Retail spot power load dispatching method considering physical congestion of distribution network

Abstract

Embodiments of the present disclosure disclose a retail spot power load scheduling method considering physical blocking of distribution network. A specific embodiment of the method comprises: in response to a distribution node being subjected to a transient impact, obtaining a steady-state voltage and a net load current; generating a loss gradient based on the node resistance reactance, the conductor temperature coefficient and the net load current; generating a bias factor based on the steady-state voltage and the net load current; generating a current limiting signal based on the loss gradient and the bias factor, and obtaining a load reduction instruction through dimension reduction mapping processing; issuing an electronic switch device to obtain a load reduction state, and comparing the load reduction state with a bus transient voltage envelope line; in response to the result meeting the line failure prevention condition, generating a driving instruction to drive a split excitation trip coil to separate the contacts. The embodiment can differentially regulate the loads of each node, quickly limit the current under impact working conditions, and prevent the line from being damaged due to overload and overheating.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to a method for dispatching retail spot power loads taking into account physical congestion in distribution networks. Background Technology

[0002] Currently, the retail spot electricity market contains price-sensitive flexible loads (e.g., electric vehicle charging stations, smart home appliances). When electricity prices fall, these devices may start simultaneously, easily leading to line overload. To ensure the safe operation of the distribution network and avoid physical congestion caused by line overload, it is usually necessary to current-limit or reduce the load at nodes (e.g., reduce power operation) or disconnect the load (e.g., directly cut off power). For the scheduling and current limiting of distribution network node loads, the common approach is to issue unified scheduling instructions to the node loads based on fixed electrical thresholds (e.g., uniform rated current upper limit) or commercial economic algorithms to regulate the power supply.

[0003] However, when using the above method to dispatch power load, the following technical problems often arise: First, the service life of conductors varies in different regions, and the degree of insulation aging also differs. Heat dissipation conditions and heat accumulation are easily affected by the environment. Although a unified scheduling command logically limits the load capacity, it is difficult to take into account the actual state of each piece of hardware (e.g., heat resistance).

[0004] Secondly, declining spot electricity prices can easily trigger high-concurrency surges, potentially causing communication link disruptions or delays in dispatch commands. Some switching devices may fail to respond promptly to limiting actions, leading to circuit burnout or device damage. Furthermore, aging or poorly heated distribution wires may lack appropriate current-limiting protection. When temperatures rise sharply, this can easily cause insulation scorching or even line burnout.

[0005] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0007] Some embodiments of this disclosure provide a method, apparatus, electronic device, and computer-readable medium for dispatching retail spot electricity loads taking into account distribution network physical congestion, in order to solve one or more of the technical problems mentioned in the background section above.

[0008] In a first aspect, some embodiments of this disclosure provide a retail spot power load dispatching method considering distribution network physical congestion, including: in response to a target distribution node being in a transient impact state, filtering and denoising branch current sequences and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information; generating active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the net load current characteristic information; generating a dynamic control bias factor based on the steady-state voltage characteristic information and the net load current characteristic information; and generating a dynamic control bias factor based on the active power loss gradient information. Based on the degree information and the aforementioned dynamic control bias factor, a current-limiting scheduling signal is generated. The current-limiting scheduling signal is then subjected to dimensionality reduction mapping to obtain a load-reducing scheduling instruction. This load-reducing scheduling instruction is an instruction for the underlying electronic switching device to perform power limiting. The load-reducing scheduling instruction is then sent to the underlying electronic switching device to obtain load-reducing electrical status information. A safety envelope comparison is performed based on the load-reducing electrical status information and the local bus transient voltage data to obtain a comparison result. In response to the comparison result satisfying the underlying physical defense failure condition, a drive instruction is generated to drive the shunt trip coil to force the mechanical contacts to physically separate.

[0009] Secondly, some embodiments of this disclosure provide a retail spot power load dispatching device that takes into account distribution network physical congestion, including: a filtering and denoising unit configured to, in response to a target distribution node being in a transient impact state, perform filtering and denoising processing on branch current sequences and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information; a first generation unit configured to generate active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, and the aforementioned net load current characteristic information; a second generation unit configured to generate a dynamic control bias factor based on the aforementioned steady-state voltage characteristic information and the aforementioned net load current characteristic information; and an instruction generation unit configured to... Based on the aforementioned active power loss gradient information and the aforementioned dynamic control bias factor, a current-limiting scheduling signal is generated, and the aforementioned current-limiting scheduling signal is subjected to dimensionality reduction mapping processing to obtain a load-reducing scheduling instruction. The aforementioned load-reducing scheduling instruction is an instruction for the underlying electronic switching device to perform power limiting. The result comparison unit is configured to send the aforementioned load-reducing scheduling instruction to the aforementioned underlying electronic switching device to obtain load-reducing electrical state information, and to perform a safety envelope comparison based on the aforementioned load-reducing electrical state information and the local bus transient voltage data to obtain a comparison result. The drive control unit is configured to generate a drive instruction in response to the aforementioned comparison result satisfying the underlying physical defense failure condition, so as to drive the shunt trip coil to force the mechanical contacts to physically separate.

[0010] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, such that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any implementation of the first aspect.

[0011] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0012] The above-described embodiments of this disclosure have the following beneficial effects: The retail spot electricity load dispatching method considering distribution network physical congestion, as described in some embodiments of this disclosure, can effectively address the high-concurrency impact of flexible loads caused by spot electricity price fluctuations. It achieves refined load reduction control of flexible loads and protection of the distribution network's physical defenses without exceeding the actual carrying capacity of the underlying equipment. Specifically, the reasons for related communication congestion or physical conductor overheating and melting are: traditional methods typically use fixed electrical thresholds (or commercial economic algorithms) to issue uniform current-limiting commands to node loads, without considering the heat accumulation effect of conductors, differences in insulation aging states, and the communication and computing power bottlenecks of edge controllers. When a large amount of load surges instantaneously in specific scenarios (e.g., concentrated charging of electric vehicles during off-peak electricity price periods), uniform command control is prone to deviating from the underlying physical reality, thus causing delays or failures in the execution phase. Based on this, some embodiments of the retail spot power load dispatching method considering distribution network physical congestion in this disclosure firstly, in response to the target distribution node being in a transient impact state, filter and denoise the branch current sequence and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information. This effectively filters out the influence of transient surges or high-frequency electromagnetic interference on electrical sampling, restoring the net load level and voltage background that reflect the current true load capacity of the power grid. This reduces the risks associated with dirty data (e.g., false triggering and non-execution of actions), providing a reliable data foundation for subsequent thermodynamic and electrical simulations. Secondly, based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the aforementioned net load current characteristic information, active power loss gradient information is generated. This overcomes the limitation of relying solely on current values ​​for electrical judgment. The thermal drift characteristics of the physical resistance of the conductor material are introduced. By mapping the current parameters to a dynamic Joule heat accumulation trend through the resistance temperature coefficient, the heating risk of the target node under continuous high load can be quantified, ensuring that congestion control does not exceed the thermal tolerance physical boundary of the cable material. Then, based on the aforementioned steady-state voltage characteristics and net load current characteristics, a dynamic control bias factor is generated. This bias factor, mapped from real-time current and voltage characteristics, provides a dynamic safety buffer for macro-level scheduling. When node voltage drops or net load increases, the bias factor automatically narrows the permissible range of regulation, enabling the decision-making to adaptively adjust as operating conditions worsen. Next, based on the aforementioned active power loss gradient information and the aforementioned dynamic control bias factor, a current-limiting scheduling signal is generated. This signal is then subjected to dimensionality reduction mapping to obtain a load-limiting scheduling command, which is a power-limiting command executed by the underlying electronic switching devices. This effectively reduces bus data congestion pressure and ensures that the command characteristics match the physical switching frequency limits of the underlying devices. It also reduces the risk of delay or runaway caused by command backlog.Furthermore, the aforementioned load reduction scheduling command is sent to the underlying electronic switching devices to obtain load reduction electrical status information. A safety envelope comparison is then performed based on this load reduction electrical status information and the transient voltage data of the local bus, yielding a comparison result. Therefore, when the command is issued, physical devices may fail to reduce load properly due to mechanical aging or adhesion. By resampling the actual electrical state after device operation and comparing it with the device's transient safety envelope using gate logic, the execution residual of the underlying switches can be verified, promptly identifying the hidden risk of the command not being effectively executed. Finally, in response to the comparison result satisfying the underlying physical defense failure condition, a drive command is generated to drive the shunt trip coil, forcing the mechanical contacts to physically separate. Thus, when regulation fails and energy exceeds the limit, the system generates an escalator command to activate the trip coil. The electromagnetic attraction generated by the excitation overcomes the spring resistance, causing the contacts to physically separate. Electromagnetic work is used to prevent thermal collapse, ensuring the safety of the power grid hardware. Attached Figure Description

