Power transaction settlement model version comparison and cross validation method and storage medium

Abstract

The invention relates to the field of power transaction settlement, in particular to a power transaction settlement model version comparison and cross validation method and a storage medium. A directed acyclic graph of a plurality of versions of power settlement models is established, input nodes, output nodes and operation nodes are used as constituent parts of the directed acyclic graph, the input nodes, the output nodes and the operation nodes are defined respectively, node levels of all the nodes are defined through calculation, and under the dimension of the same node level, the node levels of all the nodes are calculated. According to the method, input nodes, output nodes and operation nodes are compared, all the nodes are classified into paired identical nodes and unpaired isolated nodes, the difference of different power settlement models is visually observed, and the purpose of comparing the difference between the different power settlement models is achieved; the problems that when multiple versions of electric power settlement models coexist at present, settlement results are likely to be different, and it is difficult for business personnel to compare the differences between different electric power settlement models and verify the settlement results are solved.

Description

Technical Field

[0001] The present invention relates to the field of power transaction settlement, and in particular to a power transaction settlement model version comparison and cross-validation method and a storage medium. Background Art

[0002] In modern society, electricity trading and electricity settlement, as key links supporting the operation of the energy economy, are becoming increasingly complex. Due to the complexity and variability of the electricity settlement process, more and more flexible and configurable settlement models are introduced into the settlement process of electricity trading. Modeling and configuration allow the settlement system to dynamically adjust its operating logic and rules according to different market conditions and business needs, so as to better adapt to the diverse electricity market environment. By abstracting the settlement logic and rules into configuration items or models, business configuration personnel can configure, adjust and update them more conveniently, reducing the maintenance cost and risk of the system.

[0003] However, as settlement rules and market policies change, settlement models may need to be updated frequently, which will result in the coexistence of multiple versions of settlement models, making them difficult to track and manage. The coexistence of multiple settlement model versions may also cause compatibility issues in the settlement system during operation, or output inconsistent settlement results, making it difficult for business configuration personnel to quickly compare the differences between different versions of the electricity settlement model, and also difficult to verify the settlement results of different versions. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method and storage medium for comparing and cross-validating versions of power trading settlement models, which solves the problem that when multiple versions of power settlement models coexist, it is easy to cause differences in settlement results, and it is difficult for business personnel to compare the differences between different power settlement models and verify the settlement results.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for comparing and cross-validating power transaction settlement model versions includes the following steps:

[0007] S1. Set up several versions of the power settlement model, and establish directed acyclic graphs of each version of the power settlement model according to the power transaction settlement process;

[0008] S2. Calculate the level to which the nodes of each power settlement model in the directed acyclic graph belong, i.e., the node level;

[0009] S3. Taking the node level of each power settlement model as a dimension, compare different power settlement models, mark the same nodes of the same dimension of different power settlement models as paired identical nodes, and mark the remaining nodes as unpaired isolated nodes;

[0010] S4, inputting the original power data into each directed acyclic graph, and calculating the operation results of the power data passing through each node to obtain the power output data of each power settlement model;

[0011] S5. Select several associated fields according to the field information of the original power data, and connect each power output data in a left-connection manner by using the associated fields as connection keys;

[0012] S6. Compare a plurality of power output data, and set a first mark on the power output data whose values ​​corresponding to the same field information are within an error range, and set a second mark on the remaining power output data.

[0013] Preferably, in step S1, the following steps are specifically included:

[0014] S11. Set several versions of the electricity settlement model, and for any electricity settlement model, set all nodes with an in-degree of 0 as input nodes of the electricity settlement model, set all nodes with an out-degree of 0 as output nodes of the electricity settlement model, and set the remaining nodes as operation nodes;

[0015] S12, associating the input node with the source table, field name and field type that the power raw data calculation depends on;

[0016] S13, associating each output node with the target table, field name and field type of the power output data;

[0017] S14. According to the power transaction settlement process, all input nodes, output nodes and operation nodes of each power transaction model are associated with a directed acyclic graph.

