Random optimization method for energy storage system in virtual power plant participating in electricity market transaction
By using Monte Carlo simulation and particle swarm algorithms in virtual power plants to optimize the configuration of energy storage systems, the problem of high and low cost of energy storage systems is solved, and the economic benefits of virtual power plants are improved and the power market transaction stability is achieved.
Patent Information
- Application Number
- CN202410971787.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-07-10
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-12
AI Technical Summary
The high cost, low efficiency and short life cycle of energy storage systems in virtual power plants have certain limitations in their application in power systems, and distributed power generation is random and volatile, affecting the stability of the power grid operation.
Monte Carlo simulation method is used to generate the uncertain scenario set of virtual power plants in the power market transaction process, and a two-layer optimization model is established based on the access location, capacity and battery type of the energy storage system, combining the particle swarm algorithm and CPLEX solver. Optimize the configuration of the energy storage system to minimize the investment and operation and maintenance costs of the energy storage system.
Effectively determine the location, capacity and battery type of the energy storage system, reduce the operating costs of virtual power plants, increase the economic benefits in power market transactions, and stabilize the output of distributed power sources, and enhance the comprehensive application of energy storage systems in virtual power plants.
Smart Images

Figure CN118940885A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of electric power information technology, and in particular to a random optimization method for an energy storage system in a virtual power plant participating in electric power market transactions. Background Art
[0002] With the large-scale access of distributed power sources such as photovoltaics and wind turbines to the distribution network, non-fossil energy power generation has become the development trend of the future power grid. It is estimated that by 2030, clean energy power generation will account for 49.2% of the country's total power generation. However, distributed power generation has great randomness and volatility, which brings risks to the operation and control of the power grid. As a new type of comprehensive energy service provider, virtual power plants can effectively manage distributed power generation and grid-connected power generation. Virtual power plants are a new type of coordinated operation and management technology for distributed power sources, which provides a new method to solve the above problems. Its basic concept is to participate in power grid dispatching and power market transactions as a controllable power generator by aggregating distributed energy sources such as distributed power sources, controllable loads and energy storage systems.
[0003] In order to solve the problem of intermittent output of distributed power sources and ensure the stability of power supply of virtual power plants, the capacity configuration of controllable units and energy storage systems must be fully considered during the planning and construction stage of virtual power plants. However, the current energy storage systems have problems such as high cost, low efficiency, and short life cycle, which leads to certain limitations in their application in power systems. In order to give full play to the role of energy storage systems in virtual power plants and improve the operating income of virtual power plants, it is necessary to optimize the configuration of energy storage systems in virtual power plants. Summary of the invention
[0004] In view of the deficiencies in the prior art, the present invention provides a random optimization method for an energy storage system in a virtual power plant participating in electricity market transactions, which solves the technical problem of how to optimize the configuration of the energy storage system in the virtual power plant.
[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a random optimization method for an energy storage system in a virtual power plant participating in power market transactions, the method comprising the following processes:
[0006] Monte Carlo simulation method is used to generate uncertainty scenario sets of distributed power generation output, load forecast and operation status when virtual power plants participate in power market transactions;
[0007] Under uncertain scenarios, an upper-level optimization model is established to optimize the planning and construction of energy storage systems based on the access location, capacity, and battery type of the energy storage system.
[0008] Establish a lower-level optimization model to optimize the hourly output of distributed power sources within the virtual power plant to minimize operating costs;
[0009] Based on the upper optimization model and the lower optimization model, a two-level planning model is established with the minimum construction and operation and maintenance cost of the energy storage system as the upper objective function and the minimum daily operation cost of the virtual power plant as the lower objective function.
[0010] The particle swarm algorithm is combined with the CPLEX solver to solve the two-layer optimization model and obtain the optimal configuration results of the energy storage system in the virtual power plant.
[0011] Furthermore, the generation of uncertainty scenario sets is specifically: using the Monte Carlo simulation method to generate multiple combination scenario sets of the virtual power plant in the process of participating in electricity market transactions, including photovoltaic output scenario sets, load forecast error scenario sets and gas turbine operating status scenario sets, and calculating the expected value after sampling, thereby converting the random optimization problem into a deterministic problem for processing.
