Energy supply system stability evaluation method and system based on energy unit and medium

Abstract

The invention relates to the technical field of energy supply system stability evaluation, in particular to an energy supply system stability evaluation method and system based on an energy unit and a medium. Acquiring operation parameter time sequence data of each energy unit; in view of fluctuation differences among energy units under different working conditions, time-level clustering is adopted to divide the energy units into unit groups, and representative vectors of the unit groups are extracted to reflect the overall operation condition of the unit groups. Then, time sequence data is divided into time periods, comprehensive clustering is carried out at the time period level, and in combination with the unit group operation condition, the distribution characteristics and the energy unit parameter similarity, an operation similarity value is constructed to measure the similarity of the system operation states in the time periods; the time periods are classified based on the operation similarity values, the time periods in the similar operation states can be accurately divided, the accuracy and typicality of target time period screening are effectively improved, and then the reliability of system stability evaluation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy supply system stability assessment, and in particular to an energy supply system stability assessment method, system and medium based on an energy unit. Background Art

[0002] With the continued 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, energy supply systems have complex dynamic characteristics. Factors such as environmental conditions, load demand fluctuations, equipment aging, and communication delays can lead to unstable system operation and even safety incidents. Therefore, it is necessary to evaluate the stability of energy supply systems.

[0003] When evaluating the stability of an energy supply system, existing technologies usually cluster time periods based on the operating parameters of energy units to select typical time periods as monitoring targets. However, since many energy sources are connected to the energy supply system, and under different operating conditions, the volatility and intermittency between different energy sources will change. Therefore, during the screening process, if the time periods are directly clustered based on the operating parameters of the energy units, it will lead to inaccurate classification, which will in turn affect the accuracy of the screening of monitoring targets and reduce the credibility of the energy supply system stability assessment. Summary of the Invention

[0004] In order to solve the technical problem that when a large number of energy sources are connected to the energy supply system, the volatility and intermittency between different energy sources will change under different operating conditions. Therefore, during the screening process, if the time periods are directly clustered according to the operating parameters of the energy units, the classification will be inaccurate, which will affect the screening accuracy of the monitoring targets and reduce the credibility of the energy supply system stability assessment. The purpose of the present invention is to provide an energy supply system stability assessment method, system and medium based on energy units. The technical solutions adopted are as follows: A method for evaluating the stability of an energy supply system based on an energy unit, comprising: Obtain time series data of various operating parameters of each energy unit in the energy supply system; Cluster all energy units at each moment to obtain unit groups. Within each unit group, analyze the similarity between the operating parameters of the energy units, the numerical characteristics, and the distribution density of the energy units to determine the representative vector of each unit group. 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. Combined with the similarity characteristics between the operating parameters of the energy units, determine the operating similarity value of the energy supply system between any two time periods. This is used to classify the time periods and select the target time period. The stability of the energy supply system within the target time period is evaluated based on the operating parameters and arrangement of the energy units within the target time period.

[0005] Furthermore, the method for obtaining the representative vector includes: All the operating parameters of each energy unit at each moment form an operating state vector; 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 to form the value as the representativeness of each energy unit within the group. In each unit group, under each operating parameter, the operating parameter values of the energy units are weighted and fused using the intra-group representativeness of the energy units to obtain the operating characteristic value 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 the distance factor. The average value of the distance factor between all energy units is negatively correlated and normalized by the product of the number of energy units to obtain the density factor of the energy unit distribution in each unit group. The operating characteristic values and density factors of each unit group under all operating parameters are combined to form the representative vector of each unit group.

[0006] Furthermore, the method for obtaining the operation similarity value includes: Analyze the similarity between the representative vectors of the unit groups in different time periods and the similarity of the energy unit distribution in different time periods to obtain the state similarity between any two time periods; In any two time periods, at each identical position 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, and the mean of the similarity indexes between all types of energy units at each identical position and time is used as the mean state similarity between the two time periods at each identical position and time. The sum of the mean state similarities at all identical position and time in the two time periods is normalized and used as the state similarity value of the energy supply system in 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 after negative correlation mapping of the first adjustment weight is used as the second adjustment weight; 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, and the resulting sum is normalized and used as the operation similarity value of the energy supply system between the two time periods.

[0007] Furthermore, the method for obtaining the state similarity includes: In any two time periods, the energy units in any two unit groups in the two time periods are compared, the number of energy unit overlaps is used as the overlap factor, and the two unit groups whose overlap factor is greater than the preset overlap threshold are regarded as a combination, thereby obtaining all the combinations corresponding to the two time periods; For any combination, calculate the cosine similarity between the representative vectors of the two unit groups as the similarity parameter; In any two time periods, the corresponding coincidence factors are weighted and fused based on the similarity parameters corresponding to the combination to obtain the state similarity between the two time periods.

