Energy supply system stability assessment method, system and medium based on energy unit

Abstract

This invention relates to the field of energy supply system stability assessment technology, specifically to a method, system, and medium for energy supply system stability assessment based on energy units. The method involves collecting time-series data of operating parameters from each energy unit. Given the fluctuation differences between energy units under different operating conditions, time-level clustering is used to divide the energy units into unit groups, and representative vectors for each unit group are extracted to reflect its overall operating status. Subsequently, the time-series data is divided into time periods, and comprehensive clustering is performed at the time-period level. Combining the operating status, distribution characteristics, and energy unit parameter similarity of the unit groups, an operating similarity value is constructed to measure the similarity of system operating states between time periods. Classifying time periods based on operating similarity values ​​allows for accurate segmentation of time periods with similar operating states, effectively improving the accuracy and typicality of target time period selection, thereby enhancing the credibility of system stability assessment.

Description

Technical Field

[0001] This invention relates to the field of energy supply system stability assessment technology, specifically to a method, system, and medium for energy supply system stability assessment based on energy units. Background Technology

[0002] With the continuous growth of global energy demand and the diversification of energy structures, energy supply systems (such as distributed energy networks, microgrids, and smart grids) are playing an increasingly prominent role in modern energy systems. These systems typically consist of a large number of heterogeneous energy units that work together to achieve efficient energy conversion, storage, and distribution. However, the dynamic characteristics of energy supply systems are complex. Environmental conditions, load demand fluctuations, equipment aging, and communication delays can all lead to instability in system operation and even safety accidents. Therefore, it is necessary to assess the stability of energy supply systems.

[0003] In assessing the stability of energy supply systems, existing technologies typically cluster time periods based on the operating parameters of energy units to select typical time periods as monitoring targets. However, since numerous energy sources are connected to the energy supply system, and the volatility and intermittency of different energy sources vary under different operating conditions, directly clustering time periods based on the operating parameters of energy units during the screening process can lead to inaccurate classification, thereby affecting the accuracy of the selection of monitoring targets and reducing the credibility of the energy supply system stability assessment. Summary of the Invention

[0004] To address the issue that energy supply systems incorporate numerous energy sources, and that the volatility and intermittency of these energy sources vary under different operating conditions, directly clustering time periods based on energy unit operating parameters during the screening process can lead to inaccurate classification, thereby affecting the accuracy of monitoring target selection and reducing the reliability of energy supply system stability assessments, this invention aims to provide a method, system, and medium for energy supply system stability assessment based on energy units. The specific technical solution adopted is as follows:

[0005] A method for stability assessment of an energy supply system based on energy units, comprising:

[0006] Acquire time-series data of various operating parameters for each energy unit in the energy supply system;

[0007] Cluster all energy units at each time point to obtain unit groups; within each unit group, analyze the similarity of operating parameters, numerical characteristics, and distribution density of energy units to determine the representative vector of each unit group;

[0008] The time series is divided into time periods. Within different time periods, the similarity of the representative vectors of the unit groups and the similarity of the distribution of energy units are analyzed. Combined with the similarity characteristics between the operating parameters of the energy units, the operating similarity value of the energy supply system between any two time periods is determined. This value is used to classify the time periods and thus select target time periods.

[0009] The stability of the energy supply system during the target time period is assessed based on the operating parameters and arrangement of the energy units.

[0010] Furthermore, the method for obtaining the representative vector includes:

[0011] The operating parameters of each energy unit at each time point are combined into an operating state vector.

[0012] In each unit group, the cosine similarity between the operating state vectors of any two energy units is used as the similarity factor between the two energy units. The standard deviation of all similarity factors corresponding to each energy unit is negatively correlated and normalized, and the value is used as the intra-group representativeness of each energy unit.

[0013] Within each unit group, under each operating parameter, the operating parameter values ​​of the energy units are weighted and fused using the in-group representativeness of the energy units to obtain the operating characteristic values ​​of each unit group under each operating parameter.

[0014] In each unit group, the Euclidean distance between the operating state vectors of any two energy units is used as a distance factor. The mean of the distance factors between all energy units is negatively correlated and the product of the mean and the number of energy units is normalized to obtain the density factor of the energy unit distribution in each unit group.

[0015] The operating characteristic values ​​and density factors of each unit group under all operating parameters are used to form the representative vector of each unit group.

[0016] Furthermore, the method for obtaining the similarity value includes:

[0017] By analyzing the similarity between the representative vectors of the unit groups and the similarity of the energy unit distribution in different time periods, the state similarity between any two time periods can be obtained.

[0018] Within any two time periods, at each same location and time, the cosine similarity between the operating state vectors of the same type of energy units is used as the similarity index between the energy units of that type. The mean of the similarity indices between all types of energy units at each same location and time is used as the mean state similarity between the two time periods at each same location and time. The normalized sum of the mean state similarity values ​​of the two time periods at all the same location and time is used as the state similarity value of the energy supply system within the two time periods.

[0019] For any two time periods, the normalized value of the state similarity between the two time periods is used as the first adjustment weight, and the value of the first adjustment weight after negative correlation mapping is used as the second adjustment weight.

[0020] The product of the state similarity between the two time periods and the first adjustment weight, and the product of the state similarity value and the second adjustment weight are added together. The normalized sum is then used as the operational similarity value of the power supply system between the two time periods.

[0021] Furthermore, the method for obtaining the state similarity includes:

[0022] Within any two time periods, compare the energy units in any two unit groups within these two time periods, take the number of overlapping energy units as the overlap factor, and take two unit groups with an overlap factor greater than a preset overlap threshold as a combination, thereby obtaining all combinations corresponding to these two time periods.