[0012] Furthermore, the photovoltaic output scenario set is shown as follows:
[0013]
[0014] In the above formula, is the actual output of photovoltaic power; ΔP PV is the photovoltaic output prediction error; P PV Contribute to photovoltaic forecasting;
[0015] The load forecast error scenario set is as follows:
[0016]
[0017] in,
[0018] σ L =P L / 100
[0019]
[0020] In the above formula, ΔP L is the error value of load prediction; μ L , σ L are the statistical mean and standard deviation of load demand respectively; P L is the predicted value of load; is the actual value of load;
[0021] The gas turbine operating status scenario set is shown in the following formula:
[0022]
[0023] In the above formula: n i is the number of the normal gas turbine; m is the number of the faulty gas turbine; q ni is the normal gas turbine shutdown probability; qm is the shutdown probability of the faulty gas turbine.
[0024] Furthermore, the optimization objectives of the upper-level optimization model include the equivalent annual investment cost, operation and maintenance cost, and virtual power plant operating cost of the energy storage system. The battery types of the energy storage system are lithium-ion batteries and lead-carbon batteries.
[0025] Furthermore, the annual investment and construction cost of the energy storage system coupled with its installed capacity can be calculated by the following formula:
[0026]
[0027] In the above formula, A is the initial investment cost coefficient of equal annual value; C ES is the cost per unit capacity of the energy storage system; C bat is the installed capacity of the energy storage system; d is the actual interest rate; N r The service life of the energy storage system;
[0028] The operation and maintenance cost of the energy storage system is mainly related to its rated power, and the rated power of the energy storage system is linearly related to the installed capacity, that is:
[0029]
[0030] In the above formula, K bat C is the ratio coefficient between the rated power of the energy storage system and its installed capacity; op is the annual operation and maintenance cost per unit capacity of the energy storage system;
[0031] The operating cost of a virtual power plant consists of an investment cost constraint and an access location constraint. The investment cost constraint is:
[0032] C ES C bat ≤C inv.max
[0033] In the above formula, C inv,max is the maximum investment cost of the energy storage system;
[0034] The access location constraints are:
[0035] ES location ∈Φ
[0036] In the above formula, ES location is the node number of the energy storage system access node; Φ is the set of nodes that can be accessed by the energy storage system.
[0037] Furthermore, the lower-level optimization model includes system power balance constraints, node power flow constraints, node voltage constraints, gas turbine output constraints, photovoltaic output constraints, and energy storage system constraints.
[0038] Furthermore, the system power balance constraint is as follows:
[0039]
[0040] In the above formula, K is the total number of system nodes; N is the total number of scenes; is the load demand of node i at time t; are the system power loss and abandoned light amount at time t in the scenario ξ respectively; is the power generation of the i-th micro gas turbine at time t in scenario ξ; are the charging power and discharging power of the energy storage system at time t respectively; is the charging power of the energy storage system at time t; is the discharge power of the energy storage system at time t; is the amount of electricity purchased by the virtual power plant from the distribution network at time t in the scenario ξ; is the electricity sold by the virtual power plant from the distribution network at time t in the scenario ξ;
[0041] The node power flow constraint is as follows:
[0042]
[0043] In the above formula, g ij , b ij are the conductance and susceptance of branches i and j respectively; θ ij,t is the phase difference between the voltage of node i and the voltage of node j at time t; N(i) is the set of all nodes connected to node i; U i,t is the voltage amplitude of node i at time t; are the active and reactive outputs of all distributed generation sources at node i at time t respectively; is the reactive load of node i at time t; U j,t is the voltage amplitude of node j at time t; is the active load of node i;
[0044] The node voltage constraint is as follows:
[0045] U i,min ≤U i,t ≤U i,max
[0046] In the above formula, U i,min , U i,max are the lower and upper limits of the voltage amplitude of node i at time t respectively;
[0047] The gas turbine output constraint is as follows:
[0048]
[0049]
[0050] In the above formula, P mti,min , P mti,max , Q mti,min , Q mti,max are the lower and upper limits of the active and reactive output amplitudes of node i at time t, respectively; is the active power generated by the i-th micro gas turbine at time t in scenario ξ; is the reactive power generated by the i-th micro gas turbine at time t in scenario ξ;
[0051] The photovoltaic output constraint is as follows:
[0052]
[0053] In the above formula, is the maximum photovoltaic output at time t; is the amount of system power abandoned at time t in the scenario ξ; is the photovoltaic output at time t in scenario ξ;
[0054] The energy storage system constraints are as follows:
[0055]
[0056]
[0057]
[0058]
[0059] In the above formula, is the maximum charging and discharging power of energy storage; is the SOC value of the energy storage system at time t; η es The charging and discharging efficiency of the energy storage system; are the minimum and maximum SOC values of the energy storage system respectively; C bat is the installed capacity of the energy storage system; Δt is the time period t.
