Electric power spot market checking method, device and system

Abstract

The invention provides an electric power spot market checking method, device and system, and the method comprises the steps: collecting multi-source heterogeneous data related to the operation of an electric power spot market, and carrying out the preprocessing of the multi-source heterogeneous data, and obtaining target multi-source heterogeneous data; constructing a market clearing check model and a risk assessment check model based on target multi-source heterogeneous data fusion; the market clearing checking model is used for checking the market clearing result; the risk assessment check model is used for assessing multiple aspects of risks existing in the electric power spot market; and solving the market clearing check model and the risk assessment check model based on a particle swarm optimization algorithm to obtain an electric power spot market check analysis result. According to the invention, all-around and high-precision checking of the electric power spot market is realized, powerful technical support is provided for healthy and orderly development of the electric power spot market, the operation quality and supervision level of the electric power spot market are improved, and safe, stable and efficient operation of the electric power spot market is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation and management, and particularly to a method, device and system for checking a power spot market. Background Art

[0002] As an important part of the power market system, the power spot market is of crucial significance for optimizing power resource allocation and promoting power market competition. However, with the rapid development of the power spot market, it faces many complex problems, which pose higher requirements for the market checking technical support system. On the one hand, the trading data of the power spot market is huge in scale and complex and changeable. There are many market participants, including various power generation enterprises (thermal power, hydropower, new energy power generation, etc.), electricity sales companies and a large number of power users. The trading data covers multiple dimensions such as electricity volume, electricity price, trading time, trading location, etc., and the requirement for data timeliness is extremely high. Traditional data processing technologies are difficult to quickly and accurately process massive and high-frequency data, resulting in difficulty in ensuring data accuracy and timeliness, and further affecting the accuracy and timeliness of market checking. For example, during peak hours, the trading data volume may increase sharply in an instant, and traditional systems may experience problems such as data congestion and processing delays, and cannot provide effective data support for the checking work in a timely manner. On the other hand, the operation rules of the power spot market are complex and there are many constraint conditions. Market clearing needs to meet various requirements such as power balance, unit operation restrictions, power grid security constraints and new energy power generation characteristic constraints. Existing checking technologies have problems such as imperfect model construction and low calculation efficiency when dealing with complex constraint conditions. For example, when considering the uncertainty and intermittency of new energy power generation, traditional checking methods are difficult to accurately simulate its impact on the market clearing result, resulting in a deviation between the checking result and the actual market operation situation, and unable to effectively discover potential market risks and operation problems.

[0003] In addition, the dynamic changes in the market environment, such as the continuous increase in the proportion of new energy access, the fluctuation of load demand and the diversification of market trading strategies, make the traditional checking technical support system lack sufficient adaptability and flexibility. Traditional technologies are difficult to quickly respond to market changes, timely adjust checking strategies and model parameters, and cannot provide comprehensive and accurate checking information and decision-making support for market regulatory agencies and market participants. Summary of the Invention

[0004] Therefore, the present invention provides a method, device and system for checking a power spot market, aiming to solve the technical problem that the prior art is difficult to ensure the safe, stable and efficient operation of the power spot market.

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

[0006] According to the first aspect of the present invention, the present invention provides a method for checking in the electricity spot market, the method comprising:

[0007] Collecting multi-source heterogeneous data related to the operation of the electricity spot market, and preprocessing the multi-source heterogeneous data to obtain target multi-source heterogeneous data;

[0008] Constructing a market clearing check model and a risk assessment check model based on the fusion of the target multi-source heterogeneous data; the market clearing check model is used to check the market clearing result; the risk assessment check model is used to evaluate various risks existing in the electricity spot market;

[0009] Solving the market clearing check model and the risk assessment check model based on the particle swarm optimization algorithm to obtain the check analysis result of the electricity spot market.

[0010] Further, the collecting of the multi-source heterogeneous data related to the electricity spot includes:

[0011] Obtaining power source related data from the power generation side, including at least one of real-time output data, power generation plan data, operation status data, and power generation cost data of multiple types of units; and,

[0012] Obtaining grid operation data from the grid side, including at least one of grid topology structure data, line power flow data, node voltage data, transformer parameter data, and security constraint data; and,

[0013] Obtaining market transaction data from the trading platform, including at least one of declaration data, transaction data, market clearing price data, ancillary service transaction data, and meteorological data affecting new energy power generation of each market entity; and,

[0014] Obtaining rule-related data from the market regulatory agency, including at least one of market rules, policies and regulations, and market supervision data.

[0015] Further, the preprocessing of the multi-source heterogeneous data includes:

[0016] For multi-source heterogeneous data of different data types, respectively constructing data filtering rules based on physical laws and statistical characteristics, and using the data filtering rules matching the data type to eliminate outliers from the multi-source heterogeneous data of the data type; wherein, the data type includes at least one of unit power generation data, new energy power generation data, grid power flow data, and market transaction data;

[0017] And filling the missing data in the multi-source heterogeneous data by using the linear interpolation algorithm.

[0018] Furthermore, the construction of the market clearing verification model and the risk assessment verification model based on the target multi-source heterogeneous data fusion includes:

[0019] Based on the target multi-source heterogeneous data fusion, considering the physical characteristics of the power system and the market operation rules, a market clearing verification model is constructed with the objective of minimizing the deviation between the market clearing result and the actual operation data. The expression formula of the objective function is:

[0020]

[0021] Where, is the output of power generation enterprise i at time t calculated by the market clearing model; is the actual output of power generation enterprise i at time t; is the electricity consumption of user j at time t calculated by the market clearing model; is the actual electricity consumption of user j at time t; T is the verification time period; n is the number of power generation enterprises; m is the number of users;

[0022] And, based on the multi-scenario simulation technology, a risk assessment verification model for performing price risk assessment, electricity quantity risk assessment, and / or new energy fluctuation risk assessment in the electricity spot market is constructed.

