Settlement model verification test method and system based on actual case backtracking

Abstract

The application provides a settlement model verification test method and system based on actual case backtracking, relates to the technical field of model testing, and obtains case data of a power market through the application, the case data including settlement basis data, settlement deviation values, correction measures and historical correction results; correlation analysis is performed on the settlement deviation values and the actual power supply state of the power market to obtain a correlation parameter set covering different power supply states; correction operation rules are obtained according to the correlation parameter set; the correction operation rules are used to generate a deviation value set; each deviation value in the deviation value set is compared with a preset deviation threshold to obtain verification results of an initial settlement model under different simulated power supply states, the preset deviation threshold including an absolute deviation threshold and a relative deviation threshold, high-precision and quantifiable verification of the settlement model under different power supply states is realized, and the accuracy and robustness of the model are improved.

Description

Technical Field

[0001] This application relates to the field of model testing technology, and in particular to a settlement model verification testing method and system based on backtracking from real-world cases. Background Technology

[0002] In the context of an increasingly complex electricity market, the accuracy of the settlement model is directly related to the economic interests of market participants and the stability of grid operation. Especially in an environment with a high proportion of renewable energy access, increased load fluctuations, and parallel trading across multiple time scales, the settlement process is highly susceptible to factors such as metering errors, deviations in the execution of dispatch instructions, and sudden changes in power supply status, which can cause the settlement results to deviate from the true value.

[0003] To address the aforementioned needs, existing mainstream solutions include settlement model verification schemes based on digital twins and rule engines. These schemes simulate the settlement process by constructing a virtual mapping environment and use predefined rules to verify the consistency of the settlement logic.

[0004] However, existing solutions have certain limitations. For example, the verification logic is disconnected from the actual power supply operation status, and the deviation injection is based only on a preset mode without fully relating to the actual power supply status changes of the power grid, resulting in significant differences between the test scenario and the actual causes of deviations. The rule base is difficult to dynamically integrate historical deviation correction experience and cannot adjust the verification strategy based on the closed-loop feedback of real cases, resulting in insufficient verification coverage when the model faces atypical or complex deviations, and making it difficult to truly reflect the deviation handling capability of the settlement model in a dynamic power system environment. Summary of the Invention

[0005] The purpose of this application is to provide a settlement model verification test method and system based on backtracking from real-world cases, in order to solve the problem that there are significant differences between the test scenarios and the causes of deviations from reality in the existing technology.

[0006] Firstly, this application provides a settlement model verification and testing method based on backtracking from real-world cases, including: Acquire case data from the electricity market, including settlement basis data, settlement deviation values, corrective measures, and historical correction results. Settlement basis data includes electricity transaction data, metering data, and dispatch instruction data. A correlation analysis was conducted between the settlement deviation value and the actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply conditions; Based on the associated parameter set, multiple simulated power supply states covering different power supply scenarios are constructed. Based on the settlement basic data, correction measures, and simulated power supply states, the operation rules of the pre-constructed initial settlement model are corrected to obtain the corrected operation rules. The settlement basis data in the target pairing data under each simulated power supply state is settled using the correction operation rules to obtain the correction results corresponding to each simulated power supply state. The correction results are compared with the historical correction results to generate a set of deviation values. The deviation values ​​in the deviation value set are compared with the preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, so as to realize the verification test of the initial settlement model. The preset deviation threshold includes absolute deviation threshold and relative deviation threshold.

[0007] Optionally, a correlation analysis is performed on the settlement deviation value and the actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply conditions, including: The settlement deviation value is linked to the actual power supply data at the corresponding time point in the electricity market to form multiple pairs of deviation status data. The actual power supply data includes power supply voltage data, power supply frequency data, and power supply load data. Based on the preset power supply load range, the deviation state data in the same power supply load range are grouped to obtain the grouping results under different power supply conditions. The actual power supply data in each grouping result is used as the power distribution parameters of the corresponding grouping result. Calculate the proportion of the settlement deviation value in each group's results to the corresponding correction value in the correction measures, and use the proportion as the deviation correction coefficient for each group's results; The distribution parameters and corresponding deviation correction coefficients of each group's results are linked and bound to form a set of associated parameters covering different power supply conditions.

[0008] Optionally, the distribution parameters and corresponding deviation correction coefficients of each group's results are correlated and bound to form a set of correlated parameters covering different power supply states, including: Assign a corresponding status identifier to each group of results to obtain a set of grouped results with identifiers. The status identifiers include the power supply voltage range, power supply frequency range, and power supply load range. Determine whether the grouped distribution parameters and deviation correction coefficients of each group with status indicators share the same status indicator. If they share the same status indicator, the group is considered to be successfully matched. The distribution parameters and deviation correction coefficients that are determined to be successfully matched within the group are combined to form multiple parameter matching pairs. A corresponding status identifier is added to each parameter matching pair to form multiple groups of identified associated units. All associated units are integrated to obtain an associated parameter set covering different power supply states.

[0009] Optionally, based on the associated parameter set, multiple simulated power supply states covering different power supply scenarios are constructed. Based on the settlement basis data, correction measures, and simulated power supply states, the calculation rules of the pre-constructed initial settlement model are modified to obtain modified calculation rules, including: Based on the distribution parameters in the associated parameter set, the fluctuation range of the distribution parameters is determined, including a first range, a second range, and a third range. Based on the preset data value interval, voltage simulation data, frequency simulation data, and load simulation data within the fluctuation range are generated, and the voltage simulation data, frequency simulation data, and load simulation data are cross-combined in multiple dimensions to form multiple sets of simulated power distribution parameters. Based on the simulated power distribution parameters of each group, construct multiple simulated power supply states covering different power supply scenarios; The settlement basis data and the corresponding time point correction measures are linked to form multiple sets of linked data; Based on the fluctuation range, the threshold range of distribution parameters for different simulated power supply states is determined. The actual power supply data that is within the threshold range of distribution parameters and is associated with the settlement basis data is selected as the target distribution parameters. The associated data corresponding to the target distribution parameters is used as the target pairing data for the corresponding simulated power supply state. Based on the target pairing data and simulated power distribution parameters for each simulated power supply state, the calculation rules of the pre-constructed initial settlement model are modified to obtain the modified calculation rules.

[0010] Optionally, based on each set of simulated power distribution parameters, multiple simulated power supply states covering different power supply scenarios are constructed, including: Assign corresponding simulation status identifiers to each group of simulated power distribution parameters to obtain the simulated parameter combination. The simulation status identifiers include simulated voltage information, simulated frequency information, and simulated load information. Match the simulation status identifier of each simulation parameter combination with the status identifier corresponding to the deviation correction coefficient. Use the deviation correction coefficient that perfectly matches each simulation parameter combination as the target correction coefficient for the corresponding simulation parameter combination. If there is no deviation correction coefficient that perfectly matches each simulation parameter combination, use the deviation correction coefficient with the highest matching degree with each simulation parameter combination as the target correction coefficient. By associating each set of simulation parameters with the corresponding target correction coefficients, multiple sets of target association groups are obtained; Add scenario description information to each target association group to form multiple scenario units. The scenario description information includes power supply pressure level, power supply stability level and power consumption period. Based on scenario units, multiple simulated power supply states covering different power supply scenarios are constructed.

[0011] Optionally, the settlement basis data in the target pairing data under each simulated power supply state is settled using the correction calculation rules to obtain the correction result corresponding to each simulated power supply state. The correction result is then compared with the historical correction result to generate a deviation value set, including: The power transaction data in the settlement base data of the target pairing data under each simulated power supply state is processed by using the modified calculation rules to obtain the calculation results corresponding to each simulated power supply state. Based on the metering data in the settlement basis data of the target pairing data under each simulated power supply state, the corresponding calculation results are supplemented and adjusted to obtain the adjusted results for each simulated power supply state. Based on the scheduling instruction data in the settlement base data of the target pairing data under each simulated power supply state, the corresponding adjusted results are processed to obtain the correction results corresponding to each simulated power supply state. Calculate the deviation between each correction result under each simulated power supply state and the historical correction result corresponding to the same time node, and integrate all deviation values ​​to obtain a deviation value set.

