Flexible resource-oriented double-layer global optimization configuration method

Abstract

The invention relates to a flexibility resource-oriented double-layer global optimization configuration method, which comprises the following steps of: establishing an optimization model comprising a planning layer and an operation layer, and taking annual minimum total cost as an optimization target by the optimization model; and solving the optimization model by adopting a simulated annealing algorithm, and outputting to obtain an optimal flexibility resource allocation scheme. Wherein the planning layer considers the investment quantity of resources with different flexibility and calculates the investment cost and the operation and maintenance cost of the resources; the operation layer considers the voltage constraint, the line load constraint and the DG installation capacity constraint, and calculates the power purchase cost, the network loss cost and the light abandoning cost of the microgrid. Compared with the prior art, the method has the advantages that on the basis of the distributed resource-containing double-layer optimization configuration model based on the PSO algorithm, the optimization model is solved by adopting the simulated annealing algorithm, so that the optimization model can be prevented from falling into local optimum as far as possible when the decision space is relatively large, and the optimal flexible resource configuration under the complex condition is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network planning, and in particular to a double-layer global optimization configuration method for flexibility resources. Background Art

[0002] At present, in the application field of distributed generation distribution network planning, most planning schemes use genetic algorithms or PSO (Particle swarm optimization) algorithms. These two algorithms perform well in solving such integer non-convex optimization problems. However, with the increase in the number of grid nodes and the richness of the types of flexible resources (mainly wind and solar energy) that can be configured in different nodes, the probability of these two models falling into the local optimal dilemma increases, which brings difficulties to the optimization problem.

[0003] On the one hand, the existing technology proposes a flexible resource operation-planning joint optimization two-layer configuration model to address the flexibility issues brought about by high-proportion flexible resources, and adds a variety of flexible resource planning schemes to the IEEE 33bus model, effectively reducing network losses and system wind and solar power abandonment rates, and improving system economy. In addition, some studies take into account the timing characteristics of distributed resource output and grid load demand, and obtain planning schemes that are more in line with actual conditions through timing simulation. However, when the decision space increases, PSO and genetic algorithms often fall into local optimal situations, making it difficult to achieve the optimal configuration of flexible resources under complex circumstances. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a two-layer global optimization configuration method for flexible resources, so as to achieve the optimal configuration of flexible resources in a larger decision space and avoid the local optimal dilemma.

[0005] The object of the present invention can be achieved by the following technical solution: A two-layer global optimization configuration method for flexibility resources, comprising the following steps:

[0006] S1. Establish an optimization model including planning layer and operation layer, and the optimization model takes the annualized minimum total cost as the optimization target;

[0007] S2. Use simulated annealing algorithm to solve the optimization model and output the optimal flexibility resource allocation plan.

[0008] Furthermore, in step S1, the planning layer considers the investment amount of different flexibility resources and calculates their investment cost and operation and maintenance cost;

[0009] The operation layer considers voltage constraints, line load constraints, and DG (Distributed Generation) installation capacity constraints, and calculates the microgrid's electricity purchase cost, network loss cost, and abandoned light cost.

[0010] Furthermore, the voltage constraints include power flow constraints and voltage upper and lower limit constraints, and the line load constraints include line power constraints and DG equipment capacity constraints.

[0011] Furthermore, the power flow constraint is specifically:

[0012]

[0013] Among them, P i is the active power of node i, Q i is the reactive power of node i, U i is the voltage amplitude of node i, g ij is the real part of the node admittance matrix, b ij is the imaginary part of the node admittance matrix, θ ij is the phase angle of the node voltage, j∈i represents all nodes j connected to node i;

[0014] The voltage upper and lower limit constraints are specifically:

[0015]

[0016] in, represents the lower voltage limit of node i, represents the voltage upper limit of node i, and S is the set of all nodes.

[0017] Furthermore, the line power constraint is specifically:

[0018]

[0019] Among them, S i represents the apparent power passing through line i, represents the upper limit of the power capacity of line i, and T is the set of all lines;

[0020] The DG equipment capacity constraints are specifically:

[0021]

[0022] Among them, S DG It is a collection of available nodes of DG devices. represents the DG equipment capacity of node i at time t, Fit the timing characteristic curves of all load nodes in the system at time t.

[0023] Furthermore, the DG installation capacity constraint is specifically:

[0024]

[0025] in, is the total capacity of node i accessing WG, is the total capacity of the PV connected to node i, The maximum capacity of the WG connected to the same node, is the maximum capacity of the PV connected to the same node, S WG is the set of nodes connected to WG, S PV A collection of nodes connected to the PV.

