Electric power spot market clearing and checking method and system

Abstract

The invention provides an electric power spot market clearing and checking method and system, and the method comprises the steps: building an electric power market checking system which comprises core checking conditions needing to be checked at different market levels; establishing a three-dimensional collaborative check function group; constructing a collaborative check model according to the electricity market check system and the three-dimensional collaborative check function group; constructing a case knowledge base containing a plurality of historical algorithm cases, matching historical algorithm cases similar to the current liquidation method in the case knowledge base, and generating a target check strategy and an initial logic verification result according to the historical algorithm cases; performing weight adjustment on the collaborative check model according to the initial logic verification result so as to generate an evaluation result for the existing clearing method by using the collaborative check model; and based on the checking strategy, executing clearing checking. According to the method and the device, a data closed loop from real-time data acquisition to check strategy generation and execution is realized, and the check efficiency and the resource utilization rate are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity spot market, and in particular to a method and system for clearing and verifying electricity spot market. Background Art

[0002] Unlike other markets, the clearing of the electricity market is not only a market-level equilibrium, but is also subject to the constraints of grid operation, such as SCED constraints and node marginal electricity price conditions. Due to the exponential increase in the complexity of the current power grid system and market, it is no longer feasible to rely on manual or traditional statistical models to verify the clearing algorithm. Therefore, in recent years, the power system has begun to introduce automated testing tools to achieve batch comparison of calculation results with historical data, but lacks penetrating verification of the algorithm's internal logic; some systems have begun to integrate logical rule checking modules, but each verification dimension (logic / results / efficiency) is still in an isolated verification state. For the verification of the algorithm for clearing the electricity spot market, the existing technology generally has the following defects or deficiencies:

[0003] 1) The fragmentation of verification dimensions leads to verification blind spots. Lack of computational logic verification: Existing tools only test input-output relationships (e.g., the macro-correlation between SCUC clearing prices and load), but fail to verify the inherent coupling of constraints (e.g., the mathematical consistency between unit ramp rates and network security constraints). Efficiency and accuracy are disconnected: Fixed thresholds are used to assess computational time (e.g., requiring less than 2 hours), and no adaptive correlation model between efficiency indicators and grid size is established (e.g., benchmarking the difference between 500 and 5,000 nodes).

[0004] 2) Lack of a multi-algorithm collaborative evaluation mechanism. Fragmented verification standards: Independent verification systems were used for different clearing stages, leading to conflicting standards during cross-algorithm optimization. Dynamic coupling failure: The verification chain from unit commitment to day-ahead scheduling to real-time scheduling was not established, resulting in a lack of verification of time-period coupling constraints (e.g., the continuity of minimum unit operating time across time periods).

[0005] 3) Bottlenecks in real-time adaptability and knowledge reuse. Static assessment limitations: The current verification system uses a static test set of historical cases (e.g., the N-1 standard fault scenario), which cannot respond to dynamic topological changes caused by fluctuations in renewable energy penetration (±30%). Insufficient knowledge accumulation: Verification report generation relies on templated documents (with a manual completion efficiency of <5 cases / person-day), and there is no graphical correlation mechanism for abnormal cases (e.g., the implicit correlation between "node electricity price inversion" and "network congestion mode").

[0006] The above problems make it difficult for the existing evaluation and verification of clearing algorithms to adapt to the coordination of multiple links in the market, resulting in market operation risks and inefficient resource utilization. Summary of the Invention

[0007] To this end, the present invention provides a method and system for clearing and verification of the electricity spot market, aiming to solve the technical problem in the prior art that the evaluation and verification of the existing clearing algorithm are difficult to adapt to the connection of multi-link market collaboration, resulting in market operation risks and inefficient resource utilization.

[0008] To achieve the above objectives, the present invention adopts the following technical solutions:

[0009] According to a first aspect of the present invention, the present invention provides a method for clearing and verifying electricity spot markets, the method comprising:

[0010] Establishing a power market verification system that includes core verification conditions for different market levels; and establishing a three-dimensional collaborative verification function set that considers logic verification, accuracy, and efficiency;

[0011] Constructing a collaborative verification model based on the power market verification system and the three-dimensional collaborative verification function group; the collaborative verification model is used to perform a multi-dimensional evaluation of the clearing algorithm to obtain an evaluation result;

[0012] Construct a case knowledge base containing multiple historical algorithm cases, match historical algorithm cases similar to the current clearing algorithm in the case knowledge base, and generate a target verification strategy and initial logic verification results based on the historical algorithm cases;

[0013] The collaborative verification model is weighted according to the initial logic verification result to generate an evaluation result for an existing clearing algorithm using the collaborative verification model; and a clearing verification is performed based on the verification strategy.

