Improved lizard algorithm-based distributed power supply-containing power distribution network reconstruction method

Abstract

The invention provides a distributed power supply-containing power distribution network reconstruction method based on an improved lizard algorithm. The distributed power supply-containing power distribution network reconstruction method is characterized by comprising the following steps of S1, determining an objective function and constraint conditions of a power distribution network reconstruction problem; s2, a lizard algorithm is introduced, and improvement is carried out, specifically, the improvement comprises. And S3, reconstructing the power distribution network containing the distributed power supply based on the lizard algorithm, specifically, Tent mapping is introduced into the standard lizard algorithm, so that an initial solution is uniformly distributed in a solution space as far as possible, and the behavior of canid escaping from a predator is simulated to enhance the search capability of the algorithm, so that the algorithm is more efficient and more efficient. The method aims to optimize global search and local search capabilities in the algorithm, can have better convergence characteristics, can be used for solving a power distribution network reconstruction problem, can be optimized to obtain smaller active loss and higher voltage stability index, effectively improves the horizontal distribution of bus voltage, and has better stability.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network optimization, and in particular relates to a distribution network reconstruction method containing distributed power sources based on an improved lizard algorithm. Background Art

[0002] With the rapid growth of the domestic economy and people's demand for electricity, the distribution network at the end of the power network system that supplies electricity to grid customers has gradually become prominent due to its delayed construction, unreasonable structure, and poor power supply reliability. Therefore, it is difficult to meet the high power quality and high power supply reliability requirements of power supply users. Distribution network reconstruction is precisely one of the important methods for controlling and operating the distribution network, and it is also the most important component of the distribution network management link. Therefore, it is necessary to conduct reconstruction research in the current distribution network environment. At the same time, with the increasing proportion of electricity consumption by ordinary users in the system, the load and network structure carried by power transmission through the public distribution network have undergone great changes, and the proportion of system losses caused by line losses in the distribution network has also increased. Therefore, such problems have attracted the attention of power system researchers. In response to these problems, connecting DG to the distribution network system can not only reasonably optimize the economy of system operation, but also increase more electricity. However, DG access makes the flow in the distribution network bidirectional. DG grid connection will change the flow distribution of the traditional distribution network, resulting in changes in the load of each node in the network. The selection of DG grid connection and capacity has a great impact on the operating characteristics of the system. If the configuration is unreasonable, it will not only increase the active power loss of the system operation, affect the voltage level of each bus, but also increase the operating cost of the power grid. Therefore, it is very important to optimize the distribution network containing DG. Summary of the invention

[0003] 1. Technical problems to be solved:

[0004] Solve the problem of optimizing and reconstructing distribution networks containing distributed generation and improve the distribution network's ability to absorb DG.

[0005] 2. Technical solution:

[0006] In order to solve the above problems, the present invention provides a method for reconfiguring a distribution network containing distributed power sources based on an improved lizard algorithm, which is characterized by comprising the following steps:

[0007] Step S1: Determine the objective function and constraints of the distribution network reconstruction problem.

[0008] Step S2: Introduce the lizard algorithm and make improvements. The specific improvements include:

[0009] The Tent chaotic map is used to initialize the population generation strategy, so that the initial solution is distributed as evenly as possible in the solution space, and the generated chaotic sequence Z I k The process is:

[0010]

[0011] Where k is the population size; I is the current iteration number; to ensure the randomness of the algorithm initialization information, u takes a random number in the range of [0,1].

[0012] Further generate the initial positions of the population individuals in the search area The process is:

[0013]

[0014] In the formula, They are The maximum and minimum values ​​of the sequence.

[0015] Step S3: Reconstruct the distribution network containing distributed generation based on the lizard algorithm, specifically:

[0016] Step S31: The horned lizard hides the imitation behavior, and based on the trigonometric function combination of the selected search agent and the proportional coefficient in the palette, mutates the current optimal solution to generate a new candidate solution.

[0017] Step S32: simulating the blood spraying behavior and movement escape strategy of the lizard, and simulating the darkening of the lizard's skin to adaptively adjust the search strategy.

[0018] Step S33: Calculate the melanocyte rate MSH and replace individuals with low melanocyte rates with new and better solutions.

[0019] Step S34: Determine whether the termination condition is met. If the maximum number of iterations and accuracy conditions are met, output the optimal reconstruction structure of the distribution network corresponding to the optimal solution phasor. Otherwise, repeat steps S31 to S33 until the algorithm meets the termination condition.

