A Method and System for Automated Power Distribution Control of JP Cabinets Based on Big Data Cloud-Edge Collaboration

Abstract

This invention belongs to the field of power grid Internet of Things (IoT) data acquisition technology, specifically involving a method and system for automated power distribution control of JP cabinets based on big data cloud-edge collaboration. The method includes: a cloud server cluster extracting the equivalent heat capacity and internal thermal impedance of the cabinet based on a time-series environmental feature matrix, pre-deriving and constructing a dynamic thermal simulation formula, offline generating a mapping array and heat injection intrinsic coefficients, and distributing them; an edge computing gateway extracting the thermal conductivity admittance comprehensive factor based on the external blowing wind speed, equivalently converting the radiant heat flux intensity into the square of the radiant equivalent current, and inputting it into the dynamic thermal simulation formula to calculate and output the current predicted temperature; the edge computing gateway using the current predicted temperature to perform simulation to obtain the temperature rise rate, and initiating a variable structure flexible automated protection intervention mechanism. This invention can promptly intervene in abnormal heat accumulation processes to improve the bottom line of equipment operation safety.

Description

Technical Field

[0001] This invention relates to the field of power grid Internet of Things (IoT) data acquisition technology. More specifically, this invention relates to a method and system for automated power distribution control of JP (Power Distribution Unit) cabinets based on big data cloud-edge collaboration. Background Technology

[0002] During the operation of the power distribution network, the integrated distribution box, or JP cabinet, is a key outdoor device that integrates multiple functions such as power distribution, metering, compensation, and protection. Because the JP cabinet is exposed to the open environment for a long time, its internal electronic components generate a lot of heat when working. This heat needs to pass through the cabinet surface and be dissipated with the help of the surrounding airflow. In actual operation, this heat dissipation effect depends not only on the magnitude of the current inside the cabinet, but also on the drastic influence of sunlight, ambient temperature, and wind speed. Especially in the summer when the temperature is high and the power load is at its peak, if the heat inside the cabinet cannot be dissipated in time, it is very easy to cause the internal parts to burn out or cause serious power outages.

[0003] Chinese patent application CN113098711A discloses a cloud-edge collaborative power distribution IoT CPS management method and system. This patent application utilizes a collaborative architecture between the cloud and the edge to manage and monitor power equipment. However, this patent application focuses on macroscopic data flow and monitoring of equipment operating status, without analyzing the specific patterns of heat dissipation in complex outdoor environments. This results in control commands being unable to keep up with actual temperature changes when dealing with JP (Power Distribution Unit) cabinets, which are highly sensitive to weather conditions.

[0004] Chinese patent document CN119705204B, authorized by patent number CN119705204B, provides a cloud-edge collaborative method for energy-saving control of AC power regulators. This patent uses a complex search algorithm run in the cloud to find power-saving control parameters, which are then executed at the edge. However, the main purpose of this patent is to reduce power loss by adjusting the on / off time of the current; its calculation logic does not consider the correlation between heat accumulation inside the cabinet and external wind and sunlight, making it difficult to use for the safe control of JP (Power Switchgear) cabinets under extreme weather conditions.

[0005] Chinese patent documents with authorization announcement number CN112217879B and application publication number CN121906798A respectively relate to the allocation of computing resource weights in the distribution Internet of Things (IoT) and the coordinated scheduling of large-scale power grids. These patent documents mainly address the issues of how to allocate network signals in power systems and how to balance multiple regional power grids. However, when dealing with the heat dissipation control of a single JP cabinet, these methods cannot perceive the impact of wind speed on the cabinet's cooling rate, nor can they identify the additional heat brought to the cabinet by direct sunlight, resulting in a discrepancy between the control logic and the actual physical conditions.

[0006] In existing technologies, although some solutions attempt to control power distribution equipment through cloud and edge gateway cooperation, most of these solutions rely solely on fixed current limits or readings from the thermometer inside the cabinet to make a response. Since JP cabinets are located in complex outdoor environments, changes in wind speed directly alter the rate at which the air carries away heat from the cabinet surface. This change is highly random, and existing control methods typically assume that heat dissipation efficiency is a fixed value, which can easily lead to significant deviations when calculating temperature rise. However, if all meteorological data is transmitted to the cloud for complex model simulations in pursuit of calculation accuracy, the latency caused by network transmission can prevent timely current cutoff when the equipment is about to overheat. On the other hand, if calculations are performed solely by the tiny chip installed locally in the cabinet, the limited performance of the local chip cannot run the complex formulas involving changes in wind speed and the effects of sunlight. This results in existing control systems often acting too late due to inaccurate predictions when facing intense midday sun or sudden decreases in wind speed, or causing unnecessary power outages due to blind protection, seriously affecting the reliability of JP cabinet automated control. Summary of the Invention

[0007] To address the problem that existing cloud-edge collaborative control architectures suffer from weak edge node computing power and cloud communication delays, making it impossible to calculate in real time the drastic changes in outdoor wind speed and sunlight affecting the cabinet's actual heat dissipation capacity, resulting in a significant deviation between the system's predicted internal temperature and the actual situation, and ultimately causing the JP cabinet's automated overheat protection action to be often delayed or erroneously triggered, this invention provides solutions in the following aspects.

