Multi-region hybrid energy storage microgrid two-layer optimization configuration method

Abstract

This invention discloses a two-layer optimization configuration method for multi-regional hybrid energy storage microgrids. First, considering the characteristics of energy storage, a battery lifespan cost model and a turbine operating efficiency model are established. Based on the electricity consumption habits of each region, demand response mechanism models for residential, commercial, and industrial areas are established. Then, with annual investment cost and daily operating economics as objective functions, a two-layer optimization configuration model for the multi-regional hybrid energy storage microgrid is established. A piecewise linearization method is used to transform the nonlinear problem of the original model into a mixed-integer linear programming problem. Finally, a particle swarm optimization algorithm and a Gurobi solver are used to solve the upper and lower layers of the model, and the Shapley value method based on demand degree is used to allocate costs. The method proposed in this invention can effectively improve the lifespan of energy storage while effectively reducing the construction and operation costs of microgrids, and can allocate costs more fairly and reasonably.

Description

Technical Field

[0001] This invention belongs to the field of power system technology, specifically relating to a two-layer optimized configuration method for multi-region hybrid energy storage microgrids. Background Technology

[0002] Currently, power supply in remote areas remains a significant challenge. While the increasing proportion of renewable energy in the power system offers an effective solution, solar and wind power generation are subject to weather conditions, and varying electricity consumption habits across different areas within a microgrid lead to resource waste. Energy storage and demand-side response (DSR) can effectively mitigate the problems caused by randomness and inconsistent electricity consumption habits. However, the actual operation of energy storage and DSR within a microgrid still needs to be considered; that is, the characteristics of energy storage and user electricity consumption habits must be taken into account when establishing the system model. As the service area of ​​microgrids expands, and different regions exhibit varying electricity consumption habits and load complementarity, multi-regional cooperation in building microgrids can complement the characteristics of each load, improve the absorption capacity of renewable energy and the performance of the microgrid. However, a reasonable allocation method is also needed to distribute the economic costs of the microgrid to increase the enthusiasm of various regions to participate in its construction. Summary of the Invention

[0003] To address the aforementioned issues, this invention proposes a two-layer optimized configuration method for multi-region hybrid energy storage microgrids. This method can improve the energy utilization efficiency of the system, effectively reduce the operating cost of the system, extend the service life of system components, and reasonably distribute the economic cost of the microgrid.

[0004] This invention proposes a two-layer optimized configuration method for multi-region hybrid energy storage microgrids, and the specific design scheme is as follows:

[0005] (1) Establish a battery life cost model and a turbine operating efficiency model that take into account energy storage characteristics;

[0006] (2) Establish demand response models for residential, commercial, and industrial areas;

[0007] (3) The original nonlinear problem of the model is transformed into a mixed integer linear programming problem by using the piecewise linearization method;

[0008] (4) The two-level optimization model is solved using the particle swarm optimization algorithm and the Gurobi solver;

[0009] (5) The total economic cost of the microgrid is allocated using the Shapley value method based on demand.

[0010] Furthermore, the battery life cost model and turbine operating efficiency model in step (1) are as follows:

[0011] (1-1) Battery life model:

[0012] The lifespan of a battery is related to factors such as its depth of discharge, charge / discharge rate, and number of cycles.

[0013] S R =L R D R N R

[0014] In the formula: S R L represents the total effective discharge capacity. R D R N R These are the battery's rated cycle life, rated depth of discharge, and rated capacity, respectively.

[0015] L A =aD -b e -cD

[0016] In the formula: L A denoted as , where is the actual cycle life of the battery, D is the actual depth of discharge of the battery, and a, b, and c are curve fitting coefficients.

[0017]

[0018] In the formula: I R and I es For rated and actual discharge current; d eff and d a P represents the battery's discharge capacity under rated conditions and its discharge capacity under actual conditions, respectively. R and These are the rated and actual discharge power.

[0019] Considering the combined effects of depth of discharge and discharge rate, the amount of discharge consumed during the i-th battery discharge can be expressed as:

[0020]

[0021]

[0022] In the formula: Let be the discharge capacity of the battery under non-rated conditions converted to rated conditions during the i-th discharge process. Let c be the battery loss cost for the i-th discharge. es N represents the unit investment cost of the battery. es This refers to the battery capacity.

