Transformer substation ground subsidence early warning method, system and equipment integrating monitoring driving

Abstract

The invention discloses a transformer substation ground subsidence early warning method, system and equipment integrating monitoring driving, and relates to the related field of building subsidence monitoring, and the method comprises the steps: deploying a magnetostrictive sensor array to sense ground micro-deformation, collecting subsidence data in real time, building a self-organizing network through a low-power-consumption wireless WAPI (Wireless Local Area Network Application Program Interface), encrypting and transmitting the data to an edge computing node; performing multi-source data fusion processing on the nodes, and constructing a settlement trend prediction channel; a double-threshold early warning mechanism is set, a trend prediction value is obtained through prediction channel analysis, and multi-level early warning is triggered by comparing double thresholds based on the prediction value, so that intelligent early warning is realized. The technical problems that the early warning accuracy rate is low and potential safety hazards cannot be found in time due to the fact that a traditional transformer substation ground subsidence monitoring method is single, the multi-source data fusion capacity is weak and the real-time performance is poor are solved, and the purpose that the potential safety hazards cannot be found in time through multi-source monitoring data fusion and trend prediction is achieved. The technical effects of high-precision real-time sensing and graded intelligent early warning of land subsidence are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of building settlement monitoring, in particular to a substation ground settlement early warning method, system and equipment driven by fusion monitoring. BACKGROUND

[0002] As a key infrastructure of the power grid, the site of the substation is often in soft soil foundation, backfill land or complex environment affected by groundwater level fluctuation and surrounding construction disturbance. Long-term load, equipment vibration, earthquake and rainfall seepage and other factors are easy to cause uneven settlement of the foundation, and then may lead to uneven subsidence of the equipment foundation, change of the frame stress, stretching or even fracture of the busbar, causing serious operation accidents and threatening the power supply safety of the entire regional power grid. The traditional manual inspection and single sensor monitoring method has obvious limitations: manual measurement has long cycle and low efficiency, and cannot realize continuous real-time monitoring; and single monitoring method (such as using only a level instrument or a single inclination sensor) can only obtain isolated and one-sided information, lacks fusion perception ability of the whole process and multi-dimensional characteristics of the settlement, and is difficult to capture early and small deformation trends, with strong early warning lag. In addition, the environment of the substation is complex, and the electromagnetic interference is strong. The traditional wireless communication method faces challenges in transmission stability, safety and anti-interference ability, and it is difficult to ensure the high reliability and encryption of the monitoring data transmission to the processing center, affecting the integrity and timeliness of the monitoring data. SUMMARY

[0003] The present application provides a substation ground settlement early warning method, system and equipment driven by fusion monitoring, which solves the technical problems of low early warning accuracy and inability to timely discover safety hazards caused by single monitoring method of traditional substation ground settlement, weak multi-source data fusion ability and poor real-time performance, and achieves the technical effect of realizing high-precision real-time perception and hierarchical intelligent early warning of ground settlement through multi-source monitoring data fusion and trend prediction.

[0004] The present application provides a substation ground settlement early warning method driven by fusion monitoring, which is embedded with a multi-modal data acquisition device in the intelligent glasses, including: deploying a magnetostrictive sensor array to perceive the micro-deformation characteristics of the substation ground, collecting the substation ground settlement data in real time, constructing a self-organizing network through low-power wireless WAPI technology, and encrypting the substation ground settlement data for transmission to an edge computing node; performing fusion processing on the multi-source monitoring data at the edge computing node, constructing a settlement trend prediction channel; setting a double-threshold early warning mechanism, analyzing the substation ground through the settlement trend prediction channel, obtaining a settlement trend prediction value, comparing the settlement trend prediction value with the double-threshold early warning mechanism, triggering a multi-level early warning instruction, and intelligently warning the substation ground settlement.

[0005] The application also provides a fusion monitoring driven substation ground subsidence early warning system, the intelligent glasses are embedded with a multi-modal data acquisition device, comprising: a data acquisition module: deploying a magnetostrictive sensor array to perceive the micro-deformation characteristics of the substation ground, collecting substation ground subsidence data in real time, constructing a self-organizing network through low-power wireless WAPI technology, and encrypting and transmitting the substation ground subsidence data to an edge computing node; a data fusion module: fusing and processing multi-source monitoring data at the edge computing node to construct a subsidence trend prediction channel; an intelligent early warning module: setting a double-threshold early warning mechanism, analyzing the substation ground through the subsidence trend prediction channel to obtain a subsidence trend prediction value, comparing the subsidence trend prediction value with the double-threshold early warning mechanism, triggering a multi-level early warning instruction, and intelligently warning the substation ground subsidence.

[0006] The application also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the fusion monitoring driven substation ground subsidence early warning method.

[0007] The fusion monitoring driven substation ground subsidence early warning method, system and device provided by the application first deploy a magnetostrictive sensor array in the key area of the substation to perceive the micro-deformation of the ground and collect subsidence data in real time, then build a self-organizing network through low-power wireless WAPI technology, and transmit the data to an edge computing node after encryption. Subsequently, multi-source monitoring information including sensors, inclinometers and the like are fused and processed at the edge side to establish a subsidence trend prediction channel. Then, a double-threshold early warning mechanism is set to compare and analyze the prediction results, and different levels of early warning are triggered once the set range is exceeded, thereby achieving the technical effects of high-precision real-time perception and hierarchical intelligent early warning of ground subsidence. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings of the embodiments of the application will be briefly introduced below. In the present application, a flowchart is used to illustrate the operations performed by the system according to the embodiments of the application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. Meanwhile, other operations can be added to these processes, or a step or several steps can be removed from these processes.

[0009] Figure 1 The flowchart of the fusion monitoring driven substation ground subsidence early warning method provided by the embodiments of the application.