[0060] Furthermore, the two-level programming model is expressed as:
[0061]
[0062]
[0063] The constraints of the bilevel programming model are:
[0064]
[0065]
[0066] In the above formula, g ij , b ij are the conductance and susceptance of branches i and j respectively; P ij,t is the active power of node i and node j at time t; Q ij,t is the reactive power of node i and node j at time t; u i,t 、u j,t They are in and are the voltage amplitude of node i at time t and the voltage amplitude of node j at time t respectively; R ij,t is the cosine product of the phase difference between the voltage amplitudes at node i and node j at time t; T ij,t is the sine product of the phase difference of the voltage amplitudes at node i and node j at time t; U i,min , U i,max are the lower and upper limits of the voltage amplitude at node i at time t respectively.
[0067] By means of the above technical solution, the present invention provides a random optimization method for an energy storage system in a virtual power plant participating in power market transactions, which has at least the following beneficial effects:
[0068] 1. The random optimization method of the energy storage system in the virtual power plant proposed in the present invention can determine the location, capacity and battery type of the energy storage. First, the uncertainty of the virtual power plant in the electricity market transaction process is analyzed, and the Monte Carlo sampling and scenario reduction techniques are used to model the uncertainty for the volatility of photovoltaic output and the randomness of load demand. Secondly, under the system constraints of the virtual power plant, a two-layer optimization model is established with the goal of minimizing the investment cost of the energy storage system, and the particle swarm algorithm and second-order cone programming theory are used to solve it. The improved IEEE33 node system is used to verify that the optimization method is feasible and effectively improves the economic benefits of virtual power plants in the electricity market.
[0069] 2. The present invention can effectively solve the problem of intermittent output of distributed power sources during the planning and construction stage of the virtual power plant by optimizing the configuration of the energy storage system in the virtual power plant, ensure the stability of the power supply of the virtual power plant, and reduce the impact of the energy storage system on the virtual power plant due to problems such as high cost, low efficiency, and short life cycle, thereby improving the comprehensiveness of the application of the energy storage system in the power system, thereby giving full play to the role of the energy storage system in the virtual power plant and improving the operating income of the virtual power plant. BRIEF DESCRIPTION OF THE DRAWINGS
[0070] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0071] Figure 1 A flowchart of the present invention combining the particle swarm algorithm with CPLEX for solving the problem;
[0072] Figure 2 It is the optimal output curve diagram of the energy storage system of the present invention;
[0073] Figure 3 It is the electricity trading curve and gas turbine output curve of the virtual power plant of the present invention;
[0074] Figure 4 This is the PSO convergence characteristic curve diagram of the present invention. DETAILED DESCRIPTION
[0075] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.
[0076] Please refer to Figure 1-Figure 4 This embodiment proposes a random optimization method for energy storage systems in virtual power plants participating in electricity market transactions. First, the uncertainty of virtual power plants in the electricity market transaction process is analyzed, and the Monte Carlo sampling and scenario reduction techniques are used to model the uncertainty of the volatility of photovoltaic output and the randomness of load demand. Secondly, under the constraints of the virtual power plant system, a two-layer optimization model is established with the goal of minimizing the investment cost of the energy storage system, and the particle swarm algorithm and second-order cone programming theory are used to solve it. Finally, the modified IEEE33 node system is used as a virtual power plant to simulate and verify the above-mentioned two-layer optimization model and algorithm. The method comprises the following steps:
[0077] S1. Use the Monte Carlo simulation method to generate uncertainty scenario sets of the output, load forecast and operating status of distributed power sources of virtual power plants in the process of participating in electricity market transactions. The uncertainty scenario sets include photovoltaic output scenario sets, load forecast error scenario sets and gas turbine operating status scenario sets. In this embodiment, the Monte Carlo simulation method is used to generate multiple combined scenario sets, and the expected value is calculated after sampling, thereby converting the random optimization problem into a deterministic problem for processing. Consider the photovoltaic output, load forecast error and gas turbine operating status scenario sets respectively, where the photovoltaic output scenario set, load forecast error scenario set and gas turbine operating status scenario set are: w = {w i}i=1,2,......,nw s={s j}j=1,2......,n s ,q=q{q k}k=1,2,.....,n q , where n w ,n s ,n q is the number of corresponding scene sets, then the total number of scenes is the number of each scene n w ,n s ,n q and its probability, represented by a discrete distribution function.
[0078] The basic parameters of the distributed power supply are shown in Table 1, and the technical parameters of the gas turbine are shown in Table 2.
