Cable branch box adaptive load control method and system

Abstract

The application provides a cable branch box adaptive load control method and system, relates to the technical field of data processing, and comprises the following steps: under the constraint of a current change boundary constraint set, iteratively adjusting current distribution values of each loop with the balanced residual heat tolerance time as a target; after each iteration adjustment, recalculating expected heat generation based on the adjusted current value, driving temperature field deformation, updating dynamic temperature field topology distribution, heat flow path and residual heat tolerance time, until the current distribution value difference of two adjacent iterations is less than a set threshold, and the final current distribution value is taken as a current control adjustment amount; and according to the current control adjustment amount, corresponding control levels are matched, hierarchical response control instructions are generated and sent to an execution terminal for action response. The application can realize active optimization distribution of the loads of each loop of the cable branch box, effectively prevents local overheating, prolongs the service life of the equipment and improves the power supply safety and reliability.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to an adaptive load control method and system based on cable branch boxes. Background Technology

[0002] In the actual operation of cable branch boxes on the low-voltage side of the distribution network, the number of outgoing circuits is usually large, and the types of loads carried by each circuit may differ. For example, some circuits are residential lighting loads, while others are small power loads or charging pile loads, and their power consumption behavior has a certain degree of uncertainty in the time dimension. When pulse loads such as electric vehicle charging piles are connected, the load current of some circuits may fluctuate within a specific period of time, which will cause corresponding changes in the temperature at the cable joint, while the load level of other circuits may be relatively stable. Some existing load control methods adopt protection strategies based on fixed current or temperature thresholds, that is, when the detected value exceeds the set threshold... After the threshold is reached, tripping or load reduction operations are performed; however, under certain actual operating conditions, this method may not fully consider the continuous spatial distribution characteristics of the internal temperature field of the cable and the dynamic response process of the heat conduction path with load changes. Therefore, before load adjustment is required, the assessment of the current thermal withstand capability of each circuit may have some deviation. For example, when the temperature of a certain circuit is already at a high level but has not yet reached the protection threshold, while the adjacent circuits still have a large thermal capacity margin, if the judgment is made only based on the independent threshold, it may not be possible to coordinate the load distribution of each circuit in a timely manner, thereby affecting the refinement of the control strategy and the safety margin of the overall equipment operation to a certain extent. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an adaptive load control method and system based on cable branch boxes, which can realize the active optimization distribution of loads in each circuit of the cable branch box, effectively prevent local overheating, extend equipment life and improve power supply safety and reliability.

[0004] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: Firstly, an adaptive load control method for cable branch boxes is provided, the method comprising: Step 1: Real-time acquisition of load current values ​​and cable joint temperature values ​​for each branch circuit; Step 2: Based on the load current value and its rate of change, determine the maximum allowable increase or decrease of current in each circuit in the future, which will serve as the upper and lower limits of the adjustable current. These limits will be summarized to form a set of current change boundary constraints. Step 3: Based on the collected temperature value, load current value, and current change boundary constraint set, construct the initial grid of the cable temperature field; calculate the corresponding expected heat generation using the load current value, drive the initial grid of the cable temperature field to undergo dynamic deformation, obtain the current dynamic temperature field topology distribution, and determine the heat flow path based on the temperature gradient of the current dynamic temperature field topology distribution. Step 4: Compare the temperatures of adjacent nodes along the heat flow path to determine the degree of deviation of each location from the rated maximum allowable temperature, and map the degree of deviation to the remaining heat tolerance time of each loop. Step 5: Under the constraints of the current change boundary constraint set, with the goal of balancing the remaining heat tolerance time, iteratively adjust the current distribution value of each loop; after each iteration adjustment, recalculate the expected heat generation based on the adjusted current value and drive the temperature field deformation, update the dynamic temperature field topology distribution, heat flow path and remaining heat tolerance time, until the difference between the current distribution values ​​of two adjacent iterations is less than the set threshold, and use the final current distribution value as the current control adjustment amount. Step 6: Match the corresponding control level according to the current control adjustment amount, generate a graded response control command and send it to the execution terminal for action response.

[0005] Secondly, the adaptive load control system based on cable branch boxes includes: The data acquisition module is used to collect the load current value and cable joint temperature value of each branch circuit in real time; The current boundary constraint generation module is used to determine the maximum allowable increase or decrease of current in each circuit in the future based on the load current value and its rate of change, which serves as the upper and lower limits of the adjustable current, and is summarized to form a set of current change boundary constraints. The analysis module is used to construct an initial grid of the cable temperature field based on the collected temperature values, load current values, and current change boundary constraint sets; calculate the corresponding expected heat generation using the load current values, drive the initial grid of the cable temperature field to undergo dynamic deformation, obtain the current dynamic temperature field topology distribution, and determine the heat flow path based on the temperature gradient of the current dynamic temperature field topology distribution; The heat tolerance time assessment module is used to compare the temperature of adjacent nodes along the heat flow path, determine the degree of deviation of each location from the rated maximum allowable temperature, and map the degree of deviation to the remaining heat tolerance time of each loop. The optimization module is used to iteratively adjust the current distribution value of each loop under the constraint of the current change boundary constraint set, with the goal of balancing the remaining heat tolerance time. After each iteration adjustment, the expected heat generation is recalculated based on the adjusted current value and the temperature field is deformed. The dynamic temperature field topology distribution, heat flow path and remaining heat tolerance time are updated until the difference between the current distribution values ​​of two adjacent iterations is less than the set threshold. The final current distribution value is used as the current control adjustment amount. The execution module is used to match the corresponding control level according to the current control adjustment amount, generate hierarchical response control commands and send them to the execution terminal for action response.

[0006] Thirdly, a computing device, comprising: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0007] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0008] The above-described solution of the present invention has at least the following beneficial effects: By collecting real-time load current and cable joint temperature data from each branch circuit, and combining current values ​​and rates of change to construct a current change boundary constraint set, while simultaneously discretizing the internal space of the cable branch box into a grid to correct the temperature field grid topology, a dynamic temperature field topology distribution is effectively constructed. Based on temperature gradient analysis, the actual heat transfer path is identified. This method can capture dynamic temperature changes caused by load fluctuations, effectively locate high-temperature areas and heat transfer patterns, and improve the one-sidedness of single-point fixed threshold assessments. It allows for dynamic and holistic analysis of the cable's thermal state, helping to improve the accuracy and completeness of equipment thermal state perception. By matching the real-time current change rate of each circuit with a preset maximum allowable change rate threshold, and calculating the adjustable upper and lower limits of the current for each circuit based on the load control response cycle, a current change boundary constraint set is constructed, standardizing the operating range of a single load adjustment. Simultaneously, iterative optimization of the control logic is adopted. After each current adjustment, the temperature field distribution, heat flow path, and thermal tolerance time are updated synchronously. The control parameters are corrected through a closed-loop approach, gradually stabilizing the current distribution value. This method can reduce the potential operational risks of equipment caused by over-range adjustments, adapt to the fluctuation characteristics of different types of loads such as residential loads and charging pile pulse loads, keep the load adjustment process stable and compliant, and help improve the adaptability and operational stability of the control strategy. Attached Figure Description

[0009] Figure 1 This is a schematic flowchart of an adaptive load control method based on a cable branch box provided in an embodiment of the present invention.

[0010] Figure 2 This is a schematic diagram of an adaptive load control system based on a cable branch box provided in an embodiment of the present invention. Detailed Implementation

[0011] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0012] like Figure 1 As shown, an embodiment of the present invention proposes an adaptive load control method based on a cable branch box, the method comprising the following steps: Step 1: Real-time acquisition of load current values ​​and cable joint temperature values ​​for each branch circuit; Step 2: Based on the load current value and its rate of change, determine the maximum allowable increase or decrease of current in each circuit in the future, which will serve as the upper and lower limits of the adjustable current. These limits will be summarized to form a set of current change boundary constraints. Step 3: Based on the collected temperature value, load current value, and current change boundary constraint set, construct the initial grid of the cable temperature field; calculate the corresponding expected heat generation using the load current value, drive the initial grid of the cable temperature field to undergo dynamic deformation, obtain the current dynamic temperature field topology distribution, and determine the heat flow path based on the temperature gradient of the current dynamic temperature field topology distribution. Step 4: Compare the temperatures of adjacent nodes along the heat flow path to determine the degree of deviation of each location from the rated maximum allowable temperature, and map the degree of deviation to the remaining heat tolerance time of each loop. Step 5: Under the constraints of the current change boundary constraint set, with the goal of balancing the remaining heat tolerance time, iteratively adjust the current distribution value of each loop; after each iteration adjustment, recalculate the expected heat generation based on the adjusted current value and drive the temperature field deformation, update the dynamic temperature field topology distribution, heat flow path and remaining heat tolerance time, until the difference between the current distribution values ​​of two adjacent iterations is less than the set threshold, and use the final current distribution value as the current control adjustment amount. Step 6: Match the corresponding control level according to the current control adjustment amount, generate a graded response control command and send it to the execution terminal for action response.

[0013] In this embodiment of the invention, by real-time acquisition of load current and cable joint temperature data of each branch circuit, and combining current values ​​and rates of change to construct a current change boundary constraint set, the internal space of the cable branch box is discretized into a grid to complete the temperature field grid topology correction, effectively constructing a dynamic temperature field topology distribution, and identifying the actual heat flow transfer path based on temperature gradient analysis. This method can capture the dynamic temperature changes caused by load fluctuations, effectively locate high-temperature areas and heat transfer patterns, improve the one-sidedness of single-point fixed threshold assessment, and conduct dynamic and holistic analysis of cable thermal status, which helps to improve the accuracy and completeness of equipment thermal status perception. By matching the real-time current change rate of each circuit with the preset maximum allowable change rate threshold, and by calculating the adjustable upper and lower limits of the current of each circuit in combination with the load control response cycle, a current change boundary constraint set is constructed to standardize the operating range of a single load adjustment. At the same time, an iterative optimization control logic is adopted. After each current adjustment, the temperature field distribution, heat flow path, and heat tolerance time are updated synchronously. The control parameters are corrected through a closed-loop method, so that the current distribution value gradually tends to stabilize. This method can reduce the potential operational risks of equipment caused by over-range adjustments, adapt to the fluctuation characteristics of different types of loads such as residential loads and charging pile pulse loads, keep the load adjustment process stable and compliant, and help improve the adaptability and operational stability of the control strategy.