[0008] Furthermore, the method for obtaining the target time period includes: Based on the K-means clustering algorithm and the preset K value, cluster analysis is performed on all time periods to obtain all categories. The distance metric is the value after negative correlation mapping of the running similarity values between time periods. Multiply the ratio of the number of time periods in each category to the total number of time periods in all categories by the preset number to obtain the number of selections corresponding to each category; In each category, the average of the running similarity values between each time period and all other time periods is normalized to form the prevalence of each time period. Time periods with a prevalence greater than a preset prevalence threshold are considered ordinary time periods, and time periods with a prevalence less than or equal to the preset prevalence threshold are considered special time periods. In each category, the target quantity for the ordinary period is the product of the proportion of the quantity in the ordinary period and the selected quantity corresponding to each category. The target quantity for the special period is the product of the proportion of the quantity in the special period and the selected quantity corresponding to each category. In each category, the number of targets based on the normal period was randomly selected during the normal period, and the number of targets based on the special period was randomly selected during the special period; The time period selected from all categories is used as the target time period.

[0009] Furthermore, the 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 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, and the operating parameter time series data of all energy units in the energy supply system in each target time period are associated with the node admittance matrix to obtain a power network model; In the power network model, the operating parameter time series data of the energy unit in each target time period is processed according to the three-phase balanced system, the three-phase unbalanced system, Ohm's law, and the Joule heat loss formula to obtain the current value and voltage value of the energy supply system at each moment, as well as the power loss value of each energy unit in the energy supply system at each moment; Determine the power value of the energy supply system at each moment based on the power flow calculation method and the output power, load power and power loss values of each energy unit at each moment; Determine the limit power and critical voltage of the energy supply system within the target time period based on the current value, voltage value, and power value of the energy supply system at each moment within the target time period; The ratio of the difference between the energy supply system's limit power and the power value at each moment during the target time period 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 expressed as a percentage. The ratio of the difference between the voltage value at each moment in the target time period and the corresponding critical voltage to the voltage value at each moment is used as the voltage stability reserve coefficient at each moment, where the voltage stability reserve coefficient is a percentage; During a certain target time period, if the average value of the power angle stability reserve coefficient at all times is within the preset first stability range, and the average value of the voltage stability reserve coefficient at all times is within the preset second stability range, then the stability of the energy supply system during the target time period is considered to be good; otherwise, the stability is considered to be poor and an early warning is issued.

[0010] Furthermore, determining the limit power and critical voltage of the energy supply system within the target time period based on the current value, voltage value, and power value 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 the preset ratio; Based on the proportional load increment continuous power flow method, the current, voltage and power values of the energy supply system in each target time period are analyzed to check whether the power flow solution converges. If not, it is considered that the collapse point has been reached and the calculation is terminated. The load level corresponding to the collapse point is used as the limit power of the energy supply system in the target time period, and the voltage value corresponding to the collapse point is used as the critical voltage of the energy supply system in the target time period.

[0011] Furthermore, the method for obtaining the unit group includes: At each moment, cluster analysis is performed on the data points corresponding to the operating state vectors of all energy units based on the DBSCAN clustering algorithm to obtain all clusters, and the energy units in each cluster are regarded as a unit group.

[0012] A system for evaluating the stability of an energy supply system based on an energy unit includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and when the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor, the steps of a method for evaluating the stability of an energy supply system based on an energy unit are implemented.

[0013] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for evaluating the stability of an energy supply system based on an energy unit.

[0014] The present invention has the following beneficial effects: The energy supply system is composed of a large number of heterogeneous energy units that work together. Therefore, the present invention collects the 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 unit. Given that different energy units will produce different fluctuation characteristics under different operating conditions, the energy units at each moment are initially clustered based on the moment level to obtain unit groups. The numerical characteristics, similarity, and distribution density of the operating parameters of the energy units in each unit group are analyzed to determine the representative vector of the unit group. The representative vector is used to reflect the characteristics of the overall operating status of each unit group at each moment. In order to screen 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 the time periods, the present invention integrates the similarity of the overall operating status of the unit group, the similarity of the unit group distribution, and the similarity of the operating parameters of the energy units. This can combine the local characteristics with the overall characteristics of the energy supply system, thereby more accurately measuring the similarity between two time periods and obtaining an indicator reflecting the similarity of the operating status of the energy supply system between the two time periods - the operating similarity value. Categorizing time periods based on operational similarity values accurately divides time periods with similar operating states, effectively improving the accuracy and representativeness of target time period screening, and thus enhancing the credibility of the final stability assessment of the energy supply system. In summary, this solution adapts to system characteristics under different operating conditions, effectively focusing on critical operating periods, and improving the utilization efficiency of monitoring resources and the accuracy of assessment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flow chart of a method for evaluating the stability of an energy supply system based on an energy unit provided by one embodiment of the present invention; Figure 2 A flowchart of a method for obtaining a representative vector provided by one embodiment of the present invention; Figure 3 A flow chart of a method for obtaining a running similarity value provided by one embodiment of the present invention; Figure 4 A flow chart of a stability assessment method provided by one embodiment of the present invention; Figure 5 A system block diagram of an energy supply system stability assessment system based on energy units provided by one embodiment of the present invention; Figure 6 A schematic diagram of the system structure of an energy supply system stability assessment system based on energy units provided by one embodiment of the present invention; Figure 7 A schematic diagram of a computer-readable storage medium provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the energy supply system stability assessment method, system and medium based on the energy unit proposed by the present invention, its specific implementation method, structure, features and effects. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics of one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, 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 belongs.