[0012] Optionally, each deviation value in the deviation value set is compared with a preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, so as to realize the verification test of the initial settlement model, including: Based on different simulated power supply conditions, the deviation value set is grouped to obtain multiple subsets of deviation values; The deviation values ​​in each deviation value subset are compared with the preset deviation threshold. Deviation values ​​that are less than the preset deviation threshold are judged as qualified deviation values. The pass rate of each deviation value subset is calculated by combining the total number of deviation values ​​in each deviation value subset. The pass rate of each subset of deviation values ​​is compared with the preset pass threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions. The verification result corresponding to the pass rate being greater than or equal to the preset pass threshold is considered as passing the verification, and the verification result corresponding to the pass rate being less than the preset pass threshold is considered as failing the verification.

[0013] Secondly, this application provides a settlement model verification and testing system based on real-world case backtracking, including: The acquisition module is used to acquire case data of the electricity market. The case data includes settlement basis data, settlement deviation values, correction measures, and historical correction results. Among them, the settlement basis data includes electricity transaction data, metering data, and dispatch instruction data. The correlation module is used to perform correlation analysis on the settlement deviation value and the actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply conditions; The correction module is used to construct multiple simulated power supply states covering different power supply scenarios based on the associated parameter set. Based on the settlement basic data, correction measures and simulated power supply states, the module corrects the calculation rules of the pre-constructed initial settlement model to obtain the corrected calculation rules. The settlement module is used to process the settlement base data in the target pairing data under each simulated power supply state using the correction calculation rules, obtain the correction result corresponding to each simulated power supply state, compare the correction result with the historical correction result, and generate a set of deviation values. The comparison module is used to compare each deviation value in the deviation value set with a preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, so as to realize the verification test of the initial settlement model. The preset deviation threshold includes an absolute deviation threshold and a relative deviation threshold.

[0014] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to implement the steps of a settlement model verification test method based on real-world case backtracking, as described in the first aspect above, when executing a computer program.

[0015] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the settlement model verification test method based on real-world case backtracking as described in the first aspect above.

[0016] This application provides a settlement model verification and testing method based on real-world case backtracking. The method acquires case data from the electricity market, including settlement foundation data, settlement deviation values, correction measures, and historical correction results. The settlement foundation data includes electricity transaction data, metering data, and dispatch instruction data. A correlation analysis is performed between the settlement deviation values ​​and actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply states. Based on the correlation parameter set, multiple simulated power supply states covering different power supply scenarios are constructed. The calculation rules of the pre-constructed initial settlement model are corrected based on the settlement foundation data, correction measures, and simulated power supply states to obtain corrected calculation rules. The corrected calculation rules are used to process the settlement foundation data in the target pairing data under each simulated power supply state to obtain the correction results corresponding to each simulated power supply state. The correction results are compared with historical correction results to generate a deviation value set. Each deviation value in the deviation value set is compared with a preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply states, thereby achieving verification testing of the initial settlement model. The preset deviation thresholds include absolute deviation thresholds and relative deviation thresholds.

[0017] By systematically collecting complete case data, including settlement basis data, settlement deviation values, correction measures, and historical correction results, a verification data foundation closely coupled with the actual operation of the electricity market was constructed. A structural mapping between settlement deviations and the actual power supply status of the grid was achieved, revealing the typical characteristics and correction patterns of deviations under different power supply conditions. This expanded the verification dimension from isolated settlement logic to a level deeply coupled with the physical operating state of the power system. This enhanced the model's adaptability to complex and dynamic operating environments, improved the credibility and interpretability of the verification results, and enabled a quantitative evaluation of model performance. It not only identifies the model's performance in typical scenarios but also exposes its potential defects under marginal or composite power supply conditions, thus comprehensively measuring the model's robustness and accuracy.

[0018] Furthermore, the settlement deviation value is bound to the actual power supply data at the corresponding time node to form multiple pairs of deviation status data; and the ratio of the settlement deviation value in each group to the correction value in the correction measures is calculated as the deviation correction coefficient; finally, the power distribution parameters are associated and bound with the deviation correction coefficient to form a set of associated parameters covering different power supply states; this solves the problems of deviation injection being disconnected from power supply state and correction logic being statically fixed in the existing scheme, and supports the settlement model's ability to perform refined verification and adaptive optimization in real and dynamic power operation environments. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a settlement model verification and testing method based on real-world case backtracking, provided for an embodiment of this application; Figure 2 A flowchart illustrating a method for generating an associated parameter set provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a settlement model verification and testing system based on backtracking from real-world cases, provided in an embodiment of this application. Detailed Implementation

[0021] To address the core shortcomings of existing settlement model verification schemes based on digital twins and rule engines in the complex operating environment of the power market, such as the disconnect between verification logic and actual power supply status, rigid deviation injection modes, and difficulty in integrating historical correction experience, this application focuses on the real needs of power market settlement deviation verification. It conducts in-depth correlation analysis between settlement deviations and actual power supply data of the power system, constructs simulated power supply status covering different power supply scenarios, generates a deviation value set by comparing it with the actual correction results, and performs quantitative evaluation by combining absolute and relative deviation thresholds. This enables accurate, closed-loop, and scenario-based verification of the settlement model's ability to identify and correct deviations in complex power supply environments such as high proportion of renewable energy access and severe load fluctuations, breaking through the bottlenecks of existing technologies in terms of dynamic adaptability and verification authenticity.

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The core of this application is to provide a settlement model verification and testing method based on backtracking from real-world cases. A flowchart of one specific implementation method is shown below. Figure 1 As shown, the method includes: Step 101: Obtain case data of the electricity market. Case data includes settlement basis data, settlement deviation value, correction measures and historical correction results. Settlement basis data includes electricity transaction data, metering data and dispatch instruction data.

[0024] In this step, case data refers to various types of settlement-related data that have occurred in the electricity market in the past. Settlement basis data refers to the basic data required for the electricity market settlement process. Electricity transaction data refers to transaction-related data involving the buying and selling of electricity in the electricity market. Metering data refers to data obtained by measuring electricity consumption, supply, etc. Dispatch instruction data refers to the instruction-related data issued during the electricity dispatch process. Settlement deviation value refers to the difference between the actual settlement result and the expected or correct settlement result in past settlement processes. Corrective measures refer to the corrective means or methods adopted to address past settlement deviations. Historical correction results refer to the settlement results obtained after correcting the deviations using corrective measures.

[0025] In this embodiment of the application, case data is collected through channels such as electricity market databases and information systems. During the collection process, it is necessary to ensure that the case data includes settlement basis data, settlement deviation values, correction measures, and historical correction results. The settlement basis data includes electricity transaction data, metering data, and dispatch instruction data.

[0026] Step 102: Perform correlation analysis on the settlement deviation value and the actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply conditions.

[0027] In this step, actual power supply data refers to state-related data during the actual power supply process in the electricity market.

[0028] Step 103: Based on the associated parameter set, construct multiple simulated power supply states covering different power supply scenarios. Based on the settlement basic data, correction measures, and simulated power supply states, correct the calculation rules of the pre-constructed initial settlement model to obtain the corrected calculation rules.