[0020] In step S1, the objective function is:

[0021] OF=min(w1OF1+w2OF2) (3)

[0022] Where: w1, w2 are the weight coefficients of the two sub-objective functions, OF1 is the active power loss in the distribution network, and OF2 is the voltage stability index.

[0023] The active power loss calculation formula in the distribution network is:

[0024]

[0025] Where: B is the number of branches; P loss is the energy storage operation cost; R k represents the resistance of the kth branch; I k represents the current in the kth branch.

[0026] The calculation formula of voltage stability index is:

[0027]

[0028] Where: OVSI is the voltage stability index of each bus, excluding the relaxed nodes, N represents the number of nodes; VSI is the voltage stability index, and the calculation formula is:

[0029] VSI i+1 =|V i 4 |-4(P i+1 X k -Q i+1 R k )-4(P i+1 X k -Q i+1 R k )|V i 2 | (6)

[0030] Where: X k represents the reactance of the kth branch; V i represents the voltage of node i; P i+1 , Q i+1 represents the active and reactive power at node i.

[0031] In step S1 , the constraints include power balance constraints, node voltage constraints, branch current constraints, DG capacity constraints, and DG penetration rate constraints.

[0032] The power balance constraint is:

[0033]

[0034] Where: P s , Q s N is the active and reactive power generated by the generator at the relaxed node; DG is the number of DGs connected to the distribution network; P DG，i , Q DG，i represents the active and reactive power generated by the DG at node i.

[0035] For node i, the upper and lower limit constraints of its node voltage are expressed as:

[0036] V i,min ≤Vi ≤V i,max (8).

[0037] The constraint on the kth branch current is:

[0038] I k ≤I k,max (9)

[0039] In the formula, I k，max The maximum current allowed in the branch.

[0040] The DG capacity constraint is expressed as:

[0041]

[0042] Where P DG，min , P DG，max They represent the minimum and maximum active power generated by DG, Q DG，min , Q DG，max They represent the minimum and maximum values ​​of reactive power generated by DG respectively;

[0043] The penetration rate of DG is selected to be 10% to 60% of the total active load demand in the distribution network; the DG capacity constraint is:

[0044]

[0045] In step S31, the horned lizard hides its imitation behavior and generates a new candidate solution by mutating the current optimal solution based on the trigonometric function combination of the selected search agent and the proportional coefficient in the palette. The process expression is:

[0046]

[0047] [c1(sin(x r1 (t))-cos(x r2 (t))-(-1) s c2(cos(x r3 (t))-sin(x r4 (t)))](12)

[0048] In the formula, is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent of generation t; r1, r2, r3 and r4 are integer random numbers generated between 1 and the maximum number of the population, where r1-r4 are not equal to each other; x r1 (t), x r2 (t), x r3 (t) and x r4(t) is the selected r1, r2, r3, r4-th search agent; Max_iter represents the maximum number of iterations, σ is a binary value obtained by Algorithm 1, and c1, c2, where c1≠c2, are random numbers drawn from the inclusive standardized palette.

[0049] In step S32, the trajectory expression of the blood ejection behavior of the lizard is:

[0050]

[0051] In the formula, x i (t+1) is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent found, the current search agent is x i (t), Max_iter represents the maximum number of iterations, t is the current number of iterations, v o Set to 1, a is set to Π / 2, ε is set to 1E-6, and g is the acceleration due to gravity.

[0052] In step S32, the mobile escape strategy expression is:

[0053]

[0054] x i (t+1) is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent to generate t, walk is a random number generated between -1 and 1, c is the mean of the random numbers generated by the standard Cauchy distribution, σ is set to a number between 0 and 1, and x i (t) is the i-th search agent in the current t iteration.

[0055] In step S32, the adaptive search strategy of the lizard skin darkening behavior simulation algorithm is used to allow the algorithm to adaptively adjust the search strategy according to the quality of the current solution. The skin variable and darkening expression are:

[0056]

[0057]

[0058] Among them, Light1 and Light2 are random numbers generated between Lightening1 and Lightening2, using normalized values; Dark1 and Dark2 are random numbers generated between Darkening1Darkening2, using normalized values.

[0059] 3. Beneficial effects:

[0060] The present invention proposes a distribution network reconstruction method containing distributed power sources based on an improved lizard algorithm, introduces Tent mapping in the standard lizard algorithm to make the initial solution as evenly distributed as possible in the solution space, and simulates the behavior of horned lizards escaping from predators to enhance the search ability of the algorithm, aiming to optimize the global search and local search capabilities in the algorithm, so that the method can have better convergence characteristics, and is used for solving the distribution network reconstruction problem, which can optimize to obtain smaller active power loss and higher voltage stability index, and effectively improve the horizontal distribution of bus voltage, and the method has good stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of the present invention.