[0014] In a preferred embodiment of the present invention, the execution process of step 1 is as follows: The acquisition circuit continuously acquires the analog electrical signals output by the current sensors installed on each branch circuit at a preset sampling frequency, and simultaneously continuously acquires the analog electrical signals output by the temperature sensors installed at each cable joint location. The preset sampling frequency is set according to the fluctuation characteristics of the load carried by the cable branch box. Specifically, the root mean square value of the current change amplitude between two adjacent sampling time points in the daily load curve of the branch circuit is used as the fluctuation intensity index. When the fluctuation intensity index exceeds 15% of the rated current of the circuit, a sampling frequency of 5 to 10 times per second is adopted, that is, the sampling interval is 0.1 to 0. 2 seconds; when the fluctuation intensity index is lower than or equal to 15% of the rated current of the circuit, a sampling frequency of 1 to 2 times per second is adopted, that is, the sampling interval is 0.5 to 1 second; the above sampling frequency and 15% fluctuation intensity threshold are for illustrative purposes only. In actual engineering applications, a suitable sampling frequency can be determined based on the historical fluctuation statistics of the load type of each branch circuit to achieve a balance between the real-time data acquisition and the computational load of the control unit; after the analog signal acquisition circuit completes the acquisition of the analog signal, the analog-to-digital conversion circuit converts the acquired analog electrical signal into a digital quantity; for the analog signal output by the current sensor, the analog-to-digital conversion circuit... The conversion circuit converts the digital signal into the corresponding actual load current value in amperes based on the turns ratio parameters of the current sensor. For the analog signal output by the temperature sensor, the analog-to-digital conversion circuit converts the digital signal into the corresponding actual temperature value in degrees Celsius based on the temperature-voltage correspondence table pre-stored in the load control unit's memory. The temperature-voltage correspondence table exists in the form of a calibration curve, which is constructed as follows: in a standard constant temperature bath, within a temperature range of -40 degrees Celsius to +150 degrees Celsius, a constant temperature point is set every 5 degrees Celsius, and the voltage output by the temperature sensor is recorded at each constant temperature point. The values ​​are plotted in a two-dimensional rectangular coordinate system with temperature on the x-axis (degrees Celsius) and voltage on the y-axis (millivolts). All data points corresponding to constant temperatures are then plotted, and adjacent data points are connected sequentially using piecewise linear interpolation to obtain a continuous polygonal line composed of multiple straight line segments. This polygonal line is the temperature-voltage calibration curve. The load control unit obtains a set of real-time data in each sampling period. This set of real-time data serves as a real-time data acquisition set, containing the following elements: the load current value of each branch circuit, expressed as a floating-point number in amperes, and the temperature value at each cable joint location, expressed as a floating-point number in degrees Celsius.

[0015] The real-time acquired data set, obtained after analog-to-digital conversion and physical quantity conversion, is temporarily stored in the random access memory of the load control unit in the form of a two-dimensional data table. The data table contains the following fields: timestamp of the sampling time, accurate to milliseconds; branch loop number, a positive integer starting from 1, each number uniquely corresponding to a branch loop; load current value, floating-point number, in amperes; cable joint temperature value, floating-point number, in degrees Celsius. Each row of records in this data structure contains the complete state of a branch loop at a sampling time.

[0016] In a preferred embodiment of the present invention, step 2 is performed as follows: The load control unit's data processing and calculation circuit reads the real-time data sets from the random access memory at the current sampling time and the historical sampling time. For each branch circuit, the data processing and calculation circuit extracts the load current value at the current sampling time and records it as the current current value. At the same time, it extracts the load current value at the previous sampling time, which is immediately adjacent to the current sampling time and records it as the historical current value. The calculation process of the current change rate is as follows: subtract the historical current value from the current current value to obtain the current change amount. Divide the current change amount by the time length of the sampling period, that is, the time interval between two adjacent samplings. The quotient is the current change rate of the branch circuit in the current sampling period. A positive current change rate indicates that the load current of the circuit is increasing, and a negative current change rate indicates that the load current of the circuit is decreasing. The magnitude of the absolute value of the current change rate reflects the speed of the load current change.

[0017] To improve the stability and anti-interference capability of the current change rate calculation, the data processing and arithmetic circuit adopts a sliding window averaging method to calculate the current change rate. Specifically, it takes the load current values ​​of the five consecutive sampling moments preceding the current sampling moment. With a sampling frequency of 5 times per second, these five sampling moments correspond to the sampling data within the previous second. A linear fit is then performed on these five load current values, i.e., the best-fit line is obtained using the least squares method. The slope of this line is the current change rate at the current moment. Using the sliding window averaging method can retain the current change trend information while reducing interference from individual factors. The influence of measurement noise or sudden current spikes at the sampling time on abnormal changes in the rate of change of current is considered. The length of the sliding window, i.e., the number of sampling points involved in the fitting, is a configurable parameter, ranging from 3 to 20 sampling points. The selection criteria are as follows: if the historical load fluctuation of the branch circuit is mainly low-frequency and slow, with a fluctuation frequency below 0.5 Hz, the window length can be 10 to 20 sampling points to enhance the smoothing effect; if the branch circuit has high-frequency impulsive loads with a fluctuation frequency above 0.5 Hz, the window length should be 3 to 5 sampling points to retain the ability to respond promptly to rapid changes.

[0018] In the memory of the load control unit, a maximum allowable current change rate threshold is pre-configured for each branch circuit. This threshold, measured in amperes per second, represents the maximum allowable current change rate per unit time for that branch circuit, taking into account the cross-sectional area of ​​the cable conductor, the heat resistance rating of the insulation material, the thermal stability of the contact resistance at the joint, and the maximum withstand capability of the connected load equipment to sudden current changes. The range of this maximum allowable current change rate threshold and its determination method are as follows: For residential lighting circuits, the cross-sectional area of ​​the branch circuit cable is typically 10 to 25 square millimeters, and the insulation material is polyvinyl chloride (PVC). The range of the maximum allowable current change rate threshold is 0.5 amperes per second to 3.0 amperes per second, with a recommended value of 1.0 amperes per second to 2.0 amperes per second. The rationale is as follows: The loads of residential lighting circuits mainly consist of lighting fixtures and household appliances. Their natural load fluctuations are typically between 5% and 20% of the rated current, with current changes occurring slowly on the order of seconds to minutes. Therefore, it is not necessary to set an excessively high allowable rate of change. For small power circuits, the cross-sectional area of ​​their branch circuit cables is typically 25 to 70 square millimeters, and the insulation material is cross-linked polyethylene. The maximum allowable current change rate threshold ranges from 3.0 amps per second to 20.0 amps per second, with a recommended value of 5.0 amps per second to 10.0 amps per second. The rationale for this value is that the motor loads connected to small power circuits generate an inrush current several times the rated current at the moment of startup, which typically lasts for 0.5 to 3 seconds. Therefore, a higher allowable rate of change needs to be set to accommodate the normal start-up and shutdown conditions of the motor. For electric vehicle charging pile circuits, the cross-sectional area of ​​branch circuit cables is typically 10 to 35 square millimeters, and the maximum allowable current change rate threshold ranges from 2.0 amps per second to 15.0 amps per second, with a recommended value of 5.0 amps per second to 8.0 amps per second. The basis for this value is that the charging current of the charging pile will undergo significant step changes within a few seconds to tens of seconds according to the needs of the battery management system, but it should not be too high to avoid excessive thermal shock to the cable joints.

[0019] The absolute value of the current rate of change of each branch circuit (focusing only on the speed of change without distinguishing between rising and falling currents) is compared with the preset maximum allowable current rate of change threshold for that branch circuit. If the absolute value of the current rate of change is less than or equal to the threshold, the matching result is qualified, indicating that the load change rate of that branch circuit is within its own safe range. In this case, the upper limit of the current adjustment rate is directly taken as the maximum allowable current rate of change threshold of that branch circuit. If the absolute value of the current rate of change is greater than the threshold, the matching result is out of limit, indicating that the load change rate of that branch circuit has exceeded its own safe range. However, considering that the load control should not further aggravate the drastic load change, the upper limit of the current adjustment rate is still taken as the maximum allowable current rate of change threshold of that branch circuit. That is, the active adjustment action of the control system is not allowed to change the load current at a rate exceeding the circuit's capacity.

[0020] The load control response cycle refers to the maximum allowable time interval corresponding to the complete cycle of the load control unit from completing a current acquisition, calculation, and control decision to issuing a control command, executing the terminal to complete the action response, and then to the next data acquisition. This parameter is pre-configured in the memory of the load control unit. The value range of the load control response cycle is 1 second to 10 seconds, and its selection is based on the following: if the load type of each branch circuit connected to the cable branch box is mainly stable load, such as municipal lighting and small motors operating at constant power, the response cycle can be 5 seconds to 10 seconds to reduce the computational load and communication overhead of the control unit; if the load connected to the cable branch box includes many rapidly fluctuating loads, such as electric vehicle charging... For loads such as electrical piles and elevators that frequently start and stop, the response period should be 1 to 3 seconds to ensure timely response of load control to rapid changes. In the operation of a cable branch box in a residential area, the recommended value for the load control response period is 2 seconds. For each branch circuit, multiply the upper limit of the current adjustment rate by the length of the load control response period. The product is the maximum allowable change in current for that branch circuit within a single load control response period, in amperes. The maximum increase in current is taken as the value of the maximum change in current, indicating how many amperes the current can be increased by at most within that response period. The maximum decrease in current is also taken as the value of the maximum change in current, indicating how many amperes the current can be decreased by at most within that response period.

[0021] For each branch circuit, read the load current value at the current sampling moment. Add the maximum current increase to the current value; the sum is the adjustable upper limit of the current for that branch circuit. This adjustable upper limit indicates that, under the current operating conditions, limited by the circuit's own current change rate and the load control response cycle, the load current of that circuit can only increase to this upper limit in the next response cycle; the current cannot be set higher than this upper limit during actual control adjustment. Subtract the maximum current decrease from the current value; the difference is the adjustable lower limit of the current for that branch circuit. This adjustable lower limit indicates that the load current of that circuit can only decrease to this lower limit in the next response cycle; the current cannot be set lower than this lower limit during actual control adjustment. If the calculated adjustable lower limit is negative, i.e., when the maximum current decrease exceeds the current load current value, the adjustable lower limit is zero, because the load current cannot be negative.

