Novel power system multi-agent cooperative operation optimization method and device

Abstract

The invention discloses a novel power system multi-agent cooperative operation optimization method and device. The method comprises the following steps: constructing a support set and a fuzzy set containing multivariate uncertainty prediction errors; constructing a two-stage distribution robust collaborative operation optimization model of the novel power system in a multi-body participation multi-element market environment; the objective function in the first stage aims at maximizing the total revenue of the novel power system; the target function of the second stage is based on the optimization target of the first stage, and an optimization target of the total income expectation under the worst distribution in the fuzzy set is added; formulating constraint conditions of the first stage; in consideration of random occurrence of any prediction error in the support set, performing linear anti-radiation on the constraint condition of the first stage to obtain a constraint condition of a second stage; and solving the two-stage distribution robust collaborative operation optimization model to obtain an optimal operation scheme of each main body participating in multi-element market collaborative scheduling in the novel power system. According to the invention, the optimal operation that multiple subjects participate in multiple markets can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power system optimization, and in particular to a novel power system multi-agent collaborative operation optimization method and device. Background Art

[0002] At present, some scholars have carried out research on distributed rod optimization scheduling of energy systems, focusing on various energy systems in new power systems, taking into account the uncertainty of renewable energy output on the source side of the system. For example: Chinese patent publication number CN114865665A discloses a distributed rod scheduling optimization method and device for wind power-energy storage systems considering the uncertainty of wind power output, aiming to achieve economic operation of the system and high proportion of renewable energy consumption. Chinese patent publication number CN116667431A discloses a distributed rod operation optimization method for photovoltaic-electrochemical energy storage systems considering the uncertainty of distributed photovoltaic output at ultra-high altitudes, aiming to increase photovoltaic utilization while obtaining higher economic benefits. Chinese patent publication number CN116780649A focuses on an integrated energy system including hydrogen production, electricity, and natural gas, taking into account the uncertainty of wind power output and load demand response of the system, and then proposes a distributed rod operation optimization method and device.

[0003] Based on the above existing technical solutions, it can be seen that the existing research mainly considers the uncertainty of renewable energy output at the source end of the system and power market transactions, and designs a distributed robust scheduling optimization method for the energy system. However, on the one hand, with the active construction of my country's electricity energy market, carbon market, and green certificate market, energy systems often exist in a multi-market environment. The existing methods do not consider the market participation of the energy system or only consider participating in a single electricity market, ignoring the existence of the carbon and green certificate markets, resulting in the inability of the constructed model to guide the scheduling and operation of the energy system under the premise of participating in multiple markets, which does not conform to the actual situation; on the other hand, in addition to the uncertainty of renewable energy output at the source end of the system, the market prices of products such as electricity, carbon quotas, and green certificates in the multiple market environment where the system is located, as well as the load end load, have uncertainty characteristics. The existing methods only consider the uncertainty of renewable energy output at the source side of the system, and do not fully consider the impact of multiple uncertainties such as load side electricity load and commodity prices in the market environment on system scheduling, making the constructed model too simple and unable to be close to the actual situation, which in turn leads to poor accuracy of model solution and weak energy supply stability of the energy system. Summary of the invention

[0004] In order to solve the deficiencies in the prior art, the present invention provides a novel power system multi-subject collaborative operation optimization method and device, by constructing a two-stage distributed blue-robust collaborative operation optimization model that takes into account the source-end photovoltaic and wind power output of the new power system, the load-end electrical load, and the uncertainty of the electricity, carbon, and green certificate market prices in the system environment, thereby obtaining a collaborative operation optimization scheme for multiple subjects such as coal-fired power, photovoltaics, wind power, and energy storage in the new power system participating in electricity, carbon, and green certificate market transactions; while resisting the interference of multiple uncertainties to the stable operation of the system, the present invention can meet the complex development trends of multiple markets in the future at this stage, achieve the optimal operation of multiple subjects participating in multiple markets, improve the overall operating efficiency, and further enhance the overall wind power, photovoltaic and other renewable energy consumption level of the new power system.

[0005] The present invention adopts the following technical solution.

[0006] In a first aspect, the present invention provides a novel method for optimizing the coordinated operation of multiple agents in a power system, the method comprising:

[0007] Step 1: Obtain historical forecast data and real data of wind power output, photovoltaic output, power load, and market prices of unit electricity, carbon quotas, and green certificates to establish support sets and fuzzy sets containing multivariate uncertainty prediction errors;

[0008] Step 2: Construct a two-stage distributed robust collaborative operation optimization model for the new power system with multi-agent participation in a multi-market environment; the objective function of the first stage is to maximize the total revenue of the new power system; the objective function of the second stage is based on the optimization objective of the first stage, adding the optimization objective of the total revenue expectation under the worst distribution in the fuzzy set;

[0009] Step 3: Formulate the constraints of the first stage based on the internal data of each subject in the new power system and the external environment market data;

[0010] Step 4: Taking into account the random occurrence of any prediction error in the support set, linear projection is performed on the constraints of the first stage to obtain the constraints of the second stage;

[0011] Step 5: Solve the two-stage distributed blue-robust collaborative operation optimization model under the constraints to obtain the optimal operation plan for each subject in the new power system to participate in multi-market collaborative dispatch.

[0012] Optionally, the internal data of each subject in the novel power system includes equipment parameters and historical operation data of a coal power subject, a wind power subject, a photovoltaic subject and an electrochemical energy storage subject;

[0013] The external environment market data includes meteorological data, time-of-use electricity prices, carbon quota prices and green certificate prices.

[0014] Optionally, in step 1, the expressions of the support set and fuzzy set containing the multivariate uncertainty prediction error are as follows:

[0015]

[0016] In the formula, For the support set, represents the prediction error of the nth element in the t period, and Respectively The minimum and maximum values ​​of , with superscript n = 1, 2, ..., 6, represent the wind power output, photovoltaic output, power load, and the market price of unit power, carbon quota, and green certificate respectively; T table shows the total number of dispatch periods; in are the set of prediction errors of wind power output, photovoltaic output, electric load, and market prices of unit electricity, carbon quota, and green certificates; is a fuzzy set, Include All functions defined in ; express The dimension of n represents The index of the element in; express The expected value of is equal to the mean of its corresponding sample Indicates that based on sample data The expected absolute deviation does not exceed its corresponding empirical value It means that the global integral of the probability density function is 1.

[0017] Optionally, the objective functions of the first stage and the second stage are expressed as follows:

[0018] F1=max(F D +F T +F L -F Y )

[0019]

[0020] Where F1 is the total daily revenue of the new power system, F D is the daily income of all entities participating in spot power market transactions in the new power system, F T is the daily income of coal-fired power companies participating in carbon market transactions, F Lis the daily income of photovoltaic and wind power entities participating in green certificate market transactions, F Y is the daily operation and maintenance cost of all entities in the new power system; F2 represents the fuzzy set The highest expected value of total return under the worst distribution; Representing fuzzy sets The total return adjustment under the worst distribution in .

