Low-voltage transformer area loss reduction optimization method and device based on particle swarm optimization
By optimizing the weight and learning factor update method in the particle swarm algorithm, and adjusting the update function using perturbation vectors and update matrix, the problem of particles exceeding the search space range is solved, and the algorithm's convergence speed and calculation efficiency of low-voltage platform losses are improved.
Patent Information
- Application Number
- CN202411721081.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-28
- Publication Date
- 2025-02-14
AI Technical Summary
During the loss reduction optimization process of low-voltage table area, the existing particle swarm algorithm has defects in the update weight and learning factors, causing the particles to exceed the search space range, affecting the algorithm's convergence speed.
By optimizing the update methods of inertial weights, cognitive learning factors and social learning factors in the particle swarm algorithm, the parameter update is updated using the update factor and iteration times, and further adjust the update function through the perturbation vector and the update matrix to improve the ability of the particle flight trajectory Control and algorithm convergence speed.
It effectively reduces the occurrence of particles exceeding the search space range, speeds up the convergence speed of the algorithm, and improves the calculation efficiency of low-voltage platform area losses.
Smart Images
Figure CN119443417A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power grid monitoring, and in particular to a low-voltage area loss reduction optimization method and device based on a particle swarm algorithm. Background Art
[0002] Due to the complex topological structure of the low-voltage substation, it is not possible to simply make a judgment based on one or several parameters when optimizing the loss reduction of the low-voltage substation. The particle swarm algorithm is a relatively mature multi-dimensional optimization algorithm. When used in the loss reduction optimization process of the low-voltage substation, it can quickly calculate the loss of the low-voltage substation, and then optimize the loss reduction of the low-voltage substation. However, the existing particle swarm algorithm has defects in the way it updates weights and learning factors during the iteration process. This leads to the problem that under certain complex working conditions, due to inappropriate parameter updates, more particles exceed the scope of the search space. To address this problem, the existing technology usually adopts the method of adding restrictions, such as rebound strategies, boundary truncation, saturation restrictions, etc. to compensate for it, but the effects are not good.
[0003] In summary, there is an urgent need for a low-voltage area loss reduction optimization method based on particle swarm algorithm to calculate and optimize the low-voltage area loss. Summary of the invention
[0004] To solve the above problems, the present invention provides a low-voltage substation loss reduction optimization method and device based on a particle swarm algorithm, which uses an update factor and the current number of iterations to update the current update parameters, thereby reducing the occurrence of particles exceeding the search space range.
[0005] In a first aspect, a low voltage area loss reduction optimization method based on a particle swarm algorithm is provided, the method comprising:
[0006] Using the acquired operating parameters of the low-pressure area, an initial population is generated, wherein each particle in the initial population represents a combination of operating parameters of the low-pressure area;
[0007] Determine the particle fitness function, calculate the fitness of each particle, and determine the individual extreme value of each particle and the overall extreme value of the particle group;
[0008] Using the current update parameters to update the speed and position of the particle, and using the update factor and the current number of iterations to update the current update parameters, wherein the update parameters include: inertia weight, cognitive learning factor and social learning factor; the update factor includes part or all of the inertia weight update factor, cognitive learning update factor and social learning update factor;
[0009] When the iteration number threshold is reached or the fitness of the particle converges, the position of the optimal particle in the entire current iteration process is obtained;
[0010] The low-pressure zone loss is calculated and optimized according to the position of the optimal particles.
[0011] Furthermore, the operating parameters of the low voltage substation include: user category, line category, average load, average power factor and part or all of bus voltage.
[0012] Furthermore, the fitness function is:
[0013]
[0014] Among them, x is the low-voltage operation parameter combination, f i (x) is the objective function of the i-th working condition, k i is the weight value corresponding to the i-th objective function, and n is the total number of objective functions.
[0015] Further, the update factor includes an inertia weight update factor, a cognitive learning update factor, and a social learning update factor;
[0016] Use the update factor and the current number of iterations to update the current update parameters, including:
[0017] Use the following inertia weight update function to update the current inertia weight:
[0018]
[0019] Among them, ω ′ is the updated inertia weight, ω max is the initial maximum inertia weight, ω min is the initial minimum inertia weight, iter is the current iteration number, k ω is the inertia weight update factor;
[0020] Use the following cognitive learning factor update function to update the current cognitive learning factor:
[0021]
[0022] Among them, c 1 is the current cognitive learning factor, c ′ 1 is the updated cognitive learning factor, k c1 Update factors for cognitive learning factors;
[0023] Use the following social learning factor update function to update the current social learning factor:
[0024]
[0025] Among them, c 2 is the current social learning factor, c ′2 is the updated social learning factor, k c2 Updates the factor for the current social learning factor.