[0023]

[0024] Where: SOC es(t) represents the state of charge of the battery at time t, and σ represents the self-discharge rate of the battery. and This refers to the battery's charging and discharging power and efficiency.

[0025] (1-2) Upper reservoir and reversible turbine:

[0026]

[0027]

[0028]

[0029]

[0030]

[0031] Where: SOC ur (t) represents the state of charge of the upper reservoir, and τ represents the self-discharge rate of the pumped storage. η p , These are the water flow rate, overall efficiency, and pumping power in the pumping mode of a reversible turbine; η t , The parameters represent the water flow rate, overall efficiency, and pumping power of the reversible turbine in power generation mode; ρ is the water density, which is 1000 kg / m³. 3 g is the acceleration due to gravity, which is 9.81 m / s². 2 h represents the height difference between the upper and lower reservoirs.

[0032] Furthermore, the modeling steps for the demand response models of residential, commercial, and industrial areas in step (2) are as follows:

[0033] 1) Residential areas

[0034] Compensation cost for residents participating in demand response C re for:

[0035]

[0036] P re,dr (t)=P re (t)-DR out (t)+DR in (t)

[0037]

[0038] In the formula: P re (t) represents the original resident load, P re,dr(t) represents the residential load after participating in demand response; DR out (t) and DR in (t) represents the loads removed and added; a re The compensation coefficient for loads removed by residential users.

[0039] 2) Commercial area

[0040] Compensation fee C for business users participating in demand response com for:

[0041]

[0042] P com,dr (t)=P com (t)-ΔP com,cut (t)

[0043] Where: ΔP com,cut (t) represents the load reduction amount for commercial users during time period t; c bu (t) represents the compensation electricity price for commercial load shedding at time t, P com (t) represents the original business load, P com,dr (t) represents the commercial load after participating in demand response.

[0044] 3) Industrial areas

[0045] Compensation cost C for industrial users participating in load reduction du Represented as:

[0046]

[0047] P du,dr (t)=P du (t)-ΔP du,cut (t)

[0048] Where: ΔP du,cut (t) represents the load reduction amount for industrial users; a du and b du P is the compensation factor for industrial load reduction. du (t) represents the original industrial load, P du,dr (t) represents the industrial load after participating in demand response.

[0049] Furthermore, the piecewise linearization method described in step (3) is as follows:

[0050] 1) Linearization of battery loss cost function

[0051] Battery loss costs increase with the increase of the actual discharge power of the battery, and the relationship between the two is non-linear.

[0052] The linearized function expression for battery degradation cost is as follows:

[0053]

[0054]

[0055] 0≤P i (t)≤(P i+1 -P i X(t)

[0056] C1 = C(P1)

[0057] In the formula: C and These represent the battery loss cost function and its linear approximation function, respectively; N is the number of segments into which the curve is divided; a i P is the slope of the i-th segment; i P represents the actual discharge power of the battery in the i-th segment; i X(t) represents the actual discharge power of the battery during time period t; P1 represents the minimum discharge power of the battery; X(t) represents the discharge state of the battery during time period t; and C1 represents the battery loss cost at the minimum discharge power.

[0058] 2) Linearization of the power generation-water flow function of reversible hydro turbines

[0059] The working efficiency of reversible hydro turbine power generation mode is closely related to its water flow rate, and the relationship between the two is non-linear.

[0060]

[0061]

[0062] β j =P(Q) j )-α j Q j

[0063] Q j Z j (t)≤Q j (t)≤Q j+1 Z j (t)

[0064]

[0065] In the formula: M is the number of segments into which the curve is divided; P, These are the turbine power generation curve function and its linear approximation function, respectively; α j β j These are the slope and equivalent intercept of the j-th piecewise linear segment, respectively; Q jLet Q be the j-th segment point, where Q1 is the minimum water flow velocity of the turbine operating in power generation mode, and Q... j (t) represents the output of the j-th segment in time period t; Z j (t) represents the state of segment j in time period t, and A(t) represents the operating state of the turbine power generation mode in time period t.