[0023] For any combination, calculate the cosine similarity between the representative vectors of two unit groups as a similarity parameter;

[0024] For any two time periods, the corresponding overlap factors are weighted and fused based on the similarity parameters corresponding to the combination to obtain the state similarity between the two time periods.

[0025] Furthermore, the method for obtaining the target time period includes:

[0026] Cluster analysis was performed on all time periods based on the K-means clustering algorithm and a preset K value to obtain all categories. The distance metric is the value after negative correlation mapping of the running similarity values ​​between time periods.

[0027] Multiply the proportion of time slots in each category to the total number of time slots in all categories by a preset quantity to obtain the selection quantity for each category;

[0028] Within each category, the mean of the operational similarity values ​​between each time period and all other time periods is normalized and used as the prevalence of each time period. Time periods with a prevalence greater than a preset prevalence threshold are classified as ordinary time periods, while time periods with a prevalence less than or equal to the preset prevalence threshold are classified as special time periods.

[0029] In each category, the target number of ordinary time slots is the product of the proportion of ordinary time slots and the selection quantity corresponding to each category; the target number of special time slots is the product of the proportion of special time slots and the selection quantity corresponding to each category.

[0030] Within each category, the target quantity based on the normal time period is randomly selected during the normal time period, and the target quantity based on the special time period is randomly selected during the special time period;

[0031] Select the time period from all categories as the target time period.

[0032] Furthermore, the stability assessment of the energy supply system within the target time period based on the operating parameters and arrangement of energy units within the target time period includes:

[0033] The types of operating parameters include at least current, voltage, impedance, output power, and load power;

[0034] Based on the electrical connection relationship of various devices in the energy supply system, a node admittance matrix is ​​established. The time series data of the operating parameters of all energy units in the energy supply system within each target time period are associated with the node admittance matrix to obtain the power network model.

[0035] In the power network model, the time series data of the operating parameters of the energy unit under each target time period are processed according to the three-phase balanced system, the three-phase unbalanced system, Ohm's law and Joule's heat loss formula to obtain the current value, voltage value of the power supply system at each time, and the power loss value of each energy unit in the power supply system at each time.

[0036] Based on the power flow calculation method and the output power, load power and power loss value of each energy unit at each time, the power value of the energy supply system at each time is determined;

[0037] Based on the current, voltage and power values ​​of the energy supply system at each moment within the target time period, determine the limit power and critical voltage of the energy supply system within the target time period.

[0038] The ratio of the difference between the limit power of the power supply system within the target time period and the power value at each moment to the power value at each moment is used as the power angle stability reserve coefficient at each moment, where the power angle stability reserve coefficient is a percentage.

[0039] The voltage stability reserve coefficient at each moment is the ratio of the difference between the voltage value at each moment and the corresponding critical voltage within the target time period to the voltage value at each moment. The voltage stability reserve coefficient is a percentage.

[0040] If the average power angle stability reserve coefficient at all times is within a preset first stability range and the average voltage stability reserve coefficient at all times is within a preset second stability range within a certain target time period, then the stability of the power supply system is considered good within that target time period; otherwise, the stability is considered poor, and an early warning is issued.

[0041] Furthermore, determining the limiting power and critical voltage of the energy supply system within the target time period based on the current, voltage, and power values ​​of the energy supply system at each moment within the target time period includes:

[0042] Set the initial load level and increase the load according to a preset ratio;

[0043] The current, voltage, and power values ​​of the power supply system within each target time period are analyzed based on the proportional load increment continuous power flow method. The convergence of the power flow solution is checked. If the convergence is not achieved, the collapse point is considered to have been reached, the calculation is terminated, and the load level corresponding to the collapse point is taken as the limit power of the power supply system within the target time period. The voltage value corresponding to the collapse point is taken as the critical voltage of the power supply system within the target time period.

[0044] Furthermore, the method for obtaining the unit group includes:

[0045] At each time step, the data points corresponding to the operating state vectors of all energy units are clustered based on the DBSCAN clustering algorithm to obtain all clusters, and the energy units in each cluster are taken as a unit group.

[0046] A stability assessment system for an energy supply system based on an energy unit includes a processor and a memory. The memory stores at least one instruction, at least one program, code set, or instruction set. When the processor loads and executes the at least one instruction, at least one program, code set, or instruction set, it implements the steps of the stability assessment method for an energy supply system based on an energy unit.

[0047] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for evaluating the stability of an energy supply system based on an energy unit.

[0048] The present invention has the following beneficial effects:

[0049] The energy supply system consists of a large number of heterogeneous energy units working together. Therefore, this invention collects time-series data of the operating parameters of each energy unit in the energy supply system to comprehensively capture the dynamic operating characteristics of the energy units. Given that different energy units exhibit different fluctuations under different operating conditions, the energy units at each time point are initially clustered at the time level to obtain unit groups. Within each unit group, the numerical characteristics, similarities, and distribution density of the operating parameters of the energy units are analyzed to determine a representative vector for the unit group. This representative vector reflects the overall operating status of each unit group at each time point. To select the target data for final monitoring, the time series can be divided into time periods, and representative time periods should be selected. Therefore, all time periods can be clustered at the time period level. When clustering time periods, this invention integrates the similarity of the overall operating status of the unit groups, the similarity of the unit group distribution, and the similarity characteristics of the operating parameters of the energy units. This allows for a combination of local and overall characteristics in the energy supply system, thereby more accurately measuring the similarity between two time periods and obtaining an index reflecting the similarity of the operating status of the energy supply system between two time periods—the operating similarity value. At this point, classifying time periods based on operational similarity values ​​allows for accurate segmentation of time periods with similar operational states. This effectively improves the accuracy and typicality of target time period selection, thereby enhancing the reliability of the final stability assessment of the power supply system. In summary, this solution adapts to system characteristics under different operating conditions, effectively focuses on critical operating segments, and improves the utilization efficiency of monitoring resources and the accuracy of assessment results. Attached Figure Description

