Scheduling method of multi-park aggregation system under uncertain risk

Abstract

The invention provides a multi-energy scheduling method for a multi-park aggregation power grid under an uncertain risk, and belongs to the technical field of multi-energy complementary distributed park management and transaction. Constructing an operation model of each device in the multi-energy complementary park subsystem and a power generation cost model of each device; based on the operation model of each device and the power generation cost model of each device in the multi-energy complementary park subsystem, under the reasonable constraint of the electricity price, a revenue optimization model of the multi-park aggregation system is constructed by taking the revenue maximization of the multi-park aggregation system as a target; under the operation constraint of each energy device, constructing an optimization model of the multi-energy complementary park subsystem by taking the cost minimization of each multi-energy complementary park subsystem as a target; and integrating the income optimization model of the multi-park aggregation system and the optimization model of the multi-energy complementary park subsystem as a joint optimization model, and solving to obtain the electricity price of the multi-park aggregation system and the output plan of the equipment in each multi-energy complementary park subsystem.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of multi-energy complementary distributed park management and transaction, in particular to a scheduling method of a multi-park aggregation system under uncertain risks. BACKGROUND

[0002] As an important secondary energy, electric energy plays a core role in promoting global economic development and energy transformation. In order to improve the utilization efficiency of electric energy and optimize resource allocation, multi-energy complementary parks (Integrated Energy System, IES) integrate various energy forms such as electricity, heat and gas in the region through multi-energy coupling and coordinated scheduling, so as to realize system energy efficiency improvement, economic operation and low-carbon environmental protection goals. Multi-energy complementary parks coordinate distributed renewable energy, energy storage devices and traditional power generation units, become a key component of modern energy networks, and help to alleviate energy supply and demand contradictions and promote sustainable development.

[0003] At present, for the management of multi-energy complementary parks in a single region, centralized or hierarchical optimization control strategies are mainly adopted, and source-grid-load-storage collaborative operation is realized based on predicted data and fixed scheduling models. For example, some schemes establish wind turbine and photovoltaic power generation prediction models, and combine energy storage systems to smooth power fluctuations; or use demand side response mechanisms to adjust loads to match the intermittent output of renewable energy. In addition, existing technologies usually rely on localized control, and realize power distribution of each unit in the system through static optimization algorithms or rule bases, and have limited power interaction with the power grid to maintain supply and demand balance.

[0004] However, the above methods still have significant limitations: the management of multi-energy complementary parks in a single region highly depends on the prediction accuracy of local renewable energy, and the strong randomness of wind turbine and photovoltaic power generation easily leads to scheduling deviation, which reduces the stability of system operation; existing optimization models are mostly based on static scene design, and it is difficult to respond to the severe fluctuations of source and load in real time, resulting in power over-limit or abandoned wind turbine and photovoltaic power when interacting with the grid; current technologies focus on local regional autonomy, lack of cross-regional energy complementary mechanism, and restrict the consumption of large-scale renewable energy.

[0005] Therefore, a scheduling method of a multi-park aggregation system under uncertain risks is needed. SUMMARY

[0006] Therefore, the present application provides a scheduling method of a multi-park aggregation system under uncertain risks, constructs a revenue optimization model of a multi-park aggregation system and an optimization model of a multi-energy complementary park subsystem as a joint optimization model, and solves it to obtain the equipment output plan of the multi-energy complementary park subsystem under reasonable electricity price and balanced power resources and the corresponding cost.

[0007] To this end, the present application provides the following technical solutions: A scheduling method of a multi-park aggregation system under uncertain risks, comprising: Constructing an operation model of each device in the multi-energy complementary park subsystem and a cost model of power generation of each device; Based on the operation model of each device in the multi-energy complementary park subsystem and the cost model of power generation of each device, under the reasonable constraint of electricity price, a revenue optimization model of the multi-park aggregation system is constructed with the maximization of the revenue of the multi-park aggregation system as the target, and under the operation constraint of each energy device, an optimization model of the multi-energy complementary park subsystem is constructed with the minimization of the cost of each multi-energy complementary park subsystem as the target; Integrating the revenue optimization model of the multi-park aggregation system and the optimization model of the multi-energy complementary park subsystem as a joint optimization model; Solving the joint optimization model to obtain the electricity price of the multi-park aggregation system and the output plan of the device in each multi-energy complementary park subsystem and the corresponding cost.

