Power supply zero perception switching method and system based on phase real-time tracking

Abstract

The invention relates to the technical field of electric power, in particular to a power supply zero-sensing switching method and system based on phase real-time tracking, and the method comprises the following steps: collecting the communication connection state of each inverter node in a micro-grid group, constructing a communication topology model of the micro-grid group, and analyzing a sparse communication connection relationship among the nodes in the communication topology model, and generating a Laplacian matrix reflecting node connectivity according to the sparse communication connection relationship. According to the method, the system state can be comprehensively evaluated from the two dimensions of external alignment precision and internal stability, and misjudgment caused by a single index is avoided. Dual errors are input into a switching stability evaluation model to calculate a stability index, and a switching instruction is generated only when the index exceeds a grid-connected permission threshold value, so that quantitative decision and strict control of a grid-connected opportunity are realized, and the micro-grid and a main grid are highly synchronous in phase and frequency at the moment that a circuit breaker is closed; and transient impact and circulation at the switching-on moment are eliminated.

Description

Technical Field

[0001] This invention relates to the field of power technology, and in particular to a power supply zero-awareness switching method and system based on real-time phase tracking. Background Technology

[0002] Existing technologies rely on threshold judgments of a single voltage amplitude or frequency at the point of common coupling (PCC) in practical applications. During operation, simple parameter matching at the PCC alone cannot accurately reflect the dynamic balance within the microgrid group. This leads to situations where, although the PCC parameters meet requirements at the moment of closing, phase discrepancies may still exist between the internal inverters. This crude control strategy is prone to inrush current during switching operations due to phase misalignment, causing arcing and erosion of circuit breaker contacts, thus shortening equipment lifespan. Therefore, improvements are needed. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the existing technology and propose a power supply zero-sensory switching method and system based on real-time phase tracking.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a power supply zero-sensory switching method based on real-time phase tracking, comprising the following steps: The communication connection status of each inverter node in the microgrid group is collected, the communication topology model of the microgrid group is constructed, the sparse communication connection relationship between each node in the communication topology model is analyzed, and a Laplace matrix reflecting the node connectivity is generated based on the sparse communication connection relationship. The leader node and follower node in the microgrid group are determined. The leader node is controlled to collect the main grid voltage phase in real time. The phase interaction channel between the follower node and the neighboring node is established according to the Laplace matrix. The output phase of the follower node is adjusted by using the main grid voltage phase as a reference trajectory to obtain the real-time output phase of each inverter node. The real-time output phase is monitored in real time, the phase difference between the pilot node and the main grid voltage phase is calculated to obtain the main grid tracking error, the phase distribution of each inverter node at the same time is analyzed, the phase dispersion within the microgrid group is calculated, and the synchronization consistency error is obtained. The main network tracking error and the synchronization consistency error are input into a preset handover stability assessment model. The handover stability assessment model is used to calculate the handover stability index at the current moment. The handover stability index is then compared with a preset grid connection permit threshold. When the switching stability index is greater than the grid connection permit threshold, the system is determined to meet the zero-aware switching conditions, generates a grid connection switching command, and controls the circuit breaker at the common connection point to close according to the grid connection switching command, thus completing the zero-aware power supply switching.

[0005] Preferably, the steps of collecting the communication connection status of each inverter node in the microgrid group, constructing a communication topology model of the microgrid group, analyzing the sparse communication connection relationships between each node in the communication topology model, and generating a Laplace matrix reflecting the node connectivity based on the sparse communication connection relationships are as follows: By scanning the physical location and signal transmission path of all inverter nodes in the microgrid group through a sensor network, the unidirectional or bidirectional communication links between each inverter node are identified. Each inverter node is mapped to a vertex in a graph theory model, and the communication link is mapped to an edge in a graph theory model. A directed graph topology is constructed as the communication topology model. Analyze the adjacency matrix of the communication topology model to determine the connection weight between any two inverter nodes. Calculate the in-degree matrix of the communication topology model based on the connection weight and the adjacency matrix. Subtract the adjacency matrix from the in-degree matrix to obtain the Laplace matrix describing the internal communication structure of the microgrid group.

[0006] Preferably, the steps of determining the lead node and follower nodes in the microgrid group, controlling the lead node to collect the main grid voltage phase in real time, establishing a phase interaction channel between the follower node and its neighboring nodes based on the Laplace matrix, and using the main grid voltage phase as a reference trajectory to adjust the output phase of the follower node to obtain the real-time output phase of each inverter node are as follows: Inverter nodes in the microgrid group that are directly connected to the main grid are marked as the pilot nodes, and inverter nodes that cannot directly obtain the main grid voltage phase are marked as the follower nodes. The pilot nodes are controlled to periodically read the main grid voltage phase through a phase-locked loop. Based on the Laplace matrix, a distributed consensus protocol is initiated. The follower node is controlled to read the local phase information of the neighbor node through the phase interaction channel, calculate the local phase difference between the follower node and the neighbor node, multiply the local phase difference by a preset coupling gain coefficient to obtain a frequency compensation amount, and add the frequency compensation amount to the reference frequency of the follower node. This drives the phase of the follower node to converge to the main network voltage phase obtained by the leader node until all the real-time output phases reach dynamic equilibrium.

