Energy storage optimal configuration, regulation and control method and device

Abstract

The invention discloses an energy storage optimal configuration and regulation and control method. The method comprises the following steps: acquiring photovoltaic output, load output, energy storage selection position, capacity and charge and discharge power and time scale of energy storage; the time sequence of light output and the photovoltaic permeability factor are considered, an energy storage double-layer optimization configuration and operation model is established, the upper layer objective function of the model is the minimum energy storage configuration capacity, and the lower layer objective function is the power distribution network vulnerability and the minimum active power loss; determining constraint conditions of the energy storage double-layer optimization configuration and operation model; performing energy storage position configuration and capacity configuration on the upper-layer model by using a COA algorithm, performing energy storage charging and discharging power configuration on the lower-layer model by using a particle swarm-gravitational search hybrid algorithm PSO-GSA, and minimizing a target function of the lower layer through optimal configuration on the upper layer; according to the method, the characteristics of distributed energy output intermittency, volatility and randomness can be effectively relieved, and the photovoltaic consumption capability can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system regulation and control, and more specifically, to a method and device for optimizing energy storage configuration and regulation. Background Art

[0002] Distributed photovoltaic points are numerous and widespread, with strong randomness and volatility. After being connected to the distribution network in large numbers, a series of problems such as reverse overload of transformers and voltage over-limit will occur. On the other hand, the charging load of urban electric vehicles, the load of rural residents returning home, the load of tourism, and the seasonal load have put forward higher requirements on the carrying capacity of the distribution network. Distributed energy storage has attracted much attention as an effective technical means to smooth out the volatile source load. However, the current construction and development of distributed energy storage still has the following pain points: First, there is a lack of distributed energy storage optimization configuration methods for distributed photovoltaic consumption, economic operation of distribution networks, and high-quality and reliable power supply; second, the efficient regulation technology of distributed energy storage needs to be improved. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, the present invention provides a two-layer optimization configuration and operation control of energy storage based on the COA algorithm and the PSO-GSA algorithm, and considers the timing of light output and photovoltaic penetration factors at the same time, and establishes a two-layer mathematical model. The upper objective function is to minimize the energy storage configuration capacity, and the lower objective function is to minimize the vulnerability of the distribution network and the active network loss. By optimizing the configuration of the upper layer, the objective function of the lower layer is minimized, thereby improving the reliability and stability of the distribution network operation, effectively reducing the vulnerability of the distribution network and the active network loss, and promoting the high-proportion penetration photovoltaic consumption.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for optimizing energy storage configuration and regulation comprises the following steps:

[0006] Obtain photovoltaic output, load output, energy storage selection location, capacity, and energy storage charging and discharging power and time scale;

[0007] Considering the timing of solar output and photovoltaic penetration factors, a two-layer optimization configuration and operation model of energy storage is established. The upper layer objective function of the model is to minimize the energy storage configuration capacity, and the lower layer objective function is to minimize the vulnerability of the distribution network and the active network loss.

[0008] Determining the constraint conditions of the energy storage dual-layer optimization configuration and operation model, wherein the constraint conditions include power balance constraint, power flow constraint, distributed new energy output constraint, and energy storage constraint;

[0009] Use the COA algorithm to configure the energy storage location and capacity for the upper-layer model, and use the particle swarm-gravitational search hybrid algorithm PSO-GSA to configure the energy storage charge and discharge power for the lower-layer model. By optimizing the upper-layer configuration, the objective function of the lower layer is minimized, thereby improving the operation reliability and stability of the distribution network.

[0010] Furthermore, the method of using the COA optimization algorithm to configure the energy storage location and capacity for the upper-layer model, using the particle swarm-gravitational search hybrid algorithm PSO-GSA to configure the energy storage charge and discharge power for the lower-layer model, and minimizing the objective function of the lower layer through the optimization of the upper layer to improve the operation reliability and stability of the distribution network is as follows:

[0011] S1. Set the population size N and the number of iterations T of the COA optimization algorithm. COA is a two-stage algorithm, the first stage is P1, and the second stage is P2. Set the population size N 2 and the number of iterations T 2 of the particle swarm-gravitational search hybrid algorithm PSO-GSA, and set the individual learning factor C1 and the global learning factor C2; initialize the iteration number t = 1 of the COA algorithm.

[0012] S2. The decision variables of the upper-layer model are the installation location and capacity of the energy storage, and the objective function is to minimize the configured capacity of the energy storage. Initialize the population size i = 1 of the COA optimization algorithm. In the first stage P1 of COA, when i < N / 2, calculate the position of the i-th COA, update the COA position, and let i = i + 1. Repeat this process to update the COA position until i > N / 2, and then proceed to step S3.

[0013] S3. When i > N / 2, randomly generate the target position, calculate the COA position, update the COA position, and let i = i + 1. Repeat this process to update the COA position until i >= N, and then proceed to step S4.

[0014] S4. When i >= N, enter the second stage P2, let i = 1, calculate the position of the i-th COA in the second stage, update the COA position, and let i = i + 1. Repeat this process to update the COA position until i >= N, and then pass the result to the lower layer and proceed to step S5.

[0015] S5. The decision variables of the lower-layer PSO-GSA algorithm are the charge and discharge power of the energy storage for 24 hours, and the objective function is to minimize the active power loss and the vulnerability of the distribution network. Update the gravity G and the local optimal solution of the particle swarm, calculate the gravity and acceleration of all particles, update the speed and position. If t 2 < T 2 , t 2 = t 2 + 1. Repeat this step to update the speed and position until t2 >=T 2 , proceed to step S6;

[0016] S6. Determine whether the COA optimization algorithm COA iteration number t is less than T. If not, output the optimal solution found for the target; if so, set t=t+1 and repeat steps S2-S5.

