessing;

[0026] S2: time synchronizing and spatially aligning the preprocessed data, and combining data credibility to dynamically assign weights to realize multi-source data weighted fusion;

[0027] S3: analyzing hidden dangers through a neural network based on the fused data, and publishing early warning information based on the detected hidden danger types.

[0028] It should be noted that the substation is a core hub facility in the power system, used for voltage conversion and power transmission, to ensure efficient transmission of electric energy to the user end. At the same time, as a power grid node, the substation is used for power system stability control, power quality regulation, and in the new type of power system, it bears the function of photovoltaic, wind energy and other new energy grid connection interface. Therefore, the hidden danger monitoring and early warning of the substation is very important.

[0029] Therefore, in view of the above-mentioned problems of substation hidden danger monitoring and early warning, through the steps of S1-S3, through the space dimension expansion, data dimension fusion and algorithm accuracy improvement of satellite and ground cooperative monitoring technology, the transformation from "passive repair" to "active defense" is realized, which is an important technical breakthrough for building the safety line of the new type of power system.

[0030] Embodiment 2, refer to Figure 1 In an embodiment of the present application, based on the above-mentioned embodiment, a substation hidden danger monitoring and early warning method is provided.

[0031] In the present application, the step S1 of obtaining substation hidden danger monitoring data comprises the following steps A1-A2:

[0032] A1: obtaining satellite data;

[0033] A2: obtaining ground data.

[0034] Specifically, in the A1 step, satellite data collection is performed by periodically acquiring high-definition images of the substation and the surrounding area using a high-resolution optical satellite to obtain macro information.

[0035] The macro information includes information such as surrounding topography, building distribution, and vegetation coverage.

[0036] The synthetic aperture radar satellite is used to monitor the substation and the surrounding area under all-weather conditions to obtain ground deformation information.

[0037] It should be noted that the synthetic aperture radar satellite is particularly suitable for monitoring geological changes that may affect the safety of the substation.

[0038] In the A2 step, ground data collection is performed by installing built-in sensors on key equipment inside the substation to collect real-time equipment operating state data, and by deploying environmental sensors around the substation to obtain surrounding environmental data.

[0039] The built-in sensors include temperature sensors, vibration sensors, and partial discharge sensors.

[0040] The environmental sensors include temperature and humidity sensors, wind speed sensors, and image sensors.

[0041] In the embodiments of the present application, the data preprocessing in step S1 includes the following steps B1-B2:

[0042] B1: Preprocessing the acquired satellite data;

[0043] B2: Preprocessing the acquired ground data.

[0044] Specifically, in the B1 step, the high-resolution optical satellite image is subjected to radiation correction to eliminate radiation differences caused by environmental factors.

[0045] The environmental factors include atmospheric scattering, solar elevation angle, and other factors.

[0046] Geometric correction is performed to correct image geometric distortion caused by systematic error factors.

[0047] The systematic error factors include satellite attitude, earth curvature, and other factors.

[0048] The synthetic aperture radar satellite data is subjected to denoising processing to remove coherent speckle noise and improve image quality.

[0049] In the B2 step, the temperature, vibration, and other sensor data are subjected to denoising by a filtering algorithm to remove abnormal values caused by electromagnetic interference, sensor self-error, and other factors.

[0050] Grayscale and enhance image sensor data to improve image clarity and feature extraction accuracy.

[0051] In the embodiment of the present application, the time synchronization in step S2 includes aligning the satellite data and the ground data in time based on their timestamp information.

[0052] Specifically, based on the satellite impact shooting time and the ground sensor recording time, an interpolation algorithm is used to align the timestamp of the ground data to the time point of the satellite data;

[0053] The conditions for difference calculation include: if the time difference between ground data and satellite data is ≤10min, then direct difference is used;

[0054] If the time difference between ground data and satellite data is greater than 10 minutes, the data of this period will be discarded to avoid error accumulation.