[0007] Preferably, the steps of monitoring the real-time output phase, calculating the phase difference between the pilot node and the main grid voltage phase to obtain the main grid tracking error, analyzing the phase distribution of each inverter node at the same time, calculating the phase dispersion within the microgrid group, and obtaining the synchronization consistency error are as follows: All real-time output phases are synchronously acquired within a preset sampling period. The phase value of the navigation node is extracted as the reference tracking phase. The absolute difference between the reference tracking phase and the main grid voltage phase is calculated. The absolute difference of multiple consecutive sampling periods is subjected to moving average filtering to obtain the main grid tracking error. The average value of all the real-time output phases is calculated as the average phase of the microgrid group. The sum of squared deviations between each real-time output phase and the average phase of the microgrid group is calculated. The sum of squared deviations is normalized to obtain the synchronization consistency error.

[0008] Preferably, the step of inputting the main network tracking error and the synchronization consistency error into a preset handover stability assessment model, calculating the handover stability index at the current moment using the handover stability assessment model, and comparing the handover stability index with a preset grid connection permit threshold specifically includes: The main network tracking error and the synchronization consistency error are input into the pre-constructed handover stability assessment model. The main network tracking error is mapped through the logarithmic saturation function in the handover stability assessment model, and the synchronization consistency error is mapped through the Gaussian exponential decay function in the handover stability assessment model. The handover stability index, which characterizes the current grid connection risk level, is obtained by weighted calculation. The calculation formula for the switching stability assessment model is as follows: ,in This represents the switching stability index. This represents the preset gain coefficient of the external tracking weights. Represents the logarithmic function with base 10. The preset baseline constant representing the mainnet tracking accuracy. This indicates the mainnet tracking error. This indicates a preset small positive number to prevent the denominator from being zero. The preset gain coefficient representing the internal consistency weight. This represents an exponential function with base e. The preset adjustment factor represents the uniform convergence rate. This indicates the synchronization consistency error; Read the preset grid connection permit threshold from the memory, determine whether the calculated switching stability index is greater than the grid connection permit threshold, and if the condition is met, confirm that the system is in a stable grid connection state.

[0009] Preferably, when the switching stability index is greater than the grid connection permit threshold, the system is determined to meet the zero-aware switching conditions, a grid connection switching command is generated, and the circuit breaker at the point of common coupling is closed according to the grid connection switching command to complete the zero-aware power supply switching. The specific steps are as follows: When the switching stability index is detected to be greater than the grid connection permit threshold for N consecutive control cycles, it is confirmed that the microgrid group and the main grid have reached a pre-synchronization state, the grid connection switching instruction with a high priority flag is generated, and the grid connection switching instruction is sent to the intelligent circuit breaker controller of the point of common connection. After receiving the grid connection switching command, the intelligent circuit breaker controller performs a closing operation to close the circuit breaker at the common connection point and broadcasts a grid connection confirmation signal. After receiving the grid connection confirmation signal, each inverter node in the microgrid group switches from islanded control mode to grid-connected current control mode and uses the Laplace matrix to maintain the power distribution balance after grid connection, thus completing the zero-aware power supply switching.

[0010] Preferably, before calculating the handover stability index, the method further includes a step of adaptively calibrating the sensitivity parameters of the handover stability assessment model, specifically: Collect historical switching data of the microgrid group, analyze the amplitude of circulating current impact during the historical switching process, and identify the sources of deviations that cause the amplitude of the circulating current impact; If the circulating current impact amplitude in the historical switching data mainly originates from phase tracking deviation, then the value of the preset reference constant for the main grid tracking accuracy is reduced; if the circulating current impact amplitude in the historical switching data mainly originates from internal inverter circulating current, then the value of the preset adjustment factor for the consistency convergence rate is increased.

[0011] Preferably, after controlling the pilot node to periodically read the main grid voltage phase through a phase-locked loop, the method further includes a step of detecting anomalies in the main grid voltage phase, specifically: Monitor the frequency change rate of the main grid voltage phase, determine whether the frequency change rate exceeds a preset grid fluctuation threshold, and if the frequency change rate does not exceed the grid fluctuation threshold, input the main grid voltage phase as a valid reference value into the distributed consensus protocol. If the frequency change rate exceeds the power grid fluctuation threshold, the main grid is determined to be in a transient disturbance state. The historical main grid phase collected at the previous moment is linearly extrapolated to generate a virtual main grid reference phase. The virtual main grid reference phase is used to replace the main grid voltage phase to participate in the phase iteration of the follower node.