[0062] Figure 2 This is the improvement of the bus voltage distribution level in the example of the present invention.

[0063] Figure 3 It is a comparison chart of the convergence of various algorithms in the examples of the present invention. DETAILED DESCRIPTION

[0064] The present invention is described in detail below through the accompanying drawings and embodiments.

[0065] like Figure 1 As shown, a method for reconfiguring a distribution network containing distributed power sources based on an improved lizard algorithm is characterized in that it includes the following steps:

[0066] Step S1: Determine the objective function and constraints of the distribution network reconstruction problem.

[0067] Step S2: Introduce the lizard algorithm and make improvements. The specific improvements include:

[0068] The Tent chaotic map is used to initialize the population generation strategy, so that the initial solution is distributed as evenly as possible in the solution space, and the generated chaotic sequence The process is:

[0069]

[0070] Where k is the population size; I is the current iteration number; to ensure the randomness of the algorithm initialization information, u takes a random number in the range of [0,1].

[0071] Further generate the initial positions of the population individuals in the search area The process is:

[0072]

[0073] In the formula, They are The maximum and minimum values ​​of the sequence.

[0074] Step S3: Reconstruct the distribution network containing distributed generation based on the lizard algorithm, specifically:

[0075] Step S31: The horned lizard hides the imitation behavior, and based on the trigonometric function combination of the selected search agent and the proportional coefficient in the palette, mutates the current optimal solution to generate a new candidate solution.

[0076] Step S32: simulating the blood spraying behavior and movement escape strategy of the lizard, and simulating the darkening of the lizard's skin to adaptively adjust the search strategy.

[0077] Step S33: Calculate the melanocyte rate MSH and replace individuals with low melanocyte rates with new and better solutions.

[0078] Step S34: Determine whether the termination condition is met. If the maximum number of iterations and accuracy conditions are met, output the optimal reconstruction structure of the distribution network corresponding to the optimal solution phasor. Otherwise, repeat steps S31 to S33 until the algorithm meets the termination condition.

[0079] In one embodiment, a mathematical model for distribution network reconstruction is established with the goal of minimizing active power loss and maximizing voltage stability index; the distribution network reconstruction problem containing DG can be defined as a DG optimization problem, which minimizes active power loss in the distribution network and improves the voltage stability index under all operating constraints;

[0080] Therefore, the overall objective function can be written as:

[0081] OF=min(w1OF1+w2OF2) (3)

[0082] Where: w1, w2 are the weight coefficients of the two sub-objective functions, is the active power loss in the distribution network, and is the voltage stability index.

[0083] Among them, the calculation formula for active power loss in the distribution network is:

[0084]

[0085] Where: B is the number of branches; P loss is the energy storage operation cost; R k represents the resistance of the kth branch; I k represents the current in the kth branch.

[0086] The calculation formula of voltage stability index is:

[0087]

[0088] Where: OVSI is the voltage stability index of each bus (except for the relaxed nodes), N represents the number of nodes; VSI is the voltage stability index, and the calculation formula is:

[0089] VSI i+1 =|V i 4 |-4(P i+1 X k -Q i+1 R k )-4(P i+1 X k -Q i+1 R k )|V i 2 | (6)

[0090] Where: X k represents the reactance of the kth branch; V i represents the voltage of node i; P i+1 , Q i+1 represents the active and reactive power at node i.

[0091] In step S1: determine the boundary constraints for solving the distribution network reconstruction problem; the constraints for normal operation of the distribution network include power balance constraints, node voltage constraints, branch current constraints, DG capacity constraints and DG penetration constraints.

[0092] For radial distribution networks, the power balance constraint is:

[0093]

[0094] Where: P s , Q s N is the active and reactive power generated by the generator at the relaxed node; DG is the number of DGs connected to the distribution network; P DG，i , Q DG，i represents the active and reactive power generated by the DG at node i.

[0095] For node i, the upper and lower limit constraints of its node voltage can be expressed as:

[0096] V i,min ≤V i ≤V i,max (2).

[0097] The constraint on the branch current of the kth branch is:

[0098] I k ≤I k,max (3)

[0099] In the formula, I k，maxThe maximum current allowed in the branch.