[0013] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0014] Figure 1 This is a flowchart of some embodiments of the retail spot electricity load dispatching method taking into account distribution network physical congestion according to this disclosure; Figure 2 These are schematic diagrams of some embodiments of a retail spot power load dispatching device that takes into account physical congestion in the distribution network, based on this disclosure. Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure; Figure 4 This is a midday valley-triggered centralized access diagram according to one embodiment of this disclosure; Figure 5 This is a double low-price window that triggers two impact graphs according to one embodiment of the present disclosure; Figure 6 This is a diagram of slow price reduction and delayed concurrent access according to one embodiment of this disclosure; Figure 7 This is a flash price and short-pulse load diagram according to one embodiment of the present disclosure; Figure 8 This is a batch access diagram under a low-cost platform according to one embodiment of this disclosure; Figure 9 This is a load inertia diagram of the price recovery process according to one embodiment of this disclosure; Figure 10This is a diagram of slight, frequent price reductions and clustered access according to one embodiment of this disclosure; Figure 11 This is a diagram showing the early morning low-price window shift according to one embodiment of this disclosure; Figure 12 This is a graph showing evening low prices and delayed response according to one embodiment of this disclosure; Figure 13 This is a two-level price reduction corresponding to a tiered access diagram according to one embodiment of this disclosure. Detailed Implementation

[0015] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0016] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0017] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0018] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0019] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0020] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] refer to Figure 1 The flowchart 100 illustrates some embodiments of a retail spot electricity load dispatching method considering distribution network physical congestion according to this disclosure. This retail spot electricity load dispatching method considering distribution network physical congestion includes the following steps: Step 101: In response to the target distribution node being in a transient impact state, the branch current sequence and three-phase bus voltage data are filtered and denoised to obtain steady-state voltage characteristic information and net load current characteristic information.

[0022] In some embodiments, the executor (e.g., an electronic device) of the above-described retail spot electricity load dispatching method considering distribution network physical congestion can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed in the hardware devices listed above. It can be implemented as multiple software programs or software modules to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0023] In some embodiments, the execution entity may, in response to a target distribution node being in a transient impact state, perform filtering and denoising processing on the branch current sequence and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information. The target distribution node may be a low-voltage side transformer bus node in the distribution network topology. For example, a low-voltage side outgoing node of a distribution transformer with a rated capacity of 800kVA and a downstream charging pile communication gateway. The transient impact state may be a non-steady-state condition where, within a preset sampling period, the first-order difference value of the branch current exceeds the allowable ramp-up rate of the equipment's current carrying capacity. For example, within a 20-millisecond sampling step, the instantaneous current increment exceeds the rated margin by 15%. The steady-state voltage characteristic information may be characteristic information representing the effective value of the grid fundamental voltage after bandpass filtering to remove switching frequency harmonics. The net load current characteristic information may be characteristic information representing the effective value of the grid fundamental current after bandpass filtering to remove switching frequency harmonics.

[0024] As an example, raw sampling data can be obtained from current transformers and voltage transformers via an industrial Ethernet bus. After filtering out high-frequency harmonics from the inverter using a Butterworth bandpass filter, the effective value of the fundamental current and the root mean square value of the bus voltage can be calculated as net load current characteristic information and steady-state voltage characteristic information, respectively.

[0025] In some optional implementations of certain embodiments, the execution entity may further include the following steps before the step of responding to the target distribution node being in a transient impact state: The first step is to obtain the downward fluctuation signal of the retail spot electricity price. This signal can be a control message issued by the power trading center server. This control message is triggered when the marginal electricity price at the node falls more than a preset threshold compared to the benchmark price. For example, if a surge in solar power generation in summer causes the midday spot electricity price to drop to 0.1 yuan / kWh, the trading system will push a price reduction flag frame (electricity price floating-point number and event stamp) to the edge gateway.

[0026] As an example, the electricity spot trading platform interface can be called via a dedicated line at a polling cycle (e.g., every 5 minutes). The returned messages are then parsed to extract the real-time floating-point electricity price. Comparing this with the local historical electricity price, if the difference exceeds a preset threshold, the message is identified as a downward fluctuation signal in the retail spot electricity price.

[0027] The second step involves determining the real-time gradient of the branch current sequence in response to the aforementioned downward fluctuation signal in retail spot electricity prices. This real-time gradient can be the ratio of the difference in the effective value of the branch current within adjacent sampling periods to the time step size. For example, with a sampling step size of 10 milliseconds, when the current climbs from 50A to 150A, the ramp-up slope can be calculated as 10000A / s.

[0028] The third step involves determining that the target power distribution node is in a transient impact state in response to the aforementioned real-time gradient exceeding a preset concurrent access threshold. This concurrent access threshold can be a pre-set upper limit for allowable current differential. This upper limit can be determined based on the conductor cross-sectional area and thermal time constant. For example, based on the cable heat dissipation curve safety boundary specification of "1590 A / s," this value corresponds to the extreme load-bearing scenario of 50 7kW charging piles simultaneously starting charging within the same second.

[0029] Step 102: Based on the received resistance and reactance parameter information and conductor resistance temperature coefficient information corresponding to the target distribution node, as well as the net load current characteristic information, generate active power loss gradient information.

[0030] In some embodiments, the aforementioned executing entity can generate active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the aforementioned net load current characteristic information. The aforementioned resistance and reactance parameter information can be conductor impedance parameters sent by a cloud topology server, which can characterize the inherent impedance properties of the line below the target distribution node. For example, it could be a constant array of resistance and reactance values ​​for a 500-meter-long, 120-square-millimeter cross-section overhead conductor at 20 degrees Celsius. The aforementioned conductor resistance temperature coefficient can be attribute information characterizing the drift of conductor material resistivity with increasing temperature. For example, it could be a fixed percentage constant (e.g., 0.00393 / ℃) for the resistance of a copper core conductor increasing by 1 degree Celsius. The aforementioned active power loss gradient information can be a slope parameter characterizing the change of Joule heat power of the distribution line over time. For example, it could be the first-order time derivative of the conductor heat power increasing at a rate of 100 watts per second.

[0031] In some optional implementations of certain embodiments, the execution entity may generate active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the received net load current characteristic information. This may include the following steps: The first step is to extract the effective value sequence from the above net load current characteristic information.

[0032] The second step involves generating dynamic temperature rise parameters based on the aforementioned effective value sequence and the aforementioned conductor resistance temperature coefficient. These dynamic temperature rise parameters can be the temperature difference between the actual operating temperature of the physical conductor and the ambient temperature under a specific current load.

[0033] As an example, the effective value sequence can be input into the physical thermodynamic cumulative function. Based on the resistance value of the previous sampling period and the temperature coefficient of resistance of the wire, the real-time resistance value at the current temperature is estimated. Then, the Joule heating is calculated using this real-time resistance value and the current effective value. The heating is then converted based on a preset heat dissipation time constant to generate a dynamic temperature rise parameter.

[0034] The third step involves performing temperature compensation processing on the aforementioned dynamic temperature rise parameters to obtain corrected resistance parameters. These corrected resistance parameters can be the real-time resistance value of the conductor after temperature correction. For example, when the conductor temperature reaches 60 degrees Celsius, the 0.5-ohm resistance calibrated at 20 degrees Celsius is updated to 0.58 ohms using the temperature drift equation.

[0035] As an example, the temperature coefficient of conductor resistance can be multiplied by the dynamic temperature rise parameter to obtain the impedance drift ratio. This ratio is then applied to the base resistance component to complete the temperature compensation correction, resulting in the corrected resistance parameter. The reactance component of the aforementioned resistance-reactance parameter can be output as an auxiliary parameter.

[0036] The fourth step is to determine the product of the square of the above effective value sequence and the above corrected resistance parameter to obtain the transient active power loss. This transient active power loss can be the instantaneous power consumed by the conductor due to heating caused by resistance. For example, based on Joule's law, the calculated power of the conductor heating at a certain millisecond is 3000 watts.

[0037] The fifth step is to determine the rate of change of the transient active power loss within a preset time sliding window to obtain the active power loss gradient information. The preset time sliding window can be a logical data block with a fixed time span in memory, which can be shifted and updated with the latest sampling time. For example, the window length can be set to 500 consecutive sampling points, and the sliding step size can be 1 sampling point.