[0017] Furthermore, in step S2, the position of the i-th COA is calculated using equation (14), and the COA position is updated using equation (17):

[0018]

[0019]

[0020] In step S3, the target position is randomly generated using equation (15), the COA position is calculated using equation (16), and the COA position is updated using equation (17):

[0021] Iguana G :Iguana G =lb j +r×(Iguana j -I×x i,j ),j=1,2,...,m, (15)

[0022]

[0023]

[0024] In step S4, the position of the i-th COA in the second stage is calculated using equations (18)-(19), and the COA position is updated using equation (20):

[0025]

[0026] X i,P2 :x i,j,P2 =x i,j +(1-2r)×(lb j,local +r×(ub j,local -lb j.local )),i=1,2,...,N,j=1,2,...,m, (19)

[0027]

[0028] Among them, X i , X i,p1 , X i,p2 They represent the new position of the i-th COA, the new position of the i-th COA in the first stage P1, the new position of the i-th COA in the second stage P2, and x respectively.i,j,p1 is the jth decision variable of the ith COA, P1 represents the first stage of COA, P2 represents the second stage of COA; r is a random number in [0,1]; F i 、F i,P1 、F i,P2 、F IguanaG They represent the objective function value, the objective function value of the P1 stage, the objective function value of the P2 stage, and the Iguana objective function value respectively; Iguana represents the best member position; Iguana j represents the jth decision variable; I is an integer randomly selected from {1, 2}; Iguana G Indicates the position of Iguana on the ground, which is randomly generated; Iguana j,G represents the jth decision variable; F IguanaG is the value of the objective function; is the layer function (maximum integer function); ubj and lbj represent the upper and lower bounds of the j-th decision variable; ubj,loca and lbj,local represent the local upper and lower bounds of the j-th decision variable.

[0029] Furthermore, in step S5, equation (22) is used to update the local optimal solution of gravity G and particle swarm, equations (23)-(24) are used to calculate the gravity and acceleration of all particles, and equations (25)-(26) are used to update the velocity and position:

[0030]

[0031]

[0032]

[0033] v ipso，t+1 =wv ipso，t +c 1 ×rand×a ipso，t +c 2 ×rand×(gbest-x ipso，t ) (25)

[0034] x ipso，t2+1 =x ipso，t2 +v ipso，t2+1 (26)

[0035] In the formula, Indicates the drop coefficient; G 0 represents the initial value; t 2 is the number of iterations; maxiter is the maximum number of iterations; F igsa (t) represents the individual igsa at the tth 2The total force of the iteration, randjgsa is a random number (0, 1); M ii (t) is the tth 2 The mass of the particle in iteration i, a i (t 2 ) represents the tth 2 The gravitational force of particle i in iteration 1; w represents the weighting function; v ipso,t2 represents the ipsoth particle t 2 The speed of generation time; c 1 represents the weight factor; rand is a random number between 0 and 1; c2 represents the weight factor; gbest is the best solution; v ipso,t2 Indicates the tth 2 The speed of the particle in the generation; v ipso,t2+1 Indicates the tth 2 +1 generation particle speed; x ipso,t2 t 2 The optimal position of the particle in the generation; wv ipso,t2 represents the exploration ability of the particle, c 1 ×rand×a i (t 2 ) represents the individual capacity of the particle, c 2 ×rand×(gbest-x ipso,t2 ) represents the cooperative ability of particles; x ipso,t2+1 represents the ipsoth particle t 2 +1 generation position.

[0036] Furthermore, the upper objective function of the model is to minimize the energy storage configuration capacity, and the lower objective function is to minimize the vulnerability of the distribution network and the active network loss.

[0037] Furthermore, the upper objective function of the model is:

[0038]

[0039]

[0040] Where: Indicates the i store Energy storage store The charge / discharge power at the moment; F 1 is the total energy storage capacity.

[0041] Furthermore, the lower objective function of the model is:

[0042] F 2 =min(z 1 f 1 +z 2 f 2 ) (8)

[0043] Where: f 1 Indicates active network loss; f 1 Indicates the vulnerability of the distribution network; 1 +z 2 =1 and z 1 ,z 2 ∈(0,1);

[0044] in,

[0045]

[0046] Where: N branch Indicates the number of branches; r ibranch Indicates i branch Resistance of the branch; Pi branch Indicates i branch Active power loss of the branch; Qi branch Indicates i branch Reactive power loss of branch circuit; Ui branch Indicates i branch The voltage of the branch;

[0047]

[0048] Where: f 2 represents the vulnerability index of the distribution network; N represents the total number of nodes, Indicates t time The voltage deviation of the inode node at the moment; f 2 The smaller it is, the safer the distribution network operation is.

[0049] An energy storage optimization configuration and control device, comprising:

[0050] Data acquisition module, used to obtain photovoltaic output, load output, energy storage selection location, capacity, and energy storage charging and discharging power and time scale;

[0051] The model building module is used to consider the timing of solar power output and photovoltaic penetration factors, and establish a two-layer energy storage optimization configuration and operation model. The upper layer objective function of the model is to minimize the energy storage configuration capacity, and the lower layer objective function is to minimize the vulnerability of the distribution network and the active network loss.

[0052] A constraint determination module, used to determine the constraint conditions of the energy storage dual-layer optimization configuration and operation model, wherein the constraint conditions include power balance constraint, power flow constraint, distributed new energy output constraint, and energy storage constraint;

[0053] The model solving module is used to use the COA algorithm COA to configure the energy storage location and capacity of the upper model, and use the particle swarm-gravitational search hybrid algorithm PSO-GSA to configure the energy storage charging and discharging power of the lower model. By optimizing the upper layer, the objective function of the lower layer is minimized, thereby improving the reliability and stability of the distribution network operation.