[0026] Furthermore, after the current update parameter is updated using the update factor and the current iteration number, the method further includes:
[0027] Generate a disturbance vector according to the motion trend of the particle in the current time step, wherein the elements in the disturbance vector include: the disturbance amount of inertia weight update, the disturbance amount of cognitive learning factor update and the disturbance amount of social learning factor update;
[0028] The update function is updated using the disturbance vector, wherein the update function includes: an inertia weight update function, a cognitive learning factor update function and a social learning factor update function.
[0029] Furthermore, a disturbance vector is generated according to the motion trend of the particle in the current time step, including:
[0030] Cluster the particles according to their positions at the current time step;
[0031] The moving speed of each type of particles is synthesized, and the synthesized speed is assigned to the corresponding cluster center;
[0032] The cluster centers with speed attributes are weighted and merged to obtain a disturbance vector, where the weight value of each cluster center is inversely proportional to the average distance between this cluster center and other cluster centers.
[0033] Furthermore, the update function is updated using the disturbance vector, including:
[0034] A reference coordinate system is established with the highest dimensions of the three update functions, and the three update functions are mapped into the reference coordinate system;
[0035] Map the perturbation vector to the reference coordinate system and apply it to the three update functions to extract the nonlinearly related eigenvectors of the three update functions after perturbation, and use the eigenvectors to generate the update matrix;
[0036] Update the update function using the update matrix.
[0037] Furthermore, the update function is updated using the update matrix, including:
[0038] The update function is updated using the update matrix using the following formula:
[0039]
[0040] Among them, P = [L1, L2...Ln] is the update matrix, |P| is the value of the determinant of the update matrix, L1 is the first eigenvector, L2 is the second eigenvector, and Ln is the nth eigenvector.
[0041] In a second aspect, a low-voltage area loss reduction optimization device based on a particle swarm algorithm is provided, the device comprising: an initial population generation unit, a fitness determination unit, an update unit, an optimal solution determination unit and a calculation unit; wherein:
[0042] An initial population generating unit is used to generate an initial population using the acquired operation parameters of the low-pressure station area, wherein each particle in the initial population represents a combination of operation parameters of the low-pressure station area;
[0043] A fitness determination unit, used to determine the particle fitness function, calculate the fitness of each particle, and determine the individual extreme value of each particle and the overall extreme value of the particle group;
[0044] An updating unit, used to update the velocity and position of the particle using the current update parameters, and to update the current update parameters using the update factor and the current iteration number, wherein the update parameters include: inertia weight, cognitive learning factor and social learning factor; the update factor includes part or all of the inertia weight update factor, cognitive learning update factor and social learning update factor;
[0045] The optimal solution determination unit obtains the position of the optimal particle in the entire current iteration process when the iteration number threshold is reached or the fitness of the particle converges;
[0046] The calculation unit is used to calculate and optimize the low-pressure area loss according to the position of the optimal particle.
[0047] Furthermore, the fitness function is
[0048]
[0049] Among them, x is the low-voltage operation parameter combination, f i (x) is the objective function of the i-th working condition, k i is the weight value corresponding to the i-th objective function, and n is the total number of objective functions.
[0050] Further, the update factor includes an inertia weight update factor, a cognitive learning update factor, and a social learning update factor;
[0051] The updating unit is specifically used to update the current inertia weight using the following inertia weight updating function:
[0052]
[0053] Among them, ω′ is the updated inertia weight, ω max is the initial maximum inertia weight, ω min is the initial minimum inertia weight, iter is the current iteration number, k ω is the inertia weight update factor;
[0054] Use the following cognitive learning factor update function to update the current cognitive learning factor:
[0055]
[0056] Among them, c 1 is the current cognitive learning factor, c ′ 1 is the updated cognitive learning factor, k c1 Update factors for cognitive learning factors;
[0057] Use the following social learning factor update function to update the current social learning factor:
[0058]
[0059] Among them, c 2 is the current social learning factor, c ′ 2 is the updated social learning factor, k c2 is the current social learning factor update factor, and iter is the current iteration number.