[0029] In this step, "multiple simulated power supply states" refers to simulated power supply states covering various power supply scenarios, constructed based on a set of associated parameters. The pre-built initial settlement model refers to a model designed and built in advance for electricity market settlement before verification testing. The calculation rules of the initial settlement model refer to the rules followed by the pre-built initial settlement model when performing settlement calculations. The revised calculation rules refer to the new calculation rules obtained after revising the calculation rules of the initial settlement model. Target pairing data refers to the associated data corresponding to each simulated power supply state used to revise the calculation rules of the initial settlement model.

[0030] Step 104: Use the correction calculation rules to perform settlement processing on the settlement basis data in the target pairing data under each simulated power supply state to obtain the correction result corresponding to each simulated power supply state. Compare the correction result with the historical correction result to generate a deviation value set.

[0031] In this step, the correction result refers to the settlement result obtained after performing settlement processing on the settlement basis data using the correction operation rules.

[0032] Step 105: Compare each deviation value in the deviation value set with the preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, so as to realize the verification test of the initial settlement model. The preset deviation threshold includes absolute deviation threshold and relative deviation threshold.

[0033] In this step, the preset deviation threshold refers to a pre-set threshold used to determine whether a deviation value is acceptable. The absolute deviation threshold refers to the threshold within the preset deviation thresholds used to measure the magnitude of the absolute deviation. The relative deviation threshold refers to the threshold within the preset deviation thresholds used to measure the magnitude of the relative deviation. The verification result refers to the result obtained by comparing the deviation value with the preset deviation threshold to determine whether the initial settlement model is qualified under different simulated power supply conditions. The verification test refers to the process of verifying the performance and accuracy of the initial settlement model by obtaining the verification results.

[0034] The embodiments of this application address the problems in the prior art where the settlement model verification lacks scenario-based support and the model rules do not match the actual power supply status.

[0035] This application provides a specific embodiment, such as Figure 2 As shown, step 102 involves performing a correlation analysis on the settlement deviation value and the actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply states. This specifically includes the following steps: Step 201: Bind the settlement deviation value with the actual power supply data at the corresponding time point in the electricity market to form multiple pairs of deviation status data. The actual power supply data includes power supply voltage data, power supply frequency data, and power supply load data.

[0036] In this step, actual power supply data refers to data reflecting the power supply status during the actual power supply process in the electricity market. Power supply voltage data refers to data recording the voltage magnitude during actual power supply. Power supply frequency data refers to data recording the frequency magnitude during actual power supply. Power supply load data refers to data recording the load magnitude during actual power supply. Settlement deviation value refers to the difference between the actual settlement result and the correct settlement result in past settlement processes. Deviation status data refers to paired data formed by binding the settlement deviation value with the actual power supply data at the corresponding time point.

[0037] In this embodiment of the application, the specific time node of each settlement deviation value is first determined, and then the actual power supply data that completely matches the time node is extracted from the power market operation monitoring system. Then, each settlement deviation value is bound to the actual power supply data of the extracted corresponding time node to ensure the time consistency and data integrity of each pair of bound data. Finally, multiple pairs of deviation status data are formed, and each pair of deviation status data includes a settlement deviation value and a set of corresponding actual power supply data.

[0038] Step 202: Based on the preset power supply load range, group the deviation state data in the same power supply load range to obtain the grouping results under different power supply conditions, and use the actual power supply data in each grouping result as the power distribution parameters of the corresponding grouping result.

[0039] In this step, the preset power supply load range refers to the pre-defined interval used to divide the power supply load range. The grouping result refers to the data set formed by grouping the deviation data within the same preset power supply load range. The distribution parameters refer to the actual power supply data included in each grouping result.

[0040] In this embodiment of the application, based on the value range of the preset power supply load, all deviation state data are traversed, and the value range of the preset power supply load to which the power supply load data in each pair of deviation state data belongs is determined one by one. Deviation state data belonging to the same value range of the preset power supply load are grouped together, thereby obtaining the grouping results under different power supply states. Then, the actual power supply data included in each grouping result is determined as the power distribution parameters corresponding to the grouping result.

[0041] Step 203: Calculate the proportion of the settlement deviation value in each group's results to the corresponding correction value in the correction measures, and use the proportion as the deviation correction coefficient for each group's results.

[0042] In this step, corrective measures refer to the corrective means or methods adopted to address past settlement deviations. The corresponding correction values ​​in the corrective measures refer to the specific values ​​used to correct settlement deviations. The deviation correction coefficient refers to the proportion of the corresponding correction values ​​in the aforementioned corrective measures used as a correction reference coefficient for the grouping results.

[0043] In this embodiment of the application, the correction measures corresponding to each group of results are first extracted from the case data, and the specific correction values ​​used to correct the settlement deviation under the corresponding power supply state are determined. Then, for each group of results, the ratio of each settlement deviation value in the group to the correction value in the corresponding correction measure is calculated. All individual ratios in the group are screened to exclude abnormal ratio values ​​that are obviously beyond the reasonable range. The arithmetic mean of the remaining valid individual ratios is then calculated, and the calculated arithmetic mean is determined as the deviation correction coefficient of the group of results.

[0044] Step 204: Associate and bind the distribution parameters and corresponding deviation correction coefficients of each group's results to form an associated parameter set covering different power supply states.

[0045] In this embodiment of the application, for each group of grouping results, the corresponding power distribution parameters of the group are associated and bound with the calculated deviation correction coefficient of the group to ensure that the power distribution parameters and deviation correction coefficient of each group can accurately reflect the settlement deviation related characteristics under the power supply state. Then, all the data pairs that have been associated and bound are integrated to finally form a set of associated parameters covering different power supply states.

[0046] This application addresses the problems in the prior art where historical settlement deviations are disconnected from actual power supply conditions and where there is a lack of specific correction parameters for different power supply conditions. It provides accurate scenario-based data support for the subsequent construction of various simulated power supply conditions and the correction of the initial settlement model's calculation rules, ensuring that subsequent model correction and verification have clear actual data basis.

[0047] For example, in the settlement model verification project of electricity market A, the specific data from January to June are first extracted from the operational database of electricity market A as follows: January's settlement deviation value is 30,000 yuan, corresponding to actual power supply data of 220V voltage, 50.02Hz frequency, and 800MW load; February's settlement deviation value is 50,000 yuan, corresponding to actual power supply data of 225V voltage, 50.05Hz frequency, and 1200MW load; March's settlement deviation value is 20,000 yuan, corresponding to actual power supply data of 222V voltage, 50.03Hz frequency, and 800MW load. z. Power supply load: 900MW; April settlement deviation value: 60,000 yuan, corresponding to actual power supply data of power supply voltage: 226V, power supply frequency: 50.06Hz, power supply load: 1300MW; May settlement deviation value: 40,000 yuan, corresponding to actual power supply data of power supply voltage: 221V, power supply frequency: 50.04Hz, power supply load: 850MW; June settlement deviation value: 40,000 yuan, corresponding to actual power supply data of power supply voltage: 224V, power supply frequency: 50.05Hz, power supply load: 1100MW. The monthly settlement deviation value is linked with the actual power supply data of that month to form 6 pairs of deviation status data.

[0048] The preset power supply load range can be 500 to 1000MW for low load and 1000 to 1500MW for medium load. Six pairs of deviation status data are traversed. The power supply load data for January, March, and May are between 500 and 1000MW and are classified as low load group results. The power supply load data for February, April, and June are between 1000 and 1500MW and are classified as medium load group results. At the same time, the power supply voltage, frequency, and load data in each group are determined as the distribution parameters of the corresponding group.

[0049] Two sets of corresponding correction measures were extracted from the database. The settlement base coefficient is a dimensionless coefficient used to characterize the intensity of the initial settlement base adjustment in power settlement business. Its physical meaning is the scaling ratio of the initial settlement amount under a specific power supply state in order to eliminate settlement deviations caused by systematic errors of metering equipment or delays in dispatch response.

