Method and system for evaluating black-start capability of grid-forming new energy and energy storage device

Abstract

A method and system for evaluating the black start capability of networked new energy and energy storage devices. According to the real-time operation state of the power grid, the fault nodes of the power grid are detected; and the topology of the power grid is reconstructed according to the fault nodes. In the reconstructed power grid topology, the recovery priority score of the load node is calculated, and the real-time output capacity of each new energy and energy storage device is calculated; the reconstructed power grid topology is converted into a weighted graph, and the edge weight and node weight of the weighted graph are corrected according to the recovery priority, output capacity and path cost of the intermediate node. Based on the corrected weighted graph, the black start path is searched and deployed in the simulation environment for simulation; the operation data of each new energy and energy storage device in each simulation process is recorded, and the black start capability of each device is evaluated. The application can accurately reflect the actual connectivity and operation constraints when the power grid is powered off, and provide more realistic operation conditions for the black start capability evaluation of each new energy and energy storage device.

Description

Technical Field

[0001] This invention belongs to the field of new power systems and new energy technology, specifically relating to a method and system for evaluating the black start capability of grid-connected new energy and energy storage devices. Background Technology

[0002] In the absence of external power sources, renewable energy devices (such as wind or solar power) or energy storage devices can start up and provide initial power to support the startup of critical infrastructure in the power grid. The black-start capability of renewable energy or energy storage devices ensures that power supply can be restored even without traditional power sources, ensuring emergency recovery and stable operation of the power grid.

[0003] Patent application CN118487269A establishes and selects key evaluation indicators, builds a black-start decision model, simplifies the input-output relationship, and finally ranks black-start schemes using weighted evaluation and fuzzy evaluation methods to select the optimal scheme. This assesses the black-start capability of energy storage systems after large-scale power outages, ensuring rapid grid restoration. Patent application CN114862150A proposes a black-start capability assessment method for distribution networks based on distributed generation. It utilizes an entropy-weighted fuzzy comprehensive evaluation model, acquires and standardizes evaluation indicators, calculates subjective and objective weights, and combines fuzzy evaluation methods to ultimately derive the black-start capability assessment value of distributed generation. Patent application CN115906615A analyzes key technical factors for the initial black-start of new energy sources, defines the spatiotemporal support capability for black-start, and combines LSTM neural networks to model time-series data to assess the black-start capability of new energy systems. This method considers the spatiotemporal volatility of new energy units, optimizes black-start schemes, and improves grid recovery efficiency.

[0004] The aforementioned studies have proposed different technical approaches and evaluation methods for assessing the black-start capability of combined renewable energy and energy storage power generation systems. However, some shortcomings remain for grid-connected renewable energy and energy storage devices. The methods primarily focus on the individual roles of renewable energy and energy storage devices in black-start, but the complexity of the power grid structure is also a significant factor influencing black-start capability. Changes in grid topology, coordination among different types of distributed energy sources, and load variations and distribution also affect the black-start process. Therefore, the black-start assessment of grid-connected renewable energy needs to consider the interconnectivity and collaborative operation between different power sources to achieve a more comprehensive evaluation of black-start capability. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this application provides a method and system for evaluating the black-start capability of grid-connected new energy and energy storage devices, dynamically and accurately assessing the black-start capability of each device. The method first performs fault detection on the power grid and reconstructs the grid topology. Then, based on the reconstructed topology, it calculates the recovery priority of load nodes and the real-time output capability of the new energy / energy storage devices, combining these with the path costs of intermediate nodes to form a weighted graph, and searches for the optimal black-start path based on this graph. Finally, it performs a black-start operation in a simulation environment, records device operating data, and calculates indicators such as start-up time, available power, voltage / frequency support capability, and impact intensity to generate a comprehensive score.

[0006] The first aspect of this application discloses a method for evaluating the black-start capability of grid-type new energy and energy storage devices, the method employing the following technical solution:

[0007] Based on the real-time operating status of the power grid, a two-layer fault judgment method is adopted during different load periods to detect faulty nodes in the power grid; and the connection relationship of power grid nodes is updated according to the faulty nodes to obtain a reconstructed power grid topology.

[0008] In reconstructing the power grid topology, the recovery priority score of each load node is calculated, and the real-time output capacity of each new energy source and energy storage device is calculated.

[0009] The reconstructed power grid topology is transformed into a weighted graph to restore priority, output capacity, and path cost of intermediate nodes, and the edge weights and node weights of the weighted graph are corrected; based on the corrected weighted graph, the optimal black start path is searched.

[0010] Deploy the black-start path in the simulation environment to perform black-start simulation; record the operating data of each new energy source and energy storage device during each simulation; evaluate the black-start capability of a single new energy source / energy storage device based on the operating data.

