Power distribution network tower abnormity monitoring method and system

Abstract

The invention discloses a power distribution network tower abnormity monitoring method and system, and relates to the technical field of tower monitoring, and the method comprises the steps: arranging a multi-mode sensor on a power distribution network tower, and collecting the state data of the tower; constructing a power distribution network tower digital twinborn model, inputting tower state data, and predicting the future state of the tower; performing regular inspection on the tower by using an unmanned aerial vehicle, and collecting dynamic three-dimensional data of the tower; and the tower state data, the tower future state prediction result and the tower dynamic three-dimensional data are fused, and tower abnormity is monitored and early warning is performed. The invention also discloses a method. Through the application of the multi-source data fusion technology, the accuracy and reliability of anomaly detection are effectively improved, professional guidance is provided for maintenance personnel based on a grading early warning mechanism of a comprehensive anomaly index, the fault processing efficiency is improved, the accuracy and timeliness of tower anomaly monitoring are remarkably improved, and the safety and reliability of the tower are improved. And predictive maintenance and intelligent management are also realized, and the safety and reliability of the power distribution network are greatly enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of pole tower monitoring, and in particular to a method and system for monitoring abnormality of distribution network pole towers. Background Art

[0002] As a key infrastructure in the power system, the safe and stable operation of distribution network towers is crucial to ensuring power supply. With the continuous growth of electricity demand and the rapid development of smart grids, the monitoring technology of distribution network towers is also evolving. Traditional distribution network tower monitoring methods mainly rely on manual inspections and regular maintenance, which has many problems such as low efficiency, poor real-time performance, and insufficient accuracy. In recent years, with the rise of emerging technologies such as the Internet of Things, big data, and artificial intelligence, distribution network tower monitoring technology has gradually developed in the direction of intelligence, automation, and refinement. The application of new technologies such as multimodal sensing technology, digital twin technology, and drone inspection technology has provided new ideas and methods for abnormal monitoring of distribution network towers.

[0003] However, the existing distribution network tower abnormality monitoring technology still has some shortcomings. First, a single monitoring method is difficult to fully reflect the actual status of the tower, resulting in low reliability and accuracy of the monitoring results. Secondly, traditional data analysis methods are difficult to handle massive, multi-dimensional monitoring data and cannot fully mine the valuable information contained in the data. Furthermore, the existing monitoring systems often lack the ability to predict the future status of the tower, making it difficult to achieve active prevention and timely intervention. In addition, the existing technology needs to be improved in the integration of multi-source heterogeneous data, making it difficult to achieve a comprehensive and accurate assessment of the tower status. Summary of the invention

[0004] In view of the problems existing in the existing distribution network tower abnormality monitoring technology, such as single monitoring mode, insufficient prediction capability, and inability to integrate data, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: through the comprehensive application of multimodal sensing, digital twins, drone inspection, data fusion and other technologies, comprehensive, accurate and real-time monitoring and early warning of distribution network towers are achieved.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a distribution network tower abnormality monitoring method, comprising: arranging multimodal sensors on the distribution network towers to collect tower status data; constructing a digital twin model of the distribution network towers, and predicting the future status of the towers; using drones to conduct regular inspections of the towers to collect dynamic three-dimensional data of the towers; integrating the tower status data, the tower future status prediction results and the tower dynamic three-dimensional data to monitor tower abnormalities and issue early warnings.

[0007] As a preferred solution of the distribution network tower abnormality monitoring method of the present invention, the multimodal sensor includes a vibration sensor, a temperature sensor, an acoustic wave sensor and a current sensor; the vibration sensor is used to collect vibration frequency, amplitude and acceleration data of the tower; the temperature sensor is used to collect temperature data of key parts of the tower; the key parts of the tower include wire connection points and transformers; the acoustic signal around the tower is collected by the acoustic wave sensor; the acoustic signal includes discharge sound, mechanical friction sound and wind noise; the current data of the power line on the tower is collected by the current sensor; the current data includes current intensity, phase and harmonic components; the tower status data includes vibration frequency, amplitude and acceleration data, temperature data, acoustic signals, and current data.

