Power distribution network voltage control method, device and equipment based on working condition state

Abstract

The application provides a power distribution network voltage control method, device and equipment based on working condition state, and relates to the technical field of power distribution network control. The method comprises the following steps: identifying candidate working condition states of the power distribution network in a current control period based on photovoltaic output states and voltage states of the power distribution network in the current control period; each working condition state corresponds to a control target; performing de-bouncing confirmation on the candidate working condition states in the current control period based on the candidate working condition states of the power distribution network in historical control periods, so as to obtain a final working condition state of the power distribution network in the current control period; calculating active power instructions and reactive power instructions of the power distribution network in the current control period based on a PID parameter set corresponding to the final working condition state; and distributing the active power instructions and the reactive power instructions of the power distribution network in the current control period based on a battery state of charge and a battery converter reactive power demand. The application can realize stable, continuous and self-adaptive adjustment performance in the full working condition range of power distribution network voltage control.

Description

Technical Field

[0001] This invention relates to the field of power distribution network control technology, and in particular to a power distribution network voltage control method, device, and equipment based on operating conditions. Background Technology

[0002] With the increasing penetration rate of distributed photovoltaic (PV) power in distribution networks, the operating characteristics of these networks have undergone profound changes. The randomness and volatility of PV output cause frequent changes in the power distribution of distribution networks, leading to increasingly prominent voltage exceeding-limit problems, especially in rural distribution networks where loads are dispersed, line impedance is high, and the power imbalance between PV output and load demand often results in significant fluctuations in bus voltage.

[0003] Currently, the nine-zone diagram method is widely used for reactive power and voltage control in distribution networks. This method divides the operating plane into nine zones based on measured voltage and reactive power values, with each zone corresponding to a fixed set of control strategies. However, the inherent drawback of the nine-zone diagram method is that its fixed zone boundaries cannot adapt to changes in system characteristics under different photovoltaic output and load conditions. The coefficients in the control command formulas are mostly empirical values, lacking a systematic tuning basis for specific operating conditions. Another inherent drawback is that each control cycle independently determines the zone and executes the corresponding strategy based on the current measured value, lacking the ability to remember historical judgments. When the voltage measurement value changes slightly near the zone boundary due to noise or short-term fluctuations, the control strategy may switch back and forth between two zones, leading to frequent switching of compensation devices such as capacitors, increasing equipment wear and reducing system stability. To address this issue, existing solutions have attempted to mitigate the phenomenon by introducing fuzzy boundaries or dynamically adjusting the zoning thresholds of the nine-zone diagram. However, these solutions still rely on current or historical voltage measurements for zoning adjustments, essentially remaining a mode of independently determining the zone and executing the strategy accordingly in each control cycle. They only change the specific location of the zone without altering the instantaneous nature of the zoning determination itself or the resulting strategy jump characteristics.

[0004] Furthermore, in existing active and reactive power compensation devices, active and reactive power commands are typically generated by a unified optimization model or decision logic, and are subject to the same capacity constraints. When the active power output capacity of battery energy storage decreases due to exceeding the state of charge limit or module failure, the reactive power compensation command is proportionally reduced along with the active power reduction due to the same capacity constraint. The actual demand for reactive power compensation in the distribution network depends on voltage deviation and power factor targets, and is not directly related to the active power output capacity of the battery. This proportional reduction method under coupled constraints lacks physical basis. In actual operation, when the battery is fully charged and can no longer absorb active power, the reactive power compensation capacity is also reduced due to the same constraint. At this time, the voltage exceeding the limit problem cannot be effectively suppressed, making it difficult to maintain the continuity of reactive power compensation. Summary of the Invention

[0005] This invention provides a distribution network voltage control method, apparatus, and equipment based on operating conditions to solve the problem of frequent switching of distribution network voltage control strategies near the boundary and difficulty in maintaining the continuity of reactive power compensation.

[0006] In a first aspect, embodiments of the present invention provide a distribution network voltage control method based on operating conditions, comprising: Based on the photovoltaic output status and voltage status of the distribution network in the current control cycle, candidate operating conditions of the distribution network in the current control cycle are identified; where each operating condition corresponds to a control target. Based on the candidate operating conditions of the distribution network in the historical control cycle, the candidate operating conditions of the current control cycle are de-jittered and confirmed to obtain the final operating condition of the distribution network in the current control cycle. Based on the PID parameter set corresponding to the final operating condition, the active power command and reactive power command of the distribution network in the current control cycle are calculated. Based on the battery state of charge and the reactive power demand of the battery converter, the active power command and reactive power command of the distribution network in the current control cycle are allocated.

[0007] Secondly, embodiments of the present invention provide a distribution network voltage control device based on operating conditions, comprising: The status recognition module is used to identify candidate operating conditions of the distribution network in the current control cycle based on the photovoltaic output status and voltage status of the distribution network in the current control cycle; where each operating condition corresponds to a control target. The debouncing confirmation module is used to confirm the candidate operating conditions of the distribution network in the current control cycle based on the candidate operating conditions of the distribution network in the historical control cycle, so as to obtain the final operating condition of the distribution network in the current control cycle. The instruction calculation module is used to calculate the active power instruction and reactive power instruction of the distribution network in the current control cycle based on the PID parameter set corresponding to the final operating condition. The instruction allocation module is used to allocate active power instructions and reactive power instructions to the distribution network in the current control cycle based on the battery state of charge and the reactive power demand of the battery converter.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0009] This invention identifies candidate operating conditions based on photovoltaic power output and voltage status, enabling the control strategy to adaptively adjust to changes in actual operating conditions, eliminating strategy mismatch issues at fixed partition boundaries under different operating conditions. By debouncing and confirming the current candidate state based on historical candidate states, the operating condition determination possesses time-dimensional state memory capabilities, eliminating strategy jumps near boundaries caused by measurement noise and short-term fluctuations, avoiding frequent switching of compensation devices, reducing equipment wear, and improving system stability. By independently calculating active and reactive power commands based on the PID parameter set corresponding to the final operating condition, active and reactive power regulation can employ matching control parameters under each operating condition, achieving condition-adaptive optimization of regulation quality. By allocating active power commands based on battery state of charge and reactive power commands based on converter reactive power demand, active and reactive power are allocated separately at the execution level according to their independent physical bases, avoiding the problem of reactive power being reduced along with active power when it is limited, ensuring the effectiveness of voltage regulation, thereby achieving stable, continuous, and adaptive regulation performance across the entire operating range of distribution network voltage control. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the implementation of the power distribution network voltage control method based on operating conditions provided in this embodiment of the invention. Figure 2 This is a schematic diagram of the structure of the distribution network voltage control device based on operating conditions provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0012] See Figure 1 The document illustrates a flowchart of the implementation of a power distribution network voltage control method based on operating conditions, as provided in an embodiment of the present invention. Details are as follows: Step 101: Based on the photovoltaic output status and voltage status of the distribution network in the current control cycle, identify the candidate operating conditions of the distribution network in the current control cycle; wherein, each operating condition corresponds to a control target.

[0013] In this embodiment, the operating conditions of the distribution network are determined by two key physical quantities: the photovoltaic output status reflects the power balance between distributed power sources and loads, and the voltage status reflects whether the power quality is up to standard. Different operating conditions formed by the combination of these two dimensions require differentiated voltage control strategies. Specifically, the three possible photovoltaic output statuses and three possible voltage statuses combine to form nine regular operating conditions, plus one independent emergency operating condition, for a total of ten candidate operating conditions. Each operating condition corresponds to a specific control objective, such as whether the battery needs to absorb or release active power, or whether the capacitor needs to generate or absorb reactive power. The complete combination space formed by the photovoltaic output status and voltage status ensures that any measured operating condition falls into a unique candidate state, eliminating any coverage blind spots.