[0021] Optionally, the fuzzy set The total return adjustment under the worst distribution in The calculation formula is as follows:

[0022]

[0023] In the formula, It is the daily profit adjustment amount of all entities participating in spot power market transactions in the new power system. It is the daily income adjustment amount of coal-fired power entities participating in carbon market transactions. It is the daily income adjustment amount of photovoltaic and wind power entities participating in green certificate market transactions. It is the daily operation and maintenance cost adjustment of all entities in the new power system.

[0024] Optionally, the constraints of the first stage include interactive constraints on the participation of various entities in the new power system in the spot power market, interactive constraints on the participation of coal-fired power entities in the carbon market, interactive constraints on the participation of photovoltaic and wind power entities in the green certificate market, operating constraints on coal-fired power entities, operating constraints on photovoltaic and wind power entities, operating constraints on electrochemical energy storage entities and / or power node balance constraints.

[0025] Optionally, the expression of the interaction constraint of the coal-fired power entity participating in the carbon market is as follows:

[0026]

[0027] Where, CA i represents the total daily carbon quota of the i-th coal-fired power entity; CE i represents the daily carbon dioxide emissions of the i-th coal-fired power entity; μ CA represents the carbon quota benchmark value of coal-fired power generation units; μ CE represents the unit carbon emission coefficient of coal-fired power units of the i-th coal-fired power entity; represents the electricity sales of the i-th coal-fired power entity at time t; It represents the electric energy sent by the i-th coal-fired power entity to the electrochemical energy storage entity at the t-th time; I represents the number of coal-fired power entities, and T represents the total number of scheduling periods.

[0028] Optional interactive constraints for PV and wind power entities participating in the green certificate market include:

[0029]

[0030] In the formula, and They represent the daily total number of green certificates and the rated number of green certificates of the j-th photovoltaic entity, represents the electricity sales of the j-th photovoltaic entity at the t-th moment, represents the electrical energy sent by the j-th photovoltaic entity to the electrochemical energy storage entity at time t, and They represent the total daily green certificates and rated green certificates of the k-th wind power entity, represents the electricity sales of the k-th wind power entity at the t-th moment; represents the electric energy sent by the k-th wind power entity to the electrochemical energy storage entity at time t; ω PV and ω WT ) represent the power generation quota ratios of photovoltaic entities and wind power entities respectively; J and K represent the number of photovoltaic entities and wind power entities respectively, and T represents the total number of scheduling periods.

[0031] Optionally, the step of solving the two-stage distributed robust collaborative operation optimization model includes:

[0032] Step 5.1: Linearize the fuzzy set and support set;

[0033] Step 5.2: Based on the linearized fuzzy set and support set, the constraints and objective functions in the two-stage distributed robust collaborative operation optimization model are converted into a compact form;

[0034] Step 5.3: Based on the strong duality theory, the min-max structure of the objective function and the semi-infinite form of the constraints in the compact form are decimalized, thereby converting the model into a mixed integer linear programming model;

[0035] Step 5.4: Call the cplex solver to solve the mixed integer linear programming model to obtain the optimal operation plan for each subject in the new power system to participate in the multi-market coordinated dispatch.

[0036] In a second aspect, the present invention provides a novel multi-agent coordinated operation optimization device for a power system, which performs the steps of any method described in the first aspect of the present invention, characterized in that the device comprises:

[0037] Acquisition data unit is used to obtain historical forecast data and real data of wind power output, photovoltaic output, electric load, and market price of unit electricity, carbon quota, and green certificate, so as to establish support set and fuzzy set containing multivariate uncertainty prediction error;

[0038] Construct a model unit to construct a two-stage distributed robust collaborative operation optimization model for a new power system with multiple subjects participating in a multi-market environment; the objective function of the first stage is to maximize the total revenue of the new power system; the objective function of the second stage is based on the optimization objective of the first stage, adding the optimization objective of the total revenue expectation under the worst distribution in the fuzzy set;

[0039] The first constraint unit is used to formulate the constraint conditions of the first stage according to the internal data of each subject in the new power system and the external environment market data;

[0040] The second constraint unit is used to take into account the random occurrence of any prediction error in the support set, perform linear projection on the constraint conditions of the first stage, and obtain the constraint conditions of the second stage;

[0041] The solving and dispatching unit is used to solve the two-stage distributed blue-robust coordinated operation optimization model under the constraints to obtain the optimal operation plan for each subject in the new power system to participate in the coordinated dispatch of multiple markets.

[0042] The beneficial effect of the present invention is that, compared with the prior art,

[0043] 1. The present invention takes into account a more comprehensive market environment for the new power system, including the power market, the carbon market and the green certificate market. In a diversified market environment, different entities in the new power system can participate in different markets based on their own conditions, relevant standards and requirements, thereby obtaining more benefits, deepening the operation and dispatching strategies of the energy system, and expanding the theoretical system of multi-agent collaborative optimization under the existing new power system.

[0044] 2. The present invention takes into account more comprehensive uncertainty interference, including the random factors of photovoltaic and wind power output at the source end of the new power system, the electric load at the load end, and the market prices of electricity, carbon, and green certificates in the system environment. On this premise, a multi-subject distributed blue-rod coordinated operation optimization scheme for coal power, photovoltaic, wind power, electrochemical energy storage, etc. in the new power system is designed; on the one hand, the present invention introduces the uncertainty of multiple market prices to ensure that each subject can achieve economic operation on the basis of effectively avoiding market risks; on the other hand, the present invention introduces the uncertainty of wind power output, photovoltaic output and electric load to improve the operating stability of each subject and the level of renewable energy consumption.

[0045] In summary, the present invention takes into account the uncertainty of photovoltaic and wind power output at the source end of the new power system, the load at the load end, and the market prices of electricity, carbon, and green certificates in the environment in which the system is located, and constructs a two-stage distributed blue-rod collaborative operation optimization model, thereby obtaining a collaborative operation optimization scheme for multiple subjects such as coal-fired power, photovoltaic, wind power, and energy storage in the new power system participating in electricity, carbon, and green certificate market transactions. While resisting the interference of multiple uncertainties to the stable operation of the system, the present invention can meet the complex development trend of multiple markets in the future at this stage, achieve the optimal operation of multiple subjects participating in multiple markets, help provide a reasonable operation strategy for multiple subjects in the new power system, enhance the enthusiasm of each subject for collaborative operation, maximize the benefits of the new power system, and promote the construction of the new power system; the effective use of the distributed blue-rod collaborative operation optimization scheme for multiple subjects of the new power system in a multiple market with multiple uncertainties can optimize the allocation of market resources, and ensure the stable operation of the new power system and the gradual improvement of renewable energy consumption under the background of a high proportion of renewable energy; the present invention can be more suitable for the real environment faced by various subjects in the current new power system, and is suitable for promotion and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of the novel power system multi-agent coordinated operation optimization method of the present invention;

[0047] Figure 2 It is a schematic diagram of the new multi-agent collaborative operation architecture of the power system in the present invention;

[0048] Figure 3 It is a schematic diagram comparing the benefits and carbon emission results of scenario 1 and scenario 2 in the simulation experiment of the present invention;

[0049] Figure 4 It is a schematic diagram comparing the load shedding and wind and solar power abandonment results of scenario 1 and scenario 3 in the simulation experiment of the present invention;