[0008] As a preferred solution of the method for monitoring abnormalities of distribution network towers of the present invention, the step of predicting the future state of the towers includes: constructing a geometric model based on the shape, size and material properties of the towers; calculating the stress state of the towers under different load conditions;

[0009]

[0010] in, is a divergence operator, which represents the divergence of the stress tensor σ, which represents the influence of the internal force on the material at position p and time t, σ(p, t) is a stress tensor, which represents the state of the force at position p and time t, f(p, t) represents the body force density function, which represents the density of the external force applied at position p and time t, and ρ represents the material density of the tower material; u(p, t) represents the displacement function at position p and time t; the node coordinates in the geometric model correspond to the stress tensor of the stress state, and a digital twin model of the distribution network tower is constructed; the historical data of the tower are collected, and the historical data and the tower state data are processed by a machine learning algorithm, and a mapping relationship between the tower state data and the parameters of the digital twin model of the distribution network tower is established, so that the digital twin model of the distribution network tower reflects the current state of the tower; based on the current state of the tower and the historical data of the tower, a time series analysis algorithm is used to predict the future state of the tower, and a prediction result of the future state of the tower is obtained.

[0011] As a preferred solution of the distribution network pole tower abnormality monitoring method of the present invention, the acquisition of dynamic three-dimensional data of the pole tower includes finding the optimal inspection path based on the pole tower geographical location information using an ant colony algorithm; dividing the inspection sub-areas according to the drone's endurance and weather conditions; formulating an inspection execution plan based on the optimal inspection path and combining the inspection sub-areas and the priorities of the pole towers; according to the inspection execution plan, the drone uses a camera to acquire all-round images of the pole tower for identifying abnormalities, and uses a laser radar to perform a three-dimensional scan of the pole tower to obtain three-dimensional data of the pole tower.

[0012] As a preferred solution of the distribution network pole tower abnormality monitoring method of the present invention, the fusion of pole tower state data, pole tower future state prediction results and pole tower dynamic three-dimensional data includes: standardizing and merging the pole tower state data through a weighted fusion algorithm to obtain initial data X(t); combining the pole tower state data with the pole tower future state prediction results, considering historical prediction errors, to obtain optimized data X′(t); combining the optimized data X′(t) with the pole tower three-dimensional data to generate the pole tower final fused data X″(t).

[0013] As a preferred solution of the distribution network tower abnormality monitoring method of the present invention, the calculation formula of the initial data X(t) is as follows:

[0014]

[0015] Among them, X(t) represents the initial data, α i represents the corresponding data weight, β i represents the adjustment parameter, represents the standardized data of the i-th sensor, express The calculation formula of the optimized data X′(t) is as follows:

[0016]

[0017] Among them, X′(t) represents the optimized data, X(t) represents the initial data, S(t+Δt) is the prediction result of the future state of the tower, γ represents the weight coefficient of the prediction result of the future state of the tower, ε(t) represents the historical prediction error, and λ represents the adjustment coefficient; the calculation formula of the final fusion data X″(t) of the tower is as follows:

[0018]

[0019] Among them, X″(t) represents the final fusion data of the tower, X′(t) represents the optimized data, F(t) represents the feature vector in the three-dimensional data of the tower, δ represents the weight coefficient of the feature vector of the three-dimensional data of the tower, η represents the adjustment coefficient, and ||F(t)|| represents the norm of F(t).