[0066] The linearization method for the pumping power-water flow function of the reversible turbine is consistent with that for the battery loss cost function.

[0067] Furthermore, the objective functions and constraints of the two-layer optimization model in step (4) include the planning layer and the running layer, as shown below.

[0068] 1) Objective function of the planning layer

[0069] minATC=min{C inv +C om +C dr}

[0070] In the formula: C inv C represents the initial investment cost of the components. om For operation and maintenance costs, C dr Cost of responding to demand.

[0071] The initial investment cost includes the investment cost of the wind turbine, reversible water turbine, and upper reservoir.

[0072]

[0073]

[0074] In the formula: R represents the capital recovery factor; r represents the annual interest rate; n is the lifespan of the system; Ω S1 Represents a collection of equipment requiring a one-time investment; N k This represents the unit capacity cost and configuration capacity of the k-th type of equipment.

[0075] Operation and maintenance costs include battery wear and tear costs and maintenance costs for various equipment.

[0076]

[0077] Where: Ω p A collection representing summer, transitional seasons, and winter; Ω S2 Represents a collection of one-time investment equipment and batteries; T d The number of typical days in season D; For the daily maintenance costs of various types of equipment, This represents the daily battery depletion cost under typical day conditions in each season.

[0078] Demand response costs include demand response costs for residential, commercial, and industrial areas.

[0079]

[0080] 2) Planning Level Constraints

[0081] 0≤N k ≤N k,max

[0082] Where: N k,max This represents the maximum capacity at which the k-th type of device can be installed.

[0083] 3) Runtime layer objective function

[0084] The objective function of the runtime layer is the daily operating cost C. day The optimal cost includes battery degradation costs and demand response costs in each region.

[0085]

[0086] 4) Runtime layer constraints

[0087] Microgrid power balance constraints

[0088]

[0089] In the formula: P wt (t) represents the power generation of the wind turbine during time period t; This refers to the battery's output and input power.

[0090] Energy storage constraints

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098] In the formula: The maximum charge and discharge power of the battery is... This represents the maximum water flow rate during turbine power generation. These represent the battery's charging and discharging status flags, respectively. These represent the status flags for charging and discharging of pumped storage hydroelectric power, respectively. This is a binary variable representing the charging and discharging status flags for the k-th type of energy storage, including both battery and pumped hydro storage. These represent the maximum and minimum states of charge for energy storage. and The state of charge at the beginning and end of a day for energy storage.

[0099] The solution method based on particle swarm optimization and Gurobi solver described in step (4) is as follows:

[0100] The planning layer uses a particle swarm optimization algorithm to optimize the system component capacity, deriving the optimal capacity configuration scheme. The operation layer uses the equipment capacity from the planning layer as a constraint, with the daily operational economic optimization as the objective function, and uses the Gurobi solver to obtain the optimal system operation scheme. The inner layer passes the optimal operation scheme to the outer layer to obtain the annual system investment cost. Through continuous iteration of particle position and velocity, the optimal system capacity configuration scheme and operation scheme are finally obtained. The solution process is as follows:

[0101] 1) System initialization: Read raw system data, initialize particle population, generate initial population P, and maximum population size P. max The maximum number of iterations is M.

[0102] 2) Call Gurobi to calculate the inner layer and obtain the optimal operating scheme for each individual in population P within the inner layer.

[0103] 3) Based on the capacity information of the outer layer and the operation scheme of the inner layer, the objective function values ​​of the outer and inner layers are obtained.

[0104] 4) Update the position and velocity of the particle population P to obtain the offspring population R, and replace P with R.

[0105] 5) Check if the maximum number of iterations is met. If not, repeat steps 2)-4). If it is met, output the global optimal target value.

[0106] Furthermore, step (5) uses the demand-based Shapley value method to allocate the total economic cost of the microgrid. The demand-based Shapley value method is as follows:

[0107] Suppose there are n regions in the microgrid, and region R i There are a total of (|S|-1)! arrangements when participating in the co-construction of a microgrid, where |S| is the number of regions in the microgrid that have reached a co-construction agreement at the time of participation. The remaining n-|S| regions have (n-|S|)! arrangements. Region R i The order of the participating combinations divided by all combinations of n regions equals region R. iThe weight of the cost allocated to the microgrid after participating in the co-construction.