[0023] Furthermore, the constraint conditions of the market clearing verification model include at least one of power balance constraint, unit operation constraint, ramp rate constraint, minimum start-stop time constraint, grid security constraint, new energy generation constraint, and new energy penetration constraint;

[0024] The expression formula of the power balance constraint is:

[0025]

[0026] Where, L t is the system network loss at time t;

[0027] The expression formula of the unit operation constraint is:

[0028]

[0029] Where, and are the minimum and maximum verification outputs of unit i at time t respectively;

[0030] The expression formula of the ramp rate constraint is:

[0031]

[0032] Where, R i,d and Ri,u are the downward and upward ramp rates of unit i, respectively;

[0033] The expression formula for the minimum start-stop time constraint is:

[0034]

[0035] where T i,on,min and T i,off,min are the minimum start-up and shutdown times of unit i, respectively; and are the durations of the current start-up and shutdown of unit i, respectively;

[0036] The expression formula for the power grid security constraint is:

[0037]

[0038] where is the checked power flow of line l at time t, F l,t,min and F l,t,max are the lower and upper limits of the power flow of line l at time t and the node voltage constraint, respectively; is the checked voltage of node k at time t, V k,t,min and V k,t,max are the lower and upper limits of the voltage of node k at time t, respectively;

[0039] The expression formula for the new energy power generation constraint is:

[0040]

[0041] where is the predicted output of new energy power generation; is the actual output of new energy power generation; ε min and ε max are the lower and upper limits of the prediction error of the new energy power generation output, respectively;

[0042] The expression formula for the new energy penetration rate constraint is:

[0043]

[0044] where is the actual power generation of new energy source s at time t, and ρ is the new energy penetration rate requirement.

[0045] Furthermore, the price risk assessment includes:

[0046] Evaluating the market price risk by combining conditional value at risk CVaR and value at risk VaR, including:

[0047] Let the market price sequence be P(t), where t = 1, 2, …, T; calculate the value-at-risk parameter VaR using a pre-designed calculation method α , and the conditional value-at-risk parameter CVaR α The calculation formula is as follows:

[0048]

[0049] where f(p) is the probability density function of price p;

[0050] By comparing the value-at-risk parameter VaR α , the conditional value-at-risk parameter CVaR α and a pre-set risk threshold, evaluate the market price risk;

[0051] The electricity quantity risk assessment includes:

[0052] Based on the electricity quantity deviation risk index, evaluate the potential electricity quantity risk hidden between electricity supply and demand, including:

[0053] Calculate the first electricity quantity deviation risk between the actual electricity generation of the power generation enterprise and the planned electricity generation, and the calculation formula is:

[0054]

[0055] where is the actual electricity generation of power generation enterprise i at time t, is the planned electricity generation of power generation enterprise i at time t; and,

[0056] Calculate the second electricity quantity deviation risk between the actual electricity consumption of users and the expected electricity consumption, and the calculation formula is:

[0057]

[0058] where is the actual electricity consumption of user j at time t, is the expected electricity consumption of user j at time t;

[0059] The new energy fluctuation risk assessment includes:

[0060] Based on the new energy fluctuation risk index, evaluate the impact of the intermittency and volatility of new energy power generation on the market. The calculation formula of the new energy fluctuation risk index is:

[0061]

[0062] where P ren (t) is the new energy power generation output sequence; t = 1, 2, …, T.

[0063] Further, the market clearing verification model and the risk assessment verification model are solved based on the particle swarm optimization algorithm to obtain the verification analysis result of the electricity spot market, including:

[0064] Generate an initial particle population, where each particle in the initial particle population represents a potential power distribution strategy; the power distribution strategy includes a power output distribution plan for power generation enterprises and a power consumption distribution plan for users.

[0065] Calculate the objective function value corresponding to the power distribution strategy represented by each particle for the market clearing verification model, and iteratively update the individual extreme value and the global extreme value according to the objective function value.

[0066] During the particle swarm iteration process, use the adaptive inertia weight strategy and / or the learning factor dynamic adjustment strategy to search for the optimal solution.

[0067] Until the convergence condition is met, the particle swarm reaches the global optimal position, and the target power distribution strategy closest to the actual operation data is obtained.

[0068] Use the risk assessment verification model to analyze various risks existing in the target power distribution strategy to obtain the verification analysis result of the electricity spot market.

[0069] Further, the adaptive inertia weight strategy includes: setting the inertia weight ω using a linear decreasing strategy, and the expression formula is:

[0070]

[0071] where k is the current iteration number, and K max is the maximum iteration number;

[0072] The learning factor dynamic adjustment strategy includes: adjusting the learning factors c1 and c2 according to the fitness value of the particle and the average fitness value of the particle swarm, and the expression formula is:

[0073]

[0074] where f i is the fitness value of particle i; is the average fitness value of the population; c 1,init and c 1,end and c 2,init and c 2,end are respectively the initial value and the final value of the learning factor; δ max is the maximum value of δ.

[0075] According to the second aspect of the present invention, the present invention provides an electricity spot market verification device, and the device includes:

[0076] A data acquisition module, configured to collect multi-source heterogeneous data related to the operation of the electricity spot market, and preprocess the multi-source heterogeneous data to obtain target multi-source heterogeneous data;

[0077] A model construction module, configured to construct a market clearing verification model and a risk assessment verification model based on the fusion of the target multi-source heterogeneous data; the market clearing verification model is used to verify the market clearing result; the risk assessment verification model is used to evaluate various risks existing in the electricity spot market;

[0078] A model solving module, configured to solve the market clearing verification model and the risk assessment verification model based on the particle swarm optimization algorithm to obtain an electricity spot market verification analysis result.