[0050] Figure 5 It is a structural principle block diagram of the novel power system multi-agent collaborative operation optimization method device in the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in the present invention are only part of the embodiments of the present invention, not all of the embodiments. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0052] Embodiment 1:

[0053] With the adjustment of the global energy structure and the continuous opening of the electricity market, the operation optimization problems of various entities in the new power system under different market scenarios are becoming increasingly complex. In a diversified market environment, the coordinated operation optimization of entities such as coal-fired power, photovoltaics, wind power, and electrochemical energy storage in the new power system must not only consider their own energy scheduling, but also coordinate and take into account the impact of participating in energy market transactions, so it becomes very difficult. However, the output of wind and light has obvious uncertainty, which together hinders the stable operation of multiple entities in the new power system. Faced with the complex coupling of multiple entities, how to achieve the optimal operation of entities such as coal-fired power, photovoltaics, wind power, and electrochemical energy storage in the new power system under a diversified market environment under the influence of the uncertainty of wind and light output is a problem that needs to be solved urgently; to this end, the present invention is based on the following Figure 2 The new power system multi-agent collaborative operation architecture shown in the figure provides a new power system multi-agent collaborative operation optimization method, constructs a distributed and robust collaborative operation model of coal-fired power, photovoltaic power, wind power, electrochemical energy storage and other entities in the new power system under a diversified market environment, and obtains the energy dispatch strategy of each entity.

[0054] Reference Figure 1 The novel power system multi-agent coordinated operation optimization method provided by the embodiment of the present invention specifically includes the following steps:

[0055] Step 1: Obtain historical forecast data and real data of wind power output, photovoltaic output, power load, unit electricity, carbon quota, and market price of green certificates to establish support sets and fuzzy sets containing multivariate uncertainty prediction errors; the specific process is as follows:

[0056] First, the present invention considers the uncertainty of the photovoltaic processing, wind power output, load-side electricity load and the market price of electricity, carbon and green certificates in the new power system in the new power system, and specifically describes the uncertainty based on the prediction error value of each of the above random factors. Among them, the prediction errors of photovoltaic, wind power output, electricity load and unit electricity, carbon quota and green certificate market price on a dispatching day can be expressed as random vectors: The corresponding calculation method is:

[0057]

[0058]

[0059] In the formula, They are the day-ahead forecasts of photovoltaic power output, wind power output, power load, unit power, carbon quota, and green certificate market prices.

[0060] They are the actual values ​​of photovoltaic processing, wind power output, electric load, unit electricity, carbon quota, and green certificate market prices. It is worth noting that it is assumed that the main body in the new power system conducts only one transaction with the carbon market and the green certificate market, but in order to facilitate the establishment and conversion of the model in the subsequent text, the transaction with the carbon market on a scheduling day is divided into T time periods, similar to the expression of wind power, photovoltaic output, electric load and electricity market price on a scheduling day. Therefore, the carbon quota and green certificate prices in each time period are equal.

[0061] Let multivariate prediction error but All possible values ​​of are constructed as a support set

[0062]

[0063] In the formula, For the support set, represents the prediction error of the nth element in the t period, and Respectively The minimum and maximum values ​​of , with superscript n = 1, 2, ..., 6, represent the wind power output, photovoltaic output, power load, and the market price of unit power, carbon quota, and green certificate respectively; T table shows the total number of dispatch periods; in are the set of prediction errors of wind power output, photovoltaic output, electric load, and market prices of unit electricity, carbon quota, and green certificates;

[0064] Furthermore, the present invention constructs a fuzzy set based on the prediction error moment information of each random factor By description The possible range of the corresponding joint probability distribution f is:

[0065]

[0066] In the formula is a fuzzy set, Include All functions defined in ; express The dimension of n represents The index of the element in; express The expected value of is equal to the mean of its corresponding sample Indicates that based on sample data The expected absolute deviation does not exceed its corresponding empirical value It means that the global integral of the probability density function is 1.

[0067] Step 2: Construct a two-stage distributed robust collaborative operation optimization model for the new power system with multi-agent participation in a multi-market environment; the objective function of the first stage is to maximize the total revenue of the new power system; the objective function of the second stage is based on the optimization objective of the first stage, adding the optimization objective of the total revenue expectation under the worst distribution in the fuzzy set;

[0068] Specifically, based on the support set and fuzzy sets To describe the multivariate uncertainty, the present invention will take into account the multivariate market transactions and construct a two-stage distributed robust collaborative operation optimization model for a new power system with multi-subject participation in a multi-market environment, where the objective function of the first stage is expressed as follows:

[0069] The expressions of the objective functions of the first and second stages are as follows:

[0070] F1=max(F D +F T +F L -F Y )

[0071]

[0072] Where F1 is the total daily revenue of the new power system, F D is the daily income of all entities participating in spot power market transactions in the new power system, F T is the daily income of coal-fired power companies participating in carbon market transactions, F L is the daily income of photovoltaic and wind power entities participating in green certificate market transactions, F Y is the daily operation and maintenance cost of all entities in the new power system; F2 represents the fuzzy set The highest expected value of total return under the worst distribution; Representing fuzzy sets The total return adjustment under the worst distribution in;

[0073] In F D middle, (constant) and (variables) represent the electricity price and electricity sales volume of the i-th coal-fired power entity at time t; (constant) and (variables) represent the electricity sales price and electricity sales volume of the jth photovoltaic entity at the tth moment; (constant) and (variables) represent the electricity sales price and electricity sales volume of the k-th wind power entity at the t-th moment; (constant) and (variables) represent the electricity selling price and electricity selling (discharging) of the lth electrochemical energy storage entity at the tth moment; I, J, K and L represent the number of coal power entities, photovoltaic entities, wind power entities and electrochemical energy storage entities respectively; in F T middle, (constant) and CA i (variable), CE i (variables) represent the daily carbon quota price, daily total carbon quota, and total carbon emissions of the i-th coal-fired power entity; in F L middle, (constant), (variable), (variables) represent the daily green certificate price, daily total green certificates, and rated green certificate quantity of the j-th photovoltaic entity; (constant) and (variable), (variables) represent the daily green certificate price, daily total green certificate, and rated green certificate quantity of the k-th wind power entity; Y Medium, SD i (constant) and SU i (constant) represent the shutdown and startup costs of the i-th coal-fired power entity, (variable) and (variable) is a 0-1 variable, which represents the shutdown and startup status of the i-th coal-fired power entity at the t-th time; (constant), (constant), (constant) and (constant) represents the operation and maintenance cost per unit output (charging and discharging) of the i-th coal power entity, the j-th photovoltaic entity, the k-th wind power entity and the l-th electrochemical energy storage entity respectively; and It represents the discharge (sale) and charge capacity of the lth electrochemical energy storage entity at the tth moment.