[0012] The present invention also provides a system comprising: The communication topology modeling module is used to collect the communication connection status of each inverter node in the microgrid group, construct the communication topology model of the microgrid group, parse the sparse communication connection relationship between each node in the communication topology model, and generate a Laplace matrix reflecting the node connectivity based on the sparse communication connection relationship. The phase tracking control module is used to determine the lead node and follower node in the microgrid group, control the lead node to collect the main grid voltage phase in real time, establish the phase interaction channel between the follower node and neighboring nodes according to the Laplace matrix, and use the main grid voltage phase as a reference trajectory to traction adjust the output phase of the follower node to obtain the real-time output phase of each inverter node. The error monitoring and analysis module is used to monitor the real-time output phase in real time, calculate the phase difference between the pilot node and the main grid voltage phase to obtain the main grid tracking error, analyze the phase distribution of each inverter node at the same time, calculate the phase dispersion within the microgrid group, and obtain the synchronization consistency error. The handover stability assessment module is used to input the main network tracking error and the synchronization consistency error into a preset handover stability assessment model, calculate the handover stability index at the current moment using the handover stability assessment model, and compare the handover stability index with a preset grid connection permit threshold. The grid-connected switching execution module is used to determine that the system meets the zero-aware switching conditions when the switching stability index is greater than the grid-connected permit threshold, generate a grid-connected switching command, and control the circuit breaker at the common connection point to close according to the grid-connected switching command, thereby completing the zero-aware power supply switching.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are as follows: This invention collects the communication connection status of each inverter node in a microgrid cluster and analyzes the sparse communication connection relationships between nodes. It constructs a communication topology model and generates a Laplace matrix reflecting node connectivity, thus characterizing the complex network structure within the microgrid. This provides a solid mathematical foundation for distributed collaborative control and solves the problem of communication path planning between large-scale nodes. By establishing a hierarchical architecture of leader and follower nodes, and using the leader node to lock onto the main grid voltage phase in real time as a reference trajectory, and guiding follower nodes to interact with neighboring nodes through a phase interaction channel, it enables the traction adjustment of the follower node's output phase. This allows the phase of the entire microgrid cluster to quickly converge to the main grid reference without global communication, reducing communication bandwidth consumption and improving system response speed. Simultaneously calculating the leader node's tracking error to the main grid and the synchronization consistency error within the microgrid during real-time monitoring, it comprehensively evaluates the system status from both external alignment accuracy and internal stability dimensions, avoiding misjudgments caused by a single indicator. The stability index is calculated by switching the dual error input to the stability assessment model, and a switching command is generated only when the index exceeds the grid connection permit threshold. This enables quantitative decision-making and strict control over the timing of grid connection, ensuring that the microgrid and the main grid are highly synchronized in phase and frequency at the moment the circuit breaker closes, eliminating transient impacts and circulating currents at the moment of closing, and preventing voltage flicker from interfering with sensitive loads. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] Please see Figure 1 This invention provides a technical solution: a power supply zero-sensory switching method based on real-time phase tracking, comprising the following steps: The communication connection status of each inverter node in the microgrid group is collected, a communication topology model of the microgrid group is constructed, the sparse communication connection relationship between each node in the communication topology model is analyzed, and a Laplace matrix reflecting the node connectivity is generated based on the sparse communication connection relationship. The leader node and follower node in the microgrid group are determined. The leader node is controlled to collect the main grid voltage phase in real time. The phase interaction channel between the follower node and the neighboring node is established according to the Laplace matrix. The output phase of the follower node is traction-adjusted using the main grid voltage phase as a reference trajectory to obtain the real-time output phase of each inverter node. Real-time monitoring of the real-time output phase, calculation of the phase difference between the pilot node and the main grid voltage phase to obtain the main grid tracking error, analysis of the phase distribution of each inverter node at the same moment, calculation of the phase dispersion within the microgrid group to obtain the synchronization consistency error; The main network tracking error and synchronization consistency error are input into the preset handover stability assessment model. The handover stability assessment model is used to calculate the handover stability index at the current moment. The handover stability index is then compared with the preset grid connection permit threshold. When the switching stability index is greater than the grid connection permit threshold, the system is determined to meet the zero-aware switching conditions, a grid connection switching command is generated, and the circuit breaker at the common connection point is closed according to the grid connection switching command to complete the zero-aware power supply switching.

[0017] In this embodiment, the steps of collecting the communication connection status of each inverter node in the microgrid cluster, constructing a communication topology model of the microgrid cluster, analyzing the sparse communication connection relationships between each node in the communication topology model, and generating a Laplace matrix reflecting the connectivity of the nodes based on the sparse communication connection relationships are as follows: scanning the physical location and signal transmission path of all inverter nodes in the microgrid cluster through a sensor network, identifying the one-way or two-way communication links between each inverter node, mapping each inverter node to a vertex in a graph theory model, mapping the communication links to edges in a graph theory model, constructing a directed graph topology structure as the communication topology model, analyzing the adjacency matrix of the communication topology model, determining the connection weight between any two inverter nodes, calculating the in-degree matrix of the communication topology model based on the connection weight and the adjacency matrix, and subtracting the adjacency matrix from the in-degree matrix to obtain the Laplace matrix describing the internal communication structure of the microgrid cluster.

