Distribution area micro-grid distributed power supply-energy storage robust collaborative planning method considering load side resource flexibility

Abstract

The invention discloses an area micro-grid distributed power supply-energy storage robust collaborative planning method considering load side resource flexibility, and belongs to the technical field of power system planning. Comprising the steps that energy storage characteristics of load side electric vehicles and heating ventilation air conditioners are considered, and a cluster electric vehicle generalized energy storage aggregation model and a heating ventilation air conditioner generalized energy storage aggregation model are established; considering the influence of source-load uncertainty, and according to the cluster electric vehicle generalized energy storage aggregation model and the heating ventilation air conditioner generalized energy storage aggregation model, establishing a zone area micro-grid robust collaborative planning model considering the load side resource flexibility; and solving the zone area micro-grid robust collaborative planning model considering the load side resource flexibility by adopting a nested column and a constraint generation algorithm. According to the collaborative planning model established by the method, the flexible adjustment potential of load side resources can be fully mined, the energy storage configuration scale is remarkably reduced, and economic and efficient operation of the micro-grid can be ensured.

Description

Technical Field

[0001] This invention relates to the field of power system planning technology, and in particular to a robust collaborative planning method for distributed generation and energy storage in a distribution microgrid that takes into account the flexibility of load-side resources. Background Technology

[0002] The large-scale development of new energy sources has brought significant opportunities to the power industry, but it has also brought unprecedented challenges. Achieving efficient utilization of new energy sources has become an important way for the power system to achieve low-carbon transformation. As a basic unit of the power system, distribution microgrids play a pioneering role in addressing the challenges of new energy sources. However, the inherent volatility and uncertainty of new energy sources may lead to prominent problems for microgrids, such as insufficient energy storage backup capacity, reduced voltage quality, and source-load imbalance, which seriously threaten the stability of system operation.

[0003] Currently, significant work has been done in the planning of distributed generation and energy storage. However, research on microgrid planning, particularly on the energy storage characteristics of load-side flexibility resources, is insufficient. The impact of aggregating diverse flexibility resources on microgrid planning has not been fully explored. Furthermore, from the perspective of overall microgrid planning, existing technologies mainly focus on the coordinated planning of new energy sources and energy storage, neglecting the adjustment potential of diverse flexibility resources and failing to fully consider the importance of load-side resources participating in the coordinated planning of microgrid generation and energy storage. This, to some extent, limits the improvement of microgrid operating efficiency and energy utilization efficiency. Therefore, a robust coordinated planning method for distributed generation and energy storage in distribution microgrids that considers the flexibility of load-side resources is needed to fully exploit the flexibility of distribution microgrid resources and ensure the economical and efficient operation of the microgrid. Summary of the Invention

[0004] The purpose of this invention is to propose a robust collaborative planning method for distributed generation and energy storage in a transformer substation microgrid that considers load-side resource flexibility, including:

[0005] Considering the energy storage characteristics of load-side electric vehicles and HVAC systems, a generalized energy storage aggregation model for clustered electric vehicles and a generalized energy storage aggregation model for HVAC systems are established.

[0006] Considering the impact of source-load uncertainty, a robust collaborative planning model for the distribution microgrid, taking into account the resource flexibility of the load side, is established based on the generalized energy storage aggregation model of clustered electric vehicles and the generalized energy storage aggregation model of HVAC.

[0007] A nested column and constraint generation algorithm is used to solve the robust collaborative planning model of a distribution microgrid that takes into account the resource flexibility of the load side.

[0008] Furthermore, the generalized energy storage aggregation model for clustered electric vehicles is as follows:

[0009]

[0010] In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These represent the upper and lower power limits of the cluster EV, respectively. These represent the upper and lower limits of the energy consumption of the cluster EVs, respectively. This refers to the actual power of the cluster EVs; Let be the electrical energy of the cluster EV at time t; Δt represents the change in the electrical energy of the cluster EVs when EVs are connected to or disconnected from the grid; Δt is the time interval.

[0011] Furthermore, the generalized energy storage aggregation model for HVAC systems is as follows:

[0012]

[0013] In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These are the upper and lower power limits for the cluster HVAC system, respectively. and These are the upper and lower limits of power consumption for the cluster HVAC system; This represents the actual power of the cluster HVAC system. Let be the electrical energy of the cluster HVAC at time t; β be the dissipation coefficient; and Δt be the time interval.