[0011] Furthermore, the dual-layer fault determination method includes:

[0012] For each power grid node, preliminary detection of operating parameters is performed during different load periods; when the preliminary detection results exceed the trigger threshold, fault detection is triggered.

[0013] For a power grid node that triggers fault detection, calculate the instantaneous change value of the operating parameters; if the instantaneous change value exceeds the fault detection threshold, the power grid node is determined to be a fault node.

[0014] The fault detection threshold is the sum of the upper limit of the trigger threshold and the offset. The offset is the product of the standard deviation of the operating parameters during the load period corresponding to the fault detection and the offset weight. The offset weight is the ratio of the instantaneous change value of the parameter to the interval used.

[0015] Furthermore, the step of obtaining the reconstructed power grid topology map includes:

[0016] Obtain the location, equipment type, connection lines, and line attributes of all nodes in the power grid, and construct the power grid topology using an undirected graph structure;

[0017] The edge weights of the power grid topology diagram are the weighted sum of multiple factors, including line impedance, line capacity, line length, voltage level, and restoration cost; wherein, the line capacity and voltage level are taken as reciprocals in the weighted calculation.

[0018] A fault status marker is set for the faulty node, and the edge connected to the faulty node is marked as disconnected. In the power grid topology diagram, the topology connection relationship is updated according to the fault status marker and disconnection marker to obtain the reconstructed power grid topology diagram.

[0019] Furthermore, the recovery priority of the load node is calculated as a weighted sum of the recovery urgency, power demand level, and recovery difficulty of the load node;

[0020] The urgency of recovery is assigned according to load type; the power demand level is expressed as the ratio of the maximum power demand of a load node to the maximum single-node power demand among all load nodes.

[0021] The recovery difficulty is the baseline value of 10 minus the weighted sum of the three deduction items.

[0022] The deduction items include equipment status score and available backup power capacity of load nodes; the recovery time is the ratio of the estimated recovery time of the load node to the maximum tolerable outage time.

[0023] Further, the real-time output capabilities of new energy sources and energy storage devices in the reconstructed power grid topology are calculated; including:

[0024] For wind power devices in new energy sources, an instantaneous available power is calculated using a wind speed-power conversion model based on power curves and meteorological parameters, and grid operation state constraints are introduced for real-time correction.

[0025] For photovoltaic devices in new energy sources, the theoretical output is calculated based on the light intensity and then corrected according to the module temperature.

[0026] For energy storage devices, the actual discharge power that the energy storage system can provide in real time is calculated based on the real-time state of charge, rated capacity and maximum charge and discharge power, and is expressed as the minimum power constraint after discharge efficiency correction.

[0027] The minimum power constraint is the minimum between the maximum allowable discharge power and the available power determined by the state of charge. The available power determined by the state of charge is calculated as the product of the rated capacity and a fraction. The numerator of the fraction is the current available capacity, and the denominator is the discharge time window.

[0028] Furthermore, the discharge time window is dynamically corrected by combining load forecasting and safety constraints; the corrected discharge time window is a weighted sum of the theoretical maximum discharge time window, the energy-load ratio, and the temperature safety function.

[0029] The energy load ratio is the ratio of the available energy of the energy storage device to the predicted load.

[0030] Furthermore, the weighted graph includes load nodes, power generation nodes, and intermediate nodes;

[0031] The weight of a load node is represented as the reciprocal of its recovery priority; the weight of a generator node is represented as the reciprocal of its real-time output capability; and the weight of an intermediate node is represented as the reciprocal of its path cost.

[0032] For the edges between load nodes and power generation nodes, the edge weights are adjusted by the comprehensive value of the endpoints; based on the node weights in the weighted graph and the adjusted edge weights, the shortest path is searched as the optimal black start path.

[0033] Furthermore, the step of correcting the edge weights through the endpoint comprehensive value includes:

[0034] For the edge between the load node and the power generation node, the endpoint comprehensive value is calculated as the product of the node balancing term and the path cost suppression term.

[0035] The node balancing term is a weighted sum of the normalized value of recovery priority and the normalized value of real-time output capability;

[0036] The path cost suppression term is 1 minus the path cost normalization value;

[0037] The corrected edge weight is the ratio of the original edge weight in the power grid topology diagram to the comprehensive value of the endpoint; during the calculation, the comprehensive value of the endpoint is adjusted by the influence intensity.

[0038] Furthermore, during multiple simulated black start processes, the operating data of new energy and energy storage devices are recorded in real time; for each new energy and energy storage device, individual evaluation indicators are calculated, including average start-up time, average available start-up power, voltage and frequency support capability, and influence intensity coefficient.

[0039] Based on the individual evaluation indicators and their reference values, calculate the comprehensive score of the black start capability of each new energy / energy storage device under the optimal black start path.