[0008] Further, the cost of power generation of each device comprises: The cost of power generation and the cost of carbon emission.

[0009] Further, the objective function of the revenue optimization model of the multi-park aggregation system is:

[0010] Wherein, represents the revenue of the multi-park aggregation system, T represents the scheduling period, and N represents the total number of multi-energy complementary park subsystems; represents the selling electricity price of the power grid at the current time, represents the buying electricity price of the power grid at the current time, represents the buying electricity price of the multi-park aggregation system from the power grid, represents the selling electricity price of the multi-park aggregation system to the power grid; represents the electricity energy sold by the multi-park aggregation system to the power grid, represents the electricity energy bought by the multi-park aggregation system from the power grid; is the electricity energy purchased by the multi-energy complementary park subsystem from the multi-park aggregation system, is the electricity energy sold by the multi-energy complementary park subsystem to the multi-park aggregation system.

[0011] Further, the objective function of the optimization model of the multi-energy complementary park subsystem is:

[0012] Wherein, represents the total cost of each multi-energy complementary park subsystem; interaction cost between the multi-energy complementary park subsystem and the multi-park aggregation system, operation cost of the gas turbine, operation cost of the energy storage, cost of flexible interruptible load; carbon emission cost, compensation of compensation mechanism.

[0013] Further, the constraint condition of the profit optimization model of the multi-park aggregation system comprises:

[0014]

[0015] wherein, total electric energy of the multi-park aggregation system traded with the power market after aggregating the purchased and sold electric energy of each multi-energy complementary park subsystem, positive value represents purchased electric energy, and negative value represents sold electric energy; purchased electric energy of the multi-park aggregation system in the power market, required electric energy of the multi-energy complementary park subsystem.

[0016] Further, the constraint condition of the optimization model of the multi-energy complementary park subsystem comprises:

[0017] wherein, W total dispatching period, traded electric energy of each multi-energy complementary park subsystem with the multi-park aggregation system, output of the gas turbine, output of the battery, output of the biomass power generation. load demand at the current time; output of the wind turbine, output of the photovoltaic.

[0018] Further, the solving process of the profit optimization model of the multi-park aggregation system and the optimization model of the multi-energy complementary park subsystem comprises: generating a classical scenario set of the wind turbine and photovoltaic output based on historical data observation values of the wind turbine and photovoltaic; evaluating the classical scenario set by the TOPSIS method to obtain a poor scenario of the wind turbine and photovoltaic output; taking the poor scenario of the wind turbine and photovoltaic output as a boundary for solving to obtain the output plan of the equipment in each multi-energy complementary park subsystem.

[0019] Further, the wind turbine and photovoltaic based on historical data observations generate a wind turbine and photovoltaic output classic scene set, comprising: The kernel density estimation method is used to determine the marginal cumulative distribution function of the wind turbine output and the marginal cumulative distribution function of the photovoltaic output; Based on the marginal cumulative distribution function of the wind turbine output and the marginal cumulative distribution function of the photovoltaic output, an edge joint distribution model is constructed; The classic scene set of wind turbine and photovoltaic output is obtained by the edge joint distribution model.

[0020] Further, the compensation incentive comprises:

[0021]

[0022] Wherein, The clean energy and park load correlation coefficient is represented, ranging from [-1, 1]; The covariance of the uncertainty output of the clean energy and the regional load is represented; The current clean energy consumption is represented, The current load is represented; The standard deviation of the clean energy consumption in the current period is represented; The standard deviation of the load demand is represented, The current time grid electricity selling price is represented.