[0014] Furthermore, the robust collaborative planning model for distribution microgrids that takes into account the resource flexibility of the load side includes the objective function of the planning model, sub-models for multiple types of uncertain resources, and constraints of the planning model.

[0015] Furthermore, the objective function of the planning model is:

[0016] minC total =C inv +C op

[0017] In the formula, C total C represents the average annual total cost of the microgrid in the entire planning period of the distribution area. inv and C op These are the equivalent annual investment cost and the average annual operating cost of the system, respectively.

[0018] in:

[0019]

[0020] In the formula, For distributed wind power investment costs, For distributed photovoltaic investment costs, Let z be the energy storage investment cost, and z be the discount rate; y PV y WT and yES These are the investment periods for photovoltaic, wind power, and energy storage, respectively. and These are the unit investment cost coefficients for photovoltaic, wind power, and energy storage, respectively. and These are the configuration capacities for photovoltaic, wind power, and energy storage, respectively.

[0021]

[0022] In the formula, For the operation and maintenance costs of wind and solar generator sets, For energy storage operation and maintenance costs, For flexible resource scheduling costs, The costs of curtailing wind and solar power and the cost of purchasing electricity are represented by t and r, respectively, which are the time period number and the typical day number for each season; π r This represents the percentage of each typical day out of the total number of days in a year. and These are the photovoltaic operation and maintenance cost coefficient, wind turbine operation and maintenance cost coefficient, energy storage operation and maintenance cost coefficient, cluster EV dispatch cost coefficient, cluster HVAC dispatch cost coefficient, curtailment penalty cost coefficient, and electricity purchase price; These are typical daily photovoltaic power, wind power, energy storage charging power, and energy storage discharging power, respectively. and These represent typical daily wind and solar curtailment amounts; This represents the electricity purchased from the distribution network; Δt is the time interval.

[0023] Furthermore, the multi-type uncertainty resource sub-model includes:

[0024]

[0025] In the formula, and These represent the uncertain power outputs of EVs, HVAC systems, photovoltaic power, wind power, and conventional loads, respectively. and These are the predicted power values ​​for each unit; and These represent the maximum power fluctuation deviations of each unit; and These are the actual power fluctuation coefficients for each unit; t and r are the time period number and the typical day number for each season, respectively; T is the planned time period.

[0026] Sub-model constraints:

[0027]

[0028] In the formula, Γ EVVB ,ΓHVVB ,Γ PV ,Γ WT and Γ L This represents the uncertain budget for the corresponding unit.

[0029] Furthermore, the constraints of the planning model include:

[0030] Constraints on Distributed New Energy Planning:

[0031]

[0032] In the formula, This refers to the maximum allowable installed capacity of distributed wind turbine units. This refers to the maximum allowable installed capacity of distributed photovoltaic (PV) units.

[0033] Energy storage planning constraints:

[0034]

[0035] In the formula, t ES Energy storage duration at rated power; Configuration capacity for energy storage;

[0036] Planning and investment constraints:

[0037]

[0038] In the formula, Investment ceiling; C inv The cost of investment is the equivalent annual value.

[0039] Microgrid power balance constraints:

[0040]

[0041] In the formula, and These refer to the charging and discharging power of flexible resources; and These are the energy storage charging and discharging power, respectively. These represent typical daily photovoltaic and wind power outputs, respectively. For the purchase of electricity from the distribution network; Electricity sales volume; This is the power of a normal load. This refers to the amount of wind and solar power curtailed; among which, the interaction power between the distribution substation microgrid and the distribution network must meet the following requirements:

[0042]

[0043] In the formula, To purchase electricity from the distribution network; These are 0-1 variables, representing electricity sales and electricity purchases; This represents the maximum cross-connection power.

[0044] Operational constraints of distributed renewable energy sources:

[0045]

[0046] In the formula, and These represent typical daily photovoltaic and wind power outputs, respectively. and These represent the uncertain power output per unit of photovoltaic power, wind power output, and conventional load, respectively. These are the configuration capacities for photovoltaic and wind power, respectively.