[0010] Figure 2 The structural diagram of the fusion monitoring driven substation ground subsidence early warning system provided by the embodiments of the application.

[0011] Figure 3 A structural schematic diagram of an electronic device is provided for an embodiment of the present application.

[0012] Legend: data acquisition module 11, data fusion module 12, intelligent early warning module 13, processor 21, memory 22, input device 23, output device 24. DETAILED DESCRIPTION

[0013] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clear, the following specific embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0014] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will combine the drawings to make a further detailed description of the present application. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those skilled in the art without making creative efforts belong to the scope of protection of the present application.

[0015] In the following description, "some embodiments" are described, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subset of all possible embodiments, and can be combined with each other without conflict. The term "first\second" is only to distinguish similar objects, and does not represent a specific order of the object. The terms "include" and "have" and any variations, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art of the technology to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0016] The embodiments of the present application provide a substation ground subsidence early warning method driven by fusion monitoring. The smart glasses are embedded with a multi-modal data acquisition device, such as Figure 1 As shown, the method comprises: Deploying a magnetostrictive sensor array to perceive the micro-deformation characteristics of the substation ground, collecting real-time substation ground subsidence data, constructing a self-organizing network through low-power wireless WAPI technology, and encrypting the substation ground subsidence data for transmission to an edge computing node.

[0017] Specifically, first, the magnetostrictive sensor array is laid out in a grid according to the preset key area on the ground of the substation, each magnetostrictive sensor array includes but is not limited to a magnetostrictive sensor, an inclinometer, and an environmental temperature and humidity sensor. This sensing device is in close contact with the ground, used to sense the small deformation characteristics of the ground in real time and convert them into measurable electrical signals. Subsequently, the magnetostrictive sensor array continuously collects the ground subsidence state according to the preset sampling frequency, thereby obtaining high-time-efficiency and high-precision subsidence data. Then, in order to ensure low power consumption and high reliability of data transmission, a low-power wireless WAPI communication module is configured at each magnetostrictive sensor array, and a self-organizing network between arrays is constructed through multiple links to realize dynamic interconnection and selection between arrays. Then, the self-organizing network encrypts the collected subsidence data to form a transmittable encrypted data packet to prevent monitoring information from being stolen or tampered with during transmission. Finally, the encrypted data packet is transmitted along the optimal path determined by the self-organizing network to the edge computing node, providing a safe and reliable data source for subsequent data fusion, trend prediction, and early warning analysis, thereby ensuring that the system can stably and continuously perform subsidence monitoring tasks in a complex power operation environment.

[0018] In one possible implementation, a magnetostrictive sensor array is deployed to perceive the micro-deformation characteristics of the ground of the substation, and the ground subsidence data of the substation is collected in real time. The method comprises: Performing subsidence influence analysis on the ground area of the substation, determining a plurality of key areas according to subsidence influence parameters; performing grid analysis according to the plurality of key areas, setting a node spacing, identifying a plurality of sensor nodes based on the plurality of key areas according to the node spacing; deploying a magnetostrictive sensor array according to the plurality of sensor nodes, continuously collecting the ground of the substation through the magnetostrictive sensor array according to a preset sampling frequency, and obtaining a ground micro-deformation signal; amplifying and filtering the ground micro-deformation signal to generate a noise reduction signal for verification and packaging, and obtaining the ground subsidence data of the substation.

[0019] Specifically, first, the geological exploration data of the entire substation site is traversed to determine the settlement influence parameters of each region, including but not limited to foundation soil characteristic parameters (such as soil layer distribution, compression modulus, void ratio, water content, etc.), load distribution parameters (such as the load size and distribution position of various electrical equipment and buildings), environmental parameters (such as underground water level, environmental vibration). Subsequently, the maximum-minimum normalization method is used to process these settlement influence parameters, so that these settlement influence parameters are in the same dimension, and then the weights set by the domain experts for each settlement influence parameter according to actual business requirements and historical experience are called to weight and fuse the normalized settlement influence parameters to analyze the settlement influence of each region. If the settlement sensitivity index of a certain region calculated by weighting is greater than or equal to the preset settlement sensitivity threshold, this region is defined as a key region. By comparing all regions in the same way, multiple key regions can be obtained, which are regions prone to uneven settlement or have high potential risks. Then, the key regions are centered for grid division, and the node spacing is set according to the site size and monitoring accuracy requirements. Then, the candidate positions are marked one by one at the grid intersection points according to the set node spacing, so as to determine the specific deployment positions of multiple magnetostrictive sensors, and these positions are defined as multiple sensor nodes. Then, during the sensor deployment stage, the magnetostrictive sensors are fixedly installed on the ground or shallowly buried in the foundation structure according to the determined sensor nodes, forming a magnetostrictive sensor array covering the key regions. This magnetostrictive sensor array will continuously monitor the deformation of the ground surface at a preset sampling frequency, real-time collect ground surface micro-displacement and settlement characteristic signals, and form ground surface micro-deformation signals to ensure the continuity and high timeliness of the monitoring. Finally, since the original ground surface micro-deformation signals output by the sensor often have weak amplitude and noise interference, the signals will be amplified and filtered by the front-end processing circuit after signal acquisition to enhance the effective signals and suppress environmental noise, so as to convert the processed signals into noise-reduced settlement effective data. In order to ensure the accuracy and integrity of the data transmission process, the settlement effective data will be attached with check codes or cyclic redundancy check information, and the data frame will be packaged according to the preset communication protocol to form standardized substation ground settlement data. This substation ground settlement data includes settlement monitoring value, sampling timestamp, node number and check information fields, etc., providing high-quality input for subsequent encrypted transmission, edge computing and trend prediction.

