A switching control system and method for optical storage AC-DC hybrid microgrid

Abstract

This invention relates to the field of intelligent control technology for new energy microgrids, and discloses a switching control system and method for a photovoltaic-storage AC / DC hybrid microgrid. The method constructs an initial network model by collecting power system operating parameters and analyzing the topology and weak points of the ultra-high voltage AC transmission network. Real-time alarm thresholds of the large power grid security defense system are embedded into the model to form a dynamic monitoring framework. Based on this framework, the AC / DC bus status is monitored, and the stability margin is assessed by comparing it with grid connection and disconnection reference standards, triggering power coordination and redistribution mechanisms. The output priority of distributed power sources is calculated, and an instruction set is generated after adjusting the energy storage state of charge. Finally, the grid connection interface and power conversion device are controlled to achieve smooth switching of operating modes. This method achieves dynamic coordination between microgrid control and main grid structural risks and security defense, improving the foresight of microgrid operating mode switching and the overall grid's anti-disturbance capability.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for new energy microgrids, specifically to a switching control system and method for a photovoltaic-storage AC / DC hybrid microgrid. Background Technology

[0002] Existing control methods for switching between photovoltaic and energy storage hybrid AC / DC microgrids primarily rely on monitoring local electrical quantities, such as grid connection point voltage, frequency, or preset thresholds. These methods aim to achieve internal microgrid stability and mode switching, and their decision-making logic and control boundaries are limited to the microgrid's own system. This localized control mode fails to effectively link the microgrid's operation with the macroscopic state and security requirements of the main power grid.

[0003] The drawback of existing technical solutions lies in the fact that, as distributed units connected to the main power grid, the operation of microgrids is deeply coupled with that of the main grid, but the control strategies lack awareness and response to the main grid's operational status. Conventional methods cannot obtain information on the main grid's topology and weak points, nor can they respond in real time to changes in the state of the grid-level security defense system. This results in microgrid operation mode switching often being isolated and passive responses, unable to proactively avoid cascading problems caused by structural risks in the main grid, and unable to proactively adjust to provide support when the grid faces pressure.

[0004] The problem this invention aims to solve is how to enable microgrid operation mode switching control strategies to overcome the limitations of local information and deeply integrate into the topology analysis and real-time security defense system of the large power grid, thereby achieving a shift from passive local response to proactive wide-area collaborative defense. This requires the control method to incorporate main grid structure risk information in the early stages of decision-making and dynamically integrate real-time thresholds of the power grid security system during stability assessment. Summary of the Invention

[0005] The purpose of this invention is to provide a switching control system and method based on a photovoltaic-storage AC / DC hybrid microgrid to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a switching control method for a photovoltaic-storage AC / DC hybrid microgrid, the method comprising: Collect operating parameters of the power supply system, analyze the power distribution and potential weak links of the current power grid based on the topology data of the 750 kV and above AC transmission network, and generate an initial network structure model. By embedding real-time alarm thresholds of a large-scale power grid security and defense system and an intelligent dispatch system into the initial network structure model, a dynamic monitoring framework is formed. The voltage and phase information of the AC / DC bus connection point is obtained through a dynamic monitoring framework, compared with the reference standard of the off-grid status, and the stability margin of the current operation mode of the microgrid is determined. Based on the calculation results of the stability margin, the power coordination and redistribution mechanism of the photovoltaic-storage system is triggered; Calculate the output priority sequence of each distributed power source under the power coordination and redistribution mechanism; Based on the state of charge of the DC bus energy storage unit, the output priority sequence is modified to generate a power allocation instruction set; The power allocation instruction set is converted into control instructions for specific power units, which control the grid connection interface device of the photovoltaic-storage system, synchronously adjust the working status of the bidirectional power conversion device, and complete the switching of the AC / DC hybrid microgrid operation mode.

[0007] Preferably, the acquisition of power supply system operating parameters, based on the topology data of AC transmission networks above 750 kV, analyzes the current power distribution and potential weak links of the power grid to generate an initial network structure model, specifically including: Continuously acquire voltage level, power flow direction, line load rate and relay protection setting information from the power grid dispatching master station and local monitoring terminal, and integrate the voltage level, power flow direction, line load rate and relay protection setting information into a multi-dimensional set of operating parameters; Import the predefined electrical connection relationships and equipment parameters of the 750 kV and above AC transmission backbone network to construct a basic topology knowledge graph; The multi-dimensional set of operating parameters is mapped to the corresponding nodes and branches of the basic topological knowledge graph, and the operating status markers of the relevant branches are activated. Based on the operational status marking, nodes and lines with load rates exceeding the warning line, voltage exceeding the limit, or abnormal operation frequency of protection devices are identified in the basic topology knowledge graph and marked as potential weak links. The potential weak links are marked in the basic topology knowledge graph, and combined with real-time power flow data, an initial network structure model with the weights of the weak links and the real-time power flow direction is generated.

[0008] Preferably, the step of embedding real-time alarm thresholds of a large-scale power grid security and defense system and an intelligent dispatching system into the initial network structure model to form a dynamic monitoring framework specifically includes: From the configuration database of the large-scale power grid security and defense system and intelligent dispatch system, call the real-time operation alarm threshold set for different voltage levels and different equipment types. The alarm threshold set includes voltage upper limit, voltage lower limit, frequency upper limit, frequency lower limit, and power oscillation amplitude threshold. The alarm threshold set is associated and bound with each electrical node and line in the initial network structure model, and a one-to-one threshold comparison rule is established for each monitoring point; In the initial network structure model, a corresponding data acquisition channel and logical judgment unit are configured for each threshold comparison rule. The logical judgment unit receives the running data of the corresponding monitoring point in real time. When the operational data of the monitoring point triggers any alarm threshold, the logic judgment unit outputs an out-of-limit event identifier with a timestamp and severity level; By aggregating the out-of-limit event identifiers output by all logical judgment units and combining them with the topological relationship of the initial network structure model, the scope of event impact is assessed, forming a dynamic monitoring framework with real-time state awareness, threshold comparison, and event evaluation capabilities.

[0009] Preferably, the step of acquiring voltage and phase information at the AC / DC bus connection point through a dynamic monitoring framework, comparing it with the reference standard for the off-grid state, and determining the stability margin of the current microgrid operating mode specifically includes: The amplitude and phase angle of the AC bus voltage at the grid connection point, as well as the amplitude of the DC bus voltage, are simultaneously collected through the preset data acquisition channels in the dynamic monitoring framework. Call the grid-connected operation status reference standard database to obtain the current allowable range of grid-connected voltage amplitude, allowable range of phase synchronization deviation, and allowable range of DC voltage. Call the off-grid operation status reference standard database to obtain the current allowable range of off-grid voltage amplitude, allowable frequency fluctuation range, and allowable DC voltage range; The collected AC bus voltage amplitude and phase angle at the grid connection point are compared item by item with the grid connection operation status reference standard, and the deviation values ​​of each item are calculated. The collected DC bus voltage amplitude is compared with the allowable range of DC voltage under both grid-connected and off-grid conditions to calculate the voltage deviation. Based on the magnitude and rate of change of each deviation value, a comprehensive stability margin quantification index is calculated using a preset fuzzy evaluation rule. The stability margin quantification index is used to characterize the degree of distance between the current operating mode and the boundary state.