[0047] Load-side flexibility resource constraints:

[0048]

[0049] In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These represent the upper and lower power limits of the cluster EV, respectively. These represent the upper and lower limits of the energy consumption of the cluster EVs, respectively. This refers to the actual power of the cluster EVs; Let be the electrical energy of the cluster EV at time t;

[0050]

[0051] In the formula, These are the upper and lower power limits for the cluster HVAC system, respectively. and These are the upper and lower limits of power consumption for the cluster HVAC system; This represents the actual power of the cluster HVAC system. Let represent the electrical energy of the cluster HVAC system at time t.

[0052] The beneficial effects of this invention are as follows:

[0053] The collaborative planning model established by the method of this invention can not only fully tap the flexible adjustment potential of load-side resources and significantly reduce the scale of energy storage configuration, but also ensure the economical and efficient operation of microgrids. Attached Figure Description

[0054] Figure 1 The flowchart is for the robust collaborative planning method of distributed generation and energy storage in a transformer microgrid that takes into account the flexibility of load-side resources in this invention.

[0055] Figure 2(a) and (b) are the feasible regions for unit EV power, where (a) is when the initial capacity is less than the lower limit of battery capacity, and (b) is when the initial capacity is greater than the lower limit of battery capacity.

[0056] Figure 3 Flowchart of the algorithm for solving the robust collaborative planning model of a distribution microgrid that takes into account the flexibility of load-side resources;

[0057] Figure 4 Diagram of the microgrid system structure in the distribution area;

[0058] Figure 5 This is a comparison chart of various indicators between scenario 1 and scenario 2 of the present invention;

[0059] Figure 6 This is a comparison chart of various indicators between scenario 1 and scenario 3 of the present invention;

[0060] Figure 7 The total load curves for scenarios 1 and 4 of the present invention are shown with and without time-of-use pricing. Detailed Implementation

[0061] This invention proposes a robust collaborative planning method for distributed generation and energy storage in a microgrid that considers the flexibility of load-side resources. The invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0062] Figure 1 The flowchart of the robust collaborative planning method for distributed generation and energy storage in a transformer substation microgrid, which considers load-side resource flexibility according to the present invention, includes the following steps:

[0063] Step A: Considering the energy storage characteristics of load-side electric vehicles and HVAC systems, establish a generalized energy storage aggregation model for clustered electric vehicles and a generalized energy storage aggregation model for HVAC systems.

[0064] first, Figure 2 (a) and (b) represent the feasible regions per unit EV power, where (a) represents an initial capacity less than the lower limit of battery capacity, and (b) represents an initial capacity greater than the lower limit of battery capacity. Where T... o T d These refer to the EV's grid connection time and grid disconnection time, respectively; E i,o This refers to the initial charge level of the EV when connected to the grid; E i,ex This represents the vehicle owner's desired battery level during off-grid operation. (E) i,max and E i,min The upper and lower limits of the EV's charge and discharge battery capacity are limited by the initial capacity E of the EV at the time of grid connection. i,o With the lower boundary E of EV battery capacity i,min The comparison divides the EV feasible region into two scenarios, a and b. When E i,o Less than the lower boundary E i,minAt that time, the EV will first charge at maximum power to E i,min Then charge and discharge; when E i,o Greater than the lower boundary E i,min At this time, the EV can be directly charged and discharged. In both cases, the trajectory of the upper boundary is that it is charged at maximum power from the moment of grid connection to E. i,max Only the lower boundary is different.

[0065] Based on the feasible operating region of a single EV, the energy and power boundaries of the cluster EVs are further obtained using the Minkowski summation method, as shown below:

[0066]

[0067] In the formula, N e The total number of EVs in the cluster is shown; i, t, and r are the EV number, time period number, and typical day number for each season, respectively. These represent the maximum and minimum electrical energy per unit EV, respectively. These represent the maximum and minimum power per EV, respectively; This indicates the grid connection status of the i-th EV during time period t, with a value of 0 or 1, where 0 indicates no connection and 1 indicates connection. These represent the upper and lower limits of the energy consumption of the cluster EVs, respectively. These represent the upper and lower power limits of the cluster EVs, respectively.

[0068] In addition, the energy of the cluster EVs will change as one or more EVs are connected to and disconnected from the grid. The change in the energy of the cluster EVs when the i-th EV is connected to and disconnected from the grid is shown below.