[0079] Table 1 Distributed power supply parameters
[0080]
[0081] Table 2 Technical parameters of gas turbine
[0082]
[0083] As the main power generation unit in the virtual power plant, photovoltaic output is not only related to its own working characteristics, but also closely related to the geographical environment, light intensity and other climatic conditions at the time. Since the solar light intensity follows the Beta distribution, the distribution parameters α and β are calculated according to the average light intensity of each cycle, and then Monte Carlo sampling is used to randomly generate photovoltaic output. The distribution function of its probability density is as follows:
[0084]
[0085] In the above formula, P PV Contribute to photovoltaic prediction; PV,max is the installed capacity of photovoltaic power generation; α and β are the shape parameters of the Beta distribution. Since there are errors in the actual production process of photovoltaic power generation, the fluctuating data are randomly generated as the error value of photovoltaic output prediction. The actual photovoltaic output is the superposition of the photovoltaic predicted output and the photovoltaic output prediction error, so the photovoltaic output scenario set is shown as follows:
[0086]
[0087] In the above formula, is the actual output of photovoltaic power; ΔP PV is the photovoltaic output prediction error; P PV Contribute to photovoltaic forecasting.
[0088] Since load forecasting also has a certain error, assuming that the load forecasting error follows a Gaussian cumulative distribution, the load forecasting error scenario set is as follows:
[0089]
[0090] in,
[0091] σ L =P L / 100
[0092]
[0093] In the above formula, ΔP L is the error value of load prediction; μ L , σ L are the statistical mean and standard deviation of load demand respectively; P L is the predicted value of load; is the actual value of the load.
[0094] Considering that the probability of multiple gas turbines failing at the same time in the virtual power plant system is very small, only a single gas turbine failure shutdown scenario set is established. A two-state continuous Markov model is used to simulate the start and stop state of the gas turbine in the scheduling cycle, and an N-1 failure scenario set of a gas turbine is established, that is, a gas turbine operating state scenario set, and its failure probability is determined by the following formula:
[0095]
[0096] In the above formula, n i is the number of a normal gas turbine; m is the number of a faulty gas turbine; is the normal gas turbine shutdown probability; q m is the shutdown probability of the faulty gas turbine.
[0097] In this embodiment, the shape parameters of the Beta distribution calculated based on the light intensity are α=0.45 and β=9.18. The number of photovoltaic output scenes generated n w =4, the probability of each scenario is 0.25; the number of load forecast scenarios n s =5, the probability of each scenario is 0.2; the number of gas turbine scenarios n q
[0098] =3, the probabilities of each scene are 0.5, 0.4, and 0.1 respectively, so the total number of scenes is 60.
[0099] S2. Under the uncertainty scenario set, an upper-level optimization model is established to optimize the planning and construction of energy storage systems based on the access location, capacity, and battery type of the energy storage system. Due to the different technical indicators such as the access location and capacity of the energy storage system, the corresponding costs and benefits are also different. Therefore, it is necessary to optimize the planning and construction of the energy storage system. The optimization objectives of the upper-level model include the annual investment and construction cost of the energy storage system, the loss cost, and the operating cost of the virtual power plant.
[0100] For the sake of simplicity, the battery types available for energy storage systems are lithium-ion batteries and lead-carbon batteries. Their economic and technical parameters are shown in Table 3.
[0101] Table 3 Economic and technical parameters
[0102]
[0103] The objective function of the upper optimization model consists of three parts, as shown in the following formula:
[0104] minC={(C inv +C opm )+C dop} (5)
[0105] In the above formula, C inv C is the investment and construction cost of the energy storage system; opm is the operation and maintenance cost of the energy storage system; C dop is the daily operating cost of the virtual power plant, calculated by the underlying optimization model.
[0106] The annual investment and construction cost of the energy storage system coupled with its installed capacity can be calculated by the following formula:
[0107]
[0108] In the above formula, A is the initial investment cost coefficient of equal annual value; C ES is the cost per unit capacity of the energy storage system; C bat is the installed capacity of the energy storage system; d is the actual interest rate; N r It is the service life of the energy storage system.
[0109] The loss cost of the energy storage system mainly considers the operation and maintenance cost, and the operation and maintenance cost of the energy storage system is mainly related to its rated power, and the rated power of the energy storage system is linearly related to the installed capacity, that is:
[0110]
[0111] In the above formula, K bat C is the ratio coefficient between the rated power of the energy storage system and its installed capacity; opIt is the annual operation and maintenance cost per unit capacity of the energy storage system.