[0038] In practice, firstly, the difference between the final power value and the initial power value within a preset time sliding window is determined. Then, this difference is divided by the time span to obtain the slope, which is used as the rate of change.

[0039] Step 103: Generate a dynamic control bias factor based on steady-state voltage characteristic information and net load current characteristic information.

[0040] In some embodiments, the execution entity can generate a dynamic control bias factor based on the steady-state voltage characteristics and the net load current characteristics. This dynamic control bias factor can be a floating-point multiplicative factor used for contraction scheduling current limiting references. For example, it could be 0.75 under voltage dip conditions.

[0041] In some optional implementations of certain embodiments, the execution entity may generate a dynamic control bias factor based on the steady-state voltage characteristic information and the net load current characteristic information, which may include the following steps: The first step is to obtain the rated distribution voltage and physical current carrying capacity limit parameters of the hardware equipment corresponding to the target distribution node, as well as the physical penalty weight coefficient matrix relating the topology level and insulation aging status of the target distribution node. The rated distribution voltage can be the nominal operating voltage specified on the equipment's nameplate. The physical current carrying capacity limit parameter can be the maximum continuous current allowed to pass through the conductor without thermal melting. For example, a limit of 200A can be found in a cable manual. The physical penalty weight coefficient matrix can be a two-dimensional matrix quantifying the impact of spatial location and hardware health on the contraction of the control boundary. In this matrix, rows represent the spatial topology level characteristics of the distribution node, and columns represent the conductor environmental aging assessment dimensions. Matrix elements represent the penalty proportionality constant allocated under the intersection of a specific topology and aging dimension.

[0042] The second step involves generating a first voltage drop deviation value based on the aforementioned steady-state voltage characteristic information and the aforementioned rated distribution voltage. This first voltage drop deviation value can be the absolute difference between the actual voltage and the rated voltage. For example, when the steady-state voltage characteristic information is 360V and the rated distribution voltage is 380V, the resulting difference is 20V.

[0043] The third step involves generating an over-limit current residual sequence based on the aforementioned net load current characteristic information and the aforementioned physical current carrying capacity upper limit parameter. This over-limit current residual can be the current difference exceeding the physical carrying capacity limit. For example, when the upper limit is 200A, a sampling point of 210A is converted into a difference of 10A.

[0044] As an example, the net load current characteristic information sampling points can be compared numerically with the physical current carrying capacity limit parameter one by one. The difference exceeding the limit is retained, and the values ​​of the sampling points that do not exceed the limit are set to zero. Then, the data is reassembled according to the original timestamps to generate an over-limit current residual sequence.

[0045] The fourth step involves performing time-series integration on the aforementioned over-limit current residual sequence to obtain the second over-limit deviation value. This second over-limit deviation value can be the area integral value characterizing the intensity of accumulated heat generation during continuous current overload within the sampling window.

[0046] As an example, valid data points are extracted from the over-limit current residual sequence, and combined with a fixed ADC sampling time step, the trapezoidal integral algorithm is called to calculate the area under the sequence envelope, which is used as the second over-limit deviation value.

[0047] Fifth, based on the above physical penalty weight coefficient matrix, the above first drop deviation value and the above second over-limit deviation value are subjected to nonlinear fusion processing to obtain the dynamic control bias factor.

[0048] As an example, firstly, the executing entity can extract the maximum matrix norm of the aforementioned physical penalty weight coefficient matrix as the global physical penalty scalar. Secondly, the first fall deviation value and the second over-limit deviation value are concatenated into a one-dimensional input feature vector, and then multiplied by a preset two-dimensional electrothermal weight coefficient column vector. Next, the multiplication result is multiplied by the global physical penalty scalar. Finally, the final product is input into a local activation function (e.g., sigmoid) to obtain the dynamic control bias factor.

[0049] In some optional implementations of certain embodiments, the execution entity may obtain a physical penalty weight coefficient matrix relating the target distribution node topology level and insulation aging state, which may include the following steps: The first step is to extract the spatial hierarchy parameters of the target distribution node in the physical topology of the distribution network. These spatial hierarchy parameters can be topology identifiers representing the electrical distance or branch depth of the target node from the substation main busbar, and may include main line distance identifiers and branch level identifiers.

[0050] As an example, the standard connection table of the distribution network can be parsed to locate the physical MAC address of the target distribution node. Then, the topology depth integer vector is extracted by looking up the table as a spatial hierarchy parameter.

[0051] The second step involves generating a spatial penalty weight base for mapping the physical spread risk of distribution network congestion based on the aforementioned spatial hierarchy parameters. This spatial penalty weight base can be a spatial dimension penalty row vector representing the larger the spread of heat generation risk as the topology deepens.

[0052] As an example, firstly, spatial hierarchy parameters can be loaded into the accumulator register, and then the topology hierarchy and risk mapping static lookup table can be queried to extract the data constant sequence of the corresponding address and combine them to generate the spatial penalty weight base. The aforementioned topology hierarchy and risk mapping static lookup table can be a pre-calibrated row vector table of penalty coefficients based on theoretical analysis (or simulation calculation) of the impact range of thermal overload at different level nodes, according to a typical distribution network topology.

[0053] The third step is to extract the historical operating temperature sequence and environmental humidity characteristics of the physical conductors of the target power distribution node. The historical operating temperature sequence can be a sequence of readings from temperature sensors on the conductor surface within a fixed period.

[0054] As an example, historical operating temperature sequences can be read by accessing a local sensor database via a communication bus. Then, the current relative humidity value can be read as an ambient humidity characteristic using a generic asynchronous transceiver interface.

[0055] The fourth step involves generating a physical insulation aging derating factor based on the aforementioned historical operating temperature sequence and environmental humidity characteristics. This physical insulation aging derating factor can be a coefficient characterizing the degree of attenuation of the insulation layer's breakdown resistance under varying temperature and humidity conditions.

[0056] As an example, firstly, the arithmetic mean of the historical operating temperature series is determined. Secondly, this mean is used in a polynomial equation for insulation life assessment based on environmental humidity characteristics to obtain the physical insulation aging derating factor. The aforementioned polynomial equation for insulation life assessment can be... The above. This can be the derating factor due to physical insulation aging. (The above...) This could be the arithmetic mean of a historical operating temperature series. (The above...) This could be a characteristic of ambient humidity. (The above...) The coefficients can be derived from the breakdown data of the same type of conductor sample under accelerated aging test at temperature and humidity, and fitted by the least squares method.

[0057] The fifth step involves performing tensor product mapping on the aforementioned spatial penalty weight base and the aforementioned physical insulation aging derating coefficient to obtain the boundary penalty operator matrix. This boundary penalty operator matrix can be an unnormalized two-dimensional parametric matrix that integrates the spatial topology spread dimension and the hardware aging environment dimension. The number of rows and columns of the matrix are determined by the dimensions of the spatial weight base and the aging derating coefficient, respectively.

[0058] As an example, the spatial penalty weight base can be configured as a column vector, the physical insulation aging derating coefficient can be configured as a row vector, and the floating-point multiply-accumulate instruction set can be used to perform an outer product operation on the two to generate a boundary penalty operator matrix.

[0059] The sixth step is to normalize the boundary penalty operator matrix to obtain the physical penalty weight coefficient matrix.

[0060] Step 104: Based on the active power loss gradient information and the dynamic control bias factor, generate a current limiting scheduling signal, and perform dimensionality reduction mapping processing on the current limiting scheduling signal to obtain a load reduction scheduling instruction.

[0061] In some embodiments, the execution entity can generate a current-limiting scheduling signal based on the active power loss gradient information and the dynamic control bias factor, and perform dimensionality reduction mapping processing on the current-limiting scheduling signal to obtain a load-limiting scheduling instruction. The load-limiting scheduling instruction is an instruction for the underlying electronic switching devices to perform power limiting.

[0062] The aforementioned current-limiting dispatch signal can be a signal that limits the effective value of the downstream current at the distribution node. The aforementioned load-reducing dispatch command can be an electrical signal that controls solid-state switching devices (changing the duty cycle) to reduce load power. The aforementioned underlying electronic switching devices can be devices that directly control load switching or current limiting. For example, power electronic inverter modules in energy storage converters and smart circuit breakers.