[0069]

[0070] In the formula, Let be the electrical energy of the cluster after the i-th EV is connected to the grid and disconnected from the grid, respectively. When m EVs are connected to or disconnected from the grid, the change in electrical energy is the sum of the differences between equations (3) and (4), as shown below:

[0071]

[0072] In the formula, This represents the change in the electrical energy of the cluster of EVs when m EVs are connected to or disconnected from the grid.

[0073] A generalized energy storage aggregation model for electric vehicle (EV) clusters is established as follows:

[0074]

[0075] In the formula, This refers to the actual power of the cluster EVs; Let be the electrical energy of the cluster EV at time t; Δt is the time interval.

[0076] The upper and lower power limits of a clustered HVAC system can be obtained by summing the maximum and minimum power of a single HVAC unit:

[0077]

[0078] In the formula, N h This represents the total number of HVAC units in the cluster. These represent the maximum and minimum power per unit of HVAC; These represent the upper and lower power limits of the clustered HVAC system. Therefore, the upper and lower power limits for charging and discharging of the clustered HVAC system based on the generalized energy storage model are as follows:

[0079]

[0080] In the formula, α out α in These are the ambient temperature and the air conditioning set temperature, respectively; C p R is the energy efficiency coefficient; R is the thermal resistance. This is a power reference value; and These represent the upper and lower limits of the charging and discharging power of the cluster HVAC system.

[0081] The upper and lower limits of power consumption for clustered HVAC systems are shown below:

[0082]

[0083] In the formula, τ is the controllable time period; β is the dissipation coefficient; and C is the heat capacity. These are the cluster power values ​​used to maintain the battery level within the upper and lower limits, respectively. and These are the upper and lower limits of electrical energy for the cluster HVAC system.

[0084] Finally, a generalized energy storage aggregation model for Heating, Ventilating and Conditioning (HVAC) is established, as shown below:

[0085]

[0086] In the formula, This represents the actual power of the cluster HVAC system. Δt represents the electrical energy of the cluster HVAC system at time t; Δt is the time interval.

[0087] Step B: Considering the impact of source-load uncertainty, establish a robust collaborative planning model for the distribution microgrid that takes into account the resource flexibility of the load side, based on the generalized energy storage aggregation model of clustered electric vehicles and the generalized energy storage aggregation model of HVAC.

[0088] The objective function of the collaborative planning model is the average annual total cost C of the microgrid area over the entire planning period. total Minimize, including the equivalent annual investment cost and the system's average annual operating cost, the objective function is as follows:

[0089] minC total =C inv +C op (16)

[0090] In the formula, C inv and C op These are the equivalent annual investment cost and the average annual operating cost of the system, respectively.

[0091] Equivalent annual investment cost C inv Including distributed wind power investment costs Distributed photovoltaic investment costs and energy storage investment costs As shown below:

[0092]

[0093] In the formula, z is the discount rate; y PV y WT and y ES These are the investment periods for photovoltaic, wind power, and energy storage, respectively. and These are the unit investment cost coefficients for photovoltaic, wind power, and energy storage, respectively. and These are the configuration capacities for photovoltaic, wind power, and energy storage, respectively.

[0094] The average annual operating cost of the system, C op Including the operation and maintenance costs of wind and solar power generators Energy storage operation and maintenance costs Flexible resource scheduling cost The costs of curtailing wind and solar power and the cost of purchasing electricity As shown below:

[0095]

[0096] In the formula, r is the number of a typical day in each season; πr is the percentage of each typical day in a year. and These are the photovoltaic operation and maintenance cost coefficient, wind turbine operation and maintenance cost coefficient, energy storage operation and maintenance cost coefficient, cluster EV dispatch cost coefficient, cluster HVAC dispatch cost coefficient, curtailment penalty cost coefficient, and electricity purchase price; These are typical daily photovoltaic power, wind power, energy storage charging power, and energy storage discharging power, respectively. and These represent typical daily wind and solar curtailment amounts; This represents the electricity purchased from the distribution network; Δt is the time interval.