[0060] Furthermore, the update unit is also used to generate a disturbance vector according to the movement trend of the particle in the current time step, wherein the elements in the disturbance vector include: the disturbance amount of the inertia weight update, the disturbance amount of the cognitive learning factor update and the disturbance amount of the social learning factor update; and the update function is updated using the disturbance vector.
[0061] Furthermore, the updating unit is specifically used to cluster particles according to the particle positions at the current time step; synthesize the movement speed of each type of particles, and assign the synthesized speed to the corresponding cluster center; and perform weighted merging of the cluster centers with speed attributes to obtain a disturbance vector, wherein the weighted value of each cluster center is inversely proportional to the average distance between the cluster center and other cluster centers.
[0062] Furthermore, the updating unit is specifically used to establish a reference coordinate system with the highest dimension of the three update functions, and map the three update functions into the reference coordinate system;
[0063] Map the perturbation vector to the reference coordinate system and apply it to the three update functions to extract the nonlinearly related eigenvectors of the three update functions after perturbation, and use the eigenvectors to generate the update matrix;
[0064] Update the update function using the update matrix.
[0065] Furthermore, the updating unit is specifically configured to update the update function using the update matrix using the following formula:
[0066]
[0067] Among them, P = [L1, L2...Ln] is the update matrix, |P| is the value of the determinant of the update matrix, L1 is the first eigenvector, L2 is the second eigenvector, and Ln is the nth eigenvector.
[0068] In a third aspect, an electronic device is provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;
[0069] Memory, used to store computer programs;
[0070] The processor is used to implement the steps of the above method when executing the program stored in the memory.
[0071] In a fourth aspect, a computer storage medium is provided, wherein a computer program is stored in the computer storage medium, and when the computer program is executed by a processor, the steps of the above method are implemented.
[0072] Compared with the prior art, the present invention has the following advantages:
[0073] 1. The present invention improves the controllability of particle flight trajectories by optimizing the updating method of weights and learning factors, can reduce the occurrence of particles exceeding the search space range, and accelerates the convergence speed of the algorithm.
[0074] 2. By inputting the optimized disturbance vector, the update function presents multi-dimensional nonlinear changes, which fully exposes the defects of inertia weight, cognitive learning factor and social learning factor, and then uses the update matrix to correct the above defects to improve the controllability of particle flight and accelerate the convergence speed of the algorithm. Using this particle swarm algorithm with fast convergence speed, the loss of low-pressure station area can be quickly calculated.
[0075] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0076] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0077] Figure 1 A flow chart of a low voltage area loss reduction optimization method based on a particle swarm algorithm according to an embodiment of the present disclosure is shown;
[0078] Figure 2 A schematic diagram of a reference coordinate system according to an embodiment of the present disclosure is shown;
[0079] Figure 3 A block diagram of a low-voltage area loss reduction optimization device based on a particle swarm algorithm according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0080] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.
[0081] Figure 1 A schematic diagram of a low voltage area loss reduction optimization method based on a particle swarm algorithm according to an embodiment of the present disclosure is shown. Figure 1 As shown, a low-voltage area loss reduction optimization method based on a particle swarm algorithm in an embodiment of the present disclosure includes the following steps:
[0082] Step 101: Generate an initial population using the acquired operating parameters of the low-pressure station area;
[0083] Among them, each particle in the initial population represents a combination of low-pressure station operation parameters;
[0084] The operating parameters of the low-voltage substation include part or all of the user category, line category, average load, average power factor, and bus voltage.
[0085] Step 102: Determine the particle fitness function, calculate the fitness of each particle, and determine the individual extreme value of each particle and the overall extreme value of the particle group;
[0086] Among them, the fitness function is:
[0087]
[0088] Among them, x is the low-pressure operation parameter combination, that is, the particle, f i (x) is the objective function of the i-th working condition, k i is the weight value corresponding to the i-th objective function, n is the total number of objective functions, different working conditions correspond to different objective functions, the objective function can be selected based on actual conditions, the weight value of each objective function is selected according to the specific working conditions, and the sum of the weight values of each objective function is 1. The embodiment of the present invention does not make specific limitations on it here.