[0050] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 A flowchart illustrating a method for evaluating the stability of an energy supply system based on energy units, provided in one embodiment of the present invention;

[0052] Figure 2 A flowchart illustrating a method for obtaining a representative vector according to an embodiment of the present invention;

[0053] Figure 3 This is a flowchart of a method for obtaining similarity values ​​according to an embodiment of the present invention;

[0054] Figure 4This is a flowchart of a stability assessment method provided in one embodiment of the present invention;

[0055] Figure 5 A system block diagram of a stability assessment system for an energy supply system based on an energy unit, provided as an embodiment of the present invention;

[0056] Figure 6 A schematic diagram of the system structure of a stability evaluation system for an energy supply system based on an energy unit, provided in an embodiment of the present invention;

[0057] Figure 7 This is a schematic diagram of a computer-readable storage medium provided according to an embodiment of the present invention. Detailed Implementation

[0058] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method, system, and medium for evaluating the stability of an energy supply system based on energy units proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0060] The following detailed description, in conjunction with the accompanying drawings, illustrates the specific scheme of the energy unit-based energy supply system stability assessment method, system, and medium provided by this invention.

[0061] Please see Figure 1 The diagram illustrates a flowchart of a method for evaluating the stability of an energy supply system based on energy units, according to an embodiment of the present invention. The method includes the following steps:

[0062] Step S1: Obtain time-series data of various operating parameters for each energy unit in the energy supply system.

[0063] With the transformation of the global energy structure and technological advancements, the concept of energy supply systems has gradually expanded in recent years from traditional isolated generator sets to integrated systems encompassing multiple energy sources and energy storage forms. In energy supply systems, stability is a crucial concept. In this embodiment of the invention, the timing data of the overall operating parameters of the energy supply system and the individual timing data of the operating parameters of each energy unit are first acquired. The types of operating parameters should at least include current, voltage, impedance, output power, and load power.

[0064] Specifically, sensors can be installed on each energy unit (such as photovoltaic inverters, wind turbines, and traditional generator sets) to collect various operating parameters. Then, the parameters are standardized based on their rated values ​​(using the ratio of the actual parameter value to the rated value as the standardized value) to obtain time-series data for each operating parameter.

[0065] It should be noted that the time series data collection duration can be set to 1 hour, and the collection frequency can be set to once per second. The specific duration and frequency can be adjusted according to the implementation scenario and are not limited here. The rated value of each operating parameter can be based on the value recorded on the nameplate of the energy unit, or it can be set by yourself according to the implementation scenario and is not limited here.

[0066] Step S2: Cluster all energy units at each time point to obtain unit groups; within each unit group, analyze the similarity between the operating parameters of the energy units, their numerical characteristics, and the distribution density of the energy units to determine the representative vector of each unit group.

[0067] In this embodiment of the invention, when assessing the stability of an energy supply system, it is necessary to determine the system's operating status. However, due to factors such as environmental conditions, load demand fluctuations, equipment aging, and communication delays, the power supply from individual energy units can fluctuate intermittently. Furthermore, different energy units react differently to these factors. Therefore, even at the same moment, different energy units may exhibit different operating conditions, making it extremely complex to describe the operating status of an energy supply system at a given time. Therefore, cluster analysis can be performed on the energy units at each moment, dividing the numerous energy units into a limited number of unit groups, reducing the difficulty of subsequent analysis while retaining key feature information. Then, at the unit group level, the similarity, numerical characteristics, and distribution density of the energy unit's operating parameters are analyzed to form a macroscopic understanding of the unit group's operating status, resulting in a representative vector for the unit group, which reflects the operating condition of this group of energy units.

[0068] First, at each time point, cluster analysis is performed on all energy units in the energy supply system to obtain unit groups.

[0069] Preferably, in one embodiment of the present invention, the method for obtaining the unit group includes:

[0070] Each energy unit has an operating state vector composed of all its operating parameters at each time step. The order of the operating parameters in the operating state vectors of all energy units should be consistent.

[0071] Then, at each time step, the data points corresponding to the state vectors of all energy units are clustered based on the DBSCAN clustering algorithm to obtain all clusters, and the energy units in each cluster are taken as a unit group.

[0072] It should be noted that, in this embodiment of the invention, the neighborhood radius can be determined by... Function to obtain, where Values ​​take ( ,in (Number of types of running parameters); minimum number of points in the preset neighborhood. Pick In other embodiments of the present invention, the neighborhood radius and the minimum number of points in the neighborhood can be set according to the implementation scenario, and are not limited here; the DBSCAN clustering algorithm is a well-known technology, and the specific process will not be described in detail here.

[0073] Clustering the energy units at each time point yields a unit group, which distinguishes the operating states of numerous energy units at each time point. Next, within each unit group, the similarity and numerical characteristics of the operating parameters of the energy units can be analyzed. Combined with the distribution density of the energy units, a representative vector for each unit group is determined. This vector is used to integrate all the information and obtain a comprehensive characteristic data that can characterize the overall operating conditions of all energy units in the unit group—the representative vector.

[0074] Preferably, in one embodiment of the present invention, the method for obtaining the representative vector includes:

[0075] Please see Figure 2 The diagram illustrates a method flowchart for obtaining a representative vector in one embodiment of the present invention, which includes the following steps:

[0076] Step S201: In each unit group, analyze the similarity and numerical characteristics between the operating state vectors of the energy units, and determine the operating characteristic values ​​of each unit group under each operating parameter.

[0077] Based on the aforementioned steps, the operating state vector of each energy unit can be obtained. In each unit group, the similarity between the operating state vectors of any two energy units is measured. The cosine similarity between the operating state vectors of any two energy units is used as the similarity factor between the two energy units. The larger the similarity factor, the more similar the operating states of the two energy units are.