[0074] It is further explained that when the prediction error of photovoltaic and wind power output occurs, the coal-fired power, photovoltaic, wind power and electrochemical energy storage in the new power system need to be re-dispatched to ensure the stability of the new power system, so the daily total revenue will change accordingly; therefore, the objective function of the second stage of the present invention takes into account that the value and distribution of the prediction error are uncertain, and in the fuzzy set The expected value of the total return change is optimized under the "worst" distribution in item( represents the total benefit adjustment, which is given in detail below), thereby forming a max-min double-layer objective structure to resist uncertain interference. The specific expression is given below (Formulas (40)-(44)); the inner min minimization objective is used to dynamically search for the distribution scenario with the lowest ("worst") expected value of total benefit in the aforementioned fuzzy set; the outer max maximization objective is used to pursue the highest expected value of total benefit under a given distribution.

[0075] Step 3: Formulate the constraints of the first stage based on the internal data of each subject in the new power system and the external environment market data;

[0076] Step 4: Taking into account the random occurrence of any prediction error in the support set, linear projection is performed on the constraints of the first stage to obtain the constraints of the second stage;

[0077] Among them, the internal data of each entity in the new power system include the equipment parameters and historical operation data of coal-fired power, wind power, photovoltaic and electrochemical energy storage; the external environment market data include meteorological data, time-of-use electricity prices, carbon quota prices and green certificate prices.

[0078] It should be noted that, taking into account the uncertainty interference of photovoltaic and wind power output, the coordinated operation optimization of coal-fired power, photovoltaic, wind power and electrochemical energy storage in the new power system presents a two-stage form; in the first stage, based on the photovoltaic output forecast information, including the energy dispatch of coal-fired power, photovoltaic, wind power and electrochemical energy storage in the new power system, the relevant constraints of multi-subject participation in multi-market transactions are constructed. In the second stage, considering the prediction error in the support set In order to take into account the random occurrence of the random values ​​in the new power system, robust constraints on the re-dispatching of coal-fired power, photovoltaic power, wind power and electrochemical energy storage in the new power system are constructed to ensure the stable operation of the new power system under the interference of photovoltaic output uncertainty and to accommodate the random values ​​of photovoltaic output.

[0079] As an embodiment of the present invention, in step 3, the constraints of the first stage include the interactive constraints of various entities in the new power system participating in the spot power market, the interactive constraints of coal-fired power entities participating in the carbon market, the interactive constraints of photovoltaic and wind power entities participating in the green certificate market, the coal-fired power entity operation constraints, the photovoltaic and wind power entity operation constraints, the electrochemical energy storage entity operation constraints and / or the electric node balance constraints; specifically as follows:

[0080] ① Interactive constraints of various entities participating in the spot power market in the new power system

[0081] The present invention focuses on the day-ahead dispatching problem of each subject in the new power system. Therefore, only coal-fired power, photovoltaic power, wind power and electrochemical energy storage are considered to participate in the spot trading market. Due to the limitation of line transmission capacity, each subject participating in the spot power market transaction should meet the following constraints:

[0082]

[0083] In the formula, (Constant) represents the maximum electric power allowed to be transmitted by the transmission line.

[0084] ②Interactive constraints on coal-fired power entities participating in the carbon market

[0085] The carbon market incentivizes coal-fired power entities to achieve the goal of carbon emission reduction through cost. When the actual carbon emissions of coal-fired power entities are less than the carbon quotas obtained, they can sell excess carbon quotas to obtain income. Otherwise, they need to purchase carbon quotas to meet the assessment requirements. This model uses the benchmark method to calculate the total daily carbon quota CA of the i-th coal-fired power entity. i And is allocated to the entity free of charge, and the carbon quota constraints are as follows:

[0086]

[0087] Where, CA i represents the total daily carbon quota of the i-th coal-fired power entity; μ CA (constant) represents the carbon quota benchmark value of coal-fired power generation units. It represents the electric energy sent by the i-th coal power entity to the electrochemical energy storage entity at the t-th time.

[0088] Furthermore, based on the carbon emission coefficient of coal combustion, the daily carbon dioxide emissions of the i-th coal-fired power entity are calculated:

[0089]

[0090] In the formula, μ CE (Constant) represents the carbon emission coefficient per unit of electricity generated by the coal-fired power units of the i-th coal-fired power entity.

[0091] ③Interactive constraints on photovoltaic and wind power entities participating in the green certificate market

[0092] Similar to the carbon market, photovoltaic and wind power entities that meet the renewable energy quota sell green certificates in the green certificate market based on the actual power generation of all their own renewable energy. If they do not meet the quota requirements, they need to purchase green certificates in the green certificate market to achieve balance; the actual green certificates obtained are calculated as follows:

[0093]

[0094] In the formula, represents the total daily green certificate quantity of the j-th photovoltaic entity, represents the total daily green certificate quantity of the k-th wind power entity, (variable) represents the electrical energy sent by the j-th photovoltaic entity to the electrochemical energy storage entity at time t, (variable) represents the electric energy sent by the k-th wind power entity to the electrochemical energy storage entity at time t.

[0095] Furthermore, the number of green certificates that meet the quota requirements of the j-th photovoltaic entity and the k-th wind power entity is calculated:

[0096]

[0097] In the formula, represents the daily rated green certificate quantity of the j-th photovoltaic entity, represents the daily rated green certificate quantity of the k-th wind power entity, ω PV (constant) and ω WT (constant) represents the quota ratio of photovoltaic and wind power generation respectively.

[0098] ④ Constraints on the operation of coal-fired power plants

[0099]

[0100] In the formula, and is a 0-1 variable, representing the shutdown and startup status of the i-th coal-fired power entity at time t; and is a 0-1 variable, representing the operating status of the i-th coal-fired power entity at time t and time t-1 respectively; θ i represents the ramp coefficient of the i-th coal-fired power entity; and They represent the upper and lower limits of the power of the i-th coal-fired power entity respectively; and They represent the amount of electricity sold by the i-th coal-fired power entity to the power market at time t and time t-1 respectively; and They represent the amount of electricity delivered to the electrochemical energy storage by the i-th coal-fired power entity at time t and time t-1 respectively.

[0101] ⑤ Constraints on the operation of photovoltaic and wind power entities

[0102]

[0103] In the formula, (variable) and (variables) represent the amount of electricity sold to the electricity market and the amount of electricity delivered to the electrochemical energy storage entity by the jth photovoltaic entity at time t; (constant) represents the predicted output of the jth PV entity at time t; (variable) and (variables) represent the amount of electricity sold to the electricity market and the amount of electricity delivered to the electrochemical energy storage entity by the k-th wind power entity at time t; (constant) represents the predicted output of the kth wind power entity at the tth moment.

[0104] ⑥Operation constraints of electrochemical energy storage

[0105]

[0106]

[0107] In the formula, Φ l,t (variable) and Φ l,t-1 (variable) represents the battery storage capacity of the lth electrochemical energy storage subject at the tth moment and the t-1th moment respectively; (constant) and (constants) represent the charging and discharging efficiency of the lth electrochemical energy storage entity; Φ l,T (variable) and Φ l,0 (variable) represents the battery storage capacity of the first electrochemical energy storage subject at the end and the beginning of the day respectively; (variable) and (variable) is the charge / discharge state of the lth electrochemical element at time t, which is a 0-1 variable; (constant) and (constant) represents the rated charge and discharge power of the first electrochemical; Φ max (constant) and Φ min (constants) respectively represent the upper and lower limits of the storage capacity of the lth electrochemical energy storage entity.

