Microgrid structural component condition testing and processing methods and systems

Abstract

This invention relates to the field of electrical measurement and testing technology, and discloses a method and system for processing condition testing of microgrid structural components, including: Step 1, generating an association and mapping matrix, collecting voltage and current, and calculating power, total energy, and energy storage changes; Step 2, determining the network consistency residual threshold; Step 3, summarizing inflows, outflows, and energy storage changes into residual vectors according to the matrix, and comparing them to obtain the network-wide consistency judgment result; Step 4, calculating the health level, and using upper and lower limits and the current value to determine the controllability margin; Step 5, constructing an attribution matrix to distribute the residuals, and linking the health level and controllability margin for correction; Step 6, combining the network-wide consistency and the corrected state quantities to obtain a classification judgment; Step 7, generating a verification queue and power and current limits according to the classification, and performing continuous time window closed-loop verification. This invention achieves unified standards, clear judgment, traceability, and executable testing of microgrid structural components.

Description

Technical Field

[0001] This invention belongs to the field of electrical measurement and testing technology, specifically relating to a method and system for testing and processing the condition of microgrid structural components. Background Technology

[0002] With the widespread integration of distributed power sources, energy storage devices, and power electronic interface equipment into microgrids, the structural complexity and operational characteristics of microgrids are continuously increasing. Microgrids contain various structural components such as generation units, energy storage units, converters, switches, and loads, and their operational status directly affects the energy balance and safe operation of the microgrid. In existing technologies, the assessment of the status of microgrid structural components largely relies on the voltage, current, or power measurements of individual components, or on comparing local parameters using empirical thresholds. These methods struggle to ensure consistency in test parameters under topology changes. Furthermore, existing methods for energy-level verification often remain at the level of simple power integration or node aggregation, failing to systematically consider the impact of uncertainties such as measurement errors and timing errors on energy consistency assessments, which can easily lead to false alarms or missed alarms.

[0003] Furthermore, existing technologies lack a clear distinction between network-wide anomalies and component anomalies, often making it difficult to determine whether an anomaly originates from network metering standards, topology relationships, or performance degradation of specific structural components. After an anomaly occurs, there is a lack of deterministic classification based on test results and subsequent verification, quota, and review mechanisms, resulting in a lack of unified data links and traceability between test results and handling actions. Summary of the Invention

[0004] This invention provides a method and system for testing the status of microgrid structural components, which solves the technical problems in related technologies such as inconsistent metering and summarization standards, failure to include voltage measurement errors, current measurement errors and time errors in energy consistency determination leading to false alarms and missed alarms, difficulty in distinguishing between network-side and component-side anomalies, and lack of a closed-loop system for handling and verifying based on test results.

[0005] This invention provides a method for testing and processing the condition of microgrid structural components, comprising the following steps:

[0006] Step 1: Generate an association matrix and a mapping matrix based on the primary system wiring method and the actual position of the switch; set a time window, collect voltage data and current data of each component, and obtain power data, total energy of each component and energy change of energy storage component;

[0007] Step 2: Determine the network consistency residual threshold based on voltage measurement error, current measurement error, and time error;

[0008] Step 3: Based on the correlation matrix and mapping matrix, compare the inflow energy, outflow energy and energy change of energy storage components to form a residual vector, and compare it with the network consistency residual threshold to obtain the network-wide energy consistency judgment result;

[0009] Step 4: Calculate the efficiency deviation, temperature rise deviation, and terminal pressure retention deviation based on the commissioning baseline and current metering data, and obtain the health status by weighting; obtain the controllability margin based on the voltage, temperature, upper and lower limits of energy storage capacity and the current value.

[0010] Step 5: Construct an attribution matrix based on the electrical connection relationship between nodes and components, distribute the residual vector into a residual distribution vector, and perform a linkage correction on the health and controllability margin to obtain the corrected health and controllability margin.

[0011] Step 6: Based on the results of the overall network energy consistency assessment, the corrected health level, and the corrected controllability margin, the classification determination is obtained;

[0012] Step 7: Generate the metering side verification queue, power limit, and current limit based on the classification judgment results, and conduct closed-loop verification according to the release conditions within a continuous time window.

[0013] This invention also provides a microgrid structural component condition testing and processing system, comprising:

[0014] The topology metering module is used to generate an association matrix and a mapping matrix based on the primary system wiring method and the actual position of the switch; it sets a time window, collects voltage and current data of each component, and obtains power data, total energy of each component and energy change of energy storage component.

[0015] The threshold generation module is used to determine the network consistency residual threshold based on voltage measurement error, current measurement error, and time error.

[0016] The consistency determination module is used to compare the inflow energy, outflow energy, and energy change of energy storage components with the correlation matrix and the mapping matrix to form a residual vector, and compare it with the network consistency residual threshold to obtain the energy consistency determination result of the entire network.

[0017] The health margin module is used to calculate efficiency deviation, temperature rise deviation, and terminal pressure retention deviation based on the commissioning baseline and current metering data, and then weight them to obtain the health level; and to obtain the controllability margin based on voltage, temperature, upper and lower limits of energy storage capacity and current values.