[0079] According to a third aspect of the present invention, there is provided an electricity spot market verification system, including: a processor; a memory for storing instructions executable by the processor; the processor is configured to execute the electricity spot market verification method according to any one of the first aspects of the present invention.

[0080] The present invention adopts the above technical solutions and at least has the following beneficial effects:

[0081] Through the solution of the present invention, multi-source heterogeneous data related to the operation of the electricity spot market is collected, and the multi-source heterogeneous data is preprocessed to obtain target multi-source heterogeneous data; a market clearing verification model and a risk assessment verification model based on the fusion of the target multi-source heterogeneous data are constructed; the market clearing verification model is used to verify the market clearing result; the risk assessment verification model is used to evaluate various risks existing in the electricity spot market; the market clearing verification model and the risk assessment verification model are solved based on the particle swarm optimization algorithm to obtain an electricity spot market verification analysis result. Thus, through the construction of key technical modules, accurate verification of the electricity spot market is realized, and the stability and economy of market operation are improved; with the help of multi-source data mining algorithms, operations such as cleaning and conversion of massive and complex data are performed to ensure the accuracy and availability of the data, laying a data foundation for subsequent verification work; at the same time, the particle swarm optimization algorithm is introduced to solve the electricity spot market verification model; by simulating the search behavior of the particle swarm in the solution space, according to the adaptive inertia weight strategy and the method of dynamically adjusting the learning factor, the global search and local search capabilities of the algorithm are balanced, and the optimal solution is efficiently found. The present invention aims to achieve all-round and high-precision verification of the electricity spot market, provide strong technical support for the healthy and orderly development of the electricity spot market, improve the operation quality and supervision level of the electricity spot market, and ensure its safe, stable and efficient operation.

[0082] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and do not limit the present invention. Brief Description of the Drawings

[0083] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0084] Figure 1 The flowchart showing the power spot market checking method provided by an embodiment of the present invention;

[0085] Figure 2 The flowchart showing the solution process of the particle swarm optimization algorithm provided by an embodiment of the present invention;

[0086] Figure 3 The structural diagram showing the power spot market checking device provided by an embodiment of the present invention;

[0087] Figure 4 The entity structure diagram showing the power spot market checking system provided by an embodiment of the present invention. Detailed Embodiments

[0088] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the 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. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0089] It should be noted that, in this document, relational terms such as "first" and "second" are only used 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 term "comprising", "including" or any other variation thereof is intended to cover a non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, the elements defined by the statement "comprising..." do not exclude the presence of additional identical elements in the process, method, article or device including the said elements.

[0090] An embodiment of the present invention provides a method for checking an electricity spot market, as follows: Figure 1 As shown, it may at least include the following steps S101 to S103:

[0091] Step S101: Collect multi-source heterogeneous data related to the operation of the electricity spot market, and preprocess the multi-source heterogeneous data to obtain target multi-source heterogeneous data.

[0092] Traditional data collection means often only focus on limited data sources, with a single data format and lacking effective integration and in-depth mining capabilities, making it difficult to meet the urgent need of the electricity spot market for comprehensive, accurate, and real-time data. The lack of in-depth mining capabilities causes the data to remain on the surface, unable to extract valuable potential information from it, seriously restricting the efficient operation and scientific decision-making of the electricity spot market. To effectively overcome these problems, considering the high complexity of the operation of the electricity spot market and the core value of data in market analysis, decision-making, and risk control, it is deeply recognized that there are many limitations in traditional data collection methods. An embodiment of the present invention establishes extensive data collection interfaces to collect multi-source heterogeneous data from multiple data sources, including power generation enterprises, power grid dispatching centers, trading platforms, meteorological departments, and market supervision agencies, etc.

[0093] Specifically, the multi-source heterogeneous data includes the following aspects:

[0094] 1) Power source-related data on the power generation side, including real-time output data, generation plan data, operation status data (such as start-stop status, ramp rate, minimum stable output, maximum output, etc.), and generation cost data of various types of units (such as thermal power, hydropower, wind power, photovoltaic, etc.). In practical applications, these data can be obtained through monitoring equipment installed on the generator sets, the production management system of power generation enterprises, and the communication interfaces with power generation enterprises.

[0095] 2) Power grid operation data on the power grid side, including power grid topology structure data, line power flow data, node voltage data, transformer parameter data, and security constraint data (such as line transmission capacity limits, node voltage upper and lower limits, short-circuit capacity, etc.). In practical applications, these data can be obtained by using the power grid dispatching automation system, substation automation equipment, and power communication network.

[0096] 3) Obtain market transaction data from the trading platform, including declaration data (such as electricity volume, electricity price, trading time, etc.), transaction data, market clearing price data, ancillary service trading data, and meteorological data affecting new energy power generation (such as wind speed, wind direction, light intensity, temperature, air pressure, etc.) of each market entity. Meteorological data is closely related to new energy power generation and can be used for new energy power generation prediction and verification, and is generally obtained through the meteorological department.

[0097] 4) Obtain rule-related data from the market regulatory agency, including market rules, policies and regulations, and market supervision data, etc., which can ensure that the verification work complies with market supervision requirements.

[0098] Furthermore, in order to ensure the accuracy and integrity of multi-source heterogeneous data, it is necessary to perform data cleaning and data preprocessing on the multi-source heterogeneous data to obtain the processed target multi-source heterogeneous data. Specifically, in the embodiments of the present invention, for multi-source heterogeneous data of different data types, data filtering rules based on physical laws and statistical characteristics are constructed, and the data filtering rules matching the data types are used to eliminate outliers from the multi-source heterogeneous data of the data types. Among them, different data types can be unit power generation data, new energy power generation data, power grid power flow data, and market transaction data, etc.