[0008] In a first aspect, the present invention provides an automated power distribution control method for JP cabinets based on big data cloud-edge collaboration, including communication and transmission of feature data between a cloud server cluster and an edge computing gateway, comprising: the cloud server cluster extracting the equivalent heat capacity and internal thermal impedance of the cabinet based on a time-series environmental feature matrix, pre-deriving and constructing a dynamic thermal deduction formula, generating a mapping array and heat injection intrinsic coefficients offline and sending them to the edge computing gateway, wherein the mapping array contains a thermal conductivity admittance comprehensive factor; the edge computing gateway extracting the corresponding thermal conductivity admittance comprehensive factor from the mapping array based on the acquired external blowing wind speed, converting the radiative heat flux intensity into the equivalent square of the radiative current according to the spatial geometric incident parameters, inputting the equivalent square of the radiative current, the external ambient temperature, and the main circuit current into the dynamic thermal deduction formula, and calculating and outputting the current predicted temperature; the edge computing gateway performing time difference slope deduction and extraction calculation using the current predicted temperature of the previous discrete calculation cycle to obtain the temperature rise rate inside the metal sealed cavity, and initiating a variable structure flexible automated protection intervention action execution mechanism based on the current predicted temperature and the temperature rise rate inside the metal sealed cavity.

[0009] Preferably, the edge computing gateway reads historical wind speed sequences and historical continuous environmental temperature sequences over a fixed time span; it also reads historical load current sequences and historical continuous cabinet temperature sequences simultaneously; and it aligns and splices the historical wind speed sequences, historical continuous environmental temperature sequences, historical load current sequences, and historical continuous cabinet temperature sequences according to time characteristics to construct the time-series environmental feature matrix.

[0010] Preferably, the dynamic thermal derivation formula is: In the formula, This is the current sampling time; For historical sampling moments; The current predicted temperature; Predicting temperatures based on historical data; The thermal conductivity-admittance comprehensive factor; External ambient temperature; The intrinsic coefficients for heat injection; This represents the main circuit current at the current sampling moment; The square of the radiation equivalent current.

[0011] Preferably, the formula for calculating the thermal conductivity admittance composite factor is: In the formula, The thermal conductivity-admittance comprehensive factor; The sampling period time; The external wind speed at the current sampling moment; External wind speed at the current sampling time Real-time controlled dynamic wind speed mapping item; The equivalent heat capacity of the cabinet; This represents the surface area for heat dissipation.

[0012] Preferably, the formula for calculating the intrinsic coefficient of heat injection is: In the formula, The intrinsic coefficients for heat injection; The sampling period time; Internal heating resistance; The equivalent heat capacity of the cabinet.

[0013] Preferably, the formula for calculating the square of the radiation equivalent current is: In the formula, Represents the square of the radiation equivalent current; Represents the intensity of radiative heat flux; Represents the surface area for heat dissipation; Represents the cosine function; Represents the angle of incidence in space; Represents the spectral absorbance of the coating; This represents the reference heating impedance.

[0014] Preferably, the variable structure flexible automated protection intervention action execution mechanism is pre-configured with a stable early warning threshold, a sudden alarm threshold, a first temperature threshold, and a second temperature threshold; the first temperature threshold is limited to be greater than the second temperature threshold, and the sudden alarm threshold is greater than the stable early warning threshold.

[0015] Preferably, when the absolute value of the temperature rise rate inside the metal sealed cavity is less than the stable warning threshold, the edge computing gateway confirms that the external thermal boundary layer heat dissipation flow path is in a stable state; when the current predicted temperature is greater than the first temperature threshold, a high-level duty cycle square wave drive signal is output to the pulse width modulation peripheral channel to drive the forced cooling exhaust fan motor on the top of the distribution box to perform exhaust heat removal action; after starting the forced cooling exhaust fan motor on the top of the distribution box, in response to the detection result that the current predicted temperature shows a continuous upward trend, the edge computing gateway sends a tripping command to the tripping electric operating mechanism of the intelligent molded case circuit breaker through the underlying asynchronous serial communication bus interface to cut off the pre-configured secondary priority load power supply output branch.