[0108]

[0109] In the formula: w i For region R i Weighting of cost allocation.

[0110] Region R i The cost f of the k-th type of component of the microgrid is allocated. i k for:

[0111]

[0112] In the formula: Y = {1, 2, ..., n}; V k (S) is R i The cost of the k-th component when participating in the co-construction of a microgrid, including wind turbine cost, battery loss cost, reversible turbine cost, reservoir cost, and demand-side response cost; V k (S\i) represents the component cost when not participating in the co-construction of the microgrid.

[0113] Region R i The shared construction cost F of the microgrid that should be allocated in the microgrid i for:

[0114]

[0115] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0116] (1) In modeling energy storage, this invention considers battery loss and the efficiency of reversible turbines, which effectively improves battery life and the energy utilization rate of pumped storage power stations.

[0117] (2) When modeling the demand-side response model, this invention takes into account the user's electricity consumption habits, which effectively improves the user's enthusiasm for the demand-side response mechanism.

[0118] (3) This invention proposes a Shapley value method based on demand to reasonably allocate the economic cost of co-constructing microgrids, so as to improve the enthusiasm of each region to participate in co-constructing microgrids. Attached Figure Description

[0119] Figure 1 This is a framework diagram of the multi-region hybrid energy storage microgrid of the present invention.

[0120] Figure 2 This is a schematic diagram illustrating the linearization of the battery loss cost-power curve in an example of the present invention.

[0121] Figure 3 This is a schematic diagram illustrating the linearization of turbine power generation versus water flow rate in an example of the present invention.

[0122] Figure 4 This is a diagram of the planning-running two-layer optimization structure in an example of the present invention. Detailed Implementation

[0123] The invention is further illustrated below with reference to specific embodiments and accompanying drawings. This invention proposes a two-layer optimized configuration method for multi-region hybrid energy storage microgrids. The framework diagram of the multi-region hybrid energy storage microgrid is shown below. Figure 1 As shown, the specific implementation steps are as follows:

[0124] (1) Establish a battery life cost model and a turbine operating efficiency model;

[0125] The lifespan of a battery is related to factors such as its depth of discharge, charge / discharge rate, and number of cycles.

[0126] S R =L R D R N R

[0127] In the formula: S R L represents the total effective discharge capacity. R D R N R These are the battery's rated cycle life, rated depth of discharge, and rated capacity, respectively.

[0128] L A =aD -b e -cD

[0129] In the formula: L A denoted as , where is the actual cycle life of the battery, D is the actual depth of discharge of the battery, and a, b, and c are curve fitting coefficients.

[0130]

[0131] In the formula: I R and I es For rated and actual discharge current; d eff and d a P represents the battery's discharge capacity under rated conditions and its discharge capacity under actual conditions, respectively. R and These are the rated and actual discharge power.

[0132] Considering the combined effects of depth of discharge and discharge rate, the amount of discharge consumed during the i-th battery discharge can be expressed as:

[0133]

[0134]

[0135] In the formula: Let be the discharge capacity of the battery under non-rated conditions converted to rated conditions during the i-th discharge process. Let c be the battery loss cost for the i-th discharge. es N represents the unit investment cost of the battery. es This refers to the battery capacity.

[0136]

[0137] Where: SOC es (t) represents the state of charge of the battery at time t, and σ represents the self-discharge rate of the battery. and This refers to the battery's charging and discharging power and efficiency.

[0138] (1-2) Upper reservoir and reversible turbine:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144] Where: SOC ur (t) represents the state of charge of the upper reservoir, and τ represents the self-discharge rate of the pumped storage. η p , These are the water flow rate, overall efficiency, and pumping power in the pumping mode of a reversible turbine; η t , The parameters represent the water flow rate, overall efficiency, and pumping power of the reversible turbine in power generation mode; ρ is the water density, which is 1000 kg / m³. 3 g is the acceleration due to gravity, which is 9.81 m / s². 2 h represents the height difference between the upper and lower reservoirs.