[0050] The specific calculation logic for obtaining the corrected value based on the settlement baseline coefficient is as follows: the difference between the settlement baseline coefficients is multiplied by the corresponding benchmark settlement amount for the month to obtain the corrected value. The benchmark settlement amount refers to the original settlement amount calculated according to the initial settlement rules based on the electricity transaction data and metering data in the settlement baseline data before the retrospective verification of this application.

[0051] In the low-load group, the baseline settlement amount is set at 1.2 million yuan for January, 1.6667 million yuan for March, and 1.75 million yuan for May. The adjusted settlement base coefficient is set to 1.05 for January, corresponding to a correction of 60,000 yuan; the adjusted settlement base coefficient is set to 1.03 for March, corresponding to a correction of 50,000 yuan; and the adjusted settlement base coefficient is set to 1.04 for May, corresponding to a correction of 70,000 yuan.

[0052] In the medium load group, the base settlement amount for February, April, and June is set at 1 million yuan. The settlement base coefficient for February is adjusted to 1.1, with a corresponding correction of 100,000 yuan; the settlement base coefficient for April is adjusted to 1.12, with a corresponding correction of 120,000 yuan; and the settlement base coefficient for June is adjusted to 1.08, with a corresponding correction of 80,000 yuan.

[0053] Calculate the deviation correction factor: For the low load group, first calculate the individual proportions for each month. The proportion for January is 0.5, the proportion for March is 0.4, and the proportion for May is 0.57. After excluding proportions without abnormalities, the deviation correction factor is calculated to be 0.49. For the medium load group, the individual proportions for each month are calculated in the same way. The proportion for February is 0.5, the proportion for April is 0.5, and the proportion for June is 0.5. The deviation correction factor is 0.5.

[0054] Finally, the distribution parameters of the low-load group, including the specific data for each month, are associated with 0.49, and the distribution parameters of the medium-load group, including the specific data for each month, are associated with 0.5, thus forming a set of associated parameters.

[0055] This application provides a specific embodiment. Step 204 involves associating and binding the power distribution parameters of each group's grouping results with the corresponding deviation correction coefficients to form a set of associated parameters covering different power supply states. This specifically includes the following steps: Step 211: Assign a corresponding status identifier to each group of group results to obtain a group result set with identifiers. The status identifiers include the power supply voltage range, power supply frequency range, and power supply load range.

[0056] In this step, the grouping result status identifier refers to a unique identifier assigned to each grouping result to distinguish different power supply states. The power supply voltage range refers to the range of power supply voltage values ​​determined based on the power supply voltage data of the distribution parameters in each grouping result. The power supply frequency range refers to the range of power supply frequency values ​​determined based on the power supply frequency data of the distribution parameters in each grouping result. The power supply load range refers to the range of power supply load values ​​determined based on the power supply load data of the distribution parameters in each grouping result.

[0057] In this embodiment, for each group of results, the power supply voltage data, power supply frequency data, and power supply load data are extracted from the power distribution parameters. The maximum and minimum values ​​of these three types of data within each group are calculated to determine the power supply voltage range, power supply frequency range, and power supply load range corresponding to each group. Then, a unique status identifier for the group of results is generated based on these three ranges, and the identifier content must clearly reflect the three ranges. Finally, all group results with assigned status identifiers are integrated to obtain the group result set.

[0058] Step 212: Determine whether the grouped distribution parameters and deviation correction coefficients with status indicators in each group share the same status indicator. If they share the same status indicator, the group is considered to be successfully matched.

[0059] In this embodiment of the application, the grouped results with labels are extracted from the grouped result set. The status labels attached to the power distribution parameters and the deviation correction coefficients in each group are obtained respectively. Then, the status labels of the power distribution parameters and the status labels of the deviation correction coefficients in the same group are compared character by character to check whether the power supply voltage range, power supply frequency range and power supply load range of the two are completely consistent. If the three ranges are completely consistent, it is determined that the power distribution parameters and the deviation correction coefficients in the group share the same status label and the group matching is successful. If any range is different, it is determined that they do not share the same status label and the group matching fails.

[0060] Step 213: Combine the power distribution parameters and deviation correction coefficients that are determined to be successfully matched within the group to form multiple sets of parameter matching pairs. Add corresponding status identifiers to each parameter matching pair to form multiple sets of identified associated units. Integrate all associated units to obtain a set of associated parameters covering different power supply states.

[0061] In this step, the status identifier of the parameter matching pair refers to the status identifier of the corresponding grouping result added for each parameter matching pair.

[0062] In this embodiment, all successfully matched grouping results within a group are selected, and the power distribution parameters in each group are combined with the corresponding deviation correction coefficients to ensure that the parameters and coefficients of each group come from the same successfully matched group, thereby forming multiple parameter matching pairs. Then, a status identifier corresponding to the group is added to each parameter matching pair so that each matching pair can be associated with a specific power supply state. Finally, each parameter matching pair with the added status identifier is treated as an association unit, and all association units are integrated to obtain an association parameter set covering different power supply states.

[0063] The embodiments of this application address the shortcomings of the prior art where the correlation between power supply status-related parameters and correction coefficients is chaotic and the correspondence cannot be clearly defined, providing an accurate and orderly data foundation for subsequent construction of simulated power supply status based on the correlation parameter set.

[0064] This application provides a specific embodiment. Step 103 involves constructing multiple simulated power supply states covering different power supply scenarios based on the associated parameter set. Based on the settlement foundation data, correction measures, and simulated power supply states, the calculation rules of the pre-constructed initial settlement model are corrected to obtain the corrected calculation rules. Specifically, this includes the following steps: Step 301: Based on the distribution parameters in the associated parameter set, determine the fluctuation range of the distribution parameters, which includes a first range, a second range, and a third range.

[0065] In this step, the fluctuation range of distribution parameters refers to the value range of various distribution data determined after extracting all distribution parameters from the associated parameter set. The first range refers to the range between the minimum and maximum values ​​of all power supply voltage data in the distribution parameters. The second range refers to the range between the minimum and maximum values ​​of all power supply frequency data in the distribution parameters. The third range refers to the range between the minimum and maximum values ​​of all power supply load data in the distribution parameters.

[0066] In this embodiment, the power distribution parameters of all associated units are first extracted from the associated parameter set, and the power supply voltage data, power supply frequency data, and power supply load data are separated. The values ​​of these three types of data are statistically analyzed, and the minimum and maximum values ​​of the power supply voltage data are calculated to determine the first range. Similarly, the minimum and maximum values ​​of the power supply frequency data are calculated to determine the second range. The minimum and maximum values ​​of the power supply load data are calculated to determine the third range. The three ranges are integrated to obtain the fluctuation range of the power distribution parameters.

[0067] Step 302: Based on the preset data value interval, generate voltage simulation data, frequency simulation data, and load simulation data within the fluctuation range, and perform multi-dimensional cross-combination of voltage simulation data, frequency simulation data, and load simulation data to form multiple sets of simulated power distribution parameters.

[0068] In this step, the preset data interval refers to the pre-defined difference between two adjacent data points used to generate the simulation data. Voltage simulation data refers to the simulated voltage data generated within a first range according to the preset data interval. Frequency simulation data refers to the simulated frequency data generated within a second range according to the preset data interval. Load simulation data refers to the simulated load data generated within a third range according to the preset data interval. Simulated distribution parameters refer to the parameter combinations formed by multi-dimensional cross-combination of voltage simulation data, frequency simulation data, and load simulation data.

[0069] In this embodiment, a preset data value interval is first set; based on the first range, starting from the minimum value, multiple voltage simulation data are generated sequentially at preset intervals; similarly, frequency simulation data is generated based on the second range and preset intervals, and load simulation data is generated based on the third range and preset intervals; then each voltage simulation data is cross-combined with each frequency simulation data and each load simulation data to form multiple sets of simulated power distribution parameters.