[0018] Specifically, a sensor network consisting of fiber optic Ethernet switches or industrial-grade wireless ZigBee communication modules deployed at various physical sites in the microgrid initiates a link probe request at a preset polling period, such as every 10 milliseconds. This scans the physical location coordinates and real-time status of signal transmission paths of all inverter nodes in the microgrid cluster, identifying whether there are usable one-way or two-way data transmission communication links between the inverter nodes. If the packet loss rate between two nodes is less than 0.1% and the communication latency is less than 5 milliseconds, it is determined to be a valid link. Each inverter node is then mapped to a set of vertices in a graph theory model. ,in This represents a set containing all vertices. Represents the i-th inverter node. Representing the total number of inverter nodes in a microgrid cluster, the identified valid communication links are mapped to a set of edges in a graph theory model. ,in This represents the set containing all edges. To represent a subset relation, Representing the Cartesian product, constructing a directed graph topology containing vertices and edges as a communication topology model, and extracting the connection relationships from the communication topology model to generate an adjacency matrix. ,in Represents the adjacency matrix. This represents the element in the i-th row and j-th column of the adjacency matrix, used to characterize the connection state between the i-th node and the j-th node. If a connection exists, then... ,otherwise The adjacency matrix of the communication topology model is analyzed to determine the connection weight between any two inverter nodes. This connection weight is set based on a weighted normalization process using the reciprocal of the communication bandwidth and physical distance between the nodes. For example, when the bandwidth between node i and node j is 100 Mbps and the distance is 50 meters, the calculated weight is 0.8. If there is no connection, the weight is recorded as 0. The in-degree matrix of the communication topology model is then calculated based on the connection weights and the adjacency matrix. ,in Represents the in-degree matrix. This represents the diagonal matrix generating function. The in-degree value of the i-th node is represented by the formula: ,in The summation symbol is used to represent the summation symbol. This represents the loop variable for summation, iterating from 1 to... Subtract the adjacency matrix from the in-degree matrix by performing matrix subtraction. ,in The Laplace matrix is ​​used to describe the communication structure within the microgrid cluster. The eigenvalue distribution of this matrix will be used to determine the convergence speed of the multi-agent system, ensuring that the rank of the matrix satisfies the algebraic connectivity requirement of the connected graph.

[0019] In this embodiment, the steps of determining the lead node and follower nodes in the microgrid group, controlling the lead node to collect the main grid voltage phase in real time, establishing a phase interaction channel between the follower node and neighboring nodes based on the Laplace matrix, and using the main grid voltage phase as a reference trajectory to traction adjust the output phase of the follower node to obtain the real-time output phase of each inverter node are as follows: the inverter nodes in the microgrid group that are directly connected to the main grid are marked as lead nodes, and the inverter nodes that cannot directly obtain the main grid voltage phase are marked as follower nodes. The lead node is controlled to periodically read the main grid voltage phase through a phase-locked loop. A distributed consensus protocol is started based on the Laplace matrix. The follower nodes are controlled to read the local phase information of neighboring nodes through the phase interaction channel. The local phase difference between the follower node and the neighboring node is calculated. The local phase difference is multiplied by a preset coupling gain coefficient to obtain the frequency compensation amount. The frequency compensation amount is superimposed on the reference frequency of the follower node to drive the phase of the follower node to converge to the main grid voltage phase obtained by the lead node until all real-time output phases reach dynamic equilibrium.

[0020] Specifically, the grid connection point status registers of each inverter node are read. Inverter nodes in the microgrid group that are directly electrically connected to the main grid via the point of common coupling (PCC) and have high-precision voltage transformers are marked as lead nodes. Inverter nodes whose physical location is inside the microgrid and cannot directly obtain the main grid voltage phase through hardware sensors are marked as follow nodes. The lead nodes are controlled to discretize the main grid voltage waveform at a sampling frequency of 20kHz using a software phase-locked loop (SPLL) or a phase-locked algorithm based on a second-order generalized integrator. The main grid voltage phase is periodically read, and a distributed consensus protocol is initiated based on the Laplace matrix. This protocol requires each node to exchange data only with adjacent nodes in the topology. The follow nodes are controlled to read the local phase information of neighboring nodes through the phase interaction channel, calculate the local phase difference between the follow node's own phase and the phase of its neighboring nodes, and multiply the local phase difference by a preset coupling gain coefficient to obtain the frequency compensation. The value of this coupling gain coefficient is tuned according to Lyapunov stability theory, for example, by constructing a Lyapunov candidate function. And differentiate, where The energy scalar function of the system. This represents the system's state error vector. Represents the matrix transpose symbol. The symmetric positive definite matrix used for weighting the system state is used to ensure that the derivative is less than zero, thus deriving that the gain coefficient should be 5.0 to ensure convergence within 0.5 seconds. The frequency compensation is superimposed on the reference frequency of the follower node, which is usually set to the rated power frequency of 50Hz or 60Hz. The frequency adjustment is applied to the phase through integral operation, driving the phase of the follower node to converge to the main grid voltage phase obtained by the leader node. The above iterative process is repeated in each control cycle until the maximum deviation between all real-time output phases is less than 0.01 radians, reaching dynamic equilibrium.

[0021] In this embodiment, the steps of real-time monitoring of the real-time output phase, calculating the phase difference between the pilot node and the main grid voltage phase to obtain the main grid tracking error, analyzing the phase distribution of each inverter node at the same moment, and calculating the phase dispersion within the microgrid group to obtain the synchronization consistency error are as follows: synchronously acquiring all real-time output phases within a preset sampling period, extracting the phase value of the pilot node as the reference tracking phase, calculating the absolute difference between the reference tracking phase and the main grid voltage phase, performing moving average filtering on the absolute difference of multiple consecutive sampling periods to obtain the main grid tracking error, calculating the average value of all real-time output phases as the average phase of the microgrid group, calculating the sum of squared deviations between each real-time output phase and the average phase of the microgrid group, and normalizing the sum of squared deviations to obtain the synchronization consistency error.