[0145] (2) Establish demand response models for residential, commercial and industrial areas.

[0146] 1) Residential areas

[0147] Compensation cost for residents participating in demand response C re for:

[0148]

[0149] P re,dr (t)=P re (t)-DR out (t)+DR in (t)

[0150]

[0151] In the formula: P re (t) represents the original resident load, P re,dr (t) represents the residential load after participating in demand response; DR out (t) and DR in (t) represents the loads removed and added; a re The compensation coefficient for loads removed by residential users.

[0152] 2) Commercial area

[0153] Compensation fee C for business users participating in demand response com for:

[0154]

[0155] P com,dr (t)=P com (t)-ΔP com,cut (t)

[0156] Where: ΔP com,cut (t) represents the load reduction amount for commercial users during time period t; c bu (t) represents the compensation electricity price for commercial load shedding at time t, P com (t) represents the original business load, P com,dr (t) represents the commercial load after participating in demand response.

[0157] 3) Industrial areas

[0158] Compensation cost C for industrial users participating in load reduction du Represented as:

[0159]

[0160] P du,dr (t)=P du (t)-ΔP du,cut (t)

[0161] Where: ΔP du,cut (t) represents the load reduction amount for industrial users; a du and b duP is the compensation factor for industrial load reduction. du (t) represents the original industrial load, P du,dr (t) represents the industrial load after participating in demand response.

[0162] (3) Piecewise linearization is used to transform the original nonlinear problem of the model into a mixed integer linear programming problem;

[0163] 1) Linearization of battery loss cost function

[0164] Battery loss costs increase with the increase of the actual discharge power of the battery, and the relationship between the two is non-linear. Figure 2 This is a schematic diagram illustrating the linearization of the battery loss cost-power curve.

[0165] The linearized function expression for battery degradation cost is as follows:

[0166]

[0167]

[0168] 0≤P i (t)≤(P i+1 -P i X(t)

[0169] C1 = C(P1)

[0170] In the formula: C and These represent the battery loss cost function and its linear approximation function, respectively; N is the number of segments into which the curve is divided; a i P is the slope of the i-th segment; i P represents the actual discharge power of the battery in the i-th segment; i X(t) represents the actual discharge power of the battery during time period t; P1 represents the minimum discharge power of the battery; X(t) represents the discharge state of the battery during time period t; and C1 represents the battery loss cost at the minimum discharge power.

[0171] 2) Linearization of the power generation-water flow function of reversible hydro turbines

[0172] The operating efficiency of a reversible hydro turbine power generation system is closely related to its water flow rate, and the relationship between the two is non-linear. Figure 3 A schematic diagram illustrating the linearization of turbine power generation versus water flow.

[0173]

[0174]

[0175] β j =P(Q) j )-α j Qj

[0176] Q j Z j (t)≤Q j (t)≤Q j+1 Z j (t)

[0177]

[0178] In the formula: M is the number of segments into which the curve is divided; P, These are the turbine power generation curve function and its linear approximation function, respectively; α j β j These are the slope and equivalent intercept of the j-th piecewise linear segment, respectively; Q j Let Q be the j-th segment point, where Q1 is the minimum water flow velocity of the turbine operating in power generation mode, and Q... j (t) represents the output of the j-th segment in time period t; Z j (t) represents the state of segment j in time period t, and A(t) represents the operating state of the turbine power generation mode in time period t.

[0179] The linearization method for the pumping power-water flow function of the reversible turbine is consistent with that for the battery loss cost function.

[0180] (4) Establish a two-level optimization model, including the objective functions and constraints of the planning layer and the operation layer, as shown below. Figure 4 A two-tier optimized structure for planning and operation.

[0181] 1) Objective function of the planning layer

[0182] minATC=min{C inv +C om +C dr}

[0183] In the formula: C inv C represents the initial investment cost of the components. om For operation and maintenance costs, C dr Cost of responding to demand.

[0184] The initial investment cost includes the investment cost of the wind turbine, reversible water turbine, and upper reservoir.

[0185]

[0186]

[0187] In the formula: R represents the capital recovery factor; r represents the annual interest rate; n is the lifespan of the system; Ω S1 Represents a collection of equipment requiring a one-time investment; Nk This represents the unit capacity cost and configuration capacity of the k-th type of equipment.