[0112] The operating cost of a virtual power plant consists of an investment cost constraint and an access location constraint. The investment cost constraint is:
[0113] C ES C bat ≤C inv.max (9)
[0114] In the above formula, C inv,max It is the maximum investment cost of the energy storage system.
[0115] The access location constraints are:
[0116] ES location ∈Φ(10)
[0117] In the above formula, ES location is the node number of the energy storage system access node; Φ is the set of nodes that can be accessed by the energy storage system.
[0118] S3. Establish a lower-level optimization model that optimizes the hourly output of distributed power sources within the virtual power plant to achieve the minimum operating cost. The lower-level optimization model includes system power balance constraints, node flow constraints, node voltage constraints, gas turbine output constraints, photovoltaic output constraints, and energy storage system constraints.
[0119] In this embodiment, a second-order cone programming model for optimal scheduling of virtual power plants, i.e., the lower optimization model, is established to optimize the hourly output of distributed power sources within the virtual power plant, thereby minimizing the operating cost. The clearing price of the power market is shown in Table 4. The electricity sales and purchase prices of the unbalanced power of the virtual power plant are reduced and increased by 20% on the basis of the clearing price, respectively.
[0120] Table 4 Market clearing electricity prices
[0121]
[0122] The objective function of the layer optimization model is as follows:
[0123]
[0124] in,
[0125]
[0126]
[0127] In the above formula, β(ξ) is the probability of scenario ξ occurring; Cost of electricity generation for gas turbines; is the system loss cost; The cost of abandoned light; is the energy storage loss cost; is the electricity purchase cost and electricity sales revenue; R ξ To balance the power cost. T is the number of time intervals in the scheduling cycle; M is the total number of gas turbines; a i and b i are the cost coefficients of the i-th micro gas turbine; is the power generation of the i-th micro gas turbine at time t in scenario ξ; η loss is the unit power loss cost; are the system power loss and abandoned light amount at time t in the case of scenario ξ; C grid,t is the clearing electricity price of the distribution network at time t; N lcc The number of charge and discharge times during the energy storage life cycle; are the charging power and discharging power of the energy storage system at time t respectively; are the power purchase and sales of the virtual power plant from the distribution network at time t in the scenario ξ; C rt,t is the real-time electricity price of the distribution network at time t; is the photovoltaic output at time t in scenario ξ; The error value of load forecast in the case of scenario ξ; is the error value of the generated power at time t in the scenario ξ.
[0128] The system power balance constraint is as follows:
[0129]
[0130] In the above formula, K is the total number of system nodes; N is the total number of scenes; is the load demand of node i at time t; are the system power loss and abandoned light amount at time t in the scenario ξ respectively; is the power generation of the i-th micro gas turbine at time t in scenario ξ; are the charging power and discharging power of the energy storage system at time t respectively; is the charging power of the energy storage system at time t; is the discharge power of the energy storage system at time t; is the amount of electricity purchased by the virtual power plant from the distribution network at time t in the scenario ξ; is the amount of electricity sold by the virtual power plant from the distribution network at time t in scenario ξ.
[0131] The node power flow constraint is as follows:
[0132]
[0133] In the above formula, gij , b ij are the conductance and susceptance of branches i and j respectively; θ ij,t is the phase difference between the voltage of node i and the voltage of node j at time t; N(i) is the set of all nodes connected to node i; U i,t is the voltage amplitude of node i at time t; are the active and reactive outputs of all distributed generation sources at node i at time t respectively; is the reactive load of node i at time t; U j,t is the voltage amplitude of node j at time t; is the active load of node i.
[0134] The node voltage constraint is as follows:
[0135] U i,min ≤U i,t ≤U i,max (15)
[0136] Where: U i,min , U i,max are the lower and upper limits of the voltage amplitude at node i at time t respectively.
[0137] The gas turbine output constraint is as follows:
[0138]
[0139] In the above formula, P mti,min , P mti,max , Q mti,min , Q mti,max are the lower and upper limits of the active and reactive output amplitudes of node i at time t, respectively; is the power generation of the i-th micro gas turbine at time t in scenario ξ; is the reactive power generated by the i-th micro gas turbine at time t in scenario ξ.
[0140] The photovoltaic output constraint is as follows:
[0141]
[0142] Where: is the maximum photovoltaic output at time t; is the amount of system power abandoned at time t in the scenario ξ; is the PV output at time t in scenario ξ.