[0019] The following describes in detail a method, system and medium for evaluating the stability of an energy supply system based on an energy unit provided by the present invention with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a method flow chart of a method for evaluating the stability of an energy supply system based on an energy unit provided by an embodiment of the present invention, the method comprising the following steps: Step S1: Acquire time series data of various operating parameters of each energy unit in the energy supply system.

[0021] With the global energy transition and technological advancements, the concept of energy supply systems has gradually expanded in recent years from traditional, isolated generators to integrated systems encompassing multiple energy sources and storage systems. Stability is a crucial concept in energy supply systems. In this embodiment of the present invention, time-series data on the operating parameters of the energy supply system as a whole and for each energy unit within it is first acquired. These operating parameters should include at least current, voltage, impedance, output power, and load power.

[0022] Specifically, sensors can be installed on each energy unit (such as photovoltaic inverters, wind turbines, and traditional generator sets) to collect various operating parameters, and then normalize them based on the rated values of each operating parameter (using the ratio of the actual parameter value to the rated value as the normalized value) to obtain time series data for each operating parameter.

[0023] It should be noted that the collection duration of time series data 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 itself according to the implementation scenario and is not limited here.

[0024] Step S2: Cluster all energy units at each moment to obtain unit groups; in each unit group, analyze the similarity between the operating parameters of the energy units, the numerical characteristics and the distribution density of the energy units to determine the representative vector of each unit group.

[0025] In an embodiment of the present invention, when evaluating the stability of an energy supply system, it is necessary to determine the operating status of the energy supply system. In the energy supply system, due to the influence of environmental conditions, load demand fluctuations, equipment aging, communication delays and other factors, the power supply of each individual energy unit to the outside will fluctuate from time to time, and different energy units react differently to the influence of these factors. Even at the same moment, different energy units will have different working conditions, which makes it very complicated to describe the working conditions of an energy supply system at a certain moment. Therefore, cluster analysis can be performed on the energy units at each moment, and the numerous energy units can be divided into a limited number of unit groups to reduce the difficulty of subsequent analysis while retaining key feature information. Then, at the unit group level, the similarity of the operating parameters of the energy units, the numerical characteristics and the distribution density of the energy units are analyzed to form a macroscopic understanding of the operating status of the unit group, and a representative vector of the unit group is obtained to reflect the operating conditions of this group of energy units.

[0026] First, at each moment, all energy units in the energy supply system are clustered and analyzed to obtain unit groups.

[0027] Preferably, in one embodiment of the present invention, a method for obtaining a unit group includes: All the operating parameters of each energy unit at each moment are combined into an operating state vector. At this time, each energy unit has an operating state vector, and the order of the operating parameters in the operating state vectors of all energy units should be consistent.

[0028] Then, at each moment, cluster analysis is performed on the data points corresponding to the state vectors of all energy units based on the DBSCAN clustering algorithm to obtain all clusters, and the energy units in each cluster are regarded as a unit group.

[0029] It should be noted that, in this embodiment of the present invention, the neighborhood radius can be Function acquisition, where Value ( ,in is the number of types of operating parameters); the 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 also 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 is not described here.

[0030] The energy units at each moment are clustered to obtain unit groups, that is, the operating status of many energy units at each moment is distinguished. Next, in each unit group, the similarity characteristics and numerical characteristics of the operating parameters of the energy units can be analyzed, and combined with the distribution density of the energy units, the representative vector of each unit group can be determined 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.

[0031] Preferably, in one embodiment of the present invention, the method for obtaining the representative vector includes: See also Figure 2 , which shows a flow chart of a method for obtaining a representative vector in one embodiment of the present invention, the method includes the following steps: Step S201: In each unit group, the similarities and numerical characteristics between the operating state vectors of the energy units are analyzed to determine the operating characteristic value of each unit group under each operating parameter.

[0032] Based on the above 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, and 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.

[0033] At this time, in the unit group, each energy unit has a similarity factor with each of the remaining energy units. If the size of all similarity factors corresponding to a certain energy unit is more consistent, then the representativeness of the energy unit in the unit group will be stronger, and the reference value will be higher when constructing the representative vector of the unit group in the future. Therefore, the standard deviation of all similarity factors corresponding to each energy unit is calculated. The smaller the standard deviation, the higher the degree of consistency of all similarity factors of the energy unit, and its representativeness should be improved. Therefore, the standard deviation is negatively correlated and normalized to achieve logical relationship correction, thereby obtaining the in-group representativeness of each energy unit. The greater the in-group representativeness, the higher the reference value of the energy unit in the subsequent process. The negative correlation mapping and normalization here can be performed using the formula ,in, It represents the exponential function with the natural constant e as the base, and x represents the independent variable.