[0188] Operation and maintenance costs include battery wear and tear costs and maintenance costs for various equipment.

[0189]

[0190] Where: Ω p A collection representing summer, transitional seasons, and winter; Ω S2 Represents a collection of one-time investment equipment and batteries; T d The number of typical days in season D; For the daily maintenance costs of various types of equipment, This represents the daily battery depletion cost under typical day conditions in each season.

[0191] Demand response costs include demand response costs for residential, commercial, and industrial areas.

[0192]

[0193] 2) Planning Level Constraints

[0194] 0≤N k ≤N k,max

[0195] Where: N k,max This represents the maximum capacity at which the k-th type of device can be installed.

[0196] 3) Runtime layer objective function

[0197] The objective function of the runtime layer is the daily operating cost C. day The optimal cost includes battery degradation costs and demand response costs in each region.

[0198]

[0199] 4) Runtime layer constraints

[0200] Microgrid power balance constraints

[0201]

[0202] In the formula: P wt (t) represents the power generation of the wind turbine during time period t; This refers to the battery's output and input power.

[0203] Energy storage constraints

[0204]

[0205]

[0206]

[0207]

[0208]

[0209]

[0210]

[0211] In the formula: The maximum charge and discharge power of the battery is... This represents the maximum water flow rate during turbine power generation. These represent the battery's charging and discharging status flags, respectively. These represent the status flags for charging and discharging of pumped storage hydroelectric power, respectively. This is a binary variable representing the charging and discharging status flags for the k-th type of energy storage, including both battery and pumped hydro storage. These represent the maximum and minimum states of charge for energy storage. and The state of charge (SOC) at the beginning and end of a day for energy storage is given. To verify the effectiveness of the proposed capacity configuration method, the economic costs and carbon emissions of the following three systems are compared and analyzed.

[0212] (5) The two-level optimization model is solved using the particle swarm optimization algorithm and the Gurobi solver, specifically as follows:

[0213] The planning layer uses a particle swarm optimization algorithm to optimize the system component capacity, deriving the optimal capacity configuration scheme. The operation layer uses the equipment capacity from the planning layer as a constraint, with the daily operational economic optimization as the objective function, and uses the Gurobi solver to obtain the optimal system operation scheme. The inner layer passes the optimal operation scheme to the outer layer to obtain the annual system investment cost. Through continuous iteration of particle position and velocity, the optimal system capacity configuration scheme and operation scheme are finally obtained. The solution process is as follows:

[0214] 1) System initialization: Read raw system data, initialize particle population, generate initial population P, and maximum population size P. max The maximum number of iterations is M.

[0215] 2) Call Gurobi to calculate the inner layer and obtain the optimal operating scheme for each individual in population P within the inner layer.

[0216] 3) Based on the capacity information of the outer layer and the operation scheme of the inner layer, the objective function values ​​of the outer and inner layers are obtained.

[0217] 4) Update the position and velocity of the particle population P to obtain the offspring population R, and replace P with R.

[0218] 5) Check if the maximum number of iterations is met. If not, repeat steps 2)-4). If it is met, output the global optimal target value.

[0219] The specific model solution flowchart is as follows: Figure 4 As shown.

[0220] (6) The total economic cost of the microgrid is allocated using the Shapley value method based on demand.

[0221] Suppose there are n regions in the microgrid, and region R i There are a total of (|S|-1)! arrangements when participating in the co-construction of a microgrid, where |S| is the number of regions in the microgrid that have reached a co-construction agreement at the time of participation. The remaining n-|S| regions have (n-|S|)! arrangements. Region R i The order of the participating combinations divided by all combinations of n regions equals region R. i The weight of the cost allocated to the microgrid after participating in the co-construction.

[0222]

[0223] In the formula: w i For region R i Weighting of cost allocation.

[0224] Region R i The cost f of the k-th type of component of the microgrid is allocated. i k for:

[0225]

[0226] In the formula: Y = {1, 2, ..., n}; V k (S) is R i The cost of the k-th component when participating in the co-construction of a microgrid, including wind turbine cost, battery loss cost, reversible turbine cost, reservoir cost, and demand-side response cost; V k (S\i) represents the component cost when not participating in the co-construction of the microgrid.