[0063] In some optional implementations of certain embodiments, the execution entity may generate a current-limiting scheduling signal based on the active power loss gradient information and the dynamic control bias factor, and perform dimensionality reduction mapping processing on the current-limiting scheduling signal to obtain a load-reducing scheduling instruction, which may include the following steps: The first step is to determine the node voltage and power characteristics of the aforementioned active power loss gradient information in the physical topology of the distribution network. These node voltage and power characteristics can be a partial derivative mapping vector representing the change in voltage amplitude at a node in a specific network topology as a function of active power injection. The aforementioned active power loss gradient information can be used as the equivalent injected power increment for current node load fluctuations and participate in power flow calculations.

[0064] As an example, the active power loss gradient information is combined with the locally preset distribution network admittance matrix to perform Newton-Raphson power flow calculations and generate a Jacobian matrix that reflects the linear relationship between the power and voltage increments of all network nodes. Then, the inverse matrix (i.e., the sensitivity matrix) of this Jacobian matrix is ​​calculated, and a one-dimensional array of partial derivatives of the target distribution node voltage with respect to active power is extracted from it as the node voltage-power characteristics.

[0065] The second step involves performing spatial topology optimization and safety boundary pruning on the aforementioned node voltage and power characteristics and the aforementioned dynamic control bias factor to obtain the current-limiting scheduling signal. The aforementioned spatial topology optimization and safety boundary pruning can be a mathematical programming operation based on hardware constraints to prune the theoretical control solution set, thereby eliminating over-limit control paths.

[0066] As an example, the node voltage and power characteristics can be input into a quadratic programming optimizer to generate an initial active power reduction trajectory. Then, the amplitude of this trajectory can be multiplied and clipped using a dynamic control bias factor to output a target active power adjustment trajectory sequence as a current limiting scheduling signal.

[0067] The third step involves obtaining the instantaneous communication bandwidth margin and local controller computing power limit of the underlying electronic switching devices to be executed under the aforementioned load reduction scheduling command, as well as the physical switching frequency limit of the insulated-gate bipolar transistors (IGBTs) in the underlying electronic switching devices. The instantaneous communication bandwidth margin can be the number of bytes remaining to be transmitted per second on the communication bus. The local controller computing power limit can be the peak number of floating-point instructions that the edge computing chip can execute per second. The physical switching frequency limit can be the maximum number of switching pulses per second allowed by the IGBT without thermal breakdown.

[0068] As an example, bandwidth margin can be read via an industrial bus. A status query frame can be sent to the edge controller to obtain the local controller's computing power limit parameter. Then, the physical switching frequency limit can be obtained by looking up a table.

[0069] The fourth step involves generating a discretized sampling step size that maps to the underlying hardware's data throughput bottleneck, based on the aforementioned instantaneous communication bandwidth margin and the local controller's computing power limit. This discretized sampling step size can be a time span constant that matches the data processing limit of the underlying device per unit time. For example, a 50-millisecond downsampling period.

[0070] As an example, the instantaneous communication bandwidth margin and the local controller's computing power limit parameter can be input into a preset hardware capacity assessment function to calculate the minimum safe delay without causing packet loss or overflow. This result is then rounded up to an integer multiple of the standard clock cycle to obtain the discretized sampling step size. The aforementioned hardware capacity assessment function can be a weighted assessment model based on communication queuing delay and computation processing delay. The parameters of this function are calibrated by actual measurements on the target chip and communication bus, and can be used to map bandwidth and computing power input to the minimum safe delay. For example, the aforementioned hardware capacity assessment function could be... The above. This can be the minimum safe delay. The above can be... This can be a fixed overhead constant for low-level hardware interrupt response and protocol stack processing (which can be obtained from the manufacturer's manual or through actual testing). The above This can be the average message length of a single scheduling command. (The above...) This can be instantaneous communication bandwidth margin (communication queuing delay item). The above. This could be the average number of floating-point instructions required for a single scheduling operation. The above... This could be the local controller's maximum computing power parameter (computation processing latency). (The above...) This can be a preset margin, used to compensate for sudden fluctuations (e.g., taking...). 15% of the total.

[0071] Fifth, based on the aforementioned discretized sampling step size, the current limiting scheduling signal is subjected to time-domain dimensionality reduction processing to obtain the active power reduction action feature sequence. This time-domain dimensionality reduction processing can be a rank reduction operation on the downsampled time series based on the discretized step size. The active power reduction action feature sequence can be a discrete power difference sequence that guides hardware peak clipping after downsampling.

[0072] Step 6: Based on the aforementioned physical switching frequency limit, pulse width modulation (PWM) is applied to the aforementioned active power reduction action characteristic sequence to obtain the load reduction scheduling command. The aforementioned PWM processing can be a mapping process that converts control parameters into square waves representing the on / off duty cycles of the switching transistors.

[0073] As an example, the amplitude in the active power reduction action characteristic sequence can be linearly mapped to a duty cycle constant. Then, the square wave carrier frequency is constrained by the physical switching frequency limit, and a drive message with both duty cycle and carrier frequency constrained is output as a load reduction scheduling command. Furthermore, based on the principle that the product of duty cycle and carrier frequency does not exceed the IGBT safe operating area (SOA), if a conflict arises under the joint constraint of the two, the carrier frequency is reduced first to ensure device safety.

[0074] Step 105: Send the load reduction scheduling command to the underlying electronic switching device to obtain the load reduction electrical status information, and compare the safety envelope based on the load reduction electrical status information and the local bus transient voltage data to obtain the comparison result.

[0075] In some embodiments, the executing entity can send the load reduction scheduling command to the underlying electronic switching device to obtain load reduction electrical status information, and perform a safety envelope comparison based on the load reduction electrical status information and local bus transient voltage data to obtain a comparison result. The load reduction electrical status information can be the RMS current value and power response data resampled and fed back by the local transformer after the switching device performs limiting duty cycle adjustment. The local bus transient voltage data can be the root mean square value of the bus voltage captured by the voltage transformer within the current sampling step. The safety envelope can be a preset two-dimensional operating safety boundary coordinate set characterizing the equipment under the dual constraints of thermal stability limit and voltage drop limit. The comparison result can be a Boolean logic value (or absolute value of spatial distance) parameter characterizing whether the actual operating coordinate point exceeds the safety boundary.

[0076] In some optional implementations of certain embodiments, the execution entity may perform a safety envelope comparison based on the aforementioned load reduction electrical state information and local bus transient voltage data to obtain a comparison result, and in response to the comparison result satisfying the underlying physical defense failure condition, generate a drive command to drive the shunt trip coil to force the mechanical contacts to physically separate, which may include the following steps: The first step is to obtain the thermal stability current limit parameter and the transient voltage sag threshold. The thermal stability current limit parameter can be the maximum allowable overcurrent time integral value for preventing irreversible thermal breakdown of the conductor insulation material. The transient voltage sag threshold value can be the minimum voltage value required to prevent undervoltage tripping of the inverter or contactor control circuit.

[0077] As an example, thermal stability current limit parameters and transient voltage drop thresholds can be obtained from the factory safety configuration file.

[0078] The second step is to generate a safety envelope based on the aforementioned thermal stability current limit parameter and the aforementioned transient voltage drop threshold.

[0079] As an example, a two-dimensional state-space coordinate system is constructed in memory, with the accumulated heat as the horizontal axis and voltage as the vertical axis. Then, the thermal stability current limit parameter is mapped to the vertical boundary of the horizontal axis, and the transient voltage drop threshold is mapped to the horizontal boundary of the vertical axis. Finally, the vertical and horizontal boundaries are closed and intersected to generate a safety envelope composed of a safety coordinate lattice.

[0080] The third step is to determine the cumulative current over-limit value in the aforementioned descent electrical status information. This cumulative current over-limit value can be the area of ​​the time over which the actual current continuously exceeds the rated current, and can be used to quantify the Joule heat accumulation effect.

[0081] As an example, the load reduction electrical status information can be parsed to extract the effective value sequence of the actual output current. Then, the difference between each value in the sequence and the rated current constant is calculated one by one. Next, the trapezoidal integral algorithm is called to accumulate the positive difference values ​​over time, resulting in the cumulative value of the output current exceeding the limit.

[0082] Fourth, based on the aforementioned safety envelope, perform gate logic cross-comparison processing on the aforementioned cumulative current exceeding the limit value and the aforementioned transient voltage data of the local bus, and obtain the comparison result. The aforementioned gate logic cross-comparison processing can be an operation using hardware logic gates (or Boolean operations) to determine whether multidimensional data falls within a specific coordinate region.

