An environmentally-sensing-based power data acquisition system and method

Abstract

This invention discloses an environmentally-sensing-based power data acquisition system and method, relating to the field of power monitoring technology. The method includes: converting raw data into physical field feature vectors including heat flux concentration and eddy current intensity; constructing a dynamic heterogeneous graph network based on the physical field feature vectors; utilizing the topological relationship of the dynamic heterogeneous graph network to solve for continuous regions where the physical field distortion gradient value exceeds a gradient threshold; activating sensors within the continuous regions to form a minimum sensing subgraph; associating target power acquisition points based on the distortion coordinates of the minimum sensing subgraph using the conductor's current-carrying capacity; generating power feature dictionary atoms based on Joule's law; generating environmental feature dictionary atoms based on mechanical vibration equations; performing joint sparse coding using the power feature dictionary atoms and the environmental feature dictionary atoms; and separating and reconstructing power data from the result of the joint sparse coding.

Description

Technical Field

[0001] This application relates to the field of power monitoring technology, and more specifically, to a power data acquisition system and method based on environmental perception. Background Technology

[0002] In modern power system operation and energy management, the accurate and efficient collection of electrical energy data is of fundamental significance. It is a key support for realizing real-time monitoring of power grid status, load forecasting and analysis, energy efficiency optimization management, and fault early warning and diagnosis. With the development of smart grid and Internet of Things technologies, obtaining higher-quality electrical energy data is becoming increasingly important for ensuring the safe and stable operation of the power grid and improving energy utilization efficiency.

[0003] However, traditional power data acquisition methods still have room for improvement when facing complex environments and massive monitoring points. Existing methods generally adopt continuous, full-domain or fixed-mode data acquisition strategies, which are inefficient in resource-constrained scenarios (such as communication bandwidth and sensor power consumption), and it is difficult to balance the real-time nature of monitoring with the economic efficiency of the system. How to more intelligently coordinate sensing resources and effectively integrate environmental information to improve the efficiency of power data acquisition is a direction worthy of attention in current technology.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides an environmental perception-based power data acquisition system and method to solve the above-mentioned technical problems.

[0006] This application provides an environmental sensing-based power data acquisition system, comprising:

[0007] The physical field feature vector conversion module is used to acquire raw data collected by various sensors within the monitoring area in real time and convert the raw data into physical field feature vectors that include heat flux concentration and eddy current intensity.

[0008] A heterogeneous graph network construction module is used to construct a dynamic heterogeneous graph network based on the physical field feature vector;

[0009] The minimum sensing subgraph forming module is used to utilize the topological relationship of the dynamic heterogeneous graph network to solve the continuous region where the physical field distortion gradient value exceeds the gradient threshold; and to activate the sensors in the continuous region to form the minimum sensing subgraph.

[0010] The target power acquisition point association module is used to associate the target power acquisition point with the conductor current carrying capacity based on the distortion coordinates of the minimum sensing sub-map.

[0011] The joint sparse coding module is used to input the power data of the target power acquisition point and the environmental data of the minimum sensing sub-map into the encoder, and perform the following: generating power feature dictionary atoms based on Joule's law; generating environmental feature dictionary atoms based on mechanical vibration equations; and performing joint sparse coding using the power feature dictionary atoms and the environmental feature atoms.

[0012] The power data reconstruction module is used to separate and reconstruct the power data from the results of joint sparse coding.

[0013] This application provides a method for acquiring electrical energy data based on environmental perception, including:

[0014] The system acquires raw data collected by various sensors within the monitoring area in real time and converts the raw data into physical field feature vectors that include heat flux concentration and eddy current intensity.

[0015] A dynamic heterogeneous graph network is constructed based on the physical field feature vectors;

[0016] Using the topological relationship of the dynamic heterogeneous graph network, the continuous region where the physical field distortion gradient value exceeds the gradient threshold is solved; the sensors in the continuous region are activated to form a minimum sensing subgraph.

[0017] Based on the distorted coordinates of the minimum sensing sub-graph, the target power acquisition point is associated with the conductor's current-carrying capacity;

[0018] The power data from the target power acquisition point and the environmental data from the minimum sensing sub-map are input into the encoder, and the following steps are performed: generating power feature dictionary atoms based on Joule's law; generating environmental feature dictionary atoms based on mechanical vibration equations; and performing joint sparse coding using the power feature dictionary atoms and the environmental feature atoms.

[0019] The electrical energy data is separated and reconstructed from the results of joint sparse coding.

[0020] Based on the embodiments provided in this application, by converting raw sensor data into feature vectors reflecting the physical field state (such as heat flux concentration and eddy current intensity), and constructing a dynamic heterogeneous graph network accordingly, it is possible to intelligently identify critical continuous regions where the physical field distortion gradient exceeds a threshold. Furthermore, only sensors within this continuous region are activated to form a minimal sensing subgraph, abandoning the fixed, continuous, and global acquisition mode of sensors in non-critical regions. This "on-demand activation, local focus" sensing strategy effectively reduces sensor energy consumption and communication bandwidth usage, significantly improving the overall system efficiency in resource-constrained scenarios and optimizing the balance between real-time monitoring and economy.

[0021] This paper innovatively employs a joint sparse coding framework to process environmental data (derived from sensors in the physical field distortion region) from a minimal sensing subgraph and associated target power acquisition point data. The core of this framework lies in: generating power feature dictionary atoms based on Joule's law to accurately characterize the inherent patterns of power data; generating environmental feature dictionary atoms based on mechanical vibration equations to effectively represent the physical nature of environmental interference; and using these two types of atoms with clear physical meaning for joint sparse coding. This process can deeply explore the structural differences and coupling relationships between power data and environmental interference in the sparse domain, thereby more effectively separating and reconstructing the target power data in the coding results. This significantly reduces the interference of complex environmental factors (such as heat, eddy currents, and vibrations) on the interpretation of power data, improving the accuracy of the acquired data. This method no longer views power data acquisition in isolation but tightly couples it with the dynamic perception of the environmental physical field state, giving power data acquisition an environment-aware driven intelligence, and providing a guarantee for obtaining purer and more reliable power data in complex interference environments. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0023] Figure 1 This is a structural diagram of an optional environmentally aware-based power data acquisition system according to an embodiment of this application;

[0024] Figure 2 This is a flowchart of an optional environmentally-aware power data acquisition method according to an embodiment of this application;

[0025] Figure 3 This is a flowchart of another optional environmentally-aware power data acquisition method according to an embodiment of this application;

[0026] Figure 4 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application.

[0027] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Optionally, such as Figure 1 As shown, this application provides an environmental sensing-based power data acquisition system, comprising:

[0030] The physical field feature vector conversion module 101 is used to acquire the raw data collected by each sensor in the monitoring area in real time and convert the raw data into physical field feature vectors including heat flow concentration degree and eddy current intensity.

[0031] Heterogeneous graph network construction module 102 is used to construct dynamic heterogeneous graph networks based on physical field feature vectors;

[0032] Minimum sensing subgraph forming module 103 is used to utilize the topological relationship of the dynamic heterogeneous graph network to solve the continuous region where the physical field distortion gradient value exceeds the gradient threshold; and to activate the sensors in the continuous region to form the minimum sensing subgraph.

[0033] The target power acquisition point association module 104 is used to associate the target power acquisition point with the conductor current carrying capacity based on the distortion coordinates of the minimum sensing sub-graph.

[0034] The joint sparse coding module 105 is used to input the power data of the target power acquisition point and the environmental data of the minimum sensing sub-map into the encoder, and perform the following: generating power feature dictionary atoms based on Joule's law; generating environmental feature dictionary atoms based on mechanical vibration equations; and performing joint sparse coding using the power feature dictionary atoms and the environmental feature atoms.

[0035] The power data reconstruction module 106 is used to separate and reconstruct power data from the results of joint sparse coding.

[0036] Optionally, such as Figure 2 As shown, this application provides a method for acquiring electrical energy data based on environmental perception, including:

[0037] S201 acquires raw data collected by various sensors within the monitoring area in real time and converts the raw data into physical field feature vectors including heat flow concentration and eddy current intensity.

[0038] The sensors include temperature sensors, vibration sensors, and magnetic field sensors.

[0039] The heat flux concentration is calculated based on temperature sensor data, using the rate of heat accumulation in a local area (e.g., the product of the rate of change of temperature difference between adjacent nodes and the thermal conductivity). The eddy current intensity is calculated based on magnetic field sensor data, using the analysis of the frequency of magnetic field changes and spatial curl (e.g., the circulation integral of the magnetic field vectors at three adjacent points).

[0040] For example, a temperature sensor at a cable joint measures a temperature rise of 2°C within 1 second (50% higher than the average), while a magnetic field sensor detects a 100Hz high-frequency vortex magnetic field. The converted eigenvector is then [heat flux concentration = 0.85, eddy current intensity = 1.2] (normalized value).

[0041] In S201, the raw sensor data is converted into physical field feature vectors such as heat flow concentration (quantifying the heat accumulation trend) and eddy current intensity (characterizing electromagnetic loss). This breaks through the limitations of traditional static parameters and provides computational primitives with clear physical meaning for environmental perception by extracting the core features of dynamic physical processes (such as temperature rise rate and magnetic field curl).

[0042] S202, a dynamic heterogeneous graph network constructed based on physical field feature vectors;

[0043] In S202, a topology is established that reflects the correlation between environmental conditions in the monitoring area (network nodes represent sensor locations, and edges represent the correlation between environmental conditions).