[0020] In one possible implementation, through the construction of a self-organizing network by low-power wireless WAPI technology, the substation ground settlement data is encrypted and transmitted to the edge computing node. The method comprises: The low-power wireless WAPI technology is configured to the plurality of sensor nodes to construct an initial network link; communication quality analysis is carried out based on the initial network link, and a target sensor node set is determined according to a link quality parameter; topology arrangement is carried out according to the target sensor node set to construct a multi-hop mesh network topology result; distributed network management is carried out through the multi-hop mesh network topology result to construct a self-organizing network; substation ground subsidence data is encrypted through the self-organizing network to generate an encrypted data packet; data transmission analysis is carried out according to the multi-hop mesh network topology structure to determine a transmission path, and the encrypted data packet is transmitted to the edge computing node according to the transmission path.

[0021] Specifically, first, a low-power wireless WAPI communication component is integrated in each sensor node, and module configuration is completed in the node initialization stage, so that each sensor node has the ability to access a wireless link. Then, through the handshake and link establishment process between nodes, an initial network link covering the entire network is constructed. Subsequently, in order to ensure the reliability and stability of data transmission, the system analyzes the communication quality of this initial network link. In this process, link quality parameters such as signal strength, delay, packet loss rate, and bandwidth utilization are collected in real time, and these link quality parameters are compared with the corresponding constraint intervals. From multiple sensor nodes, a set of target sensor nodes with good communication performance is selected. After obtaining the target sensor node set, the spatial positions of each node in the target sensor node set are extracted, the coordinate information of each target sensor node is obtained, the relative distance between nodes is calculated using the Euclidean distance, and a candidate connectivity table between nodes is established based on the extracted spatial positions and the link quality parameters of the nodes. Then, under the premise that all nodes meet the communication quality requirements, the nodes are hierarchically divided according to the strategy of minimum hop count priority and link quality weighting. That is, from the candidate connectivity table, target sensor nodes with excellent communication performance and close to the edge computing nodes are preferentially selected as relay core nodes, and peripheral nodes are sequentially accessed according to distance and communication quality, thereby constructing a multi-hop hierarchical structure extending outward from the core nodes. After the hierarchical relationship is established, a mesh interconnection method is used to allocate multiple redundant paths to each layer of nodes, so that there are at least two or more selectable paths between any two nodes, thereby improving the fault tolerance and robustness of the network, and finally building the multi-hop mesh network topology result. Then, based on this multi-hop mesh network topology result, the system carries out distributed network management, including node state monitoring, link dynamic updating, energy balance scheduling, and other operations, thereby realizing the stable operation of the self-organizing network built by the multi-hop mesh network topology result. After the self-organizing network can stably operate, the ground subsidence data collected by the sensor nodes will be encrypted through the self-organizing network. During encryption, the WAPI protocol will call the pre-configured symmetric encryption or asymmetric encryption technology on the node side to encapsulate the ground subsidence data, generating encrypted data packets that meet the security transmission requirements. After obtaining the encrypted data packets, the system collects real-time quality parameters of each link in the network based on the already constructed multi-hop mesh network topology, evaluates each candidate link using the collected quality parameters, calculates the comprehensive quality score, and retrieves multiple available paths that meet the transmission requirements according to the calculated score. Finally, the sustainability of the multiple available paths is evaluated, and an optimal transmission path is spliced. The encrypted data packets from the front-end node are forwarded hop by hop according to this transmission path, and are transmitted to the edge computing node through the relay nodes one by one, thereby realizing the stable transmission and real-time aggregation of the ground subsidence monitoring data of the substation.

[0022] In a possible implementation, the method for data transmission analysis according to the multi-hop mesh network topology and determination of the transmission path comprises the following steps. Based on the multi-hop mesh network topology, real-time quality analysis is performed on the links in the self-organizing network to obtain a plurality of link real-time quality parameters; link quality evaluation is performed based on the plurality of link real-time quality parameters to obtain a plurality of link quality scores; path retrieval is performed according to the plurality of link quality scores to determine a plurality of available paths; and sustainability evaluation is performed on the plurality of available paths to determine the transmission path.

[0023] Specifically, first, according to the multi-hop mesh network topology, the links in the self-organizing network are monitored in real time through periodic handshake packets and state feedback between nodes, and quality parameters including link signal strength, time delay, bandwidth utilization, packet loss rate, and node energy consumption state are collected to form a set of real-time quality parameters of a plurality of links. Subsequently, the acquired real-time quality parameters are normalized in the same manner as described above, and the real-time quality parameters of each link are weighted and calculated using a preset weight to obtain a plurality of link quality scores of the plurality of links. Then, according to a preset quality score threshold, path retrieval is performed on the plurality of links, and links with a link quality score greater than or equal to the quality score threshold are selected from the plurality of links and defined as available paths for further selection. Then, the available paths are traversed, and for each available path traversed, the node energy consumption state corresponding to the link is extracted from the plurality of link real-time quality parameters, and the node energy consumption state is compared with an energy consumption tolerance threshold. If a link has a high score but high energy consumption, it will be determined as not having long-term availability, because such a link will cause the node to run out of battery power too soon. Therefore, the system will exclude these links from the plurality of available paths. Finally, the remaining available paths are spliced with the goal of using the least number of nodes to form a path that is optimal in terms of comprehensive performance, stability, and energy consumption balance as the final transmission path, which is used to ensure efficient and secure transmission of encrypted data packets.

[0024] The edge computing node fuses the multi-source monitoring data to construct a settlement trend prediction channel.