[0040] The second aspect of this application discloses a black-start capability assessment system for grid-type new energy and energy storage devices, which operates the black-start capability assessment method as described in the first aspect of this application. The system includes:

[0041] The topology reconfiguration module is used to detect faulty nodes in the power grid based on the real-time operating status of the power grid and the two-layer fault judgment method according to the load period; and to update the connection relationship of power grid nodes according to the faulty nodes to obtain the reconfigured power grid topology map.

[0042] Recovery capability assessment module; used to calculate the recovery priority score of each load node and the real-time output capability of each new energy source and energy storage device in the reconfiguration of the power grid topology;

[0043] The black start path search module is used to convert the reconstructed power grid topology into a weighted graph to restore priority, output capacity, and path cost of intermediate nodes, and to correct the edge weights and node weights of the weighted graph; based on the corrected weighted graph, it searches for the optimal black start path.

[0044] The black-start capability assessment module is used to deploy the black-start path in the simulation environment to perform black-start simulation; record the operating data of each new energy source and energy storage device during each simulation; and evaluate the black-start capability of a single new energy source / energy storage device based on the operating data.

[0045] Compared with the prior art, the technical solution proposed in this application has at least one of the following beneficial effects:

[0046] 1. This application acquires the real-time operating status of the power grid, performs two-level fault judgment by load period, promptly identifies faulty nodes, and updates the connection relationships of power grid nodes based on the faulty nodes, thereby generating a reconstructed power grid topology. This dynamic topology accurately reflects the actual connectivity and operational constraints of the power grid under different fault conditions, providing a more realistic black-start capability assessment for each new energy and energy storage device.

[0047] 2. Based on the reconstructed power grid topology, this application integrates the load node recovery priority, the real-time output capacity of the power generation node, and the path cost of intermediate nodes into a weighted graph. By searching for the optimal black start path, critical loads are prioritized for recovery, and the output potential of each device is fully utilized. This path optimization not only ensures the feasibility and stability of the black start process but also enables a more accurate assessment of the start-up capability and power support performance of each new energy or energy storage device under actual power grid conditions.

[0048] 3. In multiple black-start simulations, this application records in real time the start-up time, available power, voltage and frequency support capabilities, and impact intensity on the system of each new energy source and energy storage device, and comprehensively scores the black-start capability of a single new energy source or energy storage device under the optimal black-start path. By incorporating the reconstructed grid topology and path optimization results into the evaluation, end-to-end quantitative black-start capability can be obtained, providing a scientific basis for grid operation management and maintenance decisions, and also providing an operable reference for black-start scheme optimization. Attached Figure Description

[0049] Figure 1 A flowchart illustrating the black-start capability assessment method for grid-type new energy and energy storage devices. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0051] As an embodiment of this application, a specific implementation method for evaluating the black-start capability of grid-type new energy and energy storage devices is disclosed. (Refer to...) Figure 1 , Figure 1 A flowchart illustrating the black-start capability assessment method for grid-type new energy and energy storage devices.

[0052] Step 1: Obtain the real-time operating status of the power grid, determine the location and status of the power grid fault, and reconstruct the power grid topology based on the fault area.

[0053] 1.1: Deploy edge devices at each power grid node to monitor the operating status of power grid equipment and collect operating data from the power grid node;

[0054] Edge devices are deployed at each grid node to collect parameter data at a frequency of 1 second, including current, voltage, power, and frequency under different loads. The grid nodes include power generation nodes (new energy and energy storage devices), intermediate nodes (substations and distribution stations), and load nodes in the grid.

[0055] Based on real-time collected voltage and current data, the system determines whether abnormal conditions such as voltage or current drops or undervoltage have occurred. When the voltage or current exceeds a corresponding threshold, fault detection is triggered. The threshold settings for current and voltage are as follows:

[0056] The collected historical parameter data are classified according to load periods, including low load periods (below 40% of rated power), medium load periods (40%-60% of rated power), medium-high load periods (60%-80% of rated power), and high load periods (above 80% of rated power).

[0057] On the edge device, statistical values ​​of the compute node parameters, including mean and standard deviation, are generated for each load period. For each load period, fault detection trigger thresholds for current, voltage, and frequency are also defined. for , This represents the mean. Indicates standard deviation; The multiplier is set according to the requirements of the load period. For example, during low-load periods, because the load is lower and the power grid stability is stronger, It can be set to 0.5; during periods of high load, the power grid load fluctuates significantly. It can be set to 2 or 3.

[0058] When the real-time data of the power grid exceeds the fault trigger threshold The upper and lower limits are immediately triggered to activate fault detection, which includes voltage surges, current surges, and frequency anomalies.

[0059] As an optional approach in this embodiment, for a grid node that triggers a fault warning, the instantaneous changes in voltage and current are calculated. The instantaneous change value is the difference between the current voltage / current value and the voltage / current value one sampling interval ago.