[0083] As an example, firstly, the accumulated value of current exceeding the limit can be combined with the transient voltage of the local bus to form a real-time two-dimensional operating coordinate point. Then, a hardware comparator is called to load the safety envelope boundary constraints, determine whether the coordinate point is outside the boundary, and output an out-of-bounds flag as the comparison result.

[0084] Fifth, in response to the comparison results satisfying the underlying physical defense failure condition, determine the electromagnetic excitation inductance parameters of the shunt trip coil and the static holding resistance of the closing spring of the mechanical contact. The underlying physical defense failure condition can be a fault identification state where the comparison result continuously exceeds the safety boundary for a period exceeding the inherent fault tolerance delay of the switch. The electromagnetic excitation inductance parameters can be parameters (Henry constants) representing the coil's conversion efficiency of electrical energy into magnetic energy and its impedance characteristics. The static holding resistance of the closing spring can be the minimum spring tension required to maintain the circuit breaker contacts pressed tightly closed.

[0085] As an example, when the comparison result continuously exceeds the limit and triggers a timer overflow interrupt, the failure condition of the underlying physical defense is established. Then, the electromechanical nameplate parameters stored in the circuit breaker are read through the I2C interface to extract the Henry-level inductance constant and the Newton-level mechanical constant, which are used as the electromagnetic excitation inductance parameter and the static holding resistance of the closed spring, respectively.

[0086] Step 6: Perform electromagnetic tripping mapping on the aforementioned electromagnetic excitation inductance parameters and the static holding resistance of the closed spring to obtain the target tripping current parameter. The electromagnetic tripping mapping can be derived based on Ampere's circuital law to calculate the excitation energy required to overcome the spring resistance. The target tripping current parameter can be the minimum driving current required to generate the attraction force to break the spring.

[0087] As an example, the static holding resistance of the closed spring can be input into a preset electromagnetic attraction balance inverse equation. Combining the electromagnetic excitation inductance parameters and the coil turns constant, the critical excitation ampere-turns required to reach the attraction threshold can be calculated. Then, this is divided by the coil turns constant for linear mapping to generate the target tripping current parameter. The aforementioned electromagnetic attraction balance inverse equation is based on Maxwell's electromagnetic attraction formula. Its physical meaning is that when current flows through the shunt trip coil, magnetic flux is generated in the magnetic circuit, resulting in an electromagnetic attraction force on the iron core. With coil Current Magnetic circuit cross-sectional area It is proportional to the square of the working air gap. They are inversely proportional. Therefore, the formula could be: The above. It can be the vacuum permeability constant. This is used to overcome the static holding resistance of a closed spring. Electromagnetic attraction is required. Solving the above formula in reverse, we can obtain the inverse equation for the electromagnetic attraction equilibrium: The above. It can be an engineering correction factor (range of 0.3-0.8) for magnetic circuit leakage and core saturation, which can be determined by actual measurement using samples of the same type of electromagnet.

[0088] Step 7: Based on the aforementioned target tripping current parameters, a strong-cut high-level signal is generated. This strong-cut high-level signal is used to bypass the operating system kernel. The strong-cut high-level signal can be a DC trigger level (e.g., 3.3V or 5V) directly output to the microcontroller's general-purpose input / output pins. The bypassed operating system kernel can be a low-level execution path that bypasses the software task scheduling queuing mechanism and directly triggers hardware register toggles using non-maskable interrupts.

[0089] As an example, after generating the target trip current parameter, a non-maskable interrupt request is immediately triggered, suspending all regular software processes in the current operating system kernel, writing a set instruction to the specified general-purpose input / output port control register through direct memory access, and pulling the output strong-cut high-level signal high on the physical pin.

[0090] Step 8: The aforementioned high-level signal is subjected to opto-isolation drive processing to obtain a drive command. This drive command triggers the shunt trip coil to generate a transient electromagnetic attraction, forcing the mechanical contacts to physically separate. The opto-isolation drive processing can utilize optocouplers to electrically isolate the low-voltage side control signal of the microcontroller from the high-voltage circuit of the circuit breaker and perform power amplification. The drive command can be a high-current command that directly conducts the trip coil circuit after being output in a push-pull manner by a power amplifier. The transient electromagnetic attraction can be the force generated by the alternating magnetic field at the moment the coil is energized, acting on the mechanical armature. The physical separation of the mechanical contacts can be the result of the moving and stationary contacts overcoming spring resistance, displacing, and completely disconnecting, thus forcibly cutting off the main circuit.

[0091] As an example, a strong high-level signal can be injected into the LED pin of the optocoupler isolation chip. Then, the optical signal crosses the isolation barrier to activate the phototransistor on the receiving side, driving the subsequent power switch into its saturation region and outputting amplified power as the drive command. Next, the shunt trip coil circuit is momentarily closed, using the transient electromagnetic force generated by the coil to overcome spring resistance and pull open the mechanical contacts, achieving physical separation of the mechanical contacts.

[0092] Step 106: In response to the comparison result satisfying the failure condition of the underlying physical defense, a drive command is generated to drive the shunt trip coil to force the mechanical contacts to physically separate.

[0093] In some embodiments, the execution entity may generate a drive command in response to the comparison result satisfying the underlying physical defense failure condition, so as to drive the shunt trip coil to force the mechanical contacts to physically separate.

[0094] The above-described embodiments of this disclosure have the following beneficial effects: The retail spot electricity load dispatching method considering distribution network physical congestion, as described in some embodiments of this disclosure, can effectively address the high-concurrency impact of flexible loads caused by spot electricity price fluctuations. It achieves refined load reduction control of flexible loads and protection of the distribution network's physical defenses without exceeding the actual carrying capacity of the underlying equipment. Specifically, the reasons for related communication congestion or physical conductor overheating and melting are: traditional methods typically use fixed electrical thresholds (or macroeconomic algorithms) to issue uniform current-limiting commands to node loads, without considering the heat accumulation effect of conductors, differences in insulation aging states, and the communication and computing power bottlenecks of edge controllers. When a large amount of load surges instantaneously in specific scenarios (e.g., concentrated charging of electric vehicles during off-peak electricity price periods), uniform command control is prone to deviating from the underlying physical reality, thus causing delays or failures in the execution phase. Based on this, some embodiments of the retail spot power load dispatching method considering distribution network physical congestion in this disclosure firstly, in response to the target distribution node being in a transient impact state, filter and denoise the branch current sequence and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information. This effectively filters out the influence of transient surges or high-frequency electromagnetic interference on electrical sampling, restoring the net load level and voltage background that reflect the current true load capacity of the power grid. This reduces the risks associated with dirty data (e.g., false triggering and non-execution of actions), providing a reliable data foundation for subsequent thermodynamic and electrical simulations. Secondly, based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the aforementioned net load current characteristic information, active power loss gradient information is generated. This overcomes the limitation of relying solely on current values ​​for electrical judgment, introducing the thermal drift characteristics of the physical resistance of the conductor material. The current parameter is mapped to a dynamic Joule heat accumulation trend through the resistance temperature coefficient. This quantifies the heating risk of the target node under continuous high load, ensuring that congestion control does not exceed the thermal tolerance physical boundary of the cable material. Then, based on the aforementioned steady-state voltage characteristics and net load current characteristics, a dynamic control bias factor is generated. This bias factor, mapped from real-time current and voltage characteristics, provides a dynamic safety buffer for macro-level scheduling. When node voltage drops or net load increases, the bias factor automatically narrows the permissible range of regulation, enabling the decision-making to adaptively adjust as operating conditions worsen. Next, based on the aforementioned active power loss gradient information and the aforementioned dynamic control bias factor, a current-limiting scheduling signal is generated. This signal is then subjected to dimensionality reduction mapping to obtain a load-limiting scheduling command, which is a power-limiting command executed by the underlying electronic switching devices. This effectively reduces bus data congestion pressure and ensures that the command characteristics match the physical switching frequency limits of the underlying devices. It also reduces the risk of delay or runaway caused by command backlog.Furthermore, the aforementioned load reduction scheduling command is sent to the underlying electronic switching devices to obtain load reduction electrical status information. A safety envelope comparison is then performed based on this load reduction electrical status information and the transient voltage data of the local bus, yielding a comparison result. Therefore, when the command is issued, physical devices may fail to reduce load properly due to mechanical aging or adhesion. By resampling the actual electrical state after device operation and comparing it with the device's transient safety envelope using gate logic, the execution residual of the underlying switches can be verified, promptly identifying the hidden risk of the command not being effectively executed. Finally, in response to the comparison result satisfying the underlying physical defense failure condition, a drive command is generated to drive the shunt trip coil, forcing the mechanical contacts to physically separate. Thus, when regulation fails and energy exceeds the limit, the system generates an escalator command to activate the trip coil. The electromagnetic attraction generated by the excitation overcomes the spring resistance, causing the contacts to physically separate. Electromagnetic work is used to prevent thermal collapse, ensuring the safety of the power grid hardware.