[0025] Specifically, after receiving the encrypted data packet, the edge computing node first decrypts and integrity checks the encrypted data packet to ensure that the data has not been tampered with. Then, the decrypted data is indexed according to the timestamp and spatial coordinates and stored in the local time series database to ensure the efficiency of subsequent retrieval and processing. Subsequently, the edge computing node performs micro-deformation feature analysis on the data collected by the magnetostrictive sensor, performs inclination change analysis on the inclination instrument data, and compensates and corrects the analysis results using environmental temperature and humidity data, thereby eliminating errors caused by environmental disturbances. Finally, the edge computing node fuses the corrected multi-source feature parameters, and iteratively trains the deep neural network using the fused data to construct a settlement trend prediction channel. This settlement trend prediction channel enables the system to dynamically predict the future evolution trend of ground subsidence, providing reliable data support for subsequent threshold comparison and graded early warning.

[0026] In a possible implementation, the edge computing node fuses the multi-source monitoring data, constructs a settlement trend prediction channel, including: The edge computing node receives an encrypted data packet, decrypts the encrypted data packet to generate decrypted data, integrity verifies the decrypted data, and when the verification is passed, constructs a data index for the decrypted data, stores the decrypted data in a time series database according to the data index, iteratively trains the deep neural network based on the monitoring fusion feature vector, and constructs a settlement trend channel.

[0027] Specifically, first, after receiving the encrypted data packet, the edge computing node will decrypt it through the pre-installed WAPI protocol stack and extract the original decrypted data. After decryption, the data content will be verified for integrity, that is, a hash check (such as SHA-256) will be performed on the decrypted data to calculate a digest, which will be compared with the digest value in the data packet. If the comparison is consistent, it is determined that the data is complete and has not been tampered with. In addition, it will also check whether the data packet timestamp and system time deviation are within the allowed range to prevent replay attacks. When the verification is passed, the data index is constructed according to the data timestamp, node number and geographic coordinates, and the decrypted data is stored in the local time series database in order according to the index rules for subsequent retrieval and calling. Subsequently, the edge computing node traverses the time series database and performs multi-source retrieval according to different monitoring tasks to obtain the initial monitoring data set containing the outputs of the magnetostrictive sensor, the inclinometer and other sensors. In order to ensure the unity of the data, the initial monitoring data will be time and space registered, that is, in the time dimension, all kinds of data will be synchronized to the same time reference, and in the spatial dimension, they will be aligned to the unified coordinate reference system, thereby generating multi-source monitoring data with consistency. Then, the signal collected by the magnetostrictive sensor is extracted from the multi-source monitoring data, and the sensitivity coefficient of the magnetostrictive sensor is multiplied by the normalized amplitude value of the signal to obtain the displacement, and then the displacement data of the continuous time series is differentiated to obtain the settlement rate per unit time, and the displacement sequence is accumulated or integrated to obtain the cumulative settlement. By summarizing the calculated parameters, the micro-deformation characteristic parameters are formed. In addition, the signal collected by the inclinometer is also extracted from the multi-source monitoring data, and according to the calibration formula of the inclinometer, the signal is converted into the actual tilt angle. Through trend fitting and dynamic difference calculation of the actual tilt angle, the inclination change characteristic parameters such as horizontal offset angle, tilt rate and angle change gradient are extracted. Then, in order to reduce the interference of environmental factors, temperature and humidity data are extracted from the multi-source monitoring data, and temperature, humidity, micro-deformation characteristic parameters and inclination change characteristic parameters are input into the BP neural network compensation model. This BP neural network compensation model has been trained in advance using historical data through forward propagation, loss calculation, back propagation and parameter optimization, and can perform nonlinear compensation on the received characteristic parameters according to environmental factors to generate micro-deformation characteristic correction parameters and inclination change characteristic correction parameters corrected by temperature and humidity. Further, the edge computing node will splice and fuse the micro-deformation characteristic correction parameters and the inclination change characteristic correction parameters according to the pre-set vector template to construct a unified monitoring fusion feature vector, which can fully reflect the ground subsidence state.Finally, the system is based on the constructed monitoring fusion feature vector, the long short-term memory network is iteratively trained, and the training steps include forward propagation, loss calculation, back propagation and parameter optimization, etc. Thus, a settlement trend prediction channel capable of dynamic updating and self-correction is gradually formed, which provides data support for realizing early prediction and intelligent warning of settlement, and improves the accuracy of substation settlement monitoring.

[0028] In a possible implementation, the time series database is traversed for multi-source retrieval, initial monitoring data is generated, the initial monitoring data is spatio-temporally registered, and multi-source monitoring data is generated. The method comprises: Based on the decrypted data, a monitoring task is formulated. The multi-source monitoring data storage area in the time series database is traversed, multi-condition combination retrieval is performed according to the monitoring task, and an initial monitoring data set is obtained. The initial monitoring data set is uniformly processed according to a time reference, and time-unified monitoring data is obtained. The initial monitoring data set is uniformly processed according to a spatial coordinate, and spatially-unified monitoring data is obtained. The time-unified monitoring data and the spatially-unified monitoring data are subjected to data quality consistency checking, and the multi-source monitoring data is constructed.

[0029] Specifically, first, the decrypted data is content-analyzed, and monitoring demand analysis is performed in combination with the current operation condition of the substation and the monitoring target. For example, when the monitoring focus is on uneven foundation settlement, a monitoring task related to settlement is automatically formulated. Subsequently, the multi-source monitoring data storage area in the time series database is traversed, and retrieval conditions such as time range, node position, sensor type and data accuracy are set according to the formulated monitoring task, and multi-condition combination retrieval is performed according to the retrieval conditions, and an initial monitoring data set that meets the requirements is screened. Then, a unified time reference is set, such as the earliest timestamp or a certain specific time, all data are adjusted according to the reference, to ensure that each data point can be at a unified time point, eliminate the differences in sampling period and timestamp of different sensors, and obtain time-unified monitoring data. In addition, the seven-parameter method is used to convert the coordinate systems in the initial monitoring data set to the unified CGCS2000 national geodetic coordinate system, to ensure that the coordinates of each monitoring point are represented in a unified reference system, and to obtain spatially-unified monitoring data. Finally, the system performs quality consistency checking on the time-unified data and the spatially-unified data, including detecting missing values, abnormal values, etc., and interpolating, correcting or removing the data with deviations, thereby constructing multi-source monitoring data with high reliability and consistency, providing a solid data foundation for subsequent feature extraction, fusion calculation and trend prediction.