[0018] Preferably, the operation nodes include arithmetic nodes, type conversion nodes, grouping aggregation nodes, connection nodes and conditional nodes.

[0019] Preferably, the arithmetic node is used to perform mathematical operations, expression evaluation and precision conversion;

[0020] The type conversion node is used to perform type conversion operations on fields, converting between different field types;

[0021] The grouping aggregation node is used to group the data according to conditions and perform mathematical operations;

[0022] The connection node is used to perform left connection, inner connection and full connection on different data according to keyword conditions;

[0023] The conditional node is used to output the value of the variable according to the value of the input variable and the conditional expression.

[0024] Preferably, in step S2, the following steps are specifically included:

[0025] S21. Define temporary variables: the initial level I is 1, G is the directed acyclic graph of any power settlement model i;

[0026] S22, record the set of all nodes with in-degree 0 in the directed acyclic graph G as UiI, and mark the level to which all nodes in the set UiI belong as I;

[0027] S23, remove the node set UiI and its corresponding directed edges from the directed acyclic graph G;

[0028] S24, judging whether the directed acyclic graph G is empty;

[0029] If yes, go to the next step;

[0030] If not, set I=l+1 and return to step S22.

[0031] Preferably, in step S3, the following steps are specifically included:

[0032] S31. Obtain the input nodes of the directed acyclic graph corresponding to each power settlement model. If there is any input node with the same source table, field name and field type at the same node level in each directed acyclic graph, mark the input node as a paired identical node, and mark the remaining input nodes as unpaired isolated nodes.

[0033] S32, obtaining the output nodes of the directed acyclic graph corresponding to each power settlement model, if there is any output node in the same node level of each directed acyclic graph, the target table, field name and field type are the same, then the output node is marked as a paired identical node, and the remaining output nodes are marked as unpaired isolated nodes;

[0034] S33. Traverse the operation nodes of the directed acyclic graph corresponding to each power settlement model in turn. If the operation of the operation node at the same node level in each directed acyclic graph is the same, and the operation node input to the operation node is the same and is a paired identical node, then mark the operation node as a paired identical node, and mark the remaining operation nodes as unpaired isolated nodes.

[0035] The present invention also provides a computer storage medium storing a computer program, which, when executed, implements the steps of the power transaction settlement model version comparison and cross-validation method.

[0036] Compared with the prior art, the present invention provides a method and storage medium for comparing and cross-validating power transaction settlement model versions, which has the following beneficial effects:

[0037] 1. The present invention establishes a directed acyclic graph of multiple versions of the electricity settlement model, takes input nodes, output nodes and operation nodes as components of the directed acyclic graph, and defines the input nodes, output nodes and operation nodes respectively so that the directed acyclic graph conforms to the electricity settlement process, and then extracts paired identical nodes and unpaired isolated nodes of several directed acyclic graphs to achieve the purpose of comparing the differences between different electricity settlement models, and compares the power output nodes of the nodes to achieve verification of the power output data.

[0038] 2. The present invention defines the node hierarchy of each node by calculation, and then compares the input nodes, output nodes and operation nodes under the dimension of the same node hierarchy, so as to classify each node into paired identical nodes and unpaired isolated nodes, so as to intuitively observe the differences between different electricity settlement models and realize rapid comparison between different electricity trading settlement model versions.

[0039] 3. The present invention compares the numerical value of the power output data corresponding to the same field information according to the field information of the power output data. If the numerical value is within a certain error range, a first mark is set, otherwise a second mark is set, thereby directly completing the verification of the corresponding power output data more quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0041] Figure 1 It is a flow chart of the method for comparing and cross-validating versions of the power transaction settlement model of the present invention;

[0042] Figure 2 is a schematic diagram of a conventional directed acyclic graph;

[0043] Figure 3 It is a schematic diagram of a directed acyclic graph of the price calculation model version V1 of the user-side medium- and long-term transaction deviation income recovery price of the present invention;