[0099] For unit power generation data: Taking a thermal power generating unit as an example, the output of a thermal power generating unit usually has a reasonable range, which depends on the capacity and operating status of the unit. Let the rated capacity of a certain thermal power generating unit be P rated , considering a certain overload capacity, the lower limit of the output P min during normal operation is generally not lower than 0.3P rated , and the upper limit of the output P max is not higher than 1.1P rated . When the collected power generation output data P g satisfies P g <P min or P g >P max , then it can be determined that the data is an outlier and is excluded.

[0100] For new energy power generation data: Taking wind power generation as an example, filtering can be performed according to the power curve characteristics of the wind turbine. When the wind speed v ci is below the cut-in wind speed of the wind turbine, the output is basically 0; when the wind speed reaches the rated wind speed v r , the rated power P r is reached; when the wind speed is above the cut-out wind speed v co , the wind turbine stops operating and the output is also 0. Therefore, when the collected wind power output data P w does not satisfy the power curve relationship with the corresponding wind speed v, then it can be determined that the data is abnormal. For example, if v < v ci and P w >0, or v r <v < v co and P w ≠P r within a certain error range, then the data is marked as abnormal data.

[0101] For power grid power flow data, filtering can be performed according to Kirchhoff's law. That is, at a power network node, the sum of the currents flowing into the node is equal to the sum of the currents flowing out of the node, i.e.:

[0102]

[0103] where I k,in is the current flowing into node k; I k,out is the current flowing out of node k; n is the number of branches connected to the node. If the collected power flow data does not satisfy this law, or the line power flow data exceeds the thermal stability limit of the line, according to the rated current-carrying capacity I rated of the line, the line power flow I l during normal operation should satisfy |I l | ≤ I rated ; if this condition is not met, the data is considered incorrect data and is excluded.

[0104] For market transaction data, filtering can be performed according to market rules and trading logic. For example, in the electricity spot market, the electricity price usually has a reasonable range, restricted by factors such as market supply and demand and costs. Let the minimum electricity price allowed by the market be and the maximum electricity price be If the collected electricity price data P t satisfies: or If this condition is not met, the data may be abnormal. At the same time, it should be noted that in the embodiments of the present invention, for duplicate declarations or transaction data of the same market entity within the same trading period, only valid data is retained and duplicate data is removed.

[0105] In addition, when there are a small number of missing values in multi-source heterogeneous data, the embodiments of the present invention can also use the linear interpolation algorithm to fill in these missing data. Specifically, the expression formula of linear interpolation can be:

[0106]

[0107] where the data y1 at time t1 and the data y2 at time t2 are both known data in the time series data, and the missing data y at time t1 < t < t2 can be filled using the above formula.

[0108] For example, in the power generation output data, if the power generation output data at a certain moment t m is missing, and it is known that the power generation output at the previous moment t m-1 is P m-1 , and the power generation output at the next moment t m+1 is P m+1 , then the power generation output at t m is calculated by linear interpolation.The power generation output at a certain moment is as follows:

[0109]

[0110] For another example, for the grid node voltage data, if the voltage data is missing at a certain moment t at a certain monitoring point n v then, according to the linear distribution characteristic of the voltage along the line, the voltage at the monitoring point n at the moment t v can be expressed as:

[0111]

[0112] where, V n-1 and V n+1 are the voltages at the known adjacent monitoring points n - 1 and n + 1 at the moment t v respectively; d n-1,n and d n,n+1 are the electrical distances between adjacent monitoring points respectively.

[0113] It can be understood that for the missing values of other data types in the multi - source heterogeneous data, the linear interpolation algorithm similar to the above examples can be used to fill the missing values, and the present invention will not elaborate on this.

[0114] Step S102, construct a market clearing verification model and a risk assessment verification model based on the fusion of target multi - source heterogeneous data.

[0115] The market clearing verification model in the embodiment of the present invention is used to verify the market clearing result. Specifically, based on the fusion of target multi - source heterogeneous data, integrate the real - time operation data of power generation enterprises, grid topology and power flow data, transaction data of the trading platform, and meteorological data, etc., and consider the physical characteristics of the power system and market operation rules to accurately verify the rationality and accuracy of the market clearing result. The market clearing verification model takes minimizing the deviation between the market clearing result and the actual operation data as the objective function, and the expression formula is:

[0116]

[0117] where, is the output of power generation enterprise i at the moment t calculated by the market clearing model; is the actual output of power generation enterprise i at the moment t; is the electricity consumption of user j at the moment t calculated by the market clearing model; is the actual electricity consumption of user j at the moment t; T is the verification time period; n is the number of power generation enterprises; m is the number of users.

[0118] It should be noted that the constraint conditions of the market clearing verification model cover various factors, which can include power balance constraints, unit operation constraints, ramp rate constraints, minimum start-stop time constraints, power grid security constraints, new energy power generation constraints, and new energy penetration rate constraints, etc.

[0119] The expression formula for power balance constraint is:

[0120]

[0121] Among them, L t is the system network loss at time t. The power balance constraint ensures that the total power generation in the verification model is always equal to the sum of the total power consumption and the network loss, which precisely matches the power balance requirements of the actual power system.

[0122] The expression formula for unit operation constraint is:

[0123]

[0124] Among them, and are respectively the minimum and maximum verified output of unit i at time t.

[0125] The expression formula for ramp rate constraint is:

[0126]

[0127] Among them, R i,d and R i,u are respectively the downward and upward ramp rates of unit i.