[0070] Step 303: Based on the simulated power distribution parameters of each group, construct multiple simulated power supply states covering different power supply scenarios.

[0071] In this embodiment, for each set of simulated power distribution parameters, the characteristics of the voltage simulation data, frequency simulation data, and load simulation data included are analyzed. Combined with the classification standards of actual power supply scenarios in the power market, the corresponding power supply scenario type is matched for each set of simulated power distribution parameters. Then, each set of simulated power distribution parameters is bound to the corresponding scenario type to construct the corresponding simulated power supply state. Finally, all simulated power supply states are integrated to form multiple simulated power supply states covering different power supply scenarios.

[0072] Step 304: Link the settlement base data with the corresponding time point correction measures to form multiple sets of linked data.

[0073] In this embodiment of the application, all settlement basic data and all corrective measures are extracted from the case data, the specific execution time node of each corrective measure is determined, and the settlement basic data and corrective measures at the same time node are associated accordingly to ensure that each set of associated data includes the settlement basic data within a certain time period and the corrective measures used to correct the settlement deviation within that time period, thereby forming multiple sets of associated data.

[0074] Step 305: Based on the fluctuation range, determine the threshold range of distribution parameters for different simulated power supply states, select the actual power supply data that is within the threshold range of distribution parameters and is associated with the settlement basis data as the target distribution parameters, and use the associated data corresponding to the target distribution parameters as the target pairing data for the corresponding simulated power supply state.

[0075] In this step, the distribution parameter threshold range refers to the parameter range set for each simulated power supply state to filter actual power supply data, based on the fluctuation range of distribution parameters. The target distribution parameter refers to the actual power supply data that falls within the distribution parameter threshold range and is associated with the settlement basis data. Target paired data refers to the associated data corresponding to the target distribution parameter.

[0076] In this embodiment, for each simulated power supply state, a specific power distribution parameter threshold range is first set based on the corresponding simulated power distribution parameters and the fluctuation range of the power distribution parameters. Then, actual power supply data associated with the settlement basis data is extracted from the case data, and it is determined whether these data are within the power distribution parameter threshold range of a certain simulated power supply state. The actual power supply data within the range is determined as the target power distribution parameter of the simulated power supply state. Then, the associated data corresponding to the target power distribution parameter is found and determined as the target pairing data of the simulated power supply state.

[0077] Step 306: Based on the target pairing data and simulated power distribution parameters for each simulated power supply state, the calculation rules of the pre-constructed initial settlement model are modified to obtain the modified calculation rules.

[0078] In this embodiment, for each simulated power supply state, a pre-built initial settlement model is invoked. The settlement basis data in the target pairing data of the simulated power supply state is input into the initial settlement model, and settlement is performed according to the calculation rules of the initial settlement model to obtain the initial settlement result. Then, the initial settlement result is compared with the historical correction results in the target pairing data to analyze the reasons for the deviation. Then, based on the correction measures in the target pairing data and the characteristics of the simulated power distribution parameters, the calculation rules of the initial settlement model are adjusted. After adjustment, the settlement basis data is input again for verification until the deviation between the settlement result and the historical correction result meets the requirements. Finally, the correction calculation rules corresponding to the simulated power supply state are obtained. After the correction calculation rules of all simulated states are integrated, a complete set of correction calculation rules is formed.

[0079] The training process of the initial settlement model is implemented as follows: First, in the data preprocessing stage, this application employs the following quantitative rules for cleaning the raw data: For missing values, the system automatically identifies empty positions in the settlement baseline data; if the value is continuous, linear interpolation is used to fill it in; if the value is discrete, the mode of that feature is filled in. For outliers, the three-standard-deviation criterion is used; data points exceeding the mean ± three standard deviations are identified as equipment acquisition anomalies, and the mean of the observations at adjacent time points is used to reset the data.

[0080] In terms of derived feature extraction, the system constructs power supply balance features and timeliness features based on the original data. Specifically, the extraction methods are as follows: load factor features are calculated using the ratio of power supply load data to the rated capacity of the distribution network; power grid stability features are calculated using the real-time variance of the power supply frequency; and transaction time is mapped into three discrete time dimensions: peak, flat, and valley. These derived features, together with the original power transaction data, constitute the input feature vector of the initial settlement model.

[0081] Secondly, the specific structure of the architecture combining the rule engine and traditional machine learning model defined in this application is as follows: This architecture includes a rule validation layer and a numerical regression layer arranged in series. The rule validation layer, acting as a front-end engine, incorporates a Boolean logic rule library based on electricity market principles to perform hard business constraint checks on the basic data. The numerical regression layer employs a gradient boosting decision tree model, with its input being the feature vectors filtered by the rule validation layer. The output of the machine learning model serves as the basic settlement value, which is then finalized and weighted by the rule engine layer according to preset business logic.

[0082] Finally, during training, this application achieves model convergence through the following logic: the loss function is defined as the mean squared error between the model output and the historical correction results. The stochastic gradient descent algorithm is used to iteratively optimize the tree depth, learning rate, and leaf node weights of the gradient boosting decision tree model. The training objective is to make the loss function reach a preset convergence threshold on the validation set. Through the aforementioned explicit preprocessing methods, topology, and optimization algorithms, those skilled in the art can completely reproduce the training process of this initial decision model.

[0083] After constructing the initial settlement model, the calculation rules of the initial settlement model are adjusted based on the correction measures in the target pairing data and the characteristics of the simulated power distribution parameters. Since the basic model has been clearly defined through the above quantification method, the subsequent adjustments to the parameters of the numerical regression layer or the thresholds of the rule verification layer of this model have clear implementation targets and operation paths, thus enabling the realization of a complete technical solution.

[0084] This application addresses the combined shortcomings of existing technologies, such as the limited range of simulated power supply scenarios, the mismatch between model calculation rules and actual power supply conditions, and the lack of accurate data support for corrections. It ensures that the corrected calculation rules can be adapted to different power supply scenarios, laying the foundation for the accurate generation of correction results in the future.

[0085] This application provides a specific embodiment. Step 303 involves constructing multiple simulated power supply states covering different power supply scenarios based on each set of simulated power distribution parameters. This specifically includes the following steps: Step 311: Assign corresponding simulation status identifiers to each group of simulated power distribution parameters to obtain the simulated parameter combination. The simulated status identifiers include simulated voltage information, simulated frequency information, and simulated load information.

[0086] In this step, the simulated status identifier refers to a unique identifier assigned to each set of simulated power distribution parameters to distinguish different simulated power supply states. The simulated parameter combination refers to the combination of identifiers and parameters formed after assigning simulated status identifiers to each set of simulated power distribution parameters. Simulated voltage information refers to the range of simulated power supply voltage values ​​determined based on the simulated voltage data in each set of simulated power distribution parameters. Simulated frequency information refers to the range of simulated power supply frequency values ​​determined based on the simulated frequency data in each set of simulated power distribution parameters. Simulated load information refers to the range of simulated load values ​​determined based on the simulated load data in each set of simulated power distribution parameters.

[0087] In this embodiment, for each set of simulated power distribution parameters, the voltage simulation data, frequency simulation data, and load simulation data are extracted. Combined with the tolerance for small fluctuations in parameter values, the simulated voltage information, simulated frequency information, and simulated load information corresponding to each set are determined. Then, based on these three interval information, a unique simulated status identifier for the set of simulated power distribution parameters is generated. The identifier content must clearly reflect the three interval information. Finally, each set of simulated power distribution parameters is bound to the corresponding simulated status identifier to obtain the simulated parameter combination.