[0054] A computing device comprising:

[0055] one or more processing units;

[0056] A storage unit for storing one or more programs,

[0057] Wherein, when the one or more programs are executed by the one or more processing units, the one or more processing units execute the energy storage optimization configuration and control method as described above.

[0058] A computer-readable storage medium having a non-volatile program code executable by a processor, wherein the computer program, when executed by the processor, implements the steps of the above-mentioned energy storage optimization configuration and regulation method.

[0059] The present invention provides a novel method and device for quantitatively evaluating the regulation capability of a power system, which has the following beneficial effects:

[0060] The present invention proposes a dual-layer optimization strategy for energy storage using the COA algorithm and the PSO-GSA algorithm. The two algorithms proposed balance the local optimum and the global optimum, and have better optimization capabilities than existing optimization algorithms. By comparing the simulation results of Case 1, Case 2, and Case 3, the average reductions in the distribution network vulnerability index are 0.77%, 0.99%, and 0.8%, respectively, indicating that increasing energy storage can effectively enhance the reliability of distribution network operation; the average reductions in active network losses in the three cases are -5.9%, 1.96%, and 6.5%, respectively; for the energy storage system scheduling strategy, it can effectively alleviate the intermittent, volatile, and random characteristics of distributed energy output, and store excess unused electrical energy when the photovoltaic output is high; when the photovoltaic output is low and cannot meet the load requirements, the energy storage system releases electrical energy; it is beneficial to improve the photovoltaic absorption capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic diagram of a novel method for quantitatively evaluating the regulation capability of a power system provided by an embodiment of the present invention;

[0062] Figure 2 The 4-1 IEEE 33 node distribution network provided in Embodiment 1 of the present invention;

[0063] Figure 3 A typical daily load curve provided by Example 1 of the present invention;

[0064] Figure 4 A typical daily photovoltaic output curve provided in Example 1 of the present invention;

[0065] Figure 5 The energy storage scheduling strategy when the penetration rate is 30% provided in Example 1 of the present invention;

[0066] Figure 6 The energy storage scheduling strategy when the penetration rate is 50% provided in Example 1 of the present invention;

[0067] Figure 7 This is the energy storage scheduling strategy when the penetration rate is 80% provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] Example 1

[0070] This embodiment provides a method for optimizing energy storage configuration and regulation, including the following steps:

[0071] Obtain photovoltaic output, load output, energy storage selection location, capacity, and energy storage charging and discharging power and time scale;

[0072] Considering the timing of solar output and photovoltaic penetration factors, a two-layer optimization configuration and operation model of energy storage is established. The upper layer objective function of the model is to minimize the energy storage configuration capacity, and the lower layer objective function is to minimize the vulnerability of the distribution network and the active network loss.

[0073] Determining the constraints of the energy storage dual-layer optimization configuration and operation model, wherein the constraints include power balance constraints, power flow constraints, distributed new energy output constraints, and energy storage constraints;

[0074] The COA algorithm is used to configure the energy storage location and capacity of the upper model, and the particle swarm-gravitational search hybrid algorithm PSO-GSA is used to configure the energy storage charging and discharging power of the lower model. The objective function of the lower layer is minimized by optimizing the upper layer, thereby improving the reliability and stability of the distribution network operation.

[0075] Specifically, the method of using the COA optimization algorithm COA to configure the energy storage location and capacity for the upper-layer model, using the particle swarm-gravitational search hybrid algorithm PSO-GSA to configure the energy storage charge and discharge power for the lower-layer model, and minimizing the objective function of the lower layer through the optimization configuration of the upper layer to improve the operation reliability and stability of the distribution network is as follows:

[0076] S1. Set the population size N and the number of iterations T of the COA optimization algorithm COA, set the population size N2 and the number of iterations T2 of the particle swarm-gravitational search hybrid algorithm PSO-GSA, and set the individual learning factor C1 and the global learning factor C2; Initialize the number of iterations t of the COA algorithm COA to 1;

[0077] S2. The decision variables of the upper-layer model are the installation location and capacity of the energy storage, and the objective function is to minimize the configured capacity of the energy storage. Initialize the population size of the COA optimization algorithm COA; In the first stage of COA, when i < N / 2, calculate the position of the i-th COA, update the position of the COA, and let i = i + 1. Repeat updating the position of the COA in this way until i > N / 2, and then proceed to step S3;

[0078] S3. When i > N / 2, randomly generate the target position, calculate the position of the COA, update the position of the COA, and let i = i + 1. Repeat updating the position of the COA in this way until i >= N, and then proceed to step S4;

[0079] S4. When i >= N, enter the second stage. Let i = 1, calculate the position of the i-th COA in the second stage, update the position of the COA, and let i = i + 1. Repeat updating the position of the COA in this way until i >= N, and then pass the result to the lower layer and proceed to step S5;

[0080] S5. The decision variables of the lower-layer PSO-GSA algorithm are the charge and discharge power of the energy storage for 24 hours, and the objective function is to minimize the active power loss of the distribution network and the vulnerability of the distribution network. Update the gravity G and the local optimal solution gbest of the particle swarm, and the gravitational force F gsa and the acceleration a gsa , update the velocity v pso and the position x pso , if t2 < T2, then t2 = t2 + 1. Repeat this step to update the velocity and position in this way until t2 >= T2, and then proceed to step S6;

[0081] S6. Judge whether the number of iterations t of the COA optimization algorithm COA is less than T. If not, then output the optimal solution where the target is found; if so, then let t = t + 1, and repeat steps S2 - S5.