[0100] DG capacity constraint: The active power and reactive power of DG connected to the distribution network are appropriate effective values. The constraint can be expressed as:

[0101]

[0102] Where P DG，min , P DG，max They represent the minimum and maximum active power generated by DG, Q DG，min , Q DG，max They represent the minimum and maximum values ​​of the reactive power generated by DG respectively.

[0103] The connection of multiple DGs to the grid requires more advanced distribution network control technology. Otherwise, the fluctuation may lead to load shedding, grid disconnection and reduced stability of the distribution network. However, this can be avoided by limiting the penetration level of DG. The penetration rate of DG is selected to be 10% to 60% of the total active load demand in the distribution network. The calculation formula is:

[0104]

[0105] In one embodiment, in step S31, the horned lizard hides the imitation behavior, and based on the trigonometric function combination of the selected search agent and the proportional coefficient in the palette, mutates the current optimal solution to generate a new candidate solution. The process expression is:

[0106]

[0107] In the formula, is the new search agent position (horned lizard) in the solution search space of generation t+1, x best (t) is the best search agent of generation t; r1, r2, r3 and r4 are integer random numbers generated between 1 and the maximum number of the population, where r1-r4 are not equal to each other; x r1 (t), x r2 (t), x r3 (t) and x r4 (t) is the selected r1, r2, r3, r4-th search agent; Max_iter represents the maximum number of iterations (generations), σ is a binary value obtained by Algorithm 1, and c1, c2, where c1≠c2, are random numbers drawn from the containing standardized palette.

[0108] In one embodiment, in step S32, similar blood spraying behavior and mobile escape strategy are two behaviors that simulate lizards escaping from predators, which are used to enhance the search ability of the algorithm, aiming to optimize the global search and local search capabilities in the algorithm, help the algorithm jump out of the local optimum and explore the search space more effectively. The blood spraying behavior trajectory expression is:

[0109]

[0110] In the formula, x i (t+1) is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent found, the current search agent is x i (t), Max_iter represents the maximum number of iterations, t is the current number of iterations, v o is set to 1, a is set to Π / 2, ε is set to 1E-6, and g is gravity.

[0111] The mobile escape strategy expression is:

[0112]

[0113] x i (t+1) is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent to generate t, walk is a random number generated between -1 and 1, c is the mean of the random numbers generated by the standard Cauchy distribution, and σ is set to a number between 0 and 1. i (t) is the i-th search agent in the current t iteration.

[0114] In one embodiment, in step S32, an adaptive search strategy is used to simulate the darkening behavior of lizard skin, allowing the algorithm to adaptively adjust the search strategy according to the quality of the current solution. The skin variable and darkening expression are:

[0115]

[0116]

[0117] Among them, Light1 and Light2 are random numbers generated between Lighthening1 (0 value) and Lighthening2 (0.4046661 value), using normalized values; Dark1 and Dark2 are random numbers generated between Darkening1 (0.5440510 value) Darkening2 (1 value), also using normalized values.

Claims

1. A method for reconfiguring a distribution network containing distributed power sources based on an improved lizard algorithm, characterized in that: The following steps are involved: Step S1: Determine the objective function and constraints of the distribution network reconstruction problem; Step S2: Introduce the lizard algorithm and make improvements. The specific improvements include: The Tent chaotic map is used to initialize the population generation strategy, so that the initial solution is distributed as evenly as possible in the solution space, and the generated chaotic sequence The process is: In the formula, k is the population number; I is the current iteration number; to ensure the randomness of the algorithm initialization information, u takes a random number in the range of [0,1], Further generate the initial positions of the population individuals in the search area The process is: In the formula, They are The maximum and minimum values ​​of the sequence; Step S3: Reconstruct the distribution network containing distributed generation based on the lizard algorithm, specifically: Step S31: The horned lizard hides the imitation behavior, and based on the trigonometric function combination of the selected search agent and the proportional coefficient in the palette, mutates the current optimal solution to generate a new candidate solution; Step S32: simulating the blood spraying behavior and movement escape strategy of the lizard, simulating the blackening of the lizard's skin to adaptively adjust the search strategy; Step S33: Calculate the melanocyte rate MSH and replace individuals with low melanocyte rates with new and better solutions; Step S34: Determine whether the termination condition is met. If the maximum number of iterations and accuracy conditions are met, output the optimal reconstruction structure of the distribution network corresponding to the optimal solution phasor. Otherwise, repeat steps S31 to S33 until the algorithm meets the termination condition.