[0020] As a preferred solution of the distribution network tower abnormality monitoring method of the present invention, the tower abnormality monitoring and early warning includes calculating the tower comprehensive abnormality index based on the tower final fusion data X″(t), and the calculation formula of the tower comprehensive abnormality index is as follows:

[0021]

[0022] Among them, E(t) represents the comprehensive anomaly index, σ represents the activation function, X″(t) represents the final fusion data of the tower, W represents the weight matrix, T represents the transposed matrix, ρ represents the adjustment coefficient, and b represents the bias vector; when E(t) < the first threshold, the state of the distribution network tower is normal, the routine monitoring procedure is continuously executed, the frequency of tower state data collection is maintained, and the digital twin model of the distribution network tower is regularly updated; when the first threshold ≤ E(t) < the second threshold, the monitoring system determines that there is a primary anomaly in the tower state and initiates the primary warning. At this time, the tower is diagnosed abnormally. , identify potential faults, increase the sampling frequency of some multimodal sensors, and generate a checklist; when the second threshold ≤ E(t) < the third threshold, the monitoring system determines that there is an intermediate abnormality in the tower state and activates the intermediate warning. At this time, the sampling frequency of all multimodal sensors is increased, the abnormal source is located and the impact range of the abnormality is evaluated, and an abnormality report and maintenance plan are generated; when E(t) ≥ the third threshold, the monitoring system determines that the tower state is seriously abnormal, activates the advanced warning, and the monitoring system immediately issues an emergency alarm, monitors the abnormal tower in real time, and generates an emergency disposal plan and risk mitigation strategy.

[0023] Another object of the present invention is to provide a distribution network tower abnormality monitoring system, comprising: a data acquisition module, used to arrange multimodal sensors on the distribution network towers to collect tower status data; a first model construction module, used to construct a digital twin model of the distribution network towers and predict the future state of the towers; a second model construction module, used to use drones to conduct regular inspections of the towers and collect dynamic three-dimensional data of the towers; an early warning module, used to integrate the tower status data, the tower future state prediction results and the tower dynamic three-dimensional data, monitor the tower abnormalities and issue early warnings.

[0024] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the above distribution network tower abnormality monitoring method are implemented.

[0025] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the above-mentioned distribution network tower abnormality monitoring method are implemented.

[0026] Beneficial effects of the present invention: through the arrangement of multimodal sensors, all-round and multi-dimensional monitoring of the tower status is achieved, the comprehensiveness and accuracy of data collection are improved, and the digital twin model of the distribution network tower constructed in combination with the digital twin technology can accurately predict the future status of the tower, providing a scientific basis for preventive maintenance. Intelligent drone inspections not only improve the inspection efficiency, but also provide accurate spatial information for abnormal location. The application of multi-source data fusion technology effectively improves the accuracy and reliability of abnormal detection. The hierarchical early warning mechanism based on the comprehensive abnormal index realizes the refined management of the tower and improves the efficiency of fault handling. The present invention not only significantly improves the accuracy and timeliness of abnormal monitoring of towers, but also realizes predictive maintenance and intelligent management, greatly enhances the safety and reliability of the distribution network, and provides strong technical support for the stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0028] Figure 1 This is a schematic diagram of the structure of a terminal in the first embodiment of the present invention;

[0029] Figure 2 It is a schematic diagram of a flow chart in the first embodiment of the present invention;

[0030] Figure 3 FIG. 4 is a schematic diagram of the structure of a computer device in a third embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0032] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0033] Example 1

[0034] Reference Figure 1 , as an embodiment of the present invention, provides a method for monitoring abnormalities of distribution network towers.

[0035] First, the distribution network tower abnormality monitoring method provided by the present application can be applied to Figure 1 In the terminal shown. Figure 1 As shown, the terminal may include one or two ( Figure 1 Only one is shown in the figure) processor and memory for storing data, wherein the processor may include but is not limited to a processing system such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may also include a transmission device and an input and output device for communication functions. It can be understood by ordinary technicians in this field that Figure 1 The structure shown is for illustration only and does not limit the structure of the above terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0036] The memory can be used to store computer programs, such as the computer program corresponding to the distribution network tower abnormality monitoring method in this embodiment. The processor executes various functional applications and data processing by running the computer program stored in the memory, that is, implementing the above method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0037] The transmission device is used to receive or send data via a network. The above network includes a wireless network provided by a communication provider of the terminal. In one example, the transmission device includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0038] like Figure 2 As shown, the embodiment of the present invention provides a method for monitoring abnormalities of distribution network towers, and the method is applied to Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0039] S1: Multimodal sensors are deployed on distribution network towers to collect tower status data.