[0023] Advantages and positive effects of the present application: The method establishes a multi-park aggregation system model, and considers the wind turbine, photovoltaic and other renewable energy sources, and considers the biomass power generation model and carbon trading process to reduce carbon emissions. The improved distribution function is used to establish the typical scene of wind turbine photovoltaic output, to generate multiple types of scene of uncertain source and load, and to obtain the wind turbine photovoltaic output under the condition of severe scene and normal scene based on the TOPSIS method. Based on the dynamic electricity price process, a multi-agent game method between the park aggregator and the park is established, a renewable energy-load correlation coefficient is established based on the low-carbon perspective, and the clean energy consumption is further improved. Through the hierarchical robust optimization algorithm, the robust low-carbon optimization scheduling of the renewable energy uncertainty of the multi-park aggregation system is realized. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0025] Figure 1 A scheduling method flowchart of a multi-park aggregation system under uncertain risks in embodiment 1; Figure 2 An energy structure schematic diagram of a multi-energy complementary park subsystem in embodiment 2; Figure 3 A game transaction mechanism diagram of a multi-park aggregation system in embodiment 2; Figure 4 A solving method flowchart in embodiment 2. DETAILED DESCRIPTION

[0026] In order to enable personnel in the technical field to better understand the present application scheme, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the person skilled in the art without creative labor should belong to the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0028] The present application provides a scheduling method of a multi-park aggregation system under uncertain risks, first, a mathematical model of a multi-energy complementary park subsystem is constructed, then a multi-objective optimization model of a multi-park aggregation system is constructed according to the mathematical model of each multi-energy complementary park subsystem, finally, the multi-objective optimization model of the multi-park aggregation system is solved to obtain the electricity price of the multi-park aggregation system and the output strategy of each device in each multi-energy complementary park subsystem.

[0029] Embodiment 1 A scheduling method of a multi-park aggregation system under uncertain risks, as shown in Figure 1 , comprising: Step 1: Constructing a mathematical model of a multi-energy complementary park subsystem, including: operating state equation, operating constraint and climbing constraint of each energy device, cost representation model based on electricity price, and electricity price reasonable constraint.

[0030] Step 2: Based on the mathematical model of the multi-energy complementary park subsystem, under the reasonable constraint of electricity price, a revenue optimization model of the multi-park aggregation system is constructed with the goal of maximizing the revenue of the multi-park aggregation system; under the operation constraint of each energy device, an optimization model of the multi-energy complementary park subsystem is constructed with the goal of minimizing the cost of each multi-energy complementary park subsystem.

[0031] Step 3: Solve the revenue optimization model of the multi-park aggregation system and the optimization model of the multi-energy complementary park subsystem to obtain the electricity price of the multi-park aggregation system and the output plan of the devices in each multi-energy complementary park subsystem and the corresponding cost.

[0032] The electricity price published by the multi-park aggregation system and the grid electricity price are transmitted to each multi-energy complementary park subsystem. Each multi-energy complementary park subsystem iterates according to the optimal goal of the optimization model of the multi-energy complementary park subsystem, transmits the trading electricity and the output plan of each device in the multi-energy complementary park subsystem to the revenue optimization model of the multi-park aggregation system, corrects the dynamic electricity price, re-publishes the electricity price and transmits it to each multi-energy complementary park subsystem again. The multi-park aggregation system participates in the power market transaction according to the published electricity selling price and the grid electricity price, and uses the difference between the two to obtain revenue. Based on this process, iterative iteration is repeated until the optimal solution of the multi-agent is solved.

[0033] Step 3 includes: Step 31, input the historical wind turbine and photovoltaic data, based on the correlation characteristics of renewable energy, use the improved kernel distribution function to obtain the wind turbine photovoltaic correlation output probability distribution under the typical scenario, and after inverse transformation by uniform distribution, the renewable energy scenario can be obtained. Further use the TOPSIS method for evaluation, and obtain the output situation under normal and severe conditions; Step 32, based on the obtained wind turbine photovoltaic severe uncertain scenario and the established multi-agent cost model, a multi-layer robust optimization algorithm is proposed to solve the optimization problem, realize the recovery of self-energy supply while ensuring the minimum cost, and at the same time, other parks can consume renewable energy resources as much as possible to reduce carbon emission cost. Output the final scheduling plan and minimum cost.

[0034] Embodiment 2 A multi-park aggregation system scheduling method under uncertain risk, comprising: S1, construct a mathematical model of a multi-energy complementary park subsystem, including: operating state equation of each energy device, operating constraint and climbing constraint, cost representation model based on electricity price, and reasonable constraint of electricity price Each multi-energy complementary park subsystem has a structure as Figure 2The new energy sources include: wind power, photovoltaic power, combined heat and power, battery and biomass energy.