[0108] ⑦Electric node balance constraints

[0109]

[0110] In the formula, (Constant) represents the predicted electric load at time t.

[0111] It is further explained that the actual output of photovoltaic and wind power entities is often difficult to accurately predict, and there will be deviations between it and the predicted value, that is, prediction errors. Therefore, in the second stage, taking into account the occurrence of multivariate prediction errors, the linear affine strategy is used to re-dispatch the entities in the new power system:

[0112]

[0113] In the formula, Indicates the operating results of coal-fired power, photovoltaic power, wind power and electrochemical energy storage in the first stage The redispatching result obtained after adjustment based on ; Respectively represent the corresponding The adjustment rate vector.

[0114] When coal-fired power, photovoltaic power, wind power and electrochemical energy storage are redispatched to resist the uncertainty interference of wind and solar output, the total daily revenue of the new power system (i.e., formula (9)) will change accordingly. Therefore, the fuzzy set in the objective function of the second stage is The total return adjustment under the worst distribution in It can be specifically expressed as:

[0115]

[0116] In the formula, It is the daily profit adjustment amount of all entities participating in spot power market transactions in the new power system. It is the daily income adjustment amount of coal-fired power entities participating in carbon market transactions. It is the daily income adjustment amount of photovoltaic and wind power entities participating in green certificate market transactions. is the daily operation and maintenance cost adjustment of all entities in the new power system; middle, and They represent the actual electricity sales price, the day-ahead predicted electricity sales price, the actual electricity sales volume, and the day-ahead predicted electricity sales volume of the i-th coal-fired power entity at time t respectively; and They represent the actual electricity sales price, the day-ahead predicted electricity sales price, the actual electricity sales volume, and the day-ahead predicted electricity sales volume of the j-th photovoltaic entity at the t-th moment respectively; and They represent the actual electricity sales price, the day-ahead predicted electricity sales price, the actual electricity sales volume and the day-ahead predicted electricity sales volume of the k-th wind power entity at the t-th moment respectively; and They represent the actual electricity sales price, the day-ahead predicted electricity sales price, the actual discharge (sale) amount and the day-ahead predicted discharge (sale) amount of the lth electrochemical energy storage entity at the tth moment; I, J, K and L represent the number of coal-fired power entities, photovoltaic entities, wind power entities and electrochemical energy storage entities respectively; in F T middle, and They represent the actual carbon quota price, the day-ahead predicted carbon quota price, the actual daily carbon quota and the actual daily carbon emissions of the i-th coal power entity respectively; CA i and CEi They represent the daily total carbon quota and total carbon emissions of the i-th coal-fired power entity respectively; L middle, and They represent the daily green certificate price, daily total green certificates and rated number of green certificates of the jth photovoltaic entity respectively; and They represent the daily green certificate price, daily total green certificate and rated green certificate quantity of the k-th wind power entity respectively; Y middle, and They represent the operation and maintenance costs per unit output of the i-th coal power entity, the j-th photovoltaic entity, the k-th wind power entity, and the l-th electrochemical energy storage entity respectively; and They represent the actual amount of electricity and the predicted amount of electricity delivered by the i-th coal power entity to the electrochemical energy storage entity at time t respectively; and They represent the actual amount of electricity and the predicted amount of electricity delivered by the j-th photovoltaic entity to the electrochemical energy storage entity at time t, respectively; and They respectively represent the actual amount of electricity and the predicted amount of electricity delivered by the k-th wind power entity to the electrochemical energy storage entity at time t; represents the predicted charge capacity of the lth electrochemical energy storage subject at the tth moment; It represents the actual charge amount of the lth electrochemical energy storage entity at the tth moment; T represents the total number of scheduling periods.

[0117] In addition, when the prediction error occurs randomly, the re-dispatched coal power, photovoltaic, wind power and electrochemical energy storage entities still need to meet the relevant constraints to ensure the stable operation of the new power system; therefore, considering the support set The random occurrence of any prediction error in , linearly projecting the constraints of the first stage, yields the constraints corresponding to the second stage:

[0118]

[0119]

[0120] In the formula, and They represent the amount of electricity actually sold to the electricity market and the amount of electricity actually delivered to the electrochemical energy storage by the i-th coal-fired power entity at time t-1; and They represent the actual battery storage capacity of the lth electrochemical energy storage subject at the tth moment and the t-1th moment respectively; Indicates the final actual battery storage capacity of the lth electrochemical energy storage subject within the day; Represents the actual electric load at time t. The meanings of formulas (45)-(66) are similar to those of formulas (14)-(24), (27)-(34), and (36)-(38), respectively. The main difference is that formulas (45)-(66) are constraints on the re-dispatch of coal-fired power, photovoltaic power, wind power, and electrochemical energy storage when multivariate random factor prediction errors occur.

[0121] Finally, a two-stage distributed blue-robust coordinated operation optimization model for coal-fired power, photovoltaic power, wind power, and electrochemical energy storage in a new power system under a multi-market environment with formulas (9)-(13) (including formulas (40)-(44)) as targets and formulas (14)-(39) and (44)-(66) as constraints was fully constructed.

[0122] Step 5: Solve the two-stage distributed blue-robust collaborative operation optimization model under the constraints to obtain the optimal operation plan for each subject in the new power system to participate in multi-market collaborative dispatch.

[0123] In this embodiment, step 5 of solving the two-stage distributed blue stick cooperative operation optimization model includes:

[0124] Step 5.1: Linearize the fuzzy set and support set;

[0125] exist middle, With nonlinear structure, this paper introduces auxiliary variables ( express The upper limit of ) makes it linear. Therefore, and is modified to the following form:

[0126]

[0127] Step 5.2: Based on the linearized fuzzy set and support set, the constraints and objective functions in the two-stage distributed robust collaborative operation optimization model are converted into a compact form; the specific process is as follows:

[0128] S5.2.1. To facilitate the description of the solution in the following text, the fuzzy set and support set Rewritten in compact form:

[0129]

[0130] Where n is and The index of the element in . and is the coefficient vector. is the parameter vector.

[0131] S5.2.2. Rewrite the model {Formula (9)-(66)} into a condensed form as follows:

[0132]

[0133] Wherein, formula (71) is the condensed form of the objective function formula (9) (including formula (40), and x is the corresponding non-flexible resources of each subject in the new power system, including The vector composed of They are x, The coefficient vector corresponding to the term; Formula (72) is a condensed form of constraint formulas (14)-(38), and They are x, y 0 The corresponding coefficient matrix is, is a constant vector; Formula (73) is a compact form of constraint formulas (45)-(66), They are x, The corresponding coefficient matrix is, The expression of is formula (74), which represents the compact form of formula (39) (linear affine), Yes constant vector, which can be further written in compact form as follows:

[0134]

[0135] in, Does not contain A real vector of yes The corresponding coefficient vector. InEq.(45), y 0 Is The vector composed of represents the two-stage operation results of the flexibility resources. Is The vector composed of represents the third-stage rescheduling result of flexible resources. Is The adjustment rate vector composed of Is The adjustment rate vector composed of .