[0022] The adjustable upper and lower limits of current calculated for each branch circuit are compiled and organized according to the branch circuit number. For each branch circuit, its allowable current adjustment range is defined as a closed interval. The two endpoints of this interval are the adjustable lower limit of current (the starting point) and the adjustable upper limit of current (the ending point). The length of the interval is equal to the adjustable upper limit of current minus the adjustable lower limit of current, representing the adjustable current range of the circuit in the next control cycle. The allowable current adjustment ranges of all branch circuits are combined to form the current variation boundary constraint set. The complete components of the current variation boundary constraint set are as follows: if the total number of branch circuits is the same as the actual number of circuits configured in the cable branch box, the constraint set contains the same number of closed interval elements as the total number of branch circuits. Each closed interval consists of two floating-point endpoint elements, namely the adjustable lower limit and the adjustable upper limit, and is accompanied by the corresponding branch circuit number as an identifier. The physical meaning of this constraint set is: in the next load control response cycle, when the load control unit adjusts the load current of any branch circuit, the adjusted current value must fall within the current allowable adjustment range of the corresponding circuit; in other words, the current change boundary constraint set provides a multi-dimensional constraint space for subsequent load balancing iterative adjustment, with the dimension equal to the total number of branch circuits; each component of the current allocation value vector generated by any iterative adjustment step must satisfy the interval constraints of the corresponding branch circuit in this constraint set.

[0023] In a preferred embodiment of the present invention, step 3 is performed as follows: First, a geometric space description representing the three-dimensional space inside the cable branch box is established in memory. The length, width, and height dimensions of this geometric space description correspond to the actual internal space dimensions of the cable branch box, including the spatial orientation and position of each branch circuit conductor, the spatial coordinates of each cable joint node, and the positional information of boundary structures such as the box walls. Taking a six-circuit cable branch box as an example, its internal space dimensions are approximately 800 mm long, 600 mm wide, and 300 mm high. Each branch circuit conductor is a horizontally oriented copper busbar or copper core cable with a rectangular or circular cross-section, arranged at equal intervals along the width direction inside the box, with a spacing of approximately 60 mm to 100 mm. This three-dimensional geometric space description is then spatially discretized, that is, the continuous internal space is divided into multiple small volume units, and the vertex of each small volume unit is used as a mesh node, thus forming a mesh structure composed of mesh nodes and edges connecting the mesh nodes. The spatial discretization can be performed using the finite element mesh generation method, dividing the three-dimensional space into tetrahedral elements. The side length parameter of the tetrahedral element determines the mesh resolution. In areas where the temperature gradient is expected to be hundreds to thousands of degrees Celsius per meter near the conductor surface and cable joints, a side length parameter of 1 mm to 3 mm is set to ensure the accuracy of the temperature field calculation. In areas where the temperature gradient is expected to be several to tens of degrees Celsius per meter in the air region, a side length parameter of 10 mm to 20 mm is set to control the total number of mesh nodes and the amount of computation. If the internal structure of the cable branch box has relatively regular geometric features, such as a cuboid structure and the conductors of each branch circuit arranged in parallel in the horizontal or vertical direction, a structured orthogonal mesh can be used for discretization. The space is divided into cuboid elements at equal or unequal intervals along the length, width, and height directions. The mesh spacing in the conductor region is 2 mm to 5 mm, and the mesh spacing in the air region is 10 mm to 15 mm. The eight vertices of each cuboid element are the mesh nodes.

[0024] After spatial discretization, attribute values ​​are assigned to each grid node. For grid nodes whose spatial location corresponds to the installation location of a cable joint sensor, the temperature value collected by the sensor at the current sampling time is assigned to the grid node as its initial temperature value. For other grid nodes whose spatial location does not belong to the cable joint location, their initial temperature values ​​are estimated using spatial linear interpolation based on the temperature values ​​of adjacent cable joint nodes: the spatial distance between the node to be estimated and its three nearest surrounding cable joint nodes is calculated; the reciprocal of the distance is used as the weight to perform a weighted average of the known temperature values ​​of the three joint nodes to obtain the initial temperature value of the node to be estimated. For grid nodes whose spatial location corresponds to the conductor path of a branch loop cable, the load current value collected by the branch loop at the current sampling time is assigned to the grid node as its initial current parameter. The current variation boundary constraints are concentrated on the adjustable upper and lower limits of the current corresponding to the branch loop, and set as the current boundary conditions for all relevant grid nodes on the conductive path of the branch loop. Thus, the data processing and computation circuit constructs an initial mesh for the cable temperature field. This initial mesh, as a data structure, contains the following elements: multiple mesh nodes, each with a unique node number, three-dimensional spatial coordinates, initial temperature value attributes, initial current parameter attributes, and current boundary condition attributes; multiple edges, each connecting two mesh nodes, used to define the topological adjacency relationship of the mesh; multiple surface elements and volume elements, formed by multiple nodes and edges, used to define the geometry of the solution domain; the initial temperature field distribution of the mesh is composed of the measured temperature value of the cable joint and the interpolated temperature values ​​of the remaining nodes.

[0025] In the context of computational geometry and finite element analysis, a mesh is considered a manifold mesh if: every edge in the mesh is shared by exactly two face elements (or, in 3D, the common face of two face elements); the face elements surrounding each vertex must be able to form a continuous, simply connected sector around that vertex; if an edge in the mesh is shared by three or more face elements, then that edge is called a non-manifold edge; if a vertex in the mesh cannot form a continuous, simply connected sector around that vertex (e.g., there are two or more unconnected sectors), then that vertex is called a non-manifold vertex; the existence of non-manifold edges and non-manifold vertices indicates that the mesh has a topological anomaly at that location, which can lead to errors or discontinuities in subsequent mesh-based temperature distribution calculations and heat flow path tracing.

[0026] According to the pre-defined mesh topology data structure, a half-edge data structure is adopted, that is, each edge is split into two directed half edges with opposite directions, and each half edge belongs to a surface cell. Each edge in the initial mesh of the cable temperature field is traversed sequentially. For each edge, the total number of half edges is counted. If the number of half edges is not equal to 2, the edge is marked as a non-manifold edge. Similarly, each vertex is traversed sequentially. For each vertex, all half edges adjacent to the vertex are collected in counterclockwise order. It is checked whether these half edges can form a closed loop chain. If they cannot be closed, the vertex is marked as a non-manifold vertex.

[0027] For each edge and vertex marked as nonmanifold, examine the mesh cells associated with these nonmanifold topological elements. If, among the multiple face cells associated with a nonmanifold edge, some face cells are connected to the mesh body only through that nonmanifold edge, meaning that the other edges of these face cells are not shared with other cells in the mesh body, then these face cells constitute suspended cells. They are like isolated structures suspended outside the mesh body, not participating in the calculation of the continuous temperature field distribution inside the mesh body, and instead introducing unnecessary numerical errors. Similarly, if, among the set of face cells associated with a nonmanifold vertex, some face cells are connected to the mesh body only through that vertex, meaning that the other vertices of these face cells are not shared with other cells in the mesh body, then these face cells also constitute suspended cells.

[0028] Remove all identified dangling elements from the mesh data. The removal operation includes: deleting the data records of the face and volume elements corresponding to these dangling elements; releasing the memory space occupied by these elements; and updating the adjacency data of the edges and vertices originally involved with these elements. For an edge, if the number of its associated face elements decreases from 3 to 2, it is restored to the normal state; if it decreases to 1 or 0, it is updated as a boundary edge or marked as to be deleted.

[0029] During the discretization of the mesh space, especially after removing dangling elements, some vertices previously connected by dangling elements may not have the necessity to exist independently. Furthermore, due to limitations in discretization resolution or small gaps between adjacent elements, situations may arise where vertices almost overlap spatially but are recorded as two different vertices in the data structure. The process involves traversing all vertices in the first intermediate mesh. For every two different vertices, the Euclidean distance in 3D space is calculated. The Euclidean distance is calculated as follows: the difference in 3D coordinates between the two vertices is taken, the square of each difference is calculated, the three squared values ​​are added together, and the square root of this sum is taken to obtain the Euclidean distance between the two vertices. If the calculated Euclidean distance is equal to zero or less than a preset geometric tolerance value, the two vertices are considered to overlap spatially. This geometric tolerance value is one-thousandth of the side length of the smallest element in the mesh, ranging from 0.001 mm to 0.01 mm, corresponding to a side length of 1 mm to 10 mm for the smallest element. The geometric tolerance value is determined based on the following: the tolerance value should be much smaller than the minimum feature size of the mesh, on the order of one-thousandth, to ensure that only truly overlapping vertices are merged and adjacent vertices that should retain spacing are not mistakenly merged; at the same time, the tolerance value should be greater than the rounding error of floating-point operations in data processing and arithmetic circuits. The relative error of double-precision floating-point numbers is approximately on the order of 10 to the power of -15, to avoid vertices that should overlap being unrecognized due to numerical rounding errors. Vertices determined to be spatially overlapping are merged: the data record of one vertex is retained, the endpoint pointers of all edges with the merged vertex as their endpoint are updated to point to the retained single vertex, and then the data record of the merged vertex is deleted; after merging, a second intermediate mesh with duplicate vertices eliminated is obtained.

[0030] After removing dangling cells and merging duplicate vertices, the data processing and computation circuit performs a connectivity check on the entire second intermediate mesh. The connectivity check is performed as follows: The first volume cell numbered 1 in the second intermediate mesh is selected as the starting cell. A breadth-first search is then performed outwards based on the adjacency relationships of shared edges and shared vertices. The breadth-first search process involves adding the starting cell's number to a first-in-first-out search queue; removing a cell number from the head of the queue and checking all its adjacent cells; adding unvisited adjacent cell numbers to the tail of the queue and marking them as visited; repeating this process until the queue is empty. After the search is complete, all cells marked as visited constitute a connected entity. Cells not marked as visited are identified as isolated cells, meaning that although these cells exist in the mesh data structure, they do not have any connection relationships with the connected entity established through shared edges or shared vertices.

[0031] The data processing and computation circuit removes all isolated cells from the grid data, leaving a connected and complete grid body. This connected grid body satisfies all of the following conditions: each edge in the grid is shared by exactly two face cells, i.e., there are no non-manifold edges; the face cells around each vertex form a simply connected sector, i.e., there are no non-manifold vertices; there are no hanging structures; there are no isolated cells; there are no duplicate vertices; the grid that satisfies all of these conditions is the topologically complete and continuous modified temperature field grid.