[0095] Optionally, the following is the applicant's experimental test report in the experimental scenario: The experimental scenario could be: a surge in summer solar power generation leading to a midday drop in spot electricity prices, prompting the trading system to send a price reduction signal to the edge gateway; the system further calculates the real-time gradient of branch current changes, and when the gradient exceeds the concurrent access threshold set according to the physical conditions of the line, the target node is determined to enter a transient impact state. An example of an extreme load-bearing scenario is provided: "50 7kW charging piles starting up simultaneously in one second." The results in the figure illustrate the feasibility and response characteristics of this technical mechanism.

[0096] refer to Figure 4 The corresponding scenario is as follows: During the midday hours of summer, photovoltaic (PV) output is high, leading to pressure on renewable energy absorption in the distribution area, and retail spot electricity prices quickly enter a low-price range. Upon receiving this low-price information, a large number of electric vehicle charging stations with price responsiveness activate charging within a short period, causing a rapid increase in the branch current of the target transformer area. This scenario directly corresponds to the typical application scenario in the application document: "Summer PV power generation leads to a midday drop in spot electricity prices, inducing a concentrated influx of charging load."

[0097] like Figure 4 As shown in the figure, the electricity price curve first forms a clear trough, followed by a rapid increase in branch current, while the current change gradient produces a significant peak. This indicates that the system can not only acquire the market-side leading signal of "electricity price decline," but also further determine whether a high-concurrency impact on the electrical side has truly occurred through the current change gradient. Therefore, the criterion is not simply to directly initiate current limiting based on low electricity prices, but rather to combine market price changes with the actual branch current response.

[0098] Understandably, identifying transient impacts before a large influx of flexible loads causes sustained line overload allows for response time in subsequent filtering, loss gradient calculation, dynamic control biasing, and load reduction scheduling, thereby improving the ability to identify high-concurrency loads induced by low prices. The ultimate goal of the application is precisely to address the high-concurrency impact of flexible loads caused by fluctuations in spot electricity prices.

[0099] refer to Figure 5 The corresponding scenario is as follows: the target area experiences two independent low-price windows within a single day. For example, the first low-price period occurs in the morning due to increased output from new energy sources, while the second price decline occurs in the evening due to changes in local supply and demand. Price-sensitive charging loads are concentrated in both low-price periods, resulting in two independent branch current surges.

[0100] like Figure 5 As shown in the figure, two electricity price troughs can be observed, each accompanied by an increase in branch current and a peak in the current change gradient. The system recovers to a relatively stable state between the two events, indicating that the detection method is not limited to detecting only one anomaly, but can repeatedly perform the "electricity price decline - gradient calculation - impact judgment" process for consecutive market-side events.

[0101] Understandably, when low-price signals occur multiple times within a day, it's possible to determine whether transient impact conditions have been met by analyzing each actual load response individually, rather than simply dividing the entire day into a single low-price state. This facilitates more granular dynamic control of node loads.

[0102] refer to Figure 6 The corresponding scenario is that spot electricity prices do not drop instantly, but rather decrease slowly and continuously over a period of time. Different charging operators, smart home appliances, or energy storage devices have different start-up price thresholds, so the load change is small when the electricity price first begins to fall; when the electricity price drops to the economic operating threshold set by some devices, a large number of devices begin to be put into operation, resulting in a significant price response lag.

[0103] like Figure 6 As shown in the figure, the key point is not that the lowest point of electricity price coincides with the highest point of current, but that there is a time shift between the two: the electricity price drops first, and the branch current then increases rapidly, and the peak value of the current change gradient also shifts accordingly.

[0104] Understandably, the system did not use the simplistic logic of "identifying a line impact as a decrease in electricity price," but instead continued to monitor actual branch current changes after receiving a signal of a price decline. The application document specifies this two-stage judgment mechanism: first, obtaining a signal of a downward fluctuation in retail spot electricity prices, and then calculating the gradient of branch current changes. Therefore, it reduces the possibility of prematurely initiating congestion control when electricity prices change but terminal loads have not yet started up significantly, making transient impact judgments closer to the actual electrical conditions.

[0105] refer to Figure 7 The corresponding scenario is as follows: Due to factors such as a short-term surge in renewable energy output and changes in market clearing results, spot electricity prices experience a rapid, brief decline, followed by a swift recovery. Some highly automated charging devices activate immediately upon receiving the price signal, thus creating short-duration but rapidly increasing current pulses in the branch circuits.

[0106] like Figure 7 As shown in the figure, the price decline lasted for a short period of time, and the branch current did not form a long-term high plateau, but showed obvious short-term shocks; at the same time, the current change gradient formed a relatively high peak.

[0107] Understandably, this result demonstrates the significance of the proposed solution using the rate of change of current, rather than simply the current amplitude, as a parameter for identifying transient impacts. Even if the branch current does not remain high for an extended period, a rapid increase within a short sampling period will still result in a significant increase in the real-time gradient, thus enabling the identification of high-concurrency access events characterized by "short duration and rapid rise." This is beneficial in preventing the traditional static current threshold from lagging in responding to rapid transient events. The application explicitly defines the real-time gradient as the ratio of the difference in effective current values ​​between adjacent sampling periods to the time step.

[0108] refer to Figure 8 The corresponding scenario is that after a drop in spot prices, the market does not immediately recover but maintains a relatively stable low-price platform for several hours. Different users, charging pile groups, and flexible loads are put into operation sequentially due to different control strategies and start-up conditions, resulting in multiple batches of load access.

[0109] like Figure 8 As shown in the figure, the electricity price remained low for a certain period of time, while the branch current did not jump all at once, but increased in multiple stages and steps, resulting in multiple local change gradient peaks at different access stages.

[0110] Understandably, this solution doesn't assess the "duration of low prices," but rather each actual rapid load change occurring under low-price conditions. Therefore, even if electricity prices remain constant for an extended period, new transient impacts can still be re-identified through branch current gradients when a new batch of charging stations starts up. The beneficial effect is improved time resolution for batches of concurrent loads, avoiding the problem of "not sensing subsequent new loads after the initial identification" due to relying solely on price status.

[0111] refer to Figure 9 The corresponding scenario is as follows: after the low-price period ends, the spot electricity price gradually rebounds, but the electric vehicle charging piles and energy storage charging equipment that have already been started will not immediately stop operating, but will continue to complete the established charging tasks, thus showing obvious load inertia.

[0112] like Figure 9 As shown in the figure, electricity prices have begun to rise, but branch currents remain at a high level and have not decreased in tandem with prices. This indicates that market price conditions and physical load conditions are not always synchronized.

[0113] Understandably, this diagram further supports the rationale behind the technical solution's design where "market-side signals serve only as trigger points, while real-time electrical status is used as the basis for impact judgment." If congestion risk is judged solely based on price, the possibility of underestimating the high load on the line after price recovery might be underestimated. This method, however, continues to use branch currents and their gradient changes, maintaining awareness of the actual line status. This diagram demonstrates the advantage of the technical solution's shift from simple economic scheduling to a joint judgment based on "market signals + physical status," and also echoes the application's emphasis on the problem that traditional commercial economic algorithms are prone to detaching from the underlying physical reality.

[0114] refer to Figure 10 The corresponding scenario is: the retail spot price fluctuates frequently within a certain sensitive range, and multiple price-responsive loads repeatedly enter or exit the charging state according to their respective thresholds, causing multiple short-term, discrete cluster changes in the branch current.

[0115] like Figure 10 As shown, unlike a single low price trough, this graph represents multiple local price fluctuations and corresponding peaks in current change gradients. Some price changes only cause slight current changes, while at other times they produce significant high-gradient shocks.

[0116] Understandably, the system can use a preset concurrent access threshold to filter current changes of different intensities. Not all market price fluctuations are necessarily judged as transient impacts; only when the actual branch current change gradient reaches the line's physically permissible ramp-up boundary will it enter the subsequent control process. In the application document, this threshold is determined by combining the conductor cross-sectional area and the thermal time constant. This figure mainly illustrates the improvement of impact identification selectivity and the reduction of the possibility of unnecessary scheduling actions caused by minor market fluctuations.