[0044] S203 utilizes the topological relationship of a dynamic heterogeneous graph network to solve for continuous regions where the physical field distortion gradient value exceeds the gradient threshold; and activates sensors within the continuous regions to form a minimum sensing subgraph.

[0045] The gradient threshold is a critical value used to determine whether the distortion of the physical field is significant, and it is dynamically set according to the safety margin of the conductor material.

[0046] In S203, significant environmental anomaly areas are located through topology analysis, and sensors in these areas are focused (the minimum sensing subgraph, i.e., the set of sensors in the anomaly area).

[0047] S204, based on the distorted coordinates of the minimum sensing sub-graph, associate the target power acquisition point with the conductor's current carrying capacity;

[0048] In S204, the location of an abnormal environment is mapped to the power monitoring point in the electrical system that is safely associated with it (current carrying capacity is used as the basis for association).

[0049] S205, input the power data of the target power acquisition point and the environmental data of the minimum sensing sub-map into the encoder, and execute: generate power feature dictionary atoms based on Joule's law; generate environmental feature dictionary atoms based on mechanical vibration equation; and perform joint sparse coding using power feature dictionary atoms and environmental feature atoms;

[0050] The power data of the target power acquisition point includes the current and voltage time-series waveforms of the associated acquisition point (such as instantaneous values ​​at a sampling rate of 10kHz).

[0051] The environmental data of the minimum sensing submap includes data collected by sensors (temperature sensors, vibration sensors, and magnetic field sensors) within a continuous area.

[0052] In S205, separation primitives are constructed based on physical laws (Joule's law / vibration equations) to provide mathematical tools for the separation of mixed signals.

[0053] S206, separate and reconstruct the power data from the results of joint sparse coding.

[0054] In S206, clean electrical energy data, free from environmental interference, is extracted from the mixed signal.

[0055] Furthermore, a dynamic heterogeneous graph network is constructed based on the physical field feature vectors, including:

[0056] Treat each sensor as a node and associate it with the physical field feature vector of that node's location;

[0057] Among them, the node position is the geometric coordinate of the sensor in the monitoring area, which is used to construct the spatial topology;

[0058] The spatial connection weights between the generated nodes decrease exponentially as the distance between the sensors increases, and the rate of decrease is controlled by a pre-calibrated attenuation coefficient.

[0059] The pre-calibrated attenuation coefficient is an exponential factor controlling the attenuation of spatial connection weight with distance, determined through fitting historical data. For example, in a cable tunnel, the measured radius of influence of the temperature field is approximately 5 meters, and the fitted attenuation coefficient is 0.2. Therefore, when two nodes are 3 meters apart, the spatial weight = e^(-0.2×3) ≈ 0.55.

[0060] The physical field coupling weight between the nodes is generated. When the angle between the temperature gradient vector at the first node and the magnetic field gradient vector at the second node is less than 37 degrees, a coupling connection is established between the first node and the second node. The coupling strength is proportional to the product of the magnitudes of the temperature gradient vector at the first node and the magnetic field gradient vector at the second node.

[0061] Among them, in the electromagnetic-thermal coupling effect, when the angle between the temperature gradient and the magnetic field gradient is less than 37°, the eddy current thermal effect of the ferromagnetic material is significantly enhanced, so as to avoid misconnection of uncoupled nodes (such as pseudo-associations caused by random noise).

[0062] The spatial connection weight and the physical field coupling weight are fused into the final connection weight according to the dynamic ratio coefficient, which is adjusted according to the sensor distribution density within the monitoring area.

[0063] Specifically, the allocation coefficient in densely populated sensor areas leans towards spatial connectivity, while the allocation coefficient in sparsely populated sensor areas leans towards physical field coupling. For example, in substation switchgear, where sensors are densely packed, spatial weight accounts for 90%; while in overhead lines, where sensors are sparsely populated, physical field coupling weight accounts for 70%.

[0064] The connection structure of the dynamic heterogeneous graph network is used to guide the topological traversal path in the process of solving the distortion gradient of the physical field.

[0065] Based on the embodiments provided in this application, by limiting the sensor types (temperature, vibration, magnetic field) and the specific construction rules of the dynamic heterogeneous graph network (exponential decay of spatial connection weights, establishment of physical field coupling weights based on the temperature / magnetic field gradient direction angle <37°, and dynamic matching coefficients for weight fusion), the accuracy of physical field state characterization and the dynamic adaptability of network topology are significantly improved. Spatial weight decay reflects the natural law of physical quantity decay with distance; the directional restriction of coupling weights (<37°) ensures that connections are established only between nodes with strong physical field interactions, effectively filtering noise correlations; the dynamic matching coefficients can automatically adjust the weights of spatial and physical field factors according to the density of sensors, making the constructed network more realistically reflect the actual propagation and coupling characteristics of the physical field in the monitoring area, laying a solid topological foundation for subsequent accurate identification of distorted areas.

[0066] Furthermore, utilizing the topological relationships of dynamic heterogeneous graph networks, the continuous regions where the physical field distortion gradient values ​​exceed the gradient threshold are solved; sensors within these continuous regions are activated to form a minimum sensing subgraph, including:

[0067] Traverse each node in the dynamic heterogeneous graph network as the current target node;

[0068] Determine the set of adjacent nodes that are directly connected to the current target node;

[0069] Direct connection is based on the connectivity relationships within the constructed dynamic heterogeneous graph network. In this network, each sensor represents a node, and whether nodes are directly connected is determined by the final connection weight. Specifically, if the final connection weight between two nodes is not zero (i.e., an edge exists), and there are no other nodes acting as intermediaries between them (no indirect connection through a third node is required), then they are considered directly connected.

[0070] Calculate the gradient vector of the physical field feature vector of the current target node and the physical field feature vector of each neighboring node;

[0071] Among them, the physical field eigenvectors include heat flux concentration (e.g., heat flux density per unit area, unit W / m²). 2 ) and eddy current intensity (e.g., eddy current density, unit A / m) 2The gradient vector is a vector that describes the "rate of change" and "direction of change" of these two features in space: for the current target node (such as T3) and its neighboring node (such as M2), the two components of the gradient vector are the "difference in heat flux concentration" (T3 heat flux concentration - M2 heat flux concentration) and the "difference in eddy current intensity" (T3 eddy current intensity - M2 eddy current intensity), and the direction of the vector is determined by the sign and magnitude of these two differences (reflecting the trend of change).

[0072] Take the maximum value of the magnitude of all gradient vectors as the physical field distortion gradient value of the current target node;

[0073] The magnitude of the gradient vector is the size of the vector, and its calculation formula can be compared to that of a two-dimensional vector magnitude (√(difference in heat flux concentration)). 2 +Edge Intensity Difference 2 The current target node will form gradient vectors with multiple directly connected neighboring nodes. The value with the largest magnitude among these vectors is taken as the physical field distortion gradient value of that node.

[0074] Nodes whose physical field distortion gradient values ​​exceed the gradient threshold are selected as candidate distortion nodes.

[0075] In this embodiment, the gradient threshold is a critical value for determining whether a node is a candidate distorted node. It is determined by the physical field stability state of the monitoring area (based on historical data or safety standard presets). When the physical field distortion gradient value of a node exceeds the gradient threshold, it indicates that its physical field characteristics have changed drastically and there may be distortion.

[0076] For example, in high-power equipment areas of a workshop (such as near welding equipment): the background of the physical field fluctuates greatly, and the gradient threshold is set to 6.0 (corresponding to a heat flux concentration change >30W / m). 2 Or eddy current intensity variation >20A / m 2 Workshop general lighting distribution area: physical field stable, gradient threshold set to 2.5 (corresponding to heat flux concentration change <10W / m); 2 And the eddy current intensity variation is <8A / m 2 That is, it is determined to be non-distortion.

[0077] Based on the embodiments provided in this application, a specific implementation scheme for calculating the distortion gradient value of the physical field is provided (node ​​traversal, neighborhood gradient calculation, and taking the maximum modulus value as the distortion gradient value). This method fully utilizes the topological relationships (adjacent node set) of the dynamic heterogeneous graph network, ensuring that the calculation of the distortion gradient is based on the spatial variation of local physical field characteristics, rather than isolated point values. By screening candidate distortion nodes, the method initially focuses on key locations of abnormal physical field states, providing a reliable set of target nodes for subsequent accurate definition of continuous distortion regions, and avoiding the waste of resources caused by blind calculation across the entire domain.

[0078] Furthermore, a breadth-first traversal is performed starting from the candidate distorted node along the edges whose final connection weight is greater than the weight threshold value;

[0079] The weight threshold is a critical value for filtering valid connected edges. It is used to select edges with sufficiently high connection strength from the adjacent edges of candidate distorted nodes, serving as the path basis for breadth-first traversal (traversing only along edges whose final connection weight is greater than this value). Its value is related to the sensor distribution density in the monitoring area (higher density means a higher threshold to avoid redundant paths; lower density means a lower threshold to ensure traversal continuity).

[0080] For example, in densely populated sensor areas (such as inside a power distribution cabinet, 10 sensors / m²) 2 ): The weight threshold is set to 0.3 (only edges with high connection strength are retained); Sensor sparse areas (such as cable trenches, 2 / m) 2 ): The weight threshold is set to 0.1 (to allow more edges to be retained and ensure that the traversal is not interrupted).