[0014] Furthermore, the electricity market verification system includes:

[0015] The first core verification condition that must be verified for the safety-constrained unit commitment (SCUC) in the day-ahead market; the first core verification condition includes at least one of the unit start-stop logic, ramp constraint coupling, and network security constraint: and / or,

[0016] A second core verification condition that must be verified for security-constrained economic dispatch (SCED) in the real-time market; the second core verification condition includes at least one of real-time power balance, node electricity price rationality, and dynamic adaptability of network topology: and / or,

[0017] A third core verification condition that must be verified by the frequency regulation resource clearing algorithm FRAC for the frequency regulation auxiliary market; the third core verification condition includes at least one of frequency regulation capacity allocation, response rate constraint, and cross-time capacity coupling: and / or,

[0018] The fourth core verification condition that the reserve capacity clearing algorithm SRAC for the reserve auxiliary market must verify; the fourth core verification condition includes at least one of the following: the availability of reserve capacity, the rationality of its spatiotemporal distribution, and its synergy with the main energy market: and / or,

[0019] The fifth core verification condition of the multi-energy collaborative clearing algorithm MEUC for multi-energy joint clearing; the fifth core verification condition includes at least one of wind and solar prediction error processing, energy storage charging and discharging logic, and multi-energy complementarity verification.

[0020] Furthermore, the establishment of a three-dimensional collaborative verification function group considering logic verification, accuracy and efficiency includes:

[0021] Define the three-dimensional collaborative verification function group, the mathematical expression is:

[0022] Φ=[Φ L ,Φ A ,Φ E ] T

[0023] Where Φ represents the three-dimensional collaborative verification function group; T represents the vector transpose; Φ L Represents the logical verification function of the clearing algorithm; Φ A represents the accuracy evaluation function of the clearing algorithm; φ E is the efficiency evaluation function of the clearing algorithm;

[0024] The logic verification function φ L The mathematical expression is:

[0025]

[0026] Among them, N c Indicates the total number of constraints; g i (x * ) indicates that the i-th constraint condition is in the optimal solution x * The deviation of L Represents the logic tolerance threshold; represents a dummy variable function;

[0027] The accuracy evaluation function Φ A The mathematical expression is:

[0028]

[0029] Among them, α and β represent weight coefficients; P calc represents the power vector; P ref represents the system reference power vector that can be obtained from historical data; ‖·‖2 represents the quadratic norm; P base represents the reference power vector; DKL (·‖·) represents the Kullback-Leibler divergence; q LMP represents the distribution of node marginal electricity price; q bench represents the distribution of benchmark electricity prices;

[0030] The efficiency evaluation function Φ E The mathematical expression is:

[0031]

[0032] Among them, λ and represents the adjustment parameter; T represents the actual time consumed by the evaluated algorithm; M represents the peak memory usage when the evaluated algorithm is running; M cap Indicates the hardware memory capacity for loading the algorithm; T max (N) represents the maximum computation time when the number of nodes is N. The mathematical expression is: T max (N) = aN b ; a and b represent the parameters obtained by fitting the experimental data.

[0033] Furthermore, the mathematical expression of the collaborative verification model is:

[0034]

[0035] Among them, Score k represents the evaluation result of the multi-dimensional evaluation of the k-th market clearing algorithm, k∈{SCUC,SCED,FRAC,SRAC,MEUC}; m represents the verification evaluation dimension, including consideration of logic verification, accuracy and efficiency; It represents the calibration weight of the market clearing algorithm k under the mth calibration evaluation dimension. The mathematical expression is:

[0036]

[0037] in, represents the contribution of market clearing algorithm k under the mth verification evaluation dimension; It represents the total contribution of the market clearing algorithm k under the third verification evaluation dimension.

[0038] Furthermore, matching historical algorithm cases similar to the current clearing algorithm in the case knowledge base includes:

[0039] The case feature vector in the case knowledge base is defined as follows:

[0040]

[0041] in, is the feature vector of the key feature index of the jth historical algorithm case in the case knowledge base; N j represents the total number of nodes corresponding to the jth historical algorithm case; represents the load fluctuation rate corresponding to the jth historical algorithm case; represents the new energy penetration rate corresponding to the jth historical algorithm case; τ j represents the time scale; Γ j represents the abnormal or fault scene code; N case represents the total number of historical algorithm cases in the case knowledge base;

[0042] Based on the case feature vector, the similarity between the current market clearing algorithm and the historical algorithm case in the case knowledge base is queried, and the mathematical expression is:

[0043]

[0044] wherein, Sim(C q ,C j ) represents the similarity between the market clearing algorithm and the historical algorithm case; C q represents the feature vector input when checking the market clearing algorithm; w i represents the weight of the i th feature dimension in the feature vector; represents the cosine similarity between the market clearing algorithm and the historical algorithm case in the i th feature dimension; and respectively represent the numerical feature value of the market clearing algorithm q and the historical algorithm case j in the i th feature dimension;

[0045] Based on the similarity between the market clearing algorithm and the historical algorithm case, the historical algorithm case similar to the current clearing algorithm in the case knowledge base is screened.

[0046] Further, the target checking strategy and the initial logical verification result generated according to the historical algorithm case include:

[0047] The target checking strategy is generated according to the historical algorithm case and the current clearing algorithm, and the mathematical expression is:

[0048]

[0049] wherein, represents the feature of the current market clearing algorithm; represents the feature of the historical algorithm case; f(·) represents the function form of the target checking strategy; represents the initial verification logic result, and the mathematical expression is:

[0050]

[0051] Among them, N con represents the total number of all constraints; x0 represents the initial solution of the market clearing algorithm; g k (x0) represents the deviation of the kth constraint from the initial solution x0; Represents the initial value of the logical tolerance threshold under the kth constraint condition.