[0030] The double-threshold early warning mechanism is set, the substation ground is analyzed through the settlement trend prediction channel to obtain a settlement trend prediction value, the double-threshold early warning mechanism is combined based on the settlement trend prediction value to perform comparison, a multi-level early warning instruction is triggered, and intelligent early warning is performed on the substation ground settlement.

[0031] Specifically, after the edge computing node completes the fusion processing on the multi-source monitoring data and constructs the settlement trend prediction channel, the pre-configured double-threshold early warning mechanism is activated. The double-threshold early warning mechanism includes an absolute settlement amount threshold and a settlement change rate threshold. The absolute settlement amount threshold is used to measure whether the cumulative ground settlement exceeds a safety limit. The settlement change rate threshold is used to determine whether the development trend of the ground settlement is abnormally accelerated. After the system obtains the settlement trend prediction value corresponding to the current substation ground condition from the settlement trend prediction channel, the double-threshold early warning mechanism is used to analyze and compare the settlement trend prediction value, determine the early warning level to be triggered, and generate a multi-level early warning instruction. Finally, according to the multi-level early warning instruction, the local alarm device or the remote monitoring platform is used to push the operation and maintenance personnel, so that reinforcement, maintenance or shutdown and other disposal measures can be taken in time. In summary, through the comparison mechanism combined with the double thresholds, false positives and false negatives caused by single threshold judgment can be avoided, and dynamic and hierarchical monitoring of the settlement trend can be realized, thereby ensuring the early identification and effective control of the substation ground settlement risk.

[0032] In one possible implementation, a double-threshold early warning mechanism is set, the substation ground is analyzed through the settlement trend prediction channel to obtain a settlement trend prediction value, the double-threshold early warning mechanism is combined based on the settlement trend prediction value to perform comparison, a multi-level early warning instruction is triggered, and intelligent early warning is performed on the substation ground settlement, and the method comprises: A double-threshold early warning mechanism is set, and the double-threshold early warning mechanism includes an absolute settlement amount threshold and a settlement change rate threshold. The substation ground settlement data is analyzed in real time through the settlement trend prediction channel to obtain a settlement trend prediction value. The settlement trend prediction value includes a settlement amount prediction value and a settlement change rate prediction value. The settlement amount prediction value is compared with the absolute settlement amount threshold to obtain a first comparison result. The settlement change rate prediction value is compared with the settlement change rate threshold to obtain a second comparison result. A multi-level early warning instruction is triggered according to the first comparison result and the second comparison result, the multi-level early warning instruction is executed, and intelligent early warning and disposal of the substation ground settlement are realized.

[0033] Specifically, first, according to the design specification of the substation, the foundation bearing requirement and the long-term accumulated historical monitoring data, the absolute settlement threshold value is set by the field expert to indicate whether the ground settlement has exceeded the design range, and then the trend analysis is performed on the settlement data of the substation in previous years to calculate the rate of change of the settlement, and the settlement change rate threshold value is set by the field expert to identify whether the settlement rate is abnormally accelerated, and by encapsulating the absolute settlement threshold value and the settlement change rate threshold value, a double-threshold early warning mechanism is constructed. In the running process, the edge computing node analyzes the monitoring data collected and fused in real time through the settlement trend prediction channel to obtain the settlement trend prediction value in the future period, compares the settlement prediction value in the settlement trend prediction value with the absolute settlement threshold value to obtain a first comparison result, and compares the settlement change rate prediction value in the settlement trend prediction value with the settlement rate threshold value to obtain a second comparison result. Finally, the system triggers a multi-level warning instruction according to different combination relationships of the first comparison result and the second comparison result, and executes the corresponding warning strategy according to the warning level defined in the multi-level warning instruction, so as to effectively avoid the false alarm and the missed alarm caused by single threshold judgment, realize the dynamic, intelligent and hierarchical early warning and safe disposal of the ground settlement of the substation.

[0034] In a possible implementation, according to the first comparison result and the second comparison result, a multi-level warning instruction is triggered, the multi-level warning instruction is executed, and intelligent early warning and disposal of the ground settlement of the substation are realized. The method comprises the following steps: Based on the first comparison result and the second comparison result, weight analysis is performed on the ground settlement of the substation to determine a plurality of weight coefficients; the first comparison result and the second comparison result are analyzed according to the plurality of weight coefficients to trigger early warning, and a warning trigger decision matrix is constructed; based on the warning trigger decision matrix, the first comparison result and the second comparison result are determined to trigger early warning, and a trigger determination result is generated; and the multi-level warning instruction is activated according to the trigger determination result.

[0035] Specifically, after obtaining the first comparison result and the second comparison result, in order to avoid misjudgment caused by single index deviation, weight coefficients are assigned to the settlement amount and the settlement rate according to the substation operating environment, the monitoring target and the statistical law of historical overrun events, for example, in the scene of soft soil foundation or long-term load, the absolute settlement amount weight is higher, and in the short-term risk monitoring scene, the settlement rate weight is relatively increased. Then, according to the set weight coefficients, the first comparison result and the second comparison result are weighted and processed to construct a warning trigger decision matrix, which takes the overrun degree of the comparison result as the input, and maps different weight combinations to the corresponding warning level interval. Then, based on the warning trigger decision matrix, the comparison result is judged, if the weighted result is in the normal interval, no warning judgment is generated; if the weighted result is in the low risk interval, a first level warning judgment is generated; if the weighted result is in the high risk interval, a second or third level warning judgment is generated. Finally, according to the trigger judgment result, the corresponding multi-level warning instruction is automatically activated, the low level warning is usually fed back in the form of local prompt or monitoring report; the middle level warning is pushed to the monitoring center to prompt the operation and maintenance personnel to follow up, and the high level warning triggers the sound and light alarm, remote alarm and even linkage disposal mechanism, to ensure the safe operation of the substation.