[0089] Step 103: using the current update parameters to update the velocity and position of the particle, and using the update factor and the current number of iterations to update the current update parameters;
[0090] The update parameters include: inertia weight, cognitive learning factor and social learning factor; the update factor includes part or all of the inertia weight update factor, cognitive learning update factor and social learning update factor;
[0091] Specifically, the particle velocity is updated using formula (1):
[0092] v(t+1)=ωv(t)+c 1 [pbest-x(t)]+c 2 [gbest-x(t)] (2) Update the particle position using formula (2):
[0093] x(t+1)=x(t)+v(t+1) (3)
[0094] Among them, v(t+1) is the velocity of the particle in the next time step, v(t) is the velocity of the particle in the current time step, ω is the current inertia weight, and c 1 is the current cognitive learning factor, c 2 is the current social learning factor, pbest is the best historical position of the particle, gbest is the best historical position of all particles, x(t) is the position of the particle at the current time step, and x(t+1) is the position of the particle at the next time step. It should be noted that when the particle speed and position are updated for the first time, ω is the initial inertia weight, c 1 is the initial cognitive learning factor, c 2 is the initial social learning factor.
[0095] After each update of the particle's speed and position, the current inertia weight, the current cognitive learning factor, and the current social learning factor are updated, and the updated inertia weight, the updated cognitive learning factor, and the updated social learning factor are used as the current inertia weight, the current cognitive learning factor, and the current social learning factor for the next particle speed and position update.
[0096] The updating method of inertia weight, cognitive learning factor and social learning factor is:
[0097] Establish the update function of inertia weight, cognitive learning factor and social learning factor:
[0098] The update function of the inertia weight is:
[0099]
[0100] Among them, ω ′ is the updated inertia weight, ω max is the initial maximum inertia weight, ω min is the initial minimum inertia weight, iter is the current iteration number, k ω is the inertia weight update factor;
[0101] The update function of the cognitive learning factor is:
[0102]
[0103] Among them, c 1 is the current cognitive learning factor, c ′ 1 is the updated cognitive learning factor, k c1 Update factors for cognitive learning factors;
[0104] The update function of the social learning factor is:
[0105]
[0106] Among them, c 2 is the current social learning factor, c ′ 2 is the updated social learning factor, k c2 is the current social learning factor update factor, and iter is the current iteration number.
[0107] The dimension of formula (4) is 1 dimension, the dimension of formula (5) is 1 dimension, and the dimension of formula (6) is 1 dimension. The reference coordinate system is established with the highest dimension of the three update functions, which is 3 dimensions. The three update functions are mapped to the reference coordinate system. The established reference coordinate system is as follows: Figure 2 shown.
[0108] Generate a disturbance vector according to the motion trend of the particle in the current time step;
[0109] The disturbance vector is l = (a, b, c), where a is the disturbance amount of the inertia weight update function, b is the disturbance amount of the cognitive learning factor update function, and c is the disturbance amount of the social learning factor update function. Specifically, the particles are clustered according to the particle positions at the current time step, the movement speed of each type of particles is synthesized, the synthesized speed is assigned to the corresponding cluster center, and the cluster centers with speed attributes are weighted and merged to obtain the disturbance vector, where the weighted value of each cluster center is inversely proportional to the average distance between this cluster center and other cluster centers.
[0110] The disturbance vector is mapped to the reference coordinate system and acts on the three update functions. The nonlinear related eigenvectors after the three update functions are disturbed are extracted, so that Lf = λf, where L is the eigenvector, f is the update function, and λ is the eigenvalue corresponding to the eigenvector. The eigenvector is used to generate the update matrix, the update matrix P = [L1, L2…Ln], L1 is the first eigenvector, L2 is the second eigenvector, and Ln is the nth eigenvector. The update function is updated using the update matrix, and the inertia weight, cognitive learning factor, and social learning factor are updated using the updated update function.
[0111] Specifically, the updated update function is:
[0112]
[0113] Among them, P = [L1, L2...Ln] is the update matrix, and n is the number of elements in the update matrix.
[0114] Step 104: Determine whether the iteration number threshold is reached or whether the fitness of the particle converges; if so, execute step 105; otherwise, jump to step 103.