[0081] In some embodiments, a candidate distorted node may have multiple directly connected neighboring nodes, but not all edges are meaningful (some edges have too low connection weights, which may be noise or weak correlations). Therefore, only edges with a final connection weight greater than a weight threshold are traversed to ensure that the traversal path focuses on nodes with close physical field correlations.

[0082] During the traversal, record the direction of change of the physical field feature vectors of three consecutive visited nodes;

[0083] Among them, the three visited nodes refer to the three nodes that are visited consecutively during the breadth-first traversal.

[0084] Calculate the direction change angle. If the direction change angle of three consecutive nodes is less than 15 degrees, then it is marked as a valid propagation path.

[0085] The direction change angle refers to the angle between the gradient vector directions of the first two nodes and the gradient vector directions of the last two nodes in a series of three consecutive nodes. For example, if the gradient vector direction from node A to B is θ1 and the gradient vector direction from B to C is θ2, then the direction change angle is |θ1-θ2|.

[0086] If the directional change angle of three consecutive nodes is less than 15 degrees, it indicates that the direction of change of the physical field distortion is stable (without drastic changes), proving that the distortion on this path is "continuous and consistent" (caused by the same reason, such as overheating of a conductor), rather than scattered random noise, and is therefore marked as an effective propagation path.

[0087] Aggregate all nodes covered by the effective propagation paths to form a continuous region;

[0088] Verify whether the number of target node pairs satisfying the physical field coupling condition within a continuous region exceeds a set threshold; wherein, the target node pairs include temperature sensor nodes and magnetic field sensor nodes;

[0089] Among them, the target node pair refers to the node pair that meets the calculation conditions of physical field coupling weight, that is, the pair consisting of temperature sensor node and magnetic field sensor node, and the angle between the directions of their temperature gradient vector and magnetic field gradient vector is less than 37 degrees.

[0090] Establishing the physical field coupling relationship between nodes by using this condition and verifying the number of such node pairs is to ensure that the distortion in the continuous region is indeed caused by the electromagnetic-thermal coupling effect (rather than the random fluctuation of a single physical field), thereby enhancing the reliability of region determination.

[0091] In one specific implementation, the effective propagation path strength is calculated based on the following formula:

[0092]

[0093] Among them, S path The effective propagation path strength (dimensionless, 0 to 1), with larger values ​​indicating higher path reliability; θ1 and θ2 are the directional change angles (unit: °) of three consecutive nodes, reflecting the continuity of the physical field eigenvector changes (e.g., θ1 is the gradient direction angle from node A to B, and θ2 is the gradient direction angle from node B to C); w1 and w2 are the final connection weights (0 to 1) of adjacent nodes on the path, calculated by the dynamic heterogeneous graph network, reflecting the physical field coupling strength between nodes (e.g., w1 is the final connection weight from node A to B, and w2 is the final connection weight from node B to C); k dens The sensor distribution density correction factor (dimensionless, 0.7 to 1.3) is set to 1.2 for densely populated areas (>5 sensors / m²) and 0.8 for sparsely populated areas (<2 sensors / m²), adapting to different deployment scenarios; d avg Let d be the average spacing between nodes on the path (in meters). For example, if the spacing between nodes AB is 0.3m and the spacing between nodes BC is 0.5m, then d avg =0.4m; L0 is the reference distance (unit: m), which is the average installation spacing of sensors in the monitoring area (e.g., 0.5m in the workshop power distribution area) to determine the attenuation effect of distance on path reliability.

[0094] It should be noted that θ1 and θ2 are the directional changes of three consecutive nodes. Since an effective propagation path requires that the directional changes of all three consecutive nodes are less than 15 degrees (i.e., θ1 < 15° and θ2 < 15°), therefore...

[0095] It should also be noted that w1 and w2 are the final connection weights of adjacent nodes on the effective propagation path (reflecting the physical field coupling strength between nodes, with values ​​ranging from 0 to 1). The meaning is to take the average of the connection weights of two adjacent nodes, which is used to characterize the average coupling strength of the path segment. This is to take the average of the coupling strength of two consecutive adjacent edges on the path, so as to comprehensively reflect the overall correlation of the path segment.

[0096] Since w1 and w2 are both normalized weights between 0 and 1, their average value Within the range of 0 to 1, the average coupling strength of the path can be reasonably quantified. If the number of target node pairs satisfying the physical field coupling conditions in a continuous region exceeds a set threshold, then all sensors in the continuous region are activated to form a minimum sensing subgraph.

[0097] The set quantity threshold refers to the minimum number of target node pairs that need to meet the physical field coupling conditions within a continuous region. This is used to verify the effectiveness of the region (to ensure that there is sufficient electromagnetic-thermal coupling relationship within the region, proving that the distortion is systematic).

[0098] For example, in a densely populated power distribution cabinet area (containing 10 temperature nodes and 8 magnetic field nodes): the number threshold is set to 8 pairs (at least 8 pairs of coupled nodes are required to ensure strong regional correlation); in a sparsely populated cable trench area (containing 3 temperature nodes and 2 magnetic field nodes): the number threshold is set to 2 pairs (to adapt to scenarios with fewer nodes and avoid misjudgment).

[0099] Based on the embodiments provided in this application, the specific logic for continuous region identification and minimum perceptual subgraph activation is defined (breadth-first traversal, effective propagation path determination, node aggregation, and physical field coupling verification). Breadth-first traversal, based on a weight threshold, ensures that exploration occurs on paths with strong physical field correlation. The criterion for determining an effective propagation path (angle change of three consecutive points <15°) guarantees that the identified path has consistent directional propagation of physical field distortion, effectively eliminating pseudo-continuous regions formed by random fluctuations or isolated noise points. After aggregating nodes to form a continuous region, the number of temperature-magnetic field sensor node pairs satisfying the coupling condition within the region is further verified. This confirms from a physical mechanism perspective that the distortion in this region is a genuine anomaly dominated by electromagnetic-thermal effects, rather than interference from a single factor, thereby ensuring that the activated minimum perceptual subgraph has high target specificity and physical correlation credibility.

[0100] Furthermore, based on the distortion coordinates of the minimum sensing subgraph, the target power acquisition point is associated with the conductor's current-carrying capacity, including:

[0101] A set of power acquisition points for locating and monitoring the electrical connections of conductors in the area;

[0102] Extract temperature data from all temperature sensors within the smallest sensing sub-map;

[0103] Calculate the deviation between the temperature data and the regional average temperature;

[0104] Temperature sensors with a deviation value greater than three standard deviations are selected to form a high-temperature cluster.

[0105] Calculate the weighted spatial centroid coordinates of the high-temperature cluster, with the weights being the temperature deviation values;

[0106] The weighted spatial centroid coordinates are output as the thermal accumulation core coordinates, and the material parameters of the conductor associated with the thermal accumulation core coordinates are obtained.

[0107] In the electrical connectivity subnet, select the multiple power collection points closest to the coordinates of the heat accumulation core; the electrical connectivity subnet is divided according to the circuit breaker tripping logic;

[0108] In some embodiments of this application, the three power collection points closest to the coordinates of the heat accumulation core can be selected;

[0109] Invoke the conductor's temperature coefficient of resistance and safe current carrying limit;

[0110] For each of the multiple power acquisition points, obtain the real-time current value of that point; calculate the dynamic safe current carrying limit of that point based on the conductor resistance temperature coefficient, the safe current carrying limit, and the real-time temperature of the conductor at that point; calculate the real-time current carrying margin of that point based on the real-time current value and the dynamic safe current carrying limit.

[0111] The sampling point with the smallest real-time current carrying capacity is selected as the correlation target.

[0112] In one embodiment, taking substation cable joint monitoring as an example: the minimum sensing sub-map detects temperature data at the joint [85℃, 92℃, 78℃, 45℃], with an average regional temperature of 65℃ (standard deviation 15℃). High-temperature sensors (85℃, 92℃) with deviations > 45℃ are selected, and the coordinates of the heat accumulation core are calculated with the deviation value as the weight, located 2cm east of the joint, and associated with the copper core cable (α = 0.00393). Within the electrical subnet defined by the circuit breaker, the three nearest points are selected: point A (0.5m), point B (1.2m), and point C (2.0m). The dynamic safe current carrying limit is calculated: when the temperature at point A is 92℃, the limit is corrected from 900A to 0; when the temperature at point B is 85℃, the limit is 760A; and when the temperature at point C is 78℃, the limit is 820A. Combined with the real-time current (810A at point A, 750A at point B, and 700A at point C), the current carrying margins are calculated to be -810A, 10A, and 120A, respectively. Point A, with the minimum margin (a negative value indicates severe overload), is selected as the associated target, triggering in-depth data acquisition and alarm for the electrical energy at that point. This embodiment verifies the ability of the method in weight 6 to accurately locate the most dangerous node in a high-temperature overload scenario.

[0113] Based on the embodiments provided in this application, a precise correlation is achieved through a collaborative mechanism of thermal accumulation core location and dynamic current carrying capacity assessment. First, based on temperature sensor data within the minimum sensing sub-map, temperature sensors with deviation values ​​greater than three standard deviations are selected to form a high-temperature cluster, eliminating interference from insignificant temperature rises. The spatial centroid coordinates are then calculated using a weighted average of temperature deviation values ​​to pinpoint the thermal accumulation core area, ensuring accurate physical location. Subsequently, the three power collection points closest to the thermal accumulation core are selected from the electrical connectivity sub-network (divided according to circuit breaker tripping logic), balancing spatial proximity and electrical correlation.

