Method for tracking trajectory of moving object in transformer substation based on three-dimensional digital model

Abstract

The invention provides a method for tracking the trajectory of a moving object in a transformer substation based on a three-dimensional digital model, and relates to the technical field of safety management and control, the method is applied to a server of a transformer substation management and control system, the transformer substation management and control system comprises the server, a plurality of controllers and a plurality of laser point cloud sensors, and the server is connected with the plurality of controllers simultaneously. Each controller is connected with a laser point cloud sensor. The method comprises the following steps: acquiring first position data transmitted by a first controller; according to a preset transformer substation space structure, calculating the probability that the moving target moves to the scanning area of each sensor; if it is determined that the probability that the moving target moves to the second scanning area is larger than or equal to a preset threshold value, an acquisition instruction is sent to a second controller; receiving change data transmitted by the second controller; and generating a continuous motion state of the three-dimensional moving target according to the change data. According to the invention, the data volume transmitted during trajectory tracking of the moving object in the transformer substation based on the three-dimensional digital model can be effectively reduced.

Description

Technical Field

[0001] The present application relates to the technical field of safety management and control, and specifically to a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model. Background Art

[0002] In the power system, substations are the core nodes of power transmission and conversion, and their safety and stability directly affect the normal operation of the entire power grid. Traditional substation monitoring and early warning methods rely on manual inspections and a small number of fixed monitoring equipment, which not only limits the efficiency and coverage of monitoring, but also brings high operating costs. In addition, when faced with emergencies, the response speed of traditional methods is slow, and it is difficult to meet the requirements of modern power systems for real-time monitoring and efficient early warning of substations. Therefore, an efficient and intelligent monitoring system is needed to improve the safety and response capabilities of substations, thereby ensuring the stable operation of the power grid.

[0003] With the development of 3D laser scanning and computer vision technology, the application of 3D reconstruction technology in power systems has attracted more and more attention. By monitoring and warning substation equipment through 3D point cloud data, the spatial information of the equipment can be obtained more comprehensively and accurately. Under conditions where vision is obstructed, such as haze and smoke, laser point cloud has higher tracking stability than video surveillance. Laser point cloud is not affected by changes in illumination and can continue to work in complex or dim lighting environments such as night, shadows, and strong light, which is particularly important for all-weather monitoring of substations. Laser point cloud modeling can also accurately measure the physical parameters of the equipment (such as size, distance, and position), which is convenient for spatial modeling and subsequent analysis. However, due to the large amount of data generated by laser point cloud, delays may occur during transmission, thus affecting real-time performance, which puts higher requirements on trajectory tracking. Summary of the invention

[0004] The present application provides a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model, which can effectively reduce the amount of data transmitted when tracking the trajectory of a moving object in a substation based on a three-dimensional digital model.

[0005] In a first aspect of the present application, a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model is provided. The method is applied to a server of a substation management and control system. The substation management and control system includes the server, multiple controllers, and multiple laser point cloud sensors. The server is connected to multiple controllers at the same time, and each of the controllers is connected to one laser point cloud sensor. The method includes: Acquire first position data transmitted by a first controller, wherein the first controller sends the first position data of the first scanning area to the server if it is determined that there is a moving target in the first scanning area of ​​the first sensor based on the acquired first point cloud data collected by the first sensor, the first controller is any one of the plurality of controllers, and the first sensor is a laser point cloud sensor connected to the first controller among the plurality of laser point cloud sensors; According to the preset substation spatial structure, the probability of the mobile target moving to the scanning area of ​​each sensor is calculated; If it is determined that the probability that the mobile target moves to the second scanning area is greater than or equal to a preset threshold, a collection instruction is sent to a second controller, wherein the second controller is any one of the multiple controllers except the first controller, the scanning area of ​​the second sensor is the second scanning area, and the second sensor is a laser point cloud sensor connected to the second controller among the multiple laser point cloud sensors; receiving the change data transmitted by the second controller, wherein after receiving the acquisition instruction, the second sensor controls the second sensor to perform continuous scanning, and compares two consecutive second point cloud data point by point to find the changed data part according to the second point cloud data obtained by multiple scans, so as to obtain the change data; A three-dimensional continuous motion state of the moving object is generated according to the change data.

[0006] Optionally, the calculating, according to a preset substation spatial structure, the probability that the mobile target moves to the scanning area of ​​each sensor specifically includes: Acquire a preset substation spatial structure, wherein the substation spatial structure includes a connection structure of laser point cloud sensors based on an internal route of the substation and a connection distance between two mutually connected laser point cloud sensors, the connection structure uses the installation positions of the laser point cloud sensors as nodes, uses the internal route of the substation as links, and nodes are connected through the internal route of the substation; Determining continuous position changes of the mobile target in the substation according to the received first point cloud data obtained by multiple scans transmitted by the first controller; Determining a moving direction of the moving target according to the continuous position change; Acquire historical probabilities of different moving objects moving from the first scanning area to each of the scanning areas in historical records; The substation spatial structure, the moving direction and the historical probability are weightedly fused to calculate the probability of the moving target moving to the scanning area of ​​each sensor.

[0007] Optionally, before determining that the probability that the moving target moves to the second scanning area is greater than or equal to a preset threshold, the method further includes: The substation spatial structure, the moving direction and the historical probability are weighted and fused to calculate the probability of the moving target moving to the second scanning area, which is specifically calculated by the following formula:

[0008] Wherein, P is the probability that the mobile target moves to the second scanning area, σ is the Sigmoid function, f1=f(θ), f2=f(d), f3=f(h), w i is the weight coefficient, when i=1, w1 is the direction weight coefficient, when i=2, w2 is the distance weight coefficient, when i=3, w3 is the history weight coefficient, α is the direction sensitivity, θ is the angle between the moving direction and the moving direction, the moving direction is the direction from the center of the first scanning area to the center of the second scanning area, θ0 is the expected angle based on the first scanning area, d 12 is the distance between the first scanning area and the second scanning area, τ is a small constant, γ is the distance sensitivity coefficient, λ is the time decay factor, T is the total number of historical records, P t (1→2) is the transition probability from the first scanning area to the second scanning area recorded at the tth time.