[0018] The attribution linkage module is used to construct an attribution matrix based on the electrical connection relationship between nodes and components, distribute the residual vector into a residual distribution vector, and perform linkage correction on health and controllability margin to obtain the corrected health and controllability margin.

[0019] The classification determination module is used to obtain the classification determination based on the network-wide energy consistency judgment result, the corrected health status, and the corrected controllability margin.

[0020] The closed-loop processing module is used to generate metering-side verification queues, power limits, and current limits based on the classification judgment results, and to perform closed-loop verification according to the release conditions within a continuous time window.

[0021] The beneficial effects of this invention are as follows: This invention constructs an association matrix and mapping matrix based on the primary system wiring method and the actual switch location, unifying metering and topology standards. Through deterministic calculations of voltage, current, power, and energy within a time window, and by transmitting voltage measurement errors, current measurement errors, and time errors to the energy layer, a network consistency residual threshold is established. Based on the node energy conservation residual and attribution matrix, traceable allocation of residuals to the component side is achieved, and health and controllability margins are adjusted accordingly. Using a rule combining network-wide consistency with component thresholds, normal operation, network-wide inconsistency, local degradation, and coupling anomalies are distinguished, further generating a metering-side verification queue and power and current limits, and performing closed-loop verification within a continuous time window. This comprehensively improves the clarity, comparability, and verifiability of microgrid structural component status testing, reduces false alarms and missed alarms, and facilitates the location of network and component responsibility boundaries. Attached Figure Description

[0022] Figure 1 This is a flowchart of the microgrid structure component condition testing and processing method of the present invention. Detailed Implementation

[0023] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0024] like Figure 1 As shown, the microgrid structural component condition testing and processing method includes the following steps:

[0025] Step 1: Generate an association matrix and a mapping matrix based on the primary system wiring method and the actual position of the switch; set a time window, collect voltage data and current data of each component, and obtain power data, total energy of each component and energy change of energy storage component;

[0026] Step 2: Determine the network consistency residual threshold based on voltage measurement error, current measurement error, and time error;

[0027] Step 3: Based on the correlation matrix and mapping matrix, compare the inflow energy, outflow energy and energy change of energy storage components to form a residual vector, and compare it with the network consistency residual threshold to obtain the network-wide energy consistency judgment result;

[0028] Step 4: Calculate the efficiency deviation, temperature rise deviation, and terminal pressure retention deviation based on the commissioning baseline and current metering data, and obtain the health status by weighting; obtain the controllability margin based on the voltage, temperature, upper and lower limits of energy storage capacity and the current value.

[0029] Step 5: Construct an attribution matrix based on the electrical connection relationship between nodes and components, distribute the residual vector into a residual distribution vector, and perform a linkage correction on the health and controllability margin to obtain the corrected health and controllability margin.

[0030] Step 6: Based on the results of the overall network energy consistency assessment, the corrected health level, and the corrected controllability margin, the classification determination is obtained;

[0031] Step 7: Generate the metering side verification queue, power limit, and current limit based on the classification judgment results, and conduct closed-loop verification according to the release conditions within a continuous time window.

[0032] In one embodiment of the present invention, an association matrix and a mapping matrix are generated based on the primary system wiring method and the actual position of the switch; a time window is set, and voltage data and current data of each component are collected to obtain power data, total energy of each component, and energy change of the energy storage component, including:

[0033] Step 11: Based on the primary system wiring method and the actual location of the switches, generate an association matrix with nodes as rows and branches as columns. The elements of the association matrix are the inflow, outflow, and unconnected status of the corresponding branches at the nodes. Nodes represent electrical equipotential connection points formed by conductors that are directly connected to each other and have the same potential. Generate a mapping matrix with components as rows and branches as columns. The elements of the mapping matrix indicate whether the corresponding component is connected to a branch. The primary system wiring method refers to the connection and separation relationship of primary power equipment at the physical and electrical level, which is determined by the port connection status of primary equipment such as buses, branches, switches, transformers, and converters. Components refer to microgrid structural components.

[0034] Step 12: Set the start and end times of the time window and the sampling interval, and establish an integer multiple relationship between the two to ensure that the sampling points are aligned with the time window boundary. Collect voltage and current data for each component in the order of sampling points, and form a sampling sequence in the order of sampling points.

[0035] Step 13: Within the same time window, multiply the voltage data and current data at each sampling point to obtain the power data; multiply the power data at each sampling point by the sampling interval and accumulate them in the order of sampling points to obtain the total energy of each component; read the current energy storage capacity of the energy storage component at the beginning and end of the time window respectively and subtract them to obtain the energy change of the energy storage component. The current energy storage capacity has the same energy dimension as the total energy of each component.

[0036] Through the above process, this embodiment generates an association matrix and a mapping matrix based on the primary system wiring method and the actual position of the switch, thereby achieving consistency between the metering caliber and the topology. By calculating power and energy at sampling points within a unified time window, it can ensure the dimensional alignment and traceability of energy statistics and changes in energy storage. Ultimately, it provides a stable and verifiable metering basis for subsequent network consistency determination and state quantity calculation, and reduces caliber ambiguity and statistical deviation.