[0088] Step 312: Match the simulation status identifier of each simulation parameter combination with the status identifier corresponding to the deviation correction coefficient. Take the deviation correction coefficient that perfectly matches each simulation parameter combination as the target correction coefficient for the corresponding simulation parameter combination. If there is no deviation correction coefficient that perfectly matches each simulation parameter combination, take the deviation correction coefficient with the highest matching degree with each simulation parameter combination as the target correction coefficient.

[0089] In this step, the target correction factor refers to the deviation correction factor that successfully matches the combination of simulation parameters and is used for subsequent construction of the simulated power supply state. The matching degree refers to the similarity between the simulated state identifier of the simulation parameter combination and the state identifier corresponding to the deviation correction factor.

[0090] In this embodiment, all deviation correction coefficients and their corresponding status identifiers are first extracted from the set of associated parameters. For each set of simulated parameters, the three interval information in its simulated status identifier is compared with the three interval information in the status identifier corresponding to each deviation correction coefficient. The overlap of voltage interval, frequency interval, and load interval is calculated respectively. Then, the voltage interval overlap is multiplied by the voltage influence weight to obtain the voltage-related product, the frequency interval overlap is multiplied by the frequency influence weight to obtain the frequency-related product, and the load interval overlap is multiplied by the load influence weight to obtain the load-related product. Finally, these three products are summed and expressed as a percentage to obtain the matching degree. The voltage influence weight, frequency influence weight, and load influence weight are determined by performing correlation analysis and normalization on historical case data. The specific setting method and calculation logic are as follows: First, sample sequences are extracted from the acquired electricity market case data. Let the sample size be... ,extract A deviation sequence is formed from the historical settlement deviation values. And extract the power supply voltage sequence at the corresponding time point. Power supply frequency sequence and power supply load sequence .

[0091] Next, the Pearson correlation coefficient between each power supply parameter sequence and the deviation sequence is calculated. The calculation formula is shown in formula (1): (1) in, Indicates the first Power supply parameters, These correspond to the correlation coefficients of voltage, frequency, load, and settlement deviation, respectively. This is the average value of the parameter sequence. This represents the average value of the settlement deviation series. Correlation coefficient. absolute value The larger the value, the greater the actual impact of the parameter on settlement deviations.

[0092] Finally, normalization calculations are performed based on the absolute values ​​of the correlation coefficients of each parameter to obtain the influence weights. Voltage influence weight. Frequency affects weight Load impact weight The calculation logic is shown in formula (2): (2) This ensures Taking data from January to June as an example, if we calculate the absolute value of the correlation coefficient between voltage and deviation... The absolute value of the correlation coefficient between frequency and deviation The absolute value of the correlation coefficient between load and deviation Substituting this into the above logic, we can obtain: the weight of voltage influence. Frequency affects weight Load impact weight .

[0093] If a certain deviation correction coefficient corresponds to a state identifier that completely overlaps with the three intervals of the simulation state identifier, then the deviation correction coefficient is determined as the target correction coefficient for the corresponding simulation parameter combination; if no perfectly matching deviation correction coefficient exists, then the deviation correction coefficient with the highest matching degree is selected and determined as the target correction coefficient.

[0094] Step 313: Associate each set of simulation parameter combinations with the corresponding target correction coefficients to obtain multiple target association groups.

[0095] In this step, the target association group refers to the combined unit formed by associating each group of simulation parameters with the corresponding target correction coefficient.

[0096] In this embodiment, all simulation parameter combinations are traversed. For each set of simulation parameter combinations, its corresponding target correction coefficient is found, and the two are correlated to ensure that the simulation state identifier of the simulation parameter combination matches the state identifier of the target correction coefficient to the highest degree in each correlation. After the correlation is completed, each set of simulation parameter combinations and its corresponding target correction coefficient form an independent target correlation group. After all correlation groups are integrated, multiple target correlation groups are obtained.

[0097] Step 314: Add scene description information to each target association group to form multiple scene units. The scene description information includes power supply pressure level, power supply stability level, and power consumption period.

[0098] In this step, scenario description information refers to the information added to each target association group to describe the characteristics of the simulated power supply scenario corresponding to that group. A scenario unit refers to the target association group after adding scenario description information. Power supply pressure level refers to the level describing the load pressure of the power supply system, determined based on simulated load information. Power supply stability level refers to the level describing the frequency stability of the power supply system, determined based on simulated frequency information. Electricity consumption period refers to the common electricity consumption period corresponding to the simulated load information.

[0099] In this embodiment of the application, for each target association group, the simulated load information and simulated frequency information in its simulated parameter combination are first extracted.

[0100] If the simulated power supply load range falls within the normal low load range of the electricity market, the power supply pressure level is set to low; if it falls within the normal medium load range, the power supply pressure level is set to medium; if it falls within the normal high load range, the power supply pressure level is set to high. If the simulated power supply frequency range fluctuation is within the stable fluctuation range recognized by the electricity market, the power supply stability level is set to excellent; if it falls within the normal fluctuation range, the power supply stability level is set to good; if it exceeds the normal fluctuation range, the power supply stability level is set to poor. If the simulated power supply load range corresponds to the peak load period in the electricity market, the power consumption period is set to peak segment; if it corresponds to the flat load period, the power consumption period is set to flat segment; if it corresponds to the low load period, the power consumption period is set to valley segment.

[0101] The determined power supply pressure level, power supply stability level, and power consumption period are used as scene description information and added to the corresponding target association group to form multiple scene units.

[0102] Step 315: Based on scenario units, construct multiple simulated power supply states covering different power supply scenarios.

[0103] In the embodiments of this application, for each group of scene units, the combination of simulation parameters, target correction coefficients and scene description information included therein are integrated, and these three parts of information are treated as a whole and mapped to a specific simulated power supply scene. The simulated power supply scenes corresponding to all scene units are integrated to obtain a variety of simulated power supply states covering different power supply scenes.

[0104] This application addresses the combined shortcomings of existing technologies, such as the lack of clear parameter characteristics in simulated power supply scenarios, mismatch between correction coefficients and scenarios, and incomplete scenario descriptions. It enables simulated power supply states to reflect specific parameter characteristics, and provides corresponding correction basis and scenario descriptions, thus offering more accurate scenario-based support for subsequent correction of the initial settlement model's calculation rules.

[0105] This application provides a specific embodiment. Step 104 involves using correction calculation rules to perform settlement processing on the settlement basis data in the target pairing data under each simulated power supply state to obtain the correction result corresponding to each simulated power supply state. The correction result is then compared with the historical correction result to generate a deviation value set. The specific steps include the following: Step 401: Use the modified calculation rules to process the power transaction data in the settlement basis data of the target pairing data under each simulated power supply state, and obtain the calculation results corresponding to each simulated power supply state.

[0106] In this embodiment, for each simulated power supply state, the correction operation rule corresponding to the simulated power supply state is first determined, and then the power transaction data in the settlement basis data is extracted from the target pairing data of the simulated power supply state. The extracted power transaction data is input into the corresponding correction operation rule, and the operation is processed according to the operation logic set by the rule to ensure that the operation process conforms to the characteristics of the simulated power supply state, and finally the operation result corresponding to each simulated power supply state is obtained.

[0107] Step 402: Based on the metering data in the settlement basis data of the target pairing data under each simulated power supply state, supplement and adjust the corresponding calculation results to obtain the adjusted results for each simulated power supply state.

[0108] In this embodiment of the application, for each simulated power supply state, metering data is first extracted from the settlement basic data in its target pairing data; then the difference between the metering data and the corresponding calculation results is analyzed; finally, according to the actual meaning of the metering data, the corresponding calculation results are supplemented and adjusted to ensure that the adjusted results are more in line with the metering situation of the actual power supply scenario, so as to obtain the adjusted results corresponding to each simulated power supply state.