[0227] Region R i The shared construction cost F of the microgrid that should be allocated in the microgrid i for:

[0228]

[0229] To verify the effectiveness of the two-layer optimization configuration method for multi-region hybrid energy storage microgrids, we compare and analyze the following various types of alliance forms.

[0230] Table 1. Economic Costs of Each Alliance

[0231]

[0232] Table 2 Cost Allocation between Conventional and Demand-Based Shapley Value Methods

[0233]

[0234] Table 1 shows the economic costs of microgrids under each alliance. As can be seen from Table 1, compared to independently constructed systems 1, 2, and 3, the economic costs of jointly constructed systems 4, 5, 6, and 7 are lower than the sum of their individual economic costs. Compared to the conventional Shapley value method which allocates costs based on total economic cost, the demand-based Shapley value method reduces the allocated costs for regions R2 and R3 by $600 and $600 respectively, while increasing the allocated cost for region R1 by $1200. This is because region R1 has a larger total load and a greater peak-to-valley difference, resulting in a higher allocated cost. This demonstrates the effectiveness of the demand-based Shapley value method and can also increase the enthusiasm of regions R2 and R3 to participate in the co-construction of microgrids.

[0235] The above embodiments are used to explain the present invention, but not to limit the present invention. Any modifications and changes made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A two-layer optimized configuration method for multi-region hybrid energy storage microgrids, characterized in that, Includes the following steps: (1) Establish a battery life cost model and a turbine operating efficiency model that take into account energy storage characteristics; (2) Establish demand response models for residential, commercial, and industrial areas, including: 1) Residential areas; Compensation costs for residents participating in demand response for: ； ； ； In the formula: For the original residents' load, For residential load following demand response; and For loads being moved out and moved in; The compensation coefficient for loads removed by residential users; 2) Commercial areas; Compensation fees for business users participating in demand response for: ； ； In the formula: for Load reduction for commercial users during specific time periods; In order to be in Compensation price for commercial load reduction at any time For the original commercial load, To participate in business loads following demand response; 3) Industrial areas; Compensation fees for industrial users participating in load reduction Represented as: ； ； In the formula: Load reduction for industrial users; and This is the compensation coefficient for industrial load reduction; For the original industrial load, To participate in industrial loads following demand response; (3) The original nonlinear problems of the battery life cost model and the turbine working efficiency model are transformed into mixed integer linear programming problems by using the piecewise linearization method; (4) A two-level optimization model is established with the minimum annual investment cost as the upper-level objective and the minimum daily operating economic cost as the lower-level objective. The particle swarm optimization algorithm and the Gurobi solver are used to solve the two-level optimization model to obtain the optimal system capacity configuration scheme and operation scheme. The two-level optimization model includes objective functions for the planning layer and the operation layer, as shown below. 1) Objective function of the planning layer; ； In the formula: The initial investment cost of the components, For operation and maintenance costs, Cost of responding to demand; The initial investment cost includes the investment cost of the wind turbine, reversible water turbine, and upper reservoir; ； ； In the formula: Represents the capital recovery factor; Represents the annual interest rate; For the system's lifespan; Represents a collection of equipment requiring a one-time investment; , Representing the Unit capacity cost and configuration capacity of this type of equipment; Operating and maintenance costs include battery wear and tear costs and maintenance costs for various equipment. ； In the formula: A collection representing summer, the transitional season, and winter; This represents a collection of one-time investment equipment and batteries; for The number of typical days in a given season; For the daily maintenance costs of various types of equipment, The daily battery depletion cost under typical days in each season; Demand response costs include demand response costs for residential, commercial, and industrial areas; ； 2) Objective function of the runtime layer; The objective function of the runtime layer is daily operating cost. The optimal cost includes battery degradation costs and demand response costs in each region. ； (5) The total economic cost of the microgrid, including wind turbine cost, battery loss cost, reversible turbine cost, upper reservoir cost, and demand-side response cost, is allocated using the demand-based Shapley value method. Assume that the microgrid has a total of A region, obtain the region The shared construction costs of microgrids that should be allocated within a microgrid .