[0037] In one embodiment of the present invention, step 1 further includes:

[0038] Step 21: Within the predetermined time window and sampling interval, compare the sampling timestamps of all sampling points within the time window one by one in the order of sampling points. If the adjacent time difference is not equal to the sampling interval, or the same timestamp is recorded repeatedly, or the time is reversed, then the sampling point is marked as an invalid sample; where the sampling timestamp refers to the time identifier that corresponds one-to-one with each sampling point; an invalid sample refers to a sampling point that does not meet the isochronous sampling condition.

[0039] Step 22: For locations marked as invalid samples, delete their corresponding voltage and current data, and then fill them using equal-duration copying: Use the voltage and current data of the immediately preceding valid sampling point, copying and filling them at sampling intervals to maintain the same number of sampling points within the time window. Simultaneously, generate a filling index, recording the sequence number and timestamp of the filled sampling point for subsequent traceability. Equal-duration copying and filling refers to replacing the current invalid point with a fixed value based on the measurement value of the previous valid point without changing the sampling time, thus maintaining the equal intervals and length of the sampling time series.

[0040] Step 23: Based on the filled voltage and current data, recalculate the power data, total energy of each component, and energy change of the energy storage component as in Step 13, and bind the filled index to the time window one by one.

[0041] Through the above process, this embodiment achieves uniform sampling standards for voltage and current data within the time window by performing consistency verification on the sampling timestamps; by performing equal-duration copying and filling on invalid samples, the sampling sequence length and time interval remain unchanged and a traceable filling index is formed; finally, under the unified standard, the power data, the total energy of each component, and the energy change of the energy storage component are recalculated, providing a stable and verifiable metrological basis for subsequent network energy consistency determination and state quantity calculation, and reducing statistical deviations introduced by sampling anomalies.

[0042] In one embodiment of the present invention, determining the network consistency residual threshold based on voltage measurement error, current measurement error, and time error includes:

[0043] Step 31: After determining the time window and sampling interval, set the voltage measurement error, current measurement error, and time error, and process them point by point within the same time window. The voltage measurement error and current measurement error are the upper bounds of the uncertainty or equivalent error of the metering device after calibration; the time error is the upper bound of the sampling time deviation introduced by the timing and triggering links. At each sampling point, multiply the voltage data by the voltage measurement error, multiply the current data by the current measurement error, and add the two products to obtain the power measurement error. This processing is used to uniformly transfer the uncertainties of voltage and current to the power level as input for subsequent energy errors.

[0044] Step 32: Within the same time window, multiply the power measurement error at each sampling point by the sampling interval point by point and accumulate them in the sampling order to obtain the accumulated power error value. Simultaneously, multiply the time error by the accumulated absolute value of the power data within the time window and add this to the accumulated power error value to obtain the energy measurement error of the total energy of each component. For energy storage components, take the measurement error of the current energy storage capacity at the beginning and end of the time window respectively and add them together as the energy measurement error of the energy change of the energy storage component. The measurement error of the current energy storage capacity value is the upper bound of the reading error of the energy storage capacity reading under the metering caliber. The above processing is used to consistently map the time error and sensing error to the energy caliber of the time window.

[0045] Step 33: Based on the correlation matrix and mapping matrix, the error quantities are directionally summarized at the node side. The energy measurement errors of the total energy flowing into and out of each component are combined directionally, and summed with the energy measurement errors of the energy changes in the energy storage components within the node to obtain the node energy conservation residual measurement error. The sum of the absolute values ​​of the total energy flowing into and out of each component at that node is used as a normalization factor, and the ratio of the node energy conservation residual measurement error to the normalization factor is used as the node residual threshold. Within the same time window, the largest of all node residual thresholds is taken as the network consistency residual threshold, and it is bound to the time window, sampling interval, correlation matrix, and mapping matrix. The above processing is used to obtain a dimensionless threshold that is independent of node size and can be compared across nodes, for direct reference in subsequent consistency comparisons.

[0046] Through the above process, this embodiment achieves a unified caliber of uncertainty from power to energy by transmitting voltage measurement error, current measurement error, and time error at the sampling point level; by summarizing at the node side according to the correlation matrix and mapping matrix and normalizing by the sum of the absolute values ​​of inflow and outflow energy, a node residual threshold independent of scale can be obtained; finally, the largest of the residual thresholds of each node is used as the network consistency residual threshold and the binding with time window and topology is completed, providing a verifiable test criterion for subsequent consistency comparison, and reducing false alarms and missed alarms caused by measurement and timing errors.

[0047] In one embodiment of the present invention, a residual vector is formed by comparing the inflow energy, outflow energy, and energy change of energy storage components based on the correlation matrix and the mapping matrix, and then compared with the network consistency residual threshold to obtain the network-wide energy consistency judgment result, including:

[0048] Step 41: Based on the correlation matrix and mapping matrix, determine the inflow component set, outflow component set, and energy storage component set for each node. Within the time window, sum the total energy of each corresponding component and the energy change of the energy storage component according to the sets to obtain the inflow energy, outflow energy, and energy storage change. The inflow component set and outflow component set are used to identify components that are connected to the node via branches and are either fed into or out of the node in the specified direction. The energy storage component set is used to identify energy storage components whose energy content changes within the time window and are connected to the node. These three sets correspond one-to-one with the time window, correlation matrix, and mapping matrix, serving as the input for subsequent calculations.