[0052] Furthermore, the weight adjustment of the collaborative verification model according to the initial logic verification result includes:

[0053] The target weight of the collaborative verification model is calculated using the initial logic verification results. The mathematical expression is:

[0054]

[0055] in, represents the calibration weight of the market clearing algorithm k under the mth calibration evaluation dimension; represents the contribution of market clearing algorithm k under the mth verification evaluation dimension; δ represents the case similarity The adjustment parameters of Represents the characteristics of the current market clearing algorithm; Represents the characteristics of historical algorithm cases; Indicates the initial validation logic result

[0056] The target weight is used to adjust the weight of the collaborative verification model.

[0057] Furthermore, the method further comprises:

[0058] Based on the rule reasoning and case comparison for the check anomaly of the clearing algorithm, the case features in the case knowledge base are updated to optimize the target check strategy. The mathematical expression is:

[0059]

[0060] in, updating case features in the case knowledge base; is the original case feature in the case knowledge base; γ i Represents the learning parameters of the i-th feature dimension; represents the i-th characteristic component of the j-th historical algorithm case; Indicates the feature vector constructed by the newly added data.

[0061] Furthermore, the method further comprises:

[0062] Establish dynamic association rules and collaborative optimization mechanisms between different verification and evaluation dimensions, including:

[0063] Define the coupling relationship between the logic verification function and the accuracy evaluation function. The mathematical expression is:

[0064]

[0065] Among them, Φ L Represents the logical verification function of the clearing algorithm; Φ A Denotes the accuracy evaluation function of the clearing algorithm; D KL represents the Kullback-Leibler divergence; g i represents the deviation of the i-th constraint; x represents the optimal solution;

[0066] Given a logic verification threshold, a Pareto frontier optimization model is constructed, and the mathematical expression is:

[0067]

[0068] Among them, λ1 and λ2 represent weight coefficients; θ represents the set of adjustable parameters; Φ E is the efficiency evaluation function of the clearing algorithm;

[0069] By solving the coupling relationship between the logic verification function and the accuracy evaluation function and the Pareto frontier optimization model, the internal correlation and dynamic balance of the verification function are performed to achieve adaptive optimization of the verification strategy.

[0070] According to a second aspect of the present invention, the present invention provides a power spot market clearing verification system, the system comprising:

[0071] A verification model building module is used to establish a power market verification system, which includes core verification conditions that must be verified at different market levels; and to establish a three-dimensional collaborative verification function group that considers logic verification, accuracy and efficiency;

[0072] a multi-dimensional collaborative evaluation module, configured to construct a collaborative verification model based on the power market verification system and the three-dimensional collaborative verification function group; the collaborative verification model is configured to perform a multi-dimensional evaluation of the clearing algorithm to obtain an evaluation result;

[0073] A verification strategy generation module is used to build a case knowledge base containing multiple historical algorithm cases, match historical algorithm cases similar to the current clearing algorithm in the case knowledge base, and generate a target verification strategy and initial logic verification results based on the historical algorithm cases;

[0074] An algorithm verification execution module is used to adjust the weight of the collaborative verification model according to the initial logic verification result, so as to generate an evaluation result for the existing clearing algorithm using the collaborative verification model; and to perform clearing verification based on the verification strategy.

[0075] The present invention adopts the above technical solution and has at least the following beneficial effects:

[0076] The present invention provides a method and system for clearing and verification of a power spot market, comprising establishing a power market verification system, wherein the power market verification system includes core verification conditions that must be verified at different market levels; and establishing a three-dimensional collaborative verification function group that takes into account logic verification, accuracy, and efficiency; constructing a collaborative verification model based on the power market verification system and the three-dimensional collaborative verification function group; constructing a case knowledge base containing multiple historical algorithm cases, matching historical algorithm cases similar to the current clearing algorithm in the case knowledge base, and generating a target verification strategy and an initial logic verification result based on the historical algorithm case; adjusting the weight of the collaborative verification model based on the initial logic verification result to generate an evaluation result for the existing clearing algorithm using the collaborative verification model; and executing clearing verification based on the verification strategy. Through the present invention, a data closed loop is achieved from real-time data acquisition to verification strategy generation and execution, which greatly improves verification efficiency and resource utilization.

[0077] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0079] Figure 1 A schematic diagram showing a flow chart of a method for clearing and verifying electricity spot markets provided by an embodiment of the present invention is shown;

[0080] Figure 2 A schematic structural diagram of a power spot market clearing verification system provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0081] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0082] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, the elements defined by the phrase "comprising..." do not exclude the presence of other identical elements in the process, method, article, or device comprising the elements.

[0083] The embodiment of the present invention provides a method for checking the clearing of the electricity spot market, such as Figure 1 As shown, it may at least include the following steps S101 to S104:

[0084] Step S101: Establishing a power market verification system, which includes core verification conditions that must be verified at different market levels; and establishing a three-dimensional collaborative verification function group that considers logic verification, accuracy, and efficiency.