[0082] Among them, in step S2, use equation (14) to calculate the position of the i-th COA, and use equation (17) to update the position of the COA:

[0083]

[0084]

[0085] In step S3, the target position is randomly generated using equation (15), the COA position is calculated using equation (16), and the COA position is updated using equation (17):

[0086] Iguana G :Iguana G =lb j +r×(Iguana j -I×x i,j ),j=1,2,...,m, (15)

[0087]

[0088]

[0089] In step S4, the position of the i-th COA in the second stage is calculated using equations (18)-(19), and the COA position is updated using equation (20):

[0090]

[0091] X i,P2 :x i,j,P2 =x i,j +(1-2r)×(lb j,local +r×(ub j,local -lb j.local )),i=1,2,...,N,j=1,2,...,m, (19)

[0092]

[0093] Among them, X i , X i,p1 , X i,p2 They represent the new position of the i-th COA, the new position of the i-th COA in the first stage P1, the new position of the i-th COA in the second stage P2, and x respectively. i,j,p1 is the jth decision variable of the ith COA, P1 represents the first stage of COA, P2 represents the second stage of COA; r is a random number in [0,1]; F i 、F i,P1 、F i,P2 、F IguanaG They represent the objective function value, the objective function value of the P1 stage, the objective function value of the P2 stage, and the Iguana objective function value respectively; Iguana represents the best member position; Iguana jrepresents the jth decision variable; I is an integer randomly selected from {1, 2}; Iguana G Indicates the position of Iguana on the ground, which is randomly generated; Iguana j,G represents the jth decision variable; F IguanaG is the value of the objective function; is the layer function (maximum integer function); ubj and lbj represent the upper and lower bounds of the j-th decision variable; ubj,loca and lbj,local represent the local upper and lower bounds of the j-th decision variable.

[0094] In step S5, the gravity G and the local optimal solution gbest of the particle swarm are updated using equation (22), and the gravity F of all particles is calculated using equations (23)-(24). gsa and acceleration a gsa , Equations (25)-(26) update speed v pso and position x pso :

[0095]

[0096]

[0097]

[0098] v ipso，t+1 =wv ipso，t +c 1 ×rand×a ipso，t +c 2 ×rand×(gbest-x ipso，t ) (25)

[0099] x ipso，t2+1 =x ipso，t2 +v ipso，t2+1 (26)

[0100] The established energy storage dual-layer optimization configuration and operation model includes an upper-layer model and a lower-layer model, where:

[0101] The objective function of the upper model is to minimize the energy storage configuration capacity:

[0102] The rated capacity of energy storage is the maximum charge / discharge capacity in one cycle.

[0103]

[0104]

[0105] Where: Indicates the i store Energy storage storeThe charge / discharge power at the moment; F 1 is the sum of the energy storage capacity;

[0106] The objective function of the lower model is to minimize the vulnerability of the distribution network and the active network loss:

[0107] F 2 =min(z 1 f 1 +z 2 f 2 ) (13)

[0108] Where: f 1 Indicates active network loss; f 1 Indicates the vulnerability of the distribution network; 1 +z 2 =1 and z 1 ,z 2 ∈(0,1).

[0109] 1) Active network loss

[0110]

[0111] Where: N branch Indicates the number of branches; r ibranch Indicates i branch Resistance of the branch; Pi branch Indicates i branch Active power loss of the branch; Qi branch Indicates i branch Reactive power loss of branch circuit; Ui branch Indicates i branch The voltage of the branch.

[0112] 2) Distribution network vulnerability

[0113] The distribution network vulnerability index refers to the safety of grid operation when a high proportion of renewable energy and energy storage are connected. The risk resistance of the grid is judged by analyzing the voltage deviation standard value. The higher the vulnerability, the more unstable the voltage and the worse the risk resistance.

[0114]

[0115] Where: U inode Indicates the voltage of the node inode, U Nnode Indicates the rated voltage of node Nnode; 0.07 indicates the allowable voltage deviation, as shown in Table 1-1:

[0116] Table 1-1 Voltage deviation standard values

[0117]

[0118]

[0119]

[0120] Where: f 2 represents the vulnerability index of the distribution network; N represents the total number of nodes, ΔU inode,t Indicates t time The voltage deviation of the inode node at the moment; f 2 The smaller it is, the safer the distribution network operation is.

[0121] The constraints of the model are as follows:

[0122] 1) Power balance constraints

[0123] Power balance means that the supply and demand of power are equal during the normal operation of the system, that is, the output power is equal to the input power. If the power is unbalanced during the operation of the system, it will cause fluctuations in the power system, that is, voltage fluctuations or frequency fluctuations, resulting in unstable operation of the system.

[0124]

[0125] Where P load Refers to the power required by the connected load; P i Refers to the power generated by a high proportion of renewable energy; P loss Refers to the power lost during the transmission process when outputting power to the load; P grid It refers to the power lost when a high proportion of renewable energy is integrated into the power grid.

[0126] 2) Power flow constraints

[0127]

[0128] Where: P inode is the active power of the node inode; Q inode is the reactive power of the node inode; U inode is the voltage amplitude of the node inode; U jnode is the voltage amplitude of node jnode; G inodejnode is the conductance of the inode-th row and jnode-th column in the node admittance matrix; B inodejnode is the susceptance of the inode-th row and jnode-th column in the node admittance matrix; θ inodejnode is the voltage phase angle between node inode and node jnode.

[0129] 3) Distributed renewable energy output constraints

[0130] P DG,min ≤P DG，t ≤P DG,max (19)

[0131] Where: P DG,min is the minimum value of distributed photovoltaic output; P DG,t t time Distributed photovoltaic output at any time; P DG,max It is the maximum value of distributed photovoltaic output.