[0109] Step 403: Based on the scheduling instruction data in the settlement base data of the target pairing data under each simulated power supply state, perform calculations on the corresponding adjusted results to obtain the correction results corresponding to each simulated power supply state.

[0110] In this embodiment of the application, for each simulated power supply state, scheduling instruction data is extracted from the settlement basic data in its target pairing data; then the impact of the scheduling instruction data on the adjusted result is analyzed, and then the adjustment logic corresponding to the scheduling instruction data is integrated into the calculation and processing of the adjusted result. The calculation is supplemented according to the correction calculation rules of the simulated power supply state, and finally the correction result corresponding to each simulated power supply state is obtained.

[0111] Step 404: Calculate the deviation between each correction result under each simulated power supply state and the historical correction result corresponding to the same time node, and integrate all deviation values ​​to obtain a deviation value set.

[0112] In this embodiment, for each correction result under each simulated power supply state, the historical correction result at the same time point as the correction result is found from the case data, and the deviation value between the two is calculated to be equal to the correction result minus the historical correction result; then, the correction results under all simulated power supply states are traversed, and the corresponding deviation value is calculated one by one. All the calculated deviation values ​​are classified and organized according to the simulated power supply state to ensure that each deviation value can correspond to a specific simulated power supply state and time point. Finally, all deviation values ​​are integrated to obtain a deviation value set.

[0113] This application addresses the combined shortcomings of existing technologies, such as the lack of hierarchical logic in settlement processing, the lack of time correspondence between results and historical data comparisons, and the disordered integration of deviation values. It ensures that settlement processing closely matches actual data characteristics and that deviation values ​​accurately reflect the model correction effect, providing orderly and reliable data support for subsequent verification of the initial settlement model.

[0114] This application provides a specific embodiment. Step 105 involves comparing each deviation value in the deviation value set with a preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, thereby realizing the verification test of the initial settlement model. The specific steps include the following: Step 501: Group the deviation value set according to different simulated power supply states to obtain multiple subsets of deviation values.

[0115] In this embodiment, all types of simulated power supply states are first sorted out, and the identifier or feature corresponding to each simulated power supply state is determined. Then, all deviation values ​​in the deviation value set are traversed, and each deviation value is assigned to a corresponding group according to the simulated power supply state identifier or feature associated with it, ensuring that each deviation value belongs to only one group. After the grouping is completed, each group is a subset of deviation values ​​for the corresponding simulated power supply state. All deviation value subsets together cover all deviation values ​​in the deviation value set, and there is no overlap or repetition.

[0116] Step 502: Compare the deviation values ​​in each deviation value subset with the preset deviation threshold, and determine the deviation values ​​that are less than the preset deviation threshold as qualified deviation values. Combine the total number of deviation values ​​in each deviation value subset to calculate the pass rate of each deviation value subset.

[0117] In this step, the acceptable deviation value refers to the deviation value in each subset of deviation values ​​that, after comparison with the preset deviation threshold, is less than the preset deviation threshold. The total number of deviation values ​​refers to the total number of deviation values ​​included in each subset of deviation values. The pass rate refers to the proportion of acceptable deviation values ​​in each subset of deviation values ​​to the total number of deviation values.

[0118] In this embodiment, for each subset of deviation values, a preset deviation threshold corresponding to the subset is first determined; each deviation value in the subset is compared with the preset deviation threshold one by one; if a deviation value is less than both the absolute deviation threshold and the relative deviation threshold, the deviation value is determined to be a qualified deviation value; the number of qualified deviation values ​​in the subset is counted, and then the total number of deviation values ​​in the subset is determined; the pass rate can be obtained by calculating the ratio of the number of qualified deviation values ​​to the total number of deviation values; the pass rate for each subset of deviation values ​​is calculated according to this logic, providing a quantitative basis for the subsequent verification result determination.

[0119] Step 503: Compare the pass rate of each subset of deviation values ​​with the preset pass threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions. The verification result corresponding to a pass rate greater than or equal to the preset pass threshold is a pass verification, and the verification result corresponding to a pass rate less than the preset pass threshold is a fail verification.

[0120] In this step, the preset pass threshold refers to a pre-set pass rate standard used to determine whether the verification result passes under simulated power supply conditions. The verification result refers to the conclusion obtained by comparing the pass rate of each subset of deviation values ​​with the preset pass threshold, which determines whether the initial settlement model passes the verification under the corresponding simulated power supply conditions.

[0121] In this embodiment, a preset pass threshold is first determined. For each subset of deviation values, the calculated pass rate is compared with the preset pass threshold. If the pass rate of the subset is greater than or equal to the preset pass threshold, it indicates that the settlement accuracy of the initial settlement model under the simulated power supply condition meets the requirements, and the corresponding verification result is "passed verification". If the pass rate of the subset is less than the preset pass threshold, it indicates that the settlement accuracy of the initial settlement model under the simulated power supply condition does not meet the requirements, and the corresponding verification result is "failed verification". Then, all subsets of deviation values ​​are traversed, and the corresponding verification results are determined one by one. Finally, the set of verification results of the initial settlement model under all simulated power supply conditions is obtained.

[0122] The embodiments of this application ensure that the advantages and disadvantages of the initial settlement model under various power supply scenarios can be clearly identified, providing a clear direction for subsequent model optimization.

[0123] Figure 3 This is a schematic diagram illustrating a specific implementation of a settlement model verification and testing system based on real-world case backtracking, as provided in this application. (Refer to...) Figure 3 The system may include: The acquisition module 21 is used to acquire case data of the electricity market. The case data includes settlement basis data, settlement deviation value, correction measures and historical correction results. Among them, the settlement basis data includes electricity transaction data, metering data and dispatch instruction data.

[0124] The correlation module 22 is used to perform correlation analysis on the settlement deviation value and the actual power supply data of the electricity market to obtain a set of correlation parameters covering different power supply conditions.

[0125] The correction module 23 is used to construct multiple simulated power supply states covering different power supply scenarios based on the associated parameter set, and to correct the calculation rules of the pre-constructed initial settlement model based on the settlement basic data, correction measures and simulated power supply states to obtain the corrected calculation rules.

[0126] The settlement module 24 is used to perform settlement processing on the settlement base data in the target pairing data under each simulated power supply state using the correction calculation rules, to obtain the correction result corresponding to each simulated power supply state, and to compare the correction result with the historical correction result to generate a deviation value set.

[0127] The comparison module 25 is used to compare each deviation value in the deviation value set with a preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, so as to realize the verification test of the initial settlement model. The preset deviation threshold includes an absolute deviation threshold and a relative deviation threshold.

[0128] This application provides a settlement model verification test system based on real-case backtracking to implement the aforementioned settlement model verification test method based on real-case backtracking. Therefore, the specific implementation of the settlement model verification test system based on real-case backtracking can be found in the embodiment section of the settlement model verification test method based on real-case backtracking mentioned above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0129] This application also provides an electronic device, including: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of a settlement model verification test method based on real-world case backtracking as described above.

[0130] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the above-described methods for verifying and testing a settlement model based on backtracking from real-world cases.

[0131] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0132] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the settlement model verification test method based on real-case backtracking.