[0032] For each branch loop, the data processing and calculation circuit extracts the initial current parameters, i.e., the load current values, of all grid nodes on the conductive path of the cable in that branch loop from the corrected temperature field grid. According to the physical principle of Joule's law, when current flows through a conductor with a certain resistance value, the conductor will generate heat. The heat generated per unit time is proportional to the conductor resistance value and the square of the current flowing through it. The calculation process for the expected heat generation is as follows: First, determine the resistance value per unit length of the cable conductor in that branch loop. This resistance value per unit length is determined based on the resistivity of the cable conductor material and the cross-sectional area of ​​the conductor. The calculation method is to divide the resistivity of the material at the current temperature by the cross-sectional area of ​​the conductor. Taking copper conductor as an example, the resistivity of copper at 20 degrees Celsius is approximately 0.0172 ohms·mm² / m. At 90 degrees Celsius, due to the effect of the temperature coefficient of resistance, the temperature coefficient of resistance of copper is approximately 0.00393 ohms·mm² / m, and the resistivity rises to approximately 0.0220 ohms·mm² / m. For a copper core cable with a cross-sectional area of ​​25 square millimeters, under operating conditions with a conductor temperature of approximately 70 degrees Celsius, dividing the resistivity at this temperature by 25 square millimeters yields a resistance value of approximately 0.00082 ohms per meter.

[0033] Multiply the resistance value per unit length by the square of the load current value of the branch circuit. The specific implementation of the square calculation is to load the load current value into a general-purpose register of the processor, and use the processor's built-in hardware multiplier to perform a multiplication operation between the register value and its own value to obtain the square value. Multiply the product of the resistance value per unit length and the square value by the total length of the conductive path in meters to obtain the expected heat generation of the cable conductor of the branch circuit per unit time under the current load current in watts. For example, for a copper core cable with a cross-sectional area of ​​25 square millimeters and a length of 2 meters, with a load current of 50 amperes: the resistance value per unit length of 0.00082 ohms per meter multiplied by the square of 50 amperes (2500), and then multiplied by 2 meters, yields an expected heat generation of approximately 4.1 watts.

[0034] The data processing and calculation circuit loads the expected heat generation of the cable conductors in each branch loop as an internal heat source condition into the modified temperature field grid. Specifically, for each grid node located on the conductive path of each branch loop in the modified temperature field grid, the expected heat generation corresponding to that conductive path is evenly distributed according to the number of grid nodes on that conductive path. For example, if a branch loop conductive path contains 200 grid nodes and the expected total heat generation is 80 watts, then 80 is divided by 200, and each node is allocated a heat generation share of 0.4 watts, which serves as the internal heat source for these nodes.

[0035] Simultaneously, the data processing and computing circuit sets and corrects the boundary conditions of the temperature field grid; for the grid nodes corresponding to the outer wall surface of the cable branch box, convective heat transfer boundary conditions are set; the convective heat transfer boundary conditions involve two preset parameters: convective heat transfer coefficient and ambient air temperature; the convective heat transfer coefficient ranges from 5 watts per square meter·degree Celsius to 25 watts per square meter·degree Celsius, and its value is based on the following: when the cable branch box is installed in an indoor environment without forced ventilation, i.e., under natural convection conditions, the convective heat transfer coefficient is 5 to 10 watts per square meter·degree Celsius; when the cable branch box is installed outdoors and has natural wind or the box is equipped with... With ventilation holes, the convective heat transfer coefficient is 10 to 25 watts per square meter per degree Celsius. The ambient air temperature is determined based on the installation environment of the cable distribution box: 25 to 35 degrees Celsius for indoor distribution rooms; and 40 to 50 degrees Celsius for outdoor boxes in summer. For the grid nodes corresponding to the air areas inside the cable distribution box that are not occupied by conductors, the thermal conductivity parameters of the air are set, where the thermal conductivity of air is approximately 0.026 watts per meter per degree Celsius, which is the value at a temperature of approximately 30 degrees Celsius under normal pressure. The specific heat capacity of air is approximately 1005 joules per kilogram per degree Celsius.

[0036] Under the combined influence of the aforementioned internal heat source and boundary conditions, the data processing and computation circuit performs a steady-state heat conduction solution process. In the steady-state solution mode, the solution process is based on the principle of heat balance: when each grid node reaches a balance between the heat input from the heat source, the heat transferred between it and its adjacent nodes through heat conduction, and the heat dissipated between it and the external environment through convection, the temperature value of that node is its steady-state temperature value. In actual engineering, due to the presence of different material media inside the cable branch box, including copper or aluminum conductors, insulating materials, air, etc., different media have different thermal conductivity. The thermal conductivity of copper is approximately 401 watts per meter·degree Celsius, the thermal conductivity of polyvinyl chloride insulation material is approximately 0.19 watts per meter·degree Celsius, and the thermal conductivity of air is approximately 0.026 watts per meter·degree Celsius, resulting in a difference of orders of magnitude in the heat conduction rate between different grid nodes. This difference causes a redistribution of the temperature values ​​of each grid node in the corrected temperature field grid. This redistribution of temperature values, where the grid node positions remain unchanged but the temperature attribute values ​​at each node change, is referred to as dynamic deformation of the temperature field in this method.

[0037] After the temperature values ​​are redistributed, the temperature value stored on each grid node reflects the actual temperature at that node's location under the combined effects of the current load current heating, heat conduction from adjacent nodes, and convective heat dissipation from the external environment. This corrected temperature field grid is called a dynamic temperature field topology distribution. The topology distribution emphasizes that the node connections of the grid remain unchanged, but the temperature attribute values ​​of each node have been updated according to the heating and heat transfer processes. As a data structure, this dynamic temperature field topology distribution contains the following elements: a complete set of all grid nodes, each node with an updated temperature value attribute; and the edge connections between nodes, which remain unchanged.

[0038] For each grid node in the dynamic temperature field topology distribution, the data processing and computation circuit performs spatial gradient analysis. The analysis process is as follows: determine the adjacent nodes of the current node in three-dimensional space along three mutually perpendicular coordinate axes, namely the length direction, width direction, and height direction; along the length direction, subtract the temperature value of the current node from the temperature value of the adjacent node in front to obtain the temperature difference in the length direction, and divide the temperature difference by the spatial distance between the current node and the adjacent node, i.e., the straight-line distance between the centers of the two nodes, to obtain the spatial rate of temperature change component in the length direction; along the width direction, subtract the temperature value of the current node from the temperature value of the adjacent node to the right to obtain the temperature difference in the width direction, and divide the temperature difference by the spatial distance between the current node and the adjacent node to the right to obtain the spatial rate of temperature change component in the width direction; along the height direction, subtract the temperature value of the current node from the temperature value of the adjacent node above to obtain the temperature difference in the height direction, and divide the temperature difference by the spatial distance between the current node and the adjacent node above to obtain the spatial rate of temperature change component in the height direction.

[0039] The three components obtained above are combined into a three-dimensional vector. The direction of this vector is determined by the magnitude relationship between the components: the ratio of each component to the square root of the sum of the squares of the three components is calculated, and the ratio is converted into an angle with each coordinate axis using inverse cosine calculation. The three angles together describe the spatial direction of the vector. The length of this vector, i.e., the magnitude, is calculated by taking the square root of the sum of the squares of the three components. The direction of this vector represents the steepest temperature change at the current node, and the length of this vector represents the steepness of the temperature change along this direction.

[0040] The above spatial gradient analysis is performed on each of the grid nodes in the dynamic temperature field topology distribution, generating a temperature gradient vector for each grid node. The set of temperature gradient vectors of all grid nodes constitutes the temperature gradient distribution. This temperature gradient distribution is a dataset containing a ternary combination record for each grid node: node number, used to identify which grid node the gradient vector belongs to; gradient direction, represented by the angle with the three coordinate axes, three floating-point values; gradient magnitude, one floating-point value, in degrees Celsius per meter.

[0041] The process iterates through all node records in the temperature gradient distribution, comparing the temperature gradient magnitude of each node to find the node with the largest temperature gradient magnitude. The direction of the temperature gradient of this node, i.e., the spatial direction pointed to by the gradient vector of this node, is determined as the dominant direction of heat transfer. This dominant direction of heat transfer represents the most important direction of heat transfer in the entire internal space of the cable branch box. Simultaneously, the process iterates through the temperature value attributes of all grid nodes in the dynamic temperature field topology distribution, comparing them one by one to find the grid node with the highest temperature value and the grid node with the lowest temperature value. The number of the grid node with the highest temperature value is determined as the starting node identifier of the heat flow path, and the number of the grid node with the lowest temperature value is determined as the ending node identifier of the heat flow path. The grid node with the highest temperature value is usually located near the cable joint on the conductive path of the branch circuit with the largest load current and the largest expected heat generation, and is the heat source inside the cable branch box. The grid node with the lowest temperature value is usually located near the outer wall of the box or in the air area far away from the heat source conductor, and is the exit point where heat is finally dissipated to the external environment.

[0042] The data processing and computation circuit takes the starting node of the path as the current search starting point. In the grid adjacency relationship data structure of the dynamic temperature field topology distribution, it searches for all adjacent nodes directly connected to the starting node of the path through grid edges, that is, all first-level neighbor nodes of the starting node. For these first-level neighbor nodes, the following two conditions are used for filtering, and both conditions must be satisfied simultaneously: The first condition is the closest directional condition. The calculation process of the angle between each neighbor node and the dominant heat transfer direction is as follows: First, calculate the spatial direction vector of the neighbor node relative to the starting node of the path, which is obtained by subtracting the coordinates of the starting node from each component of the spatial coordinates of the neighbor node. The heat transfer direction is determined by the following steps: The dot product between the spatial direction vector and the dominant heat transfer direction is calculated, which is the sum of the products of the components of the two vectors. This dot product is then divided by the product of the lengths of the two vectors to obtain the cosine of the angle. The lengths of the two vectors are calculated as the square root of the sum of the squares of their respective three components. The cosine value is converted into an angle value using inverse cosine calculation. The smallest angle value indicates that the direction of the neighboring node is closest to the dominant heat transfer direction. Among all first-level neighboring nodes, the neighboring node with the smallest angle is selected as the optimal candidate for direction. The second condition is the temperature downhill condition: the current temperature value of the neighboring node must be lower than the current temperature value of the path's starting node. This condition ensures that the direction of heat transfer is from the high-temperature node to the low-temperature node, because heat always spontaneously transfers from a high-temperature object to a low-temperature object without external work. A neighboring node that satisfies both of the above conditions is determined as the first downstream node. The data processing and computation circuit establishes a directed connection from the path's starting node to the first downstream node in the heat flow path data structure.