[0036] In the foregoing, with reference to Figure 1 The fusion monitoring driven substation ground settlement warning method according to the embodiment of the application is described in detail. Next, the fusion monitoring driven substation ground settlement warning system according to the embodiment of the application will be described with reference to Figure 2 The fusion monitoring driven substation ground settlement warning system according to the embodiment of the application is described in detail. Next, the fusion monitoring driven substation ground settlement warning system according to the embodiment of the application will be described with reference to

[0037] The fusion monitoring driven substation ground settlement warning system according to the embodiment of the application is used to solve the technical problems of low warning accuracy and inability to discover safety hazards in time caused by single monitoring means of traditional substations, weak multi-source data fusion capability and poor real-time performance, and achieves the technical effect of realizing high-precision real-time perception and hierarchical intelligent warning of ground settlement through multi-source monitoring data fusion and trend prediction. The fusion monitoring driven substation ground settlement warning system comprises a data acquisition module 11, a data fusion module 12 and an intelligent warning module 13.

[0038] The data acquisition module 11: deploy a magnetostrictive sensor array to perceive the micro-deformation characteristics of the substation ground, collect substation ground settlement data in real time, build a self-organizing network through low-power wireless WAPI technology, and encrypt the substation ground settlement data and transmit it to the edge computing node; the data fusion module 12: fuse the multi-source monitoring data at the edge computing node, and build a settlement trend prediction channel; the intelligent early warning module 13: set a double-threshold early warning mechanism, analyze the substation ground through the settlement trend prediction channel, obtain a settlement trend prediction value, compare the settlement trend prediction value based on the double-threshold early warning mechanism, trigger a multi-level early warning instruction, and intelligently warn the substation ground settlement.

[0039] Next, the specific configuration of the data acquisition module 11 will be described in detail. As described above, the magnetostrictive sensor array is deployed to perceive the micro-deformation characteristics of the substation ground, and the substation ground settlement data is collected in real time. The data acquisition module 11 can further include: traversing the substation ground area to analyze the settlement influence, determining a plurality of key areas according to the settlement influence parameters; performing grid analysis according to the plurality of key areas, setting node spacing, identifying a plurality of sensor nodes based on the plurality of key areas according to the node spacing; deploying a magnetostrictive sensor array according to the plurality of sensor nodes, continuously collecting the substation ground through the magnetostrictive sensor array according to a preset sampling frequency, and obtaining a ground micro-deformation signal; amplifying and filtering the ground micro-deformation signal to generate a noise reduction signal for verification and packaging, and obtaining the substation ground settlement data.

[0040] Among them, through the low-power wireless WAPI technology to build a self-organizing network, the substation ground settlement data is encrypted and transmitted to the edge computing node, and the data acquisition module 11 can further include: configuring low-power wireless WAPI technology to the plurality of sensor nodes to build an initial network link; based on the initial network link, analyze the communication quality, determine a target sensor node set according to the link quality parameters; according to the target sensor node set, perform topology arrangement to build a multi-hop mesh network topology result; through the multi-hop mesh network topology result, perform distributed network management to build a self-organizing network; through the self-organizing network, the substation ground settlement data is encrypted to generate an encrypted data packet; according to the multi-hop mesh network topology structure, perform data transmission analysis to determine a transmission path, and transmit the encrypted data packet to the edge computing node according to the transmission path.

[0041] The data transmission analysis according to the multi-hop mesh network topology determines the transmission path, and the data collection module 11 can further include: performing real-time quality analysis on the links in the self-organizing network based on the multi-hop mesh network topology to obtain a plurality of link real-time quality parameters; performing link quality evaluation based on the plurality of link real-time quality parameters to obtain a plurality of link quality scores; performing path retrieval according to the plurality of link quality scores to determine a plurality of available paths; and performing sustainability evaluation on the plurality of available paths to determine the transmission path.

[0042] In the following, the specific configuration of the data fusion module 12 will be described in detail. As described above, the multi-source monitoring data is fused and processed at the edge computing node to construct a settlement trend prediction channel, and the data fusion module 12 can further include: receiving an encrypted data packet at the edge computing node, decrypting the encrypted data packet to generate decrypted data; performing integrity verification on the decrypted data, and when the verification is passed, constructing a data index on the decrypted data, and storing the decrypted data to a time series database according to the data index; performing multi-source retrieval on the time series database to generate initial monitoring data, and performing time-space registration on the initial monitoring data to generate multi-source monitoring data; performing magnetostriction analysis on the multi-source monitoring data to obtain micro-deformation characteristic parameters, and performing inclination change analysis on the multi-source monitoring data to obtain inclination change characteristic parameters; calling environmental temperature and humidity data to compensate and correct the micro-deformation characteristic parameters and the inclination change characteristic parameters to generate micro-deformation characteristic correction parameters and inclination change characteristic correction parameters; and fusing and calculating the micro-deformation characteristic correction parameters and the inclination change characteristic correction parameters through the edge computing node to construct a monitoring fusion feature vector; and performing settlement iterative training based on the monitoring fusion feature vector to construct a settlement trend channel.