[0128] The expression formula for minimum start-stop time constraint is:

[0129]

[0130] Among them, T i,on,min and T i,off,min are respectively the minimum start-up and shutdown times of unit i; and are respectively the current start-up and shutdown durations of unit i. The minimum start-stop time constraint can ensure that the unit is verified under the conditions that meet the actual operation limitations.

[0131] The expression formula for power grid security constraint is:

[0132]

[0133] Among them, is the verified power flow of line l at time t, F l,t,min and F l,t,maxThey are the lower and upper limits of the power flow of line l at time t and the node voltage constraint, respectively. is the voltage to be checked at node k at time t, V k,t,min and V k,t,max are the lower and upper limits of the voltage at node k at time t, respectively. The power grid security constraint can ensure the security and stability of the power grid during the checking process and prevent situations such as line overload and node voltage over-limit.

[0134] The expression formula for the new energy power generation constraint is:

[0135]

[0136] Among them, is the predicted output of new energy power generation; is the actual output of new energy power generation; ε min and ε max are the lower and upper limits of the prediction error of new energy power generation output, respectively. The new energy power generation constraint takes into account the uncertainty and intermittency of new energy power generation and prevents the prediction error of new energy power generation output from being too large.

[0137] The expression formula for the new energy penetration rate constraint is:

[0138]

[0139] Among them, is the actual power generation of new energy power source s at time t, and ρ is the requirement for the new energy penetration rate. The new energy penetration rate constraint ensures that the characteristics and accommodation conditions of new energy power generation are reasonably considered during the checking process.

[0140] Furthermore, the risk assessment and checking model in the embodiments of the present invention is used to comprehensively assess and check various risks existing in the electricity spot market. Among them, the various risks can at least include price risk assessment, electricity quantity risk assessment, and new energy fluctuation risk assessment.

[0141] For the price risk assessment, the embodiments of the present invention adopt a method combining conditional value at risk CVaR and value at risk VaR to assess the market price risk. Specifically: Let the market price sequence be P(t), t = 1, 2,..., T; the value at risk parameter VaR α can be solved by using a pre-designed calculation method. Among them, the pre-designed calculation method includes historical simulation method, parametric method, or Monte Carlo simulation method, etc. The conditional value at risk parameter CVaR α The calculation formula is:

[0142]

[0143] Among them, f(p) is the probability density function of price p.

[0144] Furthermore, by comparing the risk value parameter VaR α , the conditional risk value parameter CVaR α and the preset risk threshold, the market price risk can be evaluated to determine whether the market price risk is within the controllable range.

[0145] For the electricity risk assessment, the embodiment of the present invention introduces an electricity deviation risk index, and based on the electricity deviation risk index, evaluates the potential electricity risk hidden between electricity supply and demand. Specifically: calculate the first electricity deviation risk between the actual electricity generation of the power generation enterprise and the planned electricity generation, and the calculation formula is:

[0146]

[0147] where is the actual electricity generation of power generation enterprise i at time t, is the planned electricity generation of power generation enterprise i at time t; and calculate the second electricity deviation risk between the actual electricity consumption of the user and the expected electricity consumption, and the calculation formula is:

[0148]

[0149] where is the actual electricity consumption of user j at time t, is the expected electricity consumption of user j at time t.

[0150] Furthermore, the stability and reliability of power generation enterprises and users in terms of electricity supply and demand can be evaluated by calculating the electricity deviation risk, and it can be determined whether there are potential electricity risk hidden dangers.

[0151] For the new energy fluctuation risk assessment, the embodiment of the present invention considers the impact of the intermittency and volatility of new energy power generation on the market, and introduces a new energy fluctuation risk index. Based on the new energy fluctuation risk index, evaluate the impact of the intermittency and volatility of new energy power generation on the market. The calculation formula of the new energy fluctuation risk index is:

[0152]

[0153] where P ren (t) is the new energy power generation output sequence; t = 1, 2,..., T.

[0154] Furthermore, by analyzing the new energy fluctuation risk index, the impact degree of new energy power generation on the stability of the power market can be evaluated, and it can be determined whether corresponding risk response measures need to be taken.

[0155] Step S103, solve the market clearing verification model and the risk assessment verification model based on the particle swarm optimization algorithm to obtain the verification analysis result of the electricity spot market.

[0156] The Particle Swarm Optimization (PSO) algorithm is an optimization algorithm based on swarm intelligence, which is inspired by the collective behavior of bird flocks or fish schools. In model solving, the PSO algorithm searches for the optimal solution by simulating the movement of particles in the search space. Specifically, as Figure 2 shown, the solution of the Particle Swarm Optimization algorithm in the embodiments of the present invention may at least include the following steps S201 to S205:

[0157] Step S201: Generate an initial population of particles, where each particle in the initial population of particles represents a potential power distribution strategy.

[0158] In the embodiments of the present invention, each particle may represent a set of power distribution strategies, including the output distribution plan of power generation enterprises and the power consumption distribution plan of users. The position vector of the particle represents the specific value of the solution, and the velocity vector determines the moving direction and speed of the particle in the search space. In each iteration, the particle updates its velocity and position according to its own historical best position (personal best, pbest) and the global best position (global best, gbest) in the population. The velocity update formula can be expressed as:

[0159]

[0160] where is the velocity of particle i in the d-th dimension at the k-th iteration; ω is the inertia weight; c1 and c2 are learning factors; r1 and r2 are random numbers between [0, 1], is the position of particle i in the d-th dimension at the k-th iteration). Then, the position update formula can be expressed as:

[0161] Step S202: Calculate the objective function value corresponding to the power distribution strategy represented by each particle for the market clearing verification model, and iteratively update the personal extreme value and the global extreme value according to the objective function value.