[0136] Step 5.3: Based on the strong duality theory, the min-max structure of the objective function and the semi-infinite form of the constraints in the compact form are decimalized to convert the model into a mixed integer linear programming model; the specific process is as follows:

[0137] S5.3.1, Objective function min-max structure elimination:

[0138] Based on the strong duality theory, the objective function formula (71) is To convert; it should be emphasized that the variable is The constraint condition during dual transformation is formula (69);

[0139]

[0140] In the formula, θ, α n and β n They correspond to as well as middle The dual variable of .

[0141] Obviously, It's about variables Convex optimization, so the strong duality holds, and the model {Formula (71)-(74)} can be transformed into:

[0142]

[0143] st formula (72)-(74), (77), (78)

[0144] At this time, the min-max structure in the objective function is eliminated.

[0145] S5.3.2, Semi-infinite form constraint elimination:

[0146] Substituting formula (74) into formula (73) and (77) respectively, and substituting formula (75) into formula (73), we can obtain:

[0147]

[0148] Given that The number of is huge, so the constraint formulas (80) and (81) can be regarded as semi-infinite forms and need to be further eliminated.

[0149] First, formulas (80) and (81) can be rewritten equivalently as follows:

[0150]

[0151] At this time, the left sides of the two constraint inequalities can be regarded as and is the objective function of the variable, and the corresponding constraint is formula (70); based on the strong duality theory, formulas (82) and (83) can be transformed into:

[0152]

[0153] ξ≥O (87)

[0154]

[0155] ∈≤0 (91)

[0156] where the matrix ξ is the dual variable corresponding to Transform formula (82) into formula (84)-(87). The vector ∈ is the dual variable corresponding to Transform formula (83) into formula (88)-(91). Obviously, formula (82) and (83) are about For convex optimization, the strong duality holds.

[0157] At this point, the semi-infinite form in the constraints is eliminated.

[0158] S5.3.3. Finally, a standard directly solvable mixed integer linear programming model is formed: objective function: formula (79); constraints: formulas (72), (78), (83)-(90); that is, the final expression of the mixed integer linear programming model is as follows:

[0159]

[0160] ξ≥O

[0161]

[0162] ∈≤0

[0163] Step 5.4: Call the cplex solver to solve the mixed integer linear programming model to obtain the optimal operation plan for each subject in the new power system to participate in the multi-market coordinated dispatch. The specific process is as follows:

[0164] S5.4.1. Initialization:

[0165] Initialization time quantity T = 24, indicating that a scheduling day is divided into 24 time periods. Initialization number of coal power entities i = 5, number of photovoltaic entities j = 3, number of wind power entities k = 3, number of electrochemical energy storage entities l = 4

[0166] Initialize the day-ahead forecast PV output data for the tth period Wind turbine output data Electric load data

[0167] Initialize the parameter values ​​in the model objective function, including μn , δ n .

[0168] Initialize the parameter values ​​in the model constraints, including

[0169] S5.4.2. Set variables:

[0170] Set and define the variables for the tth period, including x, y 0 ,θ,α n , β n ,ξ, ∈.

[0171] S5.4.3. Call the Yalmip toolkit to write and store constraints based on the built model:

[0172] Set up a storage space Eq, and store the equality and inequality constraints between variables and parameters in the constraint formulas (72), (78), (83)-(90) in the constructed model in the space Eq.

[0173] S5.4.4. Write and store the objective function:

[0174] Set the storage space Obj, and based on the variable settings and parameter assignments, write and store the objective function in Obj.

[0175] S5.4.5. Call the cplex solver to solve the model:

[0176] Call the cplex solver, input the constraint storage space Eq and the objective function storage space Obj, solve and obtain the optimal value of the variable.

[0177] In summary, the optimization problem of distributed blue-rod coordinated operation of multiple subjects in a new power system under a multiple market taking into account multiple uncertainties proposed in the embodiments of the present invention is more in line with the complex reality and can promote the construction and development of the new power system; specifically, based on different market trading mechanisms, different subjects such as coal-fired power, photovoltaic power, wind power and electrochemical energy storage in the new power system dispatch their own electric energy to participate in market transactions and carry out cooperation among subjects, which can improve resource utilization efficiency and use the market to achieve a high proportion of renewable energy consumption and improve the overall benefits of all subjects; constructing a two-stage distributed blue-rod coordinated operation optimization model can ensure that the multi-subject operation of the new power system can resist the interference of multiple uncertainties and maintain stable operation of the system under prediction errors.

[0178] The effectiveness of the novel power system multi-agent collaborative operation optimization method provided by the embodiment of the present invention is verified by a specific experimental example below.

[0179] Three scenarios were set up in the simulation experiment for comparison, highlighting the advantages of the method of the present invention in considering the participation of electricity, carbon and green certificate markets and multivariate uncertainties to dispatch the coordinated operation of multiple subjects in the new power system. The specific scenarios are as follows:

[0180] Scenario 1: In the new power system, multiple entities such as coal-fired power, photovoltaic power, wind power, and energy storage participate in the electricity, carbon, and green certificate market transactions, and take into account the photovoltaic and wind power output at the source end, the load at the load end, and the uncertainty of the electricity, carbon, and green certificate market prices in the system environment. The designed distributed blue-robust optimization method is implemented in the operation and dispatch of multiple entities such as coal-fired power, photovoltaic power, wind power, and energy storage.

[0181] Scenario 2: In the new power system, multiple entities such as coal-fired power, photovoltaic power, wind power, and energy storage only participate in power market transactions, and take into account the output of photovoltaic power and wind power at the source end, the load at the load end, and the uncertainty of power market prices in the system environment, and implement the designed distributed blue-robust optimization method in the operation and dispatch of multiple entities such as coal-fired power, photovoltaic power, wind power, and energy storage. Specifically, compared with scenario 1, the objectives, constraints, and prediction errors related to carbon and green certificate market transactions are removed.

[0182] Scenario 3: In the new power system, multiple entities such as coal-fired power, photovoltaic power, wind power, and energy storage participate in the electricity, carbon, and green certificate market transactions. Only the uncertainty of photovoltaic and wind power output at the source is taken into account, and the designed distributed blue-robust optimization method is implemented in the operation and dispatch of multiple entities such as coal-fired power, photovoltaic power, wind power, and energy storage. Specifically, compared with scenario 1, the prediction errors related to the multi-market transaction price and electricity load are removed.

[0183] Comparison 1: Comparison between Scenario 1 and Scenario 2 verifies the advantages of multiple entities in the new power system participating in electricity, carbon and green certificate market transactions at the same time compared to only participating in electricity market transactions.

[0184] like Figure 3 As shown in the figure, the benefits in scenario 1 are greater than those in scenario 2. This shows that the new power system that operates under the premise of participating in the multi-market transactions of electricity, carbon and green certificates can obtain more benefits. On the one hand, the coal-fired power entities further increase their own income by participating in the carbon market transactions and selling excess carbon quotas to the outside; on the other hand, the photovoltaic and wind power entities that meet the renewable energy quotas sell green certificates in the green certificate market according to the actual power generation of all their own renewable energy to further profit. Therefore, the new power system can increase its own benefits by participating in the multi-market transactions of electricity, carbon and green certificates at the same time, which is in line with the principle of economic operation. Moreover, the carbon emissions in scenario 1 are significantly smaller than those in scenario 2. Carbon trading, as a market-based approach to carbon emission reduction, will effectively promote the resource allocation of carbon emissions. At the same time, when the carbon quota is the same, the larger the proportion of green certificates, the greater the renewable energy power generation, and the total carbon emissions will continue to decline. Therefore, the deep integration of the three markets can promote the low-carbon development of the entire society.