[0016] Preferably, when the current predicted temperature is less than the first temperature threshold and the temperature rise rate inside the metal sealed cavity is greater than the sudden alarm threshold, the edge computing gateway determines that the external convective thermal boundary layer is in an aerodynamic viscous hindrance physical state; and dynamically lowers the fan start trigger threshold benchmark value to the second temperature threshold to start the top exhaust fan motor with a preset maximum duty cycle.

[0017] In a second aspect, the present invention provides an automated power distribution control system for JP cabinets based on big data cloud-edge collaboration, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned automated power distribution control method for JP cabinets based on big data cloud-edge collaboration is implemented.

[0018] By adopting the above technical solution, the above-mentioned JP cabinet automated power distribution control method based on big data cloud-edge collaboration is generated into a computer program and stored in a memory for loading and execution by a processor. Terminal devices are then made based on the memory and processor for convenient use.

[0019] The beneficial effects of this invention are as follows:

[0020] This invention controls a cloud server cluster to extract the equivalent heat capacity and internal thermal impedance of the cabinet, and pre-derives and constructs a dynamic thermal simulation formula to solve the high-computational-resource-consuming modeling task offline in the background. The edge computing gateway extracts the thermal conductivity and admittance comprehensive factor based on the external blowing wind speed, converts the radiant heat flux intensity into the square of the radiant equivalent current, and combines the external ambient temperature and the main circuit current input to calculate and output the current predicted temperature in the dynamic thermal simulation formula. The edge computing gateway uses the current predicted temperature to perform time difference calculation to obtain the temperature rise rate, and combines the current predicted temperature and temperature rise rate as dual parameters to activate the variable structure flexible automatic protection intervention mechanism. This collaborative control architecture uses mathematical equivalent transformation operations to homogenize and reduce the dimension of cross-physical field environmental disturbance variables, enabling the constrained microcontroller to call multi-source meteorological characteristics based on conventional instructions for thermodynamic look-ahead evolution. This overcomes the environmental response limitations of traditional methods that use constant heat dissipation coefficients and a single static temperature upper limit, and executes protection measures before dangerous high temperatures arrive to improve the bottom line of power distribution equipment operation safety. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the automated power distribution control method for JP cabinets based on big data cloud-edge collaboration in this invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] This invention discloses an automated power distribution control method for JP cabinets based on big data cloud-edge collaboration, referring to... Figure 1 This includes steps S1-S3:

[0025] S1. The cloud server cluster identifies thermodynamic features based on the time-series environmental feature matrix, and derives and constructs a dynamic thermal derivation formula by combining the unsteady boundary layer theory, and then adds a thermal conductance and admittance mapping array and thermal injection eigencoefficients.

[0026] It should be noted that, considering the physical configuration limits of the computing power resources of the built-in microprocessor computing unit of the edge computing gateway, the underlying microcontroller node has a potential delay risk of computation cycle timeout when performing real-time matrix solving of nonlinear fluid dynamics partial differential equations. This invention transforms the continuous fluid thermodynamic equations with high floating-point computation complexity into a discretized one-dimensional feature addressing mapping array by allocating internal storage space and offline computing resources of the cloud server cluster. By configuring the local addressing read-only memory space of the edge computing gateway to reduce the machine instruction cycle execution response time, the physical simulation mechanism model is mapped and configured as a linear read-only query process, so that the automated power distribution control thermal simulation scheme meets the constraints of the underlying microcontroller hardware resources.

[0027] Specifically, the control edge computing gateway microprocessor uses an external physical asynchronous serial communication bus interface to periodically read historical wind speed sequences and historical continuous environmental temperature sequences from the sensors at the micro weather station installed on top of the distribution box, with a fixed time span as the single historical data aggregation and archiving cycle. Simultaneously, the control edge computing gateway microprocessor reads historical load current sequences from the Hall current sensors in the main busbar copper busbar area and historical continuous temperature sequences inside the cabinet from the internal temperature sensors. The control edge computing gateway microprocessor aligns and splices the historical wind speed sequences, historical continuous environmental temperature sequences, historical load current sequences, and historical continuous temperature sequences inside the cabinet according to time characteristics to construct a time-series environmental feature matrix. This time-series environmental feature matrix is ​​then losslessly uploaded to the cloud server cluster database via the built-in wireless communication baseband RF module.

[0028] Furthermore, the logical computing unit of the cloud server cluster is controlled to retrieve the static parameters of the three-dimensional spatial geometric features and the static parameters of the inherent thermal conductivity of the metal cabinet surface of the distribution box, which are pre-configured in the cloud server cluster database. Combined with the uploaded time-series environmental feature matrix, thermodynamic feature parameters are identified, and the equivalent heat capacity and internal thermal impedance of the cabinet are extracted. During the offline configuration phase, the cloud server is controlled to pre-derive and construct a dynamic thermal simulation formula adapted to the hardware computing resource characteristics of the underlying edge computing microcontroller node, based on the conventional lumped parameter thermodynamic model calculation rules, the classical Nusselt convection heat transfer criterion formula, and the first-order Euler numerical discretization approximation expansion rule. The derivation process is as follows:

[0029] The first step is to construct the thermodynamic differential equation of the cavity in the time domain based on the principle of the first law of thermodynamics. The equivalent transient temperature rise rate inside the sealed metal cavity depends on the algebraic difference between the total input heat power and the total dissipated heat power.