[0085] First, it is necessary to confirm the clearing conditions of the electricity spot market. Since the spot market includes two stages, day-ahead and intraday, in order to expand the scope of application, the embodiment of the present invention incorporates the clearing conditions of the ancillary service market. The verification conditions that need to be tested at the market level are shown in Table 1 below:

[0086] Table 1 (Core market conditions to be verified for electricity market clearing)

[0087] Market level condition Full name Core verification conditions Day-ahead market SCUC Safety restraint unit combination Unit start and stop logic, ramp constraint coupling, network security constraints (such as line flow) Live Market SCED Safety-constrained economic dispatch Real-time power balance, reasonableness of node electricity price (LMP), dynamic adaptability of network topology FM auxiliary market FRAC Frequency modulation resource clearing algorithm Frequency regulation capacity allocation, response rate constraints, and cross-time capacity coupling Backup auxiliary market SRAC Spare capacity clearing algorithm Availability of spare capacity, rationality of spatial and temporal distribution, and synergy with the main energy market Multi-energy joint clearing MEUC Multi-energy collaborative clearing algorithm (including new energy) Wind and solar forecast error processing, energy storage charging and discharging logic, and multi-energy complementarity verification

[0088] Furthermore, at the algorithm level, the embodiment of the present invention establishes a three-dimensional collaborative verification function group that takes into account logic verification, accuracy, and efficiency. The mathematical expression can be:

[0089] Φ=[Φ L ,Φ A ,Φ E ] T (1)

[0090] Where Φ represents the three-dimensional collaborative verification function group; T represents the vector transpose; Φ L Represents the logical verification function of the clearing algorithm; Φ A represents the accuracy evaluation function of the clearing algorithm; Φ E is the efficiency evaluation function of the clearing algorithm.

[0091] Specifically, with respect to the logic verification function, since the clearing of the electricity market must obey the constraints of boundary conditions, the embodiment of the present invention can start from the constraint satisfaction and construct the following logic verification function:

[0092]

[0093] Among them, N c Indicates the total number of constraints; g i (x * ) indicates that the i-th constraint condition is in the optimal solution x * The deviation of L Indicates the logic tolerance threshold, that is, if the threshold is exceeded, the logic verification is considered to have failed. The typical reference value is 10 -6 ; Represents a dummy variable function, which takes the value 1 when the conditions in the brackets are met, otherwise it takes the value 0.

[0094] For the accuracy evaluation function, the embodiment of the present invention can start from the core power and electricity price and construct the following metric function based on multi-dimensional deviation:

[0095]

[0096] Among them, α and β represent the weight coefficients of the corresponding items respectively; P calc represents the calculated power vector; P ref represents the system reference power vector that can be obtained from historical data; ‖·‖2 represents the quadratic norm; P base Denotes the reference power vector, usually the total load; D KL (·‖·) represents the Kullback-Leibler divergence, which is used to measure the difference in the distribution of the two variables in the brackets; q LMP represents the distribution of node marginal electricity price; q bench Represents the distribution of benchmark electricity prices.

[0097] For the efficiency evaluation function, the embodiment of the present invention adopts elastic efficiency to construct it, and its mathematical expression is:

[0098]

[0099] Among them, λ and Indicates the adjustment parameter (reference value is λ=2.5, ); T represents the actual time consumed by the evaluated algorithm; M represents the peak memory usage when the evaluated algorithm is running; M cap Indicates the hardware memory capacity for loading the algorithm; T max (N) represents the maximum computation time when the number of nodes is N. The calculation formula can be:

[0100] T max (N) = aN b (5)

[0101] Among them, parameters a and b can be obtained by fitting experimental data.

[0102] Step S102 , constructing a collaborative verification model based on the power market verification system and the three-dimensional collaborative verification function group; the collaborative verification model is used to perform a multi-dimensional evaluation of the clearing algorithm to obtain an evaluation result.

[0103] After completing the function construction of the above three verification and evaluation dimensions, the embodiment of the present invention can obtain a collaborative verification model for the power market clearing algorithm, and the mathematical expression is:

[0104]

[0105] Among them, Score k represents the evaluation result of the multi-dimensional evaluation of the k-th market clearing algorithm, k∈{SCUC,SCED,FRAC,SRAC,MEUC}; m represents the verification evaluation dimension, including consideration of logic verification, accuracy and efficiency; It represents the calibration weight of the market clearing algorithm k under the mth calibration evaluation dimension, which can be determined by the analytic hierarchy process (AHP). The mathematical expression can be:

[0106]

[0107] in, represents the contribution of market clearing algorithm k under the mth verification evaluation dimension; It represents the total contribution of the market clearing algorithm k under the third verification evaluation dimension.

[0108] Step S103: construct a case knowledge base containing multiple historical algorithm cases, match historical algorithm cases similar to the current clearing algorithm in the case knowledge base, and generate a target verification strategy and initial logic verification results based on the historical algorithm cases.