[0078] At this point, within the unit group, each energy unit shares a similarity factor with every other remaining energy unit. The more consistent the magnitudes of all similarity factors for a given energy unit, the stronger its representativeness within the unit group. This results in higher reference value when constructing the representative vector of the unit group later. Therefore, the standard deviation of all similarity factors for each energy unit is calculated. A smaller standard deviation indicates a higher degree of consistency among the similarity factors for that energy unit, thus increasing its representativeness. Therefore, the standard deviation is negatively correlated and normalized to correct logical relationships, thereby obtaining the intra-group representativeness of each energy unit. A higher intra-group representativeness indicates higher reference value for that energy unit in subsequent processes. This negative correlation mapping and normalization can be performed using the formula... ,in, Let x represent an exponential function with the natural constant e as the base, and let x represent the independent variable.

[0079] Finally, within each unit group, under each operating parameter, the operating parameter values ​​of the energy unit are weighted and fused using the in-group representativeness of the energy unit: for any given energy unit, the in-group representativeness of that energy unit is used... After normalization, the function is multiplied by the numerical value of the operating parameter of the energy unit to obtain a weighted parameter. The larger the weighted parameter, the larger the value of the energy unit under that operating parameter, and the higher its representativeness. This method can be used to obtain the weighted parameter of each energy unit in the unit group under each operating parameter. The sum of the weighted parameters of all energy units under each operating parameter is normalized and used as the operating characteristic value of the unit group under each operating parameter. The operating characteristic value represents the overall operating characteristics of all energy units in the entire unit group under each operating parameter. Normalization is a technique well known to those skilled in the art. The choice of normalization function can be linear normalization or standard normalization, etc., and the specific normalization method is not limited here.

[0080] Step S202: In each unit group, analyze the distribution density of energy units and determine the density factor of energy unit distribution in each unit group.

[0081] Euclidean distance can reflect the spatial distribution of the operating state vector of an energy unit, which helps to understand the distribution density of the operating state of the energy units in a unit group; while the number of energy units in a unit group can reflect the quantity distribution density of the energy units.

[0082] Therefore, within each unit group, the Euclidean distance between the operating state vectors of any two energy units is used as a distance factor. The smaller the distance factor, the closer the spatial distance between them, and the higher the distribution density of the operating states of the energy units. Simultaneously, the larger the number of energy units in a unit group, the denser the distribution of energy units. Therefore, the mean of the distance factors between all energy units is negatively correlated to correct the logical relationship, and then normalized by the product with the number of energy units to obtain the density factor of the energy unit distribution in each unit group. The larger the density factor, the better the aggregation of energy units within the unit group, and the higher the consistency of the operating states of the numerous energy units in that unit group.

[0083] It should be noted that the negative correlation mapping here can be handled using the formula. ,in, Let x represent an exponential function with the natural constant e as the base, and let x represent the independent variable. Normalization is a technique well known to those skilled in the art. The choice of normalization function can be linear normalization or standard normalization, etc., and the specific normalization method is not limited here.

[0084] Step S203: In each unit group, combine the density factor and the operating feature value under each operating parameter to obtain the representative vector of each unit group.

[0085] Based on step S201, the operating characteristic value of each unit group under each operating parameter can be obtained, which is used to reflect the overall operating characteristics of all energy units in the entire unit group under each operating parameter; based on step S202, the density factor of the distribution of energy units in each unit group is obtained, which can effectively compensate for the loss of some operating conditions caused by fitting the operating characteristics of the unit group from the operating conditions of many energy units.

[0086] Therefore, the representative vector of each unit group can be composed of the operating feature values ​​and density factors under all operating parameters. The order of parameters in the representative vectors of all unit groups should be consistent. The specific order is not specified here. In this embodiment of the present invention, they can be arranged in the order of operating parameter 1, operating parameter 2, operating parameter 3, ..., density factor.

[0087] Step S3: Divide the time series into time periods; within different time periods, analyze the similarity of the representative vectors of the unit groups and the similarity of the energy unit distribution, and combine the similarity characteristics between the operating parameters of the energy units to determine the operating similarity value of the energy supply system between any two time periods, which is used to classify the time periods and thus select target time periods.

[0088] In this embodiment of the invention, it is necessary to select representative time periods in the time series to conduct stability assessments of the energy supply system within the selected time periods. Therefore, the accuracy and representativeness of the time period selection are crucial. First, the time series can be divided into time periods. Before selecting target time periods, all time periods can be clustered to group time periods with similar operating conditions into one category. Then, the target time period is selected within each category to ensure that time periods with each operating condition characteristic are included, avoiding omissions or overlooking certain operating conditions. The aforementioned steps clustered the energy units at each moment and obtained the representative vector for each unit group at each moment. If the grouping of energy units remains largely unchanged over a period of time, and the operating conditions of energy units corresponding to two time periods are approximately similar, the similarity between unit groups within these two time periods can approximate the similarity between the operating conditions of the energy supply system. Conversely, if the similarity is low, the similarity between the operating parameters of each energy unit at various moments within the time period needs to be used to describe the similarity between the operating conditions of the energy supply system between these two time periods.

[0089] Therefore, in the clustering process of this step, based on the calculation results of the previous steps, the similarity between the representative vectors of the unit groups in different time periods and the similarity of the distribution of energy units are analyzed as an indicator for calculating clustering. At the same time, the similarity characteristics between the operating parameters of energy at each moment in different time periods are combined as a supplement to obtain the operating similarity value of the energy supply system between any two time periods. Based on this indicator, all time periods are classified and the target time period is finally selected.