[0009] Optionally, before comparing two consecutive second point cloud data point by point to find a changed data portion based on the second point cloud data obtained by multiple scans and obtaining the changed data, the method further includes: The background point cloud data is sent to the second controller so that the second controller compares the background point cloud data with the real-time point cloud data, and uses a background difference algorithm to extract the point cloud data corresponding to the mobile target in the real-time point cloud data, thereby obtaining the second point cloud data.

[0010] Optionally, the step of comparing two consecutive second point cloud data point by point to find a changed data portion based on the second point cloud data obtained by multiple scans to obtain the changed data specifically includes: The second controller initially aligns the first sub-data with the second sub-data, where the first sub-data and the second sub-data are any two adjacent second point cloud data among the plurality of second point cloud data; The second controller queries multiple groups of first data points and second data points in the first sub-data and the second sub-data according to the initial registration result, wherein the first data point is any point in the first sub-data, and the second data point is a point in the second sub-data that is closest to the first data point; The second controller calculates a rigid body transformation between the first data point and the second data point so that the first sub-data and the second sub-data coincide with each other; The second controller continuously iterates and updates the calculation until the rigid body transformation error converges below the threshold, completing the precise point cloud registration; After the second controller completes the precise point cloud registration, the moved or newly added data points can be found by comparing the spatial position changes of each data point in the first sub-data and the second sub-data to obtain the change data.

[0011] Optionally, the second controller calculates a rigid body transformation between the first data point and the second data point so that the first sub-data overlaps with the second sub-data, specifically including: The second controller calculates the first sub-data {p1, p 2, …,p n}, calculate the first centroid of the second sub-data {q1,q2,…,q n The second centroid of} is calculated by the following formula:

[0012] Among them, c p is the first centroid, c q is the second centroid, {p1,p 2, …,p n} is the first sub-data, {q1,q2,…,q n} is the second sub-data; The second controller calculates the covariance matrix, specifically by the following formula:

[0013] Where H is the covariance matrix, c p is the first centroid, c q is the second centroid, {p1,p 2, …,p n} is the first sub-data, {q1,q2,…,q n} is the second sub-data; The second controller constructs a rigid body transformation matrix based on the covariance matrix, as follows:

[0014] Wherein, T is the rigid body transformation matrix, H is the covariance matrix, E is the diagonal matrix, c p is the first centroid, c q is the second centroid; The second controller applies the rigid body transformation matrix to the first sub-data, updates the first data point, and obtains an updated first data point, as follows:

[0015] Among them, p i ' is the first data point after update, H is the covariance matrix, E is the diagonal matrix, p i is the first data point.

[0016] Optionally, generating the three-dimensional continuous motion state of the moving target according to the change data specifically includes: By extracting the centroid position of each change data, a preliminary trajectory of the moving target is obtained; Accumulate the change data obtained from multiple scans into a global coordinate system to form a cumulative point cloud; Based on the accumulated point cloud, surface reconstruction is performed to obtain the target three-dimensional model; According to the time series, a dynamic three-dimensional dynamic model is formed.

[0017] In a second aspect of the present application, a device for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model is provided. The device is a server, and the device includes an acquisition module, a processing module, and a sending module, wherein: The acquisition module is used to acquire first position data transmitted by a first controller, wherein the first controller sends the first position data of the first scanning area to the server if it is determined that there is a moving target in the first scanning area of ​​the first sensor based on the acquired first point cloud data collected by the first sensor, the first controller is any one of a plurality of controllers, and the first sensor is a laser point cloud sensor connected to the first controller among a plurality of laser point cloud sensors; The processing module is used to calculate the probability of the mobile target moving to the scanning area of ​​each sensor according to the preset substation spatial structure; The sending module is used for sending a collection instruction to a second controller if it is determined that the probability of the mobile target moving to the second scanning area is greater than or equal to a preset threshold, wherein the second controller is any one of the multiple controllers except the first controller, the scanning area of ​​the second sensor is the second scanning area, and the second sensor is a laser point cloud sensor connected to the second controller among the multiple laser point cloud sensors; The acquisition module is used to receive the change data transmitted by the second controller, wherein after the second sensor receives the acquisition instruction, the second sensor is controlled to perform continuous scanning, and according to the second point cloud data obtained by multiple scans, two consecutive second point cloud data are compared point by point to find the changed data part, so as to obtain the change data; The processing module is used to generate a three-dimensional continuous motion state of the moving target according to the change data.

[0018] Optionally, the acquisition module is used to acquire a preset substation spatial structure, wherein the substation spatial structure includes a connection structure of laser point cloud sensors based on an internal route of the substation and a connection distance between two mutually connected laser point cloud sensors, the connection structure uses the installation position of the laser point cloud sensor as a node, the internal route of the substation as a link, and the nodes are connected through the internal route of the substation; The processing module is used to determine the continuous position change of the mobile target in the substation according to the first point cloud data obtained by multiple scans transmitted by the first controller; The processing module is used to determine the moving direction of the moving target according to the continuous position change; The acquisition module is used to acquire the historical probability of different moving objects moving from the first scanning area to each of the scanning areas in the historical records; The processing module is used to perform weighted product fusion on the spatial structure of the substation, the moving direction and the historical probability, and calculate the probability that the moving target moves to the scanning area of ​​each sensor.

[0019] Optionally, the processing module is used to perform weighted product fusion on the spatial structure of the substation, the moving direction and the historical probability to calculate the probability of the moving target moving to the second scanning area, specifically by the following formula:

[0020] Wherein, P is the probability that the mobile target moves to the second scanning area, σ is the Sigmoid function, f1=f(θ), f2=f(d), f3=f(h), w i is the weight coefficient, when i=1, w1 is the direction weight coefficient, when i=2, w2 is the distance weight coefficient, when i=3, w3 is the history weight coefficient, α is the direction sensitivity, θ is the angle between the moving direction and the moving direction, the moving direction is the direction from the center of the first scanning area to the center of the second scanning area, θ0 is the expected angle based on the first scanning area, d 12is the distance between the first scanning area and the second scanning area, τ is a small constant, γ is the distance sensitivity coefficient, λ is the time decay factor, T is the total number of historical records, P t (1→2) is the transition probability from the first scanning area to the second scanning area recorded at the tth time.