[0014] Step 102: Based on the candidate operating conditions of the distribution network in the historical control cycle, perform debouncing confirmation on the candidate operating conditions of the current control cycle to obtain the final operating conditions of the distribution network in the current control cycle.

[0015] In this embodiment, measurement noise is unavoidable in voltage transformers, current transformers, and analog-to-digital converters in the distribution network. Furthermore, instantaneous load fluctuations and short-term pulsations in photovoltaic output can cause single measurements to deviate from the actual operating conditions. If the candidate operating conditions identified in each control cycle in step 101 are directly used as the final state to execute the control strategy, measurement noise and short-term disturbances will directly lead to frequent switching of the control strategy, causing repeated switching of the compensation device. This not only reduces equipment lifespan but may also cause system oscillations. Therefore, this step introduces a debouncing confirmation mechanism with state memory capabilities.

[0016] The core idea of ​​debouncing confirmation is: a single occurrence of a candidate state change is considered a possible disturbance or noise and is not accepted; only when the same candidate state appears consistently and stably is the change in operating condition confirmed. Specifically, for non-urgent candidate operating conditions, the system maintains a debouncing counter. When the candidate operating condition identified in the current control cycle is the same as that in the previous control cycle, the debouncing counter increments; when the counter accumulates to a preset value of 2, meaning the same candidate state has been identified for two consecutive control cycles, the candidate state is then confirmed as the final operating condition of the current control cycle. If the candidate operating condition of the current control cycle differs from that of the previous control cycle, the debouncing counter is reset, maintaining the final operating condition of the previous control cycle unchanged.

[0017] When debouncing confirms a state transition, the system automatically loads the control parameters corresponding to the new state and clears the integral accumulation of the previous state to prevent unnecessary overshoot caused by historical integral values ​​in the new state. Simultaneously, for the highest priority emergency conditions, debouncing is skipped and the system is directly confirmed as the final operating state to ensure the fastest possible response in the event of severe voltage exceedances.

[0018] Step 103: Based on the PID parameter set corresponding to the final operating condition, calculate the active power command and reactive power command of the distribution network in the current control cycle.

[0019] In this embodiment, the impact mechanisms and response characteristics of active and reactive power on voltage differ in distribution network voltage control. Specifically, active power affects voltage by changing the voltage drop across the line, and its response speed is limited by the power response capability of the battery converter. Reactive power, on the other hand, directly affects voltage by compensating for reactive power flow in the line, and its response speed is limited by capacitor switching and the reactive power output capability of the converter. Therefore, this step assigns active and reactive power commands to two completely independent PID controllers for calculation. The input quantities, control parameters, and output limits of the two controllers are set independently and do not affect each other. Furthermore, the integral terms of the two controllers are selectively cleared to zero during state transitions and frozen when capacity constraints are triggered, avoiding control quality deterioration caused by integral saturation.

[0020] Step 104: Based on the battery state of charge and the reactive power demand of the battery converter, allocate the active power command and reactive power command of the distribution network in the current control cycle.

[0021] In this embodiment, the active power command P generated in step 103 is the total active power command of the entire battery pack, which needs to be undertaken by multiple battery modules; the reactive power command Q is the total reactive power command of the system, which needs to be output collaboratively by the capacitors and the battery converter. This step decomposes the system-level command into specific commands for each execution unit based on the battery state of charge and the reactive power demand of the battery converter, ensuring that the decomposition process fully considers the real-time state and physical constraints of each execution unit.

[0022] This invention identifies candidate operating conditions based on photovoltaic power output and voltage status, enabling the control strategy to adaptively adjust to changes in actual operating conditions, eliminating strategy mismatch issues at fixed partition boundaries under different operating conditions. By debouncing and confirming the current candidate state based on historical candidate states, the operating condition determination possesses time-dimensional state memory capabilities, eliminating strategy jumps near boundaries caused by measurement noise and short-term fluctuations, avoiding frequent switching of compensation devices, reducing equipment wear, and improving system stability. By independently calculating active and reactive power commands based on the PID parameter set corresponding to the final operating condition, active and reactive power regulation can employ matching control parameters under each operating condition, achieving condition-adaptive optimization of regulation quality. By allocating active power commands based on battery state of charge and reactive power commands based on converter reactive power demand, active and reactive power are allocated separately at the execution level according to their independent physical bases, avoiding the problem of reactive power being reduced along with active power when it is limited, ensuring the effectiveness of voltage regulation, thereby achieving stable, continuous, and adaptive regulation performance across the entire operating range of distribution network voltage control.

[0023] In one possible implementation, based on the photovoltaic output status and voltage status of the distribution network in the current control cycle, candidate operating conditions of the distribution network in the current control cycle are identified, including: If the photovoltaic output is excessive and the bus voltage is overvoltage, the candidate operating condition is identified as the first operating condition, where the control objective is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is excessive and the bus voltage is normal, the candidate operating condition is identified as the second operating condition, where the control objective is for the battery to gradually absorb active power and for the capacitor to maintain the power factor. If the photovoltaic output is excessive and the bus voltage is undervoltage, the candidate operating condition is identified as the third operating condition, where the control objective is for the battery to release active power to raise voltage and for the capacitor to generate more reactive power. If the photovoltaic output is balanced and the bus voltage is overvoltage, the candidate operating condition is identified as the fourth operating condition, where the control objective is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is balanced and the bus voltage is normal, the candidate operating condition is identified as the fifth operating condition, where the control objective is for the battery to be in standby or fine-tuned and for the capacitor to maintain the power factor. If the photovoltaic output is balanced and the bus voltage is undervoltage, the candidate operating condition is identified as the sixth operating condition. The control objective of this condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the photovoltaic output is insufficient and the bus voltage is overvoltage, the candidate operating condition is identified as the seventh operating condition. The control objective of this condition is for the battery to absorb active power to lower the voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is insufficient and the bus voltage is normal, the candidate operating condition is identified as the eighth operating condition. The control objective of this condition is for the battery to gradually release active power and for the capacitor to maintain the power factor. If the photovoltaic output is insufficient and the bus voltage is undervoltage, the candidate operating condition is identified as the ninth operating condition. The control objective of this condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the bus voltage reaches the emergency overvoltage threshold or the emergency undervoltage threshold, the candidate operating condition is identified as an emergency operating condition. The control objective of this condition is for both the battery and the capacitor to respond unconditionally at full power.

[0024] In this embodiment, the ten operating states in the state space are composed of nine normal states formed by combining two dimensions: photovoltaic output state (excess / balance / insufficient) and voltage state (overvoltage / normal / undervoltage), plus an independent emergency state S6. The candidate conditions and control objectives for each state are as follows: S11 (Photovoltaic overcapacity - overvoltage): and Control objectives: Batteries absorb active power and reduce voltage; capacitors absorb excess reactive power. S12 (Photovoltaic excess - normal voltage): and Control objectives: Gradually absorb active power from the battery, maintain power factor using the capacitor. S13 (PV excess - undervoltage): and Control objectives: Battery releases active power to boost voltage, capacitors increase reactive power. S21 (Photovoltaic balance - overvoltage): and Control objectives: Batteries absorb active power and reduce voltage; capacitors absorb excess reactive power. S22 (Photovoltaic balance - normal voltage): and Control objectives: Battery standby or fine-tuning, capacitor maintains power factor. S23 (PV balance - undervoltage): and Control objectives: Battery releases active power to boost voltage, capacitors increase reactive power. S31 (Insufficient photovoltaic power - Overvoltage): and Control objectives: The battery absorbs active power, causing voltage reduction; the capacitor absorbs excess reactive power, causing voltage loss. S32 (Insufficient photovoltaic power - Normal voltage): and Control objectives: Gradually release active power from the battery, maintain power factor using the capacitor. S33 (Insufficient Photovoltaic Power - Undervoltage): and Control objective: Battery releases active power to boost voltage, capacitors increase reactive power. S6 (Voltage Emergency): Candidate condition is... or Regardless of the photovoltaic output, S6 has the highest priority—once the candidate condition is met, debouncing is immediately skipped, and a direct forced transition from any state to S6 is initiated. Control objective: Both the battery and capacitor achieve full-power unconditional response. , .in This is a system-level active power command (a positive value indicates that the battery absorbs active power from the grid, and a negative value indicates that the battery releases active power to the grid). This is a system-level reactive power command (positive values ​​indicate the generation of capacitive reactive power, negative values ​​indicate the absorption of capacitive reactive power / generation of inductive reactive power), both expressed in kW / kVar. The above ten states constitute a complete state space—any set of... All measured values ​​fall within the unique state candidate conditions, and there are no coverage blind spots.