[0040] Specifically, multimodal sensors include vibration sensors, temperature sensors, acoustic wave sensors, and current sensors. The vibration sensor collects the vibration frequency, amplitude, and acceleration data of the pole tower to monitor the structural stability of the pole tower and detect abnormal vibrations; the temperature sensor collects the temperature data of key parts of the pole tower, including the surface temperature of the wire connection points, transformers, and other electrical equipment, to monitor overheating and prevent fire hazards; the acoustic wave sensor collects the acoustic signals around the pole tower, including discharge sound, mechanical friction sound, and wind noise, to detect corona discharge, loose parts, and abnormal environmental noise; the current sensor collects the current data of the power line on the pole tower, including the current intensity, phase, and harmonic components, to monitor the load condition and detect electrical faults.

[0041] S2: Build a digital twin model of distribution network towers and predict the future status of the towers.

[0042] Specifically, a geometric model is constructed using 3D modeling software, including the shape (geometric features of structural components such as main poles, cross arms, and diagonal braces), dimensions (height, cross-sectional area, cross arm length, etc.), and material properties (elastic modulus, Poisson's ratio, density, etc.) of the tower;

[0043] Furthermore, the stress state of the tower under different load conditions is calculated to obtain the stress distribution and deformation of each part of the tower. The calculation formula for the stress state of the tower under different load conditions is as follows:

[0044]

[0045] in, is the divergence operator, which represents the divergence of the stress tensor σ, which represents the influence of the internal force on the material at position p and time t. σ(p,t) is the stress tensor, which represents the state of the force at position p and time t. f(p,t) represents the body force density function, which represents the density of the external force applied at position p and time t. ρ represents the material density, which reflects the mass density of the tower material. u(p,t) represents the displacement function at position p and time t.

[0046] Furthermore, according to the geometric model of the tower and the stress state of the tower under different load conditions, the node coordinates in the geometric model are corresponded to the stress tensor of the stress state to construct a digital twin model of the distribution network tower.

[0047] Furthermore, historical tower data is collected, and machine learning algorithms are used to process historical tower data and tower status data, including removing outliers and noise, processing missing data, and performing feature extraction, etc., to establish a mapping relationship between tower status data and distribution network tower digital twin model parameters, so that the distribution network tower digital twin model can reflect the current status of the tower.

[0048] Furthermore, based on the real-time tower status data, the digital twin model of the distribution network tower is continuously updated. The tower status data is converted into the parameters of the digital twin model of the distribution network tower using the mapping relationship established previously. The output of the digital twin model of the distribution network tower is compared with the actual observation data, and the digital twin model of the distribution network tower is calibrated.

[0049] Furthermore, the calibrated digital twin model of the distribution network tower is used for simulation. Based on the current state of the tower and the historical data of the tower, a time series analysis algorithm is used to predict the future state of the tower to obtain the prediction result of the future state of the tower.

[0050] Preferably, the digital twin model combines the physical characteristics of the tower with real-time data, which can more accurately reflect the actual status of the tower. By predicting the future status of the tower, it can identify potential problems and failure risks in advance, enabling staff to take preventive maintenance measures, avoid sudden failures, and reduce maintenance costs. Accurate predictions help reduce the occurrence of unexpected power outages and system failures, and improve the stability and reliability of the entire distribution network.

[0051] S3: Use drones to conduct regular inspections of towers and collect dynamic 3D data of towers.

[0052] Specifically, the distribution of pole towers is transformed into a graph theory problem. Each pole tower is regarded as a node. Heuristic information is defined, such as the distance between pole towers and their importance. The ant colony algorithm is used to consider multi-objective optimization, such as the shortest path length and priority visit to important pole towers. Through multiple iterations, the globally optimal inspection path is found.