[0114] For each candidate point, a dynamic safe current-carrying limit is calculated: the conductor resistance temperature coefficient and the safe current-carrying limit are called, and combined with the real-time temperature of the conductor at that point, a current-carrying capacity threshold suitable for actual operating conditions is obtained; then, the current-carrying margin (the difference between the dynamic limit and the actual current) is calculated using the real-time current value, and finally, the point with the smallest margin is selected as the associated target. Environmental anomalies (overheating core) are directly mapped to the most vulnerable nodes in electrical safety (such as overload points where the current-carrying margin approaches zero or is negative), avoiding misjudgments caused by the separate analysis of temperature and current data in traditional methods, and optimizing computational efficiency by limiting the number of candidate points.

[0115] Furthermore, in the steps of invoking the conductor resistance temperature coefficient and safe current carrying limit, mechanical deformation compensation and electromagnetic skin effect analysis are integrated, such as... Figure 3 As shown, it specifically includes:

[0116] S301, based on the conductor mechanical vibration spectrum collected by the vibration sensor in the minimum sensing subgraph, identifies the characteristic harmonic components formed by the combined action of electromagnetic force and mechanical stress;

[0117] In some embodiments, the time-domain signal of the mechanical vibration of the conductor is acquired from the vibration sensor of the minimum sensing subgraph and converted into a spectrum (horizontal axis is frequency, vertical axis is amplitude) by fast Fourier transform. Harmonics generated by electromagnetic force are usually related to the current frequency (e.g., the second harmonic of the 50Hz fundamental wave at 100Hz, due to the periodic electromagnetic repulsion generated by alternating current), while harmonics generated by mechanical stress are related to the thermal expansion and contraction of the conductor and structural resonance (e.g., the 30Hz natural frequency harmonic caused by temperature changes). By comparing the reference spectra of pure electromagnetic force (without temperature stress) and pure mechanical stress (without current), frequency components with significantly enhanced amplitude under the combined action of the two forces (e.g., the superimposed harmonics of 100Hz and 30Hz) are selected as characteristic harmonic components.

[0118] In S301, by identifying the characteristic harmonics that interact, the key signals of electromagnetic-thermal-mechanical multi-physics coupling are accurately captured. These harmonics are the "comprehensive fingerprints" of the interaction between the conductor's current-carrying state (electromagnetic force) and environmental stress (mechanical stress), providing multi-dimensional input for subsequent microscopic deformation inversion and avoiding the one-sidedness of single-physics field analysis.

[0119] S302, based on the amplitude-frequency characteristics and phase difference of the characteristic harmonic components, the microscopic deformation distribution pattern of the conductor near the coordinates of the thermal accumulation core is inverted.

[0120] In some embodiments, amplitude-frequency characteristics refer to the variation of the amplitude of a characteristic harmonic with frequency (e.g., 0.05 mm amplitude for a 100 Hz harmonic and 0.03 mm amplitude for a 30 Hz harmonic), and phase difference refers to the time delay of different characteristic harmonics (e.g., the 100 Hz harmonic leads the 30 Hz harmonic by 10 ms). By establishing a "characteristic harmonic-deformation" mapping model (which can be trained based on finite element simulation data), amplitude-frequency characteristics are transformed into deformation amplitude (the larger the amplitude, the more severe the deformation), and phase difference is transformed into deformation direction (phase lead corresponds to stretching in a certain direction, and lag corresponds to compression). Finally, the microscopic deformation distribution of the conductor around the coordinates of the heat accumulation core is inverted (e.g., local stretching of 0.02 mm due to electromagnetic force and compression of 0.01 mm due to thermal stress).

[0121] In the S302, deformation is not directly measured using contact strain sensors (which are susceptible to environmental interference and have installation limitations). Instead, non-invasive microscopic deformation monitoring is achieved through a two-dimensional inversion of the vibration spectrum, involving both amplitude and phase. This inverse logic from signal to physical state reduces hardware costs and captures minute deformations within conductors that are difficult to measure directly (such as micrometer-level bending), providing accurate deformation data for subsequent coordinate compensation and parameter correction.

[0122] S303 utilizes a microscopic deformation distribution model to perform three-dimensional spatial offset compensation on the coordinates of the heat accumulation core, generating deformation-compensated heat source coordinates, and simultaneously correcting the rate of change of the equivalent cross-sectional area of ​​the conductor at the deformation-compensated heat source coordinates.

[0123] In some embodiments, the microscopic deformation distribution pattern shows that the coordinates of the heat accumulation core (originally calculated as X = 10m, Y = 2m, Z = 0.5m) shift due to deformation (e.g., stretching 0.05m along the X-axis and bending 0.02m along the Z-axis). After three-dimensional spatial offset compensation, the deformation-compensated heat source coordinates are corrected to (10.05m, 2m, 0.52m). Simultaneously, based on the degree of stretching / compression of the conductor in the deformation pattern (e.g., local elongation leading to a smaller cross-sectional area), the equivalent cross-sectional area change rate (e.g., original cross-sectional area 100mm²) is calculated. 2 It has now become 95mm 2 (Change rate -5%).

[0124] In S303, the coordinates of the thermal accumulation core are a key reference point for subsequent current-carrying capacity analysis. However, conductor deformation can cause deviations between its actual and calculated positions, and the equivalent cross-sectional area will also change due to deformation (affecting current-carrying capacity). This step eliminates the interference of deformation on physical parameters through coordinate compensation and cross-sectional area correction, ensuring that subsequent resistivity calculations and current-carrying capacity analyses based on these coordinates are grounded in the actual physical state and avoiding virtual errors caused by deformation.

[0125] S304: Extract real-time temperature data from the temperature sensor at the coordinates of the deformation compensation heat source, and calculate the dynamic resistivity under the influence of deformation by combining the resistance-temperature characteristics of the conductor material.

[0126] In some embodiments, the real-time temperature (e.g., 85°C) is obtained from a temperature sensor (e.g., T5) corresponding to the coordinates of the deformation-compensated heat source (e.g., 10.05m), and the resistance-temperature characteristic formula of the conductor material (e.g., copper) is called. Simultaneously, the equivalent cross-sectional area change rate (e.g., -5%) obtained in the third step is introduced, because the change in cross-sectional area affects the current density, which in turn indirectly affects the resistance (under the same current, a smaller cross-sectional area results in a larger current density, and the resistance increases accordingly). Finally, the dynamic resistivity including the deformation effect is calculated (e.g., 3% higher than the resistivity considering only temperature).

[0127] In S304, dynamic resistivity is calculated by coupling "temperature + deformation" as two parameters, making the resistance parameters more consistent with the actual state of the conductor. This provides high-precision basic data for subsequent skin effect analysis and current carrying capacity assessment, and is a key link in realizing the closed loop of "environmental perception - accurate acquisition of power data".

[0128] Based on the embodiments provided in this application, the preliminary processing of mechanical deformation compensation and electromagnetic skin effect analysis (identifying characteristic harmonics, inverting deformation modes, spatial offset compensation, equivalent cross-sectional area correction, and dynamic resistivity calculation) is integrated into the conductor parameter calling step. By analyzing the characteristic harmonic components in the vibration spectrum (the combined effect of electromagnetic force and mechanical stress), the microscopic deformation distribution pattern of the conductor near the heat accumulation core is inverted. Based on this, three-dimensional spatial offset compensation is performed on the heat source coordinates (generating deformation-compensated heat source coordinates), and the rate of change of the equivalent cross-sectional area of ​​the conductor at that location is simultaneously corrected, significantly improving the accuracy of heat source positioning and the realism of the conductor's geometric state characterization. Combining temperature data to calculate the dynamic resistivity under the influence of deformation provides more accurate conductor state parameters for subsequent precise analysis of the skin effect and current carrying capacity, reducing the interference of mechanical deformation on electrical parameter evaluation.

[0129] Furthermore, the method also includes:

[0130] Obtain the current harmonic spectrum monitored at the target power acquisition point, and select the dominant harmonic frequency that is associated with the frequency of the vibration characteristic harmonic component.

[0131] In some embodiments, the current change data over a period of time is first collected by a current sensor at the target power acquisition point (e.g., 1000 current values ​​are recorded per second). Then, the current harmonic spectrum is obtained by signal processing methods (e.g., converting the change of current over time into current components of different frequencies, similar to decomposing sound into different tones). This spectrum contains current components of multiple frequencies (e.g., the fundamental wave of 50Hz, the second harmonic of 100Hz, the third harmonic of 150Hz, etc.).

[0132] At the same time, from the identified vibration characteristic harmonic components (i.e. vibration frequencies caused by electromagnetic force and mechanical stress, such as 100Hz and 200Hz), current harmonics that are the same as or multiples of the frequencies in the current harmonic spectrum are identified (for example, if the vibration characteristic harmonic is 100Hz, the 100Hz harmonic in the current spectrum is selected; if the vibration characteristic harmonic is 200Hz, the 200Hz harmonic in the current spectrum is selected). These selected current harmonics are the dominant harmonic frequencies.