[0097] Microgrids face uncertainties in actual operation, such as distributed wind and solar power output, load demand, and flexibility resources. The planning model needs to consider the impact of source-load uncertainty. A box-type uncertainty set is constructed to model multiple types of uncertain resources, as shown below:

[0098]

[0099] In the formula, T represents the planning time period; and These represent the uncertain power outputs of EVs, HVAC systems, photovoltaic power, wind power, and conventional loads, respectively. and These are the predicted power values ​​for each unit; and These represent the maximum power fluctuation deviations of each unit; and These are the actual power fluctuation coefficients for each unit. The values ​​can be selected according to the actual scenario. The larger the value, the greater the error between the actual value and the predicted value, and the more conservative the optimization scheme. Conversely, the smaller the value, the smaller the error, and the closer the scheme is to the deterministic model result.

[0100] The conservatism of uncertainty is controlled using an uncertain budget, with the following constraints:

[0101]

[0102] In the formula, Γ EVVB ,Γ HVVB ,Γ PV ,Γ WT and Γ L This represents the uncertain budget for the corresponding unit.

[0103] The constraints for setting the planning model mainly include distributed renewable energy planning constraints, energy storage planning constraints, planning investment constraints, microgrid power balance constraints, distributed renewable energy operation constraints, and load-side flexibility resource constraints.

[0104] The planning constraints for distributed renewable energy are as follows:

[0105]

[0106] In the formula, This represents the maximum allowable installed capacity of distributed wind turbine units. This represents the maximum capacity that can be installed in a distributed photovoltaic (PV) system.

[0107] Energy storage planning constraints are as follows:

[0108]

[0109] In the formula, t ES The energy storage duration at rated power is set to 4 hours.

[0110] The planned investment constraints are as follows:

[0111]

[0112] In the formula, This is the upper limit for investment.

[0113] The power balance constraints of microgrids are shown below:

[0114]

[0115] In the formula, and These refer to the charging and discharging power of flexible resources; and These are the energy storage charging and discharging power, respectively. These represent typical daily photovoltaic and wind power outputs, respectively. For the purchase of electricity from the distribution network; Electricity sales volume; This is the power of a normal load. This refers to the amount of wind and solar power curtailment. Specifically, the interaction power between the distribution substation microgrid and the distribution network must meet the following requirements:

[0116]

[0117] In the formula, These are 0-1 variables, representing electricity sales and electricity purchases; This represents the maximum interactive power of the tie line.

[0118] The operational constraints of distributed renewable energy sources are as follows:

[0119]

[0120]

[0121] In the formula, and These represent typical daily photovoltaic and wind power outputs, respectively. and These represent the uncertain power output per unit of photovoltaic power, wind power output, and conventional load, respectively.

[0122] Load-side flexibility resource constraints:

[0123] Cluster electric vehicle constraints:

[0124]

[0125] In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These represent the upper and lower power limits of the cluster EV, respectively. These represent the upper and lower limits of the energy consumption of the cluster EVs, respectively. This refers to the actual power of the cluster EVs; Let be the electrical energy of the cluster EV at time t.

[0126] Cluster HVAC constraints:

[0127]

[0128] In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These are the upper and lower power limits for the cluster HVAC system, respectively. and These are the upper and lower limits of power consumption for the cluster HVAC system; This represents the actual power of the cluster HVAC system. Let represent the electrical energy of the cluster HVAC system at time t.

[0129] Step C: The nested column-and-constraint generation (NC&CG) algorithm is used to solve the robust collaborative planning model of the distribution microgrid that takes into account the flexibility of load-side resources.

[0130] Figure 3 The flowchart illustrates the algorithm for solving a robust collaborative planning model for a distribution microgrid that considers load-side resource flexibility. Based on source-load uncertainties, the objective function is reconstructed into a two-stage robust collaborative planning model. The planning stage aims to minimize the annual equivalent investment cost of the microgrid; the operation stage aims to minimize the average annual operating cost of the system.

[0131]

[0132] In the formula, x is a continuous variable vector in the planning stage, representing the planned capacity of each unit; y is a continuous variable vector in the operation stage, representing the actual power of each unit in the system; z is a decision variable, describing the operating status of each unit; and u is an uncertainty variable, representing the output fluctuation of each unit.