[0026] Furthermore, the objective function of the optimization model is specifically:

[0027] min C=C inv +C ope +C loss +C pur +C cur

[0028] Where C is the total annual cost of DG equipment, C inv is the DG investment cost, C o[e is the operation and maintenance cost, C loss is the network loss cost, C pur is the electricity purchase cost, C cur The cost of abandoned light.

[0029] Furthermore, the DG investment cost is specifically:

[0030]

[0031] in, is the unit investment cost of node i connecting to DG device k, is the number of DG devices k connected to node i;

[0032] The operation and maintenance costs are specifically:

[0033]

[0034] in, is the unit operation and maintenance cost of node i connected to DG device k at time t.

[0035] Furthermore, the network loss cost is specifically:

[0036]

[0037] Among them, c loss is the unit cost of network loss, I i,s,tis the current of line i at time t under scenario s, r i is the resistance of line i;

[0038] The electricity purchase cost is specifically:

[0039]

[0040] Among them, c pur is the unit electricity purchase cost, The amount of electricity purchased by the main network at time t in scenario s;

[0041] The cost of abandoned light is specifically:

[0042]

[0043] Among them, c cur is the unit power curtailment cost, is the amount of power abandoned at time t in scenario s.

[0044] Furthermore, the step S2 specifically includes the following steps:

[0045] 01) Input grid data and define the initial temperature, target temperature, cooling rate, and maximum number of new solution explorations of the simulated annealing algorithm;

[0046] 02) Define the initial solution of flexibility resources;

[0047] 03) Determine whether the current temperature is higher than the target temperature. If so, execute step 04). If not, output the optimal solution.

[0048] 04) Determine whether the number of cycles for finding a solution at the current temperature is greater than the set maximum number of new solution explorations. If so, execute step 10); if not, execute step 05);

[0049] 05) Randomly try a new solution;

[0050] 06) Calculate the investment cost and operation and maintenance cost of the new solution;

[0051] 07) Determine whether the new solution is feasible in terms of timing. If so, execute step 08). If not, return to execute step 04);

[0052] 08) Calculate the new network loss cost, power purchase cost and abandoned light cost;

[0053] 09) Determine whether the total cost of the new solution is less than the total cost corresponding to the current optimal solution, or determine whether the total cost of the new solution meets the Metropolis condition. If so, update the optimal solution, otherwise execute step 10);

[0054] 10) Update temperature;

[0055] 11) Repeat steps 03) to 10) until the temperature is less than or equal to the target temperature.

[0056] Compared with the prior art, the present invention has the following advantages:

[0057] The present invention constructs an optimization model including a planning layer and an operation layer, wherein the operation layer considers voltage constraints, line load constraints, and DG installation capacity constraints, and calculates the power purchase cost, network loss cost, and abandoned light cost of the microgrid; the planning layer considers the investment quantity of different flexibility resources, and calculates its investment cost and operation and maintenance cost; the optimization model takes the annualized minimum total cost as the optimization target, and solves and calculates it through a simulated annealing algorithm. Every time a new cycle starts, the planning layer generates a new trial solution and calculates the corresponding investment and operation and maintenance costs. Thereafter, the trial solution enters the operation layer. The operation layer tests whether the trial solution is feasible in time according to the time sequence output of the flexibility resources. If feasible, the relevant costs are calculated, and the solution is returned to the planning layer to determine whether it is the current optimal solution. This can avoid the optimization model from falling into local optimality as much as possible when the decision space is large, and achieve the optimal flexibility resource configuration under complex conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0059] Figure 2 This is a schematic diagram of the process of solving the optimization model in Example 2. DETAILED DESCRIPTION

[0060] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0061] Embodiment 1

[0062] like Figure 1 As shown, a two-layer global optimization configuration method for flexible resources includes the following steps:

[0063] S1. Establish an optimization model including planning layer and operation layer, and the optimization model takes the annualized minimum total cost as the optimization target;

[0064] S2. Use simulated annealing algorithm to solve the optimization model and output the optimal flexibility resource allocation plan.

[0065] This embodiment constructs an optimization model, the main contents of which are:

[0066] The operation layer considers voltage constraints, line load constraints, and DG installation capacity constraints, calculates the microgrid's electricity purchase cost, network loss cost, and abandoned light cost, and strives to minimize the sum of the three; the planning layer considers the investment amount of different flexibility resources and calculates their investment cost and operation and maintenance cost;

[0067] Take the annual minimum total cost as the optimization goal and optimize the configuration of flexibility resources.

[0068] The objective function is as follows:

[0069] min C=C inv +C ope +C loss +C pur +C cur

[0070] Where C is the total annual cost of DG equipment, C inv is the DG investment cost, C ope is the operation and maintenance cost, C losx is the network loss cost, C pur is the electricity purchase cost, C cur The cost of abandoned light.