[0053] Furthermore, considering the endurance of UAVs, clustering algorithms (such as K-means) are used to group the towers. According to weather forecast data, the sub-area divisions are dynamically adjusted to avoid unfavorable weather areas, and overlapping areas are designed to ensure seamless connection between areas.

[0054] Furthermore, factors such as power supply range, load size, and geographical location are considered to evaluate the importance of poles and towers. The inspection cycle, inspection frequency, and inspection priority are adjusted according to the importance of the poles and towers and historical abnormal records to ensure that high-risk towers are inspected in a timely manner.

[0055] Furthermore, the camera on the drone is used to collect all-round images of the tower, the lidar on the drone is used to perform three-dimensional scanning to obtain three-dimensional data of the tower, YOLOv8 is used to identify the components of the tower, and computer vision algorithms are used to process the all-round images of the tower in real time to identify surface anomalies of the tower and distinguish different types of anomalies such as cracks, corrosion, and deformation.

[0056] Preferably, by optimizing the inspection path and automating the inspection, the efficiency and coverage of the inspection are greatly improved, the need for manual inspection is reduced, and the three-dimensional data of the tower is collected in real time, providing rich data support for subsequent analysis. The inspection execution plan can be formulated according to the importance of the tower and the optimal inspection path to achieve the optimal allocation of resources; by considering weather factors, the reliability of the inspection work in various environments is ensured.

[0057] S4: Integrate tower status data, tower future status prediction results and tower dynamic 3D data to monitor tower anomalies and issue warnings.

[0058] Specifically, the multimodal sensor data is standardized, the prediction results of the digital twin model of the distribution network tower are converted into a unified format, the historical prediction errors are calculated, the tower status data are integrated, and key geometric features are extracted, such as the tower inclination angle and crossarm deformation.

[0059] Furthermore, the tower state data are standardized and merged through a weighted fusion algorithm to obtain the initial data X(t); the tower state data are combined with the tower future state prediction results, and the historical prediction errors are considered to obtain the optimized data X′(t); the optimized data X′(t) is combined with the tower three-dimensional data to generate the final fused data X″(t).

[0060] Furthermore, the calculation formula of the initial data X(t) is as follows:

[0061]

[0062] Among them, X(t) represents the initial data, α i Represents the corresponding data weight, indicating the importance of each sensor data, β i Represents the adjustment parameter, which is used to make nonlinear adjustments to the data. Represents the standardized data of the i-th sensor, including four types of data, express The norm of is used for normalization.

[0063] Further, Including vibration sensor data, temperature sensor data, sound wave sensor data and current sensor data.

[0064] Furthermore, the calculation formula of the optimized data X′(t) is as follows:

[0065]

[0066] Among them, X′(t) represents the optimized data, X(t) represents the initial data, S(t+Δt) is the prediction result of the future state of the tower, γ represents the weight coefficient of the prediction result of the future state of the tower, ε(t) represents the historical prediction error, which represents the deviation between the past prediction value and the actual measurement value, and λ represents the adjustment coefficient, which is used to adjust the historical prediction error.

[0067] Furthermore, the calculation formula of the final tower fusion data X″(t) is as follows:

[0068]

[0069] Among them, X″(t) represents the final fusion data of the tower, X′(t) represents the optimized data, F(t) represents the feature vector in the three-dimensional data of the tower, including the shape, size, material properties, structural deformation and surface damage of the tower; δ is the weight coefficient of the feature vector of the three-dimensional data of the tower, η represents the adjustment coefficient, which is used to perform nonlinear adjustment on the feature vector in the three-dimensional data of the tower, and ||F(t)|| represents the norm of F(t).

[0070] Furthermore, monitoring the abnormality of the tower and issuing an early warning includes calculating the comprehensive abnormality index of the tower based on the final fusion data X″(t) of the tower. The calculation formula of the comprehensive abnormality index of the tower is as follows:

[0071]

[0072] Among them, E(t) represents the comprehensive anomaly index, which indicates the abnormality degree of the current tower, σ represents the activation function, X″(t) represents the final fusion data of the tower, W represents the weight matrix, which indicates the weight distribution of each data feature of the tower, T represents the transposed matrix, ρ represents the adjustment coefficient, which is used for nonlinear adjustment of X″(t), and b represents the bias vector.