[0132] 4) Energy storage constraints

[0133]

[0134] Where: P store,min,jstore is the minimum value of the charge / discharge power of the jth energy storage; P store,jstore (t) is the energy storage t of the jth store time Charge / discharge power at each moment; P store,max,jstore is the maximum value of the charge / discharge power of the jth energy storage; P store,jstore (t time )≥0 indicates energy storage discharge; P store,jstore (t time )<0 indicates energy storage charging; SOC j (t) is the state of charge of the energy storage.

[0135] The COA optimization algorithm in this application is described as follows:

[0136] COA optimization algorithm initialization process

[0137] The COA Optimization Algorithm (COA) is a population-based metaheuristic algorithm that uses COAs as population members. The position of each COA in the search space determines the value of the decision variable. Therefore, the position of the COA in the COA represents a candidate solution to the problem.

[0138] 1) Initialize the position of COA in the search space

[0139] a represents the particle search capability and is the control variable of the particle swarm algorithm. The speed of the particle is determined by a.

[0140] X i :x ij =lb j +r×(ub j -lb j ),i=1,2,...,N,j=1,2,...,m (21)

[0141] X i represents the position of the i-th COA in the search space; x ijrepresents the jth decision variable of the ith COA; N represents the number of COAs; m represents the number of decision variables; r is a random number in [0,1]; ub j and lb j Represents the upper and lower bounds of the decision variables.

[0142] 2) The COA population in COA is represented by matrix X.

[0143]

[0144] 3) Placing candidate solutions in the decision variables results in different evaluations of the problem's objective function. These values ​​are represented by F:

[0145]

[0146] F is the vector of the objective function, F i is the objective function value obtained based on the i-th COA.

[0147] In metaheuristic algorithms, such as the proposed COA algorithm, the value of the objective function is a measure of the quality of the candidate solutions. Therefore, the member of the population that leads to the best value of the objective function is called the best member of the population. Since the candidate solutions are updated during the algorithm iterations, the best member of the population is also updated in each iteration.

[0148] COA Optimization Algorithm Mathematical Model of COA

[0149] The process of updating the position of candidate solutions in COA is based on modeling two behaviors of COA. These behaviors include: (1) the strategy of COA when attacking Iguana; (2) the strategy of proboscis COA to escape from predators. Therefore, the COA population is updated in two different stages.

[0150] Phase 1: Hunting and attack strategies against Iguana (exploration phase)

[0151] COA capture is generally done by half of the COA climbing a tree to chase the target, while the other half waits on the ground for the target to fall; the COA's position as it rises from the tree is mathematically simulated.

[0152]

[0153] When the Iguana falls, it is placed at a random position in the search space, and based on this random position, the COA moves in the search space.

[0154] Iguana G :Iguana G =lb j +r×(Iguana j -I×x i,j),j=1,2,...,m, (25)

[0155]

[0156] For each new position of the COA, if the value of the objective function is improved, the updating process is accepted, otherwise, the proboscis bear remains in its previous position.

[0157]

[0158] Phase 2: The process of escaping from a predator (exploitation phase)

[0159] The COA position in the search space is modeled and when the predator attacks the COA, the COA flees from its position, which indicates the COA’s ability to exploit in local exploration.

[0160] To simulate this behavior, a random position is generated near each COA.

[0161]

[0162] X i,P2 :x i,j,P2 =x i,j +(1-2r)×(lb j,local +r×(ub j,local -lb j.local )),i=1,2,...,N,j=1,2,...,m, (29)

[0163] If the newly calculated position improves the value of the objective function, then the position is acceptable.

[0164]

[0165] Among them, X i , X i,p1 , X i,p2 They represent the new position of the i-th COA, the new position of the i-th COA in the first stage P1, the new position of the i-th COA in the second stage P2, and x respectively. i,j,p1 is the jth decision variable of the ith COA, P1 represents the first stage of COA, P2 represents the second stage of COA; r is a random number in [0,1]; F i 、F i,P1 、F i,P2 、F IguanaG They represent the objective function value, the objective function value of the P1 stage, the objective function value of the P2 stage, and the Iguana objective function value respectively; Iguana represents the best member position; Iguana j represents the jth decision variable; I is an integer randomly selected from {1, 2}; IguanaG Indicates the position of Iguana on the ground, which is randomly generated; Iguana j,G represents the jth decision variable; F IguanaG is the value of the objective function; is the layer function (maximum integer function); ubj and lbj represent the upper and lower bounds of the j-th decision variable; ubj,loca and lbj,local represent the upper and lower bounds of the j-th decision variable; t is the number of iterations.

[0166] PSO-GSA Algorithm Model

[0167] The interaction force of particles in the GSA universe, which is proportional to the mass of the particles and inversely proportional to the distance between them, is modeled as follows.

[0168]

[0169] In the formula, F igsajgsa (t) represents the tth 2 The gravitational force of particle jgsa on particle igsa in the iteration; G(t 2 ) represents the tth 2 The value of the universal gravitation of the iteration; M pigsa (t) Active gravitational mass; M ajgsa (t) passive gravitational mass; R igsajgsa (t) represents the Euclidean distance between igsa and jgsa; ξ represents a constant; x jgsa (t) represents the tth 2 The position of jgsa; x igsa (t) represents the tth 2 On behalf of the location of igsa.

[0170]

[0171] In the formula, Indicates the drop coefficient; G 0 represents the initial value; t 2 is the number of iterations; maxiter is the maximum number of iterations.

[0172]

[0173] In the formula, F igsa (t) represents the individual igsa at the tth 2 The resultant force of the iteration, randjgsa is a random number in (0, 1).