[0055] The time alignment accuracy of the interpolated data must reach ±1s, and the data consistency must be verified through timestamp verification. For periodically missing data, historical data from the same period will be used to fill in the gaps, and the filling ratio shall not exceed 10% of the total data volume.

[0056] In an optional implementation, time synchronization can also be achieved through network clock synchronization based on the NTP protocol. By deploying a network time protocol server to calibrate the time of ground sensor nodes, the ground sensor clock error is controlled through periodic time calibration, and the pulse per second information is synchronously accessed at the satellite data receiving end. The GPS output 1PPS signal is used to trigger the timestamp mark of the satellite data acquisition device, thereby achieving microsecond-level time base unification of satellite and ground data.

[0057] In another optional implementation, time synchronization can also be achieved by constructing a time estimation model including satellite orbit parameters and combining it with a drift supplement algorithm of a real-time clock chip built into the ground sensor.

[0058] In the embodiment of the present application, the spatial alignment in step S2 includes converting the satellite data and the ground data into a unified geographic coordinate system.

[0059] Specifically, the original coordinate system of the satellite image and the GPS positioning coordinates of the ground sensor are accurately matched through GIS software;

[0060] After spatial alignment, all data are converted to the same projection coordinate system, and data association relationships are established through spatial indexing. For areas with complex terrain, digital elevation models are additionally introduced for elevation correction to ensure three-dimensional spatial consistency.

[0061] In an optional embodiment, spatial alignment can also be achieved by deploying high-precision ground control points around the substation, obtaining the WGS84 coordinates of the control points through GNSS0RTK measurement, and performing polynomial fitting on the GCPs as reference points when performing geometric correction on the satellite image, thereby achieving spatial alignment between the satellite image and the ground sensor.

[0062] In the embodiment of the present application, the weighted fusion of multi-source data implemented in step S2 includes adopting a weighted fusion algorithm, performing weight distribution calculation according to the credibility of satellite data and ground data, assigning different weights to different types of data, and performing weighted fusion based on the assigned weights to obtain multi-source data fusion.

[0063] Specifically, for equipment temperature monitoring, ground temperature sensor data has high credibility and is assigned a high weight;

[0064] For monitoring of surrounding terrain changes, satellite data has obvious advantages and high credibility, so it is given high weight;

[0065] Among them, the system calculates data credibility based on data source characteristics, historical data performance, and monitoring target matching;

[0066] The data source characteristics include satellite data and ground data;

[0067] The historical data performance includes statistical sensor historical error rate and data integrity rate;

[0068] The matching degree of the monitoring target includes: in equipment status monitoring, the ground sensor directly contacts the equipment, so the matching degree is high; in terrain deformation monitoring, the satellite SAR data has obvious advantages in penetration and periodic monitoring, so the matching degree is high;

[0069] Credibility levels include high credibility, medium credibility, and low credibility;

[0070] Among them, high credibility means data error rate less than 5%, strong real-time performance, and direct relevance to the monitoring target;

[0071] Medium credibility means the data error rate is 5%-15%, and it is an auxiliary monitoring target;

[0072] Low confidence refers to error rate > 15% or non-directly relevant data;

[0073] The weight distribution calculation method includes goal-oriented weighting, dynamic adjustment mechanism and weighted fusion;

[0074] The target-oriented empowerment includes assigning higher weights to directly relevant data sources according to monitoring objectives;

[0075] The dynamic adjustment mechanism comprises presetting a basic weight through historical data training or expert experience, and dynamically fine-tuning the weight in combination with real-time data quality;

[0076] The weighted fusion comprises a weighted fusion formula, which is expressed as:

[0077] S r =∑(W i ·D i )

[0078] wherein W i is a data source reliability weight, and D i is a preprocessed data value.

[0079] In the embodiment of the application, the hidden danger analysis in step S3 comprises extracting spatial and temporal features in the fused data by using a convolutional neural network, and determining a hidden danger type based on a preset logic;

[0080] The hidden danger type comprises a foreign matter intrusion hidden danger, an equipment overheating hidden danger, and a structure deformation hidden danger.