[0133] The model is solved using a nested column and constraint generation (NC&CG) algorithm, which includes an inner and an outer NC&CG loop. The original problem is decomposed into a main problem (MP) and subproblems (SP). Solving the MP yields a feasible solution, which is then substituted into the SP for worst-case scenario testing, ultimately leading to the globally optimal solution. The model is first transformed into the following compact form:

[0134]

[0135] In the formula, a, b, p, C, D1, D2, D3, D4, E, F, H3, H4, c, g1, and g2 represent constant coefficient matrices or vectors.

[0136] Decomposing equation (29) yields the first-stage main problem (MP). o ) and the second-stage max-min subproblem (SP o MP o Introduce an auxiliary variable γ, and given a known variable u, solve for MP. o We can obtain the lower bound L for the planning variable x and the optimal value of the original problem. o Substitute x into SP o The upper bound U of the variable u and the optimal value of the original problem can be obtained. o Then return u to MP o Then, new variables y and z are added, and the solution is performed alternately by first relaxing and then probing, forming an outer C&CG loop, as follows:

[0137]

[0138] In the formula, * indicates that the variable is known; k m k is the current iteration number of the outer loop. o This represents the number of iterations in the outer loop's history. For the kth m Subsequent cycle SP o The known scene variables obtained by solving; and For the kth m The new variable added in the next iteration; x * To solve MP o The first-stage variables obtained, λ1, λ2, λ3, and λ4, are dual variables.

[0139] As the outer iterations proceed, each solution to SP... o The obtained scenario variable u represents the worst-case scenario. Return u as the MP value. o And add new variables and constraints, the lower bound L of the original problem's optimal value o It will get bigger and bigger, and at the same time U oIt gets smaller and smaller. This iterative solution continues until the convergence criterion (33) is satisfied, at which point the loop exits and returns the optimal solution.

[0140]

[0141] In the formula, κ is a very small positive real number.

[0142] SP o This belongs to a bilevel programming model, where the inner optimization problem (min) contains discrete decision variables, causing its constraints to not satisfy the Karush-Kuhn-Tucker (KKT) conditions. Further, SP... o Decomposed into MP i and SP i MP i Introduce an auxiliary variable γ2, and given a 0-1 variable z, solve for MP. i Get SP o The upper bound U of the optimal value i Substitute u into SP o Solving for variables i and SP yields the results. o The lower bound L of the optimal value i i returns MP i Then, new variables are added, and the solution is performed alternately, first relaxing and then probing, forming an inner C&CG loop, as follows:

[0143]

[0144] In the formula, γ2 is an auxiliary variable introduced; k n and k i These represent the current and historical iteration counts of the inner loop, respectively. For the Lagrange function with respect to k n Find the expression for the partial derivative; M is a maximal positive real number; and These are the 0-1 variables introduced during the KKT linearization process.

[0145] As the inner iterations proceed, each solution to SP... i The obtained state variable i is retained and MP is returned. i At the same time, new variables, constraints, and cutting planes are added, therefore U i It will get smaller and smaller, and at the same time L i It will grow larger and larger. Given the convergence criterion of the iteration as shown in equation (36), if the criterion is satisfied, the loop will exit and the optimal solution will be returned.

[0146]

[0147] This embodiment was tested on a microgrid system in a specific distribution area. The results show that the proposed robust collaborative planning method for distributed generation and energy storage in a distribution microgrid, which considers the flexibility of load-side resources, can not only fully tap the flexible adjustment potential of load-side resources and significantly reduce the scale of energy storage configuration, but also ensure the economical and efficient operation of the microgrid.

[0148] This embodiment uses Figure 4 Taking the microgrid system shown as an example, the effectiveness of the proposed robust source-storage collaborative planning method considering load-side resource flexibility is verified. The system is configured with 200 EVs (electric vehicles) with a charging / discharging power of 7kW and a maximum capacity of 700kWh; 300 HVAC units with a charging / discharging power of 2kW; and an air conditioning temperature range of [16, 28]℃. The energy storage model and EV state of charge range are both set to [0.15, 0.85]. Other relevant parameters in the microgrid planning process are shown in Table 1. The proposed model is solved using MATLAB 2019b version and the YALMIP toolbox by calling the CPLEX solver.