[0049] Step 42: For each node, subtract the outflow energy and the change in stored energy from the inflow energy, and take the absolute value as the residual numerator. The node energy conservation residual is obtained by comparing the ratio of the residual numerator to the normalization factor in step 33. The node energy conservation residual is a dimensionless quantity used to characterize the degree of energy conservation deviation of the node within the stated time window. The normalization factor is derived from the sum of the absolute values ​​of the total energy flowing into and out of all components of the node within the same time window, ensuring comparability between nodes of different scales.

[0050] Step 43: Arrange the node energy conservation residuals of all nodes according to node number to form a residual vector. Compare the node energy conservation residuals with the corresponding node residual thresholds and record the comparison results. Take the maximum value in the residual vector and compare it with the network consistency residual threshold to generate a network-wide energy consistency judgment result. The network-wide energy consistency judgment result includes consistent and inconsistent results. The comparison results are bound to the time window, sampling interval, correlation matrix, and mapping matrix, and used as the reference for subsequent classification and handling steps.

[0051] Through the above process, this embodiment achieves consistency in direction and dimension in energy conservation verification by aggregating and summarizing the inflow, outflow, and energy storage changes at the node side; by converting the energy conservation residuals of the nodes into dimensionless residuals using a normalization factor, objective comparisons can be made across nodes; finally, the energy consistency judgment result of the entire network is obtained by comparing the maximum value of the residual vector with the network consistency residual threshold, providing a verifiable test criterion for subsequent state classification and coordinated response.

[0052] In one embodiment of the present invention, efficiency deviation, temperature rise deviation, and terminal voltage retention deviation are calculated based on the commissioning baseline and current metering data, and a weighted health score is obtained; controllability margin is obtained based on the upper and lower limits of current, port voltage, temperature, and energy storage capacity, as well as the current value, including:

[0053] Step 51: Based on the commissioning baseline and current metering data, distinguish between input energy and output energy according to the mapping matrix. Use the ratio of output energy to input energy as the current efficiency. Divide the absolute difference between the current efficiency and the baseline efficiency by the baseline efficiency to obtain the efficiency deviation, which is used to characterize the degree of deviation of the component's energy conversion performance relative to the commissioning baseline within the time window. Within the time window, use the average value of the temperature sampling sequence as the current temperature. Divide the absolute difference between the current temperature and the baseline temperature by the allowable temperature range to obtain the temperature rise deviation, which is used to characterize the degree of deviation of the component's thermal state relative to the commissioning baseline. Within the time window, use the average value of the port voltage sampling sequence as the current port voltage. Divide the absolute difference between the current port voltage and the baseline port voltage by the allowable voltage range to obtain the terminal voltage retention deviation, which is used to characterize the degree of deviation of the port voltage stability relative to the commissioning baseline. The commissioning baseline includes baseline efficiency, baseline temperature, and baseline port voltage, as well as the corresponding allowable range. The current metering data consists of voltage data, current data, temperature sequence, and port voltage sequence acquired at sampling intervals within a time window, and is bound to the time window, sampling interval, and mapping matrix to ensure uniform input.

[0054] Step 52: According to the weights set at commissioning and summed to one, the efficiency deviation, temperature rise deviation, and terminal pressure retention deviation are weighted and summed to obtain the weighted deviation; the weighted deviation is subtracted from one and truncated within the range of zero and one to obtain the health score. The health score is a dimensionless quantity, with a value between zero and one, used to characterize the overall deviation of the component from the commissioning baseline within this time window, and is bound to the time window and weight settings for verification.

[0055] Step 53: Calculate the uplink margin and downlink margin based on the current upper and lower limits and their current values, the port voltage upper and lower limits and their current values, the temperature upper and lower limits and their current values, and the energy storage capacity upper and lower limits and their current values, respectively. The smaller of the two is taken as the constraint margin. The smallest of all constraint margins is taken as the controllability margin. The upper and lower limits are derived from the hard constraint settings of the commissioning baseline. The current value is the representative value of the corresponding physical quantity within the time window. The upper limit related quantity is the maximum value within the window, and the lower limit related quantity is the minimum value within the window. The controllability margin is a dimensionless quantity, ranging from zero to one, used to characterize the minimum adjustable margin of the component under the combined action of various hard constraints.

[0056] Through the above process, this embodiment achieves a deterministic weighted characterization of health by using the commissioning baseline as a unified reference and calculating three types of deviations within the same time window; by uniformly converting the upper and lower limits of current, port voltage, temperature, and energy storage capacity with the current values ​​into constraint margins, a controllability margin independent of equipment rating can be obtained; finally, standardized measurement results of health and controllability margins are formed, providing a verifiable and comparable electrical parameter basis for subsequent attribution linkage and classification determination.