[0010] Preferably, the step of triggering the power coordination and reallocation mechanism of the photovoltaic-storage system based on the stability margin calculation results specifically includes: The stability margin quantification is received in real time, and a start threshold and an emergency threshold are set. When the stability margin quantification index is lower than the activation threshold but higher than the emergency threshold, it is determined to be in preventive adjustment mode, triggering the preventive power coordination and redistribution process; When the stability margin quantification index is lower than the emergency threshold, it is determined to be in emergency control mode, triggering the emergency power coordination and redistribution process. In the preventive power coordination and redistribution process, with the goal of maintaining the current operating mode, the output power correction of the maximum power point tracking stage of the photovoltaic array in the photovoltaic-storage system is initiated, and the pre-adjustment of the four-quadrant operating capability of the energy storage converter is initiated. In the emergency power coordination and redistribution process, with the goal of quickly switching to a safe operating mode, preparations are made for the graded disconnection of critical loads, and the power flow direction setting of the energy storage converter is forcibly changed.

[0011] Preferably, the output priority sequence of each distributed power source under the computational power coordination and redistribution mechanism specifically includes: The system obtains the actual output of the photovoltaic array in the photovoltaic-storage system at the current moment, the real-time state of charge and adjustable power range of the energy storage unit, and the upper and lower limits of the output of the controllable micro-power source. Obtain the current net load demand of the microgrid and the stability margin quantification from the dynamic monitoring framework; In the preventive adjustment mode, with the main objectives of smoothing net load fluctuations and delaying the decline in stability margin, an objective function is established, and the optimal planned output of each distributed power source in the next scheduling cycle is obtained by solving it. In emergency control mode, the main objectives are to make up for the power deficit or absorb the excess power and restore the stability margin as quickly as possible. An objective function is established and the emergency adjustment output of each distributed power source is obtained by solving it. The optimal planned output or emergency adjustment output obtained by the solution is compared with the current actual output of each distributed power source to calculate the amount and direction of power adjustment required for each power source. Based on the magnitude of the power adjustment, the speed of the adjustment, and the level of equipment operating costs, a multi-attribute decision-making method is used to calculate the output priority sequence of each distributed power source in this adjustment.

[0012] Preferably, the step of modifying the output priority sequence and generating a power allocation instruction set by combining the state of charge of the DC bus energy storage unit specifically includes: Real-time monitoring of the current state of charge of all energy storage units on the DC bus, and acquisition of their rated capacity and maximum charging and discharging power limits; Determine the overall power balance status of the microgrid. If it is a power deficit state, check whether the current state of charge of each energy storage unit is higher than the discharge protection lower limit. If it is a power surplus state, check whether the current state of charge of each energy storage unit is lower than the charging protection upper limit. Based on the power output priority sequence, the charging and discharging feasibility of each energy storage unit is checked sequentially. For energy storage units that are ranked high in the output priority sequence but whose state of charge does not meet the feasibility of charging and discharging, their output command in this adjustment will be forcibly set to zero, and they will be temporarily removed from the list of schedulable resources in this adjustment. Based on the updated list of schedulable resources, power adjustment tasks are reallocated to ensure that the total adjustment amount matches the demand, forming the final power allocation instruction set that includes specific power values, direction of change, and execution sequence.

[0013] Preferably, the step of converting the power allocation instruction set into control instructions for specific power units to control the grid connection interface device of the photovoltaic-storage system specifically includes: The power allocation instruction set is analyzed to include a portion of the instructions for the grid-connected interface device. The power allocation instruction includes a target power value, a power change rate limit, and an execution time window. Convert the target power value into reference values ​​for active and reactive current of the converter in the grid-connected interface device; Based on the power change rate limit, the change trajectory of active and reactive current reference values ​​is planned to generate a smooth current reference trajectory. Based on the current reference trajectory, a model predictive control method is used to calculate the optimal switching state combination of the converter switching devices in the grid-connected interface device at the next moment in each control cycle. The optimal switching state combination is converted into a drive pulse and sent to the power switching device drive circuit of the grid-connected interface device. The voltage and current at the grid connection point are monitored synchronously, the actual current value is compared with the current reference trajectory, and the internal model parameters of the model predictive control are corrected through closed-loop feedback.

[0014] Preferably, the synchronous adjustment of the bidirectional power conversion device's operating state to complete the switching of the AC / DC hybrid microgrid's operating mode specifically includes: The power distribution instruction set is analyzed for some instructions of the bidirectional power conversion device between the DC bus and the AC bus. These instructions include the power transmission direction, power transmission magnitude, and DC voltage regulation target. Based on the requirements of power transmission direction and magnitude, the control mode of the bidirectional power conversion device is set to constant power control or DC voltage control. In constant power control mode, the commanded power value is used as the setpoint for the outer loop power controller, and the controller calculates the inner loop current reference value. In DC voltage control mode, the DC voltage regulation target of the command is used as the given value of the outer loop voltage controller, and the inner loop current reference value is calculated by the controller. Based on the calculated inner loop current reference value, space vector modulation technology is used to generate the switching signal of the bidirectional power converter, and control it to achieve bidirectional energy flow and DC voltage stability. During the transient process of mode switching, the timing of the control command output of the grid-connected interface device and the bidirectional power conversion device is coordinated to ensure the synchronization and smooth transition of power on the AC side and the DC side, so that the microgrid as a whole can operate in the target mode.

[0015] Preferably, the present invention also includes a switching control system based on a photovoltaic-storage AC / DC hybrid microgrid. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the switching control method based on a photovoltaic-storage AC / DC hybrid microgrid as described above.

[0016] Compared with the prior art, the beneficial effects of the present invention are: Based on the analysis of the topology and weak points of the ultra-high voltage transmission network, an initial model of the microgrid is generated, introducing the topology and risk information of the main grid into the foundational layer of microgrid control. This allows the microgrid's operational strategy to be built upon a deep understanding of the structural characteristics and potential risk points of the local grid it connects to. Control decisions not only respond to local events but also proactively consider the carrying capacity and fault propagation paths of specific areas of the main grid. This enables the microgrid to proactively adjust to structural risks in the main grid during grid-connected to off-grid or reverse switching. The microgrid's control behavior shifts from passively responding to local electrical quantity exceedances to actively avoiding or mitigating cascading risks caused by weak points in the main grid, enhancing the microgrid's forward-looking support role as a grid-friendly unit for the stability of the main grid.