[0071] ①DG investment cost

[0072] The DG investment costs are as follows:

[0073]

[0074] in, is the unit investment cost of node i connecting to DG device k, is the number of DG devices k connected to node i.

[0075] ②Operation and maintenance costs

[0076] The operation and maintenance costs are as follows:

[0077]

[0078] in, is the unit operation and maintenance cost of node i connected to DG device k at time t.

[0079] ③ Network loss cost

[0080] The network loss cost is as follows:

[0081]

[0082] Among them, c loss is the unit cost of network loss, I i,s,t is the current of line i at time t under scenario s, r i is the line resistance.

[0083] ④ Electricity purchase cost

[0084] The cost of purchasing electricity is as follows:

[0085]

[0086] Among them, c pur is the unit electricity purchase cost, is the amount of electricity purchased by the main grid at time t in scenario s.

[0087] ⑤ Cost of abandoned light

[0088] The cost of abandoning light is as follows:

[0089]

[0090] Among them, c cur is the unit power curtailment cost, is the amount of power abandoned at time t in scenario s.

[0091] The constraints are as follows:

[0092] ① Flow constraints

[0093]

[0094] Among them, P i is the active power of node i, Q i is the reactive power of node i, U i is the voltage amplitude of node i, g ij is the real part of the node admittance matrix, b ij is the imaginary part of the node admittance matrix, θ ij is the phase angle of the node voltage, j∈i represents all nodes j connected to node i.

[0095] ② Voltage upper and lower limit constraints

[0096]

[0097] in, represents the lower voltage limit of node i, represents the voltage upper limit of node i, and S is the set of all nodes.

[0098] The above power flow constraints and voltage upper and lower limit constraints together constitute voltage constraints.

[0099] ③ Line power constraints

[0100]

[0101] Among them, S i represents the apparent power passing through line i, represents the upper limit of the power capacity of line i, and T is the set of all lines.

[0102] ④DG equipment capacity constraints

[0103]

[0104] Among them, S DG It is a collection of available nodes of DG devices. represents the DG equipment capacity of node i at time t, Fit the timing characteristic curves of all load nodes in the system at time t.

[0105] The above line power constraints and DG equipment capacity constraints together constitute the line load constraints.

[0106] ⑤DG installation capacity constraints

[0107] Although WG (wind power) and PV (photovoltaic) DG devices have a good balancing effect on distributed power generation and less environmental pollution, too many connections at the same node will affect its safety and stability, so this constraint is set.

[0108]

[0109] in, is the total capacity of node i accessing WG, is the total capacity of the PV connected to node i, The maximum capacity of the WG connected to the same node, is the maximum capacity of the PV connected to the same node, S WG is the set of nodes connected to WG, S PV A collection of nodes connected to the PV.

[0110] Embodiment 2

[0111] This embodiment designs a program product, which is used to implement the method steps described in the first embodiment when executed by a computer, such as Figure 2 As shown, the program product uses a simulated annealing algorithm to solve the optimization model. The specific process includes:

[0112] 01) Input grid data and define the initial temperature, target temperature, cooling rate, and maximum number of new solution explorations of the simulated annealing algorithm;

[0113] 02) Define the initial solution of flexibility resources;

[0114] 03) Determine whether the current temperature is higher than the target temperature. If so, execute step 04). If not, output the optimal solution.

[0115] 04) Determine whether the number of cycles for finding a solution at the current temperature is greater than the set maximum number of new solution explorations. If so, execute step 10); if not, execute step 05);

[0116] 05) Randomly try a new solution;

[0117] 06) Calculate the investment cost and operation and maintenance cost of the new solution;

[0118] 07) Determine whether the new solution is feasible in terms of timing. If so, execute step 08). If not, return to execute step 04);

[0119] 08) Calculate the new network loss cost, power purchase cost and abandoned light cost;

[0120] 09) Determine whether the total cost of the new solution is less than the total cost corresponding to the current optimal solution, or determine whether the total cost of the new solution meets the Metropolis condition. If so, update the optimal solution, otherwise execute step 10);

[0121] 10) Update temperature;

[0122] 11) Repeat steps 03) to 10) until the temperature is less than or equal to the target temperature.

[0123] In summary, this scheme proposes a two-layer optimization configuration scheme for flexible resources based on the simulated annealing algorithm. On the basis of the two-layer optimization configuration model containing distributed resources based on the PSO algorithm, the simulated annealing algorithm is adopted. It can avoid the optimization model from falling into the local optimum as much as possible when the decision space is large, and realize the optimal flexible resource configuration in complex situations.