[0060] ;

[0061] in, and These are the instantaneous changes in voltage and current, respectively. and These are the current voltage value and the current current value, respectively. The sampling interval is denoted as .

[0062] When the instantaneous change value exceeds the fault detection threshold If this occurs, a fault is confirmed at a power grid node. Fault detection threshold. ; Fault detection trigger threshold The upper limit; The offset is set as the product of the standard deviation of current and voltage for the corresponding load period and the offset weight; the offset weight is the ratio of the instantaneous change in voltage and current to the sampling interval.

[0063] In this embodiment, This is a preliminary fault detection threshold with relatively lenient standards, allowing for a rapid response to sudden changes in current and voltage. However, because the power grid may experience brief fluctuations or disturbances under certain circumstances (such as equipment switching or power fluctuations), once the parameter value exceeds... This does not necessarily mean that a real fault has occurred. To ensure that grid node fault detection does not trigger unnecessary fault responses due to short-term fluctuations or errors, an offset has been added in this implementation. By setting stricter standards for fault confirmation, false alarms can be reduced.

[0064] Once the edge device completes fault detection, it can confirm that a specific power grid node where the edge device is located has failed, thus confirming the location of the power grid fault.

[0065] 1.2: Construct a power grid topology diagram and reconstruct the power grid topology based on node failure conditions.

[0066] 1.2.1; Obtain the location coordinates, equipment types, connection lines, and attributes (such as impedance, voltage level, etc.) of all nodes in the power grid based on the GIS system.

[0067] An undirected graph structure is used to construct the power grid topology. Nodes in the graph represent power equipment (such as substations, distribution stations, new energy devices, energy storage devices, and load nodes), and edges represent connecting lines. Each edge is assigned a weight, which is influenced by parameters such as line impedance, line capacity, line length, voltage level, and recovery cost. The smaller the edge weight, the higher the priority of the line selection.

[0068] For nodes and The edge weights of the two Represented as:

[0069] ;

[0070] in, This represents the impedance magnitude of the line. For line capacity, For line length, For voltage level, This indicates the cost of restoring the line (related to repair costs and repair time). , , , and In this embodiment, the objectives of power grid topology reconfiguration are adjusted to assign weights to various parameters. For example, when rapid recovery is the priority objective, the weights of recovery cost and path length are increased, and low-cost, easily recoverable lines are selected for connection.

[0071] 1.2.2; For the faulty node identified in step 1, its position in the topology is retained in the power grid topology diagram, but a fault status mark is set to indicate that the node is currently unavailable; in addition, the status of the edges connected to the faulty node is also updated and marked as disconnected; the topology is updated.

[0072] As an optional step in this embodiment, based on the updated power grid topology, a steady-state simulation is performed using the AC power flow calculation method to verify that the updated power grid topology can operate normally.

[0073] Step 2: Assess the recovery priority of load nodes and the output capacity of new energy and energy storage devices.

[0074] 2.1: Based on the recovery needs of load nodes in the power grid, different load nodes are assigned priorities.

[0075] For each load node in the updated power grid topology, a priority score is calculated based on its recovery urgency, power demand level, and recovery ease, as follows:

[0076] ;

[0077] In the formula, For load nodes Priority score, , and They are respectively load nodes The urgency of recovery, power requirement level, and ease of recovery; , and The weighting coefficients for urgency, power demand level, and ease of recovery can be set and adjusted according to the actual situation.

[0078] Furthermore, the recovery urgency is graded and scored according to the load type and the importance of the service object. An example is given in Table 1 of this embodiment.

[0079] Table 1. Examples of Emergency Score for Recovery

[0080]

[0081] The power demand level is represented by the proportion of the node's maximum load power, and the calculation method is as follows:

[0082] ;

[0083] In the formula, For load nodes Maximum power demand, This represents the maximum single-node power demand among all load nodes.

[0084] The ease of recovery is determined by considering the recovery time, equipment status, and backup power availability; the calculation method is as follows:

[0085] ;

[0086] In the formula, The standardized score for the time required for recovery is the ratio of the estimated recovery time of a load node to the longest tolerable outage time. The equipment status score is obtained based on the diagnostic results of the load node, with a value of [0.10]. The worse the equipment status, the higher the score. The availability of backup power is quantified by the available backup power capacity of the load node. If there is no available backup power nearby, the availability of backup power is 0. , and They are respectively , and The weighting coefficients can be set and adjusted according to actual conditions.

[0087] 2.2: Calculate the real-time output capacity of new energy sources and energy storage devices in the updated power grid topology.

[0088] For wind turbines, photovoltaic arrays, and energy storage systems, real-time data on operating status and environmental conditions are collected and preprocessed. Table 2 provides some examples of the collected data.