[0034] Finally, in each unit group, under each operating parameter, the energy unit's group representativeness is used to perform weighted fusion on the energy unit's operating parameter values: for any energy unit, the energy unit's group representativeness is used to After the function is normalized, it 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 numerical value of the energy unit under the operating parameter, and the higher the representativeness. In this way, the weighted parameter of each energy unit in the unit group under each operating parameter can be obtained. 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 characterizes the overall operating characteristics of all energy units in the entire unit group under each operating parameter. Normalization is a technical means well known to those skilled in the art. The choice of normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0035] Step S202: In each unit group, the distribution density of the energy units is analyzed to determine the density factor of the energy unit distribution in each unit group.

[0036] The Euclidean distance can reflect the distribution of the operating state vector of the energy unit in space, which helps to understand the distribution density of the operating state of the energy units in the unit group; and the number of energy units in the unit group can reflect the quantity distribution density of the energy units.

[0037] Therefore, in each unit group, the Euclidean distance between the operating state vectors of any two energy units is used as the distance factor. The smaller the distance factor, the closer the spatial distance between the two, and the higher the distribution density of the energy unit's operating state. At the same time, the greater the number of energy units in the unit group, the denser the distribution of the number of energy units. Therefore, the mean of the distance factors between all energy units is negatively correlated to achieve logical relationship correction, and normalized by the product of the number of energy units to obtain the density factor of the energy unit distribution in each unit group. At this time, the larger the density factor, the better the clustering of the energy units in the unit group, and the higher the consistency of the operating conditions of the many energy units in the unit group.

[0038] It should be noted that the negative correlation mapping process here can be processed using the formula ,in, represents an exponential function with the natural constant e as the base, and x represents an independent variable; normalization is a technical means well known to those skilled in the art, and the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0039] Step S203: In each unit group, the density factor and the operating characteristic value under each operating parameter are integrated to obtain a representative vector of each unit group.

[0040] 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 energy unit distribution in each unit group is obtained. This value can effectively make up for the situation where some operating conditions are lost due to fitting the operating characteristics of the unit group from the operating conditions of many energy units.

[0041] Therefore, here, the operating characteristic values and density factors of each unit group under all operating parameters can be combined to form the representative vector of each unit group. The order of the parameters in the representative vectors of all unit groups should be kept 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.

[0042] 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 screen the target time period.

[0043] In an embodiment of the present invention, it is necessary to screen out representative time periods in the time series, so as to evaluate the stability of the energy supply system in the screened time periods, so the accuracy and representativeness of the time period screening are very important. First, the time series can be divided to obtain time periods. Before screening the target time periods, all time periods can be clustered and analyzed, so that time periods with similar working condition characteristics are clustered into one category. Then, screening the target time periods in each category of time periods can ensure that time periods with each working condition characteristic can be selected, avoiding the possibility of missing or ignoring certain working conditions. The above steps cluster the energy units at each moment and obtain the representative vector of each unit group at each moment. If the grouping of energy units does not change substantially within a period of time, and the operating conditions of the energy units corresponding to the two time periods are also approximately similar, the similarity between the unit groups in the two time periods can be used to approximately replace the similarity between the operating conditions of the energy supply system. On the contrary, if the similarity is low, it is necessary to describe the similarity between the operating parameters of each energy unit at each moment in the time period to describe the similarity between the operating conditions of the energy supply system between the two time periods.

[0044] 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 energy unit distribution are analyzed as an indicator for calculating the clustering basis; at the same time, combined with the similarity characteristics between the operating parameters of the energy at each moment in different time periods as a supplement, the operating similarity value of the energy supply system between any two time periods is obtained, and then all time periods are classified based on this indicator, and finally the target time period is screened out.

[0045] The time series is divided. Specifically, in this embodiment of the present invention, the time series may be divided into time periods of 10 minutes, thereby obtaining all time periods.

[0046] Then, in different time periods, the similarities between the representative vectors of the unit groups and the similarities in the distribution of energy units are analyzed, and combined with the similar characteristics between the operating parameters of the energy units at each moment, the operating similarity value of the energy supply system between any two time periods is determined.

[0047] Preferably, in one embodiment of the present invention, the method for obtaining the running similarity value includes: See also Figure 3 , which shows a flow chart of a method for obtaining a similarity value in one embodiment of the present invention, the method comprising the following steps: 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.