[0025] In this embodiment, the Finite State Machine (FSM) addresses the core issue of deterministically selecting the correct control strategy under multivariable composite operating conditions. For example, in the case of overvoltage, when photovoltaic power is excessive, the battery needs to absorb active power to reduce voltage (S11). When photovoltaic power is insufficient, the battery should prioritize power supply to the load rather than absorbing active power (S31). The control directions for these two scenarios are drastically different. The ten states of the FSM explicitly name each composite operating condition and map it to a set of independently tuned PID parameters, solving the problem that single-variable threshold rules (such as "high voltage → active power absorption") cannot distinguish between opposite strategies. The specific parameter selection principles for each state will be explained in subsequent steps; here, we will first explain the measurement parameters upon which the FSM relies.

[0026] in This is the power margin, used to determine the photovoltaic output status. It should be based on the actual installed capacity of photovoltaic power in the distribution area. Setting, typical value (Basis: The fluctuation amplitude of photovoltaic output typically does not exceed 10% of the installed capacity. This value has been verified in multiple distribution network voltage control projects; see the parameter setting instructions in the following examples for details.) If Unknown, then As a configurable parameter, it is set on-site: On-site maintenance personnel estimate it based on the historical maximum photovoltaic output observed in the transformer area or the inverter nameplate parameters. , and then calculate Alternatively, it can be directly configured according to 10% to 20% of the rated capacity of the transformer in the distribution area. (e.g., 200kVA transformer substation) (Can be set to 2~4kW). This value should not be modified by non-maintenance personnel to prevent accidental operation. Rated bus voltage, overvoltage threshold (Based on GB / T12325 "Power Quality - Voltage Deviation" for 10kV and below distribution networks, the positive deviation limit is +7%, with a conservative value of +5%), undervoltage threshold. (Take -5% for symmetry), Emergency Overpressure Threshold (Based on the equipment insulation withstand overvoltage limit specified in IEC 60038 + 10%), emergency undervoltage threshold (Based on the voltage negative deviation limit of -10% specified in GB / T12325).

[0027] Among the ten FSM states mentioned above, the State of Charge (SOC) refers to the ratio of the battery's current remaining charge to its rated capacity, and is the core constraint on the battery's charging and discharging capabilities. The upper and lower limits of SOC (95% / 5%) are physical boundaries—the battery can no longer absorb power when it is close to full charge, and can no longer release power when it is close to depletion.

[0028] Impact of SOC on the State Machine: SOC constraints do not change the state identification results of the FSM; they only affect the execution layer by limiting active power commands. In states where the battery needs to absorb active power (S11, S12, S21, S22, etc.), if the average SOC of all online modules is ≥95% (or the upper limit set on-site), active power commands will be... Forced zeroing, reactive power instruction Keep the parameters of the FSM state unchanged and initiate auxiliary measures (such as sending a power reduction command to the photovoltaic inverter and initiating curtailment). In states requiring the battery to release active power (S13, S23, S33, etc.), if the average SOC ≤ 5% (or the lower limit set on-site), the same applies. Forced to zero, The voltage remains unchanged and an alarm is triggered. This design ensures that the reactive power channel continues to operate independently under actual voltage conditions when the SOC exceeds the limit.

[0029] In one possible implementation, based on the candidate operating condition states of the distribution network in historical control cycles, the candidate operating condition states of the current control cycle are debouncing and confirmed to obtain the final operating condition state of the distribution network in the current control cycle. This includes: if the candidate operating condition state of the current control cycle is an emergency operating condition, then the candidate operating condition state is taken as the final operating condition state of the current control cycle; if the candidate operating condition state of the distribution network in the previous control cycle is the same as the candidate operating condition state of the current control cycle, then the candidate operating condition state is taken as the final operating condition state of the distribution network in the current control cycle; otherwise, the final operating condition state of the previous control cycle is maintained as the final operating condition state of the distribution network in the current control cycle.

[0030] In this embodiment, each control cycle, the FSM performs the following two steps: (1) reads the current measurement value and determines a unique candidate state according to the above candidate conditions. The calculation logic in this step is indeed a conditional branch, but its output is only a suggestion, not the final state. (2) Debouncing confirmation (except for S6): Compared with the previous cycle Comparison. If two consecutive control cycles Same and If the current FSM state is true, then perform a state transition (current state → ...). Otherwise, the current state remains unchanged. S6 does not undergo debouncing—once... Immediate transfer. The debouncing mechanism is the core difference between FSM and simple condition classification: condition classification outputs independently each cycle, and measurement noise directly causes output jumps; while FSM maintains state memory and only transfers after "the new state is confirmed for two consecutive cycles", and naturally has hysteresis characteristics.

[0031] In one possible implementation, based on the PID parameter set corresponding to the final operating condition, the active power command and reactive power command of the distribution network in the current control cycle are calculated, including: The voltage deviation is used as the input to the active power PID controller, and the power imbalance is used as the feedforward term. Based on the active power PID parameter set corresponding to the final operating condition, the active power command of the distribution network in the current control cycle is obtained. The reactive power deviation is used as the input to the reactive power PID controller. Based on the reactive power PID parameter set corresponding to the final operating condition, the reactive power command of the distribution network in the current control cycle is obtained.

[0032] In this embodiment, when the FSM state is officially switched, the following actions are triggered: (a) Loading the PID parameter set corresponding to the new state ( , , , , , , (b) Clear the integral accumulation of the active PID. (to prevent the integral residual from the previous state from causing overshoot in the new state); (c) reactive PID integral Handling: If the new and old states are... Different parameters Synchronous reset; if same, It remains unchanged.