[0089] Table 2 Examples of collected data

[0090]

[0091] The maximum available output capacity of each wind and solar node is calculated by combining physical and empirical models. Specifically:

[0092] 2.2.1: For wind power, an instantaneous available power is calculated using a wind speed-power conversion model based on power curves and meteorological parameter corrections, with dynamic constraints introduced for real-time correction. This is represented as:

[0093] ;

[0094] in, This represents the real-time instantaneous available power of wind power, indicating the maximum active power that the wind turbine can output under the current weather conditions. This is a function for the wind turbine power curve, provided by the wind turbine manufacturer. The function output is the wind speed. Rated power ratio below; This is the air density correction factor. , and These are the current air density and the standard air density, respectively. This is the wind direction deviation correction factor. , This is the angle between the current wind direction and the direction of the wind turbine's main shaft.

[0095] The dynamic constraint correction coefficient is the minimum constraint formed by multiple sub-constraints of the power grid operating state; the dynamic constraint correction coefficient is expressed as:

[0096] ;

[0097] in, , , , and These are voltage constraints, line capacity constraints, frequency stability constraints, fault isolation constraints, and energy storage coordination constraints.

[0098] It should be noted that the selection of sub-constraints can be adaptively adjusted according to the actual situation of the power grid.

[0099] 2.2.2: For photovoltaics, according to The theoretical output is calculated and then corrected based on the component temperature. Real-time light intensity (unit: W / m2). For photovoltaic module efficiency, The total area of ​​the photovoltaic modules; the method for correcting the theoretical output based on the module temperature is as follows:

[0100] ;

[0101] in, For the corrected photovoltaic output, For temperature coefficient, and These are the reference temperature and the current actual component temperature, respectively.

[0102] 2.2.3: For energy storage systems, the actual discharge power that the energy storage system can provide in real time is calculated based on the real-time state of charge, rated capacity, and maximum charge / discharge power; expressed as:

[0103] ;

[0104] in, For real-time actual discharge power, The rated maximum discharge power of the energy storage system; and The current state of charge and the minimum permissible state of charge threshold; For the discharge time window, and This refers to the capacity and discharge efficiency of the energy storage device.

[0105] Traditionally, discharge time windows are set by the manufacturer or preset to a fixed duration. However, during the black start phase of a power outage, energy storage systems face challenges such as large load fluctuations, complex operating environments, and variable equipment states. A fixed discharge time window may be insufficient to meet continuous power supply needs when load demand surges or energy storage status declines. Therefore, this embodiment dynamically adjusts the discharge time window based on energy storage status, load demand, and grid conditions to avoid resource waste or power shortages caused by a fixed time window. The specific steps are as follows:

[0106] At the moment of power restoration, the real-time state of charge (SOC), state of health (SOH), and battery temperature of the energy storage device are acquired. and ambient temperature ;

[0107] Using historical load and environmental parameters, a time-series model is used to predict load demand at the time of power restoration; assuming the power restoration time... Load demand at that moment According to the previous Historical load data at a given time point is used as the basis for prediction; this process is expressed as:

[0108] ;

[0109] in, The time series prediction model selected for this embodiment.

[0110] As an optional approach in this embodiment, the time series prediction model can be selected from gated recurrent units or other neural network prediction methods.

[0111] Set minimum safe SOC threshold , The SOC safety lower limit; and the temperature safety threshold are... , and These are the lower and upper limits of temperature safety, respectively, which are set according to the battery material.

[0112] Assuming the maximum discharge power of the energy storage device is The possible dosage is The theoretical maximum discharge time window is ;in, , This refers to the rated capacity of the energy storage device.

[0113] Combining load forecasting and safety constraints, a weighted fusion dynamic correction of the discharge time window is adopted; expressed as:

[0114] ;

[0115] in, This is the corrected discharge time window. As a temperature safety function, the discharge time is reduced when the temperature approaches the threshold. , and For the weighting coefficients, satisfying Adjustments should be made based on the phased needs of the black start phase. For example, if the load power fluctuates drastically in the initial stage of black start, it is necessary to increase... The value of .

[0116] 3. Based on the load node recovery priority and the real-time output capacity of new energy and energy storage devices, the power grid topology is converted into a weighted graph, and the edges and nodes in the weighted graph are corrected.

[0117] 3.1: In a weighted graph, nodes include load nodes. Power generation nodes (New energy, energy storage) and intermediate nodes (Substations, distribution stations);

[0118] For load nodes, based on recovery priority Assign weights to it , represented as ; To prevent small constants from being divided by zero.

[0119] For power generation nodes, based on real-time output capacity Assign weights to it , represented as ;in, This includes the corrected instantaneous available wind power, the corrected photovoltaic output, and the actual discharge power that the energy storage system can provide.