[0081] Specifically, the fused data is analyzed by using the convolutional neural network, the fused satellite image and the ground equipment state data are taken as inputs of the CNN, potential hidden dangers are identified by using the trained model, and it is determined whether there are foreign matter intrusion, equipment overheating or structure deformation hidden dangers by analyzing distance changes between buildings and the transformer substation in the satellite image, abnormal fluctuations of the ground equipment temperature and vibration.

[0082] The working steps of the CNN model are as follows:

[0083] The input layer: the fused satellite image and the ground sensor data are converted into a unified format multi-dimensional feature matrix, wherein the satellite image comprises high-resolution optical images for extracting macro spatial features such as terrain, buildings and vegetation, and synthetic aperture radar deformation data are converted into a gray-scale image to represent ground displacement; the ground data comprises sensor numerical data and image data which are normalized and spliced into a feature vector;

[0084] The feature extraction layer: satellite image features are extracted by a CNN convolution layer, target detection and deformation gradient, and ground data features are extracted by a full connection layer or one-dimensional convolution to extract time series anomalies, and image data features are extracted by two-dimensional convolution to extract local features;

[0085] The fusion feature layer: the feature vectors of the satellite and the ground data are spatiotemporally aligned, and are combined into a comprehensive feature vector by weighted fusion to highlight the advantages of different data types;

[0086] The classification decision layer: the comprehensive features are input into a full connection layer, a hidden danger type probability is output by a Softmax classifier, a hidden danger judgment logic is set, and the hidden danger type is determined.

[0087] It should be noted that in the hidden danger type judgment logic, the foreign matter intrusion hidden danger judgment logic includes satellite data features, buildings, trees, construction machinery, and substation distance shortening in optical images, and data dumping caused by landslides;

[0088] In ground data features, image sensors capture abnormal objects entering the warning area, or damaged fences, water accumulation, and foreign matter accumulation;

[0089] The convolutional neural network superimposes satellite macro position changes and ground image real-time monitoring results. If non-fixed objects enter the safety distance, it is determined that the hidden danger type is foreign matter intrusion.

[0090] The device overheating hidden danger judgment logic includes ground data features, temperature sensor values exceeding the normal threshold of the device, and no significant abnormalities in vibration and partial discharge data;

[0091] In satellite data features, surrounding building construction blocks air vents, high humidity affects heat dissipation efficiency, or vegetation and building distribution changes cause direct sunlight to intensify;

[0092] The convolutional neural network excludes device internal faults by continuously exceeding the threshold temperature and deteriorating environmental heat dissipation conditions, and determines that the hidden danger type is a device overheating hidden danger.

[0093] The structure deformation hidden danger judgment logic includes satellite data features, SAR monitoring of ground surface deformation, and optical image terrain anomalies;

[0094] In ground data features, vibration sensors detect periodic abnormal vibrations, and device appearance images show structural cracks or positions;

[0095] The convolutional neural network couples ground deformation data with device vibration and appearance features, and the deformation trend continues, determining that the hidden danger type is a structure deformation risk.

[0096] In the embodiments of the present application, the step S3 of issuing a warning message based on the detected hidden danger type includes generating corresponding warning information according to the hidden danger type and severity when a hidden danger is detected.

[0097] In an optional embodiment, for device overheating hidden dangers, the warning information includes device location, current temperature, normal temperature range, extent of exceeding the normal range, and recommended cooling measures, etc.

[0098] In another optional embodiment, it is pushed to relevant operation and maintenance personnel in time through SMS, email, and substation monitoring system pop-up windows, etc.

[0099] Embodiment 3, refer to Figure 2The above is a schematic scheme of a substation hidden danger monitoring and early warning method. It should be noted that the technical scheme of the substation hidden danger monitoring and early warning system belongs to the same concept as the technical scheme of the substation hidden danger monitoring and early warning method described above. The technical scheme of the substation hidden danger monitoring and early warning system in this embodiment is not described in detail, and the description of the technical scheme of the substation hidden danger monitoring and early warning method described above can be referred to.