[0033] In each FSM state, active power commands and reactive power commands are calculated by independent PID controllers, and their decision-making processes do not affect each other: Active power PID controller: Voltage deviation and power imbalance Introduction As a feedforward term, it is used to implement preventative control before the voltage exceeds the limit—for example, under S12 (photovoltaic excess - normal voltage). but Feedforward allows the battery to absorb active power earlier, preventing it from operating only after the voltage has risen to overvoltage. The control law is: PID parameter naming convention: subscript This represents the active power channel parameters (as opposed to reactive power channel parameters). ), / / These represent proportion, integral, and derivative, respectively. For example... The proportional gain of the active PID controller. The feedforward coefficients are taken from a typical range. The parameters are selected based on the photovoltaic prediction accuracy: the upper limit is used when the prediction accuracy is high (reliability feedforward), and the lower limit is used when the prediction accuracy is low (conservative feedforward). PID parameters are determined by looking up a table based on the FSM state. The parameter selection principle is: for overvoltage states (S11, S21, S31), a higher proportional gain is used. To accelerate active power absorption; under voltage conditions (S13, S23, S33), a higher proportional gain is taken ( To accelerate active power release; under normal voltage conditions (S12, S22, S32), a lower gain ratio is used. To achieve gradual adjustment, but the feedforward term ensures preventative action; in emergency state S6, the maximum proportional gain is taken ( And freeze the integral. When the FSM state changes, the accumulated integral of the previous state is frozen. Zeroing out prevents overshoot caused by integral saturation. This is necessary when capacity constraints are triggered. When constrained, the integral term of the active PID controller is frozen (accumulation stops) and resumes only after the constraint is lifted. Output limiting: ,in For the amplitude limiting function (if Then take ,like Then take Otherwise take ), The total active power of the battery pack is set. The input to the reactive power PID controller is the reactive power deviation. ,in A target power factor of 0.95 is a commonly used value for reactive power compensation assessment in distribution networks. It is selected based on relevant standards such as the "Technical Principles for Reactive Power Compensation Configuration in Power Systems." In actual projects, it can be configured according to the assessment requirements of the distribution area. The values ​​between [a certain range]. The control law is: The parameters are determined by looking up a table based on the FSM status: the higher gain ratio is used for overvoltage and undervoltage states. ) Use auxiliary voltage regulation, and take the standard gain when the voltage is normal. To maintain power factor. Output limiting: , .

[0034] Functional instructions Determined solely by voltage deviation, reactive power command Determined solely by reactive power deviation and power factor targets, these two are decoupled at the decision-making level. When battery capacity limitations lead to... When it cannot be fully executed, Unaffected.

[0035] In one possible implementation, based on the battery state of charge and the reactive power demand of the battery converter, the active power command and reactive power command of the distribution network in the current control cycle are allocated. This includes: for each online battery module, calculating the state of charge of the battery module and dividing it by its internal resistance to obtain the weight of the battery module; normalizing the weights of each battery module, and allocating the active power command of the distribution network in the current control cycle to each battery module based on the normalized weights; aiming to minimize the error between the actual reactive power output of the capacitor and the reactive power command, selecting the number of capacitor switching groups according to the nearest neighbor principle, allocating the reactive power command of the distribution network in the current control cycle to the capacitors, and allocating the remaining reactive power error to the battery converter.

[0036] In this embodiment, system-level active power command according to Weights are assigned to each battery module: Read all online modules and .right A first-order low-pass filter (cutoff frequency 10Hz) is applied to suppress measurement noise. Weights are then calculated. Normalization yields the distribution coefficients. Initial sub-instructions for each module Health constraints: The module's power limit is [limited to] [specific value]. ( (Rated power of the module). The module triggers local and remote alarms, prompting for replacement. Excess power is redistributed to unrestricted modules (maximum 3 iterations). Note: The linear conversion relationship between the power limit and SOH mentioned above is an engineering simplification assumption—strictly speaking, SOH reflects capacity decay (current usable capacity / rated capacity), while power capability depends on internal resistance growth and temperature rise limits; the two are not strictly a 1:1 correspondence. This scheme uses proportional conversion as a conservative approximation. In actual engineering, if a power-SOH mapping curve provided by the battery manufacturer is available, it can replace the linear conversion.

[0037] reactive power command The load is shared by capacitors and battery converters. Capacitors take priority: capacitors are switched in groups (e.g., 1kVar per group), and the number of groups switched on is selected according to the "nearest neighbor" principle to ensure optimal output. and That is, minimizing the error of the target value. , direction and Same. Residual error Powered by the battery converter: The reactive power handled by the battery converter is constrained by the converter's remaining capacity: if ,but The reactive power output of the capacitor is unaffected by the battery's state of charge (SOC), ensuring the continuity of basic reactive power compensation. (Note:) and Using numerical values ​​in calculations, The chosen values ​​are approximately reasonable for the engineering project under the voltage level of the distribution network. If the capacitor is already at full capacity ( Furthermore, the battery converter cannot provide residual reactive power due to capacity constraints, resulting in insufficient total reactive power output. If the system records a reactive power deficit alarm, it will prioritize adjusting the active power based on the voltage deviation in the next cycle or report it to the master station.

[0038] In one possible implementation, before allocating the active power command and reactive power command of the distribution network in the current control cycle based on the battery state of charge and the reactive power demand of the battery converter, the following steps are also included: If the sum of the available active power upper limits of all online battery modules is less than the active power command of the distribution network in the current control cycle, the following corrections are performed: the active power command is corrected to the sum of the available active power upper limits, while maintaining the original direction; the reactive power command remains unchanged, and the reactive power borne by the capacitor remains unchanged; if the battery converter needs to bear reactive power and its remaining capacity is insufficient, the reactive power borne by the battery converter is reduced to its remaining capacity; a power change rate limit is applied to the corrected active power command; if the change in the current cycle correction value relative to the actual value of the previous cycle exceeds the maximum ramp rate allowed by the battery module, the corrected active power command is adjusted according to the maximum ramp rate; the integral term of the active power PID controller is frozen.

[0039] In this embodiment, when the sum of the available power limits of all online battery modules Less than At that time, the capacity constraint is executed: (1) Active power instruction correction: (Active power is reduced to the actual usable capacity of the battery). (2) The reactive power command remains unchanged, and the capacitor is reactive. (3) If the battery converter needs to provide reactive power ( )and Then only the reactive power portion handled by the battery is reduced: The reactive power of the capacitor remains unchanged. (4) Power change rate limit: Corrected Compared to the previous cycle The rate of change must not exceed the maximum allowable ramp rate of the battery. (Typical values ​​for sodium-ion batteries) ,Right now (The actual value is set according to the battery manufacturer's manual). If it exceeds the limit, the ramp rate will be limited. (5) Integral freeze: When the capacity constraint is triggered, the integral term of the active PID is frozen (accumulation stops) and waits for the limit to be reached. Unfreeze after returning to normal to prevent integral saturation. (6) The corrected instructions are directly issued and executed without feedback to the FSM or PID controller.

[0040] One possible implementation also includes: recording the final operating condition state of the distribution network in each control cycle, the candidate operating condition state, whether an operating condition state transition occurs, and the reason for the operating condition state transition.

[0041] In this embodiment, the "current state, candidate state, transition status, and transition reason" are recorded in each control cycle. The system's current state at any given time and the historical path of state transitions are all queryable and auditable, supporting fault tracing and maintenance diagnosis of the distribution network voltage control strategy.

[0042] In one specific embodiment, the method provided by the present invention includes the following steps: Step 1: Data Acquisition and Operating Condition Identification 1.1 Acquire the three-phase voltage and three-phase current of the low-voltage bus in the distribution network at a sampling frequency of not less than 1 kHz. Calculate the effective value of the bus voltage after noise reduction using Kalman filtering. Active power of transformer area load and reactive power Photovoltaic power output 1.2 Collect the State of Charge (SOC), State of Health (SOH), and Internal Resistance of each battery module at a sampling frequency of not less than 100Hz. SOC is calibrated using the ampere-hour integration method plus open-circuit voltage. SOH is defined as the ratio of the current maximum available capacity to the rated capacity. 1.3 Capacitor bank status is collected: number of capacitor banks in operation, rated capacity. Current output is reactive power. .

[0043] 1.4 Operating Condition Identification – FSM Transfer Rules: The following three steps are executed every control cycle.