[0017] A dynamic monitoring framework embedding real-time alarm thresholds into the large power grid security defense system enables microgrids to synchronize their own operational status assessment criteria with the overall security assessment results from the large power grid dispatch center in real time. The reference standard for microgrid stability margin is no longer static or isolated but dynamically correlated with the real-time security level of the main grid. When the main grid security defense system issues specific risk warnings or adjustment commands, this threshold can be dynamically adjusted, thereby proactively and precisely triggering power coordination and mode switching preparations within the microgrid. This mechanism achieves information closure and command coordination between the microgrid's autonomous control logic and the grid-level centralized security defense system, making the microgrid's operating mode switching a distributed and automated link in executing the large power grid's overall security defense strategy, enhancing the coordination and response speed of the multi-layered defense system of the power grid under complex fault conditions. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating the working principle of the switching control method for a hybrid AC / DC microgrid based on photovoltaic and energy storage described in this invention. Figure 2A flowchart for generating the initial network structure model; Figure 3 A flowchart for determining the stability margin of the current operating mode of a microgrid; Figure 4 A real-time monitoring graph of the stability margin of a photovoltaic-storage AC / DC hybrid microgrid; Figure 5 This is a power balance analysis diagram under multiple states of a photovoltaic-storage microgrid. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1This invention provides a switching control method for a hybrid AC / DC microgrid based on photovoltaic-storage. The method includes: acquiring power supply system operating parameters, including voltage, current, power, and protection status, through a data acquisition system; combining this with a pre-built topology database reflecting the connection relationships of AC transmission networks above 750 kV; analyzing the acquired parameters to identify the current power distribution characteristics and potential weaknesses of the power grid; and forming an initial network structure model containing this information. Various real-time operational alarm thresholds defined in a large-scale power grid intelligent dispatch and security defense system are embedded into the corresponding monitoring nodes of this initial network structure model, thereby constructing a dynamic monitoring framework with real-time status comparison capabilities. This dynamic monitoring framework continuously acquires voltage and phase information at the connection points between the microgrid and the main grid, i.e., the AC / DC bus connection points, and compares this measured information with pre-stored technical standard reference values ​​for both grid-connected and off-grid operation states. The stability margin of the current microgrid operating mode is quantitatively calculated by evaluating the degree of deviation. Based on the calculated stability margin quantification results, the system autonomously decides and triggers a control mechanism for power coordination and redistribution within the photovoltaic-storage system. Under this mechanism, based on the current network status and power characteristics, the priority sequence for output adjustment of each distributed power source, such as photovoltaic arrays, energy storage units, and backup generators, is calculated. Combined with the real-time state of charge of the energy storage units on the DC bus, this output priority sequence is recalibrated to generate a power allocation instruction set containing specific power values ​​and execution timing. The control center converts this power allocation instruction set into direct control instructions for the underlying power electronic equipment. These instructions regulate the grid-connected interface devices of the photovoltaic-energy storage system and simultaneously adjust the operating status of the bidirectional power conversion devices connected to the AC / DC bus. By coordinating the control of these two types of devices, a smooth and rapid switching of the AC / DC hybrid microgrid operation mode is achieved.

[0021] In one embodiment of the present invention, see [reference] Figure 2The process of collecting operating parameters of the power supply system involves data synchronization between the main station of the power grid energy management system and local monitoring terminals deployed at substations and grid connection points. This continuously acquires information on voltage levels, power flow direction, line load rates, and the settings and actions of relay protection devices. This information is integrated into a multi-dimensional set of operating parameters. Simultaneously, a predefined set of electrical connection relationships and nameplate parameters covering all lines, transformers, busbars, and switchgear of the AC transmission backbone network at voltage levels above 750 kV is imported from the power grid planning database. A basic topology knowledge graph is constructed using graph theory. The real-time acquired multi-dimensional set of operating parameters is mapped to the corresponding nodes and branches in this basic topology knowledge graph, and the operating status markers of relevant branches are activated or updated based on parameter values. Based on these operating status markers, an operating status analysis algorithm in the basic topology knowledge graph identifies specific nodes and lines where the load rate consistently exceeds the preset warning line, voltage amplitude exceeds limits, or the associated protection devices have shown abnormal recent operating frequencies. These identified components are marked as potential weak points. Finally, in the visualization and data layer of the basic topology knowledge graph, these potential weak links are highlighted. At the same time, the power flow direction of each branch is calculated by combining the real-time collected power flow data, generating an initial network structure model that includes network topology connections and also adds the weight attributes of weak links and real-time power flow direction data.

[0022] The process of embedding real-time alarm thresholds in a large-scale power grid security and defense system and intelligent dispatch system involves retrieving a set of real-time operating alarm thresholds for different voltage level nodes and different capacity types of lines from the system's remote configuration server or local cache database. This set specifically includes multiple parameters such as upper voltage limit, lower voltage limit, upper frequency limit, lower frequency limit, and power oscillation amplitude threshold. Through a data association interface, each threshold in this alarm threshold set is associated one-to-one with the corresponding electrical node and line object in the initial network structure model, establishing a unique threshold comparison rule for each monitoring point in the model. Within the software logic layer of the initial network structure model, an independent data acquisition channel and logical judgment unit are configured for each established threshold comparison rule. This logical judgment unit receives real-time operating data from monitoring points in the physical sensing layer through its corresponding data acquisition channel. When the operating data value received by a logical judgment unit exceeds any of its bound alarm thresholds, the unit immediately generates and outputs an over-limit event identifier containing the precise time of the event, monitoring point location information, and an over-limit severity level indicator. The system's central processor aggregates the over-limit event identifiers output in real time from all logical judgment units. Combining the topological location relationships of the monitoring points corresponding to these event identifiers in the initial network structure model, it analyzes and evaluates the electrical range and set of devices that may be affected by a single event or related events, thereby forming a dynamic monitoring framework that integrates real-time status perception, automatic threshold comparison, and event impact range assessment.

[0023] In practical implementation, the acquisition of power supply system operating parameters is completed through the dispatch data network and the synchronous phasor measurement unit installed in the hub substation. Voltage level, power flow direction, line load rate, and relay protection setting information are acquired at a rate of 50 frames / second and packaged into a multi-dimensional operating parameter set. In some embodiments, the multi-dimensional operating parameter set is transmitted to a local analysis server through a secure encrypted channel. The server's memory pre-stores electrical connection relationships and equipment parameters imported from the provincial power grid model library. These parameters include line impedance, transformer turns ratio, and rated capacity. Based on this data, a basic topology knowledge graph is constructed using a graph database. Nodes in the basic topology knowledge graph represent buses, and edges represent lines or transformers. The line load rate values ​​in the real-time multi-dimensional operating parameter set are mapped to the "current load" attribute field of the corresponding edge in the basic topology knowledge graph, and the voltage measurement values ​​are mapped to the "voltage amplitude" attribute field of the corresponding node. This process activates the operating status flags of the relevant branches, which include "normal," "overload," "overlimit," and "protection action."

[0024] In practical implementation, when identifying weak links based on operational status markers, the system traverses the attributes of all edges and nodes in the basic topology knowledge graph. The identification logic determines that edges with a load rate attribute exceeding 85% for 30 consecutive seconds are warning edges; nodes with voltage amplitude attributes exceeding 1.05 times or falling below 1000 kV by 0.95 times are over-limit nodes; and devices with a related protection signal action count greater than 2 in the past 10 minutes are abnormal devices. These identified edges, nodes, and devices are uniformly marked as potential weak links. In the visualization interface and topology data file of the basic topology knowledge graph, elements marked as potential weak links are highlighted in red. Simultaneously, real-time active power flow values ​​for the corresponding branches are extracted from a multi-dimensional set of operational parameters and assigned to the corresponding edges in the basic topology knowledge graph, generating an initial network structure model with weak link weight attributes and real-time power flow vectors.