[0044] Figure 4 It is a schematic diagram of the directed acyclic graph of version V2 of the user-side medium- and long-term transaction deviation profit recovery price calculation model of the present invention. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0046] Those skilled in the art can understand that all or part of the steps in the following embodiments can be completed by instructing the relevant hardware through a program, so the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0047] In order to solve the problem that when multiple versions of power settlement models coexist, it is easy to cause differences in settlement results, and it is difficult for business personnel to compare the differences between different power settlement models and verify the settlement results, this embodiment provides a power transaction settlement model version comparison and cross-validation method, which includes the following steps:

[0048] S1. Set up several versions of the power settlement model, and establish directed acyclic graphs of each version of the power settlement model according to the power transaction settlement process; in order to further illustrate the process of establishing corresponding directed acyclic graphs according to the power transaction settlement process and the power settlement model, as shown in Figure 2 As shown, it is a schematic diagram of a conventional directed acyclic graph. Figure 3 and Figure 4 It is a schematic diagram of a directed acyclic graph of power settlement established according to the power transaction settlement process and the power settlement model, corresponding to the user-side medium- and long-term transaction deviation income recovery price calculation model versions V1 and V2, that is, two power settlement models. In step S1, it specifically includes the following steps:

[0049] S11. Set up several versions of electricity settlement models, and for any electricity settlement model, set all nodes with an in-degree of 0 as input nodes of the electricity settlement model, set all nodes with an out-degree of 0 as output nodes of the electricity settlement model, and set the remaining nodes as operation nodes; the operation nodes include arithmetic nodes, type conversion nodes, grouping and aggregation nodes, connection nodes and conditional nodes; arithmetic nodes can perform mathematical operations such as addition, subtraction, multiplication, division, and exponentiation, and can also perform expression evaluation, precision conversion, etc.; type conversion nodes can perform field type conversion operations and convert between different field types; grouping and aggregation nodes can group data according to conditions and sum, average, find maximum and minimum values, count, remove duplicate counts, etc.; connection nodes can left-join, inner-join, fully-join, etc. different data according to keyword conditions; conditional nodes can determine the value of output variables based on the values ​​of input variables and conditional expressions.

[0050] S12, associating the input node with the source table, field name and field type that the power raw data calculation depends on;

[0051] S13, associating each output node with the target table, field name and field type of the power output data;

[0052] S14. According to the electricity trading settlement process, all input nodes, output nodes and operation nodes of each electricity trading model will be associated with a directed acyclic graph. For any nodes Xij and Xik in any version of the electricity settlement model i, if there is a directed edge from node Xij to node Xik, it means that the operation input of node Xik depends on the output of node Xij operation. If node Xij is an input node, it means that node Xik depends on the input field specified by node Xij; if node Xik is an output node, it means that the operation result of node Xij is output to the output target field specified by node Xik.

[0053] like Figure 3 and Figure 4 As shown, the input user-side medium- and long-term transaction deviation revenue recovery price calculation model versions V1 and V2, in model version V1, all actual electricity consumption participates in the calculation of the user-side medium- and long-term transaction deviation revenue recovery price, in model version V2, the actual electricity consumption during the period when the unified node price of the market is lower than the transaction price of electricity in the power grid purchasing market participates in the calculation of the user-side medium- and long-term transaction deviation revenue recovery price.

[0054] The input nodes of model version V1 are the monthly centralized competitive transaction price, medium- and long-term transaction deviation adjustment coefficient, actual electricity consumption during the period, and the unified settlement point electricity price field from the settlement basic table. They are integer types, and the medium- and long-term transaction deviation adjustment coefficient, monthly centralized competitive transaction price, actual electricity consumption during the period, and unified settlement point electricity price during the day are floating point types; the total electricity node is a group aggregation node, and the SUM function represents the sum of variables, and the total electricity = ∑ actual electricity consumption during the period; the total electricity fee node of the period is an arithmetic node, and the total electricity fee of the period = actual electricity consumption during the period * unified settlement point electricity price during the day; the total fee node is a group aggregation node, and the total fee = ∑ total electricity fee during the period; the weighted average comprehensive monthly market value on the day before the market is the average monthly market value. The total electricity price node is a conditional node and an output node. The ROUND(a,b) function means that a is rounded to b decimal places. If the total electricity quantity>0, the monthly weighted average comprehensive electricity price of the market on the day before = total cost / total electricity quantity, retaining six decimal places, otherwise the monthly weighted average comprehensive electricity price of the market on the day before = 0; the medium- and long-term transaction deviation income recovery price node is a conditional node and an output node. If the monthly centralized competitive transaction price>the monthly weighted average comprehensive electricity price of the market on the day before, the medium- and long-term transaction deviation income recovery price = (monthly centralized competitive transaction price-monthly weighted average comprehensive electricity price of the market on the day before)*medium- and long-term transaction deviation adjustment coefficient, otherwise the medium- and long-term transaction deviation income recovery price = 0.