[0048] In view of the need to analyze the changes in energy unit grouping within a period of time to characterize the similarity of the state of the energy supply system between two time periods, in any two time periods, we first compare the overlap of energy units in any two unit groups in the two time periods (the two unit groups come from different time periods), and use the number of overlaps of energy units as the overlap factor. The two unit groups with an overlap factor greater than the preset overlap threshold are regarded as a combination. In this way, multiple pairs of combinations corresponding to the two time periods can be obtained.

[0049] Then, for any combination, the cosine similarity between the representative vectors of the two unit groups is calculated as the similarity parameter. The larger the similarity parameter is, the higher the degree of similarity of the operating conditions between the two unit groups in the combination is. On this basis, the degree of similarity of the operating conditions of the energy supply system between the two time periods will also be higher.

[0050] Finally, within any two time periods, the corresponding coincidence factors are weighted and fused based on the similarity parameters of the combinations: the coincidence factors of all combinations are combined using The function is normalized to serve as the coincidence weight. The similarity parameter corresponding to each combination is then multiplied by the corresponding coincidence weight to obtain a weighted similarity parameter. A larger weighted similarity parameter indicates greater similarity in the operating conditions between the two unit groups in the combination. Furthermore, a higher similarity in the distribution of energy unit types indicates a higher state similarity between the energy supply systems. The sum of the weighted similarity parameters for all combinations within the two time periods is normalized to serve as the state similarity between the two time periods. A higher state similarity indicates a more consistent overall operating state of the energy supply systems within the two time periods.

[0051] It should be noted that the preset overlap threshold is set to half of the number of energy unit types rounded up.

[0052] Step S302: within any two time periods, analyzing the similarity characteristics between the operating parameters of each energy unit at the same position and time, and determining the state similarity value between the two time periods.

[0053] In any two time periods, at each same position moment, the cosine similarity between the operating state vectors of energy units of the same type is used as the similarity index between the energy units of the same type. 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 indexes between all types of energy units at each same position moment is used as the mean state similarity between the two time periods at each same position moment. The larger the mean state similarity, the higher the consistency of the operating state of the energy supply system in the two time periods at the same position moment.

[0054] Finally, the normalized sum of the state similarity means at all the same position moments in the two time periods is used as the state similarity value of the energy supply system in the two time periods. The larger the index is, the higher the similarity of the energy supply system in the two time periods is.

[0055] It should be noted that the same position moment in the embodiment of the present invention specifically refers to: the first moment in two time periods is the same position moment, and similarly, the second moment in two time periods is also the same position moment.

[0056] Step S303: for any two time periods, based on the state similarity value between the two time periods and the state similarity between the two time periods, determine the operation similarity value of the energy supply system between the two time periods.

[0057] Based on the above analysis, it can be seen that if the grouping of energy units basically does not change within a period of time, and the operating conditions of the energy units corresponding to the two time periods are also approximately similar, the proportion of state similarity between the two time periods should be increased; otherwise, it is necessary to use the similarity between the operating parameters of each energy unit at each moment in the time period to describe the similarity between the operating conditions of the energy supply system between the two time periods, that is, to increase the proportion of state similarity values.

[0058] Therefore, for any two time periods, the normalized value of the state similarity between the two time periods is used as the first adjustment weight. The greater the state similarity, the greater the first adjustment weight. The value after negative correlation mapping of the first adjustment weight is used as the second adjustment weight. The smaller the first adjustment weight, the greater the second adjustment weight. The negative correlation mapping here adopts the formula , where x represents an independent variable (in this embodiment of the present invention, it represents the first adjustment weight).

[0059] 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 time, if the first adjustment weight is larger, the proportion of state similarity is increased; conversely, if the first adjustment weight is smaller, the proportion of state similarity value is increased, which is in line with the above analysis logic. Therefore, the normalized value of the obtained sum is used as the operation similarity value of the energy supply system between the two time periods. The operation similarity value at this time can accurately reflect the similarity of the operation state of the energy supply system between the two time periods, thereby effectively improving the accuracy of subsequent clustering analysis, thereby improving the accuracy and representativeness of the target time period screening.

[0060] After obtaining the operation similarity value of the energy supply system between any two time periods, all time periods can be classified based on the index, and the target time period can be filtered in each category.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining the target time period includes: Based on the K-means clustering algorithm and the preset K value, cluster analysis is performed on all time periods to obtain all categories. At this time, the energy supply system has similar operating conditions in the time period of each category. The distance metric is the value after the negative correlation mapping of the operating similarity values between time periods. The negative correlation mapping here can be used using the formula ,in, 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 as the preset K value based on the elbow method.

[0062] The number of time periods to be filtered in each category can then be determined: the ratio of the number of time periods in each category to the total number of time periods in all categories is multiplied by the preset number to obtain the selected number of time periods to be filtered in each category.