[0073] When E(t) < the first threshold (0.2), the state of the distribution network tower is normal. The conventional monitoring procedure is continuously executed, the frequency of collecting tower status data is maintained, and the digital twin model of the distribution network tower is regularly updated.

[0074] When the first threshold value (0.2) ≤ E(t) < the second threshold value (0.4), the monitoring system determines that there is a primary abnormality in the tower state and initiates a primary warning. At this time, the tower is diagnosed for abnormality, potential faults are identified, the sampling frequency of some multimodal sensors is increased, and a checklist is generated;

[0075] When the second threshold value (0.4) ≤ E(t) < the third threshold value (0.7), the monitoring system determines that there is an intermediate abnormality in the tower state and initiates an intermediate warning. At this time, the sampling frequency of all multimodal sensors is increased, the abnormal source is located, the impact range of the abnormality is evaluated, and an abnormality report and maintenance plan are generated;

[0076] When E(t) ≥ the third threshold (0.7), the monitoring system determines that the tower status is seriously abnormal and activates the advanced warning. The monitoring system immediately issues an emergency alarm, monitors the abnormal tower in real time, and generates emergency response plans and risk mitigation strategies.

[0077] In summary, through multi-source data, a comprehensive and accurate assessment of the tower status can be made, and multiple factors can be comprehensively considered to reduce false alarms and missed alarms, improve the accuracy of early warnings, and take appropriate treatment measures according to the degree of abnormality to achieve reasonable allocation of resources. The system can automatically perform status assessment, abnormality detection and early warning, reduce human intervention, and prevent more serious faults from occurring by promptly discovering and handling primary abnormalities. For serious abnormalities, the system can respond quickly and provide emergency measures to reduce the impact time of the fault. The tower abnormality monitoring system also has prediction and early warning capabilities, which significantly improves the reliability and safety of the distribution network.

[0078] Example 2

[0079] Reference Figure 1-2 , is an embodiment of the present invention, and provides a distribution network tower abnormality monitoring system, including: a data acquisition module, used to arrange multimodal sensors on the distribution network towers to collect tower status data; a first model construction module, used to build a distribution network tower digital twin model and predict the future state of the tower; a second model construction module, used to use drones to conduct regular inspections of the towers and collect dynamic three-dimensional data of the towers; an early warning module, used to integrate the tower status data, the tower future state prediction results and the tower dynamic three-dimensional data, monitor the tower abnormalities and issue early warnings.

[0080] For the specific definition of the distribution network tower abnormality monitoring system, please refer to the definition of the distribution network tower abnormality monitoring method mentioned above, which will not be repeated here. Each module in the above-mentioned distribution network tower abnormality monitoring system can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0081] Example 3

[0082] Reference Figure 3 , which is the third embodiment of the present invention. Based on the first two embodiments, the embodiment of the present invention provides a computer device, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 3As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium.

[0083] The database of the computer device is used to store action detection data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in any of the above-mentioned sparse tensor operation acceleration method embodiments are implemented.

[0084] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0085] In one embodiment, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in any of the above-mentioned sparse tensor operation acceleration method embodiments are implemented.

[0086] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0087] Example 4

[0088] As an embodiment of the present invention, a method for monitoring abnormalities of distribution network towers is provided. In order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through simulation experiments.

[0089] Specifically, in order to verify the effectiveness of constructing a digital twin model of distribution network towers, using intelligent drones for regular inspections, and integrating multi-source data to monitor tower anomalies and issue early warnings, the researchers selected 100 10kV distribution network towers under the jurisdiction of a provincial power company as test objects and conducted a 6-month field test.