[0120] For intermediate nodes, based on path cost Assign weights to it , represented as Among them, path cost This is a comprehensive calculation of the costs of main transformer input, time, voltage over-limit, and power shortage at intermediate nodes.

[0121] 3.2: Ensure the edges in the weighted graph are consistent with the edges in the power grid topology graph; introduce node weight corrections to the edge weights of the power grid topology graph. The specific implementation method is as follows:

[0122] The recovery priority of load nodes, the real-time output capability of power generation nodes, and the path cost of intermediate nodes are all normalized to [0,1]. The comprehensive value of the endpoints is calculated using a correction function and expressed as:

[0123] ;in, For load nodes and power generation node The combined value of the endpoints between them For load nodes The normalized value of the recovery priority. For power generation nodes The normalized value of the real-time output capability. It is a regulating factor used to balance "supply capacity" and "load value".

[0124] The edge weight representation of the weighted graph is then modified as follows: ;in, For load nodes and power generation node Corrected edge weights between; To at the load node and power generation node Original edge weights in a power grid topology graph; To affect the intensity, This is used to measure the sensitivity of node characteristics to edge weight adjustments.

[0125] 3.3: Based on the node weights and corrected edge weights in the weighted graph, the path with the minimum recovery cost is searched using the shortest path search algorithm, and this path is taken as the optimal black start path.

[0126] As an optional approach in this step, you can choose a shortest path search algorithm such as Dijkstra's shortest path algorithm to find the optimal black start path based on the weighted graph.

[0127] In a typical power restoration scenario, the factors influencing edge weights in a power grid topology include line impedance, line capacity, line length, voltage level, and restoration cost. However, in a black start scenario in a distribution network, path selection depends not only on line conditions but also on the value of the path endpoints (i.e., load nodes and power generation nodes). For example, high-priority loads deserve to be restored as quickly as possible, and nodes with greater available power can drive the restoration of more loads.

[0128] This embodiment introduces node weight correction in the edge weight definition of the weighted graph, which reduces the path search cost of high-value nodes and gives priority to paths to high-priority loads or high-output power sources during the search, that is, pursuing the "cost-effectiveness" of black start recovery.

[0129] 4. Based on the optimal black start path, execute the black start strategy in the simulation environment, evaluate the black start capability of new energy and energy storage devices based on the recovery status.

[0130] As one implementation method of this step, the start-up and shutdown sequence of new energy sources (wind power, photovoltaic) and energy storage devices is determined based on the optimal black start path. The power thresholds and start-up conditions required for each device are clearly defined to ensure that the start-up process meets the requirements for power system safety and stability.

[0131] As one implementation method for this step, multiple simulations are performed. Start-stop commands are sent through the control system to adjust the output power of each device, gradually restoring the grid load until stability is achieved. During the black start process, the start-up time, ramp-up process, and synchronization characteristics of the new energy and energy storage devices are monitored in real time.

[0132] 4.1: During multiple simulated black start processes, the operating data of each new energy and energy storage device is recorded in real time through edge devices, including but not limited to the actual output power curve, voltage offset, frequency offset, power margin, and start-up success status (i.e., whether it is successfully connected to the grid).

[0133] After multiple simulations, for each new energy and energy storage device, individual evaluation indicators are calculated, including average start-up time, average available start-up power, voltage and frequency support capability, and influence intensity coefficient.

[0134] Furthermore, the average startup time is ; For the first Startup time of the simulation ; This indicates the time point at which the new energy or energy storage device receives the start-up command. This indicates the point in time when the output power reaches a stable grid-connected level. The number of simulations.

[0135] Furthermore, the average starting power that can be provided is ; For the first The simulation can provide startup power. .

[0136] Furthermore, the voltage and frequency support capabilities are ;in, For voltage support capability, , The average value after summing the maximum voltage deviations of all single simulations; For frequency support capability, , The average value after summing the maximum frequency deviations of all single simulations; and The weighting coefficients for voltage support capability and frequency support capability satisfy... .

[0137] Furthermore, the influence intensity coefficient is i.e., power generation node The mean value of the strength coefficient is affected in multiple simulations; the first The influence intensity coefficient of the second simulation is ;

[0138] in, This represents the power change during the black start process. ; For nodes In the The power received during the second simulation of the start command. They are nodes In the The simulation stabilized the output power; In the optimal recovery path, excluding nodes In addition, the sum of power changes of other new energy / energy storage nodes.

[0139] As an optional approach in this step, the average startup success rate of new energy / energy storage nodes across all simulations can also be calculated, which is the ratio of the number of successful grid connection attempts to the total simulation coefficients.