[0185] Experimental results show that compared with only participating in the electricity market, the new power system proposed in this invention can achieve the improvement of the overall economic and environmental benefits of the system by dispatching and operating while participating in electricity and carbon green certificate market transactions.

[0186] Comparison 2: Comparison of scenario 1 and scenario 3 verifies the advantages of implementing the designed distributed blue-rod optimization method in the operation and dispatch of multiple entities such as coal-fired power, photovoltaic power, wind power, and energy storage, while considering the uncertainty of photovoltaic and wind power output at the source end, the load at the load end, and the uncertainty of the electricity, carbon, and green certificate markets in the system environment. This method is compared with only considering the uncertainty of photovoltaic and wind power output at the source end, and the uncertainty of the photovoltaic and wind power output at the load end, and the electricity, carbon, and green certificate markets at the system environment.

[0187] like Figure 4 As shown in the figure, the load shedding value in scenario 1 is smaller than that in scenario 3. This shows that in any period of time, in addition to considering the uncertainty of wind power and photovoltaic output at the source end, if the uncertainty of electricity load at the load end and the uncertainty of electricity, carbon and green certificate market prices under the multi-market are also considered, and the new power system is dispatched according to the distributed blue-robust optimization strategy in scenario 1, the energy supply stability of the coordinated operation of wind power, photovoltaic power, coal power and electrochemical energy storage can be guaranteed. In addition, the amount of wind abandonment and light abandonment in scenario 1 are much smaller than those in scenario 3. This means that the proposed model can greatly improve the absorption of renewable energy in the new power system in the random environment described, while the model corresponding to scenario 3 cannot achieve the same results as the proposed model. It is precisely because of the existence of the flexible resource adjustment rate corresponding to the wind power, photovoltaic output, electricity load and market price forecast errors that the operation stability of the new power system and the absorption of a high proportion of renewable energy are guaranteed.

[0188] The experimental results show that taking into account the uncertainty of wind power and photovoltaic output at the source end, the uncertainty of electricity load at the load end, and the random interference of uncertainty in electricity, carbon and green certificate market prices under multiple markets, the model proposed in this invention can ensure that multiple subjects in the new power system can achieve operational economy, stability and renewable energy absorption on the basis of effectively avoiding market risks.

[0189] The beneficial effect of the present invention is that, compared with the prior art,

[0190] 1. The present invention takes into account a more comprehensive market environment for the new power system, including the power market, the carbon market and the green certificate market. In a diversified market environment, different entities in the new power system can participate in different markets based on their own conditions, relevant standards and requirements, thereby obtaining more benefits, deepening the operation and dispatching strategies of the energy system, and expanding the theoretical system of multi-agent collaborative optimization under the existing new power system.

[0191] 2. The present invention takes into account more comprehensive uncertainty interference, including the random factors of photovoltaic and wind power output at the source end of the new power system, the electric load at the load end, and the market prices of electricity, carbon, and green certificates in the system environment. On this premise, a multi-subject distributed blue-rod coordinated operation optimization scheme for coal power, photovoltaic, wind power, electrochemical energy storage, etc. in the new power system is designed; on the one hand, the present invention introduces the uncertainty of multiple market prices to ensure that each subject can achieve economic operation on the basis of effectively avoiding market risks; on the other hand, the present invention introduces the uncertainty of wind power output, photovoltaic output and electric load to improve the operating stability of each subject and the level of renewable energy consumption.

[0192] In summary, the present invention takes into account the uncertainty of photovoltaic and wind power output at the source end of the new power system, the load at the load end, and the market prices of electricity, carbon, and green certificates in the environment in which the system is located, and constructs a two-stage distributed blue-rod collaborative operation optimization model, thereby obtaining a collaborative operation optimization scheme for multiple subjects such as coal-fired power, photovoltaic, wind power, and energy storage in the new power system participating in electricity, carbon, and green certificate market transactions. While resisting the interference of multiple uncertainties to the stable operation of the system, the present invention can meet the complex development trend of multiple markets in the future at this stage, achieve the optimal operation of multiple subjects participating in multiple markets, help provide a reasonable operation strategy for multiple subjects in the new power system, enhance the enthusiasm of each subject for collaborative operation, maximize the benefits of the new power system, and promote the construction of the new power system; the effective use of the distributed blue-rod collaborative operation optimization scheme for multiple subjects of the new power system in a multiple market with multiple uncertainties can optimize the allocation of market resources, and ensure the stable operation of the new power system and the gradual improvement of renewable energy consumption under the background of a high proportion of renewable energy; the present invention can be more suitable for the real environment faced by various subjects in the current new power system, and is suitable for promotion and application.

[0193] Embodiment 2:

[0194] like Figure 5 As shown, the present invention provides a novel power system multi-agent coordinated operation optimization device, which is used to implement the steps of the method in the above embodiment 1, and the device specifically includes:

[0195] Acquisition data unit is used to obtain historical forecast data and real data of wind power output, photovoltaic output, electric load, and market price of unit electricity, carbon quota, and green certificate, so as to establish support set and fuzzy set containing multivariate uncertainty prediction error;

[0196] Construct a model unit to construct a two-stage distributed robust collaborative operation optimization model for a new power system with multiple subjects participating in a multi-market environment; the objective function of the first stage is to maximize the total revenue of the new power system; the objective function of the second stage is based on the optimization objective of the first stage, adding the optimization objective of the total revenue expectation under the worst distribution in the fuzzy set;

[0197] The first constraint unit is used to formulate the constraint conditions of the first stage according to the internal data of each subject in the new power system and the external environment market data;

[0198] The second constraint unit is used to take into account the random occurrence of any prediction error in the support set, perform linear projection on the constraint conditions of the first stage, and obtain the constraint conditions of the second stage;

[0199] The solving and dispatching unit is used to solve the two-stage distributed blue-robust coordinated operation optimization model under the constraints to obtain the optimal operation plan for each subject in the new power system to participate in the coordinated dispatch of multiple markets.

[0200] The novel power system multi-subject collaborative operation optimization device provided in the embodiment of the present invention and the novel power system multi-subject collaborative operation optimization method provided in Example 1 are based on the same technical concept, and can produce the beneficial effects as described in Example 1. For the contents not described in detail in this embodiment, please refer to Example 1.