[0022] Specifically, within a preset sampling period, such as 50 microseconds, the instantaneous values ​​of all real-time output phases are synchronously acquired through a global synchronization clock. The phase value of the pilot node is extracted from the memory data block as the reference tracking phase. The absolute difference between the reference tracking phase and the main grid voltage phase acquired by the hardware port is calculated. Due to the potential measurement noise in a single sampling, a moving average filter is applied to the absolute difference of multiple consecutive sampling periods, such as the most recent 50 periods. A first-in-first-out queue of length 50 is constructed. Each time new data passes through, the oldest data is removed and the average value of the data in the queue is calculated to obtain a main grid tracking error that can smoothly reflect the external synchronization accuracy. At the same time, the phase registers of all nodes are traversed, and the average value of all real-time output phases is calculated as the average phase of the microgrid group. The sum of squares of the deviation between each real-time output phase and the average phase of the microgrid group is calculated, which is the square of the difference between the phase value of each node and the average phase of the microgrid group, and then accumulated. The sum of squares of the deviation is normalized, divided by the total number of nodes and the square root to obtain the standard deviation, or divided by the preset maximum allowable phase angle deviation square value, for example... ,in The constant of pi is approximately 3.14159, and 18 is the denominator constant. The resulting synchronization consistency error can quantify the degree of dispersion between power sources within a microgrid. The smaller the value, the better the internal synchronization.

[0023] In this embodiment, the steps of inputting the main network tracking error and synchronization consistency error into a preset handover stability assessment model, calculating the handover stability index at the current moment using the handover stability assessment model, and comparing the handover stability index with a preset grid connection permit threshold are as follows: The main network tracking error and synchronization consistency error are input into the pre-constructed handover stability assessment model. The main network tracking error is mapped using a logarithmic saturation function in the handover stability assessment model, and the synchronization consistency error is mapped using a Gaussian exponential decay function in the handover stability assessment model. A weighted calculation is then performed to obtain the handover stability index characterizing the current grid connection risk level. The calculation formula for the handover stability assessment model is: ,in This indicates a switch to the stability index. This represents the preset gain coefficient of the external tracking weights. Represents the logarithmic function with base 10. The preset baseline constant representing the mainnet tracking accuracy. Indicates mainnet tracing error. This indicates a preset small positive number to prevent the denominator from being zero. The preset gain coefficient representing the internal consistency weight. This represents an exponential function with base e. The preset adjustment factor represents the uniform convergence rate. This indicates the synchronization consistency error. The preset grid connection permission threshold is read from the memory, and it is determined whether the calculated switching stability index is greater than the grid connection permission threshold. If the condition is met, the system is confirmed to be in a stable grid connection state.

[0024] Specifically, the main grid tracking error and synchronization consistency error are input into a pre-constructed handover stability assessment model. The main grid tracking error is mapped using a logarithmic saturation function in the handover stability assessment model. The nonlinear characteristics of the logarithmic function amplify the sensitivity of the low-error segment and suppress the fluctuation impact of the high-error segment. The synchronization consistency error is mapped using a Gaussian exponential decay function in the handover stability assessment model. The rapid decay characteristics of the exponential function are used to strictly penalize the internal dispersion. The handover stability index, which characterizes the current grid connection risk level, is obtained through weighted calculation. The calculation formula of the handover stability assessment model is as follows: ,in, This represents the switching stability index, which typically ranges from 0 to 1. A larger value indicates a more stable system, making it suitable for grid-connected switching operations. This represents the preset gain coefficient of the external tracking weight. This coefficient is calculated by the analytic hierarchy process (AHP) based on the main grid's tolerance to grid-connected inrush current. For example, in a weak grid environment, the main grid is sensitive to phase deviation, and this value is set to 0.6. This represents the base-10 logarithmic function, used to map ratios to the logarithmic domain, smoothing the numerical range. 1 is an offset constant to prevent the logarithm from being negative. The preset reference constant represents the main grid tracking accuracy. This constant is set according to the allowable voltage deviation value specified in the national grid connection standard GB / T 12325, for example, it is set to 0.05 radians, and serves as a reference for error evaluation. This represents the main grid tracking error. This value is derived from the moving average deviation between the pilot node and the main grid phase in the preceding steps. The unit is radians, and it reflects the overall ability of the microgrid to follow the main grid. This represents a preset small positive number to prevent the denominator from being zero. It is usually taken as the minimum value of the computer's floating-point precision, for example... This is used to maintain computational stability when the error is zero. The preset gain coefficient representing the internal consistency weight, which is related to... Complementary, for example, a value of 0.4, focuses on assessing whether circulating current risks arise within the microgrid due to phase asynchrony; This represents an exponential function with a base of approximately 2.71828 (the natural constant e), used to construct the decay curve of a Gaussian distribution. The preset adjustment factor represents the rate of convergence of the consensus. This factor controls how fast the exponential function decays. It is determined through simulation experiments. For example, setting it to 100 means that when the consensus error increases slightly, the exponential term will decay rapidly to 0, reflecting the strict limitation on the internal circulation. The synchronization consistency error is derived from the internal phase standard deviation or normalized variance calculated in the previous steps, and the unit is radians. The preset grid connection permit threshold in the memory is read. This threshold is set to 0.85, which is an empirical value derived from the statistical analysis of a large number of historical successful grid connection cases. The calculated switching stability index is then checked to see if it is greater than the grid connection permit threshold. If the condition is met, the system is confirmed to be in a stable grid connection state.