[0057] In one embodiment of the present invention, an attribution matrix is ​​constructed based on the electrical connection relationship between nodes and components, the residual vector is distributed into a residual distribution vector, and the health and controllability margin are adjusted in a linked manner to obtain the adjusted health and controllability margin, including:

[0058] Step 61: Determine the electrical connection relationships between nodes and components based on the association matrix and mapping matrix, and construct the attribution matrix. Summate the row elements of each node and add them to the non-negative infinitesimal quantity as the denominator. Divide each element of the row by the denominator to complete row normalization, and bind the attribution matrix to the time window one by one. Here, the electrical connection relationship refers to the direct connection relationship between the component metering port and the target node under the topology determined by the primary system wiring method and the actual position of the switch. The rows of the attribution matrix correspond to each node, and the columns correspond to each component. The matrix elements are the apportionment weights of the corresponding component to the energy conservation residual of the corresponding node. The non-negative infinitesimal quantity is a constant not less than zero, used to avoid the denominator being zero and to maintain the row normalization property.

[0059] Step 62: Multiply the transpose of the attribution matrix with the attribution matrix, then invert the result under the condition of invertibility. Multiply the result with the transpose of the attribution matrix, and then multiply with the residual vector to obtain the residual allocation vector. Set the components of the residual allocation vector that are less than zero to zero, and use them as inputs for linkage correction. The residual vector is composed of the energy conservation residuals of each node according to the node number. Each component of the residual allocation vector is used to quantify the deterministic contribution intensity of the corresponding component to the residual, and is bound to the time window and the attribution matrix one by one.

[0060] Step 63: Subtract the product of the preset sensitivity coefficient and the corresponding component of the residual distribution vector from the health score, and truncate the result within the range of zero and one to obtain the corrected health score; process the controllability margin according to the same rule to obtain the corrected controllability margin. The preset sensitivity coefficient is a non-negative constant used to set the linkage tightening amplitude; the corrected health score and the corrected controllability margin are dimensionless quantities, taking values ​​between zero and one, and are bound to the time window, correlation matrix, mapping matrix, and attribution matrix one by one, serving as inputs for subsequent classification determination and handling closed loop.

[0061] This embodiment achieves deterministic allocation of node-side residuals to component-side by constructing a row-normalized attribution matrix under topological caliber; by performing non-negative truncation on the allocation results, it can suppress negative contributions without physical meaning; finally, it tightens the health and controllability margins with linkage correction rules, forming a correction state quantity consistent with the time window and topology, providing a verifiable and comparable measurement basis for subsequent classification and handling.

[0062] In one embodiment of the present invention, the classification determination process includes:

[0063] For each component, the corrected health level is compared with the health level threshold, and the corrected controllability margin is compared with the controllability margin threshold, forming two comparison results. The comparison results are only of two types: greater than or equal to the corresponding threshold or less than the corresponding threshold. These are used as input markers for subsequent judgments. The health level threshold and the controllability margin threshold are preset dimensionless constants with values ​​ranging from zero to one.

[0064] For each component, a judgment is made based on the overall network energy consistency assessment result and two comparison results. This judgment serves as the basis for subsequent handling and closed-loop verification. Specifically, if the overall network energy consistency assessment result is consistent and both comparison results are greater than or equal to the corresponding threshold, it is considered normal, indicating that network-side energy conservation is satisfied. If the overall network energy consistency assessment result is inconsistent and both comparison results are greater than or equal to the corresponding threshold, it is considered inconsistent across the entire network, indicating that the network-side energy conservation verification has failed. However, the component-level state variables are not below the threshold, indicating that the problem is primarily located at the network-side testing caliber level. This triggers priority verification in the metering-side verification queue. The network testing caliber includes topology, metering... The data aggregation link is as follows: When the overall network energy consistency judgment result is consistent and any comparison result is less than the corresponding threshold, it is judged as local degradation, indicating that the network-side energy conservation is satisfied, but the component-level state quantity deviates outside the threshold, indicating that the problem is located at the component-side capability or stability level, which is used to trigger the handling and limit setting for that component; when the overall network energy consistency judgment result is inconsistent and any comparison result is less than the corresponding threshold, it is judged as coupling anomaly, indicating that the network side and component side simultaneously trigger abnormal conditions, indicating that there is a coupling anomaly between the network layer and the component layer, which serves as the basis for simultaneously triggering the handling of metering-side verification and component-side limit, and enters the continuous time window closed-loop review process. The above judgment rules are a deterministic one-to-one correspondence and do not introduce statistical learning or prediction steps.

[0065] The classification determination is bound to time window, health threshold, controllability margin threshold, network consistency residual threshold, corrected health and corrected controllability margin, and time window-level classification records are generated, which serve as the sole reference for subsequent processing and closed-loop review.

[0066] Through the above process, this embodiment achieves comparable judgment criteria between different components and different time windows by thresholding continuous state variables to form comparison results; by combining network layer consistency conclusions with component layer state variables with deterministic rules, it can distinguish four types of results: normal, network-wide inconsistency, local degradation, and coupling anomalies; finally, the classification judgment and threshold are bound to the state variables within the time window to ensure the traceability and verifiability of subsequent processing links.