[0201] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0202] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0203] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0204] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0205] 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, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A novel multi-agent coordinated operation optimization method for power system, characterized in that: Methods include: Step 1: Obtain historical forecast data and real data of wind power output, photovoltaic output, power load, and market prices of unit electricity, carbon quotas, and green certificates to establish support sets and fuzzy sets containing multivariate uncertainty prediction errors; Step 2: Construct a two-stage distributed robust collaborative operation optimization model for the new power system with multi-agent participation in a multi-market environment; the objective function of the first stage is to maximize the total revenue of the new power system; the objective function of the second stage is based on the optimization objective of the first stage, adding the optimization objective of the total revenue expectation under the worst distribution in the fuzzy set; Step 3: Formulate the constraints of the first stage based on the internal data of each subject in the new power system and the external environment market data; Step 4: Taking into account the random occurrence of any prediction error in the support set, linear projection is performed on the constraints of the first stage to obtain the constraints of the second stage; Step 5: Solve the two-stage distributed blue-robust collaborative operation optimization model under the constraints to obtain the optimal operation plan for each subject in the new power system to participate in multi-market collaborative dispatch.

2. The novel multi-agent coordinated operation optimization method for power system according to claim 1 is characterized in that: The internal data of each subject in the novel power system includes equipment parameters and historical operation data of the coal power subject, the wind power subject, the photovoltaic subject and the electrochemical energy storage subject; The external environment market data includes meteorological data, time-of-use electricity prices, carbon quota prices and green certificate prices.

3. The novel multi-agent coordinated operation optimization method for power system according to claim 1 is characterized in that: In step 1, the expressions of the support set and fuzzy set containing multivariate uncertainty prediction errors are as follows: In the formula, For the support set, represents the prediction error of the nth element in the t period, and Respectively The minimum and maximum values ​​of , with superscript n = 1, 2, ..., 6, represent the wind power output, photovoltaic output, power load, and the market price of unit power, carbon quota, and green certificate respectively; T table shows the total number of dispatch periods; in are the set of prediction errors of wind power output, photovoltaic output, electric load, and market prices of unit electricity, carbon quota, and green certificates; is a fuzzy set, Include All functions defined in ; express The dimension of n represents The index of the element in; express The expected value of is equal to the mean of its corresponding sample Indicates that based on sample data The expected absolute deviation does not exceed its corresponding empirical value It means that the global integral of the probability density function is 1.

4. The novel power system multi-agent coordinated operation optimization method according to claim 1 is characterized in that: The expressions of the objective functions of the first stage and the second stage are as follows: F1=max(F D +F T +F L -F Y ) Where F1 is the total daily revenue of the new power system, F D is the daily income of all entities participating in spot power market transactions in the new power system, F T is the daily income of coal-fired power companies participating in carbon market transactions, F L is the daily income of photovoltaic and wind power entities participating in green certificate market transactions, F Y is the daily operation and maintenance cost of all entities in the new power system; F2 represents the fuzzy set The highest expected value of total return under the worst distribution in; Representing fuzzy sets The total return adjustment under the worst distribution in .

5. The novel multi-agent coordinated operation optimization method for power system according to claim 4 is characterized in that: The fuzzy set The total return adjustment under the worst distribution in The calculation formula is as follows: In the formula, It is the daily profit adjustment amount of all entities participating in spot power market transactions in the new power system. It is the daily income adjustment amount of coal-fired power entities participating in carbon market transactions. It is the daily income adjustment amount of photovoltaic and wind power entities participating in green certificate market transactions. It is the daily operation and maintenance cost adjustment of all entities in the new power system.

6. The novel multi-agent coordinated operation optimization method for power system according to claim 1 is characterized in that: The constraints of the first stage include the interactive constraints of various entities in the new power system participating in the spot power market, the interactive constraints of coal-fired power entities participating in the carbon market, the interactive constraints of photovoltaic and wind power entities participating in the green certificate market, the operating constraints of coal-fired power entities, the operating constraints of photovoltaic and wind power entities, the operating constraints of electrochemical energy storage entities and / or power node balance constraints.

7. The novel multi-agent coordinated operation optimization method for power system according to claim 6 is characterized in that: The expression of the interactive constraints of coal-fired power entities participating in the carbon market is as follows: Where, CA i represents the total daily carbon quota of the i-th coal-fired power entity; CE i represents the daily carbon dioxide emissions of the i-th coal-fired power entity; μ CA represents the carbon quota benchmark value of coal-fired power generation units; μ CE represents the unit carbon emission coefficient of coal-fired power units of the i-th coal-fired power entity; represents the electricity sales of the i-th coal-fired power entity at time t; It represents the electric energy sent by the i-th coal-fired power entity to the electrochemical energy storage entity at the t-th time; I represents the number of coal-fired power entities, and T represents the total number of scheduling periods.

8. The novel multi-agent coordinated operation optimization method for power system according to claim 6 is characterized in that: The interactive constraints for photovoltaic and wind power entities participating in the green certificate market include: In the formula, and They represent the daily total number of green certificates and the rated number of green certificates of the j-th photovoltaic entity, represents the electricity sales of the j-th photovoltaic entity at the t-th moment, represents the electrical energy sent by the j-th photovoltaic entity to the electrochemical energy storage entity at time t, and They represent the total daily green certificates and rated green certificates of the k-th wind power entity, represents the electricity sales of the k-th wind power entity at the t-th moment; represents the electric energy sent by the k-th wind power entity to the electrochemical energy storage entity at time t; ω PV and ω WT ) represent the power generation quota ratios of photovoltaic entities and wind power entities respectively; J and K represent the number of photovoltaic entities and wind power entities respectively, and T represents the total number of scheduling periods.

9. The novel multi-agent coordinated operation optimization method for power system according to claim 1 is characterized in that: The steps of solving the two-stage distributed robust cooperative operation optimization model include: Step 5.1: Linearize the fuzzy set and support set; Step 5.2: Based on the linearized fuzzy set and support set, the constraints and objective functions in the two-stage distributed robust collaborative operation optimization model are converted into a compact form; Step 5.3: Based on the strong duality theory, the min-max structure of the objective function and the semi-infinite form of the constraints in the compact form are decimalized, thereby converting the model into a mixed integer linear programming model; Step 5.4: Call the cplex solver to solve the mixed integer linear programming model to obtain the optimal operation plan for each subject in the new power system to participate in the multi-market coordinated dispatch.

10. A novel power system multi-agent coordinated operation optimization device, which runs the novel power system multi-agent coordinated operation optimization method according to any one of claims 1 to 9, characterized in that: The device includes: Acquisition data unit is used to obtain historical forecast data and real data of wind power output, photovoltaic output, electric load, and market price of unit electricity, carbon quota, and green certificate, so as to establish support set and fuzzy set containing multivariate uncertainty prediction error; Construct a model unit to construct a two-stage distributed robust collaborative operation optimization model for a new power system with multiple subjects participating in a multi-market environment; the objective function of the first stage is to maximize the total revenue of the new power system; the objective function of the second stage is based on the optimization objective of the first stage, adding the optimization objective of the total revenue expectation under the worst distribution in the fuzzy set; The first constraint unit is used to formulate the constraint conditions of the first stage according to the internal data of each subject in the new power system and the external environment market data; The second constraint unit is used to take into account the random occurrence of any prediction error in the support set, perform linear projection on the constraint conditions of the first stage, and obtain the constraint conditions of the second stage; The solving and dispatching unit is used to solve the two-stage distributed blue-robust coordinated operation optimization model under the constraints to obtain the optimal operation plan for each subject in the new power system to participate in the coordinated dispatch of multiple markets.

Images