[0025] Optionally, for monitoring line power oscillations, the logic judgment unit receives a 10-cycle sequence of active power from the line, calculates the oscillation mode of the sequence, and outputs an over-limit event identifier containing the oscillation frequency and attenuation coefficient if the mode damping ratio is lower than 0.03. The central event processor aggregates all over-limit event identifiers output by the logic judgment units, analyzes the set of all downstream substation nodes that may be affected by a voltage over-limit event identifier of a bus, and the set of parallel-operating lines that may be affected by a power oscillation over-limit event identifier of a line, based on the electrical connection relationships defined in the initial network structure model. The impact range of each event is evaluated through a topology traversal algorithm, ultimately forming a dynamic monitoring framework that can perceive voltage, frequency, and power oscillation status in real time and automatically perform threshold comparison and event impact range analysis.

[0026] In practical implementation, the quantization calculation of the running status marker can be achieved through an evaluation function, which is defined as:

[0027] in: The quantitative score representing the running status marker, The load factor is the weighting factor. This represents the real-time load rate of the line. This is the voltage deviation weighting coefficient. This represents the percentage deviation of the node's voltage amplitude from its rated value. When When the value falls below a preset threshold, the corresponding element in the basic topological knowledge graph is activated as a potential weak link. In some embodiments, the weight coefficient... and Determined based on equipment type and historical operating data.

[0028] In one embodiment of the present invention, see [reference] Figure 3The process involves acquiring and comparing information through a dynamic monitoring framework. Utilizing a pre-set high-speed data acquisition channel within the framework, the amplitude and instantaneous phase angle of the AC bus voltage at the grid connection point, as well as the amplitude signal of the DC bus voltage, are synchronously acquired at a fixed sampling frequency. The system calls the grid-connected operation status reference standard database stored locally. Based on the current microgrid's access capacity and protocol, it obtains the allowable range of the AC bus voltage amplitude at the grid connection point, the allowable synchronous deviation range between its voltage phase and the main grid voltage phase, and the allowable range of the DC bus voltage in grid-connected mode. Simultaneously, the system calls the off-grid operation status reference standard database to obtain the allowable range of AC voltage amplitude, the allowable fluctuation range of the system frequency, and the allowable range of the DC bus voltage in off-grid mode that the microgrid should maintain during independent operation. The real-time acquired AC bus voltage amplitude and phase angle data at the grid connection point are compared item by item with the corresponding allowable ranges retrieved from the grid-connected operation status reference standard database to calculate the percentage of voltage amplitude deviation and the absolute value of phase angle deviation. The real-time collected DC bus voltage amplitude is compared with the allowable DC voltage range specified in the reference standards for both grid-connected and off-grid operation states, and the deviation from the boundary values ​​of each range is calculated. Based on the magnitude of the calculated deviation values ​​and the rate of change of these deviation values ​​within the recent time window, a comprehensive stability margin quantification index ranging from 0 to 1 is calculated using a pre-set evaluation algorithm employing fuzzy logic rules. The value of this stability margin quantification index directly characterizes how close the current microgrid operation mode is to its stable boundary state.

[0029] In its implementation, the microgrid connects to the 10 kV distribution network through a common connection point. The data acquisition channel in the dynamic monitoring framework, with a sampling frequency of 10 kHz, synchronously acquires the amplitude, phase angle, and DC bus voltage amplitude at the connection point. The AC bus voltage amplitude is measured by a voltage transformer and filtered; the phase angle is extracted in real-time from the grid voltage waveform using a phase-locked loop circuit; and the DC bus voltage amplitude is directly acquired by a high-precision voltage sensor. These acquired raw data are converted from analog to digital to form data packets containing timestamps of voltage amplitude, phase angle, and DC voltage, which are then used for subsequent comparison and judgment.

[0030] In some embodiments, the comparison calculation process is specifically carried out as follows: The system compares the instantaneous amplitude of the 10 kV AC bus voltage (10.2 kV) with the upper limit of 10.5 kV and the lower limit of 9.5 kV in the grid-connected operation status reference standard, and calculates that the voltage amplitude deviation is a positive deviation of 0.2 kV, expressed as +2%. The difference between the phase angle of the collected AC bus voltage and the main grid reference phase angle (-0.3 degrees) is compared with the allowable phase synchronization deviation range of -0.5 degrees to +0.5 degrees, and the absolute value of the phase angle deviation is calculated to be 0.3 degrees. The collected DC bus voltage amplitude (800 V) is compared with the allowable range of grid-connected DC voltage (675 V to 825 V) and the allowable range of off-grid DC voltage (675 V to 825 V), respectively, and calculates that its deviation from the upper limit of the range is -25 V and its deviation from the lower limit of the range is +125 V. In another operating scenario, if the measured AC bus voltage amplitude is 9.3 kV, the calculated voltage amplitude deviation is a negative deviation of -0.2 kV, which is -2%.

[0031] Optionally, the stability margin quantification index is calculated using a pre-defined fuzzy evaluation rule set. This rule set defines the mapping relationship between various deviation values ​​and their rates of change, and the stability margin. For example, one rule defines that if the voltage amplitude deviation percentage is within ±1% and the rate of change is less than 0.1% / second, a high stability margin factor is contributed; if the voltage amplitude deviation percentage is between ±2% and 3% or the rate of change is greater than 0.5% / second, a low stability margin factor is contributed. Phase angle deviation and DC voltage deviation are also factored according to similar rules. The comprehensive quantification calculation is achieved through the following formula:

[0032] in: This represents the calculated stability margin quantification index, with a value range of [0,1]. The factor representing the voltage amplitude deviation obtained after fuzzy evaluation. The factor representing the phase angle deviation obtained after fuzzy evaluation. The factor representing the DC voltage deviation obtained after fuzzy evaluation; , , These are the weighting coefficients assigned to the voltage amplitude, phase angle, and DC voltage, respectively. , , The phase synchronization weight has different values ​​depending on whether the microgrid is currently in grid-connected or off-grid mode. In grid-connected mode, the phase synchronization weight... The value is relatively high; in off-grid mode, the equivalent weight corresponding to frequency stability will be adjusted. The calculated value is... The closer the value is to 1, the further the current operating mode is from the boundary state of voltage, phase, or DC voltage, and the higher the stability. The closer the value is to 0, the closer the current operating state is to or has reached the stability boundary.

[0033] In practical implementation, data comparison reflects the differences in calculation results under different operating conditions. When the microgrid is operating stably and connected to the grid, assuming the measured voltage amplitude deviation is +0.5%, the phase angle deviation is +0.2 degrees, and the DC voltage deviation is +1%, the rates of change for each parameter are very low. Through the above fuzzy evaluation rules and formulas, a quantitative index of stability margin can be obtained. The value is 0.92. When the microgrid is subjected to external disturbances, assuming the voltage amplitude fluctuates to +2.5%, the phase angle deviation increases to -0.7 degrees, and the DC voltage changes to +3% at a relatively rapid rate, a stability margin quantification index can be obtained through the same evaluation process. Decreased to 0.65. Stability margin quantification. The continuous calculation and output provide a direct triggering basis for the subsequent power coordination mechanism.