[0030]

[0031] In the formula, The equivalent heat capacity of the cabinet is expressed in joules per degree Celsius (J / °C). The internal temperature of the cavity is measured in degrees Celsius (°C). For continuous time extrapolation, the dimension variable is measured in seconds (s). This is the calculus of the rate of temperature change in a continuous space, with dimensions in degrees Celsius per second (°C / s). The total input thermal power is expressed in watts (W). The total heat dissipation power is expressed in watts (W). The differential thermodynamic equation of the cavity characterizes the rate of change of the internal temperature of the sealed metal cavity with time: when the total input heat power is greater than the total dissipation heat power, the algebraic difference is positive, which causes the internal temperature of the cavity to rise with the increase of the dimension variable of continuous time extension, reflecting the process of heat accumulation inside the distribution box.

[0032] The second step involves decomposing the total input heat power and the total dissipated heat power based on thermodynamic conversion characteristics. Specifically, the total input heat power is equivalent to the internal heating resistance. Multiply by the square of the main circuit current. Square of the radiation equivalent current The linear superposition of the two ,Right now Simultaneously, according to Newton's law of cooling in classical heat transfer, the total dissipated heat power is equivalent to the heat dissipation surface area. Multiply by the convective heat dissipation coefficient With the internal temperature of the cavity and external ambient temperature algebraic difference The product of the three is Substituting all the analytical terms into the cavity thermodynamic differential equation, the continuous-time differential equation is derived as follows:

[0033]

[0034] In the formula, The internal heating resistance is in ohms (Ω). It is the main circuit current, and its dimension is ampere (A). It is the square of the radiation equivalent current, with dimensions in the square of amperes (A²). The convective heat dissipation coefficient is expressed in watts per square meter per degree Celsius (W / (m²·°C)). The heat dissipation surface area is measured in square meters (m²). The internal temperature of the cavity is measured in degrees Celsius (°C). Let be the external ambient temperature, with the dimension of degrees Celsius (°C). This continuous-time differential equation reflects the direct influence of specific physical parameters on the temperature rise: an increase in the main circuit current or an increase in the square of the radiation equivalent current will lead to an increase in the total input heat power, thereby increasing the rate of change of the cavity internal temperature with the continuous-time extrapolation dimension variable. Conversely, an increase in the algebraic difference between the cavity internal temperature and the external ambient temperature will lead to an increase in the total dissipated heat power, thereby decreasing the rate of change of the cavity internal temperature with the continuous-time extrapolation dimension variable.

[0035] The third step involves introducing an aerodynamic boundary layer hydrodynamic compensation term into the continuous-time differential equation to reduce the boundary error of the intrinsic parameter model. Specifically, this involves adjusting the traditional constant convective heat dissipation coefficient... Replace with a dynamic wind speed mapping term that is controlled in real time by external sweeping wind speed. The equation of state for calculating the interaction of heat flow and fluid flow is derived as follows:

[0036]

[0037] In the formula, The external wind speed is measured in meters per second (m / s). For the speed of external sweeping wind The dynamic mapping term of wind speed under real-time control has the dimension of watts per square meter in degrees Celsius (W / (m²·°C)). The state equation for the calculation of the flow-heat cross-action reflects the nonlinear effect of wind speed on heat dissipation: an increase in the external blowing wind speed will lead to an increase in the dynamic mapping term of wind speed under real-time control of the external blowing wind speed, which characterizes the thinning of the microscopic physical thickness of the thermal boundary layer on the surface of the distribution box metal cabinet, thereby enhancing the ability to radiate and diffuse heat to the external natural environment.

[0038] The fourth step involves performing a first-order Euler numerical discretization approximation expansion based on the hardware computing resource characteristics of the edge computing gateway microprocessor unit. Specifically, the sampling period is set, and the differential integral term of the temperature change rate in the continuous space is... Linear expansion to the current predicted temperature Compared with historical predicted temperatures algebraic difference Divide by sampling period time Substituting the first-order discrete approximate algebraic terms into the state equation for calculating the interaction between flow and heat, and adding discrete time series subscripts to all physical state variables, the discretized numerical calculation equation is derived as follows:

[0039]

[0040] In the formula, This is the current sampling time; For historical sampling moments; The current predicted temperature is expressed in degrees Celsius (°C). Historical predicted temperatures, in degrees Celsius (°C); The sampling period time; The current in the main circuit at the current sampling time is expressed in amperes (A). The external wind speed at the current sampling moment is measured in meters per second (m / s). This discretized numerical calculation equation transforms the differential evolution process in the continuous time domain into a differential form that can be executed within the clock cycle of the microcontroller. This allows the calculation of the current predicted temperature to depend on the historical predicted temperature and the discrete sensor readings of the current period, thus realizing the executability of the algorithm in the digital control system.