[0043] After connecting the first downstream node, the data processing and computation circuit sets this first downstream node as the new current node and repeats the same search process: among the first-level neighbor nodes of the current node, it filters according to the closest direction condition and the temperature downslope condition to determine the next downstream node, and adds a directed connection record from the current node to the next downstream node in the node sequence of the heat flow path; the next downstream node is updated as the current node, and the search continues downstream; the angle calculation in this search process is still performed in the aforementioned manner; this search-connection-forward iterative process continues until one of the following termination conditions is met: Termination condition 1: The current node is the path termination node, at which point the heat flow path has successfully connected from the highest temperature starting end to the lowest temperature termination end, forming a complete heat flow path; Termination condition 2: The temperature values ​​of all first-level neighbor nodes of the current node are not lower than the temperature value of the current node, that is, the current node is the local lowest temperature point, there are no lower temperature downstream nodes to continue to advance, and the current node has not yet reached the global path termination node. At this time, the current node is taken as the actual path termination node, and the heat flow path ends at this position.

[0044] The resulting directed node sequence, consisting of the path's starting node, intermediate nodes arranged in descending temperature order, and the path's ending node, is the heat flow path. This heat flow path is stored as a data structure containing the following elements: a sequence of grid node numbers arranged in progression order, with the first element being the path's starting node number and the last element being the path's ending node number; and a connection record between every two adjacent nodes, including the upstream node number, downstream node number, the temperature difference between the two nodes, and the connection direction vector between the two nodes. The significance of the heat flow path is that it represents the primary path formed inside the cable branch box under the current load current heating condition, from the hottest hotspot area to the coldest heat dissipation area, through which heat is transferred via thermal conduction. The temperature values ​​of adjacent nodes along this path decrease sequentially, reflecting the gradual temperature decay process as heat is transferred from the high-temperature end to the low-temperature end along this path.

[0045] In a preferred embodiment of the present invention, step 4 is performed as follows: Following the progression order of the directed node sequence stored in the heat flow path, starting from the path's starting node number and proceeding sequentially to the path's ending node number, the current temperature value of each grid node on the path is extracted. Starting from the path's starting node, for each pair of adjacent nodes on the heat flow path, i.e., the upstream node and its directly downstream node, the temperature value of the upstream node is subtracted from the temperature value of the downstream node to obtain a temperature difference between adjacent nodes. Since the heat flow path is constructed in a direction of progressively decreasing temperature, the temperature value of the upstream node must be higher than that of the downstream node, and the temperature difference between adjacent nodes is always positive. The above difference extraction operation is repeated for all pairs of adjacent nodes to obtain a set of adjacent node temperature differences arranged in the path progression order; this set of difference sequences is the temperature variation sequence.

[0046] The temperature variation sequence is an ordered data sequence containing floating-point values ​​arranged in the order of the heat flow path. Each value represents the temperature drop between adjacent node pairs, in degrees Celsius. For example, if the heat flow path contains 5 nodes, numbered from the first to the fifth node in the order of advancement, then the temperature variation sequence contains 4 elements: the temperature difference between the first and second nodes, the temperature difference between the second and third nodes, the temperature difference between the third and fourth nodes, and the temperature difference between the fourth and fifth nodes. A difference of 4.5 degrees Celsius in the temperature variation sequence indicates that the temperature drops by 4.5 degrees Celsius for each grid node interval, reflecting strong thermal resistance or significant heat dissipation in that segment. A difference of 0.3 degrees Celsius indicates that the temperature drop is gradual, reflecting relatively smooth heat conduction in that segment.

[0047] Traverse each grid node on the heat flow path and read the current temperature value of that grid node; at the same time, read the rated maximum allowable temperature value from the preset parameter table in the load control unit memory; the rated maximum allowable temperature refers to the highest temperature limit that can be reached at each location inside the cable branch box under long-term continuous operation conditions, in degrees Celsius.

[0048] The preset parameter table is a structured parameter lookup table stored in the read-only memory of the load control unit. It serves as the core parameter benchmark for the entire thermal safety assessment and load regulation calculation. Using the branch circuit number as a unique index, it binds fixed parameters such as cable material, insulation class, rated maximum allowable temperature, and additional temperature rise margin for each circuit. Calibration and data entry are completed before leaving the factory, and the table can be adapted and updated during operation based on on-site cable modifications. The specific construction steps are as follows: Circuit information survey and classification: All branch circuits in the cable branch box are statistically analyzed one by one, and the basic information of each circuit, such as circuit number, rated voltage, cable model, insulation material type, cable cross-section, and load application (general lighting, power, charging pile, etc.), is recorded. Two core circuit types are distinguished: polyvinyl chloride (PVC) insulated circuits and cross-linked polyethylene (XLPE) insulated circuits. Reference temperature parameter calibration: The long-term continuous operation reference temperature of the two types of mainstream insulated cables is determined: the long-term allowable operating temperature of PVC insulated cables is 70℃, and the long-term allowable operating temperature of XLPE insulated cables is 90℃.

[0049] Superimposed standard safety margin values: To mitigate risks from uncontrollable operating conditions such as long-term equipment aging, ambient temperature fluctuations, instantaneous load impacts, and joint overheating, a standard safety redundancy of 5% is uniformly reserved, and the maximum allowable temperature range of the basic rated value is determined: 65℃~75℃ for PVC insulated cables, with a recommended value of 70℃ for engineering projects; 85℃~95℃ for cross-linked polyethylene insulated cables, with a recommended value of 90℃ for engineering projects; Additional temperature rise parameter entry for joints: Considering the contact resistance of cable terminals and joints, which will generate continuous localized additional heating, a separate temperature rise margin parameter for joint positions is added. The joint position is uniformly set to have an additional temperature rise tolerance space of 15℃~25℃ relative to the conductor body, and is included in the special position parameters in the parameter table. For actual operating scenarios, parameter matching and value assignment are completed. Taking a six-branch cable distribution box in a residential area as an example: when all circuits use PVC insulated cables, the rated maximum allowable temperature of all circuits is uniformly fixed at 70℃; when small power circuits and charging pile circuits use cross-linked polyethylene insulated cables, the temperature of such circuits is fixed at 90℃, and the temperature of other ordinary civilian circuits is fixed at 70℃; all parameters are generated into a preset parameter table in the format of circuit number, insulation type, rated maximum temperature, and joint temperature rise margin.

[0050] Thermal safety margin (or simply safety margin) refers to the remaining controllable space between the real-time operating temperature of each grid node and each branch circuit in a cable branch box and the critical over-temperature temperature under current operating conditions. It is a core quantitative indicator for measuring the thermal safety status of equipment, assessing overload tolerance, and guiding load adjustment. The process involves traversing each grid node along the heat flow path and collecting and reading the current operating temperature value of each grid node in real time. Simultaneously, based on the spatial circuit to which the grid node belongs, the rated maximum allowable temperature value of the corresponding branch circuit is retrieved from a preset parameter table. The rated maximum allowable temperature refers to the highest safe temperature limit that each circuit and location inside the cable branch box can reach under long-term continuous steady-state operating conditions. It is the benchmark threshold for determining over-temperature risk and quantifying the safety margin, and is expressed in degrees Celsius.

[0051] For each grid node, the temperature deviation (quantitative value of node thermal safety margin) is calculated using a fixed formula: Temperature deviation = Rated maximum allowable temperature - Node current real-time temperature, in degrees Celsius. When the temperature deviation is positive, it means that the current temperature of the node is below the over-temperature critical value, and the thermal safety margin is sufficient. The larger the value, the greater the remaining temperature rise space, the higher the equipment operation safety, and the more load adjustment redundancy. When the temperature deviation is zero, it means that the node temperature has reached the rated over-temperature critical value and is in a critical dangerous state with no safety margin. When the temperature deviation is negative, it means that the node temperature has exceeded the limit, the safety margin has been overdrawn, and there is a great operational hazard. Load reduction, heat dissipation, and load balancing adjustment operations must be performed immediately.

[0052] For each grid node in the heat flow path, the data processing and computation circuit first determines the branch loop to which the grid node belongs, i.e., which branch loop the grid node is located on or which branch loop directly affects its heat generation; then it obtains the thermal time constant of the cable conductor in that branch loop. The thermal time constant is a physical parameter describing the magnitude of the thermal inertia of the cable conductor. It means that when a constant heating power is applied to the conductor, the time required for the conductor temperature to rise from the initial temperature to approximately 63% of the final steady-state temperature (i.e., 1 minus the reciprocal of the first power of the natural constant), expressed in seconds. A larger thermal time constant indicates stronger thermal inertia in the conductor, because the conductor has a large heat capacity but relatively poor heat dissipation conditions, resulting in a slower temperature rise but also a greater resistance to temperature drops once the temperature is reached. A smaller thermal time constant indicates weaker thermal inertia in the conductor, because the conductor has a small heat capacity but relatively good heat dissipation conditions, resulting in a more rapid temperature response to changes in current.

[0053] The range of values ​​for the thermal time constant and its determination method are as follows: This value is obtained by dividing the conductor's heat capacity by its thermal resistance. The heat capacity is equal to the conductor's mass multiplied by its specific heat capacity. For copper-core PVC insulated cables with a cross-section of 10 to 25 square millimeters, the range of the thermal time constant is 180 to 600 seconds. The basis for this is as follows: For a 10 square millimeter cross-section cable, the conductor mass is approximately 0.089 kg per meter, the specific heat capacity of copper is approximately 385 joules per kilogram per degree Celsius, and the heat capacity is approximately the product of the conductor mass per meter and the specific heat capacity, i.e., 34 joules per degree Celsius. The thermal resistance is limited by the insulation layer to 0.1 to 0.2 degrees Celsius per watt. Dividing the heat capacity by the thermal resistance yields a thermal time constant of 180 to 340 seconds. For a 25 square millimeter cross-section cable, the conductor mass is approximately 0.223 kg per meter, and the heat capacity is approximately 86 joules per meter. For copper core cables with a cross-sectional area of ​​35 to 70 square millimeters, the thermal time constant is 400 to 600 seconds under similar heat dissipation conditions. For 70 square millimeter cross-section cables, the thermal time constant ranges from 600 to 1800 seconds. This range is based on the following: for a 70 square millimeter cross-section cable, the conductor mass is approximately 0.623 kg per meter, the heat capacity is approximately 240 joules per degree Celsius, and the thermal conductivity of the cross-linked polyethylene insulation layer is similar to that of polyvinyl chloride (PVC), thus increasing the thermal time constant to 1200 to 1800 seconds. For copper core cables with a cross-sectional area of ​​95 to 120 square millimeters, the thermal time constant ranges from 1800 to 3600 seconds.