[0021] Optionally, the sending module is used to send background point cloud data to the second controller, so that the second controller compares the background point cloud data with the real-time point cloud data, and uses a background difference algorithm to extract the point cloud data corresponding to the mobile target in the real-time point cloud data, thereby obtaining the second point cloud data.

[0022] Optionally, the processing module is used to obtain a preliminary trajectory of the moving target by extracting the centroid position of each change data; The processing module is used to accumulate the change data obtained by multiple scans into a global coordinate system to form an accumulated point cloud; The processing module is used to perform surface reconstruction based on the accumulated point cloud to obtain a target three-dimensional model; The processing module is used to form a dynamic three-dimensional dynamic model according to the time series.

[0023] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0024] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0025] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. This application reduces the amount of data transmission through distributed control and local computing, mainly relying on the local processing of the controller and the efficient scheduling of the server. Specifically, the controller analyzes the point cloud data locally and reports to the server only when a moving target is detected and it is calculated that the probability of it entering other sensor areas is high, avoiding the continuous transmission of large amounts of unchanged point cloud data. In addition, through probabilistic prediction, the acquisition instructions of adjacent controllers are triggered only when necessary, further reducing redundant data transmission and ensuring that only key data such as change data and trajectory updates are uploaded to the server, thereby greatly reducing the amount of data and network load.

[0026] 2. This application relies on local processing of the controller and efficient scheduling of the server, as well as reducing the amount of data transmission through distributed control and local computing. Specifically, the controller analyzes the point cloud data locally and reports to the server only when a moving target is detected and it is calculated that it has a high probability of entering other sensor areas, avoiding the continuous transmission of large amounts of unchanged point cloud data. In addition, through probabilistic prediction, the acquisition instructions of adjacent controllers are triggered only when necessary, further reducing redundant data transmission and ensuring that only critical data is uploaded to the server, thereby significantly reducing the amount of data and network load.

[0027] 3. This application gradually calculates the centroid, covariance matrix and rigid body transformation matrix of the point cloud data, realizes accurate point cloud data registration and difference detection, and effectively identifies the change data of the moving target. Through multiple iterative calculations, two consecutive point cloud frames overlap to ensure the accuracy of data alignment. This method greatly improves the registration accuracy and stability, can accurately identify moving or newly added data points, and generate clear motion trajectories and states. In particular, through point-by-point comparison and precise registration of rigid body transformation, efficient and continuous target tracking is achieved while reducing errors, meeting the high-precision monitoring requirements in the complex environment of the substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model disclosed in an embodiment of the present application; Figure 2 It is a structural schematic diagram of a substation management and control system disclosed in an embodiment of the present application; Figure 3 It is a module schematic diagram of a mobile object trajectory tracking device in a substation based on a three-dimensional digital model disclosed in an embodiment of the present application; Figure 4 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0029] Explanation of the reference numerals: 301, acquisition module; 302, processing module; 303, sending module; 401, processor; 402, communication bus; 403, user interface; 404, network interface; 405, memory. DETAILED DESCRIPTION

[0030] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0031] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0032] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0033] Substations are the core nodes of the power system. Their safety and stability are crucial to the normal operation of the power grid. Traditional monitoring methods rely on manual inspections and fixed equipment, and have problems such as low monitoring efficiency and slow response speed, making it difficult to meet real-time monitoring needs. With the development of 3D laser scanning and computer vision technology, the use of 3D point cloud data for substation monitoring and early warning has gradually become a trend. Laser point clouds have the advantages of being unaffected by light and adapting to complex environments. They can efficiently track and accurately measure equipment information under various lighting conditions, which is helpful for all-weather monitoring and equipment analysis. However, the transmission of large amounts of point cloud data may cause delays, increasing the real-time challenges of trajectory tracking.

[0034] This embodiment discloses a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model. Figure 1 , including the following steps S110-S150: S110, acquiring first position data transmitted by the first controller.

[0035] A method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model disclosed in an embodiment of the present application is applied to a server of a substation management and control system, referring to Figure 2 The substation management and control system includes a server, multiple controllers and multiple laser point cloud sensors. The server is connected to multiple controllers at the same time, and each controller is connected to a laser point cloud sensor. The laser point cloud sensors and controllers are set in different areas of the substation according to actual monitoring needs.

[0036] The server includes but is not limited to electronic devices such as mobile phones, tablet computers, wearable devices, PCs (Personal Computers), etc., and can also be a background server that runs a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model. The server can be implemented as an independent server or a server cluster consisting of multiple servers.

[0037] The controller is one of the key components in the substation control system, used to connect the laser point cloud sensor and perform localized data processing. It receives the data collected by the sensor and performs preliminary analysis and processing, such as detecting change points, judging object movement, etc., reducing unnecessary data volume, and preferentially filtering and compressing important information before transmitting it to the server. The controller can also coordinate with other controllers to jointly complete tasks such as regional monitoring and data verification to achieve efficient distributed data management.

[0038] The laser point cloud sensor is a device used for high-precision 3D scanning in substations. It generates 3D point cloud data of the equipment and environment in the substation by emitting lasers and receiving reflected signals. The sensor can operate under various light conditions and is not affected by factors such as light and weather. It is suitable for all-weather monitoring at night or in complex environments. By accurately measuring the shape, position and changes of equipment, the laser point cloud sensor provides accurate and reliable data support for equipment monitoring, anomaly detection and 3D modeling.

[0039] The first controller controls the connected first laser point cloud sensor to scan its monitoring area periodically or continuously to obtain three-dimensional point cloud data in the area. Each scan generates a frame of three-dimensional point cloud data, which includes the shape and position of objects in the monitoring area in the substation. The first controller is any one of the multiple controllers, and the first sensor is a laser point cloud sensor connected to the first controller among the multiple laser point cloud sensors; The first controller performs a difference detection between the current frame (the most recently scanned point cloud data) and the point cloud data of the previous frame to determine whether there is a moving target. The position of each point in the point cloud of the previous and next frames is compared. If the displacement of some points exceeds the set threshold, it is considered that these points may belong to the moving target. The controller determines whether the changed area is a valid moving target based on the set threshold (such as displacement distance, change speed).