[0109] Steps S101 and S102 complete a multi-dimensional evaluation of the market clearing algorithm. Based on the evaluation results, embodiments of the present invention can calibrate existing clearing algorithms as needed. First, a case knowledge base containing multiple historical algorithm cases is constructed. By comparing the characteristics of existing clearing algorithms with the case base, anomaly diagnosis of existing clearing algorithms can be performed.

[0110] Specifically, the case feature vector in the case knowledge base is defined, and the mathematical expression can be:

[0111]

[0112] in, is the characteristic vector of the key characteristic index of the jth historical algorithm case in the case knowledge base; N j represents the total number of nodes corresponding to the jth historical algorithm case; represents the load fluctuation rate corresponding to the jth historical algorithm case; represents the new energy penetration rate corresponding to the jth historical algorithm case; τ j Indicates the time scale, such as the time period corresponding to the day-ahead and real-time; Γ j Indicates the abnormal or fault scenario code, such as N-1, N-2; N case Represents the total number of historical algorithm cases in the case knowledge base.

[0113] Based on the case feature vector, we can query the similarity between the current market clearing algorithm and the historical algorithm cases in the case knowledge base. The mathematical expression can be:

[0114]

[0115] Among them, Sim(C q ,C j ) represents the similarity between the market clearing algorithm and the historical algorithm case; C q represents the characteristic vector input when calibrating the market clearing algorithm; w i Represents the weight of the i-th feature dimension in the feature vector, which can be determined by the entropy weight method; represents the cosine similarity between the market clearing algorithm and the historical algorithm case on the i-th feature dimension; and They respectively represent the numerical eigenvalues ​​of the market clearing algorithm q and the historical algorithm case j in the i-th characteristic dimension.

[0116] Furthermore, based on the similarity between the market clearing algorithm and historical algorithm cases, historical algorithm cases similar to the current clearing algorithm can be screened from the case knowledge base. In practical applications, historical algorithm cases that exceed a preset similarity threshold can be considered similar cases.

[0117] Furthermore, the embodiment of the present invention can generate and execute a target verification strategy according to the causal chain of real-time data features - case matching calculation - dynamic configuration of verification weights - three-dimensional verification execution. The mathematical expression of the target verification strategy can be:

[0118]

[0119] in, Represents the characteristics of the current market clearing algorithm; represents the characteristics of historical algorithm cases in the case knowledge base; f(·) represents the functional form of the target verification strategy, which can be obtained through machine learning training; Indicates the initial verification logic result. The mathematical expression can be:

[0120]

[0121] Among them, N con Represents the total number of all constraints; x0 represents the initial solution of the market clearing algorithm, and its value depends on the current state. If it is to verify the clearing result of the day-ahead phase, the clearing result of the previous period can be selected; if it is the initial solution of the intraday phase, the final clearing result of the day-ahead phase can be selected as the initial solution, and so on; g k (x0) represents the deviation of the kth constraint from the initial solution x0; Represents the initial value of the logical tolerance threshold under the kth constraint condition.

[0122] Step S104 , adjusting the weight of the collaborative verification model according to the initial logic verification result, so as to generate an evaluation result for the existing clearing algorithm using the collaborative verification model; and performing clearing verification based on the verification strategy.

[0123] After the generation check passes the initial logic verification result, the embodiment of the present invention can dynamically adjust the weight of the collaborative check model shown in the above formula (6). The mathematical expression can be:

[0124]

[0125] in, represents the calibration weight of the market clearing algorithm k under the mth calibration evaluation dimension; represents the contribution of market clearing algorithm k under the mth verification evaluation dimension; δ represents the case similarity The adjustment parameters of Represents the characteristics of the current market clearing algorithm; Represents the characteristics of historical algorithm cases; Indicates the initial validation logic result.

[0126] Furthermore, the target weights can be used to adjust the weights of the collaborative verification model. Through this dynamic weight adjustment mechanism, in specific application scenarios, the existing clearing algorithm can prioritize the focus of formulation (for example, new energy penetration rate >30%), and the weight of logical verification can be increased.

[0127] In an optional embodiment, rule-based reasoning (RBR) and case-based comparison (CBR) can be performed based on the verification anomalies of the clearing algorithm to update the case features in the case knowledge base to optimize the target verification strategy. The mathematical expression of this feedback mechanism can be:

[0128]

[0129] in, Update case features in the case knowledge base; is the original case feature in the case knowledge base; γ i Represents the learning parameters of the i-th feature dimension; represents the i-th characteristic component of the j-th historical algorithm case; Indicates the feature vector constructed by the newly added data.

[0130] In another optional embodiment, the embodiment of the present invention can also establish dynamic association rules and collaborative optimization mechanisms between different verification evaluation dimensions to achieve adaptive optimization of the verification strategy. Specifically, the coupling relationship between the logic verification function and the accuracy evaluation function can be described as:

[0131]

[0132] Among them, Φ L Represents the logic verification function of the clearing algorithm; φ A Denotes the accuracy evaluation function of the clearing algorithm; D KL represents the Kullback-Leibler divergence; g i represents the deviation of the i-th constraint; x represents the optimal solution.

[0133] Furthermore, under the condition of a given logic verification threshold, a Pareto frontier optimization model can be constructed as follows:

[0134]

[0135] Among them, λ1 and λ2 are weight coefficients; θ is a set of adjustable parameters, such as solver tolerance and constraint relaxation; φ E is the efficiency evaluation function of the clearing algorithm.

