Safety situation awareness method and system for frequency conversion load of charging pile

Abstract

The application provides a charging pile variable frequency load safety situation perception method and system, obtains electric energy meter voltage data connected with the charging pile; obtains the safety situation perception result of the charging pile variable frequency load according to the obtained electric energy meter voltage data and a pre-trained BP neural network model; wherein the weight optimization of the BP neural network model is carried out by using a particle swarm optimization algorithm, the inertia factor of the particle swarm optimization algorithm is obtained according to the maximum value and the minimum value of the inertia factor, the iteration number, the maximum iteration number and the transformation coefficient; the local and global search ability and the convergence speed of the particle swarm algorithm are improved, the particle swarm algorithm is prevented from falling into local optimization and premature convergence, the performance of the particle swarm algorithm is improved, and the perception accuracy and the perception efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of charging pile technology, and in particular to a method and system for sensing the safety status of variable frequency loads in charging piles. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Situational awareness is divided into three different stages, such as Figure 1 As shown: This includes: cognition: acquiring environmental information elements and integrating and processing the information; understanding: analyzing and processing the integrated and processed information to generate the current situation; prediction: predicting future environmental change trends based on the current situation.

[0004] There are many indicators that can affect the frequency conversion load status of charging piles. Selecting the status indicators with typical data and forming the data source for subsequent status link can provide reliable data support for the next step of evaluation and prediction. The construction of the indicator system needs to refer to certain principles. Specifically, the construction of the charging pile frequency conversion load status indicator system is carried out from different dimensions: (1) Hierarchical classification principle: the charging pile frequency conversion load status indicators are hierarchical. These indicators have different meanings for different charging piles and the processing is different. Therefore, they should be considered hierarchically and classified. (2) Similarity principle: there are many influencing factors to be considered in the perception of the charging pile frequency conversion load status. However, there are many similar, related and overlapping indicator data. Similar indicators should be included in the unified consideration.

[0005] The inventors discovered that existing methods for sensing the variable frequency load situation of charging piles suffer from problems such as difficulty in charge monitoring and lack of low-voltage topology, which restrict the development of variable frequency load situation sensing for charging piles. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for safety situation perception of charging pile variable frequency loads, achieving more accurate and efficient charging pile situation perception.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] The first aspect of this invention provides a method for sensing the safety status of variable frequency loads in charging piles.

[0009] A method for safety status perception of frequency conversion load in charging piles includes the following processes:

[0010] Obtain voltage data from the electricity meter connected to the charging pile;

[0011] Based on the acquired electricity meter voltage data and the pre-trained BP neural network model, the safety situation perception results of the charging pile frequency conversion load are obtained.

[0012] Among them, the particle swarm optimization algorithm is used to optimize the weights of the BP neural network model. The inertia factor of the particle swarm optimization algorithm is obtained based on the maximum and minimum values ​​of the inertia factor, the number of iterations, the maximum number of iterations, and the transformation coefficients.

[0013] As an optional implementation of the first aspect of the present invention, the inertia factor w of the particle swarm optimization algorithm is obtained based on the maximum and minimum values ​​of the inertia factor, the number of iterations, the maximum number of iterations, and the transformation coefficients, including:

[0014]

[0015] Among them, w max and w min Let T be the maximum and minimum values ​​of the inertia factor, respectively, and t be the iteration number. max β represents the maximum number of iterations, and β is the transformation coefficient of the inertia factor.

[0016] As an optional implementation of the first aspect of the present invention, in the particle swarm optimization algorithm, a threshold value is set. When the current iteration number is less than the threshold value, the position of each particle is summed with the global extreme value and averaged. When the current iteration number is greater than or equal to the threshold value, a random perturbation term is added to disturb the global extreme value.

[0017] As a further limitation of the first aspect of the present invention, k = 0.5 × (1 + rand);

[0018] Where k is a random perturbation term and rand is a random number.

[0019] As one optional implementation of the first aspect of the present invention, the particle swarm optimization algorithm includes:

[0020] Population initialization;

[0021] The fitness value of each particle is calculated using the objective function, thus obtaining the individual extreme value and the global extreme value.