[0149] Table 1 Microgrid Planning Parameters

[0150]

[0151] To verify the effectiveness of the proposed collaborative planning method, this embodiment sets up different scenarios for comparative analysis: Scenario 1 is a two-stage collaborative planning method scenario; Scenario 2 does not consider the aggregation of load-side flexibility resources, but considers source-load uncertainty and time-of-use pricing; Scenario 3 does not consider source-load uncertainty, i.e., the model adopts a deterministic method, but considers the aggregation of flexibility resources and time-of-use pricing; Scenario 4 does not consider time-of-use pricing, i.e., a uniform pricing method is used, but considers the aggregation of flexibility resources and source-load uncertainty. Specific settings are shown in Table 2.

[0152] Table 2. Case Study Scenario Settings

[0153]

[0154] Based on the different scenarios set in Table 1, the planning results obtained by solving the model under different scenarios are shown in Table 3. A comparison of the indicators for scenario 1 and scenario 2 is also provided. Figure 5 As shown, the comparison of various indicators between scenario 1 and scenario 3 is as follows: Figure 6 As shown, the total load curves of the microgrid system considering time-of-use pricing in scenarios 1 and 4 are as follows: Figure 7 As shown.

[0155] Table 3 Planning results under different scenarios

[0156]

[0157] The results lead to the following conclusions: By considering the adjustment potential of diverse and flexible resources on the load side in the planning of the microgrid in the distribution area, the planned energy storage capacity was reduced by 480 kWh, and the average annual total cost was reduced by RMB 567,372. By using a generalized energy storage method to aggregate EVs and HVACs, the planned renewable energy consumption rate increased by 2.5%, and the costs of wind and solar curtailment and electricity purchase decreased by 10.5% and 18.5%, respectively. Considering the uncertainty of source and load and the impact of time-of-use pricing, the power supply reliability increased by 4.6%, and the average annual total cost was reduced by RMB 32,000. Introducing time-of-use pricing further reduced the cost by RMB 54,500.

[0158] The results show that the method of the present invention can not only fully tap the flexible adjustment potential of load-side resources and significantly reduce the scale of energy storage configuration, but also ensure the economical and efficient operation of microgrids.

Claims

1. A robust collaborative planning method for distributed generation and energy storage in a transformer substation microgrid, considering load-side resource flexibility, characterized in that... include: Considering the energy storage characteristics of load-side electric vehicles and HVAC systems, a generalized energy storage aggregation model for clustered electric vehicles and a generalized energy storage aggregation model for HVAC systems are established. Considering the impact of source-load uncertainty, a robust collaborative planning model for the distribution microgrid, taking into account the resource flexibility of the load side, is established based on the generalized energy storage aggregation model of clustered electric vehicles and the generalized energy storage aggregation model of HVAC. A nested column and constraint generation algorithm is used to solve the robust collaborative planning model of a distribution microgrid that takes into account the resource flexibility of the load side.

2. The robust collaborative planning method for distributed generation and energy storage in a microgrid considering load-side resource flexibility as described in claim 1, characterized in that, The generalized energy storage aggregation model for the clustered electric vehicles is as follows: In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These represent the upper and lower power limits of the cluster EV, respectively. These represent the upper and lower limits of the energy consumption of the cluster EVs, respectively. This refers to the actual power of the cluster EVs; Let be the electrical energy of the cluster EV at time t; Δt represents the change in the electrical energy of the cluster EVs when EVs are connected to or disconnected from the grid; Δt is the time interval.

3. The robust collaborative planning method for distributed generation and energy storage in a distribution microgrid considering load-side resource flexibility as described in claim 1 or 2, characterized in that, The HVAC generalized energy storage aggregation model is as follows: In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These are the upper and lower power limits for the cluster HVAC system, respectively. and These are the upper and lower limits of power consumption for the cluster HVAC system; This represents the actual power of the cluster HVAC system. Let be the electrical energy of the cluster HVAC system at time t; β is the dissipation coefficient; Δt is the time interval.

4. The robust collaborative planning method for distributed generation and energy storage in a microgrid considering load-side resource flexibility as described in claim 1, characterized in that, The robust collaborative planning model for distribution microgrids that takes into account the flexibility of load-side resources includes the objective function of the planning model, sub-models for multiple types of uncertain resources, and constraints of the planning model.