[0133] There is an inherent relationship between the harmonic frequencies of current and the characteristic harmonic frequencies of vibration. For example, the electromagnetic force generated by alternating current changes with the current frequency, which in turn induces conductor vibration, and the frequencies of the two are often synchronized. By selecting the associated dominant harmonic frequency, the core frequency of the "current-vibration" coupling can be accurately located, providing a clear target for subsequent analysis of the special effects of high-frequency current on conductors (such as the skin effect) and avoiding deviations caused by analyzing irrelevant frequencies.

[0134] The dynamic resistivity, the corrected rate of change of equivalent cross-sectional area, the dominant harmonic frequency, and the conductor permeability are input into the skin effect depth calculation model to solve the equivalent skin depth of the conductor under mechanical deformation and high-frequency current coupling.

[0135] In some embodiments, the skin effect depth calculation model can calculate the surface thickness (i.e., skin depth) through which high-frequency current concentrates in a conductor. Specifically, the higher the dynamic resistivity and the higher the dominant harmonic frequency, the smaller the skin depth (the more concentrated the current is on the surface); the higher the conductor's permeability, the smaller the skin depth; and the rate of change of the equivalent cross-sectional area affects the actual conductive space of the conductor (for example, a smaller cross-section makes the effect of current concentration more pronounced). By combining these factors, the final result is the "equivalent skin depth".

[0136] Traditional calculations of the skin effect only consider the current frequency and resistance at room temperature, neglecting the impact of potential mechanical deformations (such as bending and stretching) on ​​conductivity. This step incorporates both mechanical deformation (through the equivalent cross-sectional area change rate) and dynamic resistance (including temperature effects). The calculated skin depth more accurately reflects the current distribution under actual operating conditions (both high-frequency current and deformation), providing crucial information for determining the conductor's current carrying capacity. Based on the equivalent skin depth and the corrected equivalent cross-sectional area change rate, the conductor's effective current-carrying capacity under harmonic conditions is calculated.

[0137] In one specific implementation, the expression for the skin effect depth calculation model includes:

[0138]

[0139] Where, δ eq The equivalent skin depth (unit: m) reflects the surface thickness through which high-frequency current concentrates in a conductor; ρ d The dynamic resistivity (unit: Ω·m) includes the effects of temperature and mechanical deformation (e.g., copper at 80℃ with a deformation of +5% is approximately 2.3 × 10⁻⁶). -8 Ω·m); f h The dominant harmonic frequency (unit: Hz); μ is the conductor's permeability (unit: (Ω·s) / m), an inherent property of the material (for copper, take 4π×10⁻⁶). -7 H / m, iron is taken as 2×10 -4 H / m, 1H = 1Ω·s (ohm·second); k def The deformation intensification factor (0.3 to 0.8) is used, with 0.6 taken when the deformation rate is >5%, quantifying the exacerbating effect of deformation on current concentration; ε S The equivalent cross-sectional area change rate, e.g., -0.08 indicates an 8% reduction in cross-sectional area; k temp Temperature enhancement factor (unit: °C) -1 The value ranges from 0.01 to 0.03, and the value increases by 0.005 for every 10℃ increase in the temperature difference at the heat accumulation core (e.g., 0.02 for a 40℃ temperature difference); ΔT core The temperature difference between the heat accumulation core and the environment (unit: °C) is obtained from the temperature difference between the coordinates of the heat accumulation core and the average temperature of the region.

[0140] It should be noted that f h μ is the core influencing factor of the skin effect, which is the phenomenon where high-frequency current in a conductor generates eddy currents due to electromagnetic induction, causing the current to concentrate at the conductor surface. Its essence is directly related to the current frequency and the conductor's permeability.

[0141] Specifically, f hThe higher the Hz (i.e., 1 / s) value, the stronger the electromagnetic induction, the more significant the "displacement" effect of eddy currents on the current, and the smaller the skin depth; μ (unit (Ω·s) / m) is a parameter characterizing the magnetization ability of a conductor. The larger the μ value, the stronger the induced eddy currents generated by the magnetic field in the conductor, the easier it is for the current to be confined to the surface, and the smaller the skin depth. The product of the two, f... h μ-synthesis quantifies the influence of "high-frequency electromagnetic induction intensity" on current distribution. It is a core variable in the classical skin effect theory and has a clear physical meaning.

[0142] In some embodiments, the equivalent skin depth determines the area through which the high-frequency current actually flows in the conductor (e.g., if the skin depth is small, the current flows only in a very thin layer on the surface of the conductor); the "corrected equivalent cross-sectional area change rate" reflects the actual size of the conductor's cross-section (e.g., if the cross-section is smaller, there is less space to accommodate the current).

[0143] When calculating the effective current-carrying capacity, first calculate the effective area through which the current actually flows based on the equivalent skin depth (for example, if the total cross-section of the conductor is 100 square millimeters, but the skin depth is small, the actual effective area is only 80 square millimeters). Then, combine this with the corrected equivalent cross-sectional area (for example, due to deformation, the total cross-section becomes 90 square millimeters, and the effective area is adjusted accordingly to 72 square millimeters). Finally, based on the maximum current density that the conductor material can withstand (for example, 5 amperes of current can flow per square millimeter), calculate the maximum current that the conductor can carry under the current harmonic conditions. This is the effective current-carrying capacity.

[0144] A conductor's current-carrying capacity depends not only on its cross-sectional size but also on the characteristic of high-frequency current "concentrating on the surface"—even with a sufficiently large cross-section, if the high-frequency current flows only on the surface, the surface may overheat due to excessive current density. This step, by combining the skin depth and the actual cross-sectional area, calculates the effective current-carrying capacity to accurately reflect the conductor's actual current-carrying limit under high-frequency harmonics, avoiding misjudgments caused by relying solely on the total cross-sectional size (e.g., a large total cross-section, but the surface is already overloaded).

[0145] The safe current carrying limit is dynamically updated based on the effective current carrying capacity and the conductor's allowable temperature rise threshold.

[0146] In some embodiments, the conductor allowable temperature rise threshold refers to the maximum temperature that the conductor can raise from the ambient temperature during normal operation (for example, if the ambient temperature is 30°C and the conductor's maximum allowable temperature is 70°C, then the allowable temperature rise threshold is 40°C).

[0147] During calculation, first, based on the obtained effective current-carrying capacity, estimate the heat generated by the conductor under this current-carrying condition (the larger the current, the more heat, and the more significant the temperature rise). Then, check whether the temperature rise corresponding to this heat exceeds the allowable temperature rise threshold: if it does not exceed it, it means that the current effective current-carrying capacity is safe and can be used as the "safe current-carrying limit"; if it exceeds it, it means that operating at this current-carrying capacity will cause overheating, and the safe current-carrying limit needs to be reduced (for example, from 400 amperes to 350 amperes) until the corresponding temperature rise is within the allowable range. In this way, the safe current-carrying limit is dynamically adjusted according to the actual operating conditions.

[0148] Traditional safe current-carrying limits are fixed values ​​(such as the "rated current-carrying capacity" indicated in the manual). However, in reality, factors such as conductor heat dissipation, the presence of high-frequency harmonics, and deformation can all affect the current it can withstand. This step makes the safe current-carrying limit more flexible, dynamically adjusting it based on the actual temperature rise the conductor can withstand. This avoids misjudgments caused by fixed limits (such as being limited despite the conductor actually being able to carry a larger current), prevents overload and overheating, and better aligns with actual engineering needs.

[0149] Recalculate the real-time current margin using the updated safe current carrying limits;

[0150] In some embodiments, the real-time current carrying margin indicates how much additional current a device can currently carry. It is calculated by subtracting the actual current monitored at the target power acquisition point (e.g., 300 amps) from the updated safe current carrying limit (e.g., 350 amps). The difference (50 amps) is the real-time current carrying margin. If the actual current is 360 amps, the margin is -10 amps, indicating that the safe limit has been exceeded and there is an overload risk.

[0151] Real-time current-carrying margin directly reflects the conductor's current "remaining carrying capacity." Calculations using updated safe current-carrying limits ensure that this "remaining capacity" is a true value based on current operating conditions (including high-frequency harmonics, deformation, temperature, etc.), rather than a theoretical value based on fixed limits. For example, if the conductor's safe current-carrying limit decreases due to high-frequency harmonics, the margin may be negative even if the actual current does not exceed the original fixed limit, thus providing timely warnings of overload risk.

[0152] Based on the recalculated real-time current carrying capacity, the selection results of the associated target power acquisition points are verified and corrected.

[0153] In some embodiments, first check the real-time current margin of the previously selected target power acquisition point (e.g., point A): if the margin is positive and reasonable (e.g., 50 amps), it means that the current state of this point is stable and suitable as an acquisition point; if the margin is negative (e.g., -10 amps), it means that this point is overloaded and the acquired data may be affected by abnormal conditions, so it needs to be reselected.

[0154] When reselecting, within the electrical interconnection subnet (area divided according to circuit breaker tripping logic) where the current sampling point is located, find other sampling points (such as point B) that are closer to the coordinates of the heat accumulation core, and calculate their real-time current margin (such as 20 amps). If the margin of point B is reasonable, then the target sampling point is corrected to point B. If the margin of all sampling points in the subnet is negative, it indicates that there may be an overload in the entire area, and it is necessary to expand the scope and find a suitable sampling point again.