[0022] Determine if the current iteration count is less than the threshold. If so, update the particle's position based on the global extremum; otherwise, add a perturbation term to the global extremum.

[0023] Determine if the termination condition is met. If it is, the iteration will stop; otherwise, return to the step of calculating the fitness value of each particle using the objective function.

[0024] As a further limitation of the first aspect of the invention, updating the particle position based on the global extremum includes:

[0025]

[0026] Among them, P 新 P(t) represents the particle position after the update at the t-th iteration, P(t) represents the particle position before the update at the t-th iteration, and Gbest(t) represents the global extremum at the t-th iteration.

[0027] As an optional implementation of the first aspect of the present invention, the weight optimization of the BP neural network model is performed using a particle swarm optimization algorithm, including:

[0028] Initialize various parameters, including population size, maximum and minimum values ​​of inertia factor, learning factor, number of neurons in each layer of the neural network, target error, and particle position and velocity range;

[0029] The particle swarm for initializing the particle swarm optimization algorithm can be randomly generated based on the number of neurons in the BP neural network, including the position and velocity vectors of each particle.

[0030] Input the training sample set, use the error function of the BP neural network as the fitness function to calculate the fitness of the particles, and determine the individual extreme value of the particles and the global extreme value of the population from the fitness value;

[0031] Iteratively update the position and velocity of the particles, calculate the fitness value after each iteration, and adjust the individual extreme value and global extreme value of the particles according to the fitness value until the termination condition is met;

[0032] The final globally optimal solution is used as the weights of the BP neural network.

[0033] A second aspect of the present invention provides a safety situation awareness system for the frequency conversion load of a charging pile.

[0034] A safety situation awareness system for the frequency conversion load of a charging pile includes:

[0035] The data acquisition module is configured to acquire voltage data from the electricity meter connected to the charging pile.

[0036] The situation awareness module is configured to obtain the safety situation awareness results of the charging pile's frequency conversion load based on the acquired electricity meter voltage data and the pre-trained BP neural network model.

[0037] Among them, the particle swarm optimization algorithm is used to optimize the weights of the BP neural network model. The inertia factor of the particle swarm optimization algorithm is obtained based on the maximum and minimum values ​​of the inertia factor, the number of iterations, the maximum number of iterations, and the transformation coefficients.

[0038] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the safety situation perception method for frequency conversion load of charging piles as described in the first aspect of the present invention.

[0039] A fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the safety situation perception method for frequency conversion load of charging piles as described in the first aspect of the present invention.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] 1. This invention innovatively proposes a safety situation perception method and system for the frequency conversion load of charging piles. It uses a BP neural network optimized by particle swarm optimization to perceive the safety situation of the frequency conversion load of charging piles, thereby improving the accuracy and efficiency of perception.

[0042] 2. This invention innovatively proposes a safety situation perception method and system for the variable frequency load of charging piles. Based on the maximum and minimum values ​​of the inertia factor, the number of iterations, the maximum number of iterations, and the transformation coefficient, the inertia factor of the particle swarm optimization algorithm is obtained, which improves the local and global search capabilities and convergence speed of the particle swarm algorithm, avoids the particle swarm algorithm from getting trapped in local optima and premature convergence, and improves the performance of the particle swarm algorithm.

[0043] 3. This invention innovatively proposes a safety situation perception method and system for the frequency conversion load of charging piles. A threshold is set. When the current iteration number is less than the threshold, the position of each particle is summed with the global extreme value and averaged. This appropriately reduces the overall search capability of the population while improving the local search capability. When the current iteration number is greater than or equal to the threshold, a random perturbation term is added to interfere with the global extreme value, thereby improving the global search capability of the population.

[0044] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0045] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0046] Figure 1 This is a schematic diagram illustrating the three different stages of situational awareness provided in the background technology.

[0047] Figure 2 This is a flowchart illustrating the safety situation perception method for the frequency conversion load of a charging pile provided in Embodiment 1 of the present invention.