[0044] Step 1: Candidate State Calculation. Calculate the photovoltaic-load power imbalance. Set power margin : It should be based on the actual installed capacity of photovoltaic power in the distribution area. Setting, typical value (Based on the technical solution parameter description). If unknown, This is a configurable parameter set on-site. Set the voltage threshold: Rated bus voltage, overvoltage threshold (GB / T12325), Undervoltage threshold Emergency overpressure threshold (IEC60038), Emergency Undervoltage Threshold (GB / T12325). Based on the current measurement value, a unique candidate state is determined according to the following priority. Priority 1 (Urgent): If Skip all subsequent steps. Priority 2-4: otherwise press Partitioning .

[0045] Step 2: Debouncing Confirmation. The debouncing logic reflects the state memory characteristic of FSM: a single-cycle measurement jump will not trigger a state switch; it is only confirmed when the same candidate state appears consecutively for two cycles.

[0046] Step 3: State Transition Actions. When the state transition is confirmed, the following actions are triggered: (a) Load the PID parameter set corresponding to the new state; (b) (c) If Changes have occurred. Clear to zero; otherwise, leave unchanged.

[0047] 1.5 Impact of SOC Constraints on the State Machine: SOC constraints do not change the state identification results of the FSM; they only affect the execution layer by limiting active power commands. Specifically, in states where the battery needs to absorb active power (S11, S12, S21, S22, etc.), if the average SOC of all online modules is ≥95% (or the upper limit set on-site), the active power commands will be... Force zeroing ( ), reactive power command Maintain the original value, and simultaneously activate auxiliary measures (such as sending a power reduction command to the photovoltaic inverter and initiating curtailment). In states requiring the battery to release active power (S13, S23, S33, etc.), if the average SOC ≤ 5% (or the lower limit set on-site), the active power command will also be issued. Forced zeroing, reactive power instruction The system remains unchanged and triggers an alarm. The principle behind this design is that the SOC constraint reflects the physical boundary of the battery's charging and discharging capacity and should apply to the execution level, not the decision-making level. The FSM maintains its original state without degradation, ensuring that the reactive power channel operates independently according to the actual voltage conditions (e.g., the capacitor can still generate reactive power to assist in voltage boosting under undervoltage conditions), while avoiding the complexity of control logic introduced by frequent state switching due to SOC exceeding limits.

[0048] 1.6 Boundary Value Assignment and Debouncing Window Explanation: (a) The debouncing window takes 2 control cycles ( (a) Basis: The dominant timescale of voltage fluctuations in the distribution network is on the order of seconds (PV ramp-up, load changes). Transient fluctuations on the order of 100ms are usually caused by measurement noise or short-term disturbances and should not trigger state switching. (b) Boundary value attribution: Overvoltage condition is (Strictly greater than), normal voltage is (Including the equals sign), undervoltage is (Strictly less than). Therefore (231V) and (209V) all belong to the normal voltage range. Classifying the boundary value into the "normal" range can reduce the frequency of candidate state switching near the boundary, and work with the debouncing mechanism to reduce the risk of state oscillation.

[0049] Step 2: Active and Reactive Power Decoupling Decision This step is executed once per control cycle (typically 50ms, i.e., a 20Hz control frequency). The 50ms control cycle is chosen because the PWM carrier cycle of a power electronic converter is typically... The 50ms control cycle corresponds to 250-500 carrier cycles, which is sufficient to ensure stable execution of commands. Meanwhile, the 20Hz control frequency is much higher than the Nyquist frequency of distribution network voltage fluctuations (dominated on the second level, corresponding to <1Hz), satisfying the Shannon sampling theorem. In each FSM state, active and reactive power commands are calculated by independent PID controllers, and the two do not affect each other.

[0050] 2.1 Active PID Controller: Input includes voltage deviation and power imbalance To avoid measurement noise being amplified by the differential term and causing command jitter, during calculation... Before that, first of all Perform a first-order low-pass filter (cutoff frequency 20Hz), and the filtered error signal is denoted as... Differential terms in actual use The filter can be temporarily bypassed for 100ms after a state transition to ensure a fast response; the filter is always active during steady-state operation. The control law is: .in Feedforward coefficients (typical value) ). Introduction The feedforward term is used to achieve preventative control before the voltage exceeds the limit (e.g., in state S12). but Feedforward allows the battery to absorb active power earlier. It should be noted that the feedforward term... The magnitude (up to tens of kW) may far exceed the battery's rated power. (Typical value 4kW, corresponding to the configuration of a 4kW converter in the embodiments below;) The actual value is determined by the rated power of the converter, and a higher value can be taken in larger capacity systems. At this time, the PID feedback component is numerically submerged by the feedforward component. However, the PID feedback component still plays a key role in the following two scenarios: (a) when the photovoltaic output and load are close to balanced ( (a) When the feedforward term is close to zero, the PID feedback channel dominates the control; (b) when the feedforward exceeds the limit, quilt to At this point, the effect of the feedforward term is reflected in the correctness of direction and the speed of response—a large feedforward quantity corresponds to a working condition that strongly requires charging and discharging. Limiting to full power output is precisely the desired control behavior. When finer feedback adjustment is required in engineering, it can be... Take the smaller value (e.g.) This reduces the feedforward weight. PID parameters are determined by looking up a table based on the FSM state; example values ​​are shown in Table 1. These can be tuned as needed in practice. Table 1 When the FSM state changes, the integral accumulation of the previous state Zeroing out prevents overshoot caused by integral saturation. This is necessary when capacity constraints are triggered. When constrained, the integral term of the active PID controller is frozen (accumulation stops) and resumes only after the constraint is lifted. Output limiting: ,in For the amplitude limiting function (if Then take ,like Then take Otherwise take ), The total active power of the battery pack is set.

[0051] 2.2 Reactive Power PID Controller: Input is reactive power deviation ,in: To avoid measurement noise being amplified by the differential term and causing command jitter, during calculation... Before that, first of all Perform a first-order low-pass filter (cutoff frequency 20Hz), and the filtered error signal is denoted as... Differential terms in actual use The filter can be temporarily bypassed for 100ms after a state transition to ensure a fast response; the filter is always active during steady-state operation. The control law is: The PID parameters are determined by looking up a table based on the FSM status. Example values ​​are shown in Table 2: Table 2 Output limiting: ,in .

[0052] 2.3 and The calculations are completed independently, without affecting each other. Active power commands are determined solely by voltage deviation and power imbalance, while reactive power commands are determined solely by reactive power deviation and power factor targets; these two are decoupled at the decision-making level. When battery capacity limitations lead to... When it cannot be fully executed, Unaffected.

[0053] Step 3: SOH Weighted Power Allocation and Reactive Power Sharing 3.1 Active Power Allocation: Read the active power allocation of all online battery modules. and .right A first-order low-pass filter (cutoff frequency 10Hz) is applied to suppress measurement noise. Weights are calculated. Initial sub-instructions for each module Health constraints: for The module's power limit Module rated power; otherwise Module rated power. If... If the record exceeds the limit, Forced to be set And calculate the total excess. (Only for modules that exceed the limit). Reassign to all unrestricted modules (according to their weight within the unrestricted modules), with a maximum of 3 iterations. The module triggers local and remote alarms, prompting for replacement. 3.2 Reactive power sharing: Reactive power command The reactive power is shared by the capacitor and the battery converter. The capacitor takes priority: calculating the reactive power target. The capacitors are switched in groups (e.g., 1kVar per group), and the number of groups is selected according to the "nearest neighbor" principle to ensure the actual output... and The error is the smallest. direction and Same. Considering the operating delay of the capacitor switching switch (typically one control cycle), the actual output in this cycle is... The value remains the same as the command from the previous cycle; therefore, the reactive power command assigned to the battery converter in this cycle is corrected as follows: ,in This is the actual output of the capacitor from the previous cycle. Simultaneously, the newly calculated target group number for this cycle is stored in the execution queue, to take effect in the next cycle. If the capacitor state from the previous cycle is unknown, the initial cycle is set to... Residual error Powered by the battery converter: The reactive power output of the battery converter is constrained by the converter's remaining capacity, while the reactive power output of the capacitor is unaffected by the battery's state of charge (SOC), ensuring the continuity of basic reactive power compensation. (Note:) and Using numerical values ​​in calculations, The values ​​are approximately reasonable for engineering applications under the voltage level of the distribution network. 3.3 If the capacitor is already at full capacity ( Furthermore, the battery converter cannot provide residual reactive power due to capacity constraints, resulting in insufficient total reactive power output. If the system records a reactive power deficit alarm, it will prioritize adjusting active power based on the voltage deviation in the next cycle or report to the master station. 3.4 The final values ​​of each module The data is transmitted to the BMS via the CAN bus. Send to the capacitor switching controller. Send it to the converter.