2. The method for reconfiguring a distribution network containing distributed power sources based on the improved lizard algorithm according to claim 1, characterized in that: In step S1, the objective function is: OF=min(w1OF1+w2OF2) (3) Where: w1, w2 are the weight coefficients of the two sub-objective functions, OF1 is the active power loss in the distribution network, and OF2 is the voltage stability index.

3. The method for reconfiguring a distribution network containing distributed power sources based on the improved lizard algorithm according to claim 2, characterized in that: ； The active power loss calculation formula in the distribution network is: Where: B is the number of branches; P loss is the energy storage operation cost; R k represents the resistance of the kth branch; I k represents the current of the kth branch; The calculation formula of voltage stability index is: Where: OVSI is the voltage stability index of each bus, excluding the relaxed nodes, N represents the number of nodes; VSI is the voltage stability index, and the calculation formula is: VSI i+1 =|V i 4 |-4(P i+1 X k -Q i+1 R k )-4(P i+1 X k -Q i+1 R k )|V i 2 | (6) Where: X k represents the reactance of the kth branch; V i represents the voltage of node i; P i+1 , Q i+1 represents the active and reactive power at node i.

4. The method for reconfiguring a distribution network containing distributed power sources based on the improved lizard algorithm according to claim 1, characterized in that: In step S1, the constraints include power balance constraints, node voltage constraints, branch current constraints, DG capacity constraints, and DG penetration constraints. The power balance constraint is: Where: P s , Q s N is the active and reactive power generated by the generator at the relaxed node; DG is the number of DGs connected to the distribution network; P DG，i , Q DG，i represents the active and reactive power generated by the DG at node i; For node i, the upper and lower limit constraints of its node voltage are expressed as: In i,min ≤V i ≤V i,max (8) The constraint on the kth branch current is: I k ≤I k,max (9) In the formula, I k，max The maximum current value allowed by the branch; The DG capacity constraint is expressed as: Where P DG，min , P DG，max They represent the minimum and maximum active power generated by DG, Q DG，min , Q DG，max They represent the minimum and maximum values ​​of reactive power generated by DG respectively; The penetration rate of DG is selected to be 10% to 60% of the total active load demand in the distribution network; the DG capacity constraint is:

5. The method for reconfiguring a distribution network containing distributed power sources based on the improved lizard algorithm according to claim 1, characterized in that: In step S31, the horned lizard hides its imitation behavior and generates a new candidate solution by mutating the current optimal solution based on the trigonometric function combination of the selected search agent and the proportional coefficient in the palette. The process expression is: [c1(sin(x r1 (t))-cos(x r2 (t))-(-1) s c2(cos(x r3 (t))-sin(x r4 (t)))](12) where is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent of generation t; r1, r2, r3 and r4 are integer random numbers generated between 1 and the maximum number of the population, where r1-r4 are not equal to each other; x r1 (t), x r2 (t), x r3 (t) and x r4 (t) is the selected r1, r2, r3, r4-th search agent; Max_iter represents the maximum number of iterations, σ is a binary value obtained by Algorithm 1, and c1, c2, where c1≠c2, are random numbers drawn from the inclusive standardized palette.

6. The method for reconfiguring a distribution network containing distributed power sources based on the improved lizard algorithm according to claim 1, characterized in that: In step S32, the trajectory expression of the blood ejection behavior of the lizard is: In the formula, x i (t+1) is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent found, the current search agent is x i (t), Max_iter represents the maximum number of iterations, t is the current number of iterations, v o Set to 1, a is set to Π / 2, ε is set to 1E-6, and g is the acceleration due to gravity.

7. The method for reconfiguring a distribution network containing distributed power sources based on the improved lizard algorithm according to claim 1, characterized in that: In step S32, the mobile escape strategy expression is: x i (t+1) is the new search agent position in the solution search space of the t+1th generation, x best (t) is the best search agent to generate t, walk is a random number generated between -1 and 1, c is the mean of the random numbers generated by the standard Cauchy distribution, σ is set to a number between 0 and 1, and x i (t) is the i-th search agent in the current t iteration.

8. The method for reconfiguring a distribution network containing distributed power sources based on the improved lizard algorithm according to claim 1, characterized in that: In step S32, the adaptive search strategy of the lizard skin darkening behavior simulation algorithm is used to allow the algorithm to adaptively adjust the search strategy according to the quality of the current solution. The skin variable and darkening expression are: Among them, Light1 and Light2 are random numbers generated between Lightening1 and Lightening2, using normalized values; Dark1 and Dark2 are random numbers generated between Darkening1Darkening2, using normalized values.

Images