[0136] By solving the above equations (14) and (15), the internal correlation and dynamic trade-off of the check function can be realized. The specific solution process can be achieved through the Pareto solver.

[0137] Finally, in order to enhance the understanding of those skilled in the art on the target verification strategy generation process provided in the embodiment of the present invention, a description of the implementation process is provided using anomaly diagnosis in a provincial power market as an example:

[0138] The current algorithm features are: The case library matching results are as follows: 3 similar cases were found (Sim>0.7), of which 2 cases had failed verification due to defects in the new energy compensation logic.

[0139] The verification process includes the following steps S1 to S3:

[0140] Step S1: Calculate the initial logic verification result using the above formula (11). Specifically, the total number of constraints N con =1200 (including 800 unit constraints and 400 network constraints); the initial solution x0 takes the clearing result of the same period yesterday; it is detected that 180 constraints are not satisfied (Formula (2), );

[0141] Step S2: Use the above formula (12) to dynamically generate the target verification strategy. Specifically, due to If the similarity Sim>0.7, the following policy adjustments are triggered: logic verification priority: the order of checking the new energy compensation constraint is raised from 15th to 1st; accuracy benchmark selection: a high-volatility scenario dataset from similar cases is used as a reference; the number of nodes N is determined to be 1500, and the maximum computation time in the above formula (5) is dynamically calculated.

[0142] Step S3: Execute verification. Prioritize verification of the new energy compensation logic. It is discovered that the objective function does not include the prediction error compensation term. This is marked as a logic defect and the abnormality diagnosis process is initiated.

[0143] An embodiment of the present invention provides a method for clearing and calibration of a spot electricity market, including establishing an electricity market calibration system, the electricity market calibration system including core calibration conditions that must be verified at different market levels; and establishing a three-dimensional collaborative calibration function group that takes into account logic verification, accuracy, and efficiency; constructing a collaborative calibration model based on the electricity market calibration system and the three-dimensional collaborative calibration function group; constructing a case knowledge base containing multiple historical algorithm cases, matching historical algorithm cases similar to the current clearing algorithm in the case knowledge base, and generating a target calibration strategy and initial logic verification results based on the historical algorithm cases; adjusting the weight of the collaborative calibration model based on the initial logic verification results to generate an evaluation result for the existing clearing algorithm using the collaborative calibration model; and executing clearing calibration based on the calibration strategy. Through the present invention, a data closed loop is achieved from real-time data acquisition to calibration strategy generation and execution, which greatly improves calibration efficiency and resource utilization. Specifically, it includes at least the following beneficial effects:

[0144] 1) Technical innovation: This approach implements progressive verification and dynamic weighting of computational logic, result accuracy, and operational efficiency, addressing the blind spots of traditional single-dimensional verification methods.

[0145] 2) Dynamic Adaptability: Verification standards are dynamically adjusted based on parameters such as grid size and renewable energy penetration. Verification priorities and benchmark selection are automatically optimized based on case similarity and real-time data characteristics.

[0146] 3) Performance Improvement: Verification time and computing resource usage have been reduced; with sufficient risk prevention and control capabilities, the detection rate of logical defects has been improved, and scheduling errors caused by algorithm defects have been reduced.

[0147] Further, as Figure 1 The embodiment of the present invention provides a power spot market clearing verification system, such as Figure 2 As shown, the system may include: a verification model construction module 210 , a multi-dimensional collaborative evaluation module 220 , a verification strategy generation module 230 and an algorithm verification execution module 240 .

[0148] The verification model building module 210 can be used to establish a power market verification system, which includes core verification conditions that must be verified at different market levels; and to establish a three-dimensional collaborative verification function group that considers logic verification, accuracy, and efficiency;

[0149] The multi-dimensional collaborative evaluation module 220 can be used to construct a collaborative verification model based on the power market verification system and the three-dimensional collaborative verification function group; the collaborative verification model is used to perform a multi-dimensional evaluation of the clearing algorithm to obtain an evaluation result;

[0150] The checking strategy generation module 230 can be used to construct a case knowledge base containing a plurality of historical algorithm cases, match a historical algorithm case similar to the current clearing algorithm in the case knowledge base, generate a target checking strategy and an initial logical verification result according to the historical algorithm case;

[0151] The algorithm checking execution module 240 can be used to adjust the weight of the collaborative checking model according to the initial logical verification result, generate an evaluation result for the existing clearing algorithm by using the collaborative checking model, and perform clearing checking based on the checking strategy.

[0152] It should be noted that other corresponding descriptions of the functions of the power spot market clearing checking system provided by the embodiments of the present application can be referred to the corresponding descriptions of the method shown in FIG. 8, and will not be repeated here. Figure 1

[0153] Those skilled in the art can clearly understand the specific working process of the system, device, module and unit described above, and the corresponding process in the foregoing method embodiments can be referred to for brevity.

[0154] In addition, the functional units in each embodiment of the present application can be physically independent of each other, or two or more functional units can be integrated together, and all functional units can be integrated in one processing unit. The integrated functional units can be realized in the form of hardware or software or firmware.