[0100] The embodiment also provides a substation hidden danger monitoring and early warning system, comprising: a data processing module, a data fusion module, and a hidden danger early warning module.

[0101] The data processing module is configured to acquire substation hidden danger monitoring data and perform data preprocessing. The data fusion module is configured to perform time synchronization and spatial alignment on the preprocessed data, and dynamically allocate weights in combination with data reliability to realize multi-source data weighted fusion. The hidden danger early warning module is configured to analyze hidden dangers through a neural network based on the fused data, and issue early warning information based on the detected hidden danger types.

[0102] The embodiment also provides an electronic device suitable for substation hidden danger monitoring and early warning, comprising: a memory and a processor. The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the substation hidden danger monitoring and early warning method proposed in the above embodiment.

[0103] The embodiment also provides a storage medium having a computer program stored thereon. The program is executed by a processor to implement the substation hidden danger monitoring and early warning method proposed in the above embodiment.

[0104] The storage medium proposed in the embodiment belongs to the same inventive concept as the substation hidden danger monitoring and early warning method proposed in the above embodiment. Technical details not described in detail in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0105] From the above description of the embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software and necessary general hardware. Of course, it can also be implemented by hardware. Based on such understanding, the technical scheme of the present application or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk or an optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the method of each embodiment of the present application.

[0106] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not to limit the present application. Although the present application is explained in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application, and all of them should be covered in the scope of the claims of the present application.

Claims

1. A method for monitoring and early warning of hidden dangers in a substation, characterized in that: include: Obtain substation hidden danger monitoring data and perform data preprocessing; The pre-processed data is synchronized in time and aligned in space, and weights are dynamically assigned based on data credibility to achieve weighted fusion of multi-source data; The fused data is analyzed for hidden dangers through neural networks, and early warning information is issued based on the detected hidden danger types.

2. The substation hidden danger monitoring and early warning method according to claim 1, characterized in that: The obtaining of substation hidden danger monitoring data includes obtaining satellite data and ground data.

3. The substation hidden danger monitoring and early warning method according to claim 2, characterized in that: The data preprocessing includes preprocessing the acquired satellite data; Preprocess the acquired ground data.

4. The substation hidden danger monitoring and early warning method according to claim 3, characterized in that: The time synchronization includes aligning the satellite data and the ground data according to their timestamp information; The spatial alignment includes converting satellite data and ground data into a unified geographic coordinate system.

5. The substation hidden danger monitoring and early warning method according to claim 4, characterized in that: The implementation of weighted fusion of multi-source data includes adopting a weighted fusion algorithm, performing weight distribution calculation according to the credibility of satellite data and ground data, assigning different weights to different types of data, and performing weighted fusion based on the assigned weights to obtain multi-source data fusion.

6. The substation hidden danger monitoring and early warning method according to claim 5, characterized in that: The hidden danger analysis includes extracting spatiotemporal features from the fused data using a convolutional neural network and determining the hidden danger type based on preset logic; The types of hidden dangers include foreign object intrusion hazards, equipment overheating hazards and structural deformation hazards.

7. The substation hidden danger monitoring and early warning method according to claim 6, characterized in that: The issuing of warning information based on the detected hidden danger type includes generating corresponding warning information according to the hidden danger type and severity when a hidden danger is detected.

8. A substation hidden danger monitoring and early warning system, applying the method according to any one of claims 1 to 7, characterized in that: include: Data processing module, data fusion module, hidden danger warning module; The data processing module is used to obtain substation hidden danger monitoring data and perform data preprocessing; The data fusion module is used to synchronize the time and space of the pre-processed data, and dynamically assign weights based on the data credibility to achieve weighted fusion of multi-source data; The hidden danger warning module is used to perform hidden danger analysis on the fused data through a neural network and issue warning information based on the detected hidden danger type.

9. An electronic device comprising: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the substation hidden danger monitoring and early warning method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the steps of the substation hidden danger monitoring and early warning method according to any one of claims 1 to 7.

Images