[0025] In this embodiment, when the switching stability index is greater than the grid connection permit threshold, the system is determined to meet the zero-aware switching conditions, a grid connection switching command is generated, and the circuit breaker at the point of common coupling is closed according to the grid connection switching command to complete the zero-aware power supply switching. The specific steps are as follows: when the switching stability index is detected to be greater than the grid connection permit threshold for N consecutive control cycles, it is confirmed that the microgrid group and the main grid have reached a pre-synchronization state, a grid connection switching command with a high priority flag is generated, and the grid connection switching command is sent to the intelligent circuit breaker controller at the point of common coupling. After receiving the grid connection switching command, the intelligent circuit breaker controller performs a closing operation to close the circuit breaker at the point of common coupling and broadcasts a grid connection confirmation signal. After receiving the grid connection confirmation signal, each inverter node in the microgrid group switches from islanded control mode to grid connection current control mode and uses a Laplace matrix to maintain the power distribution balance after grid connection, thus completing the zero-aware power supply switching.

[0026] Specifically, when the switching stability index is detected to be greater than the grid connection permit threshold for N consecutive control cycles, for example, when the index remains above 0.85 for 5 consecutive cycles (100 milliseconds), to eliminate misjudgments caused by signal glitches or transient interference, and to confirm that the microgrid group and the main grid have reached a pre-synchronization state, a grid connection switching command with a high priority flag is generated. This command complies with IEC 61850. The GOOSE communication protocol format sends grid connection switching commands to the smart circuit breaker controller at the point of common coupling via a high-speed fiber optic network. Upon receiving the grid connection switching command, the smart circuit breaker controller immediately drives the operating mechanism to release the energy storage spring, performs a closing operation to close the circuit breaker at the point of common coupling, and broadcasts a grid connection confirmation signal within 2 milliseconds after the closing operation is completed. Upon receiving the grid connection confirmation signal, each inverter node in the microgrid immediately adjusts its underlying control logic, smoothly switching from the islanded control mode (V / f control mode) that maintains voltage amplitude and frequency to the grid-connected current control mode (P / Q control mode) that tracks the main grid current. It also uses a Laplace matrix to maintain the power distribution balance after grid connection, completing the zero-aware power supply switching.

[0027] In this embodiment, before calculating the switching stability index, the method further includes an adaptive calibration step for the sensitivity parameters of the switching stability assessment model. Specifically, this involves: collecting historical switching data of the microgrid group, analyzing the amplitude of circulating current impacts during historical switching processes, identifying the sources of deviations that cause the amplitude of circulating current impacts, and if the amplitude of circulating current impacts in the historical switching data mainly originates from phase tracking deviations, then reducing the value of the preset reference constant for the main grid tracking accuracy; if the amplitude of circulating current impacts in the historical switching data mainly originates from internal inverter circulating currents, then increasing the value of the preset adjustment factor for the consistency convergence rate.

[0028] Specifically, historical switching data of the microgrid cluster is collected. This data is stored in a non-volatile database and includes voltage and current waveform records before and after the past one hundred grid connection operations. The circulating current impact amplitude during historical switching processes is analyzed. The fundamental and harmonic components of the impact current at the moment of closing are extracted using Fast Fourier Transform (FFT). The ratio of the peak impact current to the rated current is calculated to identify the sources of deviation leading to the circulating current impact amplitude. By comparing the phase tracking data at the moment before closing with the internal consistency data, if a positive correlation is found between the correlation coefficient of the phase tracking error and the impact amplitude, i.e., if the circulating current impact amplitude in the historical switching data mainly comes from the phase tracking deviation, the preset reference constant for the main grid tracking accuracy is reduced. The value, for example, is adjusted from 0.05 to 0.03, making the logarithmic term in the formula more sensitive to error, thereby reducing the score under the same error in future evaluations and raising the grid connection threshold. If the circulating current impact amplitude in the historical switching data mainly comes from the internal inverter circulating current, that is, the circulating current is caused by poor internal consistency data, then the preset adjustment factor of the consistency convergence rate is increased. The value can be adjusted, for example, from 100 to 120, to speed up the decay of the Gaussian function and make the model more stringent on internal inconsistencies.

[0029] In this embodiment, after the pilot node periodically reads the main grid voltage phase through a phase-locked loop, the method further includes a step of anomaly detection of the main grid voltage phase. Specifically, this involves monitoring the frequency change rate of the main grid voltage phase, determining whether the frequency change rate exceeds a preset grid fluctuation threshold, and if the frequency change rate does not exceed the grid fluctuation threshold, inputting the main grid voltage phase as a valid reference value into the distributed consensus protocol. If the frequency change rate exceeds the grid fluctuation threshold, it is determined that the main grid is in a transient disturbance state, and a virtual main grid reference phase is generated by linear extrapolation using the historical main grid phase collected at the previous moment. The virtual main grid reference phase is then used to replace the main grid voltage phase in the phase iteration of the follower node.

[0030] Specifically, the rate of change of frequency (ROCOF) of the main grid voltage phase is obtained by calculating the frequency difference between two adjacent sampling periods and dividing by the time interval. ,in Indicates the change in frequency. This represents the change in frequency over time. It determines whether the rate of frequency change exceeds a preset grid fluctuation threshold, which is set at 2 Hz / s according to the power system safety and stability guidelines. If the rate of frequency change does not exceed the grid fluctuation threshold, it indicates that the main grid is operating in a steady state, and the main grid voltage phase is input into the distributed consensus protocol as a valid reference value. If the rate of frequency change exceeds the grid fluctuation threshold, it is determined that the main grid is in a transient disturbance state or has experienced a fault. In this case, the directly sampled phase data may contain jumps or noise. Linear extrapolation is performed using the historical main grid phase collected at the previous moment, according to the formula... Generate a virtual mainnet reference phase, where This represents the predicted mainnet phase value at the next moment. This represents the main network phase value collected at the current moment. This indicates that the constant value of pi is approximately 3.14159. This represents the steady-state frequency value of the main network recorded at the moment before the fault occurred. This indicates the sampling period of the control system. The virtual main grid reference phase is used to replace the main grid voltage phase to participate in the phase iteration of the follower node, preventing main grid fluctuations from causing oscillation interference to the microgrid group.