[0090] The time sequence is divided into segments. Specifically, in this embodiment of the invention, the time sequence can be divided into segments of 10 minutes each, thereby obtaining all time segments.

[0091] Then, within different time periods, the similarity between the representative vectors of the unit groups and the similarity of the energy unit distribution are analyzed. Combined with the similarity characteristics between the operating parameters of the energy units at each time point, the operating similarity value of the energy supply system between any two time periods is determined.

[0092] Preferably, in one embodiment of the present invention, the method for obtaining similarity values ​​includes:

[0093] Please see Figure 3 The diagram illustrates a method flowchart for obtaining similarity values ​​according to an embodiment of the present invention, which includes the following steps:

[0094] Step S301: Analyze the similarity between the representative vectors of the unit groups and the similarity of the energy unit distribution in different time periods to obtain the state similarity between any two time periods.

[0095] Since the analysis of changes in energy unit grouping over a period of time is needed to characterize the similarity of the energy supply system between two time periods, the overlap of energy units in any two unit groups (the two unit groups come from different time periods) is first compared within any two time periods. The number of overlapping energy units is used as the overlap factor. Two unit groups with overlap factors greater than a preset overlap threshold are combined as a group. Thus, multiple pairs of combinations corresponding to these two time periods can be obtained.

[0096] Then, for any combination, the cosine similarity between the representative vectors of the two unit groups is calculated as a similarity parameter. The larger the similarity parameter, the higher the degree of similarity between the two unit groups in the combination. Based on this, the degree of similarity between the operating conditions of the energy supply system between the two time periods will also be higher.

[0097] Finally, within any two time periods, the corresponding overlap factors are weighted and fused based on the similarity parameters corresponding to the combinations: the overlap factors of all combinations are utilized... The function is normalized and used as the overlap weight. Then, the similarity parameter corresponding to each combination is multiplied by the overlap weight corresponding to the combination to obtain the weighted similarity parameter. The larger the weighted similarity parameter, the greater the similarity of the operating conditions between the two unit groups in that combination, and the higher the similarity of the distribution of energy unit types, indicating a higher state similarity between the energy supply systems. The normalized sum of the weighted similarity parameters of all combinations within these two time periods is used as the state similarity between these two time periods. The greater the state similarity, the more consistent the overall operating state of the energy supply system is within these two time periods.

[0098] It should be noted that the preset overlap threshold is set to half of the number of energy unit types, rounded up.

[0099] Step S302: Within any two time periods, analyze the similarity characteristics between the operating parameters of the energy unit at the same location and time, and determine the state similarity value between the two time periods.

[0100] Within any two time periods, at each same location and time, the cosine similarity between the operating state vectors of the same type of energy units is used as the similarity index between those energy units. The larger the similarity index, the more similar the operating states of the two energy units of the same type are. Then, the mean of the similarity indices between all types of energy units at each same location and time is used as the mean state similarity between the two time periods at each same location and time. The larger the mean state similarity, the higher the consistency of the operating states of the energy supply system between the two time periods at the same location and time.

[0101] Finally, the sum of the mean values ​​of state similarity at all the same locations and times in these two time periods is normalized and used as the state similarity value of the energy supply system in these two time periods. The larger this index is, the higher the degree of similarity of the energy supply system in the analysis of the similarity of the operating state at individual times.

[0102] It should be noted that, in the embodiments of the present invention, "same location time" specifically refers to: the first time in two time periods is the same location time, and similarly, the second time in two time periods is also the same location time.

[0103] Step S303: For any two time periods, determine the operating similarity value of the power supply system between the two time periods based on the state similarity value and the state similarity between the two time periods.

[0104] Based on the foregoing analysis, if the grouping of energy units remains largely unchanged over a period of time, and the operating conditions of the energy units in the two time periods are approximately similar, then the proportion of state similarity between the two time periods should be increased. Conversely, the similarity between the operating parameters of each energy unit at each moment in the time period should be used to describe the similarity between the operating conditions of the energy supply system in the two time periods, that is, to increase the proportion of state similarity value.

[0105] Therefore, for any two time periods, the normalized value of the state similarity between these two time periods is used as the first adjustment weight; the greater the state similarity, the greater the first adjustment weight. The value of the first adjustment weight after applying a negative correlation mapping is used as the second adjustment weight; the smaller the first adjustment weight, the greater the second adjustment weight. The negative correlation mapping here uses the formula... , where x represents the independent variable (in this embodiment of the invention, it represents the first adjustment weight).

[0106] Finally, the product of the state similarity between the two time periods and the first adjustment weight is added to the product of the state similarity value and the second adjustment weight. At this point, if the first adjustment weight is larger, the proportion of state similarity increases; conversely, if the first adjustment weight is smaller, the proportion of state similarity value increases. This is consistent with the aforementioned analytical logic. Therefore, the normalized sum is used as the operational similarity value of the energy supply system between the two time periods. This operational similarity value can accurately reflect the similarity of the operational state of the energy supply system between the two time periods, thus effectively improving the accuracy of subsequent cluster analysis and thereby improving the accuracy and representativeness of the target time period selection.

[0107] After obtaining the similarity values ​​of the energy supply system between any two time periods, all time periods can be classified based on this index, and target time periods can be selected in each category.

[0108] Preferably, in one embodiment of the present invention, the method for obtaining the target time period includes:

[0109] Cluster analysis is performed on all time periods using the K-means clustering algorithm and a preset K value to obtain all categories. At this point, the energy supply systems within each category have similar operating conditions. The distance metric is the negative correlation mapping between the operating similarity values ​​of time periods. This negative correlation mapping can be expressed using the formula... ,in, It represents an exponential function with the natural constant e as the base, and x represents the independent variable; at the same time, the optimal K value can be obtained based on the elbow method as the preset K value.