[0040] If the first controller determines that there is a mobile target that meets the conditions in the scanning area, the position information of the target is encapsulated as "first position data". The position data is compressed or formatted to meet the transmission requirements to ensure the efficiency and accuracy of the transmission. The data may include the center of mass position, boundary size, movement direction and speed of the target. The controller sends the position data to the server through the network. After receiving the data, the server performs global monitoring and analysis to decide whether to further dispatch other controllers or execute early warning.

[0041] S120, calculating the probability of the mobile target moving to the scanning area of ​​each sensor according to the preset substation spatial structure.

[0042] In a possible implementation, according to a preset substation spatial structure, the probability of a mobile target moving to a scanning area of ​​each sensor is calculated, specifically comprising: obtaining a preset substation spatial structure, wherein the substation spatial structure includes a connection structure of a laser point cloud sensor based on an internal route of the substation and a connection distance between two mutually connected laser point cloud sensors, the connection structure using the installation position of the laser point cloud sensor as a node and the internal route of the substation as a link, and the nodes are connected to each other through the internal route of the substation; determining the continuous position change of the mobile target in the substation according to first point cloud data obtained by multiple scans transmitted by a received first controller; determining the moving direction of the mobile target according to the continuous position change; obtaining the historical probability of different mobile objects moving from the first scanning area to each scanning area in the historical record; performing weighted product fusion on the substation spatial structure, the moving direction and the historical probability, and calculating the probability of the mobile target moving to the scanning area of ​​each sensor.

[0043] Specifically, the spatial structure information of the substation includes the installation location of the laser point cloud sensor, the connection relationship and distance between any two adjacent sensors, etc. These structural data can be obtained through 3D modeling or on-site measurement. In the spatial structure of the substation, the installation location of each laser point cloud sensor is regarded as a node, and each node is connected through the internal route of the substation to form a network graph structure. The distance between adjacent sensors is obtained, for example, defined by straight-line distance or path length. This distance information will be used to estimate the migration probability of the mobile target. The adjacent relationship between sensors can be constructed into a topological structure graph using a graph representation method to facilitate the subsequent calculation of paths and probabilities.

[0044] Through the scanning data transmitted by the first controller multiple times, the continuous position change of the mobile target in the substation is tracked and its moving direction is calculated. In multiple scans, the center of mass position of the mobile target is obtained each time, and these positions are recorded to form a trajectory sequence, and then the movement direction is calculated through adjacent position points.

[0045] Based on historical data, the frequency of different moving objects moving from one scanning area to other areas is counted to obtain the historical transition probability from the first scanning area to each target scanning area. For different target areas, the historical transition probability Ph(A→B) of moving from the current area to the adjacent area is calculated. For example, assuming that the target has 100 movements starting from area A, 40 of which move to area B, then the historical probability is 0.4.

[0046] Based on the known spatial structure, movement direction, and historical probability of the substation, the probability of the moving target reaching each sensor scanning area is calculated. The probability of the moving target moving from the first scanning area to the second scanning area is calculated. The specific calculation process is as follows: First, the direction consistency mapping function is calculated. The direction consistency mapping function is used to calculate the degree of consistency between the current direction of the target and the expected moving direction. This function is expressed using the Gaussian distribution formula and can smoothly reflect the impact of direction deviation on probability:

[0047] Among them, α is the directional sensitivity, which is used to control the degree of influence of directional consistency on probability. The larger the value, the more sensitive the system is to directional deviations, and even a small deviation will lead to a significant decrease in probability. θ is the angle between the moving direction and the moving direction. The moving direction is the direction from the center of the first scanning area to the center of the second scanning area, and θ0 is the expected angle based on the first scanning area.

[0048] The characteristics of Gaussian distribution make the formula take the maximum value when θ=θ0, and as the direction deviation |θ-θ0| increases, the value decreases exponentially. In this way, the closer the movement is to the desired direction, the higher the probability, and the further it deviates from the direction, the lower the probability. This method can smoothly control the impact of the consistency of the target direction on the final probability.

[0049] Then calculate the distance mapping function, which is used to calculate the distance effect of the target moving from the first scanning area to the second scanning area. The closer the distance, the greater the probability; the farther the distance, the smaller the probability:

[0050] Among them, d 12 is the distance between the first scanning area and the second scanning area. The larger the distance, the more difficult it is to migrate from the current area to the target area. τ is a small constant used to prevent the denominator from being zero. γ is the distance sensitivity coefficient. The larger the value, the more obvious the attenuation of the distance.

[0051] The exponential decay property makes the probability of short-distance migration greater, while the probability of long-distance migration decreases significantly with increasing distance. This can effectively limit unreasonable long-distance predictions and make the model more in line with actual conditions.

[0052] Then calculate the historical transition probability mapping function, which uses a weighted average method with time decay to reflect the probability of the target migrating from the first scanning area to the second scanning area in the historical records:

[0053] Where λ is the time decay factor, which makes the most recent records more important and the earlier records less important. T is the total number of historical records, P t (1→2) is the transition probability from the first scanning area to the second scanning area recorded at the tth time.

[0054] Weighted time averaging incorporates the idea of ​​time decay, allowing the system to focus more on recent behavior patterns while retaining long-term trends. With the effect of the time decay factor, the formula is more sensitive to recent data, which can adapt to short-term changes in the environment and improve the accuracy of predictions.

[0055] The above three mapping functions are respectively regarded as f1=f(θ), f2=f(d), and f3=f(h), and weighted product fusion is performed according to the weight coefficient:

[0056] Where P is the probability that the mobile target moves to the second scanning area, σ is the Sigmoid function, f1=f(θ), f2=f(d), f3=f(h), w i is the weight coefficient. When i=1, w1 is the direction weight coefficient. When i=2, w2 is the distance weight coefficient. When i=3, w3 is the history weight coefficient.

[0057] The product part represents the weighted product calculation result of the three factors. The Sigmoid function is used to map the result to the range of 0 to 1:

[0058] Therefore, the final formula is:

[0059] The product model ensures that the probability is high when multiple factors are satisfied, so each factor must be high to maintain a high value for the overall probability. The weight of each factor adjusts the relative importance of each factor, allowing flexible control of the influence of different factors in the fusion. After the fusion result, the Sigmoid function is used to smoothly map it to the range of 0 to 1. This mapping makes the fusion result more in line with the probability distribution requirements, and the smoothing process can avoid excessive influence of extreme values ​​on the final probability.