[0041] The fifth step involves performing identity rearrangement operations on the discretized numerical calculation equations to derive the following theoretical formula for mechanistic temperature prediction suitable for the reduced instruction set architecture of edge computing gateway microprocessors:

[0042]

[0043] In the formula, This is the current sampling time; For historical sampling moments; The current predicted temperature is expressed in degrees Celsius (°C). Historical predicted temperatures, in degrees Celsius (°C); The sampling period time; The external wind speed at the current sampling time is expressed in meters per second (m / s). External wind speed at the current sampling time The wind speed dynamic mapping term for real-time control is expressed in watts per square meter in degrees Celsius (W / (m²·°C)). The equivalent heat capacity of the cabinet is expressed in joules per degree Celsius (J / °C). The heat dissipation surface area is measured in square meters (m²). The external ambient temperature is measured in degrees Celsius (°C). The internal heating resistance is in ohms (Ω). The current in the main circuit at the current sampling time is expressed in amperes (A). The radiation equivalent current squared, with dimensions in ampere squared (A²); the theoretical formula for temperature prediction based on this mechanism uses algebraic transformations to decompose future temperature predictions into historical temperature reference terms. Convection heat dissipation environment correction item With Joule and radiative heat injection items The linear accumulation not only eliminates complex calculus operations, but also reduces the number of cycles occupied by the arithmetic logic unit in a single interrupt service by reorganizing variables.

[0044] Step 6: Within the cloud server cluster, offline parameter aggregation calculations are performed on the physical combination terms in the theoretical formula for mechanism temperature prediction. Specifically, combining the identified and extracted equivalent heat capacity and internal thermal impedance of the cabinet, the nonlinear aggregation terms containing the fluid dynamics derivation process are defined as thermal conductivity admittance comprehensive factors. A mapping array is generated offline by batch traversal solving the wind speed domain. Simultaneously, the constant aggregation terms containing the heat transfer properties of the sheet metal inside the distribution box are defined as heat injection intrinsic coefficients. A prediction execution formula that can be directly called by the edge computing gateway microcontroller is derived and constructed, as follows:

[0045]

[0046] In the formula, This is the current sampling time; For historical sampling moments; The current predicted temperature; Predicting temperatures based on historical data; The thermal conductivity-admittance comprehensive factor; External ambient temperature; The intrinsic coefficients for heat injection; This represents the main circuit current at the current sampling moment; The square of the radiation equivalent current.

[0047] The predictive execution formula transforms the continuous integral evolution process of the partial differential shunt thermal coupling equation into a microcontroller memory lookup table mapping and periodic multiplication and addition hardware instruction execution process. When the external blowing wind speed changes, it drives the thermal conductivity admittance synthesis factor to change accordingly. This allows the physical change in wind speed to be directly converted into a constraint correction of the microscopic physical properties of heat transfer on the metal surface of the equipment. At the same time, the heat injection intrinsic coefficients construct a fixed baseline level for the heating characteristics of the equipment. This transformation ensures the efficiency and accuracy of the underlying control execution.

[0048] Furthermore, the program controlling the cloud server cluster to distribute the program encapsulates the one-dimensional structure dynamic thermal conductance admittance synthesis factor array file containing floating-point nodes and the thermal injection intrinsic coefficients into a compressed configuration file package, which is then distributed across the public network communication link using the Internet of Things protocol and stored within the storage address range of the read-only flash memory chip at the bottom layer of the edge computing gateway.

[0049] S2. The edge computing gateway extracts the thermal conductivity admittance comprehensive factor based on the external blowing wind speed, converts the radiative heat flux intensity into the square of the radiative equivalent current through physical equivalent dimensionality reduction, and combines the real-time sampled variables into the dynamic thermal extrapolation formula to calculate the current predicted temperature.

[0050] It should be further added that, since there is a difference in the thermal power conversion dimension between the external direct solar radiation heat power and the internal copper busbar Joule resistance heating mechanism, in order to meet the operational constraints of the underlying hardware memory resources, this invention adopts a mathematical equivalent dimensionality reduction processing mechanism. Based on the first-order approximation principle, the scalar quantity of spatial radiation heat flux is converted into the square of the radiation equivalent current with the same physical dimension as the square of the state variable of the total current of the internal main circuit. Thus, the discrete external optical radiation interference variable is input as a homogenized parameter into the internal discretized numerical calculation equation operation system.