[0067] In one embodiment of the present invention, a metering-side verification queue, power limit, and current limit are generated based on the classification determination result, and closed-loop verification is performed within a continuous time window according to the release conditions, including:

[0068] Step 71: When the overall network energy consistency judgment result is inconsistent, all components are taken as the verification object set; when the overall network energy consistency judgment result is consistent, the verification object set is empty. The components within the verification object set are sorted from largest to smallest according to the corresponding components of the residual allocation vector to obtain the metering-side verification queue. Components classified as locally degraded or abnormally coupled are identified as the quota object set. The residual allocation vector is the component-level contribution intensity obtained from the aforementioned attribution solution. The metering-side verification queue is used to determine the verification order of the metering links.

[0069] Step 72: For each component within the limit object set, the product of the component's rated power and the corrected controllability margin is used as the candidate power. The smaller of the candidate power and the component's rated power is taken as the power limit. The product of the component's rated current and the corrected controllability margin is used as the candidate current. The smaller of the candidate current and the component's rated current is taken as the current limit. The power limit and current limit are then issued for execution and bound to time windows. The power limit and current limit are time window-level processing quantities, corresponding one-to-one with the source of the classification judgment and the corrected controllability margin.

[0070] Step 73: Within subsequent consecutive time windows, if the maximum residual of a node is not higher than the network consistency residual threshold, and the corrected health and corrected controllability margin of the limit-required component are both greater than or equal to the corresponding thresholds, the limit is lifted and the disposal flag is cleared when the number of consecutively satisfied time windows reaches a preset continuous threshold; otherwise, the current limit is maintained and the review continues. The preset continuous threshold is a positive integer used to avoid frequent switching caused by occasional fluctuations; the disposal flag records the limit triggering reason, limit value, and time window index, and is bound to the disposal result one by one.

[0071] Through the above process, this embodiment determines the metering-side verification queue by using the residual allocation vector to form a deterministic verification sequence for the test link; it quantifies the power limit and current limit with the corrected controllability margin to obtain a time window-level processing quantity consistent with the equipment capacity; and finally, it performs closed-loop verification with a unified threshold within a continuous time window and releases the processing according to the preset continuous threshold, ensuring the traceability and stable caliber of the test processing.

[0072] This invention also provides a microgrid structural component condition testing and processing system, comprising:

[0073] The topology metering module is used to generate an association matrix and a mapping matrix based on the primary system wiring method and the actual position of the switch; it sets a time window, collects voltage and current data of each component, and obtains power data, total energy of each component and energy change of energy storage component.

[0074] The threshold generation module is used to determine the network consistency residual threshold based on voltage measurement error, current measurement error, and time error.

[0075] The consistency determination module is used to compare the inflow energy, outflow energy, and energy change of energy storage components with the correlation matrix and the mapping matrix to form a residual vector, and compare it with the network consistency residual threshold to obtain the energy consistency determination result of the entire network.

[0076] The health margin module is used to calculate efficiency deviation, temperature rise deviation, and terminal pressure retention deviation based on the commissioning baseline and current metering data, and then weight them to obtain the health level; and to obtain the controllability margin based on voltage, temperature, upper and lower limits of energy storage capacity and current values.

[0077] The attribution linkage module is used to construct an attribution matrix based on the electrical connection relationship between nodes and components, distribute the residual vector into a residual distribution vector, and perform linkage correction on health and controllability margin to obtain the corrected health and controllability margin.

[0078] The classification determination module is used to obtain the classification determination based on the network-wide energy consistency judgment result, the corrected health status, and the corrected controllability margin.

[0079] The closed-loop processing module is used to generate metering-side verification queues, power limits, and current limits based on the classification judgment results, and to perform closed-loop verification according to the release conditions within a continuous time window.

[0080] It should be noted that the interval and threshold sizes are set for ease of comparison. The size of the threshold depends on the amount of sample data and the base number set by those skilled in the art for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. Furthermore, the above formulas are all dimensionless calculations, and the formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0081] The embodiments of the present invention have been described above, but the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of the present embodiments, all of which are within the protection scope of the present embodiments.

Claims

1. A method for testing and processing the condition of microgrid structural components, characterized in that, Includes the following steps: Step 1: Generate an association matrix and a mapping matrix based on the primary system wiring method and the actual switch positions; Set a time window, collect voltage and current data of each component, and obtain power data, total energy of each component and energy storage component energy change; Step 2: Determine the network consistency residual threshold based on voltage measurement error, current measurement error, and time error; Step 3: Based on the correlation matrix and mapping matrix, compare the inflow energy, outflow energy and energy change of energy storage components to form a residual vector, and compare it with the network consistency residual threshold to obtain the network-wide energy consistency judgment result; Step 4: Calculate the efficiency deviation, temperature rise deviation, and terminal pressure retention deviation based on the commissioning baseline and current metering data, and obtain the health status by weighting; obtain the controllability margin based on the voltage, temperature, upper and lower limits of energy storage capacity and the current value. Step 5: Construct an attribution matrix based on the electrical connection relationship between nodes and components, distribute the residual vector into a residual distribution vector, and perform a linkage correction on the health and controllability margin to obtain the corrected health and controllability margin. Step 6: Based on the results of the overall network energy consistency assessment, the corrected health level, and the corrected controllability margin, the classification determination is obtained; Step 7: Generate the metering side verification queue, power limit, and current limit based on the classification judgment results, and conduct closed-loop verification according to the release conditions within a continuous time window.