[0054] The specific value of the aforementioned thermal time constant can be calibrated through type testing before delivery. The calibration method is as follows: apply a constant DC heating current to the cable conductor, typically 1.0 to 1.5 times the rated current, install a thermocouple temperature sensor on the surface of the conductor, and record the conductor surface temperature every 5 seconds from the moment the power is applied, continuously recording for no less than 5 times the estimated thermal time constant. Plot the recorded time and temperature data points one by one in a two-dimensional rectangular coordinate system with time as the horizontal axis (in seconds) and temperature as the vertical axis (in degrees Celsius). Then connect each data point sequentially with a smooth spline curve to obtain the time and temperature curve of the cable conductor. From this curve, read the time value corresponding to the temperature rise reaching 63% of the final steady-state temperature rise, which is the calibrated value of the thermal time constant of the cable. The calibrated thermal time constant value is stored in the read-only memory of the load control unit with the circuit number as the index.

[0055] The calculation process for the remaining thermal withstand time of a node is as follows: Divide the temperature deviation of the grid node by the difference between the steady-state temperature that the loop can eventually reach under the same heat output and the current node temperature, which is the total remaining temperature rise potential of the loop under the current load. This yields a proportionality coefficient ranging from 0 to 1. When the temperature deviation is zero, i.e., the node has reached its rated maximum allowable temperature, the ratio is 0. When the temperature deviation equals the total remaining temperature rise potential, the ratio is 1. Multiply this proportionality coefficient by 2.0 times the thermal time constant of the loop. The resulting product is the remaining thermal withstand time of the grid node, in seconds. This 2.0 coefficient is obtained by performing a series expansion of the exponent with the natural constant as the base and taking the first two terms. The selectable values ​​are 1.5 to 3.0, and this implementation uses 2.0. The above calculation process only involves multiplication and division operations and can be completed entirely in the embedded processor of the load control unit using pure numerical calculation.

[0056] The remaining thermal withstand time of all grid nodes is grouped and categorized according to the branch loop to which each grid node belongs. The criteria for determining the affiliation are whether the grid node is located within the spatial area traversed by the conductive path of the cable in that branch loop, including the area occupied by the conductor itself and the heat-affected zone extending outward from the conductor surface by 1 to 2 times the conductor's outer diameter. For a branch loop, its conductive path typically contains dozens to hundreds of grid nodes, depending on the resolution of the discretized grid. Therefore, this branch loop will correspond to a set of data on the remaining thermal withstand time of the nodes.

[0057] In this set of node remaining heat tolerance time data, the minimum value is taken. This minimum value means that on this branch loop, there exists a grid node location with the shortest remaining time before reaching the rated maximum allowable temperature. In other words, this location is the weakest link in the loop with the most strained thermal safety margin. If no load adjustment is made, this location will be the first to reach the over-temperature state. This minimum value is taken as the remaining heat tolerance time of this branch loop, which is the most conservative thermal safety assessment result of this loop under the current load and temperature conditions. The above loop mapping and minimum value operation is performed on each branch loop to finally obtain a set of remaining heat tolerance time values ​​that correspond one-to-one with each branch loop. This set of remaining heat tolerance time values ​​is a data set containing the same number of floating-point values ​​as the total number of branch loops. Each value corresponds to the remaining heat tolerance time of a branch loop, in seconds.

[0058] In a preferred embodiment of the present invention, step 5 is performed as follows: Read the data set of remaining heat withstand time values ​​for all branch circuits; by comparing them pairwise, find the branch circuit with the largest remaining heat withstand time value and mark it as the maximum margin circuit, and record its remaining heat withstand time as the maximum remaining time value; at the same time, find the branch circuit with the smallest remaining heat withstand time value and mark it as the minimum margin circuit, and record its remaining heat withstand time as the minimum remaining time value; the long remaining heat withstand time of the maximum margin circuit indicates that the cable conductor and joint of the circuit still have sufficient time safety margin before overheating under the current load current, and its cooling requirement is not urgent, and it has the potential to increase the load; the short remaining heat withstand time of the minimum margin circuit indicates that the cable conductor and joint of the circuit are very close to overheating, and it is necessary to reduce the load current first to slow down the temperature rise rate.

[0059] The remaining heat withstand times of all branch loops are arranged into an ordered sequence from largest to smallest, forming an uneven distribution of remaining heat withstand times. This uneven distribution of remaining heat withstand times is an ordered data sequence containing the same number of elements as the total number of branch loops, arranged in descending order of remaining heat withstand time. Each element is a binary tuple consisting of the branch loop number and its corresponding remaining heat withstand time value, which is a floating-point number in seconds. The dispersion of elements in this sequence is represented by the ratio between the maximum and minimum values, i.e., using the maximum value to represent the minimum value. The ratio of the maximum remaining heat withstand time to the minimum remaining heat withstand time is used as an indicator of imbalance. If the ratio is greater than 3.0, it indicates a significant difference in thermal safety margin between different branch circuits, with the safest circuit having a remaining heat withstand time more than three times that of the most dangerous circuit, requiring substantial current redistribution. If the ratio is between 1.5 and 3.0, it indicates a moderate degree of imbalance. If the ratio is between 1.0 and 1.5, it indicates that the remaining heat withstand times of each circuit are relatively close, and the heat load distribution of each circuit is relatively balanced, requiring only minor adjustments or even no adjustments at all.

[0060] Based on the uneven distribution of remaining heat withstand time, a current adjustment strategy is determined, with the basic direction of adjustment being bidirectional: branch circuits with remaining heat withstand time lower than the overall average are marked as de-load circuits, and their current adjustment direction is to decrease; branch circuits with remaining heat withstand time higher than the overall average are marked as load-increasing circuits, and their current adjustment direction is to increase; the overall average is the arithmetic mean of the remaining heat withstand time of all circuits, calculated by adding the remaining heat withstand times of all branch circuits and then dividing by the total number of branch circuits; the total current reduced by the de-load circuits is allocated to each load-increasing circuit according to the proportional relationship between the remaining heat withstand times of each load-increasing circuit.

[0061] When determining the specific current adjustment amount for each circuit, two levels of constraints need to be considered. The first level of constraint comes from the established current change boundary constraint set: for each branch circuit, the adjusted load current value must fall within the current allowable adjustment range corresponding to that circuit, i.e., not lower than the lower limit of the adjustable current, nor higher than the upper limit of the adjustable current. The second level of constraint comes from the total load balance constraint: the sum of the adjusted current values ​​of each branch circuit should be equal to the sum of the current values ​​of each branch circuit before adjustment, provided that the total incoming load does not change; or it should meet the total load adjustment target value set by the operator, when it is necessary to actively reduce the total load to cope with the power supply capacity limit of the upper-level distribution network.

[0062] Calculate the arithmetic mean of the remaining heat withstand time for all branch circuits; subtract the remaining heat withstand time of each circuit from this mean. A positive difference indicates that the heat margin is higher than the average level, and a negative difference indicates that the heat margin is lower than the average level. Record this difference as the relative margin. For load-reducing circuits with negative relative margins, take the absolute value of the relative margin. Multiplying this absolute value by a preset current-thermal-time coupling coefficient yields the target current reduction of the load reduction circuit. The current-thermal-time coupling coefficient means how many seconds the remaining thermal withstand time of the circuit can be extended for every 1 ampere reduction in load current. The value of this coupling coefficient ranges from 2 seconds per ampere to 20 seconds per ampere. The selection criteria are as follows: for cables with a cross-section of 10 to 16 square millimeters, due to their small heat capacity and sensitivity to current changes in temperature rise, the coupling coefficient is 2 to 5 seconds per ampere; for cables with a cross-section of 70 to 120 square millimeters, due to their large heat capacity and slower response to current changes in temperature rise, the coupling coefficient is 10 to 20 seconds per ampere; and for cables with a cross-section of 25 to 50 square millimeters, the coupling coefficient is 5 to 10 seconds per ampere.

[0063] The target current reduction amounts of all load reduction circuits are summed to obtain the total reduction amount. The remaining thermal withstand time of all load increase circuits is calculated. The remaining thermal withstand time of each load increase circuit is divided by this sum to obtain the allocation ratio of each load increase circuit. The total reduction amount is multiplied by the allocation ratio of each load increase circuit to obtain the target current increment of each load increase circuit. It is checked whether the target current increment allocated to each load increase circuit exceeds the adjustable current upper limit specified in the current change boundary constraint set of that circuit. If it exceeds, the target current increment of that circuit is truncated to the maximum allowable increment corresponding to the adjustable current upper limit, and the remaining unallocated current reduction amount is redistributed proportionally to other load increase circuits that still have load increase space. The above adjustment strategy is applied to the current load current value of each branch circuit: for load reduction circuits, the target current reduction amount is subtracted from the current load current value; for load increase circuits, the target current increment is added to the current load current value. A set of adjusted current allocation values ​​for each branch circuit is obtained. This set of current distribution values ​​is a data set containing the same number of floating-point values ​​as the total number of branch circuits. Each value corresponds to the target load current value of a branch circuit after completing this round of current redistribution, and the unit is amperes.