[0115] Here the position of the optimal particle is the optimal solution.
[0116] Step 105: Determine the position of the optimal particle in the entire current iteration process, and calculate and optimize the low-voltage station area loss based on the position of the optimal particle.
[0117] Due to the large number of operation parameters in the low-pressure area, the problem of more particles exceeding the search space range in the process of using the particle swarm algorithm is more prominent. The embodiment of the present invention optimizes the inertia weight and cognitive / social learning factor parameter update method in the particle swarm algorithm, and adds a specific disturbance state to feedback adjust the state of the three process parameters, thereby improving the controllability of the particle flight trajectory and accelerating the convergence speed of the algorithm.
[0118] Based on the above method, the embodiment of the present disclosure also provides a low-voltage area loss reduction optimization device based on a particle swarm algorithm corresponding to the above method. Figure 3 A block diagram of a low voltage area loss reduction optimization device based on a particle swarm algorithm according to an embodiment of the present disclosure is shown. Figure 3 As shown, the low-voltage area loss reduction optimization device based on the particle swarm algorithm includes: an initial population generation unit 31, a fitness determination unit 32, an update unit 33, an optimal solution determination unit 34 and a calculation unit 35; wherein:
[0119] The initial population generating unit 31 is used to generate an initial population using the acquired operation parameters of the low-pressure area, wherein each particle in the initial population represents a combination of operation parameters of the low-pressure area;
[0120] A fitness determination unit 32 is used to determine a particle fitness function, calculate the fitness of each particle, and determine the individual extreme value of each particle and the overall extreme value of the particle group;
[0121] An updating unit 33 is used to update the speed and position of the particle using the current update parameters, and to update the current update parameters using the update factor and the current iteration number, wherein the update parameters include: an inertia weight, a cognitive learning factor, and a social learning factor; and the update factor includes part or all of the inertia weight update factor, the cognitive learning update factor, and the social learning update factor;
[0122] The optimal solution determination unit 34 obtains the position of the optimal particle in the current entire iteration process when the iteration number threshold is reached or the fitness of the particle converges;
[0123] The calculation unit 35 is used to calculate and optimize the low-pressure station area loss according to the position of the optimal particle.
[0124] Preferably, the fitness function is
[0125]
[0126] Among them, x is the low-voltage operation parameter combination, f i (x) is the objective function of the i-th working condition, k i is the weight value corresponding to the i-th objective function, and n is the total number of objective functions.
[0127] Preferably, the update unit 33 is also used to generate a disturbance vector according to the movement trend of the particle in the current time step, wherein the elements in the disturbance vector include: the disturbance amount of the inertia weight update, the disturbance amount of the cognitive learning factor update and the disturbance amount of the social learning factor update; and the update function is updated using the disturbance vector.
[0128] Preferably, the updating unit 33 is specifically used to cluster particles according to the particle positions at the current time step; synthesize the movement speed of each type of particles, and assign the synthesized speed to the corresponding cluster center; perform weighted merging of the cluster centers with speed attributes to obtain a disturbance vector, wherein the weighted value of each cluster center is inversely proportional to the average distance between this cluster center and other cluster centers.
[0129] Preferably, the update factor includes an inertia weight update factor, a cognitive learning update factor, and a social learning update factor;
[0130] The updating unit 33 is specifically used to update the current inertia weight using the following inertia weight updating function:
[0131]
[0132] Among them, ω ′ is the updated inertia weight, ω max is the initial maximum inertia weight, ω min is the initial minimum inertia weight, iter is the current iteration number, k ω is the inertia weight update factor;
[0133] Use the following cognitive learning factor update function to update the current cognitive learning factor:
[0134]
[0135] Among them, c 1 is the current cognitive learning factor, c ′ 1 is the updated cognitive learning factor, k c1 Update factors for cognitive learning factors;
[0136] Use the following social learning factor update function to update the current social learning factor:
[0137]
[0138] Among them, c 2 is the current social learning factor, c ′ 2 is the updated social learning factor, k c2 is the current social learning factor update factor, and iter is the current iteration number.
[0139] Preferably, the updating unit 33 is specifically used to establish a reference coordinate system with the highest dimension of the three update functions, and map the three update functions into the reference coordinate system;
[0140] Map the perturbation vector to the reference coordinate system and apply it to the three update functions to extract the nonlinearly related eigenvectors of the three update functions after perturbation, and use the eigenvectors to generate the update matrix;
[0141] Update the update function using the update matrix.