[0143] The energy storage system constraints are as follows:
[0144]
[0145] In the above formula, is the maximum charging and discharging power of energy storage; is the SOC value of the energy storage system at time t; η es The charging and discharging efficiency of the energy storage system; are the minimum and maximum SOC values of the energy storage system respectively; C bat is the installed capacity of the energy storage system; Δt is the time period t.
[0146] The electricity trading curve and gas turbine output curve of the virtual power plant are as follows: Figure 3 As shown in Figure 1, Price represents the price curve, MT1 and MT2 represent the output curves of different turbine units, and Grid represents the grid output curve. Figure 3 It can be seen that during the period from 8:00 to 15:00, the photovoltaic power generation is large and the electricity price is high. The virtual power plant sells electricity to the grid, and the virtual power plant sells the most electricity at 14:00 when the electricity price is the highest. In other time periods, the virtual power plant purchases electricity from the grid. At night, when the grid electricity price drops below the gas turbine power generation cost, the gas turbine stops working, and the virtual power plant purchases more electricity from the grid.
[0147] S4. Based on the upper-level optimization model and the lower-level optimization model, a two-level planning model is established with the minimum construction and operation and maintenance cost of the energy storage system as the upper-level objective function and the minimum daily operating cost of the virtual power plant as the lower-level objective function. Since the virtual power plant optimization scheduling model contains nonlinear quadratic constraints, second-order cone optimization is required to linearize the constraints.
[0148] The second-order cone optimization defines new variables as follows:
[0149]
[0150] R ij,t =U i,t U j,t cos(θ i,t -θ j,t ) (twenty four)
[0151] T ij,t =U i,t U j,t sin(θ i,t -θ j,t ) (25)
[0152] In the above formula, U i,t is the voltage amplitude of node i at time t; U j,t is the voltage amplitude of node j at time t; μ i,t is the square of the voltage amplitude at time t θ i,t ,θ j,t are the phase angle difference of the voltage at node i and the phase angle difference of the voltage at node j at time t; Rij,t is the cosine product of the phase difference between the voltage amplitudes at node i and node j at time t; T ij,t It is the sine product of the phase difference of the voltage amplitudes at node i and node j at time t.
[0153] Substitute the newly defined variables into equation (13) and equation (14) to obtain new expressions and their constraints.
[0154] The expression of the two-level programming model is:
[0155]
[0156] The constraints of the bilevel programming model are:
[0157]
[0158] In the above formula, g ij , b ij are the conductance and susceptance of branches i and j respectively; P ij,t is the active power of node i and node j at time t; Q ij,t is the reactive power of node i and node j at time t; u i,t 、u j,t They are in and are the voltage amplitude of node i at time t and the voltage amplitude of node j at time t respectively; R ij,t is the cosine product of the phase difference between the voltage amplitudes at node i and node j at time t; T ij,t is the sine product of the phase difference of the voltage amplitudes at node i and node j at time t; U i,min , U i,max are the lower and upper limits of the voltage amplitude at node i at time t respectively.
[0159] S5. The particle swarm algorithm is combined with the CPLEX solver to solve the two-layer optimization model to obtain the optimal configuration result of the energy storage system in the virtual power plant. In the upper optimization model, the particle swarm algorithm is used to optimize the access location, capacity and battery type of the energy storage. In the lower optimization model, CPLEX is called to minimize the operating cost during the daily scheduling cycle of the virtual power plant, optimize the charging and discharging strategy of the energy storage system, and return the scheduling result to the upper layer to calculate the fitness value. After multiple iterations, the optimal configuration result of the energy storage system in the virtual power plant is obtained.
[0160] The learning factor of the particle swarm algorithm is set to 2, the weight coefficient is set to 0.8, the population size is set to 100, and the number of iterations is set to 20. The detailed solution steps of combining the particle swarm algorithm with CPLEX are as follows: Figure 1The above model is solved, and the basic parameters of the distributed power source are shown in Table 1. The technical parameters of the gas turbine are shown in Table 2. For simplicity, the battery types available for the energy storage system are lithium-ion batteries and lead-carbon batteries. Their economic and technical parameters are shown in Table 3. Its convergence is shown in Figure 4 As shown in the figure. After 15 iterations, the fitness value of the particle swarm algorithm reaches the optimal value. At this time, the total daily operating cost of the virtual power plant is 95,386 yuan, which is 1,064 yuan lower than the initial value. This shows that the optimization configuration method in the text has convergence and can effectively reduce the daily operating cost of the energy storage system in the virtual power plant.