[0060] S130: If it is determined that the probability that the mobile target moves to the second scanning area is greater than or equal to a preset threshold, a collection instruction is sent to the second controller.

[0061] As mentioned above, the probability of the mobile target entering the second scanning area is calculated by the weighted product fusion method, and a preset probability threshold is set to determine whether the target is likely to enter the second scanning area. This threshold can be adjusted according to the error requirements and monitoring requirements of the actual environment. For example, it can be set according to the success rate and error range of the target entering the adjacent area. The threshold is adjusted according to historical data or statistical analysis to adapt to different environmental requirements. For example, in areas where the target moves frequently or the path is complex, the threshold can be appropriately lowered to detect more sensitively.

[0062] Once the probability P is calculated, compare it with the threshold P threshold For comparison: If P ≥ P threshold , which means that the target is more likely to enter the second scanning area and trigger subsequent control instructions.

[0063] If P <P threshold , the current monitoring state is maintained and no collection instruction is sent to the second controller.

[0064] If it is determined that the conditions are met, a collection instruction is sent to the second controller to activate the laser point cloud sensor in the second scanning area to monitor the target entry situation.

[0065] S140, receiving change data transmitted by the second controller.

[0066] In a possible implementation, before comparing two consecutive second point cloud data point by point to find the changed data portion based on the second point cloud data obtained by multiple scans, and obtaining the changed data, the method also includes: sending background point cloud data to a second controller, so that the second controller compares the background point cloud data with the real-time point cloud data, and uses a background difference algorithm to extract the point cloud data corresponding to the moving target in the real-time point cloud data, thereby obtaining the second point cloud data.

[0067] Specifically, in the normal static state of the substation, the laser point cloud sensor of each controller collects one or more frames of point cloud data, and after denoising, generates a stable background point cloud data. In order to improve stability, the background data can be obtained by multi-frame averaging or long-term sampling. The collected background point cloud data is sent to the second controller. The background point cloud data may contain the spatial coordinate information of static objects in the target scanning area, providing a benchmark for subsequent comparison.

[0068] When the system is operating normally, the second controller will continuously collect real-time point cloud data to detect whether there is a moving target. The second controller controls the laser point cloud sensor connected to it to collect point cloud data in the current area at a fixed frequency to generate real-time point cloud data frames. The real-time point cloud data is preprocessed, such as noise removal and coordinate alignment, to ensure that the quality of the real-time point cloud data is consistent with the background point cloud data for subsequent comparison.

[0069] The second controller aligns the real-time point cloud data to the coordinate system of the background point cloud data. Usually, the ICP (Iterative Closest Point) or NDT (Normal Distributions Transform) algorithm is used for registration to ensure that the two sets of point cloud data are in the same reference frame. Each point in the background point cloud and the real-time point cloud is compared point by point, and the Euclidean distance between each pair of corresponding points is calculated. At the same time, a distance threshold is defined. If the distance of a point exceeds the threshold, the point is regarded as a change point. All points exceeding the threshold are retained to form the change point cloud data. After background difference and screening processing, the point cloud data of the moving target is obtained, which is defined as the "second point cloud data".

[0070] After receiving the acquisition instruction, the second sensor is controlled to perform continuous scanning, and based on the second point cloud data obtained by multiple scans, two consecutive second point cloud data are compared point by point to find the changed data part, so as to obtain the changed data.

[0071] In a possible implementation, based on the second point cloud data obtained by multiple scans, two consecutive second point cloud data are compared point by point to find the changed data part and obtain the changed data, which specifically includes: the second controller initially aligns the first sub-data with the second sub-data, and the first sub-data and the second sub-data are any two adjacent second point cloud data in the multiple second point cloud data; the second controller queries multiple groups of first data points and second data points in the first sub-data and the second sub-data according to the initial alignment result, wherein the first data point is any point in the first sub-data, and the second data point is the point in the second sub-data closest to the first data point; the second controller calculates the rigid body transformation between the first data point and the second data point so that the first sub-data and the second sub-data coincide with each other; the second controller continuously iterates and updates the calculation until the rigid body transformation error converges below the threshold to complete the precise point cloud alignment; after the second controller completes the precise point cloud alignment, by comparing the spatial position changes of each data point in the first sub-data and the second sub-data, the moved or newly added data points can be found to obtain the changed data.

[0072] Specifically, first, the centroids of the first sub-data and the second sub-data need to be calculated so as to calculate the rotation relationship of the point cloud while removing the translation effect. The second controller calculates the first sub-data {p1, p 2, …,pn}, calculate the first centroid of the second sub-data {q1,q2,…,q n The second centroid of} is calculated by the following formula:

[0073] Among them, c p is the first centroid, c q is the second centroid, {p1,p 2, …,p n} is the first sub-data, {q1,q2,…,q n} is the second sub-data.

[0074] The purpose of calculating the center of mass is to decentralize the two sets of point cloud data, so as to facilitate the subsequent calculation of the rotation matrix to align the two sets of data. By subtracting the corresponding center of mass from each point, decentralized point cloud data is obtained. After decentralization, the translation offset can be eliminated, so that the calculation is focused on rotation alignment. Based on the decentralized data, the covariance matrix between the first sub-data and the second sub-data is calculated to construct the rigid body transformation matrix. It is calculated specifically by the following formula:

[0075] Where H is the covariance matrix, c p is the first centroid, c q is the second centroid, {p1,p 2, …,p n} is the first sub-data, {q1,q2,…,q n} is the second sub-data.

[0076] The calculation of the covariance matrix H involves the "similarity" of two sets of decentralized point clouds. For each pair of corresponding points p i and q i , first subtract the centroids of each point to get the decentralized points, then calculate the outer product of the two points and sum them up for all points. The covariance matrix reflects the degree of alignment between point cloud data.

[0077] Perform singular value decomposition (SVD) on the covariance matrix H and decompose H into three matrices:

[0078] Among them, U and V are orthogonal matrices, representing the rotation relationship. E is a diagonal matrix containing singular values.