[0155] The selection of the target power acquisition point directly affects the accuracy of subsequent power data. If the original acquisition point is overloaded, its data may contain abnormal interference (such as measurement deviations caused by overheating). By verifying and correcting the acquisition point through real-time current margin, it can be ensured that the power data used for subsequent encoding and reconstruction comes from a "healthy" acquisition point, thus guaranteeing the reliability of the final data and avoiding errors in the entire analysis result due to improper selection of acquisition points.

[0156] Based on the embodiments provided in this application, the skin effect analysis and dynamic updating of the safe current-carrying limit are deepened (harmonic spectrum correlation, skin depth calculation, effective current-carrying capacity calculation, safety limit update, current margin recalculation, and target point verification and correction). The dominant harmonic frequency associated with the vibration characteristic frequency in the harmonic spectrum of the current at the power acquisition point is selected. Parameters such as dynamic resistivity and corrected cross-sectional area are input into the skin effect depth calculation model to solve for the equivalent skin depth under the coupling state of mechanical deformation and high-frequency current. Based on this, the effective current-carrying capacity of the conductor under harmonic conditions is calculated, and the safe current-carrying limit is dynamically updated according to this capacity and the allowable temperature rise threshold. The updated limit is used to recalculate the current margin and verify / correct the selection results of the associated target points, making the correlation of the target power acquisition point more accurate and reliable. This fully considers the complex influence of the combined effect of high-frequency harmonic current and conductor mechanical deformation on the actual current-carrying capacity of the conductor, improving the real-time performance and accuracy of the safety assessment.

[0157] Furthermore, electrical energy feature dictionary atoms are generated based on Joule's law; environmental feature dictionary atoms are generated based on mechanical vibration equations; and joint sparse coding is performed using the electrical energy feature dictionary atoms and the environmental feature dictionary atoms, including:

[0158] Based on the current and voltage time-series data of the target power acquisition point, the heat generated per unit time in the conductor is calculated according to Joule's law as the basic power characteristic.

[0159] By integrating the equivalent cross-sectional area change rate and dynamic resistivity of the minimum perceptual subgraph, conductor deformation parameters are compensated for the basic electrical energy characteristics, and electrical energy characteristic dictionary atoms are generated.

[0160] In some embodiments, current and voltage time-series data refers to the current and voltage records of the target power acquisition point changing over time (e.g., recording once per second to obtain a series of current and voltage values). According to Joule's law, heat is generated when current passes through a conductor, and the greater the current, the higher the voltage, and the longer the time, the more heat is generated.

[0161] To calculate the heat generated per unit time, a fixed time period (e.g., 1 second) is selected. The average current and average voltage during this period are used to calculate the total heat generated by the conductor during this time. This total heat is then divided by the time (1 second) to obtain the heat generated per second (e.g., 500 joules of heat generated per second). This heat generated per second is used as the basic electrical energy characteristic because it directly reflects how much electrical energy is converted into heat energy in the conductor.

[0162] The equivalent cross-sectional area change rate refers to the proportion of change in the cross-sectional size of a conductor due to deformation (such as bending or stretching) (e.g., the cross-section becomes 8% smaller); "dynamic resistivity" is the actual value of the conductor's resistance after taking into account temperature and deformation (e.g., it is 10% greater than at room temperature).

[0163] The heat generated by a conductor depends not only on current and voltage, but also on resistance and cross-sectional area: the greater the resistance and the smaller the cross-section, the more heat is generated when current flows through it. Therefore, when compensating for the basic electrical characteristics, these two parameters need to be considered—if the equivalent cross-sectional area decreases, it means the space through which the current flows is more congested, and the actual heat generated will be more than calculated from the basic characteristics; if the dynamic resistivity increases, it will also lead to an increase in actual heat generation. By combining these two factors, the basic electrical characteristics can be adjusted (for example, if the original basic characteristic is 500 joules / second, it can be adjusted to 550 joules / second after compensation) to obtain a more accurate electrical characteristic.

[0164] Conductor deformation (such as slight bending during installation or thermal expansion and contraction during operation) alters its conductivity, thus affecting heat generation. Traditional calculations of heat generation ignore the influence of deformation, potentially leading to inaccurate electrical energy characteristics. This step compensates for deformation parameters, ensuring that the electrical energy characteristics accurately reflect the conductor's energy state under the complex conditions of both electrical energy transmission and deformation, providing more reliable foundational data for subsequent analysis.

[0165] Extract the time-domain vibration waveform of the vibration sensor in the minimum sensing sub-graph, and calculate the mechanical energy transfer efficiency based on the mechanical vibration equation;

[0166] In some embodiments, a time-domain vibration waveform refers to a curve recorded by a vibration sensor showing how the amplitude of a conductor's vibration changes over time (e.g., the amplitude is large at one moment and small at another). The mechanical vibration equation is a formula describing the vibration law of an object, reflecting how the energy of vibration is transferred from one point to another (e.g., the greater the input force, the greater the vibration amplitude; energy loss occurs during transmission, causing the amplitude to decrease).

[0167] By combining the topological connection strength of dynamic heterogeneous graph networks on the effective propagation path, network propagation loss correction is performed on environmental mechanical vibration energy to generate environmental feature dictionary atoms.

[0168] In some embodiments, "topological connectivity strength of dynamic heterogeneous graph network" refers to the tightness of the connection between nodes (sensors) in the network (e.g., high connectivity strength indicates strong physical field correlation between the locations of two sensors, resulting in low loss during vibration energy transmission); "effective propagation path" refers to the path identified in claim 5 where physical field distortion propagates continuously (e.g., the path from sensor A to B and then to C).

[0169] When correcting environmental mechanical vibration energy, energy loss is calculated along the effective propagation path based on the topological connection strength between each node (low connection strength results in high loss; high connection strength results in low loss). For example, if the connection strength is 0.8 (20% loss) when vibration energy travels from A to B, and 0.6 (40% loss) when traveling from B to C, then the total loss from A to C is 20% + 40%. After this loss correction, the original vibration energy becomes the actual vibration energy propagated to each node. This corrected vibration energy serves as the "atom of the environmental characteristic dictionary" (i.e., the basic unit of environmental characteristics).

[0170] Vibrational energy is lost during propagation due to distance, obstacles, and other factors. The topological connectivity strength of a dynamic heterogeneous graph can precisely reflect the magnitude of this loss (the loss is small between closely connected nodes). Through this correction, the environmental feature dictionary atoms not only contain the vibrational energy itself but also incorporate information about spatial propagation, which can more realistically reflect the actual state of environmental features at different locations and avoid distortion of environmental features caused by ignoring propagation losses.

[0171] Establish a cross-mapping relationship between atoms in the electrical energy feature dictionary and atoms in the environmental feature dictionary. When the difference between the harmonic frequency in the electrical energy feature dictionary atom and the resonant frequency in the environmental feature dictionary atom is less than a set difference threshold, establish an atom association.

[0172] In some embodiments, “electrical energy feature dictionary atom” is the compensated basic unit of electrical energy feature obtained in step 2 (e.g., containing a harmonic frequency of 100 Hz); “environmental feature dictionary atom” is the loss-corrected basic unit of environmental feature obtained in step 4 (e.g., containing a resonant frequency of 102 Hz, i.e., the frequency at which the conductor vibrates most violently).

[0173] The threshold value is set as a pre-defined frequency difference (e.g., 5Hz, 10Hz). When the difference (2Hz) between the harmonic frequency of the electrical energy characteristic (100Hz) and the resonant frequency of the environmental characteristic (102Hz) is less than this threshold (5Hz), it is considered that the two atoms have an intrinsic connection (e.g., a 100Hz current harmonic triggers a 102Hz vibrational resonance), thus establishing an atomic correlation. The interaction between electrical energy and the environment is often reflected in frequency correlations (e.g., a current of a specific frequency will trigger vibrations of a specific frequency). Establishing atomic correlations through frequency differences can accurately capture this intrinsic connection, "binding" the originally independent electrical energy characteristics and environmental characteristics together, providing a structured basis for subsequent joint analysis of the two, and avoiding confusion between unrelated electrical energy and environmental characteristics during analysis.

[0174] Construct a joint sparse coding objective function, and add a sparse coefficient coupling constraint term for associated atom pairs to the objective function. The sparse coefficient coupling constraint term forces the difference in sparse coefficients of associated atom pairs to not exceed a set tolerance threshold.