[0055] The input nodes of model version V2 are the monthly centralized competitive transaction price, medium- and long-term transaction deviation adjustment coefficient, actual electricity consumption during the period, power grid purchasing market electricity transaction price and unified settlement point electricity price field from the settlement basic table, which are integer types, and the medium- and long-term transaction deviation adjustment coefficient, monthly centralized competitive transaction price, actual electricity consumption during the period, power grid purchasing market electricity transaction price and unified settlement point electricity price during the day are floating point types; whether to participate in the calculation node is a conditional node. If the unified settlement point electricity price during the day is less than the power grid purchasing market electricity transaction price, whether to participate in the calculation = 1, otherwise whether to participate in the calculation = 0; the total electricity fee node for the period is an arithmetic node. The total electricity fee for the period = actual electricity consumption during the period * unified settlement point electricity price during the day; the total electricity node is a group aggregation node. GROUPSUM(a,b) means grouping by a and summing b. The total electricity will be grouped according to whether to participate in the calculation, and the actual electricity consumption during the period will be calculated. The electricity consumption is summed up; the total cost is a group aggregation node, and the total cost will be grouped according to whether it participates in the calculation, and the total electricity cost of the time period is summed up; the monthly weighted average comprehensive electricity price node of the day-ahead market is a conditional node and an output node, and the ROUND(a,b) function means that a is rounded to b decimal places. If whether to participate in the calculation = 1 and the total electricity consumption>0, then the monthly weighted average comprehensive electricity price of the day-ahead market = total cost / total electricity consumption, retaining six decimal places, otherwise the monthly weighted average comprehensive electricity price of the day-ahead market = 0; the medium- and long-term transaction deviation income recovery price node is a conditional node and an output node. If the monthly centralized competitive transaction price is greater than the monthly weighted average comprehensive electricity price of the day-ahead market, the medium- and long-term transaction deviation income recovery price = (monthly centralized competitive transaction price-monthly weighted average comprehensive electricity price of the day-ahead market)*medium- and long-term transaction deviation adjustment coefficient, otherwise the medium- and long-term transaction deviation income recovery price = 0.

[0056] S2. Calculate the level to which the nodes of each power settlement model belong in the directed acyclic graph, that is, the node level; in order to further illustrate the calculation method of the level to which the nodes in the directed acyclic graph belong, in step S2, the following steps are specifically included:

[0057] S21. Define temporary variables: the initial level I is 1, G is the directed acyclic graph of any power settlement model i;

[0058] S22, record the set of all nodes with in-degree 0 in the directed acyclic graph G as UiI, and mark the level to which all nodes in the set UiI belong as I;

[0059] S23, remove the node set UiI and its corresponding directed edges from the directed acyclic graph G;

[0060] S24, judging whether the directed acyclic graph G is empty;

[0061] If yes, go to the next step;

[0062] If not, set I=l+1 and return to step S22.