[0034] In one embodiment of the present invention, taking a hybrid AC / DC microgrid comprising a 2 MW photovoltaic array, a 1 MW / 2 MWh energy storage unit, and a 0.5 MW standby gas turbine as an example scenario, the control system receives stability margin quantification indicators calculated by a dynamic monitoring framework in real time. The values ​​of the stability margin quantification indicators are continuously monitored. The control system has a preset start threshold of 0.8 and an emergency threshold of 0.6. When the monitored stability margin quantification indicator value is 0.75, which is lower than the start threshold of 0.8 but higher than the emergency threshold of 0.6, the system determines that it is entering a preventative adjustment mode and immediately triggers a pre-programmed preventative power coordination and redistribution process. At another operating moment, when the monitored stability margin quantification indicator value drops sharply to 0.55 due to a grid fault, which is lower than the emergency threshold of 0.6, the system determines that it is entering an emergency control mode and immediately triggers another independent emergency power coordination and redistribution process.

[0035] In some embodiments, the execution of the preventive power coordination and redistribution process is specifically manifested as follows: The control system, aiming to maintain the current grid-connected operation mode of the microgrid, sends a correction command to the photovoltaic inverter, actively lowering the output reference value of the photovoltaic array's maximum power point tracking (MPPT) stage from the current 1.8 MW to 1.7 MW. Simultaneously, the control system sends a pre-adjustment command to the energy storage converter, activating its four-quadrant operation capability and adjusting its reactive power output setpoint from 0 kVAR to -50 kVAR to absorb excess reactive power. The execution of the emergency power coordination and redistribution process is specifically manifested as follows: The control system, aiming to quickly switch to a safe off-grid operation mode, initiates preparation for tiered load shedding commands, marking the 200 kW load with priority level three in the load list as a pending shedding sequence. Simultaneously, the control system forcibly changes the power flow setting value of the energy storage converter, forcing it to discharge to the AC bus at a maximum power of 1 MW from its current floating charge state.

[0036] It is understandable that the initial step in calculating the output priority sequence of each distributed power source is to obtain real-time data. From the local monitoring unit, the actual output of the photovoltaic array at the current moment is obtained as 1.8 MW; the real-time state of charge of the energy storage unit is 60%, and its maximum charge / discharge power limit is ±1 MW; the upper limit of the standby gas turbine output is 0.5 MW, and the lower limit is 0 MW. The microgrid net load demand obtained from the dynamic monitoring framework is 2.2 MW, and the real-time updated stability margin quantification index is 0.75. When the system operates in preventative adjustment mode, the control algorithm establishes an objective function with the main optimization objectives of smoothing net load fluctuations and delaying the decline in stability margin. The objective function considers minimizing the sum of squares of the deviations between the net load forecast and the power output within the next 5-minute scheduling cycle. Under constraints including photovoltaic output of 0 to 2 MW, energy storage output of -1 to 1 MW, and gas turbine output of 0 to 0.5 MW, the optimal planned output values ​​for each distributed power source are calculated by a quadratic programming solver: 1.7 MW for photovoltaic, 0.5 MW for energy storage, and 0.5 MW for gas turbine.

[0037] Optionally, when the system operates in emergency control mode and the stability margin quantification index is 0.55, the control algorithm establishes an objective function with the primary goal of quickly compensating for the system's power deficit. The objective function requires maximizing the sum of the absolute values ​​of the output adjustments of all adjustable power sources within the next 10 seconds. Through linear programming, the emergency adjustment output values ​​for each distributed power source are quickly solved: photovoltaic (PV) maintains 1.8 MW, energy storage discharges at a maximum of 1 MW, and gas turbine generates at a maximum of 0.5 MW. The optimal planned output value or emergency adjustment output value obtained from the solution is compared with the current actual output value of each distributed power source. In preventative adjustment mode, it is calculated that PV needs to be reduced by 0.1 MW, energy storage needs to be increased by 0.5 MW in discharge, and gas turbine needs to be increased by 0.5 MW in generation. In emergency control mode, it is calculated that PV needs to be adjusted by 0 MW, energy storage needs to be increased by 1 MW in discharge, and gas turbine needs to be increased by 0.5 MW in generation.

[0038] In practical implementation, the power output priority sequence is calculated using a multi-attribute decision-making method. The evaluated attributes include the absolute value of each power source's power adjustment, the speed of its power adjustment, and the cost of its operating equipment. For the preventative regulation mode, photovoltaic power adjustment is 0.1 MW, with a fast adjustment rate and low operating cost; energy storage power adjustment is 0.5 MW, with a fast adjustment rate and moderate operating cost; and gas turbine power adjustment is 0.5 MW, with a slow adjustment rate and high operating cost. A multi-attribute decision-making algorithm standardizes and weights the three attributes of these three power sources. The score calculation can be achieved using the following formula:

[0039] in: This represents the priority score of power supply i; the lower the score, the higher the priority. Represents the absolute value of the power adjustment of power supply i; This represents the maximum absolute value of the power adjustment across all power supplies, used for normalization. The rating represents the adjustment rate of power supply i; the faster the rate, the higher the rating. This represents the operating cost score of power supply i; the lower the cost, the higher the score. , , These are weighting coefficients assigned to power adjustment amount, adjustment rate, and operating cost, respectively. In preventative adjustment mode, cost and rate are given greater emphasis, while in emergency control mode, adjustment amount and rate are given greater emphasis. Based on the formula, the example might yield a power output priority sequence with energy storage having the highest priority, followed by photovoltaics, and then gas turbines.

[0040] In one embodiment of the present invention, taking a DC bus energy storage system comprising three parallel sub-units as an example scenario, the process of correcting the output priority sequence based on the state of charge (SOC) of the DC bus energy storage units is implemented. The current SOC of energy storage units A, B, and C is monitored in real-time via the communication bus of the battery management system. The monitored values ​​are 25%, 60%, and 90%, respectively. The rated capacity of each of the three energy storage units is 1 MWh, and the maximum charging power limit and maximum discharging power limit are both 0.5 MWh and 0.5 MWh, respectively, obtained from the system parameter database. Based on the microgrid power balance information obtained from the dynamic monitoring framework, the system determines that the current overall power balance of the microgrid is in a power deficit state, with a power deficit value of 0.8 MWh. In this power deficit state, the control system sequentially checks whether the current SOC of energy storage units A, B, and C is higher than their discharge protection lower limit threshold, which is set to 20%.