[0035] 1) The output power of the gas turbine is , and the calculation formula is:

[0036] wherein, represents the electric power of the gas turbine, represents the conversion efficiency of the gas turbine, represents the gas consumption of the gas turbine at the current time.

[0037] The formula of the operation power constraint and the ramp rate constraint of the gas turbine is:

[0038]

[0039] wherein, represents the lower limit of the electric power of the gas turbine, represents the upper limit of the electric power of the gas turbine, represents the lower limit of the ramp rate of the gas turbine, represents the upper limit of the ramp rate of the gas turbine.

[0040] 2) The calculation formula of the remaining electric power of the battery is:

[0041] wherein, represents the remaining electric power of the battery, represents the power at the end of the previous state, represents the charging coefficient of the battery, represents the discharging coefficient of the battery; represents the charging power of the battery, represents the discharging power of the battery.

[0042] The constraint that the battery cannot be charged and discharged at the same time is represented by the formula:

[0043] wherein, represents the charging flag quantity with a value of 0 or 1, represents the discharging flag quantity with a value of 0 or 1, ensuring that the charging and discharging processes cannot occur at the same time.

[0044] The operation constraint and the ramp constraint of the battery are represented by the formula:

[0045]

[0046]

[0047] wherein, represents the lower limit of the charging power of the battery, represents the upper limit of the charging power of the battery; represents the lower limit of the discharging power of the battery, represents the upper limit of the discharging power of the battery; represents the lower limit of the ramp rate when the battery is charging and discharging, represents the upper limit of the ramp rate when the battery is charging and discharging.

[0048] The capacity constraint when the battery is running is expressed by the formula:

[0049] wherein, represents the upper limit of the capacity of the battery, represents the lower limit of the capacity of the battery; represents the capacity of the battery.

[0050] 3) For the utilization process of biomass, through the recycling of natural resources such as straw, through reasonable blending, power generation is carried out, and the mathematical model of biomass power generation is:

[0051]

[0052] wherein, represents the power generation power of biomass, represents the power generation efficiency, represents the fuel conversion efficiency, is the net calorific value, represents the mass of the biomass fuel, represents the current biomass blending power generation time, represents the average calorific value, , represents the combustion coefficient, which is related to the combustion properties.

[0053] The operation constraint of the biomass power generation equipment is expressed by the formula:

[0054]

[0055] S2. Based on the mathematical model of the multi-energy complementary park subsystem, under the constraint of reasonable electricity price, construct the revenue optimization model of the multi-park aggregation system with the goal of maximizing the revenue of the multi-park aggregation system; under the operational constraints of each energy equipment, construct the optimization model of the multi-energy complementary park subsystem with the goal of minimizing the cost of each multi-energy complementary park subsystem.

[0056] S21, such as Figure 3 The game process illustrated is based on a mathematical model of a multi-energy complementary park subsystem. Under reasonable electricity price constraints, a revenue optimization model for the multi-park aggregation system is constructed with the goal of maximizing the revenue of the multi-park aggregation system. The multi-park aggregation system sets the purchase and sale price of electricity. The multi-energy complementary park subsystems sell surplus electricity to the multi-park aggregation system at the sale price, while the less-energy complementary park subsystems purchase the shortfall in electricity from the multi-park aggregation system at the purchase price. Based on the electricity exchange between the multi-energy complementary park subsystems, the multi-park aggregation system trades with the grid and the grid price with the electricity market, profiting from the price difference between the two.

[0057] 1) The objective function is:

[0058] in, T represents the revenue of the multi-park aggregation system, T represents the scheduling cycle, and N represents the total number of multi-energy complementary park subsystems. This indicates the current electricity price sold by the power grid. This indicates the current purchase price of electricity on the power grid. This indicates the electricity purchase price from the grid for the multi-park aggregation system. This indicates the electricity price sold to the grid by the multi-park aggregation system; This refers to the electrical energy sold to the grid by the multi-park aggregation system. This represents the electrical energy purchased from the grid by the multi-park aggregation system. The electricity purchased by the multi-energy complementary park subsystem from the multi-park aggregation system. It refers to the electricity sold by the multi-energy complementary park subsystem to the multi-park aggregation system.