[0174]

[0175] Where M ii (t) is the tth2 The mass of the particle in iteration i, a i (t 2 ) represents the tth 2 The gravitational force of particle i in iteration i.

[0176] Each particle of the particle swarm algorithm must consider the current speed, current position, and gbest's distance modification function, which is modeled as follows.

[0177] v ipso，t+1 =wv ipso，t +c 1 ×rand×a ipso，t +c 2 ×rand×(gbest-x ipso，t ) (25)

[0178] Where: w represents the weighting function; v ipso,t2 represents the ipsoth particle t 2 The speed of generation time; c 1 represents the weight factor; rand is a random number between 0 and 1; c2 represents the weight factor; gbest is the best solution; v ipso,t2 Indicates the tth 2 The speed of the particle in the generation; v ipso,t2+1 Indicates the tth 2 +1 generation particle speed; x ipso,t2 t 2 The optimal position of the particle in the generation; wv ipso,t2 represents the exploration ability of the particle, c 1 ×rand×a i (t 2 ) represents the individual capacity of the particle, c 2 ×rand×(gbest-x ipso,t2 ) represents the collaborative ability of particles.

[0179]

[0180] Where x ipso,t2+1 represents the ipsoth particle t 2 +1 generation position.

[0181] As an example, in this embodiment, an improved IEEE 33-node distribution network is used as a test object. Figure 2 As shown, there are 32 branches and 33 nodes, and the line is an overhead line type; the system base voltage is 12.66KV, the system base capacity is 10MVA, and the balance voltage is 1.00pu.

[0182] The optimal access location and capacity of energy storage under sequential photovoltaics were tested. In the simulation, sequential photovoltaics were connected to nodes 9 and 30 of the distribution network; energy storage was randomly connected to nodes 2 to 33, with a capacity of 200KW to 900KW.

[0183] Table 1 Simulation parameter settings

[0184]

[0185] This embodiment is divided into three different scenarios for comparison to verify the effectiveness of the proposed method and model. Case 1, photovoltaic access penetration rate of 30%; Case 2, photovoltaic access penetration rate of 50%; Case 3, photovoltaic access penetration rate of 80%. This embodiment gives the top 5 optimal solutions for analysis, and the optimization results under different scenarios are shown in Table 2, Table 3, and Table 4.

[0186] Table 2 Case 1: PV access penetration rate of 30%

[0187]

[0188]

[0189] As shown in Table 2, when the photovoltaic penetration rate is 30%, without energy storage, the distribution network vulnerability is 0.790pu and the network loss is 1.480MW; when energy storage is added, the average values ​​of distribution network vulnerability and network loss are 0.784pu and 1.552MW respectively; compared with before optimization, they are reduced by an average of 0.77% and -5.9% respectively.

[0190] Table 3 Case 2: PV access penetration rate of 50%

[0191]

[0192] As shown in Table 3, when the photovoltaic penetration rate is 50% and no energy storage is added, the distribution network vulnerability is 0.804pu and the network loss is 1.633MW; when energy storage is added, the average values ​​of distribution network vulnerability and network loss are 0.796pu and 1.601MW respectively; compared with before optimization, they are reduced by an average of 0.99% and 1.96% respectively.

[0193] Table 4 Case 3, PV access penetration rate of 80%

[0194]

[0195]

[0196] As shown in Table 4, when the photovoltaic penetration rate is 80%, without energy storage, the distribution network vulnerability is 0.783pu and the network loss is 2.483MW; when energy storage is added, the average values ​​of distribution network vulnerability and network loss are 0.775pu and 2.321MW respectively; compared with before optimization, they are reduced by an average of 0.8% and 6.5% respectively.

[0197] By comparing the three cases, it can be concluded that with the increase in photovoltaic penetration, the optimal configuration and scheduling of energy storage has a significant improvement on the vulnerability of the distribution network and network losses. In terms of algorithms, the COA algorithm and the PSO-GSA algorithm have stronger local and global search capabilities and are easier to achieve global optimality. The representative optimal solutions of different cases are selected, and their energy storage access location capacity, distribution network vulnerability, active network loss, and energy storage optimization scheduling are as follows:

[0198] Case 1:

[0199] Table 5 Optimal energy storage solution when the permeability is 30%

[0200]

[0201] Case 2:

[0202] Table 6 Optimal energy storage solution when the permeability is 50%

[0203]

[0204] Case 3:

[0205] Table 7 Optimal solution for energy storage when the permeability is 80%

[0206]

[0207]

[0208] Figure 7 It is the energy storage dispatch strategy when the penetration rate is 80%. The energy storage is charged when the load is low or the photovoltaic power generation is large, and discharged when the load is high. Through the method of low storage and high discharge and time-of-use electricity price, the vulnerability of the distribution network and active network loss are effectively reduced, and the high penetration rate photovoltaic consumption is promoted.

[0209] Considering the large-scale distributed energy access to the distribution network in the future, and the randomness, volatility and intermittency of distributed energy output, this application proposes energy storage optimization and interactive regulation based on COA algorithm and PSO-GSA algorithm. The COA algorithm (COA) and the particle swarm-gravitational search hybrid algorithm (PSO-GSA) balance the capabilities of local search and global search. Compared with other popular algorithms, the two algorithms used in this paper have better competitiveness through the test function IEEE CEC-2017. First, guided by the stability of distribution network planning and distribution network operation, the model is divided into two layers: the upper layer uses the COA algorithm to configure the energy storage location and capacity, and the lower layer uses the PSO-GSA algorithm to establish a model with the goal of minimizing the vulnerability of the distribution network and the active network loss. The model is solved by taking into account the timing of photovoltaic output and factors such as different photovoltaic penetration rates. Finally, the IEEE33 node example is used to verify the rationality of the proposed method in the optimal configuration and operation of energy storage in the distribution network, which can effectively reduce the vulnerability of the distribution network and the active network loss, and promote the high-proportion penetration photovoltaic consumption.