[0041] In practice, the charging and discharging feasibility of each energy storage unit is checked sequentially according to the output priority sequence, which is: Energy Storage Unit B, Energy Storage Unit A, and Energy Storage Unit C. The checking process is as follows: For Energy Storage Unit B, which ranks first in the output priority sequence, its current state of charge (SOC) is 60%, exceeding the discharge protection lower limit by 20%. Energy Storage Unit B meets the discharge feasibility condition and is retained in the schedulable resource list. For Energy Storage Unit A, which ranks second in the output priority sequence, its current SOC is 25%, exceeding the discharge protection lower limit by 20%. Energy Storage Unit A meets the discharge feasibility condition and is retained in the schedulable resource list. For Energy Storage Unit C, which ranks third in the output priority sequence, its current SOC is 90%, exceeding the discharge protection lower limit by 20%. Energy Storage Unit C meets the discharge feasibility condition and is retained in the schedulable resource list. Refer to Table 1 for the key parameters of each energy storage unit.

[0042] Table 1: Key Parameters of Each Energy Storage Unit

[0043] In another operating scenario, if the current state of charge (SOC) of energy storage unit A is 18%, which is below the discharge protection limit of 20%, energy storage unit A is deemed not to meet the discharge feasibility conditions in this power deficit adjustment. For energy storage unit A, which is ranked high in the output priority sequence but whose SOC does not meet the charging / discharging feasibility requirements, the control system forcibly sets its planned output command in this adjustment to 0 MW and temporarily removes energy storage unit A from the current schedulable resource list. The updated schedulable resource list only includes energy storage units B and C. According to the updated schedulable resource list, a power deficit of 0.8 MW needs to be reallocated. The maximum discharge power of both energy storage units B and C is 0.5 MW. Following the output priority sequence, energy storage unit B takes precedence, resulting in the final power allocation command set: energy storage unit B discharges 0.5 MW, and energy storage unit C discharges 0.3 MW. The command execution sequence is immediate synchronous execution.

[0044] In some embodiments, the process of converting a power allocation instruction set into control instructions for a grid-connected interface device (GMT) takes, for example, a power allocation instruction set requiring the GMT to output 0.8 MW of active power and absorb 0.1 MW of reactive power. The power allocation instruction set is parsed to obtain partial instructions for the GMT, which include a target power value of 0.8 MW active power and -0.1 MW reactive power, a power change rate limit of 1.2 MW per minute, and an execution time window starting at the beginning of the next control cycle. The control system converts the target power value into reference values ​​for the active and reactive currents of the converter in the GMT. Based on the current grid connection voltage of 10.2 kV, the active current reference value is calculated to be approximately 78.4 amperes, and the reactive current reference value is approximately -9.8 amperes. According to the power change rate limit of 1.2 MW per minute (0.02 MW per second), linear programming is performed on the trajectory of the active and reactive current reference values ​​from the current value to the target value, generating a current reference trajectory that smoothly changes from the current value to the target value within 2 seconds.

[0045] Understandably, based on the current reference trajectory, model predictive control (MMC) calculates the switching states in each control cycle (100 microseconds). The MMC algorithm, based on the discrete mathematical model of the converter, predicts the future value of the converter output current under all possible combinations of switching device states in the next control cycle. An evaluation function selects the switching state combination that best approximates the predicted current to the current reference trajectory as the optimal switching state combination for the next moment. The calculated optimal switching state combination, for example, a binary sequence [1,0,0,1,0,1] representing the on / off state of a certain bridge arm's upper and lower switches, is converted into drive pulses with specific pulse widths and phases. This pulse is then transmitted via optical fiber to the insulated-gate bipolar transistor (IGBT) power switch drive circuit of the grid-connected interface device. Sensors simultaneously monitoring the grid-connected point voltage and current sample at a frequency of 50 kHz. The actual current measurement value is compared with the planned current reference trajectory value at the same moment. The current error is calculated by a proportional-integral (PI) regulator, and its output is used to correct the filter inductance parameter values ​​in the prediction model within the MMC algorithm online. The correction formula is:

[0046] in: This represents the adjusted filter inductance value used for model prediction in the k-th control cycle. This represents the nominal inductance value of the filter. This represents a correction gain coefficient. This represents the current reference value for the (k-1)th control cycle. This represents the actual current measurement value during the (k-1)th control cycle. The model parameters are continuously corrected through closed-loop feedback, making the output of the model predictive control more accurate.

[0047] Optionally, in another scenario, if the power distribution command set requires the grid-connected interface device to rapidly increase its output, with the power change rate limited to 1 megawatt per second, the slope of the current reference trajectory will increase significantly. The model predictive control algorithm will calculate a series of switching states to track steeper current changes in a shorter time. The generation frequency and duty cycle of the drive pulse will change accordingly to adapt to the requirements of rapid tracking.

[0048] See Figure 4This is a real-time monitoring chart of the stability margin of a photovoltaic-storage AC / DC hybrid microgrid. It visually presents the dynamic changes in the microgrid's operating status and the logic of mode switching, serving as a core decision-making visualization tool in microgrid switching control. The chart's core function is to visualize the microgrid's stability boundaries in real time, assisting operators or automatic control systems in quickly identifying the current operational risk level; accurately triggering corresponding control strategies; and tracing the trend of stability margin changes to optimize microgrid switching control parameters. This type of chart is a crucial link in the "state perception-decision execution" closed loop of the photovoltaic-storage AC / DC hybrid microgrid, ensuring the microgrid maintains safe and stable operation during grid-connected / off-grid switching.

[0049] In one embodiment of the present invention, the bidirectional power converter is synchronously adjusted to complete the operation mode switching process. The system parses a portion of the power distribution instruction set for the bidirectional power converter connected between the DC bus and the AC bus. This portion of the instruction includes the power transmission direction, the power transmission magnitude, and the target DC voltage regulation value to be maintained in a specific mode. Based on the power transmission direction and magnitude required by the instruction, the core control mode of the bidirectional power converter is set. If the instruction requires the transmission of a specific power, it is set to a constant power control mode; if the instruction requires stabilizing the DC bus voltage, it is set to a DC voltage control mode. In constant power control mode, the power value given by the instruction is used as the input setpoint of the outer loop power controller, which calculates and generates the current reference value required by the inner loop current controller. In DC voltage control mode, the target DC voltage regulation value given by the instruction is used as the input setpoint of the outer loop DC voltage controller, which calculates and generates the current reference value required by the inner loop current controller. Based on the current reference value calculated by the inner-loop controller, space vector modulation technology is used to generate a specific switching signal sequence suitable for the topology of the bidirectional power converter. This sequence controls the operation of switching devices such as insulated-gate bipolar transistors, thereby enabling bidirectional energy flow between the AC and DC buses and maintaining the DC bus voltage at the required level. During the transient process of switching between the entire microgrid operating modes, the central controller coordinates the timing and logic interlocking relationships of the control commands output by the grid-connected interface device and the bidirectional power converter. This ensures that the power exchange between the AC and DC sides can transition synchronously and smoothly to the new equilibrium point, ultimately enabling the AC / DC hybrid microgrid to stably switch from one operating mode to another target operating mode.