Claims

1. A two-layer global optimization configuration method for flexibility resources, characterized in that: The following steps are involved: S1. Establish an optimization model including planning layer and operation layer, and the optimization model takes the annualized minimum total cost as the optimization target; S2. Use simulated annealing algorithm to solve the optimization model and output the optimal flexibility resource allocation plan.

2. A two-layer global optimization configuration method for flexible resources according to claim 1, characterized in that: In step S1, the planning layer considers the investment amount of different flexibility resources and calculates their investment cost and operation and maintenance cost; The operation layer considers voltage constraints, line load constraints, and DG installation capacity constraints, and calculates the microgrid's electricity purchase cost, network loss cost, and abandoned light cost.

3. A two-layer global optimization configuration method for flexible resources according to claim 2, characterized in that: The voltage constraints include power flow constraints and voltage upper and lower limit constraints, and the line load constraints include line power constraints and DG equipment capacity constraints.

4. A two-layer global optimization configuration method for flexible resources according to claim 3, characterized in that: The power flow constraints are specifically: Among them, P i is the active power of node i, Q i is the reactive power of node i, U i is the voltage amplitude of node i, g ij is the real part of the node admittance matrix, b ij is the imaginary part of the node admittance matrix, θ ij is the phase angle of the node voltage, j∈i represents all nodes j connected to node i; The voltage upper and lower limit constraints are specifically: in, represents the lower voltage limit of node i, represents the voltage upper limit of node i, and S is the set of all nodes.

5. A two-layer global optimization configuration method for flexible resources according to claim 3, characterized in that: The line power constraint is specifically: Among them, S i represents the apparent power passing through line i, represents the upper limit of the power capacity of line i, and T is the set of all lines; The DG equipment capacity constraints are specifically: Among them, S DG It is a collection of available nodes of DG devices. represents the DG equipment capacity of node i at time t, Fit the timing characteristic curves of all load nodes in the system at time t.

6. A two-layer global optimization configuration method for flexible resources according to claim 3, characterized in that: The DG installation capacity constraints are specifically: in, is the total capacity of node i accessing WG, is the total capacity of the PV connected to node i, The maximum capacity of the WG connected to the same node, is the maximum capacity of the PV connected to the same node, S WG is the set of nodes connected to WG, S PV A collection of nodes connected to the PV.

7. A two-layer global optimization configuration method for flexible resources according to claim 2, characterized in that: The objective function of the optimization model is specifically: min C=C inv +C ope +C loss +C pur +C cur Where C is the total annual cost of DG equipment, C inv is the DG investment cost, C ope is the operation and maintenance cost, C loss is the network loss cost, C pur is the electricity purchase cost, C cur The cost of abandoned light.

8. A two-layer global optimization configuration method for flexible resources according to claim 7, characterized in that: The DG investment cost is specifically: in, is the unit investment cost of node i connecting to DG device k, is the number of DG devices k connected to node i; The operation and maintenance costs are specifically: in, is the unit operation and maintenance cost of node i connected to DG device k at time t.

9. A two-layer global optimization configuration method for flexible resources according to claim 7, characterized in that: The network loss cost is specifically: Among them, c loss is the unit cost of network loss, I i,s,t is the current of line i at time t under scenario s, r i is the resistance of line i; The electricity purchase cost is specifically: Among them, c pur is the unit electricity purchase cost, The amount of electricity purchased by the main network at time t in scenario s; The cost of abandoned light is specifically: Among them, c cur is the unit power curtailment cost, is the amount of power abandoned at time t in scenario s.

10. A two-layer global optimization configuration method for flexible resources according to claim 2, characterized in that: The step S2 specifically includes the following steps: 01) Input grid data and define the initial temperature, target temperature, cooling rate, and maximum number of new solution explorations of the simulated annealing algorithm; 02) Define the initial solution of flexibility resources; 03) Determine whether the current temperature is higher than the target temperature. If so, execute step 04). If not, output the optimal solution. 04) Determine whether the number of cycles for finding a solution at the current temperature is greater than the set maximum number of new solution explorations. If so, execute step 10); if not, execute step 05); 05) Randomly try a new solution; 06) Calculate the investment cost and operation and maintenance cost of the new solution; 07) Determine whether the new solution is feasible in terms of timing. If so, execute step 08). If not, return to execute step 04); 08) Calculate the new network loss cost, power purchase cost and abandoned light cost; 09) Determine whether the total cost of the new solution is less than the total cost corresponding to the current optimal solution, or determine whether the total cost of the new solution meets the Metropolis condition. If so, update the optimal solution, otherwise execute step 10); 10) Update temperature; 11) Repeat steps 03) to 10) until the temperature is less than or equal to the target temperature.

Images