[0110] Then, the number of time periods to be filtered in each category can be determined: multiply the ratio of the number of time periods in each category to the total number of time periods in all categories by a preset number to obtain the number of time periods to be selected in each category.

[0111] Within each category, the more time periods that are similar to the operating conditions of a certain time period, the more likely that time period represents the general operating conditions in the current category. Conversely, the fewer similar time periods, the more likely that time period represents the special operating conditions in the current category. Therefore, within each category, the normalized value of the mean of the operating similarity between a certain time period and all other time periods is used as the generality of each time period. The greater the generality, the more likely that time period represents the general operating conditions in the current category. Thus, time periods with a generality greater than a preset generality threshold are classified as ordinary time periods, and time periods with a generality less than or equal to the preset generality threshold are classified as special time periods.

[0112] Next, within each category, the product of the proportion of the number of ordinary time slots and the corresponding selection quantity for each category is used as the target quantity for ordinary time slots; similarly, the product of the proportion of the number of special time slots and the corresponding selection quantity for each category is used as the target quantity for special time slots.

[0113] Finally, within each category, the target quantity for ordinary time periods is randomly selected from ordinary time periods, and the target quantity for special time periods is randomly selected from special time periods. All selected time periods are used as target time periods.

[0114] It should be noted that the preset quantity refers to the number of target time periods to be selected, which can be adjusted according to the implementation scenario and is not limited here; the preset general threshold can be set to 0.65, and the specific value can also be adjusted according to the implementation scenario and is not limited here; K-means clustering algorithm and elbow method are well known technologies, and the specific process will not be described in detail here.

[0115] Step S4: Conduct a stability assessment of the energy supply system within the target time period based on the operating parameters and arrangement of the energy units within the target time period.

[0116] Based on the aforementioned steps, all time periods can be accurately classified, and then representative target time periods can be selected to reflect the operating status of the energy supply system under various working conditions. Therefore, in this step, the system collapse boundary can be simulated based on the fluctuation of the operating parameters of the energy supply system within the selected target data segment, thereby conducting a stability assessment of the energy supply system within the target time period.

[0117] Preferably, in one embodiment of the present invention, a stability assessment of the energy supply system within the target time period is performed based on the operating parameters and arrangement of the energy units within the target time period, including:

[0118] Please see Figure 4 The diagram illustrates a stability evaluation method according to an embodiment of the present invention, which includes the following steps:

[0119] Step S401: Within each target time period, based on the arrangement of energy units in the energy supply system and the operating parameters, determine the current value, voltage value, and power value of the energy supply system at each moment.

[0120] Based on the electrical connection relationships of various devices in the energy supply system, a node admittance matrix is ​​established. The time-series data of the operating parameters of all energy units in the energy supply system within each target time period are associated with the node admittance matrix to obtain the power network model.

[0121] In the power network model, the time-series data of the operating parameters of the energy units under each target time period are processed according to the three-phase balanced system, the three-phase unbalanced system, Ohm's law and Joule's heat loss formula to obtain the current value, voltage value of the power supply system at each time, and the power loss value of each energy unit in the power supply system at each time.

[0122] Based on the power flow calculation method, that is, by solving the nodal power equations and the output power, load power and power loss value of each energy unit at each time step through the Newton-Raphson method, the power value of the power supply system at each time step is determined.

[0123] It should be noted that the power flow calculation method and nodal admittance matrix used in this step are well-known techniques, and the specific process will not be described in detail here.

[0124] Step S402: Based on the current, voltage, and power values ​​of the power supply system at each moment within the target time period, determine the limit power and critical voltage of the power supply system at each moment within the target time period.

[0125] By setting an initial load level, the system collapse boundary is dynamically simulated using the proportional load increment continuous power flow method.

[0126] First, the average load level (average power value) of the system during operation is taken as the initial load level of the system. This initial load level must satisfy the power flow equation constraints and be at a stable operating point. At the same time, the direction of load growth is set, that is, the load is increased by a preset proportion. .

[0127] Then, based on the proportional load increment continuous power flow method, the current, voltage, and power values ​​of the power supply system in each target time period are analyzed: through continuous adjustment Generate system operating trajectory, record voltage and power distribution at various load levels; calculate singular value decomposition of the Jacobian matrix (…). The system predicts the voltage and phase angle changes for the next load step, and then uses the Newton-Raphson method to correct the predicted values ​​to ensure power flow convergence. If convergence fails, the system is considered to have reached the collapse point, the calculation is terminated, and the load level corresponding to the collapse point is taken as the limit power of the power supply system at each moment within the target time period, and the voltage value corresponding to the collapse point is taken as the critical voltage of the power supply system at each moment within the target time period.

[0128] It should be noted that the proportional load increment continuous power flow method is a well-known technology, and the specific process will not be elaborated here.

[0129] Step S403: Within each target time period, analyze the deviation between the power value of the power supply system at each moment and the corresponding limit power, and determine the power angle stability reserve coefficient at each moment.

[0130] Within each target time period, the difference between the limit power and the power value at each moment is calculated, and the ratio of this difference to the power value at each moment is converted into a percentage, which is used as the power angle stability reserve coefficient at each moment.

[0131] It should be noted that the formula for calculating the power angle stability reserve coefficient is a well-known technique, and the specific formula is as follows:

[0132] in, Indicates the power angle stability reserve coefficient; Indicates the maximum power; This represents the power value at each moment.

[0133] Step S404: Within each target time period, analyze the deviation between the voltage value of the power supply system at each moment and the corresponding critical voltage, and determine the voltage stability reserve coefficient at each moment.

[0134] Within each target time period, the difference between the voltage value at each moment and the corresponding critical voltage is calculated, and the ratio of this difference to the voltage value at each moment is converted into a percentage, which is used as the voltage stability reserve coefficient at each moment.