[0117] refer to Figure 11 This corresponds to a scenario where changes in renewable energy output or regional load structure cause the low-price window, which is usually during the midday hours, to shift to the morning. Consequently, price-sensitive charging devices in residential or commercial areas are connected to the grid earlier.

[0118] like Figure 11 As shown, the overall response pattern of the curve still exhibits the pattern of "price decrease - branch current increase - change gradient increase", but the location of the event has shifted forward overall.

[0119] Understandably, the goal is not to prove that low electricity prices always exist in the morning, but rather to verify that the proposed method does not rely on fixed periods of low prices or manually set time windows. As long as the trading platform shows a signal of actual declining electricity prices and the gradient of branch current changes exceeds the concurrency threshold, transient impacts on the target node can be identified. This demonstrates that the method can adapt to high-concurrency access events that occur at constantly changing times under different dates, different renewable energy outputs, and different user electricity consumption behaviors, exhibiting better adaptability to operating conditions.

[0120] refer to Figure 12 The corresponding scenario is as follows: after the retail electricity price drops in the evening, some electric vehicles do not charge immediately. Instead, they are affected by the reservation charging strategy, the scheduling of charging operators, or the time set by users, and are only put into use after the price has dropped for a period of time.

[0121] like Figure 12 As shown in the figure, the low-price range appears first, while the current rise and the peak of the gradient lag significantly. This further highlights the "long-term delayed response" compared to the gradually decreasing electricity price lag access graph.

[0122] Understandably, price signals alone cannot accurately determine the actual timing of electrical surges. This technical solution, by real-time acquisition of branch currents and calculation of the gradient of changes between adjacent sampling periods, can pinpoint the true transient surge moment to the actual concentrated load startup phase, rather than simply to the phase of electricity price changes. This diagram demonstrates the improved accuracy of surge moment identification, providing a more reasonable trigger point for subsequent startup filtering, thermal risk calculation, and current limiting scheduling.

[0123] refer to Figure 13The corresponding scenario is as follows: the retail spot price first drops to the first-tier low-price range, triggering the connection of some flexible loads with high price sensitivity; then the electricity price drops further, triggering the connection of a second batch of charging loads with lower price sensitivity. Thus, a typical "two-tier price stimulus - two batches of load response" scenario is formed.

[0124] like Figure 13 As shown in the figure, the electricity price exhibits a two-level decreasing process, and the branch current correspondingly forms a two-level step-like increase, while generating two current change gradient peaks of different intensities.

[0125] Understandably, this diagram demonstrates the potential of the technical solution to differentiate between concurrent access events of varying scales. If the current gradient of the first batch of loads remains below the physical concurrency threshold, normal operation can continue. Only when the second batch of loads continues to access the system, causing the real-time gradient to reach or exceed the threshold, does the system determine that the target node is in a transient impact state and enter physical congestion control. This aligns closely with the technical principle in the application document: "setting concurrent access thresholds based on the physical carrying capacity of the line, rather than issuing dispatch instructions uniformly based solely on economic prices." Its beneficial effect is establishing a link between the control triggering conditions and the actual carrying capacity of the underlying line, thereby avoiding premature or delayed control.

[0126] It should be noted that by using the downward signal of retail spot electricity price as the pre-trigger information for high-concurrency access of flexible loads, and further utilizing the real-time change gradient of branch current and the concurrent access threshold set based on the physical carrying capacity of the line for secondary judgment, it is possible to identify transient load impacts under different price fluctuation modes, different access times, different response lags, and different concurrency scales. This provides an accurate triggering basis for subsequent electrical data filtering, active power loss gradient calculation, dynamic control bias generation, and current limiting and load reduction. This approach aims to address the high-concurrency impact of flexible loads caused by spot electricity price fluctuations and avoids the problem of traditional methods that rely solely on fixed electrical thresholds or economic algorithms while ignoring the underlying physical carrying capacity.

[0127] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a retail spot electricity load dispatching device that takes into account distribution network physical congestion. These device embodiments are similar to... Figure 1 Corresponding to the method embodiments shown, the retail spot power load dispatching device that takes into account the physical congestion of the distribution network can be specifically applied to various electronic devices.

[0128] like Figure 2As shown, the retail spot power load dispatching device 200, which takes into account physical congestion in the distribution network, includes: a filtering and denoising unit 201, a first generation unit 202, a second generation unit 203, an instruction generation unit 204, a result comparison unit 205, and a drive control unit 206. The filtering and denoising unit 201 is configured to: in response to a target distribution node being in a transient impact state, perform filtering and denoising processing on the branch current sequence and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information. The first generation unit 202 is configured to: generate active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the net load current characteristic information. The second generation unit 203 is configured to: generate a dynamic control bias factor based on the steady-state voltage characteristic information and the net load current characteristic information. The instruction generation unit 204 is configured to: generate a current-limiting scheduling signal based on the aforementioned active power loss gradient information and the aforementioned dynamic control bias factor; and perform dimensionality reduction mapping processing on the aforementioned current-limiting scheduling signal to obtain a load-reducing scheduling instruction, wherein the aforementioned load-reducing scheduling instruction is an instruction for the underlying electronic switching device to perform power limiting. The result comparison unit 205 is configured to: send the aforementioned load-reducing scheduling instruction to the aforementioned underlying electronic switching device to obtain load-reducing electrical state information; and perform a safety envelope comparison based on the aforementioned load-reducing electrical state information and the local bus transient voltage data to obtain a comparison result. The drive control unit 206 is configured to: generate a drive instruction in response to the aforementioned comparison result satisfying the underlying physical defense failure condition, to drive the shunt trip coil to force the mechanical contacts to physically separate.

[0129] It is understandable that the units described in the retail spot power load dispatching device 200, which takes into account physical congestion of the distribution network, are related to the reference... Figure 1 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the retail spot power load dispatching device 200 and its constituent units, taking into account physical congestion in the distribution network, and will not be repeated here.

[0130] The following is for reference. Figure 3 It shows a method suitable for implementing this disclosure. Figure 1 A schematic diagram of the structure of an electronic device 300 in some corresponding embodiments. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0131] like Figure 3As shown, the electronic device 300 may include a processing unit 301, which can perform various appropriate actions and processes according to a program stored in the read-only memory 302 or a program loaded from the storage device 308 into the random access memory 303. The random access memory 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, the read-only memory 302, and the random access memory 303 are interconnected via a bus 304. An input / output interface 305 is also connected to the bus 304.

[0132] Typically, the following devices can be connected to the input / output interface 305: input devices 306 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 308 including, for example, magnetic tape, hard disk, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0133] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 309, or installed from a storage device 308, or installed from a read-only memory 302. When the computer program is executed by the processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0134] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0135] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0136] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: respond to a target distribution node being in a transient impact state by filtering and denoising the branch current sequence and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information; generate active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the aforementioned net load current characteristic information; generate a dynamic control bias factor based on the aforementioned steady-state voltage characteristic information and the aforementioned net load current characteristic information; and generate a dynamic control bias factor based on the aforementioned... The power loss gradient information and the aforementioned dynamic control bias factor are used to generate a current limiting scheduling signal. The current limiting scheduling signal is then subjected to dimensionality reduction mapping to obtain a load reduction scheduling instruction. This load reduction scheduling instruction is an instruction for the underlying electronic switching device to perform power limiting. The load reduction scheduling instruction is then sent to the underlying electronic switching device to obtain load reduction electrical status information. Based on the load reduction electrical status information and the local bus transient voltage data, a safety envelope comparison is performed to obtain a comparison result. In response to the comparison result satisfying the underlying physical defense failure condition, a drive instruction is generated to drive the shunt trip coil to force the mechanical contacts to physically separate.

[0137] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0138] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0139] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a filtering and denoising unit, a first generation unit, a second generation unit, an instruction generation unit, a result comparison unit, and a drive control unit. The names of these units do not necessarily limit the specific unit; for example, the filtering and denoising unit may also be described as "in response to a target distribution node being in a transient impact state, performing filtering and denoising processing on the branch current sequence and three-phase bus voltage data to obtain steady-state voltage characteristic information and net load current characteristic information."