[0051] Specifically, during the outdoor power distribution operation protection task cycle, the underlying hardware timer of the control edge computing gateway is used to cyclically start the data acquisition and calculation interrupt service routine with a fixed clock beat. The external wind speed and radiative heat flux intensity fed back by the local micro weather station sensor are read through the external sensor communication interface of the edge computing gateway. The direct memory addressing lookup table method is used in the data pointer register of the edge computing gateway microprocessor core to extract the thermal conductance admittance comprehensive factor value corresponding to the external wind speed environment by indexing the mapping array address space pre-written in the internal read-only flash memory, thus completing the memory addressing conversion process based on the nonlinear boundary layer fluid dynamics mechanism.

[0052] Furthermore, the control edge computing gateway asynchronous serial communication bus reads the three-dimensional spatial latitude and longitude coordinate data and clock timing data output by the built-in satellite positioning receiving module; the control edge computing gateway control main program calls the solar altitude angle equation to calculate the spatial incident angle between the direct sunlight beam and the normal to the surface of the distribution box metal cabinet, and derives and constructs the radiation compensation conversion execution formula based on this spatial incident angle. The derivation process is as follows:

[0053] The first step, based on the principle of optical projection, shows that the effective heat power absorbed by the metal casing is equal to the intensity of the radiant heat flux. Multiply by heat dissipation surface area Angle with spatial incidence The cosine value and the pre-calibrated coating spectral absorbance The product of these factors is used to construct the equation for the absorption of external space radiation heat flux, which has the following form:

[0054]

[0055] In the formula, To effectively absorb heat power, the dimension is watt (W). The intensity of radiative heat flux is expressed in watts per square meter (W / m²). The surface area parameter of the metal cabinet of the distribution box is expressed in square meters (m²). 2 ); It is a cosine function; The angle of incidence in space; The pre-calibrated coating spectral absorptivity is dimensionless. This external space radiative heat flux absorption equation is used to reflect the actual heat absorption effect of solar radiation. When the intensity of radiative heat flux increases or the spatial incident angle decreases, resulting in an increase in the cosine value, the effective absorbed heat power will increase synchronously, characterizing a sharp increase in the effective absorbed heat power absorbed by the surface of the distribution box metal cabinet.

[0056] The second step, according to Joule's law, is that electrical heating power equals the product of the square of the current parameter and the circuit impedance parameter. To directly incorporate the external optical radiation heat power into the discrete electrical calculation channel, the effective absorbed heat power is equivalently set as the power value generated by the square of the radiation equivalent current passing through the reference heating impedance, thus establishing the equivalent equation. Substituting the external space radiation heat flux absorption equation from the first step into this equivalent equation and simplifying it algebraically, we extract the square of the radiation equivalent current, which includes division calibration operations, and derive the final compensation execution formula as follows:

[0057]

[0058] In the formula, The square of the radiation equivalent current; The intensity of radiative heat flux is expressed in watts per square meter (W / m²). The surface area parameter of the metal cabinet of the distribution box is expressed in square meters (m²). 2 ); It is a cosine function; The angle of incidence in space; The pre-calibrated absorbance ratio of the coating spectrum is dimensionless. The reference heating impedance, with dimensions in ohms (Ω), is pre-configured within the edge computing gateway. The compensation execution formula constructs equations for angular geometric projection and optical coating absorption mechanisms, which converts discrete external optical radiation interference variables into equivalent electrical Joule heating parameters. When the square of the equivalent radiation current increases, the radiation heat of the equivalent input is superimposed with the heating effect of the internal current, thereby effectively reducing the dimension of the cross-physical field interference variables and ensuring that the photothermal compensation algorithm meets the operating boundary constraints of the underlying microcontroller.

[0059] Furthermore, the edge computing gateway's analog-to-digital conversion sampling channel collects the main circuit current and external ambient temperature; the edge computing gateway's microprocessor core logic arithmetic unit substitutes the historical predicted temperature, external ambient temperature, main circuit current, square of radiation equivalent current, thermal conductivity admittance factor, and heat injection intrinsic coefficient into the prediction execution formula; the edge computing gateway's microprocessor single interrupt service routine sequentially executes flash memory read lookup instructions, multiplication instructions, and addition instructions to calculate and output the current predicted temperature corresponding to the physical law of aerodynamic thermal cross-coupling at the current sampling moment; the latest predicted temperature is overwritten in the random access memory space as the historical predicted temperature for the next cycle into the corresponding buffer register unit, and the edge computing gateway microcontroller is further controlled to enter a low-power state to wait for the next hardware timer clock cycle instruction to call the calculation.