[0175] In one specific implementation, the objective function of joint sparse coding is expressed as:

[0176]

[0177] Among them, J * J is the optimized minimum objective function value. * The numerical value of J directly reflects the coding quality: in a certain power distribution monitoring scenario, when J... * <0.5 (unit A) 2 +(m / s) 2 When the encoding result meets the engineering requirements (reconstruction error <5%, sparsity >80%, correlation coefficient deviation <0.1), no secondary optimization is needed; α is the sparse coefficient of electrical energy features (k×1 vector), and the elements in α represent the combination weight of each atom in D; β is the sparse coefficient of environmental features (l×1 vector), and the elements in β represent the combination weight of each atom in E; D is the electrical energy feature dictionary (n×k matrix, unit A), composed of k electrical energy feature atoms (each atom is an n×1 vector); E is the environmental feature dictionary (m×l matrix, unit m / s), composed of l environmental feature atoms;

[0178] X represents real electrical energy data (e.g., a current sampling sequence over a 10-minute period, with a dimension of n×1 and a unit of A); Y represents real environmental data (e.g., a vibration velocity sampling sequence for the corresponding time period, with a dimension of m×1 and a unit of m / s); Ω represents the associated atom pair index set, recording atom pairs with related physical meanings; for example, the atom corresponding to the 100Hz harmonic in the electrical energy characteristics (index 3) and the atom corresponding to the 100Hz resonance in the environmental vibration (index 5) form a pair (3,5), Ω={(3,5),(7,9),...}; α Ω For the energy coefficient extracted from α that is associated with environmental characteristics (e.g., the first pair of indices in Ω is (3,5), then α Ω Includes α3); β Ω The environmental coefficients extracted from β that are associated with electrical energy characteristics (corresponding to the example above, β) Ω Including β5); γ1 is the sparsity weight (value 0.2), used to balance "reconstruction accuracy" and "sparseness". The larger γ1 is, the more emphasis is placed on sparsity (which may sacrifice some reconstruction accuracy); conversely, it focuses more on reconstruction accuracy; γ2 is the association weight (value 0.15), which controls the strength of association constraints. The larger γ2 is, the more it forces α. Ω With β Ω Approximately close (e.g., the harmonic coefficient and resonance coefficient at 100Hz are almost equal), strengthening the coupling relationship between electrical energy and environmental characteristics; ||.||1 represents the L1 norm (also known as the Manhattan norm); It represents the square of the L2 norm (also known as the square of the Euclidean norm).

[0179] It should be noted that when matrix D (n×k) is multiplied by vector α (k×1), the resulting Dα has a dimension of n×1 (vector), which is exactly the same as the dimension of X (n×1). Therefore, X-Dα is a vector subtraction (subtracting corresponding elements).

[0180] Similarly, when matrix E (m×l) is multiplied by vector β (l×1), the resulting Eβ has a dimension of m×1 (vector), which is the same as the dimension of Y (m×1). Y-Eβ is also a valid vector subtraction.

[0181] It needs to be explained that α is a k×1 sparse coefficient vector and β is an l×1 sparse coefficient vector. Therefore, ||α||1 and ||β||1 are the L1 norm (Manhattan norm) of the vector, which quantizes the sum of the absolute values ​​of the vector elements.

[0182] The dimension of X-Dα is n×1 (vector), and the dimension of Y-Eβ is m×1 (vector), therefore , It is the square of the L2 norm of the vector (Euclidean norm squared), quantizing the sum of squares of the vector elements.

[0183] The classic definitions of L1 norm and L2 norm both apply to vector spaces; here, all norms apply to vectors.

[0184] In some embodiments, joint sparse coding is a data processing method for extracting key information from a large number of features (such as identifying the core features that best represent complex electrical and environmental data); the "objective function" is a rule that guides this extraction process, ensuring that the extracted information is both concise and accurately reflects the original data.

[0185] The sparsity coefficient refers to the "importance" of each feature during encoding (e.g., a larger coefficient indicates a more critical feature); associated atom pairs are the electrical energy and environmental feature atoms that establish a connection. Adding a "sparse coefficient coupling constraint term" requires that the importance of associated atom pairs cannot differ too much—for example, setting a tolerance threshold of 0.1, if the sparsity coefficient of the electrical energy atom is 0.8, then the sparsity coefficient of the environmental atom must be between 0.7 and 0.9. This ensures that the two associated atoms are "equally strong and equally weak" during encoding. This embodiment does not limit the specific value of the tolerance threshold.

[0186] Without coupling constraints, joint encoding might result in situations where "electrical features are important but environmental features are ignored," which is unrealistic (because the two are correlated). By constraining the sparsity coefficients of correlated atom pairs, we can ensure that the importance of electrical and environmental features is consistent in the encoding result, reflecting their true coupling relationship, preventing the encoding result from deviating from physical laws, and improving the accuracy of subsequent data reconstruction.

[0187] The updated safe current carrying limit is converted into an amplitude regularization factor and added to the boundary constraints of the objective function.

[0188] In some embodiments, the updated safe current carrying limit is a dynamically adjusted maximum safe current carrying capacity of the conductor (e.g., 350 amperes); the amplitude regularization factor is a parameter used to limit the size of the data (e.g., the larger the factor, the heavier the penalty for data exceeding the limit).

[0189] When converting the safe current-carrying limit into a regularization factor, the smaller the limit (indicating a smaller current the conductor can carry), the larger the regularization factor. By adding this factor to the boundary constraints of the objective function, when the amplitude of the electrical data (e.g., current magnitude) in the joint sparse coding result exceeds the safe current-carrying limit, the regularization factor will "penalize" this result, forcing the coding process to adjust, so that the amplitude of the electrical data in the final coding result will not exceed the safe current-carrying limit.

[0190] Based on the embodiments provided in this application, the specific implementation of joint sparse coding is refined (power feature compensation generation, environmental feature correction generation, cross-mapping association, coupling constraint term, and current-carrying limit regularization). After calculating the basic power features (heat generation) based on Joule's law, conductor deformation parameters (equivalent cross-sectional area change rate, dynamic resistivity) are fused for compensation, and the generated power feature dictionary atoms more realistically reflect the power loss characteristics of the deformed conductor. For environmental vibration data, after solving the mechanical energy transfer efficiency based on the mechanical vibration equation, propagation loss correction is performed by combining the dynamic graph network topology connection strength, and the generated environmental feature dictionary atoms more accurately characterize the transmission and attenuation of vibration energy in the network. A cross-mapping relationship between power atoms (harmonic frequencies) and environmental atoms (resonance frequencies) is established (association when the difference is < threshold), and a sparse coefficient coupling constraint term for associated atom pairs is added to the joint sparse coding objective function (forcing the coefficient difference not to exceed the tolerance), which mathematically strengthens the synergy of physically associated atoms in sparse representation and improves the accuracy of feature separation. The updated safe current-carrying limit is converted into an amplitude regularization factor and added to the boundary constraints, so that the coding process naturally conforms to the electrical boundary conditions for the safe operation of the conductor.

[0191] Furthermore, the power data is separated and reconstructed from the results of joint sparse coding, including:

[0192] The coefficient matrix of the joint sparse coding is analyzed, and the subset of electrical coefficients is separated according to the attribute labels of the atoms in the electrical energy feature dictionary and the atoms in the environmental feature dictionary;

[0193] By using the linear combination of the subset of energy coefficients on the energy characteristic dictionary, the current and voltage waveforms are initially reconstructed;

[0194] Extract vibrational spectrum features related to the current-carrying capacity of conductors from atoms in the environmental feature dictionary;

[0195] Based on the high-frequency attenuation characteristics in the vibration spectrum, the skin effect distortion is corrected for the preliminary reconstructed waveform;

[0196] The updated safe current carrying limit is used as the amplitude constraint boundary, and the amplitude of the corrected waveform is normalized.

[0197] The output is the final power data reconstruction result after skin effect correction and current carrying capacity constraint.

[0198] Based on the embodiments provided in this application, the enhanced steps for power data separation and reconstruction are clarified (coefficient separation, preliminary reconstruction, skin effect correction, current carrying capacity constraint, and final output). After separating the coefficients and initially reconstructing the waveform, vibrational spectrum features (especially high-frequency attenuation characteristics) related to the current carrying capacity of the conductor are extracted from environmental characteristic atoms. Based on this, skin effect distortion correction is performed on the initially reconstructed power waveform (current / voltage), effectively compensating for waveform distortion caused by the skin effect of high-frequency current. Furthermore, an updated safe current carrying limit is used as the amplitude constraint boundary for amplitude normalization processing to ensure that the final reconstructed power data is not only more accurate in waveform shape (after skin correction), but also strictly conforms to the safe current carrying capacity boundary of the conductor in the current state (considering deformation and skin effect) in amplitude range, thereby outputting a purer, more reliable final power data reconstruction result that meets engineering safety constraints.

[0199] It should be noted that the embodiments implemented on the environmental perception-based power data acquisition system side in this application can be referenced with the embodiments implemented on the environmental perception-based power data acquisition method side, and will not be described in detail here.

[0200] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described xx method is also provided, the electronic device being... Figure 4 The terminal device or server shown. This embodiment uses this electronic device as an example of a server. Figure 4 As shown, the electronic device includes a memory 402, a processor 404, and a transmission device 406. The memory 402 stores a computer program, and the processor 404 is configured to execute the steps in any of the above method embodiments through the computer program.

[0201] Optionally, in this embodiment, the aforementioned electronic device may be located in at least one of a plurality of network devices in a computer network.

[0202] Optionally, the transmission device 406 is used to receive or send data via a network. Specific examples of the network described above may include wired and wireless networks. In one example, the transmission device 406 includes a Network Interface Controller (NIC), which can be connected to other network devices and a router via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 406 is a radio frequency (RF) module used to communicate with the Internet wirelessly.

[0203] In addition, the aforementioned electronic device also includes: a display 408 for displaying target identification characters contained in the identity identifier of the identified target object; and a connection bus 410 for connecting various module components in the aforementioned electronic device.