[0063] Taking the model version V1 as an example: the four nodes of monthly centralized competitive transaction price, medium- and long-term transaction deviation adjustment coefficient, actual power consumption during the period and unified settlement point electricity price on the day before have an entry degree of 0, so they belong to the first layer. Figure 3 Remove the first-layer nodes and the corresponding directed edges, and the in-degree of the total electricity node and the total electricity fee node becomes 0, so it belongs to the second layer; Figure 3 Continue to remove the second-layer nodes and the corresponding directed edges, and the in-degree of the total cost node becomes 0, so it belongs to the third layer; Figure 3 Continue to remove the third-layer nodes and the corresponding directed edges, and the in-degree of the day-ahead monthly weighted average comprehensive electricity price node becomes 0, so it belongs to the fourth layer; Figure 3 Continue to remove the 4th layer nodes and the corresponding directed edges, and the in-degree of the medium- and long-term trading deviation income recovery price node becomes 0, so it belongs to the 5th layer; at this time Figure 3 If it is empty, the level to which the node belongs is calculated. The calculation process of the level to which the node belongs in model version V2 is similar and will not be repeated here.

[0064] S3. Taking the node level of each power settlement model as a dimension, different power settlement models are compared, and the same nodes of the same dimension of different power settlement models are marked as paired identical nodes, and the remaining nodes are marked as unpaired isolated nodes; for different types of nodes, especially operation nodes, determining whether they are the same is not only related to the calculation method, but also to the data source. There are differences in the determination of paired identical nodes. Therefore, in step S3, the following steps are specifically included:

[0065] S31. Obtain the input nodes of the directed acyclic graph corresponding to each power settlement model. If there is any input node with the same source table, field name and field type at the same node level in each directed acyclic graph, mark the input node as a paired identical node, and mark the remaining input nodes as unpaired isolated nodes.

[0066] S32, obtaining the output nodes of the directed acyclic graph corresponding to each power settlement model, if there is any output node in the same node level of each directed acyclic graph, the target table, field name and field type are the same, then the output node is marked as a paired identical node, and the remaining output nodes are marked as unpaired isolated nodes;

[0067] S33. Traverse the operation nodes of the directed acyclic graph corresponding to each power settlement model in turn. If the operation of the operation node at the same node level in each directed acyclic graph is the same, and the operation node input to the operation node is the same and is a paired identical node, then mark the operation node as a paired identical node, and mark the remaining operation nodes as unpaired isolated nodes.

[0068] like Figure 3 and Figure 4 As shown, it is obvious that in the first layer, the monthly centralized competitive transaction price, the medium- and long-term transaction deviation adjustment coefficient, the actual electricity consumption during the period, and the unified settlement point electricity price on the day before belong to the paired nodes, and the remaining nodes are unpaired isolated nodes. Business configuration personnel can quickly identify the differences defined between different model versions based on the visualization image results.

[0069] S4, inputting the original power data into each directed acyclic graph, and calculating the operation results of the power data passing through each node to obtain the power output data of each power settlement model;

[0070] In order to further illustrate the calculation method of step S4, the following Table 1 is used as power input data to be respectively imported into the user-side medium- and long-term transaction deviation income recovery price calculation model versions V1 and V2 for calculation, and the power output data of each model version is obtained as shown in Table 2 and Table 3;

[0071] Table 1 Power input data

[0072]

[0073] Table 2 Output results of the price calculation model for medium- and long-term transaction deviation income recovery on the user side, version V1

[0074]

[0075]

[0076] Table 3 Output results of the price calculation model for medium- and long-term transaction deviation income recovery on the user side, version V2

[0077]

[0078] S5. Select several associated fields according to the field information of the original power data, and use the associated fields as connection keys to connect the various power output data in a left connection manner; the results obtained after the left connection are shown in Tables 4 and 5 below:

[0079] Table 4 Left join results of the user-side medium- and long-term transaction deviation income recovery price calculation model version V1 and V2 output (same part)

[0080]

[0081] Table 5 Left join results of the user-side medium- and long-term transaction deviation income recovery price calculation model version V1 and V2 output (different parts)

[0082]

[0083]

[0084] S6. Compare several power output data, and set a first mark on the power output data whose values ​​corresponding to the same field information are within the error range, and set a second mark on the remaining power output data, as shown in Table 4. The same parts of the outputs of the user-side medium- and long-term transaction deviation income recovery price calculation model versions V1 and V2 are represented by traditional Chinese characters, which is the first mark, and the different parts are represented by italic bold characters, which is the second mark. Of course, other methods can also be used for marking, such as font size and color, or special symbols.