[0063] In each category, the more time periods that are similar to the operating conditions of a certain time period, the more it means that the time period represents the general operating conditions in the current category. Otherwise, it means that the time period represents the special operating conditions in the current category. Therefore, in each category, the normalized value of the mean of the operating similarity values between a certain time period and all other time periods is used as the prevalence of each time period. The greater the prevalence, the more likely the time period represents the general operating conditions in the current category. Therefore, the time period with a prevalence greater than the preset prevalence threshold is regarded as an ordinary time period, and the time period with a prevalence less than or equal to the preset prevalence threshold is regarded as a special time period.

[0064] Then, in each category, the product of the proportion of the number of ordinary time periods and the corresponding selected number of each category is used as the target number of ordinary time periods; similarly, the product of the proportion of the number of special time periods and the corresponding selected number of each category is used as the target number of special time periods.

[0065] Finally, in each category, the target number based on the ordinary time period is randomly selected in the ordinary time period, and the target number based on the special time period is randomly selected in the special time period, and all selected time periods are used as target time periods.

[0066] It should be noted that the preset number here refers to the number of target time periods that need to be selected in the end, 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; the K-means clustering algorithm and the elbow method are both well-known technologies, and the specific process will not be repeated here.

[0067] Step S4: Evaluate the stability 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.

[0068] Based on the above 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 further simulated according to the fluctuation of the operating parameters of the energy supply system in the screened target data segment, thereby evaluating the stability of the energy supply system in the target time period.

[0069] Preferably, in one embodiment of the present invention, based on the operating parameters and arrangement of the energy units in the target time period, the stability assessment of the energy supply system in the target time period includes: See also Figure 4 , which shows a flow chart of a method for stability evaluation in one embodiment of the present invention, the method includes the following steps: Step S401: In each target time period, based on the arrangement of energy units in the energy supply system and operating parameters, determine the current value, voltage value and power value of the energy supply system at each moment.

[0070] Based on the electrical connection relationship of various devices in the energy supply system, a node admittance matrix is established. The operating parameter time series data of all energy units in the energy supply system in each target time period are associated with the node admittance matrix to obtain the power network model.

[0071] In the power network model, the operating parameter time series data of the energy unit in each target time period are processed according to the three-phase balanced system, three-phase unbalanced system, Ohm's law and Joule's heat loss formula to obtain the current value and voltage value of the energy supply system at each moment, as well as the power loss value of each energy unit in the energy supply system at each moment.

[0072] Based on the power flow calculation method, that is, the Newton-Raphson method is used to solve the node power equation and the output power, load power and power loss values of each energy unit at each moment, to determine the power value of the energy supply system at each moment.

[0073] It should be noted that the power flow calculation method and node admittance matrix in this step are all well-known technologies, and the specific process will not be described here in detail.

[0074] Step S402: determining the power limit and critical voltage of the energy supply system at each moment in the target time period based on the current value, voltage value, and power value of the energy supply system at each moment in the target time period.

[0075] By setting the initial load level and using the proportional load increment continuous power flow method, the system collapse boundary is dynamically simulated.

[0076] First, the average value of the load level (the average value of the power value) during the operation of the energy supply system is used as the initial load level of the system. The initial load level must meet the constraints of the power flow equation and be at a stable operating point. At the same time, the load growth direction is set, that is, the load is increased by a preset ratio. .

[0077] Then, based on the proportional load increment continuous power flow method, the current value, voltage value and power value of the energy supply system in each target time period are analyzed: by continuously adjusting Generate system operation trajectory and record voltage and power distribution under various load levels; calculate the singular value decomposition of the Jacobian matrix ( ) to predict the direction of voltage and phase angle change for the next load step. The Newton-Raphson method is then used to correct the predicted values to ensure power flow convergence. If convergence does not occur, the calculation is terminated, assuming the breakdown point has been reached. The load level corresponding to the breakdown point is used as the energy supply system's limit power at each moment in the target time period, and the voltage value corresponding to the breakdown point is used as the energy supply system's critical voltage at each moment in the target time period.

[0078] 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 described here in detail.

[0079] Step S403: In each target time period, the deviation between the power value of the energy supply system at each moment and the corresponding limit power is analyzed to determine the power angle stability reserve coefficient at each moment.

[0080] In each target time period, the difference between the limit power and the power value at each moment is calculated, and the ratio of the difference to the power value at each moment is converted into a percentage as the power angle stability reserve coefficient at each moment.

[0081] It should be noted that the calculation formula for the power angle stability reserve coefficient is a well-known technology, and the specific formula is:

[0082] in, Indicates the power angle stability reserve coefficient; Indicates the limit power; Indicates the power value at each moment.

[0083] Step S404: within each target time period, analyzing the deviation between the voltage value of the energy supply system at each moment and the corresponding critical voltage, and determining the voltage stability reserve coefficient at each moment.

[0084] In each target time period, the difference between the voltage value at each moment in the target time period and the corresponding critical voltage is calculated, and the ratio of the difference to the voltage value at each moment is converted into a percentage as the voltage stability reserve coefficient at each moment.