[0054] Step 4: Prioritize reactive power protection under capacity constraints 4.1 Each module's BMS shall feed back the actual output power at a frequency of not less than 100Hz. 4.2 Calculate the sum of the available power limits of all online modules. 4.3 If (Insufficient active power capacity), execute capacity constraints. 4.4 If The original instructions remain unchanged. 4.5 Key Principle: Capacity constraint results are not fed back to the FSM or PID controller. Operating condition identification and decision-making for the next control cycle are still based on real-time grid measurements. , , , This ensures that the decision-making level always bases its decisions on the actual power grid condition, avoiding circular dependencies. The above four steps constitute a complete control loop from operating condition identification, decoupling decision-making, constraint decomposition to Q-priority guarantee, which is repeated in each control cycle.

[0055] The technical solution of the present invention will be described in detail below with reference to specific embodiments. This embodiment takes a combined active and reactive power compensation device connected to a rural 10kV / 0.4kV distribution substation as an example. The device includes four standardized sodium-ion battery modules (each module has a rated power of 2kW and a rated capacity of 4kWh), one bidirectional AC / DC converter (rated power of 4kW and rated capacity of 4kVA), a group of switching capacitors (rated capacity of 6kVar, switched in 6 groups, each group 1kVar), a data acquisition unit, and a core control unit (ARM architecture industrial controller). Each battery module communicates with the core control unit via a CAN bus, the data acquisition unit and the converter communicate with the core control unit via Ethernet, and the capacitor switching controller is connected to the core control unit via GPIO. The control cycle is set to 50ms.

[0056] I. Corresponding Step 1: Data Acquisition and Operating Condition Identification 1.1 Power Grid Data Acquisition: The data acquisition unit acquires the three-phase voltage and current of the low-voltage busbar in the distribution network in real time at a sampling frequency of 10kHz through voltage / current transformers and analog-to-digital converter chips. The preprocessing module performs Kalman filtering for noise reduction and scaling transformation on the raw data to calculate the effective value of the busbar voltage. Active power of transformer area load and reactive power Photovoltaic power output The processed data is uploaded to the core control unit via Ethernet. 1.2 Battery Status Acquisition: Each sodium-ion battery module has a built-in BMS that independently acquires the module's SOC (ampere-hour integration method plus open-circuit voltage calibration, error ≤ ±1.5%), SOH (current maximum usable capacity / rated capacity, calibrated every 100 full charge-discharge cycles, online accuracy ±5%), and internal resistance. (AC impedance method, 1kHz / 50mA excitation signal, 0.5mΩ resolution). Data is uploaded to the core control unit via CAN bus. 1.3 Capacitor Status Acquisition: The capacitor switching controller feeds back the current number of connected groups and the status of each capacitor, and uploads it to the core control unit via GPIO.

[0057] 1.4 Example of Operating Condition Identification: Let's assume... Actual installed capacity of photovoltaic power in the distribution area Then the power margin (like Unknown (Field configuration). Voltage threshold: Overvoltage undervoltage Emergency overpressure Emergency undervoltage Example A: A certain control cycle , , . (Overcapacity in photovoltaics) (Overvoltage). Identified as S11 (Photovoltaic overcapacity - overvoltage). Example B: A certain control cycle , , . (Photovoltaic balance) exist Between (voltage normal). Identified as S22 (photovoltaic balance - voltage normal). Example C: A certain control cycle , , . (Insufficient photovoltaic power) (Undervoltage). Identified as S33 (Insufficient photovoltaic power - Undervoltage).

[0058] 1.5 Example of SOC constraint on a state machine: In instance A (S11), if the average SOC of all modules is 96% (higher than the upper limit of 95%), then the system will issue an active power instruction. Forced zeroing, reactive power instruction The parameters corresponding to FSM state S11 remain unchanged (capacitors absorb excess reactive power), while a power reduction command is sent to the photovoltaic inverter. The FSM state is still identified as S11, and no degradation occurs.

[0059] 1.6 Debouncing and Boundary Attribution Example – Corresponding to Step 2 of Section 1.4 and Section 1.6: (a) In Example A The current FSM state is S22 (assuming a preceding steady state). Period 1 (Candidate S11 appears for the first time), maintain S22; second cycle Still S11 Execute transition S22→S11. (b) Boundary transition example: If the next cycle of instance A... It dropped back to 230V. Change to S12. The current FSM state is S11. And compared with the previous cycle different, The transfer will not be completed for the time being.

[0060] II. Corresponding Step 2: Active and Reactive Power Decoupling Decision 2.1 Taking instance A (S11 state, , , , For example, let the total rated active power of the battery be... The sum of the reactive power capacity of the capacitor and the converter Active PID: Limit to , (Active power absorption). The feedforward term contributed 10kW here, but was ultimately limited, resulting in an actual command of 4kW. In S12 state (preventative absorption), the parameters... The feedforward term still causes the battery to activate earlier. Reactive PID: After the limit (Reduce excess productivity).

[0061] III. Corresponding Step 3: SOH Weighted Power Allocation and Reactive Power Sharing 3.1 Taking Example A as an example For example, let the parameters of the four modules be the contents of Table 3: Table 3 .right After performing a first-order low-pass filter (cutoff frequency 10Hz), calculate the weights. Module 3 upper limit , No exceptions; the same applies to module 4. No reallocation is required. If the SOH of module 4 drops to 55%, an alarm will be triggered.

[0062] IV. Corresponding Step 4: Reactive Power Priority Guarantee under Capacity Constraints 4.1 Taking Example C as an example For example, suppose module 1 goes offline due to a fault, while modules 2, 3, and 4 are functioning normally. Maximum capacity for each module: Modules 2, 3, and 4... The upper limit is , . If the capacity constraint is not triggered, the original instruction will be maintained.

[0063] This embodiment achieves the following effects: 100% deterministic identification of FSM operating conditions with no boundary oscillations; PQ decoupling ensures that reactive power is not reduced when active power is limited, and the voltage qualification rate is ≥98%; SOH / R weighted allocation enables healthy modules to bear 48.7% of the power and aged modules to bear 11.0%, extending battery cycle life by about 30%; capacity constraints do not pollute decision inputs, and control logic is consistent; all four steps of calculation are completed within 50ms of the control cycle, and edge autonomy is achieved.

[0064] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention. The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0065] Figure 2 A schematic diagram of a distribution network voltage control device based on operating conditions provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 2 As shown, the distribution network voltage control device 2 based on operating conditions includes: The state identification module 21 is used to identify candidate operating conditions of the distribution network in the current control cycle based on the photovoltaic output state and voltage state of the distribution network in the current control cycle; wherein each operating condition corresponds to a control target; the debouncing confirmation module 22 is used to debouncing and confirming the candidate operating conditions of the distribution network in the current control cycle based on the candidate operating conditions of the distribution network in historical control cycles, so as to obtain the final operating condition of the distribution network in the current control cycle; the instruction calculation module 23 is used to calculate the active power instruction and reactive power instruction of the distribution network in the current control cycle based on the PID parameter set corresponding to the final operating condition; the instruction allocation module 24 is used to allocate the active power instruction and reactive power instruction of the distribution network in the current control cycle based on the battery state of charge and the reactive power demand of the battery converter.