[0060] S3, the edge computing gateway extracts the temperature rise rate based on sliding window differential extraction, and triggers a variable structure flexible automated protection intervention mechanism that includes dynamic advanced heat dissipation of fans and tripping of secondary priority loads based on the judgment results of the current predicted temperature and temperature rise rate crossing multiple preset threshold groups.

[0061] It should be noted that traditional power distribution protection architecture uses a single static temperature threshold as the trigger for determining the start / stop action of the cooling exhaust fan or the tripping action of the main circuit circuit breaker. When facing a midday stormless environment combined with local load surge conditions, the single static temperature threshold judgment feedback mechanism has limitations in terms of heat accumulation response delay. This invention introduces a time difference slope extraction and analysis mechanism for the current predicted temperature, and configures a priority tripping strategy to cut off secondary power supply branches under the condition of convection heat dissipation obstruction. By reducing the power supply continuity of non-core loads, the insulation safety of the overall core bus insulation medium physical support structure of the distribution box is ensured, so that the variable structure flexible control intervention action conforms to the hardware protection design logic of power grid equipment under extreme conditions.

[0062] Specifically, the control edge computing gateway monitors and extracts the current predicted temperature; the control edge computing gateway control main program uses the current predicted temperature stored in the buffer of multiple previous discrete calculation cycles within the sliding window matrix to perform first-order difference slope derivation extraction calculation to obtain the temperature rise rate inside the metal sealed cavity; the magnitude of the temperature rise rate inside the metal sealed cavity is used to characterize the accelerated change characteristics of heat accumulation, thereby triggering the control system's underlying driver hardware program to initiate a variable structure flexible automated protection intervention action execution mechanism based on the current predicted temperature and the temperature rise rate inside the metal sealed cavity.

[0063] It should be added that before executing this intervention, the control system pre-configures a stable early warning threshold, a sudden alarm threshold, a first temperature threshold, and a second temperature threshold, and limits the first temperature threshold to be greater than the second temperature threshold and the sudden alarm threshold to be greater than the stable early warning threshold in terms of numerical magnitude. The above threshold group setting parameters are derived from the engineering environment wind tunnel simulation calibration test: by applying a step surge current load in a closed wind tunnel and simultaneously cutting off the external blowing airflow, the slope of the tangent line of the temperature rise trajectory under different combinations of wind speed and current surge is recorded to calibrate the stable early warning threshold and the sudden alarm threshold. Then, combined with the accelerated thermal aging life curve of the core bus insulation medium, the steady-state temperature reaching the critical softening point of the insulation material is set as the first temperature threshold, and the temperature rise surge during the fan start-up response delay time is deducted from the first temperature threshold to calibrate the second temperature threshold.

[0064] Furthermore, it is determined whether the absolute value of the temperature rise rate inside the metal sealed cavity is less than the stable warning threshold pre-stored by the control system: when the absolute value of the temperature rise rate inside the metal sealed cavity is less than the stable warning threshold, the edge computing gateway determines that the heat dissipation flow path of the thermal boundary layer on the surface of the distribution box is in a stable state; under this stable temperature rise physical condition, if the current predicted temperature is greater than the first temperature threshold, the control edge computing gateway outputs a high-level duty cycle square wave drive signal to the pulse width modulation peripheral channel, thereby driving the forced cooling exhaust fan motor at the top of the distribution box to perform the exhaust heat action.

[0065] Furthermore, when the current predicted temperature is less than the first temperature threshold and the temperature rise rate inside the metal sealed cavity is greater than the preset sudden change alarm threshold, the control edge computing gateway determines that the thermal boundary layer on the surface of the distribution box metal cabinet is in an aerodynamic viscous resistance physical state; triggers the underlying protection intervention logic and starts the thermal advance compensation control intervention mechanism; the control edge computing gateway dynamically lowers the fan start trigger threshold benchmark value to the second temperature threshold, and starts the forced cooling exhaust fan motor at the top of the distribution box with a preset maximum duty cycle to thin the surface air viscous resistance layer; after starting the forced cooling exhaust fan motor at the top of the distribution box, if the current predicted temperature is still detected to show a non-linear rising trend, the control edge computing gateway microprocessor chip sends a tripping command to the tripping electric operating mechanism of the intelligent molded case circuit breaker through the underlying asynchronous serial communication bus interface, cutting off the pre-configured secondary priority load power supply output branch, thereby avoiding cascaded thermal failure faults in the overall power distribution network.

[0066] This invention also discloses an automated power distribution control system for JP cabinets based on big data cloud-edge collaboration, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the automated power distribution control method for JP cabinets based on big data cloud-edge collaboration according to this invention is implemented.