[0059] 2) Constraints on electricity supply and demand balance and reasonable electricity prices, including: The electricity purchased from the power grid by the multi-park aggregation system meets the supply and demand balance constraints with the electricity required by each multi-energy complementary park subsystem.

[0060] The electricity pricing strategy for multi-park aggregation systems is within a reasonable range. The electricity sales price of multi-park aggregation systems must not be lower than the floor price of electricity sales in the electricity market, and the electricity purchase price must not be higher than the ceiling price of electricity purchase in the electricity market. This not only prevents multi-park aggregation systems from monopolizing the market by manipulating electricity prices, but also ensures reasonable profit margins for multi-energy complementary parks.

[0061] The power supply and demand balance constraint and the reasonable price constraint are expressed as:

[0062]

[0063] wherein, represents the total power traded with the power market after the multi-park aggregation system aggregates the purchased and sold power of each multi-energy complementary park subsystem, and is a positive number indicating power purchase and a negative number indicating power sale. represents the power purchased by the multi-park aggregation system in the power market, represents the power required by the multi-energy complementary park subsystem.

[0064] S22, under the operation constraints of each energy equipment, an optimization model of the multi-energy complementary park subsystem is constructed with the minimum cost of each multi-energy complementary park subsystem as the target.

[0065] 1) Objective function:

[0066]

[0067]

[0068]

[0069]

[0070]

[0071] wherein, represents the total cost of each multi-energy complementary park subsystem; represents the interaction cost of the multi-energy complementary park subsystem and the multi-park aggregation system, represents the operation cost of the gas turbine, represents the operation cost of the energy storage, the cost of flexible interruptible load, represents the transfer cost coefficient; represents the energy storage cost coefficient; , , represents the cost coefficient of the gas turbine, represents the carbon emission cost, represents the carbon emission price; represents the carbon emission coefficient of the gas turbine power generation, represents the carbon emission coefficient of the biomass power generation; represents the compensation of the compensation mechanism.

[0072] 2) The operation constraints of each device should be met, and the supply and demand balance of each multi-energy complementary park subsystem should be met:

[0073] wherein, W denotes the total scheduling period, denotes the traded electric energy of each multi-energy complementary park subsystem and the multi-park aggregation system, denotes the output of the gas turbine, denotes the output of the battery, denotes the output of the biomass power generation. denotes the load demand at the current time. denotes the output of the wind turbine, denotes the output of the photovoltaic.

[0074] S23, define the correlation degree of clean energy and park load to measure the use of clean energy in multi-energy complementary park, and build a compensation mechanism, define the total cost of encouraging to join each multi-energy complementary park subsystem, to improve clean energy consumption.

[0075] In this embodiment, the clean energy includes: wind turbine and photovoltaic.

[0076] 1) The correlation degree of clean energy and park load is represented by the formula:

[0077] wherein, denotes the correlation coefficient of clean energy and park load, ranging from [-1, 1]; denotes the covariance of the uncertain output of clean energy and the regional load; denotes the current clean energy consumption, denotes the current load; denotes the standard deviation of clean energy consumption in the current period; denotes the standard deviation of load demand.

[0078] The covariance of the uncertain output of clean energy and the regional load is calculated by the formula:

[0079] The covariance reflects the synchronicity of the output of clean energy and the change of load demand. Positive value means that both increase and decrease, which is conducive to local consumption, and negative value means that the other needs to rely on external adjustment. and denote the mean of clean energy output and load.

[0080] The formula of the standard deviation of clean energy use and the standard deviation of load demand is:

[0081]

[0082] The standard deviation quantifies the uncertainty of the renewable energy output and load. High volatility requires more flexible resources such as energy storage, multi-energy complementary park transaction balancing, and system safety, and serves as the basis for the design of compensation mechanisms. represents the current time period.

[0083] 2) The compensation incentive calculation formula is:

[0084] wherein, refers to the current time grid electricity selling price.

[0085] S3, the income optimization model of the multi-park aggregation system and the optimization model of the multi-energy complementary park subsystem are solved, and the electricity price of the multi-park aggregation system and the output plan of the equipment in each multi-energy complementary park subsystem and the corresponding cost are obtained.