[0043] The data fusion module 12 can further include: performing monitoring demand analysis based on the decrypted data to formulate a monitoring task; performing multi-condition combination retrieval according to the monitoring task to obtain an initial monitoring data set by traversing a multi-source monitoring data storage area in the time series database; uniformly processing the initial monitoring data set according to a time reference to obtain time-unified monitoring data; uniformly processing the initial monitoring data set according to a spatial coordinate to obtain space-unified monitoring data; and performing data quality consistency verification on the time-unified monitoring data and the space-unified monitoring data to construct the multi-source monitoring data.

[0044] In the following, the specific configuration of the intelligent early warning module 13 will be described in detail. As described above, the double threshold early warning mechanism is set, the substation ground is analyzed through the settlement trend prediction channel to obtain a settlement trend prediction value, the double threshold early warning mechanism is combined with the settlement trend prediction value for comparison, a multi-level early warning instruction is triggered, and the substation ground settlement is intelligently warned. The intelligent early warning module 13 can further include: setting a double threshold early warning mechanism, the double threshold early warning mechanism including an absolute settlement amount threshold and a settlement change rate threshold; analyzing the substation ground settlement data in real time through the settlement trend prediction channel to obtain a settlement trend prediction value, the settlement trend prediction value including a settlement amount prediction value and a settlement change rate prediction value; comparing the settlement amount prediction value with the absolute settlement amount threshold to obtain a first comparison result, and comparing the settlement change rate prediction value with the settlement change rate threshold to obtain a second comparison result; triggering a multi-level early warning instruction according to the first comparison result and the second comparison result, executing the multi-level early warning instruction, and realizing intelligent early warning and disposal of the substation ground settlement.

[0045] According to the first comparison result and the second comparison result, a multi-level early warning instruction is triggered, the multi-level early warning instruction is executed, and intelligent early warning and disposal of the substation ground settlement are realized. The intelligent early warning module 13 can further include: performing weight analysis on the substation ground settlement based on the first comparison result and the second comparison result to determine a plurality of weight coefficients; performing early warning trigger analysis on the first comparison result and the second comparison result according to the plurality of weight coefficients to construct an early warning trigger decision matrix; performing early warning trigger determination on the first comparison result and the second comparison result based on the early warning trigger decision matrix to generate a trigger determination result; and activating the multi-level early warning instruction according to the trigger determination result.

[0046] The fusion monitoring driven substation ground settlement early warning system provided in the embodiments of the present application can execute the fusion monitoring driven substation ground settlement early warning method provided in any of the embodiments of the present application, has the function modules and beneficial effects corresponding to the execution method.

[0047] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, and each unit and module included is only divided according to the functional logic, but is not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy mutual differentiation, and do not limit the protection scope of the present application.

[0048] Based on the foregoing embodiments, the embodiments of the present application further provide an electronic device. Figure 3is a structural schematic diagram of an electronic device provided by an embodiment of the present application, and shows a block diagram of an exemplary electronic device suitable for implementing the embodiment of the present application. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the function and use range of the embodiment of the present application. The electronic device is in the form of a general computing device, and its components can include but are not limited to a processor 21, a memory 22, an input device 23, and an output device 24. The processor 21 can be one or more; the memory 22 can include a computer readable medium and at least one program product, which has a set of (at least one) program modules configured to perform the functions of the embodiments of the present application.

[0049] The memory 22 shown in the embodiment of the present application can adopt any combination of one or more computer readable media; the computer readable storage medium can be but is not limited to an infrared ray, a semiconductor system, a device or a component, or any combination of the above, for storing software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the substation ground subsidence early warning method driven by fusion monitoring in the embodiment of the present application. The processor 21 performs various function applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 22, that is, implements the above-mentioned substation ground subsidence early warning method driven by fusion monitoring.

[0050] The above specific embodiments do not constitute a limitation to the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present application should be included in the protection scope of the present application. In some cases, the actions or steps described in the present application can be executed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

Claims

1. A substation ground settlement early warning method driven by integrated monitoring, characterized in that, The method includes: A magnetostrictive sensor array is deployed to sense the micro-deformation characteristics of the substation ground, collect substation ground settlement data in real time, and construct a self-organizing network through low-power wireless WAPI technology to encrypt and transmit the substation ground settlement data to the edge computing node. The edge computing node fuses multi-source monitoring data to construct a settlement trend prediction channel; A dual-threshold early warning mechanism is set up. The substation ground is analyzed through the settlement trend prediction channel to obtain the settlement trend prediction value. The settlement trend prediction value is compared with the dual-threshold early warning mechanism to trigger multi-level early warning instructions and provide intelligent early warning for substation ground settlement.

2. The substation ground settlement early warning method driven by integrated monitoring as described in claim 1, characterized in that, Deploying a magnetostrictive sensor array to sense the micro-deformation characteristics of the substation ground and collect real-time substation ground settlement data includes the following methods: A settlement impact analysis was conducted across the entire ground area of ​​the substation, and several key areas were identified based on the settlement impact parameters. Based on the multiple key regions, a gridded analysis is performed, the node spacing is set, and multiple sensor nodes are identified according to the node spacing based on the multiple key regions. A magnetostrictive sensor array is deployed according to the multiple sensor nodes. The magnetostrictive sensor array continuously collects data on the substation ground at a preset sampling frequency to obtain surface micro-deformation signals. The surface micro-deformation signal is amplified and filtered to generate a noise-reduced signal, which is then verified and packaged to obtain the substation ground settlement data.

3. The substation ground settlement early warning method driven by integrated monitoring as described in claim 2, characterized in that, A self-organizing network is constructed using low-power wireless WAPI technology to encrypt and transmit substation ground settlement data to edge computing nodes. The method includes: Low-power wireless WAPI technology is configured on the multiple sensor nodes to establish an initial network link; Based on the initial network link, a communication quality analysis is performed, and the target sensor node set is determined according to the link quality parameters. Based on the target sensor node set, a topology arrangement is performed to construct a multi-hop mesh network topology. Distributed network management is performed using the multi-hop mesh network topology results to construct a self-organizing network; The self-organizing network is used to encrypt the substation ground settlement data to generate encrypted data packets. Data transmission analysis is performed according to the multi-hop mesh network topology to determine the transmission path, and the encrypted data packet is transmitted to the edge computing node according to the transmission path.