[0064] Replace the original load current value with the adjusted current distribution value; multiply the adjusted current distribution value by itself to obtain the square value, then multiply by the resistance per unit length of the loop and the conductive path length to recalculate the expected heat generation of each branch loop cable conductor at the adjusted current level; replace the original expected heat generation with the recalculated expected heat generation, and apply it again as a heat source condition to the modified temperature field grid according to the aforementioned process, keeping the same convective heat transfer boundary conditions, i.e., the convective heat transfer coefficient and ambient temperature remain unchanged, and run the steady-state heat conduction solution again to drive the temperature values ​​of the grid nodes to redistribute, obtaining the updated dynamic... The updated dynamic temperature field topology distribution reflects how the temperature field inside the cable branch box will respond to the current change after the load redistribution is implemented according to the current adjustment scheme, i.e., the load is reduced for circuits with tight thermal margins and increased for circuits with ample thermal margins. Since the adjusted current decreases in the circuits with higher original temperatures and increases in the circuits with lower original temperatures, the updated temperature field topology distribution should show that: the grid node temperature values ​​in the original high-temperature areas have decreased, the grid node temperature values ​​in the original low-temperature areas have increased, the overall temperature distribution tends to be more uniform, and the temperature difference between nodes has decreased.

[0065] For each grid node in the updated dynamic temperature field topology distribution, spatial gradient analysis is performed. Temperature change rate components are calculated along the three coordinate axes, and these three components are combined into a gradient vector to obtain the updated temperature gradient distribution. The direction with the largest temperature gradient amplitude is identified as the dominant heat transfer direction. The highest and lowest temperature nodes are identified as the starting and ending nodes of the updated path. Along the dominant heat transfer direction, direction matching and temperature downhill search are performed according to the aforementioned angle calculation method to obtain the updated heat flow path. The updated heat flow path may differ from the original heat flow path: the elements of its directed node sequence, i.e., the node numbers and order, may differ from the original path; for example, the path direction may shift, or the total path length may shorten. This is because as the temperature distribution becomes more uniform, the temperature difference between high-temperature and low-temperature regions decreases, reducing the driving potential difference of the heat flow path. The updated heat flow path reflects the improvement effect of current redistribution on the heat transfer pattern inside the cable branch box.

[0066] The adjusted current allocation values ​​in the current iteration are compared one by one with the current allocation values ​​in the previous iteration. For the first iteration, which is the current allocation value directly adjusted from the original current value, the current allocation value in the current iteration is compared with the original load current value. For the second and subsequent iterations, the current allocation value in the current iteration is compared with the current allocation value in the previous iteration. The calculation process for the difference in current allocation values ​​is as follows: For each branch loop, the current allocation value in the current iteration is subtracted from the current allocation value in the previous iteration or the original value, and the absolute value of the difference is taken to obtain the absolute value of the current change in that loop. The absolute values ​​of the current changes in all loops are summed to obtain the total current change. The total current change is divided by the total number of branch loops to obtain the average current change. This average current change is used as a measure of the difference in current allocation values ​​between the current iteration and the previous iteration. The average current change is compared with a pre-set convergence threshold; the convergence threshold ranges from 0.05 amperes to 0.5 amperes, and the value of this threshold is determined based on the following: The convergence threshold of 0.05 amps is set as the lower limit because there is a minimum resolvable current change in the current sensor and analog-to-digital converter circuit in the cable branch box. Taking a 12-bit analog-to-digital converter with a current sensor with a range of 200 amps as an example, the total number of quantization levels that the 12-bit analog-to-digital converter can output is 2 multiplied by itself 12 times, which is 4096 quantization levels. Dividing the range of 200 amps by 4096, we get the current resolution corresponding to each quantization level as approximately 0.049 amps. Therefore, setting the threshold at 0.05 amps as the lower limit can ensure that convergence is not misjudged due to insufficient analog-to-digital converter resolution.

[0067] The convergence threshold of 0.5 amps is set as the upper limit because, considering that a load control response cycle is 1 to 10 seconds, the number of iterations within this time window should be controlled within 20 to 50. Therefore, for load balancing scenarios with a total adjustment range on the order of 10 amps, a threshold of 0.1 to 0.5 amps is more reasonable, and convergence can be achieved within 10 to 30 iterations. In the recommended configuration of a cable branch box in a residential area, the convergence threshold is set at 0.2 amps.

[0068] If the calculated average current change is less than the convergence threshold, the iteration is considered to have converged. At this point, the current distribution scheme of each branch circuit has reached the optimal state of thermal load balance. The dataset of all branch circuit current distribution values ​​after optimization and adjustment in the current iteration round is determined as the final current control adjustment amount, that is, the target operating current value corresponding to each branch circuit. The iteration cycle terminates and enters the subsequent stage of hierarchical control command generation and execution.

[0069] If the average current change is greater than or equal to the convergence threshold, the iteration is considered not to have converged, and the current current allocation scheme still has a significant problem of unbalanced heat load. The remaining heat tolerance time dataset of each branch circuit obtained from the current iteration is used as the basic input condition for the new iteration optimization. The unbalanced state of the overall circuit thermal safety margin is reassessed, and the load current of each branch circuit is optimized again based on the difference in heat margin of each circuit to obtain a new current allocation scheme. Subsequently, based on the updated current allocation results, the heat generation of the equipment is re-deduced, the temperature field distribution is reconstructed, the heat flow path and the remaining heat tolerance time of each circuit are updated to complete a complete iteration cycle, continuously approaching the optimal solution for global thermal equilibrium. The above iteration process is continuously repeated, and the current allocation value gradually approaches an optimal solution that makes the remaining heat tolerance time of each branch circuit tend to be balanced with the increase of the number of iterations. Since each iteration searches within the feasible range limited by the current change boundary constraint set, and the direction of each current adjustment is objectively determined by the unbalanced distribution of the remaining heat tolerance time, i.e., reducing the load for those with small margins and increasing the load for those with large margins, the iteration process has the property of monotonically converging. The remaining thermal withstand time of each circuit tends to be more uniform after each iteration, and the corresponding current distribution scheme gradually approaches the optimal thermal safety equilibrium point. It can obtain the current control adjustment amount that satisfies the equilibrium target and all safety constraints within a limited number of iteration steps.

[0070] In a preferred embodiment of the present invention, step 6 is performed as follows: Obtain the current control adjustment amount, which is the target load current value of each branch circuit after load balancing optimization. For each branch circuit, subtract the current load current value of that circuit (i.e., the original load current value collected in step 1) from the current control adjustment amount of that circuit. The difference is the current regulation requirement of that branch circuit, in amperes. A positive difference indicates that the load needs to be increased, and a negative difference indicates that the load needs to be reduced. The absolute value of the difference indicates the magnitude of the current adjustment required. Compare the absolute value of the current regulation requirement of each branch circuit with the control level classification threshold table pre-stored in the load control unit's memory. The control level classification threshold table is a parameter table entered by engineers through the configuration interface during the equipment initialization phase and stored in non-volatile memory using a key-value pair structure. This table contains two sets of threshold records. Each set of records contains three fields: threshold name, threshold value, and threshold unit. The first set of thresholds is named the first tier threshold, with a value range of 1.0 amps to 5.0 amps, and a recommended value of 3.0 amps. The threshold unit is amps, and the determination is based on the fact that below this range, the heat generation change is less than 10%. The second set of thresholds is named the second tier threshold, with a value range of 10.0 amps to 20.0 amps, and a recommended value of 15.0 amps. The threshold unit is amps, and the determination is based on the fact that below this range, the adjustment can be completed within the existing load branch switching capacity.

[0071] The comparison process is as follows: If the absolute value of the current regulation demand is less than or equal to the first threshold, the matching control level is 1, i.e., fine-tuning level. The corresponding adjustment method is to achieve small power regulation by adjusting the conduction angle of the power device or adjusting the duty cycle of the inverter output. If the absolute value of the current regulation demand is greater than the first threshold and less than or equal to the second threshold, the matching control level is 2, i.e., medium regulation level. The corresponding adjustment method is to achieve medium power regulation by switching the secondary load branch or adjusting the output voltage level of the on-load tap changer. If the absolute value of the current regulation demand is greater than the second threshold, the matching control level is 3, i.e., readjustment level. The corresponding adjustment method is to achieve large power regulation by switching the backup power supply channel, starting the auxiliary heat dissipation device, or performing an emergency load transfer.

[0072] After determining the control level of each branch loop, a hierarchical response control command is generated. Each control command is a structured message containing the following information field elements: the target branch loop number; the target load current value of the loop, i.e., the current control adjustment amount, in amperes, rounded to one decimal place; the control level identifier of the loop, which takes the value 1, 2, or 3, corresponding to fine adjustment, medium adjustment, and readjustment, respectively; and the specific operation type that the loop execution terminal needs to perform, which takes the value of increasing or decreasing load depending on the sign of the current adjustment demand, and takes the value of holding if the current adjustment demand is zero.

[0073] The control commands are encoded and packaged according to the communication protocol agreed upon between the load control unit and the execution terminal. The above information fields are sequentially encoded into the register fields of the communication message, with each information field corresponding to a 16-bit register. After adding the slave address, starting address, number of registers, and cyclic redundancy check protocol fields, a complete communication message is formed. This communication message is sent to the execution terminal corresponding to each branch circuit through the communication interface circuit of the load control unit via a serial communication bus using differential signal transmission. The execution terminal then performs the corresponding load current adjustment action according to the control level identifier and operation type in the message.

[0074] like Figure 2 As shown, embodiments of the present invention also provide an adaptive load control system based on a cable branch box, comprising: The data acquisition module is used to collect the load current value and cable joint temperature value of each branch circuit in real time; The current boundary constraint generation module is used to determine the maximum allowable increase or decrease of current in each circuit in the future based on the load current value and its rate of change, which serves as the upper and lower limits of the adjustable current, and is summarized to form a set of current change boundary constraints. The analysis module is used to construct an initial grid of the cable temperature field based on the collected temperature values, load current values, and current change boundary constraint sets; calculate the corresponding expected heat generation using the load current values, drive the initial grid of the cable temperature field to undergo dynamic deformation, obtain the current dynamic temperature field topology distribution, and determine the heat flow path based on the temperature gradient of the current dynamic temperature field topology distribution; The heat tolerance time assessment module is used to compare the temperature of adjacent nodes along the heat flow path, determine the degree of deviation of each location from the rated maximum allowable temperature, and map the degree of deviation to the remaining heat tolerance time of each loop. The optimization module is used to iteratively adjust the current distribution value of each loop under the constraint of the current change boundary constraint set, with the goal of balancing the remaining heat tolerance time. After each iteration adjustment, the expected heat generation is recalculated based on the adjusted current value and the temperature field is deformed. The dynamic temperature field topology distribution, heat flow path and remaining heat tolerance time are updated until the difference between the current distribution values ​​of two adjacent iterations is less than the set threshold. The final current distribution value is used as the current control adjustment amount. The execution module is used to match the corresponding control level according to the current control adjustment amount, generate hierarchical response control commands and send them to the execution terminal for action response.