[0086] The solving target includes: The upper layer is the multi-park aggregation system, which aims to maximize the benefit under the condition of meeting the supply and demand balance, and issues a dynamic electricity price signal; The lower layer is each multi-energy complementary park subsystem, which aims to minimize its own economic cost and carbon cost based on the electricity price signal, and outputs the equipment output plan in the multi-energy complementary park subsystem.

[0087] The solving process, as shown in Figure 4 , includes: S31, the marginal cumulative distribution function of the wind turbine and photovoltaic output is constructed by the Sklar theorem, the joint probability distribution model of the wind turbine and photovoltaic output is established by using the Copula distribution function, and the typical scenario set considering the random correlation of the wind turbine and photovoltaic is generated by using the kernel density estimation method.

[0088] The time step 24 and the number of scenarios are initialized to 500, the wind turbine historical data observation value of each hour is stored in the array , the photovoltaic historical data observation value is stored in the array , and the formula is represented as:

[0089] wherein, the subscript represents time, represents the number of historical observation samples.

[0090] 1) According to Sklar theorem, preferably, F-type Copula function which can take into account the positive and negative correlation characteristics is adopted, and the edge cumulative distribution function of wind turbine and photovoltaic output is determined by kernel density estimation method, and the specific process is as follows: Based on the historical observation values of wind turbine and photovoltaic, the edge probability density function of wind turbine and photovoltaic is obtained respectively:

[0091]

[0092] Wherein, The edge probability density function of wind turbine is represented by fWT(xWT), The edge probability density function of photovoltaic is represented by fPV(xPV) ; The sampling window width is represented by h, The kernel density function is represented by K(x).

[0093] Then the kernel density is integrated to obtain the edge cumulative distribution function of wind turbine and photovoltaic output And :

[0094]

[0095] Wherein, The edge cumulative distribution function of wind turbine output is represented by FWT(xWT), The edge cumulative distribution function of photovoltaic output is represented by FPV(xPV).

[0096] 2) After converting the edge cumulative distribution of wind turbine and photovoltaic into wind turbine and photovoltaic uniform distribution u and v, the two are fitted to obtain the edge joint distribution model:

[0097] Wherein, the parameter Controls the degree of dependence between variables; The edge joint distribution model of wind turbine and photovoltaic is represented by C(u,v), The wind turbine uniform distribution is represented by u, The photovoltaic uniform distribution is represented by v.

[0098] 3) After obtaining the edge joint distribution model corresponding to each time, 500 groups of numerical pairs are sampled from it, and the formula is as follows:

[0099] Wherein, The random sampling points of wind turbine and photovoltaic are represented by (xWT, xPV).

[0100] Since the above sampling distribution is located on the uniform distribution, in order to obtain the actual fan photovoltaic output value, the corresponding inverse transformation needs to be done, further, the spline interpolation method is used to inversely transform it into the actual photovoltaic and fan output, which is expressed as:

[0101]

[0102] Wherein, represents the actual fan output, represents the actual photovoltaic output.

[0103] Obtain the corresponding 500 fan and photovoltaic scene output results at each time, splice the fan and photovoltaic scenes into a matrix, and each row corresponds to a whole 24-hour fan photovoltaic time sequence:

[0104] Use Manhattan distance to reduce scenes, calculate the representative probability of the cluster center, and then do weighted average on the output of the fan and photovoltaic at each time period of each cluster center, and finally obtain the predicted photovoltaic and fan output scene of 24 hours a day.

[0105] 4) Random sampling is performed to obtain the typical fan photovoltaic output scene required in this paper. At the same time, the scheduling of new energy output in adverse scenes is considered, and the TOPSIS method is used for scene evaluation to obtain the fan photovoltaic output in adverse scenes and ordinary scenes.

[0106] The principle of using the TOPSIS method is based on the Euclidean distance between the ideal scene solution and the non-ideal scene solution of the fan photovoltaic, and the relative closeness value of the fan photovoltaic output value is calculated. If the value is larger, it proves that it is closer to the ideal output, and thus the evaluation of the corresponding adverse scene and ideal scene is obtained.