[0079] Singular value decomposition (SVD) is a matrix decomposition method that decomposes the matrix H into the product of U, E, and V. U and V are orthogonal matrices that represent rotation information, and E is a diagonal matrix that contains the singular values ​​(scaling factors) of the covariance matrix, reflecting the scaling relationship between the two point cloud sets in different directions. The diagonal matrix E is a diagonal matrix in the singular value decomposition process, and its expression is:

[0080] During the singular value decomposition process, the diagonal matrix E contains the singular values ​​of the matrix H, σ1, σ2, and σ3, which reflect the scaling relationship between the two point cloud sets in different directions.

[0081] Combine the rotation matrix and translation vector to construct a 4x4 rigid body transformation matrix T:

[0082] The rigid body transformation matrix T is used to rotate and translate the first sub-data into the coordinate system of the second sub-data. The rotation matrix R is located in the upper left corner of the matrix, and the translation vector T is located in the upper right corner of the matrix. The rigid body transformation matrix can act on points in the form of homogeneous coordinates to achieve rotation and translation operations.

[0083] Use the rigid body transformation matrix T to transform each point of the first sub-data to obtain the updated first sub-data point, as follows:

[0084] Among them, p i ' is the first data point after update, H is the covariance matrix, E is the diagonal matrix, p i is the first data point. For each point of the first sub-data, rotation and translation transformations are applied to transform the point from the coordinate system of the first sub-data to the coordinate system of the second sub-data. In this way, the first sub-data and the second sub-data can be overlapped to achieve accurate registration.

[0085] After the rigid body transformation is applied, the registration error is checked. If the registration error does not converge below the preset threshold, the updated first sub-data is used as input again, and the above steps are repeated to continue iteration until the registration error reaches the convergence condition.

[0086] After completing the precise registration of the two frames of point cloud data, the moved or newly added points are identified by comparing the spatial position difference of each point in the first sub-data and the second sub-data. The distance of each paired point is compared. If the distance exceeds the set threshold, it is determined as a changed data point. Finally, all points exceeding the distance threshold are collected to form a changed data set, which represents the moved or newly added parts in two consecutive frames of point cloud data.

[0087] S150, generating a continuous motion state of a three-dimensional moving object according to the change data.

[0088] In one possible implementation, a continuous motion state of a three-dimensional moving target is generated based on change data, specifically including: obtaining a preliminary trajectory of the moving target by extracting the center of mass position of each change data; accumulating the change data obtained from multiple scans into a global coordinate system to form a cumulative point cloud; performing surface reconstruction based on the accumulated point cloud to obtain a three-dimensional model of the target; and forming a dynamic three-dimensional dynamic model based on a time series.

[0089] Specifically, the center of mass position of the moving target can be obtained by scanning the change data obtained multiple times, and the initial motion trajectory of the target can be generated using the center of mass position. For each frame of change data, the center of mass position of the point cloud is calculated. The center of mass position reflects the center of the point cloud data of the frame, which is the spatial position of the target at that moment.

[0090] Connect the centroids of each frame in chronological order to form a preliminary trajectory of the target. These centroids form a trajectory line representing the path of the moving target, recording the movement of the target in space. In order to generate a complete 3D model of the moving target, the change data at different time points need to be accumulated into a global coordinate system. The point cloud data obtained from each scan needs to be aligned in the global coordinate system to ensure that the change data between different frames can be accumulated into the same coordinate system. The aligned change data of each frame are accumulated together to form the cumulative point cloud of the target. The accumulated point cloud contains the position and shape information of the target at each time, providing a complete 3D data foundation.

[0091] Based on the accumulated point cloud, a complete 3D model of the moving target is generated through a surface reconstruction algorithm. Surface reconstruction is used to convert dense point cloud data into a continuous 3D surface to display the geometric shape of the target. For areas with sparse point clouds, interpolation methods can be used to fill in the missing parts to ensure the integrity of the model. Interpolation completion can eliminate discontinuous edges or holes.

[0092] By accumulating point cloud and time series information, a dynamic three-dimensional model of the moving target is formed to show its continuous motion state. Based on the time series, the accumulated point cloud data is updated frame by frame to form a continuous dynamic three-dimensional model. For each frame of point cloud data, its coordinates and time information are combined to update the three-dimensional model of the target. After each new scan obtains the change data, it is registered to the global coordinate system and the three-dimensional model is updated frame by frame. The dynamic model not only contains the morphological changes of the target, but also combines its motion trajectory. The displacement, rotation and other motion states of the target are recorded through the time series, so that the three-dimensional model can reflect the real motion process of the target. After the dynamic model is generated, a visual three-dimensional animation or continuous motion state output can be further generated for display or monitoring. The accumulated three-dimensional model and the time series trajectory are combined to show the continuous motion process of the target in the form of animation. The animation display can clearly show the moving path and morphological changes of the target. The key parameters of the three-dimensional dynamic model, including position, speed, direction and other information, are extracted and output in the form of time series for monitoring and analysis.

[0093] This embodiment also discloses a device for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model. The device is a server, and the device includes an acquisition module 301, a processing module 302, and a sending module 303, wherein: The acquisition module 301 is used to acquire first position data transmitted by the first controller, wherein the first controller sends the first position data of the first scanning area to the server if it is determined that there is a moving target in the first scanning area of ​​the first sensor based on the first point cloud data collected by the first sensor. The first controller is any one of the multiple controllers, and the first sensor is a laser point cloud sensor connected to the first controller among the multiple laser point cloud sensors.

[0094] The processing module 302 is used to calculate the probability of the mobile target moving to the scanning area of ​​each sensor according to the preset substation spatial structure.

[0095] The sending module 303 is used to send an acquisition instruction to the second controller if it is determined that the probability of the mobile target moving to the second scanning area is greater than or equal to a preset threshold, wherein the second controller is any one of the multiple controllers except the first controller, the scanning area of ​​the second sensor is the second scanning area, and the second sensor is a laser point cloud sensor connected to the second controller among the multiple laser point cloud sensors.