[0162] In the embodiments of the present invention, each particle represents a set of output distribution plans of power generation enterprises and power consumption distribution plans of users. First, these plans can be substituted into the objective function of the market clearing verification model:

[0163]

[0164] That is, calculate the sum of the output deviations of each power generation enterprise at each moment and the power consumption deviations The sum is then obtained by adding up the sums of these deviations at all times, and the sum of the deviations is the objective function value corresponding to the solution represented by the particle. For example, assume there are 3 power generation enterprises and 2 users, and the verification time period is 5 time instants. For the solution of a certain particle, calculate the sum of the squared output deviations of power generation enterprise 1 at 5 time instants, the sum of the squared output deviations of power generation enterprise 2 at 5 time instants, the sum of the squared output deviations of power generation enterprise 3 at 5 time instants, as well as the sum of the squared power consumption deviations of user 1 at 5 time instants and the sum of the squared power consumption deviations of user 2 at 5 time instants. Finally, add up the sums of these 5 parts, which is the objective function value corresponding to the particle.

[0165] Furthermore, the individual extreme value pbest and the global extreme value gbest can be updated according to the objective function value calculated above. Specifically:

[0166] Regarding the individual extreme value pbest: For each particle, its current objective function value can be compared with the individual extreme value pbest of this particle i (that is, the objective function value corresponding to the optimal solution found by this particle in history). If the current objective function value is less than pbest i , then update pbest i to the current objective function value, and update the current position of this particle (that is, the power output allocation plan of the power generation enterprise and the power consumption allocation plan of the user) to the individual optimal position pbest i .

[0167] Regarding the global extreme value gbest: After updating the individual extreme values of all particles, compare the individual extreme values pbest of all particles i , find the particle corresponding to the smallest objective function value among them, update the individual extreme value pbest of this particle i to the global extreme value gbest, and update the position of this particle to the global optimal position gbest.

[0168] Step S203, during the particle swarm iteration process, use the adaptive inertia weight strategy and / or the learning factor dynamic adjustment strategy to search for the optimal solution.

[0169] The adaptive inertia weight strategy in the embodiment of the present invention may include: setting the inertia weight ω using a linear decreasing strategy, and the expression formula is:

[0170]

[0171] where k is the current iteration number, and K max is the maximum iteration number.

[0172] It can be understood that the adaptive inertia weight strategy can be used to balance the global search and local search capabilities of the algorithm. At the initial stage of the algorithm, a relatively large inertia weight ω (such as ω max = 0.9) is set, so that the particles have strong global search capabilities and can quickly explore the entire search space; as the number of iterations increases, the inertia weight is gradually reduced (such as ω min = 0.4), which can improve the search accuracy of the particles in the local area and help find better solutions.

[0173] Furthermore, the learning factor dynamic adjustment strategy in the embodiments of the present invention includes: adjusting the learning factors c1 and c2 according to the fitness value of the particle and the average fitness value of the particle swarm, and the expression formula is:

[0174]

[0175] where, f i is the fitness value of particle i; is the average fitness value of the population; c 1,init , c 1,end , c 2,init , c 2,end are the initial value and the final value of the learning factor respectively; δ max is the maximum value of δ.

[0176] That is to say, when the fitness value of the particle is better than the average fitness value of the population, c1 is appropriately reduced and c2 is increased, so that the particle is more inclined to learn from the global optimal position; conversely, c1 is appropriately increased and c2 is reduced to enhance the exploration ability of the particle itself.

[0177] Step S204, until the convergence condition is satisfied, the particle swarm reaches the global optimal position, and the target power distribution strategy closest to the actual operation data is obtained.

[0178] Step S205, use the risk assessment and verification model to analyze various risks existing in the target power distribution strategy, and obtain the verification analysis result of the power spot market.

[0179] The convergence condition in the embodiments of the present invention can be that the particle swarm reaches the maximum number of iterations or the global optimal position does not change within a certain number of iterations. Thus, by accurately simulating its impact on the market clearing result through the market clearing verification model, quickly responding to market changes, timely adjusting the verification strategy and model parameters, and conducting multi-faceted risk analysis, comprehensive and accurate verification information and decision support are provided for market regulatory agencies and market participants.

[0180] The embodiments of the present invention provide a method for verifying the power spot market, which shows significant advantages in the field of verifying the power spot market and has at least the following advantages:

[0181] 1) In the data collection and integration phase, through multi-source heterogeneous data collection algorithms, data from power generation enterprises, power grid dispatching centers, trading platforms, etc. are widely and efficiently mined and integrated. This ensures that all types of key data can be comprehensively and accurately obtained, including detailed operation data on the power generation side, real-time operation status data on the power grid side, and market trading data on the trading platform.

[0182] 2) Filtering techniques based on physical laws and statistical characteristics are used for data cleaning and preprocessing, effectively removing outliers, incorrect data, and duplicate data, and accurately processing missing data through algorithms such as linear interpolation. This greatly improves the accuracy, integrity, and timeliness of the data, laying a solid data foundation for subsequent verification work.

[0183] 3) In the verification model construction module, the market clearing verification model and risk assessment verification model constructed based on multi-source data fusion and multi-scenario simulation techniques are ingenious. The market clearing verification model comprehensively considers various constraints such as power balance, unit operation, power grid safety, and new energy power generation. Its objective function accurately locates the minimization of the deviation between the market clearing result and actual operation data. Through rigorous constraint setting, the rationality and accuracy of the market clearing result can be accurately verified. The risk assessment verification model uses advanced risk measurement methods (such as the combination of CVaR and VaR) and combines multi-scenario simulation to comprehensively evaluate price risk, electricity volume risk, and new energy fluctuation risk.