[0107] For the obtained fan photovoltaic output data to be evaluated, there are m evaluation indexes for a day, then the fan photovoltaic output evaluation matrix general formula is constructed As follows:

[0108] Wherein, represents the fan or photovoltaic output evaluation element, which is normalized to form a standardized decision matrix form , wherein As follows:

[0109] Wherein, and respectively represent the maximum element and the minimum element; based on the above standardized matrix and , respectively, represent the distance between the output data and the ideal scenario solution and the poor scenario solution, which can be calculated by the following formula:

[0110]

[0111] wherein, and represent the ideal scenario solution and the poor scenario solution, respectively, which are represented as:

[0112]

[0113] and then calculate the relative closeness value as follows:

[0114] When the value of the closeness value is closer to 1, it indicates that the output scenario at this time is closer to the more ideal output scenario, and if the value is closer to 0, it indicates that it is closer to the poor scenario. By calculating the TOPSIS scores of different groups of output, the TOPSIS scenario scores reflect the closeness of the wind turbine and photovoltaic output to the ideal value. The higher the TOPSIS score is, the more ideal the scenario is, and the lower the TOPSIS score is, the more poor the scenario is.

[0115] S32, based on the wind turbine photovoltaic output under different probabilities, a target optimization solving method is established; the compact form of the target optimization solving method is:

[0116] wherein, represents the output plan of the equipment in the multi-energy complementary park, and the dynamic adjustment of the electricity price and maximizes the benefit, while considering the response behavior of the lower multi-energy complementary park to the electricity price. Each multi-energy complementary park subsystem optimizes its own equipment output and transaction strategy based on the electricity price published by the multi-park aggregation system, and minimizes the total cost . The electricity price decision of the multi-park aggregation system affects the interactive power of the multi-energy complementary park, and the game behavior of the multi-energy complementary park is fed back to the benefit model of the multi-park aggregation system, forming a dynamic equilibrium.

[0117] The uncertainty problem of the renewable energy output of each sub multi-energy complementary park is divided into a main problem and a sub problem. The main problem is the operation optimization problem of the multi-energy complementary park subsystem, and the sub problem is to solve the output and electricity price of the multi-energy complementary park under the wind and light uncertainty poor scenario. The main problem obtains the optimal solution when the source and load output is known, provides the lower boundary for the model, and transmits the result to the following sub problem:

[0118] wherein: represents the optimization objective function, represents group inequality variable coefficient vector constraints, represents the decision variable matrix. Through the iteration of the master problem and the sub-problem, the robust optimal solution is gradually approached, ensuring the feasibility of the scheme under all possible adverse scenarios.

[0119] The sub-problem finds the adverse scenario in the uncertainty set, provides an upper bound for the model, and passes the corresponding scenario output to the master function. Then:

[0120] wherein, represents the decision variable of the master problem, i.e., the output plan of each device and the trading electricity, etc., represents the uncertainty parameter, i.e., the wind and photovoltaic, represents the objective function coefficient vector, i.e., the economic cost and carbon emission cost.

[0121] and respectively represent the coupling matrix of the decision variable and the uncertainty parameter in the master problem, which maps the operating constraints, and respectively represent the boundary constraint matrix of the uncertainty parameter in the sub-problem, and equation constraint matrix. , , respectively represent the Lagrange dual variables, which correspond to the three constraints of the original problem respectively. The left side of the above formula corresponds to the constant coefficient matrix, which has been listed in the model establishment part, and the right side corresponds to the constraint coefficient matrix after the random optimization problem.

[0122] In order to avoid the above random optimization problem cannot be solved, the corresponding constraints are increased after convex relaxation as follows:

[0123] wherein, represents the binary variable (0-1 variable), which represents whether it is in a certain extreme scenario at a certain time period, represents a large enough constant used to relax the integer variable constraint, represents the split dual variable, which represents the compensation cost constraint under a certain time period extreme scenario.

[0124] When the iteration calculation result converges, output the electricity price at this time and the energy device scheduling plan of each multi-energy complementary park and the optimal cost.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A scheduling method for a multi-campus aggregation system under uncertain risks, characterized in that, include: Construct operation models for each device in the multi-energy complementary park subsystem and cost models for the power generation of each device. Based on the operation models of each device in the multi-energy complementary park subsystem and the cost models of each device's power generation, under the constraint of reasonable electricity prices, a revenue optimization model for the multi-park aggregation system is constructed with the goal of maximizing the revenue of the multi-park aggregation system; under the operation constraints of each energy device, an optimization model for the multi-energy complementary park subsystem is constructed with the goal of minimizing the cost of each multi-energy complementary park subsystem. The revenue optimization model of the multi-park aggregation system and the optimization model of the multi-energy complementary park subsystem are integrated into a joint optimization model; Solving the joint optimization model yields the electricity price of the multi-park aggregation system and the output plan and corresponding cost of equipment in each multi-energy complementary park subsystem.