[0085] It should be noted that the calculation formula for the voltage stability reserve coefficient is a well-known technology, and the specific formula is:

[0086] in, Indicates the voltage stability reserve factor; Indicates the voltage value at each moment; Indicates critical voltage.

[0087] Step S405: In each target time period, the stability of the energy supply system is determined based on the power angle stability reserve coefficient and the voltage stability reserve coefficient at all times.

[0088] Based on the requirements of the "Guidelines for 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%.

[0089] Therefore, within a target time period, if the average value of the power angle stability reserve coefficient at all times is within the preset first stability range, and the average value of the voltage stability reserve coefficient at all times is within the preset second stability range, the energy supply system is considered to have good stability during that target time period. Otherwise, the stability is considered poor and an early warning is issued. The preset first stability range is 15% to 20%, and the preset second stability range is 10% to 15%.

[0090] In summary, the energy supply system is composed of a large number of heterogeneous energy units that work together. Therefore, the embodiment of the present invention collects the time series data of the operating parameters of each energy unit in the energy supply system to fully capture the dynamic operating characteristics of the energy unit. Given that different energy units will produce different fluctuation characteristics under different operating conditions, the energy units at each moment are initially clustered based on the moment level to obtain unit groups. The numerical characteristics, similarity, and distribution density of the operating parameters of the energy units are analyzed in each unit group to determine the representative vector of the unit group. The representative vector is used to reflect the characteristics of the overall operating status of each unit group at each moment. In order to screen the target data for 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 the time periods, the present invention integrates the similarity of the overall operating status of the unit group, the similarity of the unit group distribution, and the similar characteristics of the operating parameters of the energy units. This can combine the local characteristics with the overall characteristics of the energy supply system, thereby more accurately measuring the similarity between two time periods and obtaining an indicator reflecting the similarity of the operating status of the energy supply system between the two time periods - the operating similarity value. At this point, classifying time periods based on operational similarity values accurately divides time periods with similar operating states, thereby effectively improving the accuracy and typicality of screening for target time periods, and thereby increasing the credibility of the final stability assessment of the energy supply system. In summary, the embodiments of the present invention can adapt to system characteristics under different operating conditions, effectively focusing on critical operating time periods, and improving the utilization efficiency of monitoring resources and the accuracy of assessment results.

[0091] The embodiment of the present invention also provides an energy supply system stability assessment system based on energy units, see Figure 5 , which 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 screening 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.

[0092] It should be noted that the system provided in the above embodiment is merely an example of the division of the above functional modules. In actual applications, the above functions can be distributed and completed by 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 supply system stability assessment system based on energy units and the energy supply system stability assessment method based on energy units provided in the above embodiment are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0093] See also Figure 6 , which shows a system structure diagram of an energy supply system stability assessment system based on an energy unit provided by an embodiment of the present invention, including a processor 600, a memory 601, a bus 602 and a communication interface 603, wherein the processor 600, the communication interface 603 and the memory 601 are connected via the bus 602; wherein the memory 601 may include a high-speed random access memory, the bus 602 may be an ISA bus, a PCI bus or an EISA bus, etc., and 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, a code set or an instruction set, and when the at least one instruction, at least one program, a code set or an instruction set is loaded and executed by the processor, the steps in a method for assessing the stability of an energy supply system based on an energy unit are implemented.

[0094] The embodiment of the present invention also provides a computer-readable storage medium corresponding to the method provided in the above embodiment, see Figure 7 , the storage medium shown is a CD, on which a computer program (ie, a program product) is stored. When the computer program is run by a processor, it will execute the method provided by any of the aforementioned embodiments.

[0095] 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 are not listed here one by one.

[0096] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for evaluating the stability of an energy supply system based on an energy unit, characterized in that: The method comprises: Obtain time series data of various operating parameters of each energy unit in the energy supply system; Cluster all energy units at each moment to obtain unit groups. Within each unit group, analyze the similarity between the operating parameters of the energy units, the numerical characteristics, and the distribution density of the energy units to determine the representative vector of each unit group. 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. Combined with the similarity characteristics between the operating parameters of the energy units, determine the operating similarity value of the energy supply system between any two time periods. This is used to classify the time periods and select the target time period. Evaluate the stability 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; All the operating parameters of each energy unit at each moment are combined into an operating state vector.

2. The energy supply system stability assessment method based on energy units according to claim 1 is characterized in that: 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 to form the value as the representativeness of each energy unit within the group. In each unit group, under each operating parameter, the operating parameter values of the energy units are weighted and fused using the intra-group representativeness of the energy units to obtain the operating characteristic value 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 the distance factor. The average value of the distance factor between all energy units is negatively correlated and normalized by the product of the number of energy units to obtain the density factor of the energy unit distribution in each unit group. The operating characteristic values and density factors of each unit group under all operating parameters are combined to form the representative vector of each unit group.