4. The substation ground settlement early warning method driven by integrated monitoring as described in claim 3, characterized in that, Data transmission analysis is performed based on the multi-hop mesh network topology to determine the transmission path. The method includes: Based on the multi-hop mesh network topology, real-time quality analysis is performed on the links in the self-organizing network to obtain multiple real-time link quality parameters. Based on the multiple real-time link quality parameters, link quality is evaluated to obtain multiple link quality scores; Path retrieval is performed based on the multiple link quality scores to determine multiple available paths; The sustainability of the multiple available paths is assessed to determine the transmission path.

5. The substation ground settlement early warning method driven by integrated monitoring as described in claim 1, characterized in that, The method involves fusing multi-source monitoring data at the edge computing node to construct a settlement trend prediction channel, including: The edge computing node receives encrypted data packets, decrypts the encrypted data packets, and generates decrypted data. The integrity of the decrypted data is verified. When the verification is successful, a data index is constructed on the decrypted data, and the decrypted data is stored in a time-series database according to the data index. The time-series database is traversed to perform multi-source retrieval, generating initial monitoring data. The initial monitoring data is then spatiotemporally registered to generate multi-source monitoring data. Magnetostrictive analysis was performed on the multi-source monitoring data to obtain micro-deformation characteristic parameters, and tilt angle change analysis was performed on the multi-source monitoring data to obtain tilt angle change characteristic parameters. The ambient temperature and humidity data are retrieved to compensate and correct the micro-deformation characteristic parameters and the tilt angle change characteristic parameters, thereby generating micro-deformation characteristic correction parameters and tilt angle change characteristic correction parameters. The edge computing node is used to fuse the micro-deformation feature correction parameters and the tilt angle change feature correction parameters to construct a monitoring fusion feature vector. Settlement iterative training is performed based on the aforementioned monitoring fusion feature vector to construct a settlement trend channel.

6. The substation ground settlement early warning method driven by integrated monitoring as described in claim 5, characterized in that, The method involves traversing the time-series database to perform multi-source retrieval, generating initial monitoring data, and then performing spatiotemporal registration on the initial monitoring data to generate multi-source monitoring data. Based on the decrypted data, a monitoring needs analysis is conducted, and monitoring tasks are formulated. Traverse the multi-source monitoring data storage area in the time series database, perform multi-condition combination retrieval according to the monitoring task, and obtain the initial monitoring dataset; The initial monitoring dataset is processed uniformly according to a time base to obtain time-uniform monitoring data; The initial monitoring dataset is processed uniformly according to spatial coordinates to obtain spatially unified monitoring data; The time-unified monitoring data and the spatial-unified monitoring data are subjected to data quality consistency verification to construct the multi-source monitoring data.

7. The substation ground settlement early warning method driven by integrated monitoring as described in claim 1, characterized in that, A dual-threshold early warning mechanism is established. The substation ground subsidence is analyzed through the settlement trend prediction channel to obtain a predicted settlement trend value. Based on this predicted value, a comparison is made with the dual-threshold early warning mechanism to trigger multi-level early warning commands, providing intelligent early warning for substation ground subsidence. The method includes: A dual-threshold early warning mechanism is set up, which includes an absolute settlement threshold and a settlement change rate threshold. The substation ground settlement data is analyzed in real time through the settlement trend prediction channel to obtain the settlement trend prediction value, which includes the settlement amount prediction value and the settlement change rate prediction value. The predicted settlement value is compared with the absolute settlement threshold to obtain a first comparison result, and the predicted settlement change rate is compared with the settlement change rate threshold to obtain a second comparison result. Based on the first comparison result and the second comparison result, a multi-level early warning command is triggered and executed to realize intelligent early warning and handling of substation ground subsidence.

8. The substation ground settlement early warning method driven by integrated monitoring as described in claim 7, characterized in that, Based on the first comparison result and the second comparison result, multi-level early warning commands are triggered and executed to achieve intelligent early warning and handling of substation ground subsidence. The method includes: Based on the first comparison result and the second comparison result, a weight analysis of the substation ground settlement is performed to determine multiple weight coefficients; The first comparison result and the second comparison result are analyzed for early warning triggering according to the multiple weighting coefficients to construct an early warning triggering decision matrix; Based on the aforementioned warning trigger decision matrix, a warning trigger determination is made between the first comparison result and the second comparison result to generate a trigger determination result; The multi-level early warning command is activated based on the trigger determination result.

9. A substation ground settlement early warning system driven by integrated monitoring, characterized in that, The system is used to implement the substation ground settlement early warning method driven by fusion monitoring as described in any one of claims 1-8, and the system comprises: Data acquisition module: Deploys a magnetostrictive sensor array to sense the micro-deformation characteristics of the substation ground, collects substation ground settlement data in real time, and constructs a self-organizing network through low-power wireless WAPI technology to encrypt and transmit the substation ground settlement data to the edge computing node; Data fusion module: performs fusion processing on multi-source monitoring data at the edge computing node to construct a settlement trend prediction channel; Intelligent early warning module: A dual-threshold early warning mechanism is set up. The substation ground is analyzed through the settlement trend prediction channel to obtain the settlement trend prediction value. Based on the settlement trend prediction value and the dual-threshold early warning mechanism, a multi-level early warning command is triggered to provide intelligent early warning of substation ground settlement.

10. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the substation ground settlement early warning method driven by fusion monitoring as described in any one of claims 1 to 8.

Images