[0090] During the six-month trial, the researchers recorded the system's performance indicators and compared them with traditional manual inspection methods. The results are shown in Table 1:

[0091] Table 1 Detailed data table of performance comparison between intelligent tower monitoring system and traditional methods for 6 months

[0092]

[0093] Preferably, the half-year average inspection efficiency of the present invention reaches 100.5 bases / day, while the traditional method is only 10.17 bases / day, which is improved by nearly 10 times. This efficiency improvement is mainly due to the application of drone technology and optimized path planning algorithms.

[0094] Furthermore, the abnormality detection rate of the present invention is always maintained above 96%, with a half-year average of 97.83%, which is 33.41% higher than the 73.33% of the traditional method. This significant improvement is mainly because the present invention integrates the tower status data, the tower future status prediction results and the tower dynamic three-dimensional data, and the monitoring data is more comprehensive.

[0095] Furthermore, the average early warning time of the present invention reaches 13.83 days, which is 728.14% earlier than the 1.67 days of the traditional method. This early warning capability provides operation and maintenance personnel with sufficient time to perform preventive maintenance, thereby effectively reducing the probability of failure.

[0096] Furthermore, the average number of failures per 100 towers in half a year of the present invention is reduced from 0.87 times to 0.17 times, a decrease of 80.46%, which directly reflects the excellent performance of the intelligent system in preventive maintenance.

[0097] In summary, the present invention shows significant advantages in efficiency, accuracy, cost control, energy efficiency, data collection capability, human resource optimization, weather adaptability and equipment maintenance. This all-round performance improvement not only greatly improves the reliability and safety of the distribution network, but also brings considerable economic benefits to power companies.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for monitoring abnormalities of distribution network towers, characterized in that: include: Multimodal sensors are deployed on distribution network towers to collect tower status data; Build a digital twin model of distribution network towers and predict the future status of towers; Use drones to conduct regular inspections of towers and collect dynamic 3D data of towers; Integrate tower status data, tower future status prediction results and tower dynamic 3D data to monitor tower anomalies and issue early warnings.

2. The method for monitoring abnormality of a distribution network tower according to claim 1, characterized in that: The multimodal sensor includes a vibration sensor, a temperature sensor, an acoustic wave sensor and a current sensor; Collecting the vibration frequency, amplitude and acceleration data of the tower through the vibration sensor; The temperature sensor is used to collect temperature data of key parts of the tower; the key parts of the tower include the wire connection point and the transformer; Acoustic signals around the tower are collected by the acoustic wave sensor; the acoustic signals include discharge sound, mechanical friction sound and wind noise; The current sensor is used to collect current data of the power line on the tower; the current data includes current intensity, phase and harmonic components; The tower status data includes vibration frequency, amplitude and acceleration data, temperature data, acoustic signals, and current data.

3. The method for monitoring abnormality of a distribution network tower according to claim 2, characterized in that: The step of predicting the future state of the tower comprises: Build geometric models using the shape, size and material properties of the tower; Calculate the stress state of the tower under different load conditions. in, is the divergence operator, which represents the divergence of the stress tensor σ, which represents the influence of the internal force on the material at position p and time t. σ(p,t) is the stress tensor, which represents the state of the force at position p and time t. f(p,t) represents the body force density function, which represents the density of the external force applied at position p and time t. ρ represents the material density of the tower material. u(p,t) represents the displacement function at position p and time t. Corresponding the node coordinates in the geometric model to the stress tensor of the stress state to construct a digital twin model of the distribution network tower; Collect historical tower data, use machine learning algorithms to process historical tower data and tower status data, establish a mapping relationship between tower status data and distribution network tower digital twin model parameters, and make the distribution network tower digital twin model reflect the current status of the tower; Based on the current state of the tower and the historical data of the tower, the time series analysis algorithm is used to predict the future state of the tower and obtain the prediction result of the future state of the tower.

4. The method for monitoring abnormality of a power distribution network tower according to claim 3, characterized in that: The acquisition of dynamic three-dimensional data of the tower includes: Based on the geographical location information of the tower, the ant colony algorithm is used to find the optimal inspection path; Divide inspection sub-areas based on drone endurance and weather conditions; Develop an inspection execution plan based on the optimal inspection path and the priorities of the inspection sub-areas and towers; According to the inspection execution plan, the drone uses the camera to collect all-round images of the tower to identify abnormalities, and uses the lidar to perform a three-dimensional scan of the tower to obtain three-dimensional data of the tower.