[0155] Those skilled in the art can understand that the integrated functional units, if realized in the form of software and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be essentially embodied in the form of a software product, and the computer software product is stored in a storage medium, which includes a plurality of instructions for causing a computing device (such as a personal computer, a server, or a network device) to execute all or part of the steps of the method described in the embodiments of the present application when the instructions are executed. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various storage media that can store program codes.

[0156] Alternatively, all or part of the steps of the foregoing method embodiments can be completed by program instruction related hardware (such as a computing device of a personal computer, a server, or a network device), and the program instruction can be stored in a computer readable storage medium. When the program instruction is executed by the processor of the computing device, the computing device executes all or part of the steps of the method described in the embodiments of the present application. ​

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that within the spirit and principles of the present invention, they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. However, these modifications or replacements do not deviate from the scope of protection of the present invention.

Claims

1. A method for clearing electricity spot market, characterized in that: The method comprises: Establishing a power market verification system that includes core verification conditions for different market levels; and establishing a three-dimensional collaborative verification function set that considers logic verification, accuracy, and efficiency; Constructing a collaborative verification model based on the power market verification system and the three-dimensional collaborative verification function group; the collaborative verification model is used to perform a multi-dimensional evaluation of the clearing algorithm to obtain an evaluation result; Construct a case knowledge base containing multiple historical algorithm cases, match historical algorithm cases similar to the current clearing algorithm in the case knowledge base, and generate a target verification strategy and initial logic verification results based on the historical algorithm cases; The collaborative verification model is weighted according to the initial logic verification result to generate an evaluation result for an existing clearing algorithm using the collaborative verification model; and a clearing verification is performed based on the verification strategy.

2. The method according to claim 1, characterized in that The electricity market verification system includes: The first core verification condition that must be verified for the safety-constrained unit commitment (SCUC) in the day-ahead market; the first core verification condition includes at least one of the unit start-stop logic, ramp constraint coupling, and network security constraint: and / or, A second core verification condition that must be verified for security-constrained economic dispatch (SCED) in the real-time market; the second core verification condition includes at least one of real-time power balance, node electricity price rationality, and dynamic adaptability of network topology: and / or, A third core verification condition that must be verified by the frequency regulation resource clearing algorithm FRAC for the frequency regulation auxiliary market; the third core verification condition includes at least one of frequency regulation capacity allocation, response rate constraint, and cross-time capacity coupling: and / or, The fourth core verification condition that the reserve capacity clearing algorithm SRAC for the reserve auxiliary market must verify; the fourth core verification condition includes at least one of the following: the availability of reserve capacity, the rationality of its spatiotemporal distribution, and its synergy with the main energy market: and / or, The fifth core verification condition of the multi-energy collaborative clearing algorithm MEUC for multi-energy joint clearing; the fifth core verification condition includes at least one of wind and solar prediction error processing, energy storage charging and discharging logic, and multi-energy complementarity verification.

3. The method according to claim 2, characterized in that The establishment of a three-dimensional collaborative verification function group considering logic verification, accuracy and efficiency includes: Define the three-dimensional collaborative verification function group, the mathematical expression is: Φ=[Φ L ,F A ,F E ] T Where Φ represents the three-dimensional collaborative verification function group; T represents the vector transpose; Φ L Represents the logical verification function of the clearing algorithm; Φ A represents the accuracy evaluation function of the clearing algorithm; Φ E is the efficiency evaluation function of the clearing algorithm; The logic verification function Φ L The mathematical expression is: Among them, N c Indicates the total number of constraints; g i (x * ) indicates that the i-th constraint condition is in the optimal solution x * The deviation of L Represents the logic tolerance threshold; represents a dummy variable function; The accuracy evaluation function Φ A The mathematical expression is: Among them, α and β represent weight coefficients; P calc represents the power vector; P ref represents the system reference power vector that can be obtained from historical data; ‖·‖2 represents the quadratic norm; P base represents the reference power vector; D KL (·‖·) represents the Kullback-Leibler divergence; q LMP represents the distribution of node marginal electricity price; q bench represents the distribution of benchmark electricity prices; The efficiency evaluation function Φ E The mathematical expression is: Among them, λ and represents the adjustment parameter; T represents the actual time consumed by the evaluated algorithm; M represents the peak memory usage when the evaluated algorithm is running; M cap Indicates the hardware memory capacity for loading the algorithm; T max (N) represents the maximum computation time when the number of nodes is N. The mathematical expression is: T max (N) = aN b ; a and b represent the parameters obtained by fitting the experimental data.

4. The method according to claim 3, characterized in that The mathematical expression of the collaborative verification model is: Among them, Score k represents the evaluation result of the multi-dimensional evaluation of the k-th market clearing algorithm, k∈{SCUC,SCED,FRAC,SRAC,MEUC}; m represents the verification evaluation dimension, including consideration of logic verification, accuracy and efficiency; It represents the calibration weight of the market clearing algorithm k under the mth calibration evaluation dimension. The mathematical expression is: in, represents the contribution of market clearing algorithm k under the mth verification evaluation dimension; It represents the total contribution of the market clearing algorithm k under the third verification evaluation dimension.