[0142] Preferably, the updating unit 33 is specifically used to update the update function using the update matrix using the following formula:
[0143]
[0144] Among them, P = [L1, L2...Ln] is the update matrix, |P| is the value of the determinant of the update matrix, and n is the number of elements in the update matrix.
[0145] Based on the same inventive concept as the above disclosed content, the present disclosure also provides an electronic device accordingly. The electronic device of the present disclosure embodiment includes at least one processor and at least one memory electrically connected, the memory is electrically connected to the processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the method as described above.
[0146] It should be noted that the electrical connection between the above-mentioned units does not necessarily mean the connection between lines. An indirect connection method can be applied to the embodiments of the present disclosure as long as the purpose of the present disclosure is achieved.
[0147] Based on the same inventive concept, the present disclosure also provides a computer storage medium, wherein a computer program is stored in the computer storage medium, and the steps of the above method are implemented when the computer program is executed by the processor. Although the present disclosure is described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure.
Claims
1. A low voltage area loss reduction optimization method based on particle swarm algorithm, characterized in that: The method comprises, Using the acquired operating parameters of the low-pressure area, an initial population is generated, wherein each particle in the initial population represents a combination of operating parameters of the low-pressure area; Determine the particle fitness function, calculate the fitness of each particle, and determine the individual extreme value of each particle and the overall extreme value of the particle group; Using the current update parameters to update the speed and position of the particle, and using the update factor and the current number of iterations to update the current update parameters, wherein the update parameters include: inertia weight, cognitive learning factor and social learning factor; the update factor includes part or all of the inertia weight update factor, cognitive learning update factor and social learning update factor; When the iteration number threshold is reached or the fitness of the particle converges, the position of the optimal particle in the entire current iteration process is obtained; The low-pressure zone loss is calculated and optimized according to the position of the optimal particles.
2. The method according to claim 1, characterized in that The operating parameters of the low voltage substation include part or all of the following: user category, line category, average load, average power factor and bus voltage.
3. The method according to claim 1, characterized in that The fitness function is: Among them, x is the low-voltage operation parameter combination, f i (x) is the objective function of the i-th working condition, k i is the weight value corresponding to the i-th objective function, and n is the total number of objective functions.
4. The method according to claim 1, characterized in that: The update factors include inertia weight update factor, cognitive learning update factor and social learning update factor; Use the update factor and the current number of iterations to update the current update parameters, including: Use the following inertia weight update function to update the current inertia weight: Among them, ω ′ is the updated inertia weight, ω max is the initial maximum inertia weight, ω min is the initial minimum inertia weight, iter is the current iteration number, k ω is the inertia weight update factor; Use the following cognitive learning factor update function to update the current cognitive learning factor: Among them, c1 is the current cognitive learning factor, c ′ 1 is the updated cognitive learning factor, k c1 Update factors for cognitive learning factors; Use the following social learning factor update function to update the current social learning factor: Among them, c2 is the current social learning factor, c ′ 2 is the updated social learning factor, k c2 Updates the factor for the current social learning factor.
5. The method according to claim 4, characterized in that After updating the current update parameter using the update factor and the current iteration number, the method further includes: Generate a disturbance vector according to the motion trend of the particle in the current time step, wherein the elements in the disturbance vector include: the disturbance amount of inertia weight update, the disturbance amount of cognitive learning factor update and the disturbance amount of social learning factor update; The update function is updated using the disturbance vector, wherein the update function includes: an inertia weight update function, a cognitive learning factor update function and a social learning factor update function.
6. The method according to claim 5, characterized in that Generate a disturbance vector based on the motion trend of the particle in the current time step, including: Cluster the particles according to their positions at the current time step; The moving speed of each type of particles is synthesized, and the synthesized speed is assigned to the corresponding cluster center; The cluster centers with speed attributes are weighted and merged to obtain a disturbance vector, where the weight value of each cluster center is inversely proportional to the average distance between this cluster center and other cluster centers.