[0066] In one possible implementation, the state recognition module 21 is specifically used for: If the photovoltaic output is excessive and the bus voltage is overvoltage, the candidate operating condition is identified as the first operating condition, where the control objective is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is excessive and the bus voltage is normal, the candidate operating condition is identified as the second operating condition, where the control objective is for the battery to gradually absorb active power and for the capacitor to maintain the power factor. If the photovoltaic output is excessive and the bus voltage is undervoltage, the candidate operating condition is identified as the third operating condition, where the control objective is for the battery to release active power to raise voltage and for the capacitor to generate more reactive power. If the photovoltaic output is balanced and the bus voltage is overvoltage, the candidate operating condition is identified as the fourth operating condition, where the control objective is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is balanced and the bus voltage is normal, the candidate operating condition is identified as the fifth operating condition, where the control objective is for the battery to be in standby or fine-tuned and for the capacitor to maintain the power factor. If the photovoltaic output is balanced and the bus voltage is undervoltage, the candidate operating condition is identified as the sixth operating condition. The control objective of this condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the photovoltaic output is insufficient and the bus voltage is overvoltage, the candidate operating condition is identified as the seventh operating condition. The control objective of this condition is for the battery to absorb active power to lower the voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is insufficient and the bus voltage is normal, the candidate operating condition is identified as the eighth operating condition. The control objective of this condition is for the battery to gradually release active power and for the capacitor to maintain the power factor. If the photovoltaic output is insufficient and the bus voltage is undervoltage, the candidate operating condition is identified as the ninth operating condition. The control objective of this condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the bus voltage reaches the emergency overvoltage threshold or the emergency undervoltage threshold, the candidate operating condition is identified as an emergency operating condition. The control objective of this condition is for both the battery and the capacitor to respond unconditionally at full power.

[0067] In one possible implementation, the debouncing confirmation module 22 is specifically used for: If the candidate operating condition in the current control cycle is an emergency operating condition, then the candidate operating condition is taken as the final operating condition in the current control cycle; if the candidate operating condition in the previous control cycle is the same as the candidate operating condition in the current control cycle, then the candidate operating condition is taken as the final operating condition in the current control cycle; otherwise, the final operating condition in the previous control cycle is maintained as the final operating condition in the current control cycle.

[0068] In one possible implementation, the instruction calculation module 23 is specifically used for: The voltage deviation is used as the input to the active power PID controller, and the power imbalance is used as the feedforward term. Based on the active power PID parameter set corresponding to the final operating condition, the active power command of the distribution network in the current control cycle is obtained. The reactive power deviation is used as the input to the reactive power PID controller. Based on the reactive power PID parameter set corresponding to the final operating condition, the reactive power command of the distribution network in the current control cycle is obtained.

[0069] In one possible implementation, the instruction allocation module 24 is specifically used for: For each online battery module, the state of charge of the battery module is calculated and divided by its internal resistance to obtain the weight of the battery module. The weights of each battery module are normalized, and the active power command of the distribution network in the current control cycle is allocated to each battery module based on the normalized weights. With the goal of minimizing the error between the actual reactive power output of the capacitor and the reactive power command, the number of capacitor switching groups is selected according to the nearest neighbor principle, the reactive power command of the distribution network in the current control cycle is allocated to the capacitors, and the remaining reactive power error is allocated to the battery converter.

[0070] In one possible implementation, the instruction allocation module 24 is further configured to: Before allocating active and reactive power commands to the distribution network in the current control cycle based on battery state of charge and reactive power demand of battery converters, if the sum of the available active power limits of all online battery modules is less than the active power command of the distribution network in the current control cycle, the following corrections are performed: the active power command is corrected to the sum of the available active power limits, while maintaining its original direction; the reactive power command remains unchanged, and the reactive power borne by the capacitors remains unchanged; if the battery converter needs to bear reactive power and its remaining capacity is insufficient, the reactive power borne by the battery converter is reduced to its remaining capacity; a power change rate limit is applied to the corrected active power command; if the change in the current cycle correction value relative to the actual value of the previous cycle exceeds the maximum ramp rate allowed by the battery module, the corrected active power command is adjusted according to the maximum ramp rate; the integral term of the active power PID controller is frozen.

[0071] In one possible implementation, a recording module is also included, for: Record the final operating condition status, candidate operating condition status, whether an operating condition transition occurs, and the reason for the operating condition transition in each control cycle of the power distribution network.

[0072] This invention identifies candidate operating conditions based on photovoltaic power output and voltage status, enabling the control strategy to adaptively adjust to changes in actual operating conditions, eliminating strategy mismatch issues at fixed partition boundaries under different operating conditions. By debouncing and confirming the current candidate state based on historical candidate states, the operating condition determination possesses time-dimensional state memory capabilities, eliminating strategy jumps near boundaries caused by measurement noise and short-term fluctuations, avoiding frequent switching of compensation devices, reducing equipment wear, and improving system stability. By independently calculating active and reactive power commands based on the PID parameter set corresponding to the final operating condition, active and reactive power regulation can employ matching control parameters under each operating condition, achieving condition-adaptive optimization of regulation quality. By allocating active power commands based on battery state of charge and reactive power commands based on converter reactive power demand, active and reactive power are allocated separately at the execution level according to their independent physical bases, avoiding the problem of reactive power being reduced along with active power when it is limited, ensuring the effectiveness of voltage regulation, thereby achieving stable, continuous, and adaptive regulation performance across the entire operating range of distribution network voltage control.

[0073] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 3 As shown, the electronic device 3 of this embodiment includes a processor 30 and a memory 31. The memory 31 stores a computer program 32. When the processor 30 executes the computer program 32, it implements the steps in the various method embodiments described above. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module / unit in the various device embodiments described above.

[0074] For example, computer program 32 may be divided into one or more modules / units, which are stored in memory 31 and executed by processor 30 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 32 in electronic device 3.

[0075] Electronic device 3 may include, but is not limited to, processor 30 and memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 3 may also include input / output devices, network access devices, buses, etc.

Claims

1. A distribution network voltage control method based on operating conditions, characterized in that, include: Based on the photovoltaic output status and voltage status of the distribution network in the current control cycle, candidate operating conditions of the distribution network in the current control cycle are identified; wherein, each operating condition corresponds to a control target. Based on the candidate operating conditions of the distribution network in the historical control cycle, the candidate operating conditions of the current control cycle are de-jittered and confirmed to obtain the final operating condition of the distribution network in the current control cycle. Based on the PID parameter set corresponding to the final operating condition, the active power command and reactive power command of the distribution network in the current control cycle are calculated. Based on the battery state of charge and the reactive power demand of the battery converter, the active power command and reactive power command of the distribution network in the current control cycle are allocated.