5. The robust collaborative planning method for distributed generation and energy storage in a microgrid considering load-side resource flexibility as described in claim 4, characterized in that, The objective function of the planning model is: minC total =C inv +C op In the formula, C total C represents the average annual total cost of the microgrid in the entire planning period of the distribution area. inv and C op These are the equivalent annual investment cost and the average annual operating cost of the system, respectively. in: In the formula, For distributed wind power investment costs, For distributed photovoltaic investment costs, Let z be the energy storage investment cost, and z be the discount rate; y PV y WT and y ES These are the investment periods for photovoltaic, wind power, and energy storage, respectively. and These are the unit investment cost coefficients for photovoltaic, wind power, and energy storage, respectively. and These are the configuration capacities for photovoltaic, wind power, and energy storage, respectively. In the formula, For the operation and maintenance costs of wind and solar generator sets, For energy storage operation and maintenance costs, For flexible resource scheduling costs, The costs of curtailing wind and solar power and the cost of purchasing electricity are represented by t and r, respectively, which are the time period number and the typical day number for each season; π r This represents the percentage of each typical day out of the total number of days in a year. and These are the photovoltaic operation and maintenance cost coefficient, wind turbine operation and maintenance cost coefficient, energy storage operation and maintenance cost coefficient, cluster EV dispatch cost coefficient, cluster HVAC dispatch cost coefficient, curtailment penalty cost coefficient, and electricity purchase price; These are typical daily photovoltaic power, wind power, energy storage charging power, and energy storage discharging power, respectively. and These represent typical daily wind and solar curtailment amounts; This represents the electricity purchased from the distribution network; Δt is the time interval.

6. The robust collaborative planning method for distributed generation and energy storage in a distribution microgrid considering load-side resource flexibility as described in claim 4, characterized in that, The multi-type uncertainty resource sub-model includes: In the formula, and These represent the uncertain power outputs of EVs, HVAC systems, photovoltaic power, wind power, and conventional loads, respectively. and These are the predicted power values ​​for each unit; and These represent the maximum power fluctuation deviations of each unit; and These are the actual power fluctuation coefficients for each unit; t and r are the time period number and the typical day number for each season, respectively; T is the planned time period. Sub-model constraints: In the formula, Γ EVVB ,Γ HVVB ,Γ PV ,Γ WT and Γ L This represents the uncertain budget for the corresponding unit.

7. The robust collaborative planning method for distributed generation and energy storage in a microgrid considering load-side resource flexibility as described in claim 4, characterized in that, The constraints of the planning model include: Constraints on Distributed New Energy Planning: In the formula, This represents the maximum allowable installed capacity of distributed wind turbine units. This refers to the maximum allowable installed capacity of distributed photovoltaic (PV) units. Energy storage planning constraints: In the formula, t ES Energy storage duration at rated power; Configuration capacity for energy storage; Planning and investment constraints: In the formula, Investment ceiling; C inv The cost of investment is the equivalent annual value. Microgrid power balance constraints: In the formula, and These refer to the charging and discharging power of flexible resources; and These are the energy storage charging and discharging power, respectively. These represent typical daily photovoltaic and wind power outputs, respectively. For the purchase of electricity from the distribution network; Electricity sales volume; This is the power of a normal load. This refers to the amount of wind and solar power curtailed; among which, the interaction power between the distribution substation microgrid and the distribution network must meet the following requirements: In the formula, To purchase electricity from the distribution network; These are 0-1 variables, representing electricity sales and electricity purchases; This represents the maximum cross-connection power. Operational constraints of distributed renewable energy sources: In the formula, and These represent typical daily photovoltaic and wind power outputs, respectively. and These represent the uncertain power output per unit of photovoltaic power, wind power output, and conventional load, respectively. These are the configuration capacities for photovoltaic and wind power, respectively. Load-side flexibility resource constraints: In the formula, t and r are the time period number and the typical day number for each of the four seasons, respectively; These represent the upper and lower power limits of the cluster EV, respectively. These represent the upper and lower limits of the energy consumption of the cluster EVs, respectively. This refers to the actual power of the cluster EVs; Let be the electrical energy of the cluster EV at time t; In the formula, These are the upper and lower power limits for the cluster HVAC system, respectively. and These are the upper and lower limits of power consumption for the cluster HVAC system; This represents the actual power of the cluster HVAC system. Let represent the electrical energy of the cluster HVAC system at time t.

Images