7. The method according to claim 6, characterized in that The update function is updated using the perturbation vector, including: A reference coordinate system is established with the highest dimensions of the three update functions, and the three update functions are mapped into the reference coordinate system; Map the perturbation vector to the reference coordinate system and apply it to the three update functions to extract the nonlinearly related eigenvectors of the three update functions after perturbation, and use the eigenvectors to generate the update matrix; Update the update function using the update matrix.
8. The method according to claim 7, characterized in that Use the update matrix to update the update function, including: The update function is updated using the update matrix using the following formula: Among them, P = [L1, L2...Ln] is the update matrix, |P| is the value of the determinant of the update matrix, L1 is the first eigenvector, L2 is the second eigenvector, and Ln is the nth eigenvector.
9. A low voltage area loss reduction optimization device based on particle swarm algorithm, characterized in that: The device comprises: an initial population generation unit, a fitness determination unit, an update unit, an optimal solution determination unit and a calculation unit; wherein: An initial population generating unit is used to generate an initial population using the acquired operation parameters of the low-pressure station area, wherein each particle in the initial population represents a combination of operation parameters of the low-pressure station area; A fitness determination unit, used to determine the particle fitness function, calculate the fitness of each particle, and determine the individual extreme value of each particle and the overall extreme value of the particle group; An updating unit, used to update the velocity and position of the particle using the current update parameters, and to update the current update parameters using the update factor and the current iteration number, wherein the update parameters include: inertia weight, cognitive learning factor and social learning factor; the update factor includes part or all of the inertia weight update factor, cognitive learning update factor and social learning update factor; The optimal solution determination unit obtains the position of the optimal particle in the entire current iteration process when the iteration number threshold is reached or the fitness of the particle converges; The calculation unit is used to calculate and optimize the low-pressure area loss according to the position of the optimal particle.
10. The device according to claim 9, characterized in that The fitness function is Among them, x is the low-voltage operation parameter combination, f i (x) is the objective function of the i-th working condition, k i is the weight value corresponding to the i-th objective function, and n is the total number of objective functions.
11. The device according to claim 9, characterized in that The update factors include inertia weight update factor, cognitive learning update factor and social learning update factor; The updating unit is specifically used to update the current inertia weight using the following inertia weight updating function: Among them, ω′ is the updated inertia weight, ω max is the initial maximum inertia weight, ω min is the initial minimum inertia weight, iter is the current iteration number, k ω is the inertia weight update factor; Use the following cognitive learning factor update function to update the current cognitive learning factor: Among them, c1 is the current cognitive learning factor, c′1 is the updated cognitive learning factor, k c1 Update factors for cognitive learning factors; Use the following social learning factor update function to update the current social learning factor: Among them, c2 is the current social learning factor, c′2 is the updated social learning factor, and k c2 is the current social learning factor update factor, and iter is the current iteration number.
12. The device according to claim 11, characterized in that The update unit is also used to generate a disturbance vector according to the movement trend of the particle in the current time step, wherein the elements in the disturbance vector include: the disturbance amount of the inertia weight update, the disturbance amount of the cognitive learning factor update and the disturbance amount of the social learning factor update; and the update function is updated using the disturbance vector.
13. The device according to claim 12, characterized in that The updating unit is specifically used to cluster particles according to the particle positions at the current time step; synthesize the movement speed of each type of particles and assign the synthesized speed to the corresponding cluster center; and perform weighted merging of the cluster centers with speed attributes to obtain a disturbance vector, wherein the weighted value of each cluster center is inversely proportional to the average distance between the cluster center and other cluster centers.
14. The device according to claim 13, characterized in that The updating unit is specifically used to establish a reference coordinate system with the highest dimension of the three update functions, and map the three update functions into the reference coordinate system; Map the perturbation vector to the reference coordinate system and apply it to the three update functions to extract the nonlinearly related eigenvectors of the three update functions after perturbation, and use the eigenvectors to generate the update matrix; Update the update function using the update matrix.
15. The device according to claim 14, characterized in that The updating unit is specifically used to update the update function using the update matrix using the following formula: Among them, P = [L1, L2...Ln] is the update matrix, |P| is the value of the determinant of the update matrix, L1 is the first eigenvector, L2 is the second eigenvector, and Ln is the nth eigenvector.
16. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; A processor, for implementing the steps of any method described in claims 1-8 when executing a program stored in a memory.
17. A computer storage medium, characterized in that: The computer storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any method of claims 1-8 are implemented.