2. The distribution network voltage control method based on operating conditions according to claim 1, characterized in that, The process of identifying candidate operating conditions of the distribution network in the current control cycle based on the photovoltaic output and voltage status of the distribution network includes: If the photovoltaic output is excessive and the bus voltage is overvoltage, the candidate operating condition is identified as the first operating condition. The control objective of this operating condition is that the battery absorbs active power to reduce voltage and the capacitor absorbs excess reactive power. If the photovoltaic output is excessive and the bus voltage is normal, the candidate operating condition is identified as the second operating condition. The control objective of this operating condition is that the battery gradually absorbs active power and the capacitor maintains the power factor. If the photovoltaic output is excessive and the bus voltage is low, the candidate operating condition is identified as the third operating condition. The control objective of this operating condition is for the battery to release active power to raise the voltage and for the capacitor to increase reactive power. If the photovoltaic output is balanced and the bus voltage is overvoltage, the candidate operating condition is identified as the fourth operating condition. The control objective of this operating condition is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is balanced and the bus voltage is normal, the candidate operating condition is identified as the fifth operating condition. The control objective of this operating condition is battery standby or fine-tuning and capacitor maintenance of power factor. If the photovoltaic output is balanced and the bus voltage is low, the candidate operating condition is identified as the sixth operating condition. The control objective of this operating condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the photovoltaic output is insufficient and the bus voltage is overvoltage, the candidate operating condition is identified as the seventh operating condition. The control objective of this operating condition is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is insufficient and the bus voltage is normal, the candidate operating condition is identified as the eighth operating condition. The control objective of this operating condition is to gradually release the active power of the battery and maintain the power factor of the capacitor. If the photovoltaic output is insufficient and the bus voltage is low, the candidate operating condition will be identified as the ninth operating condition. The control objective of this operating condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the bus voltage reaches the emergency overvoltage threshold or emergency undervoltage threshold, the candidate operating condition is identified as an emergency operating condition. The control objective of this operating condition is for both the battery and capacitor to respond unconditionally at full power.

3. The distribution network voltage control method based on operating conditions according to claim 1, characterized in that, The step of debouncing and confirming the candidate operating conditions of the distribution network in the current control period based on the candidate operating conditions of the distribution network in historical control periods, to obtain the final operating condition of the distribution network in the current control period, includes: If the candidate operating condition status of the current control cycle is an emergency operating condition, then the candidate operating condition status shall be taken as the final operating condition status of the current control cycle. If the candidate operating condition state of the distribution network in the previous control cycle is the same as the candidate operating condition state in the current control cycle, then the candidate operating condition state shall be taken as the final operating condition state of the distribution network in the current control cycle. Otherwise, the final operating condition of the previous control cycle is maintained as the final operating condition of the distribution network in the current control cycle.

4. The distribution network voltage control method based on operating conditions according to claim 1, characterized in that, The calculation of the active power command and reactive power command of the distribution network in the current control cycle based on the PID parameter set corresponding to the final operating condition includes: The voltage deviation is used as the input of the active PID controller, and the power imbalance is used as the feedforward term. Based on the active PID parameter set corresponding to the final operating condition, the active power command of the distribution network in the current control cycle is obtained. The reactive power deviation is used as the input of the reactive power PID controller. Based on the reactive power PID parameter set corresponding to the final operating condition, the reactive power command of the distribution network in the current control cycle is obtained.

5. The distribution network voltage control method based on operating conditions according to claim 1, characterized in that, The active power command and reactive power command of the distribution network in the current control cycle are allocated based on the battery state of charge and the reactive power demand of the battery converter. For each online battery module, calculate the state of charge of the battery module and divide it by its internal resistance to obtain the weight of the battery module; The weights of each battery module are normalized, and the active power command of the power distribution network in the current control cycle is allocated to each battery module based on the normalized weights. With the goal of minimizing the error between the actual reactive power output of the capacitor and the reactive power command, the number of capacitor switching groups is selected according to the nearest neighbor principle. The reactive power command of the distribution network in the current control cycle is allocated to the capacitors, and the remaining reactive power error is allocated to the battery converter.

6. The distribution network voltage control method based on operating conditions according to claim 1, characterized in that, Before allocating active and reactive power commands to the distribution network in the current control cycle based on battery state of charge and reactive power demand of the battery converter, the method further includes: If the sum of the available active power limits of all online battery modules is less than the active power command of the distribution network in the current control cycle, then the following correction is performed: The active power command is modified to the sum of the available active power upper limits, while maintaining the original direction; The reactive power command remains unchanged, and the reactive power borne by the capacitor remains unchanged; If the battery converter needs to handle reactive power and its remaining capacity is insufficient, then the reactive power handled by the battery converter will be reduced to its remaining capacity. A power change rate limit is applied to the corrected active power command. If the change in the current cycle correction value relative to the actual value of the previous cycle exceeds the maximum ramp rate allowed by the battery module, the corrected active power command is adjusted according to the maximum ramp rate. Freeze the integral term of the active PID controller.

7. The distribution network voltage control method based on operating conditions according to any one of claims 1 to 6, characterized in that, Also includes: Record the final operating condition status, candidate operating condition status, whether an operating condition transition occurs, and the reason for the operating condition transition in each control cycle of the power distribution network.

8. A distribution network voltage control device based on operating conditions, characterized in that, include: The status recognition module is used to identify candidate operating conditions of the distribution network in the current control cycle based on the photovoltaic output status and voltage status of the distribution network in the current control cycle; wherein, each operating condition corresponds to a control target; The debouncing confirmation module is used to confirm the candidate operating condition status of the distribution network in the current control cycle based on the candidate operating condition status of the distribution network in the historical control cycle, so as to obtain the final operating condition status of the distribution network in the current control cycle. The instruction calculation module is used to calculate the active power instruction and reactive power instruction of the power distribution network in the current control cycle based on the PID parameter set corresponding to the final operating condition. The instruction allocation module is used to allocate active power instructions and reactive power instructions of the distribution network in the current control cycle based on the battery state of charge and the reactive power demand of the battery converter.

9. The distribution network voltage control device based on operating conditions according to claim 8, characterized in that, The state recognition module is specifically used for: If the photovoltaic output is excessive and the bus voltage is overvoltage, the candidate operating condition is identified as the first operating condition. The control objective of this operating condition is that the battery absorbs active power to reduce voltage and the capacitor absorbs excess reactive power. If the photovoltaic output is excessive and the bus voltage is normal, the candidate operating condition is identified as the second operating condition. The control objective of this operating condition is that the battery gradually absorbs active power and the capacitor maintains the power factor. If the photovoltaic output is excessive and the bus voltage is low, the candidate operating condition is identified as the third operating condition. The control objective of this operating condition is for the battery to release active power to raise the voltage and for the capacitor to increase reactive power. If the photovoltaic output is balanced and the bus voltage is overvoltage, the candidate operating condition is identified as the fourth operating condition. The control objective of this operating condition is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is balanced and the bus voltage is normal, the candidate operating condition is identified as the fifth operating condition. The control objective of this operating condition is battery standby or fine-tuning and capacitor maintenance of power factor. If the photovoltaic output is balanced and the bus voltage is low, the candidate operating condition is identified as the sixth operating condition. The control objective of this operating condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the photovoltaic output is insufficient and the bus voltage is overvoltage, the candidate operating condition is identified as the seventh operating condition. The control objective of this operating condition is for the battery to absorb active power to reduce voltage and for the capacitor to absorb excess reactive power. If the photovoltaic output is insufficient and the bus voltage is normal, the candidate operating condition is identified as the eighth operating condition. The control objective of this operating condition is to gradually release the active power of the battery and maintain the power factor of the capacitor. If the photovoltaic output is insufficient and the bus voltage is low, the candidate operating condition will be identified as the ninth operating condition. The control objective of this operating condition is for the battery to release active power to raise the voltage and for the capacitor to generate more reactive power. If the bus voltage reaches the emergency overvoltage threshold or emergency undervoltage threshold, the candidate operating condition is identified as an emergency operating condition. The control objective of this operating condition is for both the battery and capacitor to respond unconditionally at full power.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 7.

Images