[0184] 4) In terms of the selection and implementation of optimization algorithms, the particle swarm optimization algorithm simulates the behavior of a particle swarm. Through an adaptive inertia weight strategy (such as a linear decreasing strategy) and a method of dynamically adjusting the learning factor (dynamically adjusting according to the particle fitness and the group average fitness), the global and local search capabilities are balanced. During iteration, the individual extreme value and the global extreme value are updated according to the objective function value (such as the total deviation of the market clearing verification model), guiding the particles to search for the optimal solution.

[0185] In summary, the power spot market verification method designed by the present invention effectively solves many problems faced by current power spot market verification, such as poor data quality, inaccurate models, low algorithm efficiency, and unreasonable system architecture, significantly improving the accuracy, reliability, and efficiency of verification. It has broad application prospects in the field of power market operation and supervision, and can provide solid technical support for the healthy, stable, and orderly development of the power spot market.

[0186] Furthermore, as a Figure 1 specific implementation, an embodiment of the present invention provides a power spot market verification device, as shown in Figure 3 . The device may include: a data acquisition module 310, a model construction module 320, and a model solution module 330.

[0187] A data acquisition module 310 can be used to collect multi-source heterogeneous data related to the operation of the electricity spot market, and preprocess the multi-source heterogeneous data to obtain target multi-source heterogeneous data;

[0188] A model construction module 320 can be used to construct a market clearing verification model and a risk assessment verification model based on the fusion of target multi-source heterogeneous data; the market clearing verification model is used to verify the market clearing results; the risk assessment verification model is used to evaluate various risks existing in the electricity spot market;

[0189] A model solving module 330 can be used to solve the market clearing verification model and the risk assessment verification model based on the particle swarm optimization algorithm to obtain the verification analysis results of the electricity spot market.

[0190] It should be noted that for other corresponding descriptions of each functional module involved in the electricity spot market verification device provided by the embodiments of the present invention, reference can be made to Figure 1 the corresponding descriptions of the method shown, which will not be elaborated here.

[0191] Based on the above method as Figure 1 shown and the embodiments of the device as Figure 3 shown, the embodiments of the present invention also provide an entity structure diagram of an electricity spot market verification system, as Figure 4 shown. The electricity spot market verification system may include a communication bus, a processor, a memory, and a communication interface, and may also include an input / output interface and a display device. Among them, each functional unit can complete mutual communication through the bus. The memory stores a computer program, and the processor is used to execute the program stored on the memory and execute the steps of the electricity spot market verification method in the above embodiments.

[0192] Those skilled in the art can clearly understand that the specific working processes of the above-described system, device, module, and unit can refer to the corresponding processes in the foregoing method embodiments. For the sake of brevity, they will not be elaborated here.

[0193] In addition, in each embodiment of the present invention, each functional unit may be physically independent, or two or more functional units may be integrated together, or all functional units may be integrated in a processing unit. The above integrated functional units may be implemented in the form of hardware, or in the form of software or firmware.

[0194] Those of ordinary skill in the art can understand that: when the integrated functional unit is implemented in the form of software and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, which includes a number of instructions for causing a computing device (such as a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present invention when the instructions are run. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0195] Alternatively, all or part of the steps of implementing the foregoing method embodiments can be completed by hardware related to program instructions (such as a computing device such as a personal computer, a server, or a network device), and the program instructions can be stored in a computer-readable storage medium. When the program instructions are 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 invention.

[0196] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that within the spirit and principles of the present invention, it is still possible to modify the technical solutions described in the foregoing embodiments, or equivalently replace some or all of the technical features therein; and these modifications or replacements do not cause the corresponding technical solutions to deviate from the protection scope of the present invention.

Claims

1. A method for verifying the electricity spot market, characterized in that: The method comprises: Collecting multi-source heterogeneous data related to the operation of the power spot market, and preprocessing the multi-source heterogeneous data to obtain target multi-source heterogeneous data; Constructing a market clearing verification model and a risk assessment verification model based on the target multi-source heterogeneous data fusion; the market clearing verification model is used to verify the market clearing results; the risk assessment verification model is used to evaluate various risks existing in the power spot market; The market clearing verification model and the risk assessment verification model are solved based on the particle swarm optimization algorithm to obtain the electricity spot market verification analysis results.

2. The method according to claim 1, characterized in that The collection of multi-source heterogeneous data related to electricity spot includes: Acquiring power supply related data from the power generation side, including at least one of real-time output data of multiple types of units, power generation plan data, operation status data, and power generation cost data; and, Acquiring grid operation data from the grid side, including at least one of grid topology data, line flow data, node voltage data, transformer parameter data, and safety constraint data; and, Obtaining market transaction data from the trading platform, including at least one of the declaration data of each market entity, transaction data, market clearing price data, auxiliary service transaction data, and meteorological data affecting renewable energy power generation; and, Obtain rule-related data from market regulatory agencies, including at least one of market rules, policies and regulations, and market regulatory data.

3. The method according to claim 1, characterized in that The preprocessing of the multi-source heterogeneous data includes: Constructing data filtering rules based on physical laws and statistical characteristics for multi-source heterogeneous data of different data types, and removing abnormal values ​​from the multi-source heterogeneous data of the data types using the data filtering rules matching the data types; wherein the data types include at least one of unit power generation data, new energy power generation data, power grid flow data, and market transaction data; And, a linear interpolation algorithm is used to fill in the missing data in the multi-source heterogeneous data.