[0140] 4.2: Based on the individual evaluation indicators obtained in step 4.1, calculate the comprehensive score of the black start capability for each new energy / energy storage device, expressed as:

[0141] ;

[0142] in, For nodes Overall score for black start capability; , , and These serve as reference values ​​for each individual indicator, designed according to the system objectives. , , and Let be the weighting coefficients of each term in the formula, satisfying .

[0143] 4.3: Based on the comprehensive score of the black start capability of each new energy / energy storage device, it is classified into different levels. Table 3 shows an example of level classification based on thresholds.

[0144] Table 3. Examples of Grade Classification

[0145]

[0146] As an optional embodiment of this application, the comprehensive scoring level provides a quantitative evaluation of the black start capability of each new energy / energy storage device. It can quickly identify high-capacity, low-impact key devices as starting or main nodes in the next black start path selection, while assisting in determining the auxiliary positions of medium-level devices, avoiding low-level bottleneck devices from undertaking key tasks, thereby optimizing the path sequence, improving overall reliability, reducing the search space, and accelerating path planning decisions.

[0147] This application identifies the optimal black-start path based on the changes in grid topology during power outages and the differences in demand at load nodes. The optimal black-start path verifies the startup capability of each renewable energy or energy storage device under actual system coordination conditions. It not only examines the device's average startup time, maximum available startup power, and voltage and frequency support capabilities, but also reveals the mutual influence and power coupling effects between devices, identifying potential bottlenecks and critical nodes in the path. This makes the black-start capability assessment more comprehensive, reliable, and closely reflects actual operation. Furthermore, the assessment results provide a scientific basis for device classification, startup sequence optimization, auxiliary energy storage configuration, and system-level black-start strategies, offering a reference for developing highly reliable black-start schemes and improving the rapid recovery capabilities of microgrids or regional power grids.

[0148] As an embodiment of this application, a black-start capability assessment system for grid-type new energy and energy storage devices is disclosed, along with a specific implementation of an embodiment of the black-start capability assessment method. The system includes:

[0149] The topology reconfiguration module is used to detect faulty nodes in the power grid based on the real-time operating status of the power grid and the two-layer fault judgment method according to the load period; and to update the connection relationship of power grid nodes according to the faulty nodes to obtain the reconfigured power grid topology map.

[0150] Recovery capability assessment module; used to calculate the recovery priority score of each load node and the real-time output capability of each new energy source and energy storage device in the reconfiguration of the power grid topology;

[0151] The black start path search module is used to convert the reconstructed power grid topology into a weighted graph to restore priority, output capacity, and path cost of intermediate nodes, and to correct the edge weights and node weights of the weighted graph; based on the corrected weighted graph, it searches for the optimal black start path.

[0152] The black-start capability assessment module is used to deploy the black-start path in the simulation environment to perform black-start simulation; record the operating data of each new energy source and energy storage device during each simulation; and evaluate the black-start capability of a single new energy source / energy storage device based on the operating data.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for evaluating the black-start capability of grid-type new energy and energy storage devices, characterized in that, include: Based on the real-time operating status of the power grid, a two-layer fault judgment method is adopted during different load periods to detect faulty nodes in the power grid; and the connection relationship of power grid nodes is updated according to the faulty nodes to obtain a reconstructed power grid topology. The dual-layer fault detection method includes: For each power grid node, preliminary detection of operating parameters is performed during different load periods; when the preliminary detection results exceed the trigger threshold, fault detection is triggered. For a power grid node that triggers fault detection, calculate the instantaneous change value of the operating parameters; if the instantaneous change value exceeds the fault detection threshold, the power grid node is determined to be a fault node. The fault detection threshold is the sum of the upper limit of the trigger threshold and the offset. The offset is the product of the standard deviation of the operating parameters during the load period corresponding to the fault detection and the offset weight. The offset weight is the ratio of the instantaneous change value of the parameter to the interval used. The steps for obtaining the reconstructed power grid topology map include: Obtain the location, equipment type, connection lines, and line attributes of all nodes in the power grid, and construct the power grid topology using an undirected graph structure; The edge weights of the power grid topology diagram are the weighted sum of multiple factors, including line impedance, line capacity, line length, voltage level, and restoration cost; wherein, the line capacity and voltage level are taken as reciprocals in the weighted calculation. A fault status marker is set for the faulty node, and the edge connected to the faulty node is marked as disconnected; in the power grid topology diagram, the topology connection relationship is updated according to the fault status marker and disconnection marker to obtain the reconstructed power grid topology diagram; In reconstructing the power grid topology, the recovery priority score of each load node is calculated, and the real-time output capacity of each new energy source and energy storage device is calculated. The reconstructed power grid topology is transformed into a weighted graph to restore priority, output capacity, and path cost of intermediate nodes, and the edge weights and node weights of the weighted graph are corrected; based on the corrected weighted graph, the optimal black start path is searched. The weighted graph includes load nodes, power generation nodes, and intermediate nodes; The weight of a load node is represented as the reciprocal of its recovery priority; the weight of a generator node is represented as the reciprocal of its real-time output capability; and the weight of an intermediate node is represented as the reciprocal of its path cost. For the edges between load nodes and power generation nodes, the edge weights are adjusted by the comprehensive value of the endpoints; based on the node weights in the weighted graph and the adjusted edge weights, the shortest path is searched as the optimal black start path. Deploy the black-start path in the simulation environment to perform black-start simulation; record the operating data of each new energy source and energy storage device during each simulation; evaluate the black-start capability of a single new energy source / energy storage device based on the operating data; During multiple simulated black start processes, the operation data of new energy and energy storage devices are recorded in real time; for each new energy and energy storage device, individual evaluation indicators are calculated, including average start-up time, average available start-up power, voltage and frequency support capability, and influence intensity coefficient. Based on the individual evaluation indicators and their reference values, calculate the comprehensive score of the black start capability of each new energy / energy storage device under the optimal black start path.