5. The method according to claim 1, wherein The historical algorithm cases similar to the current clearing algorithm are matched in the case knowledge base, including: The case feature vector in the case knowledge base is defined as follows: in, is the characteristic vector of the key characteristic index of the jth historical algorithm case in the case knowledge base; N j represents the total number of nodes corresponding to the jth historical algorithm case; represents the load fluctuation rate corresponding to the jth historical algorithm case; represents the new energy penetration rate corresponding to the jth historical algorithm case; τ j represents the time scale; Γ j Indicates abnormal or fault scenario code; N case represents the total number of historical algorithm cases in the case knowledge base; Based on the case feature vector, the similarity between the current market clearing algorithm and the historical algorithm cases in the case knowledge base is queried. The mathematical expression is: Among them, Sim(C q ,C j ) represents the similarity between the market clearing algorithm and the historical algorithm case; C q represents the characteristic vector input when calibrating the market clearing algorithm; e i Represents the weight of the i-th feature dimension in the feature vector; represents the cosine similarity between the market clearing algorithm and the historical algorithm case on the i-th feature dimension; and They represent the numerical eigenvalues ​​of the market clearing algorithm w and the historical algorithm case j on the i-th characteristic dimension respectively; Based on the similarity between the market clearing algorithm and the historical algorithm cases, historical algorithm cases similar to the current clearing algorithm are screened in the case knowledge base.

6. The method according to claim 1, characterized in that Generating a target verification strategy and initial logic verification results based on the historical algorithm case includes: Generate a target verification strategy based on historical algorithm cases and the current clearing algorithm. The mathematical expression is: in, Represents the characteristics of the current market clearing algorithm; represents the characteristics of historical algorithm cases; f(·) represents the functional form of the target verification strategy; Represents the initial verification logic result, the mathematical expression is: Among them, N con represents the total number of all constraints; x0 represents the initial solution of the market clearing algorithm; g k (x0) represents the deviation of the kth constraint from the initial solution x0; Represents the initial value of the logical tolerance threshold under the kth constraint condition.

7. The method according to claim 4, characterized in that The weight adjustment of the collaborative verification model according to the initial logic verification result includes: The target weight of the collaborative verification model is calculated using the initial logic verification results. The mathematical expression is: in, represents the calibration weight of the market clearing algorithm k under the mth calibration evaluation dimension; represents the contribution of market clearing algorithm k under the mth verification evaluation dimension; δ represents the case similarity The adjustment parameters of Represents the characteristics of the current market clearing algorithm; Represents the characteristics of historical algorithm cases; Indicates the initial validation logic result The target weight is used to adjust the weight of the collaborative verification model.

8. The method according to claim 1, characterized in that The method further comprises: Based on the rule reasoning and case comparison for the check anomaly of the clearing algorithm, the case features in the case knowledge base are updated to optimize the target check strategy. The mathematical expression is: in, updating case features in the case knowledge base; is the original case feature in the case knowledge base; γ i Represents the learning parameters of the i-th feature dimension; represents the i-th characteristic component of the j-th historical algorithm case; Indicates the feature vector constructed by the newly added data.

9. The method according to claim 3, characterized in that The method further comprises: Establish dynamic association rules and collaborative optimization mechanisms between different verification and evaluation dimensions, including: Define the coupling relationship between the logic verification function and the accuracy evaluation function. The mathematical expression is: Among them, Φ L Represents the logical verification function of the clearing algorithm; Φ A Denotes the accuracy evaluation function of the clearing algorithm; D KL represents the Kullback-Leibler divergence; g i represents the deviation of the i-th constraint; X represents the optimal solution; Given a logic verification threshold, a Pareto frontier optimization model is constructed, and the mathematical expression is: Among them, λ1 and λ2 represent weight coefficients; θ represents the set of adjustable parameters; Φ E is the efficiency evaluation function of the clearing algorithm; By solving the coupling relationship between the logic verification function and the accuracy evaluation function and the Pareto frontier optimization model, the internal correlation and dynamic balance of the verification function are performed to achieve adaptive optimization of the verification strategy.

10. A power spot market clearing verification system, characterized by: The system comprises: A verification model building module is used to establish a power market verification system, which includes core verification conditions that must be verified at different market levels; and to establish a three-dimensional collaborative verification function group that considers logic verification, accuracy and efficiency; a multi-dimensional collaborative evaluation module, configured to construct a collaborative verification model based on the power market verification system and the three-dimensional collaborative verification function group; the collaborative verification model is configured to perform a multi-dimensional evaluation of the clearing algorithm to obtain an evaluation result; A verification strategy generation module is used to build a case knowledge base containing multiple historical algorithm cases, match historical algorithm cases similar to the current clearing algorithm in the case knowledge base, and generate a target verification strategy and initial logic verification results based on the historical algorithm cases; an algorithm verification execution module, configured to adjust the weight of the collaborative verification model according to the initial logic verification result, so as to generate an evaluation result for the existing clearing algorithm using the collaborative verification model; And, performing a clearing check based on the check strategy.

Images