[0067] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

Claims

1. A method for automated power distribution control of JP cabinets based on big data cloud-edge collaboration, comprising communication and transmission of feature data between a cloud server cluster and an edge computing gateway, characterized in that, include: The cloud server cluster extracts the equivalent heat capacity and internal thermal impedance of the cabinet based on the time-series environmental feature matrix, pre-derives and constructs a dynamic thermal simulation formula, generates a mapping array and heat injection intrinsic coefficients offline and sends them to the edge computing gateway, and the mapping array contains a thermal conductivity and admittance comprehensive factor. The edge computing gateway extracts the corresponding thermal conductivity admittance comprehensive factor from the mapping array based on the acquired external wind speed, converts the radiative heat flux intensity into the equivalent square of the radiative current according to the spatial geometric incident parameters, and inputs the equivalent square of the radiative current, the external ambient temperature and the main circuit current into the dynamic thermal deduction formula to calculate and output the current predicted temperature. The edge computing gateway uses the current predicted temperature from the preceding discrete computing cycle to perform time difference slope deduction and extraction calculations to obtain the temperature rise rate inside the metal sealed cavity, and initiates a variable structure flexible automated protection intervention mechanism based on the current predicted temperature and the temperature rise rate inside the metal sealed cavity.

2. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 1, characterized in that, The edge computing gateway reads historical wind speed sequences and historical continuous environmental temperature sequences over a fixed time span; it also reads historical load current sequences and historical continuous cabinet temperature sequences simultaneously; and it aligns and splices the historical wind speed sequences, historical continuous environmental temperature sequences, historical load current sequences, and historical continuous cabinet temperature sequences according to time characteristics to construct the time-series environmental feature matrix.

3. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 1, characterized in that, The dynamic thermal derivation formula is as follows: ； In the formula, This is the current sampling time; For historical sampling moments; The current predicted temperature; Predicting temperatures based on historical data; The thermal conductivity-admittance comprehensive factor; External ambient temperature; The intrinsic coefficients for heat injection; This represents the main circuit current at the current sampling moment; The square of the radiation equivalent current.

4. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 1, characterized in that, The formula for calculating the thermal conductivity admittance factor is: ； In the formula, The thermal conductivity-admittance comprehensive factor; The sampling period time; The external wind speed at the current sampling moment; External wind speed at the current sampling time Real-time controlled dynamic wind speed mapping item; The equivalent heat capacity of the cabinet; This represents the surface area for heat dissipation.

5. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 1, characterized in that, The formula for calculating the intrinsic coefficient of the heat injection is: ； In the formula, The intrinsic coefficients for heat injection; The sampling period time; Internal heating resistance; The equivalent heat capacity of the cabinet.

6. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 1, characterized in that, The formula for calculating the square of the radiation equivalent current is: ； In the formula, Represents the square of the radiation equivalent current; Represents the intensity of radiative heat flux; Represents the surface area for heat dissipation; Represents the cosine function; Represents the angle of incidence in space; Represents the spectral absorbance of the coating; This represents the reference heating impedance.

7. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 1, characterized in that, The variable structure flexible automated protection intervention action execution mechanism pre-configures a stable early warning threshold, a sudden alarm threshold, a first temperature threshold, and a second temperature threshold; the first temperature threshold is limited to be greater than the second temperature threshold, and the sudden alarm threshold is greater than the stable early warning threshold.

8. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 7, characterized in that, When the absolute value of the temperature rise rate inside the metal sealed cavity is less than the stable warning threshold, the edge computing gateway confirms that the heat dissipation flow path of the external thermal boundary layer is in a stable state; when the current predicted temperature is greater than the first temperature threshold, a high-level duty cycle square wave drive signal is output to the pulse width modulation peripheral channel to drive the forced cooling exhaust fan motor at the top of the distribution box to perform the exhaust heat action. After starting the forced cooling exhaust fan motor on the top of the distribution box, in response to the detection result that the current predicted temperature shows a continuous upward trend, the edge computing gateway sends a tripping command to the tripping electric operating mechanism of the intelligent molded case circuit breaker through the underlying asynchronous serial communication bus interface, cutting off the pre-configured secondary priority load power supply output branch.

9. The JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to claim 7, characterized in that, When the current predicted temperature is less than the first temperature threshold and the temperature rise rate inside the metal sealed cavity is greater than the sudden alarm threshold, the edge computing gateway determines that the external convective thermal boundary layer is in an aerodynamic viscous stagnation physical state; and dynamically lowers the fan start trigger threshold benchmark value to the second temperature threshold to start the top exhaust fan motor with a preset maximum duty cycle.

10. A JP cabinet automated power distribution control system based on big data cloud-edge collaboration, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions, which, when executed by the processor, implement the JP cabinet automated power distribution control method based on big data cloud-edge collaboration according to any one of claims 1-9.

Images