[0161] The optimal configuration results of the energy storage system are shown in Table 5. The photovoltaic output, load forecast results and the optimal output curve of the energy storage system in the virtual power plant are shown in Table 5. Figure 2 As shown, load Forecast is the load forecast. Load is the load, PVForcast is the photovoltaic forecast output, PV is the photovoltaic output, and ESS1 and ESS2 are the energy storage system outputs.
[0162] Table 5 Energy storage system configuration results
[0163]
[0164] from Figure 2 It can be seen that the energy storage system in the virtual power plant chooses to charge during the night time when the electricity price is low, and discharges during the daytime when the electricity price is high. In the time period from 7 to 9 o'clock, the actual photovoltaic output is less than its predicted power, and the energy storage system discharges to make up for the power deviation. In the time period from 9 to 11 o'clock, the actual photovoltaic output is greater than its predicted power, and the energy storage system is charged to reduce the loss of abandoned light and reduce the impact of the volatility of photovoltaic output on the stability of the output of the virtual power plant. In the time period from 13 to 15 o'clock, the actual load demand is greater than its predicted value, resulting in a real-time power shortage of the virtual power plant. In order to reduce the unbalanced power cost of purchasing electricity from the power grid, the energy storage system discharges. Therefore, the energy storage system can adjust its own charging and discharging behavior according to the actual situation, and reduce the impact of the uncertainty of distributed power output and load forecast on the virtual power plant.
[0165] This embodiment also provides a computer device, which includes a processor, a memory, a network interface, an input device and a display screen connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory includes a storage medium and an internal memory. The storage medium can be a non-volatile storage medium or a volatile storage medium. The storage medium stores an operating system and can also store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can implement a random optimization method for an energy storage system in a virtual power plant. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the storage medium. The internal memory can also store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute a random optimization method for an energy storage system in a virtual power plant. The network interface of the computer device is used to communicate with an external server through a network connection. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the housing of the computer device, or an external keyboard, touchpad or mouse.
[0166] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, so the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0167] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the above embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.
[0168] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
Claims
1. A random optimization method for energy storage system in a virtual power plant participating in power market transactions, characterized in that: The method includes the following process: The Monte Carlo simulation method is used to generate a set of uncertain scenarios for the output, load forecast and operation status of distributed power sources when the virtual power plant participates in power market transactions. Under uncertain scenarios, an upper-level optimization model is established to optimize the planning and construction of energy storage systems based on the access location, capacity, and battery type of the energy storage system. Establish a lower-level optimization model to optimize the hourly output of distributed power sources within the virtual power plant to minimize operating costs; Based on the upper optimization model and the lower optimization model, a two-level planning model is established with the minimum construction and operation and maintenance cost of the energy storage system as the upper objective function and the minimum daily operation cost of the virtual power plant as the lower objective function. The particle swarm algorithm is combined with the CPLEX solver to solve the two-layer optimization model and obtain the optimal configuration result of the energy storage system in the virtual power plant.
2. The random optimization method for energy storage system in a virtual power plant according to claim 1, characterized in that: The generation of uncertainty scenario sets is specifically: using the Monte Carlo simulation method to generate multiple combination scenario sets of the virtual power plant in the process of participating in electricity market transactions, including photovoltaic output scenario sets, load forecast error scenario sets and gas turbine operating status scenario sets, calculating the expected value after sampling, and then converting the random optimization problem into a deterministic problem for processing.
3. The random optimization method for energy storage system in a virtual power plant according to claim 2, characterized in that: The photovoltaic output scenario set is shown as follows: In the above formula, is the actual output of photovoltaic power; ΔP PV is the photovoltaic output prediction error; P PV Contribute to photovoltaic forecasting; The load forecast error scenario set is as follows: in, s L =P L / 100 In the above formula, ΔP L is the error value of load prediction; μ L , σ L are the statistical mean and standard deviation of load demand respectively; P L is the predicted value of load; is the actual value of load; The gas turbine operating status scenario set is as follows: In the above formula: n i is the number of the normal gas turbine; m is the number of the faulty gas turbine; q ni is the normal gas turbine shutdown probability; q m is the shutdown probability of the faulty gas turbine.
4. The random optimization method for energy storage system in a virtual power plant according to claim 1, characterized in that: The optimization objectives of the upper-level optimization model include the annual investment cost, operation and maintenance cost, and virtual power plant operating cost of the energy storage system. The battery types of the energy storage system are lithium-ion batteries and lead-carbon batteries.