2. The method for evaluating the black-start capability of grid-type new energy and energy storage devices according to claim 1, characterized in that, The recovery priority of the load node is calculated as a weighted sum of the recovery urgency, power demand level, and recovery difficulty of the load node. The urgency of recovery is assigned according to load type; the power demand level is expressed as the ratio of the maximum power demand of a load node to the maximum single-node power demand among all load nodes. The recovery difficulty is the baseline value of 10 minus the weighted sum of the three deduction items. The deductions include the recovery time, equipment status score, and available backup power capacity of the load node; the recovery time is the ratio of the estimated recovery time of the load node to the maximum tolerable outage time.

3. The method for evaluating the black-start capability of grid-type new energy and energy storage devices according to claim 1, characterized in that, Calculate the real-time output capacity of new energy sources and energy storage devices in the reconstructed power grid topology; including: For wind power devices in new energy sources, an instantaneous available power is calculated using a wind speed-power conversion model based on power curves and meteorological parameters, and grid operation state constraints are introduced for real-time correction. For photovoltaic devices in new energy sources, the theoretical output is calculated based on the light intensity and then corrected according to the module temperature. For energy storage devices, the actual discharge power that the energy storage system can provide in real time is calculated based on the real-time state of charge, rated capacity and maximum charge and discharge power, and is expressed as the minimum power constraint after discharge efficiency correction. The minimum power constraint is the minimum between the maximum allowable discharge power and the available power determined by the state of charge. The available power determined by the state of charge is calculated as the product of the rated capacity and a fraction. The numerator of the fraction is the current available capacity, and the denominator is the discharge time window.

4. The method for evaluating the black-start capability of grid-type new energy and energy storage devices according to claim 3, characterized in that, The discharge time window is dynamically corrected by combining load forecasting and safety constraints; the corrected discharge time window is a weighted sum of the theoretical maximum discharge time window, energy load ratio, and temperature safety function. The energy load ratio is the ratio of the available energy of the energy storage device to the predicted load.

5. The method for evaluating the black-start capability of grid-type new energy and energy storage devices according to claim 1, characterized in that, The step of correcting edge weights based on endpoint comprehensive value includes: For the edge between the load node and the power generation node, the endpoint comprehensive value is calculated as the product of the node balancing term and the path cost suppression term. The node balancing term is a weighted sum of the normalized value of recovery priority and the normalized value of real-time output capability; The path cost suppression term is 1 minus the path cost normalization value; The corrected edge weight is the ratio of the original edge weight in the power grid topology diagram to the comprehensive value of the endpoint; during the calculation, the comprehensive value of the endpoint is adjusted by the influence intensity.

6. A black-start capability assessment system for grid-type new energy and energy storage devices, operating the black-start capability assessment method as described in any one of claims 1-5, characterized in that, The system includes: The topology reconfiguration module is used to detect faulty nodes in the power grid based on the real-time operating status of the power grid and the two-layer fault judgment method according to the load period; and to update the connection relationship of power grid nodes according to the faulty nodes to obtain the reconfigured power grid topology map. Recovery capability assessment module; used to calculate the recovery priority score of each load node and the real-time output capability of each new energy source and energy storage device in the reconfiguration of the power grid topology; The black start path search module is used to convert the reconstructed power grid topology into a weighted graph to restore priority, output capacity, and path cost of intermediate nodes, and to correct the edge weights and node weights of the weighted graph; based on the corrected weighted graph, it searches for the optimal black start path. The black-start capability assessment module is used to deploy the black-start path in the simulation environment to perform black-start simulation; record the operating data of each new energy source and energy storage device during each simulation; and evaluate the black-start capability of a single new energy source / energy storage device based on the operating data.

Images