[0133] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0134] The foregoing has provided a detailed description of the settlement model verification and testing method and system based on real-world case backtracking provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for verifying and testing a settlement model based on backtracking from real-world cases, characterized in that, include: Acquire case data from the electricity market, which includes settlement basis data, settlement deviation values, corrective measures, and historical correction results. The settlement basis data includes electricity transaction data, metering data, and dispatch instruction data. A correlation analysis was performed on the settlement deviation value and the actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply conditions; Based on the associated parameter set, multiple simulated power supply states covering different power supply scenarios are constructed. Based on the settlement basic data, correction measures, and the simulated power supply states, the operation rules of the pre-constructed initial settlement model are corrected to obtain the corrected operation rules. The settlement basis data in the target pairing data under each simulated power supply state is settled using the correction operation rules to obtain the correction result corresponding to each simulated power supply state. The correction result is compared with the historical correction result to generate a set of deviation values. The deviation values ​​in the deviation value set are compared with the preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, so as to realize the verification test of the initial settlement model. The preset deviation threshold includes an absolute deviation threshold and a relative deviation threshold.

2. The method according to claim 1, characterized in that, A correlation analysis is performed between the settlement deviation value and the actual power supply data in the electricity market to obtain a set of correlation parameters covering different power supply conditions, including: The settlement deviation value is bound to the actual power supply data at the corresponding time point in the electricity market to form multiple pairs of deviation status data. The actual power supply data includes power supply voltage data, power supply frequency data, and power supply load data. Based on the preset power supply load range, the deviation state data in the same power supply load range are grouped to obtain the grouping results under different power supply conditions. The actual power supply data in each grouping result is used as the power distribution parameters of the corresponding grouping result. Calculate the proportion of the settlement deviation value in each group's results to the corresponding correction value in the correction measures, and use the proportion as the deviation correction coefficient for each group's results; The distribution parameters and corresponding deviation correction coefficients of each group's results are linked and bound to form a set of associated parameters covering different power supply conditions.

3. The method according to claim 2, characterized in that, The distribution parameters of each group's results are correlated and bound with the corresponding deviation correction coefficients to form a set of correlated parameters covering different power supply states, including: Assign a corresponding status identifier to each group of group results to obtain a group result set with identifiers. The status identifiers include the power supply voltage range, the power supply frequency range, and the power supply load range. Determine whether the grouped distribution parameters and deviation correction coefficients of each group with status indicators share the same status indicator. If they share the same status indicator, the group is considered to be successfully matched. The distribution parameters and deviation correction coefficients that are determined to be successfully matched within the group are combined to form multiple parameter matching pairs. A corresponding status identifier is added to each parameter matching pair to form multiple groups of identified associated units. All associated units are integrated to obtain an associated parameter set covering different power supply states.

4. The method according to claim 1, characterized in that, Based on the associated parameter set, multiple simulated power supply states covering different power supply scenarios are constructed. Based on the settlement foundation data, correction measures, and the simulated power supply states, the calculation rules of the pre-constructed initial settlement model are modified to obtain modified calculation rules, including: Based on the power distribution parameters in the associated parameter set, the fluctuation range of the power distribution parameters is determined, and the fluctuation range includes a first range, a second range, and a third range; Based on a preset data value interval, voltage simulation data, frequency simulation data, and load simulation data within the fluctuation range are generated, and the voltage simulation data, frequency simulation data, and load simulation data are cross-combined in multiple dimensions to form multiple sets of simulated power distribution parameters. Based on the simulated power distribution parameters of each group, construct multiple simulated power supply states covering different power supply scenarios; The settlement basis data and the corresponding time point correction measures are linked to form multiple sets of linked data; Based on the fluctuation range, the threshold range of distribution parameters for different simulated power supply states is determined. The actual power supply data that is within the threshold range of the distribution parameters and is associated with the settlement basic data is selected as the target distribution parameter. The associated data corresponding to the target distribution parameter is used as the target pairing data for the corresponding simulated power supply state. Based on the target pairing data and simulated power distribution parameters for each simulated power supply state, the calculation rules of the pre-constructed initial settlement model are modified to obtain the modified calculation rules.

5. The method according to claim 4, characterized in that, Based on the simulated power distribution parameters of each group, multiple simulated power supply states covering different power supply scenarios are constructed, including: Each set of simulated power distribution parameters is assigned a corresponding simulated status identifier to obtain a simulated parameter combination. The simulated status identifier includes simulated voltage information, simulated frequency information, and simulated load information. Match the simulation status identifier of each simulation parameter combination with the status identifier corresponding to the deviation correction coefficient. Use the deviation correction coefficient that perfectly matches each simulation parameter combination as the target correction coefficient for the corresponding simulation parameter combination. If there is no deviation correction coefficient that perfectly matches each simulation parameter combination, use the deviation correction coefficient with the highest matching degree with each simulation parameter combination as the target correction coefficient. By associating each set of simulation parameters with the corresponding target correction coefficients, multiple sets of target association groups are obtained; Add scene description information to each target association group to form multiple scene units. The scene description information includes power supply pressure level, power supply stability level and power consumption period. Based on the scenario units, multiple simulated power supply states covering different power supply scenarios are constructed.

6. The method according to claim 1, characterized in that, The settlement basis data in the target pairing data under each simulated power supply state is settled using the aforementioned correction calculation rules to obtain the correction result corresponding to each simulated power supply state. The correction result is then compared with the historical correction result to generate a deviation value set, including: The power transaction data in the settlement basis data of the target pairing data under each simulated power supply state is processed using the modified calculation rules to obtain the calculation results corresponding to each simulated power supply state. Based on the metering data in the settlement basis data of the target pairing data under each simulated power supply state, the corresponding calculation results are supplemented and adjusted to obtain the adjusted results for each simulated power supply state. Based on the scheduling instruction data in the settlement base data of the target pairing data under each simulated power supply state, the corresponding adjusted results are processed to obtain the correction results corresponding to each simulated power supply state. Calculate the deviation between each correction result under each simulated power supply state and the historical correction result corresponding to the same time node, and integrate all deviation values ​​to obtain a deviation value set.

7. The method according to claim 1, characterized in that, The deviation values ​​in the deviation value set are compared with a preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, thereby realizing the verification test of the initial settlement model, including: Based on different simulated power supply states, the deviation value set is grouped to obtain multiple subsets of deviation values; The deviation values ​​in each deviation value subset are compared with the preset deviation threshold. Deviation values ​​that are less than the preset deviation threshold are judged as qualified deviation values. The pass rate of each deviation value subset is calculated by combining the total number of deviation values ​​in each deviation value subset. The pass rate of each subset of deviation values ​​is compared with the preset pass threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions. The verification result corresponding to a pass rate greater than or equal to the preset pass threshold is considered to have passed the verification, and the verification result corresponding to a pass rate less than the preset pass threshold is considered to have failed the verification.

8. A settlement model verification and testing system based on real-world case backtracking, characterized in that, include: The acquisition module is used to acquire case data of the electricity market. The case data includes settlement basis data, settlement deviation values, correction measures, and historical correction results. The settlement basis data includes electricity transaction data, metering data, and dispatch instruction data. The correlation module is used to perform correlation analysis on the settlement deviation value and the actual power supply data of the electricity market to obtain a set of correlation parameters covering different power supply states; The correction module is used to construct multiple simulated power supply states covering different power supply scenarios based on the associated parameter set, and to correct the calculation rules of the pre-constructed initial settlement model based on the settlement basic data, correction measures and the simulated power supply states to obtain the corrected calculation rules. The settlement module is used to perform settlement processing on the settlement base data in the target pairing data under each simulated power supply state using the correction calculation rules, to obtain the correction result corresponding to each simulated power supply state, and to compare the correction result with the historical correction result to generate a deviation value set. The comparison module is used to compare each deviation value in the deviation value set with a preset deviation threshold to obtain the verification results of the initial settlement model under different simulated power supply conditions, so as to realize the verification test of the initial settlement model. The preset deviation threshold includes an absolute deviation threshold and a relative deviation threshold.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of a settlement model verification test method based on real-world case backtracking as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables a settlement model verification test method based on real-world case backtracking as described in any one of claims 1 to 7.

Images