[0135] It should be noted that the formula for calculating the voltage stability reserve factor is a well-known technique, and the specific formula is as follows:

[0136]

[0137] in, Indicates the voltage stability reserve factor; This represents the voltage value at each moment. This represents the critical voltage.

[0138] Step S405: Within each target time period, the stability of the power supply system is determined based on the power angle stability reserve coefficient and voltage stability reserve coefficient at all times.

[0139] Based on the requirements of the "Guidelines for the Safety and Stability of Power Systems", under normal operating conditions, the power angle stability reserve coefficient should be 15%~20%, and the voltage stability reserve coefficient should be 10%~15%.

[0140] Therefore, within a certain target time period, if the average power angle stability reserve coefficient at all times is within a preset first stability range, and the average voltage stability reserve coefficient at all times is within a preset second stability range, then the stability of the power supply system is considered good within that target time period; otherwise, it is considered poorly stable, and an early warning is issued. The preset first stability range is 15%~20%; the preset second stability range is 10%~15%.

[0141] In summary, the energy supply system consists of a large number of heterogeneous energy units working together. Therefore, this invention collects time-series data of the operating parameters of each energy unit in the energy supply system to comprehensively capture the dynamic operating characteristics of the energy units. Given that different energy units exhibit different fluctuations under different operating conditions, the energy units at each time point are initially clustered at the time level to obtain unit groups. Within each unit group, the numerical characteristics, similarities, and distribution density of the operating parameters of the energy units are analyzed to determine a representative vector for the unit group. This representative vector reflects the overall operating status of each unit group at each time point. To filter the target data for final monitoring, the time series can be divided into time periods, and representative time periods should be selected. Therefore, all time periods can be clustered at the time period level. When clustering time periods, this invention integrates the similarity of the overall operating status of the unit groups, the similarity of the unit group distribution, and the similarity characteristics of the operating parameters of the energy units. This allows for a combination of local and overall characteristics in the energy supply system, thereby more accurately measuring the similarity between two time periods and obtaining an index reflecting the similarity of the operating status of the energy supply system between two time periods—the operating similarity value. At this point, classifying time periods based on operational similarity values ​​allows for accurate segmentation of time periods with similar operational states. This effectively improves the accuracy and typicality of target time period selection, thereby enhancing the reliability of the final stability assessment of the power supply system. In summary, the embodiments of this invention can adapt to system characteristics under different operating conditions, effectively focus on key operating segments, and improve the utilization efficiency of monitoring resources and the accuracy of assessment results.

[0142] This invention also provides a stability assessment system for energy supply systems based on energy units. Please refer to [link / reference]. Figure 5 The diagram shows a system block diagram, including a data acquisition module 501, used to implement step S1 in the above method embodiment; a representative vector acquisition module 502, used to implement step S2 in the above method embodiment; a target time period filtering module 503, used to implement step S3 in the above method embodiment; and a stability evaluation module 504, used to implement step S4 in the above method embodiment.

[0143] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the energy unit-based energy supply system stability assessment system and the energy unit-based energy supply system stability assessment method embodiment provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiment, which will not be repeated here.

[0144] Please see Figure 6 This diagram illustrates a system architecture of an energy unit-based energy supply system stability assessment system according to an embodiment of the present invention. It includes a processor 600, a memory 601, a bus 602, and a communication interface 603. The processor 600, communication interface 603, and memory 601 are connected via the bus 602. The memory 601 may contain a high-speed random access memory, and the bus 602 may be an ISA bus, PCI bus, or EISA bus, etc. The processor 600 may be an integrated circuit chip with signal processing capabilities. The memory 601 stores at least one instruction, at least one program, code set, or instruction set. When the processor loads and executes the at least one instruction, at least one program, code set, or instruction set, it implements the steps in an energy unit-based energy supply system stability assessment method.

[0145] This invention also provides a computer-readable storage medium corresponding to the method provided in the foregoing embodiments. Please refer to [link to relevant documentation]. Figure 7 The storage medium shown is an optical disc, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it executes the methods provided in any of the foregoing embodiments.

[0146] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), read-only memory (ROM), and other optical and magnetic storage media, which will not be elaborated here.

[0147] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0148] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for evaluating the stability of an energy supply system based on energy units, characterized in that, The method includes: Acquire time-series data of various operating parameters for each energy unit in the energy supply system; Cluster all energy units at each time point to obtain unit groups; within each unit group, analyze the similarity of operating parameters, numerical characteristics, and distribution density of energy units to determine the representative vector of each unit group; The time series is divided into time periods. Within different time periods, the similarity of the representative vectors of the unit groups and the similarity of the distribution of energy units are analyzed. Combined with the similarity characteristics between the operating parameters of the energy units, the operating similarity value of the energy supply system between any two time periods is determined. This value is used to classify the time periods and thus select target time periods. The stability of the energy supply system is assessed based on the operating parameters and arrangement of the energy units within the target time period. The operating parameters of each energy unit at each time point are combined into an operating state vector. The method for obtaining the representative vector includes: In each unit group, the cosine similarity between the operating state vectors of any two energy units is used as the similarity factor between the two energy units. The standard deviation of all similarity factors corresponding to each energy unit is negatively correlated and normalized, and the value is used as the intra-group representativeness of each energy unit. Within each unit group, under each operating parameter, the operating parameter values ​​of the energy units are weighted and fused using the in-group representativeness of the energy units to obtain the operating characteristic values ​​of each unit group under each operating parameter. In each unit group, the Euclidean distance between the operating state vectors of any two energy units is used as a distance factor. The mean of the distance factors between all energy units is negatively correlated and the product of the mean and the number of energy units is normalized to obtain the density factor of the energy unit distribution in each unit group. The running feature values ​​and density factors of each unit group under all running parameters are used to form the representative vector of each unit group; The method for obtaining similarity values ​​includes: By analyzing the similarity between the representative vectors of the unit groups and the similarity of the energy unit distribution in different time periods, the state similarity between any two time periods can be obtained. Within any two time periods, at each same location and time, the cosine similarity between the operating state vectors of the same type of energy units is used as the similarity index between the energy units of that type. The mean of the similarity indices between all types of energy units at each same location and time is used as the mean state similarity between the two time periods at each same location and time. The normalized sum of the mean state similarity values ​​of the two time periods at all the same location and time is used as the state similarity value of the energy supply system within the two time periods. For any two time periods, the normalized value of the state similarity between the two time periods is used as the first adjustment weight, and the value of the first adjustment weight after negative correlation mapping is used as the second adjustment weight. The product of the state similarity between the two time periods and the first adjustment weight, and the product of the state similarity value and the second adjustment weight are added together. The normalized sum is used as the operating similarity value of the power supply system between the two time periods. The method for obtaining the state similarity includes: Within any two time periods, compare the energy units in any two unit groups within these two time periods, take the number of overlapping energy units as the overlap factor, and take two unit groups with an overlap factor greater than a preset overlap threshold as a combination, thereby obtaining all the combinations corresponding to these two time periods. For any combination, calculate the cosine similarity between the representative vectors of two unit groups as a similarity parameter; For any two time periods, the corresponding overlap factors are weighted and fused based on the similarity parameters corresponding to the combination to obtain the state similarity between the two time periods.