2. The microgrid structural component condition testing and processing method according to claim 1, characterized in that, Generate correlation and mapping matrices to obtain power data, total energy of each component, and energy change of energy storage components, including: Step 11: Based on the primary system wiring method and the actual position of the switch, generate an association matrix with nodes as rows and branches as columns. The elements of the association matrix are the inflow, outflow, and unconnected status of the corresponding branch at the node. Nodes represent electrical equipotential connection points formed by conductors that are directly connected to each other and have the same potential. Generate a mapping matrix with components as rows and branches as columns. The elements of the mapping matrix indicate whether the corresponding component is connected to a branch. Step 12: Set the start and end times of the time window and the sampling interval, collect voltage and current data for each component, and form a sampling sequence according to the sampling point order; Step 13: Within the same time window, multiply the voltage data and current data at each sampling point to obtain the power data; multiply the power data at each sampling point by the sampling interval and accumulate them in the order of sampling points to obtain the total energy of each component; read the current value of the energy storage capacity at the beginning and end of the time window for the energy storage component and subtract them to obtain the energy change of the energy storage component.

3. The microgrid structural component condition testing and processing method according to claim 2, characterized in that, Step 1 also includes: Step 21: Compare the sampling timestamps within the time window according to the sampling point order. Sampling points with unequal time differences and sampling intervals, or those that are duplicated or in reverse order, are marked as invalid samples. Step 22: Delete the voltage and current data corresponding to invalid samples, and fill them with the voltage and current data of the previous valid sampling point by copying them with equal duration, keeping the number of sampling points unchanged and generating a filling index; Step 23: Based on the filled voltage and current data, recalculate the power data, total energy of each component, and energy change of the energy storage component according to step 13, and bind the filled index to the time window one by one.

4. The microgrid structural component condition testing and processing method according to claim 2, characterized in that, The network consistency residual threshold is determined based on voltage measurement error, current measurement error, and time error, including: Step 31: After the time window and sampling interval are determined, set the voltage measurement error, current measurement error and time error. At each sampling point, multiply the voltage data by the voltage measurement error and the current data by the current measurement error, and add the two products to obtain the power measurement error. Step 32: Within the same time window, multiply the power measurement error of each sampling point by the sampling interval and accumulate them in the sampling order to obtain the power error accumulation value; multiply the time error by the accumulation of the absolute value of the power data within the time window and add it to the power error accumulation value to obtain the energy measurement error of the total energy of each component; for energy storage components, take the sum of the measurement errors of the current value of the energy storage capacity at the beginning and end of the time window as the energy measurement error of the energy change of the energy storage component. Step 33: Based on the correlation matrix and mapping matrix, sum the energy measurement error of the total energy flowing into and out of each component and the energy measurement error of the energy change of the energy storage component within the node to obtain the node energy conservation residual measurement error; use the sum of the absolute values ​​of the total energy flowing into and out of each component as the normalization factor, and use the ratio of the node energy conservation residual measurement error to the normalization factor as the node residual threshold; take the largest of all node residual thresholds as the network consistency residual threshold, and bind it to the time window, sampling interval, correlation matrix and mapping matrix one by one.

5. The microgrid structural component condition testing and processing method according to claim 4, characterized in that, The residual vector is generated to obtain the energy consistency judgment result of the entire network, including: Step 41: Determine the inflow component set, outflow component set, and energy storage component set for each node based on the correlation matrix and mapping matrix. Within the time window, sum the total energy of each corresponding component and the energy change of the energy storage component according to the set to obtain the inflow energy, outflow energy, and energy storage change. Step 42: For each node, subtract the outflow energy from the inflow energy and then subtract the change in stored energy, and take the absolute value as the residual numerator. The node energy conservation residual is obtained by the ratio of the residual numerator to the normalization factor in step 33. Step 43: Arrange the node energy conservation residuals of all nodes according to the node number to form a residual vector. Compare the node energy conservation residuals with the corresponding node residual thresholds and record the comparison results. Take the maximum value in the residual vector and compare it with the network consistency residual threshold to generate the network-wide energy consistency judgment result. The network-wide energy consistency judgment result includes consistency and inconsistency.

6. The microgrid structural component condition testing and processing method according to claim 1, characterized in that, The efficiency deviation, temperature rise deviation, and terminal pressure retention deviation are calculated, weighted, to obtain the health score and the controllability margin, including: Step 51: Based on the commissioning baseline and current metering data, distinguish between input energy and output energy according to the mapping matrix. Use the ratio of output energy to input energy as the current efficiency. Divide the absolute difference between the current efficiency and the baseline efficiency by the baseline efficiency to obtain the efficiency deviation. Within the time window, use the average value of the temperature sampling sequence as the current temperature. Divide the absolute difference between the current temperature and the baseline temperature by the allowable temperature range to obtain the temperature rise deviation. Within the time window, use the average value of the port voltage sampling sequence as the current port voltage. Divide the absolute difference between the current port voltage and the baseline port voltage by the allowable voltage range to obtain the terminal voltage retention deviation. Step 52: According to the weights set at the time of commissioning, the efficiency deviation, temperature rise deviation and terminal pressure retention deviation are weighted and summed to obtain the weighted deviation. The weighted deviation is subtracted from one and truncated within the range of zero and one to obtain the health status. Step 53: Calculate the uplink margin and downlink margin based on the current upper and lower limits and current values, the port voltage upper and lower limits and current values, the temperature upper and lower limits and current values, and the energy storage capacity upper and lower limits and current values, respectively, and take the smaller of the two as the constraint margin; take the smallest of all constraint margins as the controllability margin.