[0031] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A power supply zero-aware switching method based on real-time phase tracking, characterized in that, Includes the following steps: The communication connection status of each inverter node in the microgrid group is collected, the communication topology model of the microgrid group is constructed, the sparse communication connection relationship between each node in the communication topology model is analyzed, and a Laplace matrix reflecting the node connectivity is generated based on the sparse communication connection relationship. The leader node and follower node in the microgrid group are determined. The leader node is controlled to collect the main grid voltage phase in real time. The phase interaction channel between the follower node and the neighboring node is established according to the Laplace matrix. The output phase of the follower node is adjusted by using the main grid voltage phase as a reference trajectory to obtain the real-time output phase of each inverter node. The real-time output phase is monitored in real time, the phase difference between the pilot node and the main grid voltage phase is calculated to obtain the main grid tracking error, the phase distribution of each inverter node at the same time is analyzed, the phase dispersion within the microgrid group is calculated, and the synchronization consistency error is obtained. The main network tracking error and the synchronization consistency error are input into a preset handover stability assessment model. The handover stability assessment model is used to calculate the handover stability index at the current moment. The handover stability index is then compared with a preset grid connection permit threshold. When the switching stability index is greater than the grid connection permit threshold, the system is determined to meet the zero-aware switching conditions, generates a grid connection switching command, and controls the circuit breaker at the common connection point to close according to the grid connection switching command, thus completing the zero-aware power supply switching.

2. The power supply zero-aware switching method based on real-time phase tracking according to claim 1, characterized in that, The specific steps for collecting the communication connection status of each inverter node in the microgrid group, constructing the communication topology model of the microgrid group, analyzing the sparse communication connection relationships between each node in the communication topology model, and generating a Laplace matrix reflecting the node connectivity based on the sparse communication connection relationships are as follows: By scanning the physical location and signal transmission path of all inverter nodes in the microgrid group through a sensor network, the unidirectional or bidirectional communication links between each inverter node are identified. Each inverter node is mapped to a vertex in a graph theory model, and the communication link is mapped to an edge in a graph theory model. A directed graph topology is constructed as the communication topology model. Analyze the adjacency matrix of the communication topology model to determine the connection weight between any two inverter nodes. Calculate the in-degree matrix of the communication topology model based on the connection weight and the adjacency matrix. Subtract the adjacency matrix from the in-degree matrix to obtain the Laplace matrix describing the internal communication structure of the microgrid group.

3. The power supply zero-aware switching method based on real-time phase tracking according to claim 1, characterized in that, The specific steps for determining the leader node and follower nodes in the microgrid group, controlling the leader node to collect the main grid voltage phase in real time, establishing a phase interaction channel between the follower node and its neighboring nodes based on the Laplace matrix, and using the main grid voltage phase as a reference trajectory to adjust the output phase of the follower node to obtain the real-time output phase of each inverter node are as follows: Inverter nodes in the microgrid group that are directly connected to the main grid are marked as the pilot nodes, and inverter nodes that cannot directly obtain the main grid voltage phase are marked as the follower nodes. The pilot nodes are controlled to periodically read the main grid voltage phase through a phase-locked loop. Based on the Laplace matrix, a distributed consensus protocol is initiated. The follower node is controlled to read the local phase information of the neighbor node through the phase interaction channel, calculate the local phase difference between the follower node and the neighbor node, multiply the local phase difference by a preset coupling gain coefficient to obtain a frequency compensation amount, and add the frequency compensation amount to the reference frequency of the follower node. This drives the phase of the follower node to converge to the main network voltage phase obtained by the leader node until all the real-time output phases reach dynamic equilibrium.

4. The power supply zero-aware switching method based on real-time phase tracking according to claim 1, characterized in that, The specific steps for real-time monitoring of the real-time output phase, calculating the phase difference between the pilot node and the main grid voltage phase to obtain the main grid tracking error, analyzing the phase distribution of each inverter node at the same moment, calculating the phase dispersion within the microgrid group, and obtaining the synchronization consistency error are as follows: All real-time output phases are synchronously acquired within a preset sampling period. The phase value of the navigation node is extracted as the reference tracking phase. The absolute difference between the reference tracking phase and the main grid voltage phase is calculated. The absolute difference of multiple consecutive sampling periods is subjected to moving average filtering to obtain the main grid tracking error. The average value of all the real-time output phases is calculated as the average phase of the microgrid group. The sum of squared deviations between each real-time output phase and the average phase of the microgrid group is calculated. The sum of squared deviations is normalized to obtain the synchronization consistency error.