[0085] The present invention also provides a computer storage medium storing a computer program, which, when executed, implements the steps of the power transaction settlement model version comparison and cross-validation method.

[0086] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for comparing and cross-validating versions of a power trading settlement model, characterized in that: The steps include: S1. Set up several versions of the power settlement model, and establish directed acyclic graphs of each version of the power settlement model according to the power transaction settlement process; S2. Calculate the level to which the nodes of each power settlement model in the directed acyclic graph belong, i.e., the node level; S3. Taking the node level of each power settlement model as a dimension, compare different power settlement models, mark the same nodes of the same dimension of different power settlement models as paired identical nodes, and mark the remaining nodes as unpaired isolated nodes; S4, inputting the original power data into each directed acyclic graph, and calculating the operation results of the power data passing through each node to obtain the power output data of each power settlement model; S5. Select several associated fields according to the field information of the original power data, and connect each power output data in a left-connection manner by using the associated fields as connection keys; S6. Compare a plurality of power output data, and set a first mark on the power output data whose values ​​corresponding to the same field information are within an error range, and set a second mark on the remaining power output data.

2. The comparison and cross-validation method according to claim 1, characterized in that: In step S1, the following steps are specifically included: S11. Set several versions of the electricity settlement model, and for any electricity settlement model, set all nodes with an in-degree of 0 as input nodes of the electricity settlement model, set all nodes with an out-degree of 0 as output nodes of the electricity settlement model, and set the remaining nodes as operation nodes; S12, associating the input node with the source table, field name and field type that the power raw data calculation depends on; S13, associating each output node with the target table, field name and field type of the power output data; S14. According to the power transaction settlement process, all input nodes, output nodes and operation nodes of each power transaction model are associated with a directed acyclic graph.

3. The comparison and cross-validation method according to claim 2, characterized in that: The operation nodes include arithmetic nodes, type conversion nodes, grouping aggregation nodes, connection nodes and conditional nodes.

4. The comparison and cross-validation method according to claim 3, characterized in that: The arithmetic node is used to perform mathematical operations, expression evaluation and precision conversion; The type conversion node is used to perform type conversion operations on fields, converting between different field types; The grouping aggregation node is used to group the data according to conditions and perform mathematical operations; The connection node is used to perform left connection, inner connection and full connection on different data according to keyword conditions; The conditional node is used to output the value of the variable according to the value of the input variable and the conditional expression.

5. The comparison and cross-validation method according to claim 1, characterized in that: In step S2, the following steps are specifically included: S21. Define temporary variables: the initial level I is 1, G is the directed acyclic graph of any power settlement model i; S22, record the set of all nodes with in-degree 0 in the directed acyclic graph G as UiI, and mark the level to which all nodes in the set UiI belong as I; S23, remove the node set UiI and its corresponding directed edges from the directed acyclic graph G; S24, judging whether the directed acyclic graph G is empty; If yes, go to the next step; If not, set I=l+1 and return to step S22.

6. The comparison and cross-validation method according to claim 1, characterized in that: In step S3, the following steps are specifically included: S31. Obtain the input nodes of the directed acyclic graph corresponding to each power settlement model. If there is any input node with the same source table, field name and field type at the same node level in each directed acyclic graph, mark the input node as a paired identical node, and mark the remaining input nodes as unpaired isolated nodes. S32, obtaining the output nodes of the directed acyclic graph corresponding to each power settlement model, if there is any output node in the same node level of each directed acyclic graph, the target table, field name and field type are the same, then the output node is marked as a paired identical node, and the remaining output nodes are marked as unpaired isolated nodes; S33. Traverse the operation nodes of the directed acyclic graph corresponding to each power settlement model in turn. If the operation of the operation node at the same node level in each directed acyclic graph is the same, and the operation node input to the operation node is the same and is a paired identical node, then mark the operation node as a paired identical node, and mark the remaining operation nodes as unpaired isolated nodes.

7. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed, the steps of the power transaction settlement model version comparison and cross-validation method according to any one of claims 1 to 6 are implemented.

Images