[0210] Example 2

[0211] This embodiment provides an energy storage optimization configuration and control device, including:

[0212] Data acquisition module, used to obtain photovoltaic output, load output, energy storage selection location, capacity, and energy storage charging and discharging power and time scale;

[0213] The model building module is used to consider the timing of solar power output and photovoltaic penetration factors, and establish a two-layer energy storage optimization configuration and operation model. The upper layer objective function of the model is to minimize the energy storage configuration capacity, and the lower layer objective function is to minimize the vulnerability of the distribution network and the active network loss.

[0214] A constraint determination module, used to determine the constraint conditions of the energy storage dual-layer optimization configuration and operation model, wherein the constraint conditions include power balance constraint, power flow constraint, distributed new energy output constraint, and energy storage constraint;

[0215] The model solving module is used to use the COA algorithm COA to configure the energy storage location and capacity of the upper model, and use the particle swarm-gravitational search hybrid algorithm PSO-GSA to configure the energy storage charging and discharging power of the lower model. By optimizing the upper layer, the objective function of the lower layer is minimized, thereby improving the reliability and stability of the distribution network operation.

[0216] A computing device is also provided, comprising:

[0217] one or more processing units;

[0218] A storage unit for storing one or more programs,

[0219] Among them, when the one or more programs are executed by the one or more processing units, the one or more processing units execute a method for optimizing energy storage configuration and control in this embodiment; it should be noted that the computing device may include but is not limited to a processing unit and a storage unit; those skilled in the art can understand that the computing device including a processing unit and a storage unit does not constitute a limitation on the computing device, and may include more components, or a combination of certain components, or different components, for example, the computing device may also include input and output devices, network access devices, buses, etc.

[0220] A computer-readable storage medium having a non-volatile program code executable by a processor, wherein the computer program, when executed by the processor, implements the steps of a method for optimizing energy storage configuration and regulation in the present embodiment; it should be noted that the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof; the program contained on the readable medium may be transmitted using any appropriate medium, including, but not limited to, wireless, wired, optical cable, RF, etc., or any suitable combination thereof. For example, the program code for performing the operation of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and also conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, as an independent software package, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0221] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0222] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing energy storage configuration and regulation, characterized in that: It includes the following steps: Obtain the photovoltaic output, load output, the selected location and capacity of the energy storage, as well as the charging and discharging power and time scale of the energy storage; Considering the time series of the light output and the photovoltaic penetration rate factor, establish a two-layer optimal configuration and operation model for the energy storage. The upper-layer objective function of the model is to minimize the energy storage configuration capacity, and the lower-layer objective function is to minimize the vulnerability and active power loss of the distribution network; Determine the constraint conditions of the two-layer optimal configuration and operation model for the energy storage. The constraint conditions include power balance constraints, power flow constraints, distributed new energy output constraints, and energy storage constraints; Use the COA algorithm (COA) to configure the energy storage location and capacity for the upper-layer model, and use the particle swarm-gravitational search hybrid algorithm (PSO-GSA) to configure the charging and discharging power of the energy storage for the lower-layer model. By optimizing the upper layer, the objective function of the lower layer is minimized, thereby improving the operation reliability and stability of the distribution network.

2. The energy storage optimization configuration and control method according to claim 1, characterized in that: The method of using the COA optimization algorithm (COA) to configure the energy storage location and capacity for the upper-layer model, and using the particle swarm-gravitational search hybrid algorithm (PSO-GSA) to configure the charging and discharging power of the energy storage for the lower-layer model, and minimizing the objective function of the lower layer by optimizing the upper layer, thereby improving the operation reliability and stability of the distribution network is as follows: S1. Set the population size N and the number of iterations T of the COA optimization algorithm (COA). COA is a two-stage algorithm. The first stage is P1, and the second stage is P2. Set the population size N2 and the number of iterations T2 of the particle swarm-gravitational search hybrid algorithm (PSO-GSA). Set the individual learning factor C1 and the global learning factor C2; Initialize the number of iterations t of the COA algorithm (COA) to t = 1; S2. The decision variables of the upper-layer model are the installation location and capacity of the energy storage, and the objective function is to minimize the energy storage configuration capacity. Initialize the population size i of the COA optimization algorithm (COA) to i = 1; In the first stage P1 of COA, when i < N / 2, calculate the position of the i-th COA, update the COA position, and let i = i + 1. Repeat this process to update the COA position until i > N / 2, and then proceed to step S3; S3. When i > N / 2, randomly generate the target position, calculate the COA position, update the COA position, and let i = i + 1. Repeat this process to update the COA position until i >= N, and then proceed to step S4; S4. When i >= N, perform the second stage P2. Let i = 1, calculate the position of the i-th COA in the second stage, update the COA position, and let i = i + 1. Repeat this process to update the COA position until i >= N, and then pass the result to the lower layer and proceed to step S5; S5. The decision variables of the lower-layer PSO-GSA algorithm are the charging and discharging power of the energy storage for 24 hours, and the objective function is to minimize the active power loss and vulnerability of the distribution network. Update the gravity G and the local optimal solution of the particle swarm, calculate the gravitational force and acceleration of all particles, update the velocity and position. If t2 < T2, then t2 = t2 + 1. Repeat this step to update the velocity and position until t2 >= T2, and then proceed to step S6; S6. Determine whether the number of iterations t of the COA optimization algorithm (COA) is less than T. If not, then output the optimal solution where the target is found; If yes, set t=t+1 and repeat steps S2-S5.