2. The method according to claim 1, characterized in that, The cost of generating electricity from each of the aforementioned devices includes: Electricity generation costs and carbon emission costs.

3. The method according to claim 1, characterized in that, The objective function of the revenue optimization model for the multi-park aggregation system is: in, The value represents the revenue of the multi-park aggregation system, T represents the scheduling cycle, and N represents the total number of multi-energy complementary park subsystems. This indicates the current electricity price sold by the power grid. This indicates the current purchase price of electricity on the power grid. This indicates the electricity purchase price from the grid for the multi-park aggregation system. This indicates the electricity price sold to the grid by the multi-park aggregation system; This refers to the electrical energy sold to the grid by the multi-park aggregation system. This represents the electrical energy purchased from the grid by the multi-park aggregation system. The electricity purchased by the multi-energy complementary park subsystem from the multi-park aggregation system. It refers to the electricity sold by the multi-energy complementary park subsystem to the multi-park aggregation system.

4. The method according to claim 1, characterized in that, The objective function of the optimization model for each multi-energy complementary park subsystem is: in, This represents the total cost of each multi-energy complementary park subsystem; This represents the interaction cost between the multi-energy complementary park subsystem and the multi-park aggregation system. This indicates the operating cost of a gas turbine. This indicates the operating cost of energy storage. The cost of flexible interruption loads; Indicates carbon emission costs, This refers to compensation under a compensation mechanism.

5. The method according to claim 1, characterized in that, The constraints of the revenue optimization model for the multi-park aggregation system include: in, This represents the total electricity traded with the electricity market after the multi-park aggregation system aggregates the electricity purchases and sales of each multi-energy complementary park subsystem. A positive number indicates electricity purchase, while a negative number indicates electricity sales. This indicates the amount of electricity purchased by the multi-park aggregation system in the electricity market. This indicates the electricity required by the multi-energy complementary park subsystem.

6. The method according to claim 1, characterized in that, The constraints of the optimization model for the multi-energy complementary park subsystem include: in, W Indicates the total scheduling cycle. This represents the traded electrical energy between the multi-energy complementary park subsystems and the multi-park aggregation system. This indicates the output power of the gas turbine. Indicates the output power of the storage battery. This indicates the power output of biomass power generation. This indicates the load demand at the current moment; Indicates the output of the fan. This indicates the output of photovoltaic power.

7. The method according to claim 1, characterized in that, The solution process for the revenue optimization model of the multi-park aggregation system and the optimization model of the multi-energy complementary park subsystem includes: A set of classic scenarios for wind turbine and solar power output is generated based on historical data observations of wind turbines and solar power. The TOPSIS method was used to evaluate the classic scenario set to obtain the severe scenarios for wind turbine and photovoltaic power output. Using the harsh scenarios of wind turbine and photovoltaic power output as the limit for the solution, the power output plan of the equipment in each multi-energy complementary park subsystem is obtained.

8. The method according to claim 7, characterized in that, The set of classic scenarios for wind turbine and photovoltaic power output, generated based on historical data observations of wind turbines and photovoltaics, includes: The edge cumulative distribution function of wind turbine output and the edge cumulative distribution function of photovoltaic output are determined by kernel density estimation method; An edge joint distribution model is constructed based on the edge cumulative distribution function of the wind turbine output and the edge cumulative distribution function of the photovoltaic output; The classic scenario set for obtaining wind turbine and photovoltaic power output is randomly selected using the aforementioned edge joint distribution model.

9. The method according to claim 7, characterized in that, The compensation incentives include: in, This represents the correlation coefficient between clean energy and park load, ranging from [-1, 1]. This represents the covariance between the uncertain output of clean energy and the regional load. This indicates the current amount of clean energy consumed. Indicates the current load; This represents the standard deviation of clean energy consumption during the current period. The standard deviation of load demand This refers to the current electricity price on the power grid.

Images