3. The energy supply system stability assessment method based on energy units according to claim 2 is characterized in that: The method for obtaining the operation similarity value includes: Analyze the similarity between the representative vectors of the unit groups in different time periods and the similarity of the energy unit distribution in different time periods to obtain the state similarity between any two time periods; In any two time periods, at each identical position 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, and the mean of the similarity indexes between all types of energy units at each identical position and time is used as the mean state similarity between the two time periods at each identical position and time. The sum of the mean state similarities at all identical position and time in the two time periods is normalized and used as the state similarity value of the energy supply system in 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 after negative correlation mapping of the first adjustment weight is used as the second adjustment weight; 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, and the resulting sum is normalized and used as the operation similarity value of the energy supply system between the two time periods.

4. The energy supply system stability assessment method based on energy units according to claim 3 is characterized in that: The method for obtaining the state similarity includes: In any two time periods, the energy units in any two unit groups in the two time periods are compared, the number of energy unit overlaps is used as the overlap factor, and the two unit groups whose overlap factor is greater than the preset overlap threshold are regarded as a combination, thereby obtaining all the combinations corresponding to the two time periods; For any combination, calculate the cosine similarity between the representative vectors of the two unit groups as the similarity parameter; In any two time periods, the corresponding coincidence factors are weighted and fused based on the similarity parameters corresponding to the combination to obtain the state similarity between the two time periods.

5. The energy supply system stability assessment method based on energy units according to claim 1 is characterized in that: The method for obtaining the target time period includes: Based on the K-means clustering algorithm and the preset K value, cluster analysis is performed on all time periods to obtain all categories. The distance metric is the value after negative correlation mapping of the running similarity values between time periods. Multiply the ratio of the number of time periods in each category to the total number of time periods in all categories by the preset number to obtain the number of selections corresponding to each category; In each category, the average of the running similarity values between each time period and all other time periods is normalized to form the prevalence of each time period. Time periods with a prevalence greater than a preset prevalence threshold are considered ordinary time periods, and time periods with a prevalence less than or equal to the preset prevalence threshold are considered special time periods. In each category, the target quantity for the ordinary period is the product of the proportion of the quantity in the ordinary period and the selected quantity corresponding to each category. The target quantity for the special period is the product of the proportion of the quantity in the special period and the selected quantity corresponding to each category. In each category, the number of targets based on the normal period was randomly selected during the normal period, and the number of targets based on the special period was randomly selected during the special period; The time period selected from all categories is used as the target time period.

6. The energy supply system stability assessment method based on energy units according to claim 1 is characterized in that: The 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 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, and the operating parameter time series data of all energy units in the energy supply system in each target time period are associated with the node admittance matrix to obtain a power network model; In the power network model, the operating parameter time series data of the energy unit in each target time period is processed according to the three-phase balanced system, the three-phase unbalanced system, Ohm's law, and the Joule heat loss formula to obtain the current value and voltage value of the energy supply system at each moment, as well as the power loss value of each energy unit in the energy supply system at each moment; Determine the power value of the energy supply system at each moment based on the power flow calculation method and the output power, load power and power loss values of each energy unit at each moment; Determine the limit power and critical voltage of the energy supply system within the target time period based on the current value, voltage value, and power value of the energy supply system at each moment within the target time period; The ratio of the difference between the energy supply system's limit power and the power value at each moment during the target time period 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 expressed as a percentage. The ratio of the difference between the voltage value at each moment in the target time period and the corresponding critical voltage to the voltage value at each moment is used as the voltage stability reserve coefficient at each moment, where the voltage stability reserve coefficient is a percentage; During a certain target time period, if the average value of the power angle stability reserve coefficient at all times is within the preset first stability range, and the average value of the voltage stability reserve coefficient at all times is within the preset second stability range, then the stability of the energy supply system during the target time period is considered to be good; otherwise, the stability is considered to be poor and an early warning is issued.

7. The method for evaluating the stability of an energy supply system based on an energy unit according to claim 6, characterized in that: The determining of the limit power and critical voltage of the energy supply system within the target time period based on the current value, voltage value, and power value 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 the preset ratio; Based on the proportional load increment continuous power flow method, the current, voltage and power values of the energy supply system in each target time period are analyzed to check whether the power flow solution converges. If not, it is considered that the collapse point has been reached and the calculation is terminated. The load level corresponding to the collapse point is used as the limit power of the energy supply system in the target time period, and the voltage value corresponding to the collapse point is used as the critical voltage of the energy supply system in the target time period.

8. The energy supply system stability assessment method based on energy units according to claim 1 is characterized in that: The method for obtaining the unit group includes: At each moment, cluster analysis is performed on the data points corresponding to the operating state vectors of all energy units based on the DBSCAN clustering algorithm to obtain all clusters, and the energy units in each cluster are regarded as a unit group.

9. A functional system stability assessment system based on energy units, characterized in that: It includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and when the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor, the steps of the energy supply system stability assessment method based on an energy unit as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the energy supply system stability assessment method based on an energy unit as described in any one of claims 1 to 8 are implemented.

Images