3. The energy storage optimization configuration and control method according to claim 2 is characterized in that: In step S2, the position of the i-th COA is calculated using equation (14), and the COA position is updated using equation (17): In step S3, the target position is randomly generated using equation (15), the COA position is calculated using equation (16), and the COA position is updated using equation (17): Iguana G :Iguana G =lb j +r×(Iguana j -I×x i,j ),j=1,2,...,m, (15) In step S4, the position of the i-th COA in the second stage is calculated using equations (18)-(19), and the COA position is updated using equation (20): X i,P2 :x i,j,P2 =x i,j +(1-2r)×(lb j,local +r×(ub j,local -lb j.local )),i=1,2,...,N,j=1,2,...,m, (19) Among them, X i , X i,p1 , X i,p2 They represent the new position of the i-th COA, the new position of the i-th COA in the first stage P1, the new position of the i-th COA in the second stage P2, and x respectively. i,j,p1 is the jth decision variable of the ith COA, P1 represents the first stage of COA, P2 represents the second stage of COA; r is a random number in [0,1]; F i 、F i,P1 、F i,P2 、F IguanaG They represent the objective function value, the objective function value of the P1 stage, the objective function value of the P2 stage, and the Iguana objective function value respectively; Iguana represents the best member position; Iguana j represents the jth decision variable; I is an integer randomly selected from {1, 2}; Iguana G Indicates the position of Iguana on the ground, which is randomly generated; Iguana j,G represents the jth decision variable; F IguanaG is the value of the objective function; is the layer function (maximum integer function); ubj and lbj represent the upper and lower bounds of the j-th decision variable; ubj,loca and lbj,local represent the local upper and lower bounds of the j-th decision variable.

4. The energy storage optimization configuration and control method according to claim 2 is characterized in that: In step S5, equation (22) is used to update the local optimal solution of gravity G and particle swarm, equations (23)-(24) are used to calculate the gravity and acceleration of all particles, and equations (25)-(26) are used to update the velocity and position: v ipso，t+1 =wv ipso，t +c1×rand×a ipso，t +c2×rand×(gbest-x ipso，t ) (25) x ipso，t2+1 =x ipso，t2 +v ips o,t2+1 (26) In the formula, represents the drop coefficient; G0 represents the initial value; t2 represents the number of iterations; maxiter represents the maximum number of iterations; F igsa (t) represents the resultant force on individual igsa at the t2th iteration, randjgsa is a random number (0, 1); M ii (t) is the mass of particle i at the t2th iteration, a i (t2) represents the gravitational force of particle i at the t2th iteration; w represents the weighting function; v ipso,t2 represents the speed of the ipso-th particle at generation t2; c1 represents the weight factor; rand is a random number between 0 and 1; c2 represents the weight factor; gbest is the best solution; v ipso,t2 represents the velocity of the particle at the t2th generation; v ipso,t2+1 represents the velocity of the particle at the t2+1 ​​generation; x ipso,t2 is the optimal position of the particle at generation t2; wv ipso,t2 Represents the exploration ability of the particle, c1×rand×a i (t2) represents the individual ability of the particle, c2×rand×(gbest-x ipso,t2 ) represents the cooperative ability of particles; x ipso,t2+1 Represents the position of the ipso-th particle at generation t2+1.

5. The energy storage optimization configuration and control method according to claim 1, characterized in that: The upper objective function of the model is to minimize the energy storage configuration capacity, and the lower objective function is to minimize the vulnerability of the distribution network and the active network loss.

6. The energy storage optimization configuration and control method according to claim 5, characterized in that: The upper objective function of the model is: Where: Indicates the i store Energy storage store The charge / discharge power at the moment; F1 is the total energy storage capacity.

7. The energy storage optimization configuration and control method according to claim 5, characterized in that: The lower objective function of the model is: F2=min(z1f1+z2f2) (3) Where: f1 represents active network loss; f1 represents the vulnerability of the distribution network; z1+z2=1 and z1,z2∈(0,1); in, Where: N branch Indicates the number of branches; r ibranch Indicates i branch Resistance of the branch; Pi branch Indicates i branch Active power loss of the branch; Qi branch Indicates i branch Reactive power loss of branch circuit; Ui branch Indicates i branch The voltage of the branch; Where: f2 represents the distribution network vulnerability index; N represents the total number of nodes, Indicates t time The voltage deviation of the inode node at the moment; the smaller f2 is, the safer the distribution network operation is.

8. An energy storage optimization configuration and control device, characterized in that: include: Data acquisition module, used to obtain photovoltaic output, load output, energy storage selection location, capacity, and energy storage charging and discharging power and time scale; The model building module is used to consider the timing of solar power output and photovoltaic penetration factors, and establish a two-layer energy storage optimization configuration and operation model. The upper layer objective function of the model is to minimize the energy storage configuration capacity, and the lower layer objective function is to minimize the vulnerability of the distribution network and the active network loss. A constraint determination module, used to determine the constraint conditions of the energy storage dual-layer optimization configuration and operation model, wherein the constraint conditions include power balance constraint, power flow constraint, distributed new energy output constraint, and energy storage constraint; The model solving module is used to use the COA algorithm COA to configure the energy storage location and capacity of the upper model, and use the particle swarm-gravitational search hybrid algorithm PSO-GSA to configure the energy storage charging and discharging power of the lower model. By optimizing the upper layer, the objective function of the lower layer is minimized, thereby improving the reliability and stability of the distribution network operation.

9. A computing device comprising: one or more processing units; A storage unit for storing one or more programs, Wherein, when the one or more programs are executed by the one or more processing units, the one or more processing units execute the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium having a processor-executable non-volatile program code, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images