4. The method according to claim 1, characterized in that: The construction of a market clearing verification model and a risk assessment verification model based on the target multi-source heterogeneous data fusion includes: Based on the target multi-source heterogeneous data fusion, the physical characteristics of the power system and the market operation rules are considered, and the market clearing verification model is constructed with minimizing the deviation between the market clearing result and the actual operation data as the objective function; the expression formula of the objective function is: in, The output of power generation company i at time t calculated by the market clearing model; is the actual output of power generation enterprise i at time t; The electricity consumption of user j at time t calculated by the market clearing model; is the actual power consumption of user j at time t; T is the verification time period; n is the number of power generation companies; m is the number of users; Furthermore, based on multi-scenario simulation technology, a risk assessment verification model is constructed for performing price risk assessment, quantity risk assessment and / or new energy fluctuation risk assessment in the electricity spot market.

5. The method according to claim 4, characterized in that The constraints of the market clearing verification model include at least one of power balance constraints, unit operation constraints, ramp rate constraints, minimum start and stop time constraints, power grid security constraints, new energy power generation constraints, and new energy penetration rate constraints; The power balance constraint is expressed as: Among them, L t is the system network loss at time t; The expression formula of the unit operation constraint is: in, and are the minimum and maximum check outputs of unit i at time t respectively; The climbing rate constraint is expressed as: Among them, R i,d and R i,u are the downward and upward climbing rates of unit i respectively; The minimum start-stop time constraint is expressed as: Among them, T i,on,min and T i,off,min are the minimum startup and shutdown time of unit i respectively; and are the startup and shutdown duration of unit i respectively; The expression formula of the power grid security constraint is: in, is the check flow of line l at time t, F l,t,min and F l,t,max are the lower and upper limits of the power flow and the node voltage constraints of line l at time t, respectively; is the calibration voltage of node k at time t, V k,t,min and V k,t,max are the lower and upper limits of the voltage at node k at time t, respectively; The expression formula of the renewable energy power generation constraint is: in, To forecast the output of renewable energy power generation; Actual contribution to renewable energy power generation; min and ε max are the lower and upper limits of the forecast error of renewable energy power generation output, respectively; The expression formula of the new energy penetration rate constraint is: in, is the actual power generation of the new energy source s at time t, and ρ is the new energy penetration requirement.

6. The method according to claim 4, characterized in that The price risk assessment includes: Combining conditional value at risk (CVaR) and value at risk (VaR) to assess market price risk includes: Assume that the market price sequence is P(t), t=1,2,…,T; calculate the risk value parameter VaR using the preset calculation method α , conditional value at risk parameter CVaR α The calculation formula is: Where f(p) is the probability density function of price p; By comparing the risk value parameter VaR α 、The conditional value at risk parameter CVaR α and preset risk thresholds to assess market price risks; The electricity risk assessment includes: Based on the power deviation risk index, the power risk hazards between power supply and demand are evaluated, including: Calculate the first electricity deviation risk between the actual power generation of the power generation enterprise and the planned power generation. The calculation formula is: in, is the actual power generation of power generation enterprise i at time t, is the planned power generation of power generation enterprise i at time t; and Calculate the second power deviation risk between the user's actual power consumption and expected power consumption. The calculation formula is: in, is the actual power consumption of user j at time t, is the expected power consumption of user j at time t; The new energy volatility risk assessment includes: The impact of intermittency and volatility of renewable energy generation on the market is evaluated based on the renewable energy volatility risk index. The calculation formula of the renewable energy volatility risk index is: Among them, P ren (t) is the output sequence of renewable energy power generation; t=1,2,…,T.

7. The method according to claim 1, characterized in that The market clearing verification model and the risk assessment verification model are solved based on the particle swarm optimization algorithm to obtain the power spot market verification analysis results, including: Generate an initial particle population, each particle in the initial particle population represents a potential power distribution strategy; the power distribution strategy includes a power distribution plan for power generation enterprises and a power consumption distribution plan for users; Calculating the objective function value corresponding to the power allocation strategy represented by each particle for the market clearing verification model, and iteratively updating the individual extreme value and the global extreme value according to the objective function value; In the particle swarm iteration process, the adaptive inertia weight strategy and / or the learning factor dynamic adjustment strategy are used to search for the optimal solution; Until the convergence condition is met, the particle swarm reaches the global optimal position and obtains the target power distribution strategy that is closest to the actual operation data; The risk assessment verification model is used to analyze various risks existing in the target power distribution strategy to obtain a power spot market verification analysis result.

8. The method according to claim 7, characterized in that The adaptive inertia weight strategy includes: using a linear decreasing strategy to set the inertia weight ω, which is expressed as: Among them, k is the current iteration number, K max is the maximum number of iterations; The learning factor dynamic adjustment strategy includes: adjusting the learning factors c1 and c2 according to the fitness value of the particle and the average fitness value of the particle group, which is expressed as: Among them, f i is the fitness value of particle i; is the average fitness value of the group; c 1,init 、c 1,end 、c 2,init 、c 2,end are the initial and final values ​​of the learning factor respectively; δ max is the maximum value of δ.

9. A power spot market verification device, characterized in that: The device comprises: A data acquisition module, used to collect multi-source heterogeneous data related to the operation of the power spot market, and pre-process the multi-source heterogeneous data to obtain target multi-source heterogeneous data; A model building module, used to build a market clearing verification model and a risk assessment verification model based on the target multi-source heterogeneous data fusion; the market clearing verification model is used to verify the market clearing results; the risk assessment verification model is used to evaluate various risks existing in the power spot market; The model solving module is used to solve the market clearing verification model and the risk assessment verification model based on the particle swarm optimization algorithm to obtain the power spot market verification analysis results.

10. A power spot market verification system, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the electricity spot market calibration method according to any one of claims 1 to 8.

Images