[0204] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. An environment-aware based electric energy data acquisition system, characterized in that, The method comprises the following steps: a physical field feature vector conversion module is used to acquire original data collected by each sensor in a monitoring area in real time, and convert the original data into a physical field feature vector comprising heat flow concentration and eddy current intensity; a heterogeneous graph network construction module is used to construct a dynamic heterogeneous graph network based on the physical field feature vector; a minimum perception subgraph formation module is used to solve a continuous area in which a physical field distortion gradient value exceeds a gradient threshold value by using a topological relationship of the dynamic heterogeneous graph network, and activate sensors in the continuous area to form a minimum perception subgraph; a target electric energy collection point association module is used to associate a target electric energy collection point according to distortion coordinates of the minimum perception subgraph by using conductor current-carrying capacity; a joint sparse coding module is used to input electric energy data of the target electric energy collection point and environmental data of the minimum perception subgraph into an encoder, and perform the following operations: generate electric energy feature dictionary atoms based on Joule's law; generate environmental feature dictionary atoms based on a mechanical vibration equation; and perform joint sparse coding by using the electric energy feature dictionary atoms and the environmental feature dictionary atoms; an electric energy data reconstruction module is used to separate and reconstruct the electric energy data from a result of joint sparse coding.

2. An environment-aware based electric energy data acquisition method, characterized in that, The method comprises the following steps: acquire original data collected by each sensor in a monitoring area in real time, and convert the original data into a physical field feature vector comprising heat flow concentration and eddy current intensity; construct a dynamic heterogeneous graph network based on the physical field feature vector; solve a continuous area in which a physical field distortion gradient value exceeds a gradient threshold value by using a topological relationship of the dynamic heterogeneous graph network, and activate sensors in the continuous area to form a minimum perception subgraph; associate a target electric energy collection point according to distortion coordinates of the minimum perception subgraph by using conductor current-carrying capacity; input electric energy data of the target electric energy collection point and environmental data of the minimum perception subgraph into an encoder, and perform the following operations: generate electric energy feature dictionary atoms based on Joule's law; generate environmental feature dictionary atoms based on a mechanical vibration equation; and perform joint sparse coding by using the electric energy feature dictionary atoms and the environmental feature dictionary atoms; separate and reconstruct the electric energy data from a result of joint sparse coding.

3. The environment-aware based power data collection method of claim 2, wherein, Each sensor comprises a temperature sensor, a vibration sensor and a magnetic field sensor; the dynamic heterogeneous graph network is constructed based on the physical field feature vector, which comprises the following steps: each sensor is taken as a node and a physical field feature vector of a position of the node is associated; a spatial connection weight between nodes is generated, the spatial connection weight decreases exponentially with an increase of a sensor distance, and a decrease rate is controlled by a pre-set decay coefficient; a physical field coupling weight between nodes is generated, a coupling connection is established between a first node and a second node when a direction angle between a temperature gradient vector at the first node and a magnetic field gradient vector at the second node is less than 37 degrees, and a coupling strength is proportional to a product of a modulus of the temperature gradient vector at the first node and a modulus of the magnetic field gradient vector at the second node; the spatial connection weight and the physical field coupling weight are fused into a final connection weight according to a dynamic proportioning coefficient, and the dynamic proportioning coefficient is adjusted according to a sensor distribution density in the monitoring area. The connection structure of the dynamic heterogeneous graph network is used to guide the topological traversal path in the physical field distortion gradient solving process.

4. The environment-aware based power data collection method of claim 3, wherein, The topological relationship of the dynamic heterogeneous graph network is used to solve the continuous area of the physical field distortion gradient value exceeding the gradient threshold value. The sensors in the continuous area are activated to form a minimum perception subgraph, including: Each node in the dynamic heterogeneous graph network is traversed as a current target node; A set of adjacent nodes directly connected to the current target node is determined; The gradient vector of the physical field feature vector of the current target node and the physical field feature vector of each adjacent node is calculated; The maximum value of the modulus of all gradient vectors is taken as the physical field distortion gradient value of the current target node; The nodes with physical field distortion gradient values exceeding the gradient threshold value are screened as candidate distortion nodes.

5. The environmental perception-based electric energy data acquisition method according to claim 4, characterized in that, The candidate distortion nodes are used to start breadth-first traversal along edges with connection weights greater than a weight threshold value; During the traversal process, the change direction of the physical field feature vectors of three consecutive visited nodes is recorded; The direction change angle is calculated, and if the direction change angles of the three consecutive nodes are all less than 15 degrees, the nodes are marked as valid propagation paths; All the nodes covered by the valid propagation paths are aggregated to form the continuous area; It is verified whether the number of target node pairs in the continuous area that satisfy the physical field coupling condition exceeds a set number threshold value; wherein the target node pair includes a temperature sensor node and a magnetic field sensor node; If the number of target node pairs in the continuous area that satisfy the physical field coupling condition exceeds the set number threshold value, all the sensors in the continuous area are activated to form a minimum perception subgraph.

6. The environment-aware based power data collection method of claim 5, wherein, According to the distortion coordinates of the minimum perception subgraph, a target electric energy acquisition point is associated through the conductor current-carrying capacity, including: An electric energy acquisition point set electrically connected to the conductors in the monitoring area is located; Temperature data of all temperature sensors in the minimum perception subgraph is extracted; The deviation value of the temperature data from the average temperature of the area is calculated; Temperature sensors with deviation values greater than three times the standard deviation are screened to form a high-temperature cluster; The weighted spatial barycentric coordinates of the high-temperature cluster are calculated, with the weight being the temperature deviation value; The weighted spatial barycentric coordinates are output as the heat aggregation core coordinates, and the material parameters of the conductors associated with the heat aggregation core coordinates are obtained; In the electrically connected subnetwork, a plurality of electric energy acquisition points closest to the heat aggregation core coordinates are screened; wherein the electrically connected subnetwork is divided according to the circuit breaker tripping logic; The conductor resistance temperature coefficient and the safe current-carrying limit value are called; For each of the plurality of electric energy acquisition points, the real-time current value of the point is obtained; the dynamic safe current-carrying limit value of the point is calculated according to the conductor resistance temperature coefficient, the safe current-carrying limit value, and the real-time temperature of the conductor at the point; the real-time current-carrying margin of the point is calculated according to the real-time current value and the dynamic safe current-carrying limit value; The acquisition point with the smallest real-time current-carrying margin is selected as the associated target.

7. The environment-aware based power data collection method of claim 6, wherein, In the step of calling the conductor resistance temperature coefficient and the safe current-carrying limit value, mechanical deformation compensation and electromagnetic skin effect analysis are fused and executed, specifically including: Identify a characteristic harmonic component formed by the joint action of electromagnetic force and mechanical stress based on the mechanical vibration frequency spectrum of the conductor collected by the vibration sensor in the minimum perception subgraph; Invert the micro-deformation distribution pattern of the conductor near the thermal aggregation core coordinates according to the amplitude-frequency characteristics and phase difference of the characteristic harmonic component; Generate deformation compensation heat source coordinates by using the micro-deformation distribution pattern to compensate for the three-dimensional spatial offset of the thermal aggregation core coordinates, and simultaneously correct the equivalent cross-sectional area change rate of the conductor at the deformation compensation heat source coordinates; Extract real-time temperature data of the temperature sensor at the deformation compensation heat source coordinates, and calculate the dynamic resistivity under the influence of deformation in combination with the resistance temperature characteristics of the conductor material.

8. The environment-aware based power data collection method of claim 7, wherein, The method further comprises: Obtain the current harmonic frequency spectrum monitored by the target power collection point, and select a dominant harmonic frequency associated with the vibration characteristic harmonic component frequency; Input the dynamic resistivity, the corrected equivalent cross-sectional area change rate, the dominant harmonic frequency, and the magnetic permeability of the conductor into the skin effect depth calculation model to solve the equivalent skin depth of the conductor under the coupling state of mechanical deformation and high-frequency current; Calculate the effective current-carrying capacity of the conductor under harmonic working conditions based on the equivalent skin depth and the corrected equivalent cross-sectional area change rate; Dynamically update the safe current-carrying limit value according to the effective current-carrying capacity and the allowable temperature rise threshold of the conductor; Recalculate the real-time current-carrying margin using the updated safe current-carrying limit value; Verify and correct the selection results associated with the target power collection point according to the recalculated real-time current-carrying margin. 9.The environment-aware based electric energy data collection method of claim 7, wherein, Generate power feature dictionary atoms based on Joule's law and environment feature dictionary atoms based on mechanical vibration equations; Joint sparse coding is performed using the power feature dictionary atoms and the environment feature dictionary atoms, including: For the current and voltage time series data of the target power collection point, calculate the heat generation of the conductor per unit time as the basic power feature according to Joule's law; Compensate the basic power feature for conductor deformation parameters by fusing the equivalent cross-sectional area change rate of the minimum perception subgraph and the dynamic resistivity to generate the power feature dictionary atoms; Extract the time-domain vibration waveform of the vibration sensor in the minimum perception subgraph, and solve the mechanical energy transmission efficiency based on the mechanical vibration equation; Combine the topological connection strength of the dynamic heterogeneous graph network on the effective propagation path to correct the network propagation loss of the environmental mechanical vibration energy and generate the environment feature dictionary atoms.

10. The environment-aware based power data collection method of claim 9, wherein, The method further comprises: Establish a cross-mapping relationship between the power feature dictionary atoms and the environment feature dictionary atoms, and establish atom association when the difference between the harmonic frequency in the power feature dictionary atom and the resonance frequency in the environment feature dictionary atom is less than a set difference threshold; Construct a target function for joint sparse coding, and add a sparse coefficient coupling constraint term to the target function, which forces the difference between the sparse coefficients of the associated atom pair to be less than a set tolerance threshold; Convert the updated safe current-carrying limit value into an amplitude regularization factor and add it as a boundary constraint condition to the target function.

Images