[0096] The acquisition module 301 is used to receive the change data transmitted by the second controller, wherein after the second sensor receives the acquisition instruction, the second sensor is controlled to perform continuous scanning, and based on the second point cloud data obtained by multiple scans, two consecutive second point cloud data are compared point by point to find the changed data part to obtain the changed data.

[0097] The processing module 302 is used to generate a continuous motion state of a three-dimensional moving object according to the change data.

[0098] In a possible implementation, the acquisition module 301 is used to obtain a preset substation spatial structure, wherein the substation spatial structure includes a connection structure of a laser point cloud sensor based on an internal route of the substation and a connection distance between two mutually connected laser point cloud sensors, and the connection structure uses the installation position of the laser point cloud sensor as a node and the internal route of the substation as a link, and the nodes are connected through the internal route of the substation.

[0099] The processing module 302 is used to determine the continuous position change of the mobile target in the substation according to the first point cloud data obtained by multiple scans and transmitted by the first controller.

[0100] The processing module 302 is used to determine the moving direction of the moving target according to the continuous position change.

[0101] The acquisition module 301 is used to acquire the historical probabilities of different moving objects moving from the first scanning area to each scanning area in the historical records.

[0102] The processing module 302 is used to perform weighted product fusion on the spatial structure of the substation, the moving direction and the historical probability, and calculate the probability of the moving target moving to the scanning area of ​​each sensor.

[0103] In a possible implementation, the processing module 302 is used to perform weighted product fusion on the spatial structure of the substation, the moving direction, and the historical probability, and calculate the probability of the moving target moving to the second scanning area, which is specifically calculated by the following formula:

[0104] Where P is the probability that the mobile target moves to the second scanning area, σ is the Sigmoid function, f1=f(θ), f2=f(d), f3=f(h), w i is the weight coefficient. When i=1, w1 is the direction weight coefficient. When i=2, w2 is the distance weight coefficient. When i=3, w3 is the history weight coefficient. α is the direction sensitivity. θ is the angle between the moving direction and the moving direction. The moving direction is the direction from the center of the first scanning area to the center of the second scanning area. θ0 is the expected angle based on the first scanning area. d 12 is the distance between the first scanning area and the second scanning area, τ is a small constant, γ is the distance sensitivity coefficient, λ is the time decay factor, T is the total number of historical records, P t (1→2) is the transition probability from the first scanning area to the second scanning area recorded at the tth time.

[0105] In one possible implementation, the sending module 303 is used to send background point cloud data to the second controller so that the second controller compares the background point cloud data with the real-time point cloud data, and uses a background difference algorithm to extract the point cloud data corresponding to the moving target in the real-time point cloud data, thereby obtaining second point cloud data.

[0106] In a possible implementation, the processing module 302 is configured to obtain a preliminary trajectory of the moving target by extracting the centroid position of each change data.

[0107] The processing module 302 is used to accumulate the change data obtained from multiple scans into a global coordinate system to form an accumulated point cloud.

[0108] The processing module 302 is used to perform surface reconstruction based on the accumulated point cloud to obtain a target three-dimensional model.

[0109] The processing module 302 is used to form a dynamic three-dimensional dynamic model according to the time series.

[0110] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0111] This embodiment also discloses an electronic device, referring to Figure 4 The electronic device may include: at least one processor 401 , at least one communication bus 402 , a user interface 403 , a network interface 404 , and at least one memory 405 .

[0112] The communication bus 402 is used to realize the connection and communication between these components.

[0113] The user interface 403 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0114] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0115] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 401 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented separately through a chip.

[0116] Among them, the memory 405 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 405 may also be at least one storage device located away from the aforementioned processor 401. The memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface 403 module, and an application program for a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model.

[0117] exist Figure 4In the electronic device shown, the user interface 403 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 401 can be used to call an application program stored in the memory 405 for a method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model. When executed by one or more processors 401, the electronic device executes one or more methods in the above-mentioned embodiments.

[0118] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0119] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0123] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory 405, including a number of instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory 405 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0124] The present application also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores instructions. When executed by one or more processors 401, the electronic device executes one or more of the methods described in the above embodiments.

[0125] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model, characterized in that: The method is applied to a server of a substation management and control system, wherein the substation management and control system comprises the server, multiple controllers and multiple laser point cloud sensors, wherein the server is connected to multiple controllers at the same time, and each of the controllers is connected to one laser point cloud sensor, and the method comprises: Acquire first position data transmitted by a first controller, wherein the first controller sends the first position data of the first scanning area to the server if it is determined that there is a moving target in the first scanning area of ​​the first sensor based on the acquired first point cloud data collected by the first sensor, the first controller is any one of the plurality of controllers, and the first sensor is a laser point cloud sensor connected to the first controller among the plurality of laser point cloud sensors; According to the preset substation spatial structure, the probability of the mobile target moving to the scanning area of ​​each sensor is calculated; If it is determined that the probability that the mobile target moves to the second scanning area is greater than or equal to a preset threshold, a collection instruction is sent to a second controller, wherein the second controller is any one of the multiple controllers except the first controller, the scanning area of ​​the second sensor is the second scanning area, and the second sensor is a laser point cloud sensor connected to the second controller among the multiple laser point cloud sensors; receiving the change data transmitted by the second controller, wherein after receiving the acquisition instruction, the second sensor controls the second sensor to perform continuous scanning, and compares two consecutive second point cloud data point by point to find the changed data part according to the second point cloud data obtained by multiple scans, so as to obtain the change data; A three-dimensional continuous motion state of the moving object is generated according to the change data.

2. A method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model according to claim 1, characterized in that: The calculating, according to the preset substation spatial structure, the probability that the mobile target moves to the scanning area of ​​each sensor specifically includes: Acquire a preset substation spatial structure, wherein the substation spatial structure includes a connection structure of laser point cloud sensors based on an internal route of the substation and a connection distance between two mutually connected laser point cloud sensors, the connection structure uses the installation positions of the laser point cloud sensors as nodes, uses the internal route of the substation as links, and nodes are connected through the internal route of the substation; Determining continuous position changes of the mobile target in the substation according to the received first point cloud data obtained by multiple scans transmitted by the first controller; Determining a moving direction of the moving target according to the continuous position change; Acquire historical probabilities of different moving objects moving from the first scanning area to each of the scanning areas in historical records; The substation spatial structure, the moving direction and the historical probability are weightedly fused to calculate the probability of the moving target moving to the scanning area of ​​each sensor.