[0140] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0141] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for dispatching retail spot electricity loads considering physical congestion in the distribution network, characterized in that, include: In response to the target distribution node being in a transient impact state, the branch current sequence and three-phase bus voltage data are filtered and denoised to obtain steady-state voltage characteristic information and net load current characteristic information. Based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the net load current characteristic information, active power loss gradient information is generated. Based on the steady-state voltage characteristic information and the net load current characteristic information, a dynamic control bias factor is generated; Based on the active power loss gradient information and the dynamic control bias factor, a current limiting scheduling signal is generated, and the current limiting scheduling signal is subjected to dimensionality reduction mapping processing to obtain a load reduction scheduling instruction. The load reduction scheduling instruction is an instruction for the underlying electronic switching devices to perform power limiting. The step of obtaining the load reduction scheduling instruction includes: Determine the node voltage and power characteristics of the active power loss gradient information in the physical spatial topology of the distribution network; Spatial topology optimization and safety boundary pruning are performed on the node voltage and power characteristics and the dynamic control bias factor to obtain the current limiting scheduling signal; The instantaneous communication bandwidth margin and local controller computing power limit parameters of the underlying electronic switching device to be executed by the load reduction scheduling command are obtained, as well as the physical switching frequency limit of the insulated gate bipolar transistor in the underlying electronic switching device. Based on the instantaneous communication bandwidth margin and the local controller's computing power limit parameter, a discretized sampling step size is generated to map the underlying hardware data throughput bottleneck; Based on the discretized sampling step size, the current limiting scheduling signal is subjected to time-domain dimensionality reduction processing to obtain the active power reduction action feature sequence. Based on the physical switching frequency limit, the active power reduction action characteristic sequence is subjected to pulse width modulation processing to obtain the load reduction scheduling command. The load reduction scheduling command is sent to the underlying electronic switching device to obtain load reduction electrical status information, and a safety envelope comparison is performed based on the load reduction electrical status information and the local bus transient voltage data to obtain the comparison result; In response to the comparison result satisfying the underlying physical defense failure condition, a drive command is generated to drive the shunt trip coil to force the mechanical contacts to physically separate.

2. The retail spot power load dispatching method considering distribution network physical congestion according to claim 1, characterized in that, Prior to the step of responding to a transient impact state at the target distribution node, the method further includes: Obtain signals of downward fluctuations in retail spot electricity prices; In response to the downward fluctuation signal of the retail spot electricity price, the real-time change gradient of the branch current sequence is determined; In response to the real-time change gradient being greater than a preset concurrent access threshold, it is determined that the target power distribution node is in a transient impact state.

3. The retail spot power load dispatching method considering distribution network physical congestion according to claim 1, characterized in that, The step of generating active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the net load current characteristic information, includes: Extract the effective value sequence from the net load current feature information; Based on the effective value sequence and the temperature coefficient of the conductor resistance, a dynamic temperature rise parameter is generated; Based on the dynamic temperature rise parameter, the resistance and reactance parameter is subjected to temperature compensation processing to obtain the corrected resistance parameter; The transient active power loss is obtained by multiplying the square of the effective value sequence with the corrected resistance parameter. The rate of change of the transient active power loss within a preset time sliding window is determined to obtain the active power loss gradient information.

4. The retail spot electricity load dispatching method considering distribution network physical congestion according to claim 1, characterized in that, The step of generating the dynamic control bias factor based on the steady-state voltage characteristic information and the net load current characteristic information includes: Obtain the rated distribution voltage and physical current carrying capacity of the hardware device corresponding to the target distribution node, as well as the physical penalty weight coefficient matrix associated with the topology level and insulation aging state of the target distribution node; Based on the steady-state voltage characteristic information and the rated distribution voltage, a first drop deviation value is generated; Based on the net load current characteristic information and the physical current carrying capacity upper limit parameter, an over-limit current residual sequence is generated; The second over-limit deviation value is obtained by performing time-series integration on the over-limit current residual sequence. Based on the physical penalty weight coefficient matrix, the first drop deviation value and the second over-limit deviation value are subjected to nonlinear fusion processing to obtain the dynamic control bias factor.

5. The retail spot power load dispatching method considering distribution network physical congestion according to claim 4, characterized in that, The step of associating the physical penalty weighting coefficient matrix of the target distribution node topology level and insulation aging state includes: Extract the spatial hierarchy parameters of the target distribution node in the physical topology of the distribution network; Based on the aforementioned spatial hierarchy parameters, a spatial penalty weight base for mapping the physical spread risk of distribution network congestion is generated; Extract the historical operating temperature sequence and ambient humidity characteristics of the physical conductors of the target power distribution node; Based on the historical operating temperature sequence and the environmental humidity characteristics, a physical insulation aging derating factor is generated. The boundary penalty operator matrix is ​​obtained by performing tensor product mapping on the spatial penalty weight base and the physical insulation aging derating coefficient. The boundary penalty operator matrix is ​​normalized to obtain the physical penalty weight coefficient matrix.

6. The retail spot electricity load dispatching method considering distribution network physical congestion according to claim 1, characterized in that, The steps of comparing the safety envelope based on the load reduction electrical state information and the local bus transient voltage data to obtain the comparison result, and generating a drive command in response to the comparison result satisfying the underlying physical defense failure condition to drive the shunt trip coil to force the mechanical contacts to physically separate, include: Obtain the thermal stability current limit parameter and the transient voltage drop threshold; Based on the thermal stability current limit parameter and the transient voltage drop threshold, a safety envelope is generated. Determine the cumulative current over-limit value in the descent electrical status information; Based on the safety envelope, the cumulative value of current exceeding the limit and the transient voltage data of the local bus are subjected to gate logic cross comparison processing to obtain the comparison result; In response to the comparison result satisfying the underlying physical defense failure condition, the electromagnetic excitation inductance parameters of the shunt trip coil and the static holding resistance of the closing spring of the mechanical contact are determined. Electromagnetic tripping mapping is performed on the electromagnetic excitation inductance parameters and the static holding resistance of the closed spring to obtain the target tripping current parameters; Based on the target tripping current parameter, a strong high-level signal is generated, wherein the strong high-level signal is used to bypass the operating system kernel; The high-level signal is photoelectrically isolated and driven to obtain a drive command, which triggers the shunt trip coil to generate a transient electromagnetic attraction force, forcing the mechanical contacts to physically separate.

7. Retail spot power load dispatching devices that take into account physical congestion in the distribution network, including: The filtering and denoising unit is configured to filter and denoise the branch current sequence and three-phase bus voltage data in response to the target power distribution node being in a transient impact state, so as to obtain steady-state voltage characteristic information and net load current characteristic information. The first generation unit is configured to generate active power loss gradient information based on the received resistance and reactance parameter information and conductor resistance temperature coefficient corresponding to the target distribution node, as well as the net load current characteristic information. The second generation unit is configured to generate a dynamic control bias factor based on the steady-state voltage characteristic information and the net load current characteristic information. The instruction generation unit is configured to generate a current-limiting scheduling signal based on the active power loss gradient information and the dynamic control bias factor, and to perform dimensionality reduction mapping processing on the current-limiting scheduling signal to obtain a load-reducing scheduling instruction. The load-reducing scheduling instruction is an instruction for the underlying electronic switching devices to perform power limiting. The step of obtaining the load-reducing scheduling instruction includes: Determine the node voltage and power characteristics of the active power loss gradient information in the physical spatial topology of the distribution network; Spatial topology optimization and safety boundary pruning are performed on the node voltage and power characteristics and the dynamic control bias factor to obtain the current limiting scheduling signal; The instantaneous communication bandwidth margin and local controller computing power limit parameters of the underlying electronic switching device to be executed by the load reduction scheduling command are obtained, as well as the physical switching frequency limit of the insulated gate bipolar transistor in the underlying electronic switching device. Based on the instantaneous communication bandwidth margin and the local controller's computing power limit parameter, a discretized sampling step size is generated to map the underlying hardware data throughput bottleneck; Based on the discretized sampling step size, the current limiting scheduling signal is subjected to time-domain dimensionality reduction processing to obtain the active power reduction action feature sequence. Based on the physical switching frequency limit, the active power reduction action characteristic sequence is subjected to pulse width modulation processing to obtain the load reduction scheduling command. The result comparison unit is configured to send the load reduction scheduling command to the underlying electronic switching device to obtain load reduction electrical status information, and to perform a safety envelope comparison based on the load reduction electrical status information and the local bus transient voltage data to obtain the comparison result; The drive control unit is configured to generate a drive command in response to the comparison result satisfying the underlying physical defense failure condition, so as to drive the shunt trip coil to force the mechanical contacts to physically separate.

8. An electronic device, characterized in that, include: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the retail spot power load dispatching method that takes into account distribution network physical congestion as described in any one of claims 1-6.

9. A computer-readable medium, characterized in that, It stores a computer program, wherein when the program is executed by a processor, it implements the retail spot power load dispatching method as described in any one of claims 1-6, taking into account the physical congestion of the distribution network.

Images