[0048] Figure 3 This is a schematic diagram of the prediction results provided in Embodiment 1 of the present invention;

[0049] Figure 4 This is a schematic diagram of the safety situation perception system for the frequency conversion load of a charging pile provided in Embodiment 2 of the present invention. Detailed Implementation

[0050] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0051] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0052] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0053] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0054] Example 1:

[0055] like Figure 2 As shown, Embodiment 1 of the present invention provides a method for safety situation awareness of the frequency conversion load of charging piles. Based on the combination of particle swarm optimization algorithm and BP neural network, it can monitor the voltage status of all charging piles under a transformer. Utilizing the communication resources between the concentrator and the electricity meter, the concentrator performs edge computing to directly collect the electricity meter voltage data and determine whether the charging pile is experiencing low voltage. This avoids the data traffic waste caused by the main station system constantly calling the electricity meter voltage data and can save the communication costs incurred by the concentrator uploading data to the electricity information collection system. Through the routine monitoring of charging pile voltage, low voltage management is accelerated, the efficiency of power grid operation and the quality of service are improved, and high-quality power grid development is strongly supported.

[0056] Specifically, it includes the following processes:

[0057] Obtain voltage data from the electricity meter connected to the charging pile;

[0058] Based on the acquired electricity meter voltage data and the pre-trained BP neural network model, the safety situation perception results of the charging pile frequency conversion load are obtained.

[0059] Particle Swarm Optimization (PSO) is an evolutionary algorithm based on swarm evolution and an optimization algorithm based on iteration and swarm intelligence. Assuming the particle swarm size is N and the search space dimension is D, where D also represents the number of network weights, the position of the i-th particle in the D-dimensional space is denoted as P. i =(p i1 ,p i2 ,…,p iD Let i = 1, 2, ..., N, and let V be the velocity and position of the i-th particle. i =(v i1 ,v i2 ,…,v iD Then the best extremum that the i-th particle can currently find in the D-dimensional search space is Pbest. i =(pbest i1 pbest i2 ,…,pbest iD );

[0060] The optimal position of the population globally is denoted as Gbes = (gbest1, gbest2, ..., gbest) D In the D-dimensional solution space, in each iteration, the velocity and position of each particle change with its own individual extreme position and the current global optimal position. The expressions for velocity update and position update are as shown in equations (1) and (2):

[0061] V i (t+1)=wV i (t)+c1r1(Pbest i (t)-P i (t))+c2r2(Gbest(t)-P i (t)) (1)

[0062] P i (t+1)=P i (t)+V i (t+1) (2)

[0063] Where t is the number of iterations, w is the inertia factor balancing global and local search capabilities, c1 and c2 are learning factors, generally positive constants, ranging from [0, 4], and r1 and r2 represent random numbers between 0 and 1; V i ∈[-v max ,v max ], P i ∈[-p max ,p max ], v max and p maxThese represent the maximum value of the particle velocity and the maximum value of the particle position, respectively, and are used to limit the particle velocity and position.

[0064] It should be noted that: Equation (1) is the standard particle swarm optimization algorithm (SPSO), and the population size N is determined according to the requirements. Generally, the population size N is 20 to 40; when w = 0, the speed depends only on the current position and the historical best position, and the speed itself has no memory; when w is large, the algorithm has a strong global search capability; when w is small, the algorithm has a strong local search capability; the maximum speed of the particle v max Determines the accuracy of the region between the current position and the optimal position, when v max When v is too large, the particle risks exceeding its minimum value. max When the size is too small, the particle will only search within the local minimum region, and the particle will get stuck in a local optimum.

[0065] Because the PSO algorithm suffers from low optimization accuracy, poor local optimization ability, and a tendency to premature convergence, many researchers have proposed improvements to enhance its local and global search capabilities and convergence speed. These improvements aim to prevent the algorithm from getting trapped in local optima and converging prematurely, thereby improving its overall performance. The improvements to the Particle Swarm Optimization algorithm in this embodiment are shown below:

[0066] (1) Improve the inertia factor of the PSO algorithm

[0067] The optimization process of the PSO algorithm is nonlinear, so this paper presents a strategy for decreasing the inertia factor. The mathematical expression of the improved inertia factor is shown in equation (3):

[0068]

[0069] Among them, w max and w min These represent the maximum and minimum values ​​of the inertia factor, respectively; t represents the number of iterations, T max β represents the maximum number of iterations; β represents the transformation coefficient that changes the inertia factor.