5. The method for monitoring abnormality of a distribution network tower according to claim 4, characterized in that: The fused tower state data, tower future state prediction results and tower dynamic three-dimensional data include: The tower status data is standardized and merged through the weighted fusion algorithm to obtain the initial data X(t); Combine the tower state data with the tower future state prediction results, consider the historical prediction error, and obtain the optimized data X′(t); The optimized data X′(t) is combined with the three-dimensional data of the tower to generate the final fused data X″(t) of the tower.

6. The method for monitoring abnormality of a distribution network tower according to claim 5, characterized in that: The calculation formula of the initial data X(t) is as follows: Among them, X(t) represents the initial data, α i represents the corresponding data weight, β i represents the adjustment parameter, represents the standardized data of the i-th sensor, express The norm of ; The calculation formula of the optimized data X′(t) is as follows: Among them, X′(t) represents the optimized data, X(t) represents the initial data, S(t+Δt) is the prediction result of the future state of the tower, γ represents the weight coefficient of the prediction result of the future state of the tower, ε(t) represents the historical prediction error, and λ represents the adjustment coefficient; The calculation formula of the final tower fusion data X″(t) is as follows: Among them, X″(t) represents the final fusion data of the tower, X′(t) represents the optimized data, F(t) represents the feature vector in the three-dimensional data of the tower, δ represents the weight coefficient of the feature vector of the three-dimensional data of the tower, η represents the adjustment coefficient, and ||F(t)|| represents the norm of F(t).

7. The method for monitoring abnormality of a distribution network tower according to claim 6, characterized in that: The monitoring of tower abnormalities and early warning include: Based on the final fusion data X″(t) of the tower, the tower comprehensive abnormality index is calculated. The calculation formula of the tower comprehensive abnormality index is as follows: Among them, E(t) represents the comprehensive abnormality index, σ represents the activation function, X″(t) represents the final fusion data of the tower, W represents the weight matrix, T represents the transposed matrix, ρ represents the adjustment coefficient, and b represents the bias vector; When E(t) is less than the first threshold, the state of the distribution network tower is normal, and the regular monitoring procedure is continuously executed, the frequency of collecting tower state data is maintained, and the digital twin model of the distribution network tower is regularly updated; When the first threshold ≤ E(t) < the second threshold, the monitoring system determines that there is a primary abnormality in the tower state and initiates a primary warning. At this time, the tower is diagnosed for abnormality, potential faults are identified, the sampling frequency of some multimodal sensors is increased, and a checklist is generated; When the second threshold ≤ E(t) < the third threshold, the monitoring system determines that there is an intermediate abnormality in the tower state and initiates an intermediate warning. At this time, the sampling frequency of all multimodal sensors is increased, the abnormal source is located, the impact range of the abnormality is evaluated, and an abnormality report and maintenance plan are generated; When E(t) ≥ the third threshold, the monitoring system determines that the tower status is seriously abnormal and activates the advanced warning. The monitoring system immediately issues an emergency alarm, conducts real-time monitoring of the abnormal tower, and generates emergency response plans and risk mitigation strategies.

8. A system using the distribution network tower abnormality monitoring method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module is used to arrange multi-modal sensors on the distribution network towers to collect tower status data; The first model building module is used to build a digital twin model of the distribution network tower and predict the future state of the tower; The second model building module is used to use drones to conduct regular inspections of the towers and collect dynamic three-dimensional data of the towers; The early warning module is used to integrate the tower status data, the tower future status prediction results and the tower dynamic three-dimensional data to monitor tower abnormalities and issue early warnings.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the distribution network tower abnormality monitoring method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distribution network tower abnormality monitoring method according to any one of claims 1 to 7 are implemented.

Images