[0050] In practical implementation, the microgrid includes a bidirectional power converter connecting a 750V DC bus and a 380V three-phase AC bus. The process of synchronously adjusting the operating status of the bidirectional power converter to complete the mode switching begins with parsing a portion of the power distribution instruction set related to the bidirectional power converter. The parsed instructions explicitly state that the power transmission direction is from the DC bus to the AC bus, the power transmission magnitude is 300 kW, and the DC voltage regulation target is 755V. Based on the requirements of the power transmission direction and magnitude, and considering the current microgrid mode switching target of switching from off-grid to grid-connected, the system sets the control mode of the bidirectional power converter to DC voltage control mode, with stabilizing the DC bus voltage as the primary objective.

[0051] In some embodiments, under DC voltage control mode, the control system uses the commanded DC voltage regulation target of 755 volts as the input setpoint of the outer loop voltage controller. The deviation between the DC voltage regulation target of 755 volts and the measured value of 750 volts from the DC bus voltage sensor is 5 volts. The outer loop voltage controller employs a proportional-integral (PI) regulation algorithm, which calculates an inner loop current reference value based on the 5-volt voltage deviation. The inner loop current reference value represents the active current component that needs to be absorbed from or injected into the AC side to eliminate the voltage deviation. Assuming that at another moment, the power distribution command set requires the bidirectional power converter to transmit 200 kW of power to the DC side in constant power control mode, with the power transmission direction reversed, the control system sets the control mode of the bidirectional power converter to constant power control mode. In constant power control mode, the control system uses the commanded power value of 200 kW as the input setpoint of the outer loop power controller. The outer loop power controller directly calculates the inner loop current reference value required by the inner loop current controller based on the 200 kW power command.

[0052] It is understandable that, based on the calculated inner-loop current reference value, the control system uses space vector modulation technology to generate specific switching signals for the bidirectional power converter. The implementation of space vector modulation technology first transforms the three-phase inner-loop current reference value from a stationary coordinate system to a rotating coordinate system. The error between the transformed current reference value and the actual current feedback value is calculated by the current regulator, which outputs the corresponding voltage reference vector. The space vector modulation algorithm selects two adjacent effective vectors and the zero vector from eight preset basic voltage vectors based on the position of the voltage reference vector on the complex plane, and calculates their respective action times. The formula for calculating the action time can be expressed as:

[0053] in: This represents the duration of the first effective voltage vector. This represents the duration of the second effective voltage vector. The switching period represents the space vector modulation algorithm. Represents the modulation ratio. This represents the angle between the voltage reference vector and the first effective voltage vector. Based on the calculated... and By allocating the on and off times of the corresponding switching devices, a pulse width modulation wave is generated, thereby controlling the insulated gate bipolar transistor in the bidirectional power converter to achieve bidirectional energy flow and DC voltage stability.

[0054] See Figure 5 This is a power balance analysis chart for a photovoltaic-storage microgrid under multiple states, showcasing the power interaction relationships between photovoltaic power generation, energy storage, the grid, and load demand in different operating scenarios. It serves as a core visualization tool for the energy dispatch strategy of the photovoltaic-storage system. The chart intuitively illustrates the energy dispatch strategy of the photovoltaic-storage microgrid, and its core functions include verifying the rationality of the "source-storage-grid-load" power matching under different scenarios; evaluating whether the charging and discharging strategy of the energy storage system is optimal; and optimizing the grid interaction strategy. This type of chart is a key analytical tool for the photovoltaic-storage microgrid energy management system, helping to achieve the goals of "maximizing photovoltaic absorption, minimizing grid dependence, and stabilizing load power supply."

[0055] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0056] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A switching control method for a photovoltaic-storage AC / DC hybrid microgrid, characterized in that, The method includes: Collect operating parameters of the power supply system, analyze the power distribution and potential weak links of the current power grid based on the topology data of the 750 kV and above AC transmission network, and generate an initial network structure model. By embedding real-time alarm thresholds of a large-scale power grid security and defense system and an intelligent dispatch system into the initial network structure model, a dynamic monitoring framework is formed. The voltage and phase information of the AC / DC bus connection point is obtained through a dynamic monitoring framework, compared with the reference standard of the off-grid status, and the stability margin of the current operation mode of the microgrid is determined. Based on the calculation results of the stability margin, the power coordination and redistribution mechanism of the photovoltaic-storage system is triggered; Calculate the output priority sequence of each distributed power source under the power coordination and redistribution mechanism; Based on the state of charge of the DC bus energy storage unit, the output priority sequence is modified to generate a power allocation instruction set; The power allocation instruction set is converted into control instructions for specific power units, which control the grid connection interface device of the photovoltaic-storage system, synchronously adjust the working status of the bidirectional power conversion device, and complete the switching of the AC / DC hybrid microgrid operation mode.

2. The switching control method based on a photovoltaic-storage AC / DC hybrid microgrid according to claim 1, characterized in that, The process of collecting power supply system operating parameters, based on the topology data of AC transmission networks above 750 kV, analyzes the current power distribution and potential weak points of the power grid to generate an initial network structure model, specifically including: Continuously acquire voltage level, power flow direction, line load rate and relay protection setting information from the power grid dispatching master station and local monitoring terminal, and integrate the voltage level, power flow direction, line load rate and relay protection setting information into a multi-dimensional set of operating parameters; Import the predefined electrical connection relationships and equipment parameters of the 750 kV and above AC transmission backbone network to construct a basic topology knowledge graph; The multi-dimensional set of operating parameters is mapped to the corresponding nodes and branches of the basic topological knowledge graph, and the operating status markers of the relevant branches are activated. Based on the operational status marking, nodes and lines with load rates exceeding the warning line, voltage exceeding the limit, or abnormal operation frequency of protection devices are identified in the basic topology knowledge graph and marked as potential weak links. The potential weak links are marked in the basic topology knowledge graph, and combined with real-time power flow data, an initial network structure model with the weights of the weak links and the real-time power flow direction is generated.

3. The switching control method for a photovoltaic-storage hybrid AC / DC microgrid according to claim 2, characterized in that, The process of embedding real-time alarm thresholds of a large-scale power grid security and defense system and intelligent dispatching system into the initial network structure model to form a dynamic monitoring framework specifically includes: From the configuration database of the large-scale power grid security and defense system and intelligent dispatch system, call the real-time operation alarm threshold set for different voltage levels and different equipment types. The alarm threshold set includes voltage upper limit, voltage lower limit, frequency upper limit, frequency lower limit, and power oscillation amplitude threshold. The alarm threshold set is associated and bound with each electrical node and line in the initial network structure model, and a one-to-one threshold comparison rule is established for each monitoring point; In the initial network structure model, a corresponding data acquisition channel and logical judgment unit are configured for each threshold comparison rule. The logical judgment unit receives the running data of the corresponding monitoring point in real time. When the operational data of the monitoring point triggers any alarm threshold, the logic judgment unit outputs an out-of-limit event identifier with a timestamp and severity level; By aggregating the out-of-limit event identifiers output by all logical judgment units and combining them with the topological relationship of the initial network structure model, the scope of event impact is assessed, forming a dynamic monitoring framework with real-time state awareness, threshold comparison, and event evaluation capabilities.