[0075] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0076] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0077] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0078] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for adaptive load control based on cable branch boxes, characterized in that, The method includes: Step 1: Real-time acquisition of load current values ​​and cable joint temperature values ​​for each branch circuit; Step 2: Based on the load current value and its rate of change, determine the maximum allowable increase or decrease of current in each circuit in the future, which will serve as the upper and lower limits of the adjustable current. These limits will be summarized to form a set of current change boundary constraints. Step 3: Based on the collected temperature value, load current value, and current change boundary constraint set, construct the initial grid of the cable temperature field; calculate the corresponding expected heat generation using the load current value, drive the initial grid of the cable temperature field to undergo dynamic deformation, obtain the current dynamic temperature field topology distribution, and determine the heat flow path based on the temperature gradient of the current dynamic temperature field topology distribution. Step 4: Compare the temperatures of adjacent nodes along the heat flow path to determine the degree of deviation of each location from the rated maximum allowable temperature, and map the degree of deviation to the remaining heat tolerance time of each loop. Step 5: Under the constraints of the current change boundary constraint set, with the goal of balancing the remaining heat tolerance time, iteratively adjust the current distribution value of each loop; after each iteration adjustment, recalculate the expected heat generation based on the adjusted current value and drive the temperature field deformation, update the dynamic temperature field topology distribution, heat flow path and remaining heat tolerance time, until the difference between the current distribution values ​​of two adjacent iterations is less than the set threshold, and use the final current distribution value as the current control adjustment amount. Step 6: Match the corresponding control level according to the current control adjustment amount, generate a graded response control command and send it to the execution terminal for action response.

2. The adaptive load control method based on cable branch box according to claim 1, characterized in that, Based on the load current value and its rate of change, determine the maximum allowable increase or decrease in current for each circuit in the future, which serves as the upper and lower limits for current adjustment. These limits are then summarized to form a set of current change boundary constraints, including: The load current values ​​of each branch circuit collected in real time are compared in time series to obtain the current change rate of each branch circuit; The current change rate is matched with the preset maximum allowable current change rate threshold of each branch circuit, and the upper limit of the current adjustment rate is determined based on the matching result. Based on the upper limit of the current adjustment rate and the preset load control response period, the maximum increase and maximum decrease of current in each branch circuit within the load control response period are obtained. Increase the current value by the maximum current increase to obtain the adjustable upper limit of the current for each branch circuit; decrease the current value by the maximum current decrease to obtain the adjustable lower limit of the current for each branch circuit. The upper and lower adjustable current limits of all branch circuits are collected to form a set of current change boundary constraints represented by the allowable adjustment range of current for each branch circuit.

3. The adaptive load control method based on cable branch box according to claim 2, characterized in that, Step 3 includes: The internal space of the cable branch box is discretized into multiple grid nodes. The temperature value of each cable joint is assigned to the initial temperature value of the corresponding grid node. The load current value is assigned to the initial current parameter of the corresponding conductor. The adjustable upper limit and adjustable lower limit of the current of each branch circuit in the current change boundary constraint set are set as the current boundary conditions of the corresponding conductor to construct the initial grid of the cable temperature field. The initial temperature field mesh of the cable is subjected to topology integrity testing to obtain the corrected temperature field mesh; Based on the initial current parameters of the conductors corresponding to each branch loop in the corrected temperature field grid, determine the expected heat generation of the cable conductors in each branch loop when current passes through them; The expected heat generation is applied as a heat source condition to the modified temperature field grid, driving the redistribution of temperature values ​​of each grid node in the modified temperature field grid, resulting in a dynamic temperature field topology distribution that reflects the current heating state. Spatial gradient analysis is performed on the temperature values ​​of each grid node in the dynamic temperature field topology distribution to extract the direction and steepness of temperature change of each grid node, thus forming a temperature gradient distribution. Based on the temperature gradient distribution, the direction of the most drastic temperature change is identified as the dominant direction of heat transfer. A path is then established from the high-temperature grid node to the low-temperature grid node along the dominant direction of heat transfer to obtain the heat flow path.

4. The adaptive load control method based on cable branch box according to claim 3, characterized in that, The initial temperature field mesh of the cable is topologically inspected to obtain a corrected temperature field mesh, including: Traverse all edges and vertices in the initial mesh of the cable temperature field, and identify non-manifold edges and non-manifold vertices in the initial mesh of the cable temperature field; The suspension elements associated with non-manifold edges and non-manifold vertices are removed from the initial mesh of the cable temperature field to obtain the first intermediate mesh with the suspension structure removed; Vertex overlap detection is performed on the first intermediate grid, and multiple vertices with overlapping spatial positions in the first intermediate grid are merged into a single vertex to obtain the second intermediate grid with duplicate vertices eliminated. Isolated cells in the second intermediate grid that are not connected to other cells through shared edges are detected and removed to obtain a corrected temperature field grid with a complete and continuous topology.

5. The adaptive load control method based on cable branch box according to claim 4, characterized in that, Based on the temperature gradient distribution, the direction of the most drastic temperature change is identified as the dominant heat transfer direction. A path is then established along this dominant heat transfer direction from the high-temperature grid node to the low-temperature grid node to obtain the heat flow path, which includes: The temperature gradient direction and magnitude of each grid node are extracted from the temperature gradient distribution. The direction with the largest temperature gradient magnitude among all grid nodes is taken as the dominant direction of heat transfer. In the dynamic temperature field topology distribution, the grid node with the highest temperature value is determined as the path start node, and the grid node with the lowest temperature value is determined as the path end node. Starting from the path start node, along the dominant heat transfer direction, search for the first downstream node that is adjacent to the path start node and has a lower temperature value than the path start node, and connect the path start node to the first downstream node. Take the first downstream node as the current node, continue along the dominant direction of heat transfer, search for the next downstream node that is adjacent to the current node and has a lower temperature value than the current node, and connect them. Repeat the above search and connection process until the path termination node is reached to obtain the heat flow path.

6. The adaptive load control method based on cable branch box according to claim 5, characterized in that, By comparing the temperatures of adjacent nodes along the heat flow path, the degree of deviation of each location from the rated maximum allowable temperature is determined, and the degree of deviation is mapped to the remaining heat tolerance time of each loop, including: Along the heat flow path, starting from the path start node, the temperature values ​​of every two adjacent grid nodes on the heat flow path are extracted sequentially to form a temperature variation sequence composed of the temperature difference between adjacent nodes. From the temperature change sequence, the current temperature value of each grid node is extracted point by point, and the current temperature value of each grid node is compared with the preset rated maximum allowable temperature to obtain the temperature deviation of each grid node. Based on the temperature deviation and the thermal time constant of the cable conductor of the branch circuit to which each grid node belongs, the withstand time required for each grid node to reach the rated maximum allowable temperature under the current temperature condition is calculated, which is used as the node's remaining thermal withstand time. The remaining thermal withstand time of each node is mapped to the corresponding branch loop according to the relationship between each grid node and each branch loop. The minimum value of the remaining thermal withstand time of multiple grid nodes belonging to the same branch loop is taken to obtain the remaining thermal withstand time of each branch loop.

7. The adaptive load control method based on cable branch box according to claim 6, characterized in that, Step 5 includes: Compare the remaining heat withstand time of each branch circuit, identify the branch circuit corresponding to the maximum and minimum remaining heat withstand time, and form an uneven distribution of remaining heat withstand time. Within the allowable adjustment range of the current in each branch circuit defined by the current change boundary constraint set, the direction and amount of current adjustment to reduce the imbalance are determined according to the uneven distribution of the remaining heat withstand time. The load current value of each branch circuit is then adjusted to obtain the adjusted current distribution value of each branch circuit. Based on the adjusted current distribution values ​​of each branch circuit, the expected heat generation of the cable conductors of each branch circuit is re-determined, driving the temperature field deformation to obtain the updated dynamic temperature field topology distribution. Based on the updated dynamic temperature field topology distribution, a new spatial gradient analysis is performed to determine the updated temperature gradient distribution, and the updated heat flow path is determined based on the updated temperature gradient distribution. By comparing the temperatures of adjacent nodes along the updated heat flow path, the remaining heat tolerance time of each branch loop is re-determined, and the updated remaining heat tolerance time of each branch loop is obtained. The adjusted current distribution values ​​of each branch circuit are compared with the current distribution values ​​of the previous iteration. If the difference is less than a set threshold, the adjusted current distribution values ​​of each branch circuit are used as the current control adjustment amount; otherwise, the updated remaining heat tolerance time of each branch circuit is used as the input of the comparison for the next iteration.

8. A cable branch box adaptive load control system, wherein the system implements the method as described in any one of claims 1 to 7, characterized in that, include: The data acquisition module is used to collect the load current value and cable joint temperature value of each branch circuit in real time; The current boundary constraint generation module is used to determine the maximum allowable increase or decrease of current in each circuit in the future based on the load current value and its rate of change, which serves as the upper and lower limits of the adjustable current, and is summarized to form a set of current change boundary constraints. The analysis module is used to construct the initial mesh of the cable temperature field based on the collected temperature values, load current values, and current change boundary constraint sets; The expected heat generation is calculated using the load current value, which drives the initial grid of the cable temperature field to undergo dynamic deformation, resulting in the current dynamic temperature field topology distribution. The heat flow path is then determined based on the temperature gradient of the current dynamic temperature field topology distribution. The heat tolerance time assessment module is used to compare the temperature of adjacent nodes along the heat flow path, determine the degree of deviation of each location from the rated maximum allowable temperature, and map the degree of deviation to the remaining heat tolerance time of each loop. The optimization module is used to iteratively adjust the current distribution value of each loop under the constraints of the current variation boundary constraint set, with the goal of balancing the remaining thermal withstand time. After each iteration adjustment, the expected heat generation is recalculated based on the adjusted current value and the temperature field is deformed. The dynamic temperature field topology distribution, heat flow path and remaining heat tolerance time are updated until the difference between the current allocation values ​​of two adjacent iterations is less than the set threshold. The final current allocation value is then used as the current control adjustment amount. The execution module is used to match the corresponding control level according to the current control adjustment amount, generate hierarchical response control commands and send them to the execution terminal for action response.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images