2. The method for evaluating the stability of an energy supply system based on energy units according to claim 1, characterized in that, The method for obtaining the target time period includes: Cluster analysis was performed on all time periods based on the K-means clustering algorithm and a preset K value to obtain all categories. The distance metric is the value after negative correlation mapping of the running similarity values ​​between time periods. Multiply the proportion of time slots in each category to the total number of time slots in all categories by a preset quantity to obtain the selection quantity for each category; Within each category, the mean of the operational similarity values ​​between each time period and all other time periods is normalized and used as the prevalence of each time period. Time periods with a prevalence greater than a preset prevalence threshold are classified as ordinary time periods, while time periods with a prevalence less than or equal to the preset prevalence threshold are classified as special time periods. In each category, the target number of ordinary time slots is the product of the proportion of ordinary time slots and the selection quantity corresponding to each category; the target number of special time slots is the product of the proportion of special time slots and the selection quantity corresponding to each category. Within each category, the target quantity based on the normal time period is randomly selected during the normal time period, and the target quantity based on the special time period is randomly selected during the special time period; Select the time period from all categories as the target time period.

3. The method for evaluating the stability of an energy supply system based on energy units according to claim 1, characterized in that, The stability assessment of the energy supply system within the target time period, based on the operating parameters and arrangement of energy units, includes: The types of operating parameters include at least current, voltage, impedance, output power, and load power; Based on the electrical connection relationship of various devices in the energy supply system, a node admittance matrix is ​​established. The time series data of the operating parameters of all energy units in the energy supply system within each target time period are associated with the node admittance matrix to obtain the power network model. In the power network model, the time series data of the operating parameters of the energy unit under each target time period are processed according to the three-phase balanced system, the three-phase unbalanced system, Ohm's law and Joule's heat loss formula to obtain the current value, voltage value of the power supply system at each time, and the power loss value of each energy unit in the power supply system at each time. Based on the power flow calculation method and the output power, load power and power loss value of each energy unit at each time, the power value of the energy supply system at each time is determined; Based on the current, voltage and power values ​​of the energy supply system at each moment within the target time period, determine the limit power and critical voltage of the energy supply system within the target time period. The ratio of the difference between the limit power of the power supply system within the target time period and the power value at each moment to the power value at each moment is used as the power angle stability reserve coefficient at each moment, where the power angle stability reserve coefficient is a percentage. The voltage stability reserve coefficient at each moment is the ratio of the difference between the voltage value at each moment and the corresponding critical voltage within the target time period to the voltage value at each moment. The voltage stability reserve coefficient is a percentage. If the average power angle stability reserve coefficient at all times is within a preset first stability range and the average voltage stability reserve coefficient at all times is within a preset second stability range within a certain target time period, then the stability of the power supply system is considered good within that target time period; otherwise, the stability is considered poor, and an early warning is issued.

4. The method for evaluating the stability of an energy supply system based on energy units according to claim 3, characterized in that, The determination of the limiting power and critical voltage of the energy supply system within the target time period, based on the current, voltage, and power values ​​of the energy supply system at each moment within the target time period, includes: Set the initial load level and increase the load according to a preset ratio; The current, voltage, and power values ​​of the power supply system within each target time period are analyzed based on the proportional load increment continuous power flow method. The convergence of the power flow solution is checked. If the convergence is not achieved, the collapse point is considered to have been reached, the calculation is terminated, and the load level corresponding to the collapse point is taken as the limit power of the power supply system within the target time period. The voltage value corresponding to the collapse point is taken as the critical voltage of the power supply system within the target time period.

5. The method for evaluating the stability of an energy supply system based on energy units according to claim 1, characterized in that, The method for obtaining the unit group includes: At each time step, the data points corresponding to the operating state vectors of all energy units are clustered based on the DBSCAN clustering algorithm to obtain all clusters, and the energy units in each cluster are taken as a unit group.

6. A functional system stability assessment system based on energy units, characterized in that, The system includes a processor and a memory, the memory storing at least one instruction, at least one program, code set, or instruction set, wherein when the at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor, it implements the steps of the energy unit-based energy supply system stability assessment method as described in any one of claims 1-5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the energy unit-based energy supply system stability assessment method as described in any one of claims 1 to 5.

Images