4. The switching control method based on a photovoltaic-storage AC / DC hybrid microgrid according to claim 3, characterized in that, The process of acquiring voltage and phase information at AC / DC bus connection points through a dynamic monitoring framework, comparing it with reference standards for off-grid status, and determining the stability margin of the microgrid's current operating mode specifically includes: The amplitude and phase angle of the AC bus voltage at the grid connection point, as well as the amplitude of the DC bus voltage, are simultaneously collected through the preset data acquisition channels in the dynamic monitoring framework. Call the grid-connected operation status reference standard database to obtain the current allowable range of grid-connected voltage amplitude, allowable range of phase synchronization deviation, and allowable range of DC voltage. Call the off-grid operation status reference standard database to obtain the current allowable range of off-grid voltage amplitude, allowable frequency fluctuation range, and allowable DC voltage range; The collected AC bus voltage amplitude and phase angle at the grid connection point are compared item by item with the grid connection operation status reference standard, and the deviation values ​​of each item are calculated. The collected DC bus voltage amplitude is compared with the allowable range of DC voltage under both grid-connected and off-grid conditions to calculate the voltage deviation. Based on the magnitude and rate of change of each deviation value, a comprehensive stability margin quantification index is calculated using a preset fuzzy evaluation rule. The stability margin quantification index is used to characterize the degree of distance between the current operating mode and the boundary state.

5. The switching control method for a photovoltaic-storage AC / DC hybrid microgrid according to claim 4, characterized in that, The power coordination and reallocation mechanism of the photovoltaic-storage system, triggered based on the stability margin calculation results, specifically includes: The stability margin quantification is received in real time, and a start threshold and an emergency threshold are set. When the stability margin quantification index is lower than the activation threshold but higher than the emergency threshold, it is determined to be in preventive adjustment mode, triggering the preventive power coordination and redistribution process; When the stability margin quantification index is lower than the emergency threshold, it is determined to be in emergency control mode, triggering the emergency power coordination and redistribution process. In the preventive power coordination and redistribution process, with the goal of maintaining the current operating mode, the output power correction of the maximum power point tracking stage of the photovoltaic array in the photovoltaic-storage system is initiated, and the pre-adjustment of the four-quadrant operating capability of the energy storage converter is initiated. In the emergency power coordination and redistribution process, with the goal of quickly switching to a safe operating mode, preparations are made for the graded disconnection of critical loads, and the power flow direction setting of the energy storage converter is forcibly changed.

6. The switching control method for a photovoltaic-storage AC / DC hybrid microgrid according to claim 5, characterized in that, The output priority sequence of each distributed power source under the computational power coordination and redistribution mechanism specifically includes: The system obtains the actual output of the photovoltaic array in the photovoltaic-storage system at the current moment, the real-time state of charge and adjustable power range of the energy storage unit, and the upper and lower limits of the output of the controllable micro-power source. Obtain the current net load demand of the microgrid and the stability margin quantification from the dynamic monitoring framework; In the preventive adjustment mode, with the main objectives of smoothing net load fluctuations and delaying the decline in stability margin, an objective function is established, and the optimal planned output of each distributed power source in the next scheduling cycle is obtained by solving it. In emergency control mode, the main objectives are to make up for the power deficit or absorb the excess power and restore the stability margin as quickly as possible. An objective function is established and the emergency adjustment output of each distributed power source is obtained by solving it. The optimal planned output or emergency adjustment output obtained by the solution is compared with the current actual output of each distributed power source to calculate the amount and direction of power adjustment required for each power source. Based on the magnitude of the power adjustment, the speed of the adjustment, and the level of equipment operating costs, a multi-attribute decision-making method is used to calculate the output priority sequence of each distributed power source in this adjustment.

7. The switching control method for a photovoltaic-storage hybrid AC / DC microgrid according to claim 6, characterized in that, The process of combining the state of charge of the DC bus energy storage unit to modify the output priority sequence and generate a power allocation instruction set specifically includes: Real-time monitoring of the current state of charge of all energy storage units on the DC bus, and acquisition of their rated capacity and maximum charging and discharging power limits; Determine the overall power balance status of the microgrid. If it is a power deficit state, check whether the current state of charge of each energy storage unit is higher than the discharge protection lower limit. If it is a power surplus state, check whether the current state of charge of each energy storage unit is lower than the charging protection upper limit. Based on the power output priority sequence, the charging and discharging feasibility of each energy storage unit is checked sequentially. For energy storage units that are ranked high in the output priority sequence but whose state of charge does not meet the feasibility of charging and discharging, their output command in this adjustment will be forcibly set to zero, and they will be temporarily removed from the list of schedulable resources in this adjustment. Based on the updated list of schedulable resources, power adjustment tasks are reallocated to ensure that the total adjustment amount matches the demand, forming the final power allocation instruction set that includes specific power values, direction of change, and execution sequence.

8. The switching control method for a photovoltaic-storage AC / DC hybrid microgrid according to claim 7, characterized in that, The device for converting the power allocation instruction set into control instructions for specific power units and controlling the grid connection interface of the photovoltaic-storage system specifically includes: The power allocation instruction set is analyzed to include a portion of the instructions for the grid-connected interface device. The power allocation instruction includes a target power value, a power change rate limit, and an execution time window. Convert the target power value into reference values ​​for active and reactive current of the converter in the grid-connected interface device; Based on the power change rate limit, the change trajectory of active and reactive current reference values ​​is planned to generate a smooth current reference trajectory. Based on the current reference trajectory, a model predictive control method is used to calculate the optimal switching state combination of the converter switching devices in the grid-connected interface device at the next moment in each control cycle. The optimal switching state combination is converted into a drive pulse and sent to the power switching device drive circuit of the grid-connected interface device. The voltage and current at the grid connection point are monitored synchronously, the actual current value is compared with the current reference trajectory, and the internal model parameters of the model predictive control are corrected through closed-loop feedback.

9. A switching control method for a photovoltaic-storage hybrid AC / DC microgrid according to claim 8, characterized in that, The synchronous adjustment of the bidirectional power conversion device's operating state to complete the switching of the AC / DC hybrid microgrid's operating mode specifically includes: The power distribution instruction set is analyzed for some instructions of the bidirectional power conversion device between the DC bus and the AC bus. These instructions include the power transmission direction, power transmission magnitude, and DC voltage regulation target. Based on the requirements of power transmission direction and magnitude, the control mode of the bidirectional power conversion device is set to constant power control or DC voltage control. In constant power control mode, the commanded power value is used as the setpoint for the outer loop power controller, and the controller calculates the inner loop current reference value. In DC voltage control mode, the DC voltage regulation target of the command is used as the given value of the outer loop voltage controller, and the inner loop current reference value is calculated by the controller. Based on the calculated inner loop current reference value, space vector modulation technology is used to generate the switching signal of the bidirectional power converter, and control it to achieve bidirectional energy flow and DC voltage stability. During the transient process of mode switching, the timing of the control command output of the grid-connected interface device and the bidirectional power conversion device is coordinated to ensure the synchronization and smooth transition of power on the AC side and the DC side, so that the microgrid as a whole can operate in the target mode.

10. A switching control system based on a photovoltaic-storage AC / DC hybrid microgrid, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the switching control method for a hybrid AC / DC microgrid based on photovoltaic storage as described in any one of claims 1 to 9.

Images