7. The microgrid structural component condition testing and processing method according to claim 1, characterized in that, Construct the attribution matrix to obtain the corrected health score and the corrected controllability margin, including: Step 61: Determine the electrical connection relationship between nodes and components based on the correlation matrix and mapping matrix, and construct the attribution matrix; sum the row elements of each node and add them to the non-negative small quantity as the denominator, divide each element of the row by the denominator to complete row normalization, and bind the attribution matrix to the time window one by one; wherein, the rows of the attribution matrix correspond to each node, the columns correspond to each component, and the matrix elements are the apportionment weights of the energy conservation residuals of the corresponding components to the corresponding nodes. Step 62: Multiply the transpose of the attribution matrix with the attribution matrix and then invert the result. Multiply the result with the transpose of the attribution matrix and then with the residual vector to obtain the residual distribution vector. Set the components of the residual distribution vector that are less than zero to zero and use them as input for linkage correction. Step 63: Subtract the product of the preset sensitivity coefficient and the corresponding component of the residual distribution vector from the health score, and truncate it within the range of zero and one to obtain the corrected health score; process the controllability margin according to the same rule to obtain the corrected controllability margin.

8. The microgrid structural component condition testing and processing method according to claim 1, characterized in that, The process of classification determination includes: For each component, the corrected health level is compared with the health level threshold, and the corrected controllability margin is compared with the controllability margin threshold, resulting in two comparison results; The determination is based on the overall network energy consistency assessment results. When the overall network energy consistency assessment result is consistent and both comparison results are greater than or equal to the corresponding thresholds, it is determined to be normal; when the overall network energy consistency assessment result is inconsistent and both comparison results are greater than or equal to the corresponding thresholds, it is determined to be inconsistent across the entire network; when the overall network energy consistency assessment result is consistent and either is less than the corresponding threshold, it is determined to be local degradation; when the overall network energy consistency assessment result is inconsistent and either is less than the corresponding threshold, it is determined to be coupling anomaly.

9. The microgrid structural component condition testing and processing method according to claim 8, characterized in that, Generate metering-side verification queues, power limits, and current limits, and conduct closed-loop verification according to the release conditions within a continuous time window, including: Step 71: When the overall network energy consistency judgment result is inconsistent, take all components as the verification object set; when the overall network energy consistency judgment result is consistent, the verification object set is empty. Sort the components in the verification object set from largest to smallest according to the corresponding components of the residual allocation vector to obtain the metering side verification queue. And determine the components that are classified as local degradation or coupling abnormality as the quota object set. Step 72: For each component in the limit object set, the product of the component's rated power and the corrected controllability margin is used as the candidate power. The candidate power is compared with the component's rated power, and the smaller one is taken as the power limit. The product of the component's rated current and the corrected controllability margin is used as the candidate current. The candidate current is compared with the component's rated current, and the smaller one is taken as the current limit. The power limit and current limit are issued for execution and bound to the time window one by one. Step 73: In subsequent consecutive time windows, if the maximum residual of a node is not higher than the network consistency residual threshold and the corrected health and corrected controllability margin of the regulated component are both greater than or equal to the corresponding threshold, the regulated limit is lifted and the disposal mark is cleared when the number of consecutively satisfied time windows reaches the preset consecutive threshold; otherwise, the current regulated limit is maintained and the review continues.

10. A microgrid structural component condition testing and processing system, characterized in that, The microgrid structure component condition testing and processing method as described in any one of claims 1-9 includes: The topology metering module is used to generate an association matrix and a mapping matrix based on the primary system wiring method and the actual position of the switch; it sets a time window, collects voltage and current data of each component, and obtains power data, total energy of each component and energy change of energy storage component. The threshold generation module is used to determine the network consistency residual threshold based on voltage measurement error, current measurement error, and time error. The consistency determination module is used to compare the inflow energy, outflow energy, and energy change of energy storage components with the correlation matrix and the mapping matrix to form a residual vector, and compare it with the network consistency residual threshold to obtain the energy consistency determination result of the entire network. The health margin module is used to calculate efficiency deviation, temperature rise deviation, and terminal pressure retention deviation based on the commissioning baseline and current metering data, and then weight them to obtain the health level; and to obtain the controllability margin based on voltage, temperature, upper and lower limits of energy storage capacity and current values. The attribution linkage module is used to construct an attribution matrix based on the electrical connection relationship between nodes and components, distribute the residual vector into a residual distribution vector, and perform linkage correction on health and controllability margin to obtain the corrected health and controllability margin. The classification determination module is used to obtain the classification determination based on the network-wide energy consistency judgment result, the corrected health status, and the corrected controllability margin. The closed-loop processing module is used to generate metering-side verification queues, power limits, and current limits based on the classification judgment results, and to perform closed-loop verification according to the release conditions within a continuous time window.

Images