5. The random optimization method for energy storage system in a virtual power plant according to claim 4, characterized in that: The annual investment and construction cost of the energy storage system coupled with its installed capacity can be calculated by the following formula: In the above formula, A is the initial investment cost coefficient of equal annual value; C ES is the cost per unit capacity of the energy storage system; C bat is the installed capacity of the energy storage system; d is the actual interest rate; N r The service life of the energy storage system; The operation and maintenance cost of the energy storage system is mainly related to its rated power, and the rated power of the energy storage system is linearly related to the installed capacity, that is: In the above formula, K bat C is the ratio coefficient between the rated power of the energy storage system and its installed capacity; op is the annual operation and maintenance cost per unit capacity of the energy storage system; The operating cost of a virtual power plant consists of an investment cost constraint and an access location constraint. The investment cost constraint is: C ES C bat ≤C inv.max In the above formula, C inv,max is the maximum investment cost of the energy storage system; The access location constraints are: IS location ∈Φ In the above formula, ES location is the node number of the energy storage system access node; Φ is the set of nodes that can be accessed by the energy storage system.
6. The random optimization method for energy storage system in a virtual power plant according to claim 1, characterized in that: The lower-level optimization model includes system power balance constraints, node power flow constraints, node voltage constraints, gas turbine output constraints, photovoltaic output constraints, and energy storage system constraints.
7. The random optimization method for energy storage system in a virtual power plant according to claim 6, characterized in that: The system power balance constraint is as follows: In the above formula, K is the total number of system nodes; N is the total number of scenes; is the load demand of node i at time t; are the system power loss and abandoned light amount at time t in the scenario ξ respectively; is the power generation of the i-th micro gas turbine at time t in scenario ξ; are the charging power and discharging power of the energy storage system at time t respectively; is the charging power of the energy storage system at time t; is the discharge power of the energy storage system at time t; is the amount of electricity purchased by the virtual power plant from the distribution network at time t in the scenario ξ; is the electricity sold by the virtual power plant from the distribution network at time t in the scenario ξ; The node power flow constraint is as follows: In the above formula, g ij , b ij are the conductance and susceptance of branches i and j respectively; θ ij,t is the phase difference between the voltage of node i and the voltage of node j at time t; N(i) is the set of all nodes connected to node i; U i,t is the voltage amplitude of node i at time t; are the active and reactive outputs of all distributed generation sources at node i at time t respectively; is the reactive load of node i at time t; U j,t is the voltage amplitude of node j at time t; is the active load of node i; The node voltage constraint is as follows: IN i,min ≤U i,t ≤U i,max In the above formula, U i,min , U i,max are the lower and upper limits of the voltage amplitude of node i at time t respectively; The gas turbine output constraint is as follows: In the above formula, P mti,min , P mti,max , Q mti,min , Q mti,max are the lower and upper limits of the active and reactive output amplitudes of node i at time t, respectively; is the active power generated by the i-th micro gas turbine at time t in scenario ξ; is the reactive power generated by the i-th micro gas turbine at time t in scenario ξ; The photovoltaic output constraint is as follows: In the above formula, is the maximum photovoltaic output at time t; is the amount of system power abandoned at time t in the scenario ξ; is the photovoltaic output at time t in scenario ξ; The energy storage system constraints are as follows: In the above formula, is the maximum charging and discharging power of energy storage; is the SOC value of the energy storage system at time t; η es The charging and discharging efficiency of the energy storage system; are the minimum and maximum SOC values of the energy storage system respectively; C bat is the installed capacity of the energy storage system; Δt is the time period t.
8. The random optimization method for energy storage system in a virtual power plant according to claim 1, characterized in that: The expression of the two-level programming model is: The constraints of the bilevel programming model are: In the above formula, g ij , b ij are the conductance and susceptance of branches i and j respectively; P ij,t is the active power of node i and node j at time t; Q ij,t is the reactive power of node i and node j at time t; u i,t 、u j,t They are in and are the voltage amplitude of node i at time t and the voltage amplitude of node j at time t respectively; R ij,t It is the cosine product of the phase difference between the voltage amplitudes at node i and node j at time t; T ij,t It is the sine product of the phase difference of the voltage amplitudes at node i and node j at time t; U i,min , U i,max are the lower and upper limits of the voltage amplitude at node i at time t respectively.
9. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, can implement the random optimization method for the energy storage system in a virtual power plant as described in any one of claims 1 to 8.
10. A computer device, characterized in that: It includes a processor and a memory, the memory is used to store a computer program, and when the computer program is executed by the processor, it implements the random optimization method of the energy storage system in the virtual power plant as described in any one of claims 1 to 8.
Citation Information
Cited By
Commercial building hybrid energy storage system planning method considering virtual energy storage effect
CN119990609A