5. The power supply zero-aware switching method based on real-time phase tracking according to claim 1, characterized in that, The steps of inputting the main network tracking error and the synchronization consistency error into a preset handover stability assessment model, calculating the handover stability index at the current moment using the handover stability assessment model, and comparing the handover stability index with a preset grid connection permit threshold are as follows: The main network tracking error and the synchronization consistency error are input into the pre-constructed handover stability assessment model. The main network tracking error is mapped through the logarithmic saturation function in the handover stability assessment model, and the synchronization consistency error is mapped through the Gaussian exponential decay function in the handover stability assessment model. The handover stability index, which characterizes the current grid connection risk level, is obtained by weighted calculation. The calculation formula for the switching stability assessment model is as follows: ,in This represents the switching stability index. This represents the preset gain coefficient of the external tracking weights. Represents the logarithmic function with base 10. The preset baseline constant representing the mainnet tracking accuracy. This indicates the mainnet tracking error. This indicates a preset small positive number to prevent the denominator from being zero. The preset gain coefficient representing the internal consistency weight. This represents an exponential function with base e. The preset adjustment factor represents the uniform convergence rate. This indicates the synchronization consistency error; The system reads the preset grid connection permit threshold from the memory, determines whether the calculated switching stability index is greater than the grid connection permit threshold, and confirms that the system is in a stable grid connection state if the condition is met.

6. The power supply zero-aware switching method based on real-time phase tracking according to claim 1, characterized in that, When the switching stability index is greater than the grid connection permit threshold, the system is determined to meet the zero-aware switching conditions, a grid connection switching command is generated, and the circuit breaker at the point of common coupling is closed according to the grid connection switching command to complete the zero-aware power supply switching. The specific steps are as follows: When the switching stability index is detected to be greater than the grid connection permit threshold for N consecutive control cycles, it is confirmed that the microgrid group and the main grid have reached a pre-synchronization state, the grid connection switching instruction with a high priority flag is generated, and the grid connection switching instruction is sent to the intelligent circuit breaker controller of the point of common connection. After receiving the grid connection switching command, the intelligent circuit breaker controller performs a closing operation to close the circuit breaker at the common connection point and broadcasts a grid connection confirmation signal. After receiving the grid connection confirmation signal, each inverter node in the microgrid group switches from islanded control mode to grid-connected current control mode and uses the Laplace matrix to maintain the power distribution balance after grid connection, thus completing the zero-aware power supply switching.

7. The power supply zero-aware switching method based on real-time phase tracking according to claim 5, characterized in that, Before calculating the handover stability index, the method further includes a step of adaptively calibrating the sensitivity parameters of the handover stability assessment model, specifically: Collect historical switching data of the microgrid group, analyze the amplitude of circulating current impact during the historical switching process, and identify the sources of deviations that cause the amplitude of the circulating current impact; If the circulating current impact amplitude in the historical switching data mainly originates from phase tracking deviation, then the value of the preset reference constant for the main grid tracking accuracy is reduced; if the circulating current impact amplitude in the historical switching data mainly originates from internal inverter circulating current, then the value of the preset adjustment factor for the consistency convergence rate is increased.

8. The power supply zero-aware switching method based on real-time phase tracking according to claim 3, characterized in that, After controlling the pilot node to periodically read the main grid voltage phase through a phase-locked loop, the method further includes a step of detecting anomalies in the main grid voltage phase, specifically: Monitor the frequency change rate of the main grid voltage phase, determine whether the frequency change rate exceeds a preset grid fluctuation threshold, and if the frequency change rate does not exceed the grid fluctuation threshold, input the main grid voltage phase as a valid reference value into the distributed consensus protocol. If the frequency change rate exceeds the power grid fluctuation threshold, the main grid is determined to be in a transient disturbance state. The historical main grid phase collected at the previous moment is linearly extrapolated to generate a virtual main grid reference phase. The virtual main grid reference phase is used to replace the main grid voltage phase to participate in the phase iteration of the follower node.

9. The system of the power supply zero-aware switching method based on real-time phase tracking according to any one of claims 1-8, characterized in that, include: The communication topology modeling module is used to collect the communication connection status of each inverter node in the microgrid group, construct the communication topology model of the microgrid group, parse the sparse communication connection relationship between each node in the communication topology model, and generate a Laplace matrix reflecting the node connectivity based on the sparse communication connection relationship. The phase tracking control module is used to determine the lead node and follower node in the microgrid group, control the lead node to collect the main grid voltage phase in real time, establish the phase interaction channel between the follower node and neighboring nodes according to the Laplace matrix, and use the main grid voltage phase as a reference trajectory to traction adjust the output phase of the follower node to obtain the real-time output phase of each inverter node. The error monitoring and analysis module is used to monitor the real-time output phase in real time, calculate the phase difference between the pilot node and the main grid voltage phase to obtain the main grid tracking error, analyze the phase distribution of each inverter node at the same time, calculate the phase dispersion within the microgrid group, and obtain the synchronization consistency error. The handover stability assessment module is used to input the main network tracking error and the synchronization consistency error into a preset handover stability assessment model, calculate the handover stability index at the current moment using the handover stability assessment model, and compare the handover stability index with a preset grid connection permit threshold. The grid-connected switching execution module is used to determine that the system meets the zero-aware switching conditions when the switching stability index is greater than the grid-connected permit threshold, generate a grid-connected switching command, and control the circuit breaker at the common connection point to close according to the grid-connected switching command, thereby completing the zero-aware power supply switching.

Images