3. A method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model according to claim 2, characterized in that: Before determining that the probability that the moving target moves to the second scanning area is greater than or equal to a preset threshold, the method further includes: The substation spatial structure, the moving direction and the historical probability are weighted and fused to calculate the probability of the moving target moving to the second scanning area, which is specifically calculated by the following formula: ； Wherein, P is the probability that the mobile target moves to the second scanning area, σ is the Sigmoid function, f1=f(θ), f2=f(d), f3=f(h), w i is the weight coefficient, when i=1, w1 is the direction weight coefficient, when i=2, w2 is the distance weight coefficient, when i=3, w3 is the history weight coefficient, α is the direction sensitivity, θ is the angle between the moving direction and the moving direction, the moving direction is the direction from the center of the first scanning area to the center of the second scanning area, θ0 is the expected angle based on the first scanning area, d 12 is the distance between the first scanning area and the second scanning area, τ is a small constant, γ is the distance sensitivity coefficient, λ is the time decay factor, T is the total number of historical records, P t (1→2) is the transition probability from the first scanning area to the second scanning area recorded at the tth time.

4. The method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model according to claim 1, characterized in that: Before comparing two consecutive second point cloud data point by point to find a changed data portion based on the second point cloud data obtained by multiple scans and obtaining the changed data, the method further includes: The background point cloud data is sent to the second controller so that the second controller compares the background point cloud data with the real-time point cloud data, and uses a background difference algorithm to extract the point cloud data corresponding to the mobile target in the real-time point cloud data, thereby obtaining the second point cloud data.

5. The method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model according to claim 1, characterized in that: The step of comparing two consecutive second point cloud data point by point to find a changed data portion based on the second point cloud data obtained by multiple scans to obtain the changed data specifically includes: The second controller initially aligns the first sub-data with the second sub-data, where the first sub-data and the second sub-data are any two adjacent second point cloud data among the plurality of second point cloud data; The second controller queries multiple groups of first data points and second data points in the first sub-data and the second sub-data according to the initial registration result, wherein the first data point is any point in the first sub-data, and the second data point is a point in the second sub-data that is closest to the first data point; The second controller calculates a rigid body transformation between the first data point and the second data point so that the first sub-data and the second sub-data coincide with each other; The second controller continuously iterates and updates the calculation until the rigid body transformation error converges below the threshold, completing the precise point cloud registration; After the second controller completes the precise point cloud registration, the moved or newly added data points can be found by comparing the spatial position changes of each data point in the first sub-data and the second sub-data to obtain the change data.

6. A method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model according to claim 5, characterized in that: The second controller calculates the rigid body transformation between the first data point and the second data point so that the first sub-data and the second sub-data overlap, specifically including: The second controller calculates the first sub-data {p1, p 2, …,p n }, calculate the first centroid of the second sub-data {q1,q2,…,q n The second centroid of} is calculated by the following formula: ； Among them, c p is the first centroid, c q is the second centroid, {p1,p 2, …,p n } is the first sub-data, {q1,q2,…,q n } is the second sub-data; The second controller calculates the covariance matrix, specifically by the following formula: ； Where H is the covariance matrix, c p is the first centroid, c q is the second centroid, {p1,p 2, …,p n } is the first sub-data, {q1,q2,…,q n } is the second sub-data; The second controller constructs a rigid body transformation matrix based on the covariance matrix, as follows: ； Wherein, T is the rigid body transformation matrix, H is the covariance matrix, E is the diagonal matrix, c p is the first centroid, c q is the second centroid; The second controller applies the rigid body transformation matrix to the first sub-data, updates the first data point, and obtains an updated first data point, as follows: ； Among them, p i ' is the first data point after update, H is the covariance matrix, E is the diagonal matrix, p i is the first data point.

7. The method for tracking the trajectory of a moving object in a substation based on a three-dimensional digital model according to claim 1, characterized in that: Generating the three-dimensional continuous motion state of the moving target according to the change data specifically includes: By extracting the centroid position of each change data, a preliminary trajectory of the moving target is obtained; Accumulate the change data obtained from multiple scans into a global coordinate system to form a cumulative point cloud; Based on the accumulated point cloud, surface reconstruction is performed to obtain the target three-dimensional model; According to the time series, a dynamic three-dimensional dynamic model is formed.

8. A moving object trajectory tracking device in a substation based on a three-dimensional digital model, characterized in that: The device is a server, and comprises an acquisition module (301), a processing module (302), and a sending module (303), wherein: The acquisition module (301) is used to acquire first position data transmitted by a first controller, wherein the first controller sends the first position data of the first scanning area to the server if it is determined that there is a moving target in the first scanning area of ​​the first sensor based on the acquired first point cloud data collected by the first sensor, the first controller is any one of a plurality of controllers, and the first sensor is a laser point cloud sensor connected to the first controller among a plurality of laser point cloud sensors; The processing module (302) is used to calculate the probability of the mobile target moving to the scanning area of ​​each sensor according to the preset substation spatial structure; The sending module (303) is used to send a collection instruction to a second controller if it is determined that the probability of the mobile target moving to the second scanning area is greater than or equal to a preset threshold, wherein the second controller is any one of the plurality of controllers except the first controller, the scanning area of ​​the second sensor is the second scanning area, and the second sensor is a laser point cloud sensor connected to the second controller among the plurality of laser point cloud sensors; The acquisition module (301) is used to receive the change data transmitted by the second controller, wherein after the second sensor receives the acquisition instruction, the second sensor is controlled to perform continuous scanning, and based on the second point cloud data obtained by multiple scans, two consecutive second point cloud data are compared point by point to find the changed data part, so as to obtain the change data; The processing module (302) is used to generate a three-dimensional continuous motion state of the moving target according to the change data.

9. An electronic device, characterized in that: The electronic device comprises a processor (401), a communication bus (402), a memory (405), a user interface (403), a network interface (404) and a memory (405), wherein the memory (405) is used to store instructions, the user interface (403) and the network interface (404) are both used to communicate with other devices, and the processor (401) is used to execute the instructions stored in the memory (405) so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images