2. The method for dual-layer optimized configuration of multi-region hybrid energy storage microgrids according to claim 1, characterized in that: The battery life cost model and turbine operating efficiency model in step (1) are as follows: (1-1) Battery life model: The lifespan of a battery is related to its depth of discharge, charge / discharge rate, and number of cycles. ； In the formula: The total effective discharge capacity, , , These are the battery's rated cycle life, rated depth of discharge, and rated capacity, respectively. ； In the formula: This refers to the actual cycle life of the battery. This represents the actual depth of battery discharge. These are the curve fitting coefficients; ； In the formula: and These are the rated and actual discharge currents; and These represent the battery's discharge capacity under rated conditions and its discharge capacity under actual conditions, respectively. and Rated and actual discharge power; Taking into account the effects of both discharge depth and discharge rate, the first The amount of discharge consumed by the secondary battery can be expressed as: ； ； In the formula: For the battery in the first The discharge process transforms non-rated conditions into the discharge quantity under rated conditions. For the first Battery loss cost per discharge The unit investment cost of the battery, Battery capacity; ； In the formula: For the battery State of charge at time t, This refers to the battery's self-discharge rate. , and , For the battery's charging and discharging power and charging and discharging efficiency; (1-2) Upper reservoir and reversible turbine: ； ； ； ； ； In the formula: The state of charge of the upper reservoir. The self-discharge rate of pumped hydro storage. , , These are the water flow rate, overall efficiency, and pumping power in the pumping mode of a reversible turbine; , , The parameters are: water flow rate, overall efficiency, and pumping power of the reversible turbine in power generation mode; The density of water is 1000 kg / m³. 3 ; The acceleration due to gravity is 9.81 m / s². 2 ; This represents the height difference between the upper and lower reservoirs.

3. The two-layer optimized configuration method for multi-region hybrid energy storage microgrids according to claim 2, characterized in that: The two-layer optimization model in step (4) includes constraints for the planning layer and the operation layer, as shown below; 1) Planning-level constraints; ； In the formula: For installation of the first The maximum capacity of this type of equipment; 2) Runtime layer constraints; Microgrid power balance constraints ； In the formula: For wind turbines Power generation during a given time period; , For the battery's output and input power; Energy storage constraints; ； ； ； ； ； ； ； In the formula: , The maximum charge and discharge power of the battery is... ; This represents the maximum water flow rate during turbine power generation. , These represent the battery's charging and discharging status flags, respectively. , These represent the status flags for charging and discharging of pumped storage hydroelectric power, respectively. , For the first The charging and discharging status flags for various types of energy storage are binary variables, including both battery and pumped hydro storage. , These represent the maximum and minimum states of charge for energy storage. and The state of charge at the beginning and end of a day for energy storage.

4. The two-layer optimized configuration method for multi-region hybrid energy storage microgrids according to claim 3, characterized in that: The solution method based on particle swarm optimization and Gurobi solver in step (4) is as follows: The planning layer uses a particle swarm optimization algorithm to optimize the capacity of system components, deriving the optimal capacity configuration scheme. The operation layer uses the equipment capacity of the planning layer as a constraint, with the daily operational economic optimization as the objective function, and uses the Gurobi solver to obtain the optimal system operation scheme. The inner layer passes the optimal operation scheme to the outer layer to obtain the annual investment cost of the system. Through continuous iteration of particle position and velocity, the optimal system capacity configuration scheme and operation scheme are finally obtained. The solution process is as follows: 1) System initialization: Read the original system data, initialize the particle population, generate the initial population P, the maximum population size Pmax, and the maximum number of iterations M; 2) Call Gurobi to calculate the inner layer and obtain the optimal operating scheme for each individual in population P within the inner layer; 3) Based on the capacity information of the outer layer and the operation plan of the inner layer, the objective function values ​​of the outer and inner layers are obtained; 4) Update the position and velocity of the particle population P to obtain the offspring population R, and replace P with R; 5) Check if the maximum number of iterations is met. If not, repeat steps 2)-4). If it is met, output the global optimal target value.

Images