[0070] (2) Improve the population diversity of the PSO algorithm

[0071] This embodiment improves the diversity of the population by maintaining global search capability in the early stages of the algorithm while appropriately enhancing local search capability; in the later stages of the algorithm, it appropriately enhances global search capability to prevent the algorithm from getting trapped in local minima.

[0072] The specific improvement strategy is: setting a threshold T. When the current number of iterations t is less than T, the average is obtained by summing the position of each particle and the global extremum. This currently properly reduces the global overall search capability of the population while improving the local search capability of the population; when t is greater than or equal to T, a random disturbance term k is added to disturb the global extremum, so as to appropriately improve the global search capability of the population. The expression of k is shown in formula (4):

[0073] k=0.5×(1+rand) (4)

[0074] The specifically improved PSO algorithm is as follows:

[0075] Step 1: Initialize the population, and randomly generate a population with a scale of N;

[0076] Step 2: Calculate the fitness value of each particle through the objective function, and obtain the individual extremum and the global extremum;

[0077] Step 3: Count and determine whether the current number of iterations is less than the threshold T. If the result is yes, update the position of the particle according to the global extremum, that is:

[0078]

[0079] If the current number of iterations is greater than the threshold T, add a disturbance term to the global extremum, that is:

[0080]

[0081] Then according to P i (t+1)=P i (t)+V i (t+1) to update.

[0082] Step 4: Determine whether the termination condition is satisfied. If the condition is satisfied, the iteration stops; otherwise, return to Step 2.

[0083] The PSO algorithm and the BP neural network algorithm are two different optimization algorithms, and these two algorithms are also applicable to different optimization problems. In the PSO algorithm, populations cooperate with each other and multiple particles perform synchronous search in a multi-dimensional space. If one particle falls into a local minimum, it needs to jump out of the local minimum driven by other particles;

[0084] The BP neural network algorithm has only one particle, so it is easier to fall into a local optimal solution. The optimization of weights by this algorithm starts from a certain point in the multi-dimensional space and performs local search through gradient information. Therefore, combining these two algorithms can effectively improve the classification accuracy, generalization capability and learning capability of the network;

[0085] The global search capability of the PSO algorithm can optimize the topology and connection weights of the BP neural network; in addition, the error function of the BP neural network will be used as the fitness function of the PSO algorithm.

[0086] The main steps of the BP neural network optimized based on the improved particle swarm optimization algorithm are as follows:

[0087] Step 1: Initialize various parameters, including population size, maximum and minimum values ​​of inertia factor, learning factor, number of neurons in each layer of the neural network, target error, and particle position and velocity range, and construct the neural network;

[0088] Step 2: Initialize the particle swarm for the PSO algorithm: The position and velocity vectors of each particle can be randomly generated based on the number of neurons in the BP neural network;

[0089] Step 3: Input the training sample set, use the error function of the BP neural network as the fitness function to calculate the fitness of the particles, and determine the individual extreme value of the particles and the global extreme value of the population based on the fitness value;

[0090] Step 4: Iteratively update the position and velocity of the particles, and calculate the fitness value after each iteration. Adjust the individual extreme value and global extreme value of the particles according to the fitness value until the termination condition is met.

[0091] Step 5: Use the final globally optimal solution as the weights of the BP neural network and input it into the test samples to test the network and obtain the situational awareness results.

[0092] Experiments were conducted using the collected data, and the results are as follows: Figure 3 As shown in the figure. It can be seen that the method described in this embodiment has a good predictive effect, and the prediction results are close to the actual values. The prediction results can be obtained from the prediction data, where MAE = 0.0286 and RMSE = 0.0312. It can be seen that the situation prediction based on the method described in this embodiment has high accuracy, and the prediction results are basically consistent with the actual situation results, which has a certain degree of feasibility.

[0093] The predicted and expected values ​​obtained by the method described in this embodiment are shown in Table 1. It can be seen that the prediction results of the method described in this embodiment are basically consistent with the actual situation results, demonstrating its feasibility.

[0094] Table 1: Predicted Values ​​and Expected Values

[0095] 1 0.49 0.51 8 0.68 0.71 2 0.50 0.45 9 0.71 0.67 3 0.53 0.50 10 0.54 0.50 4 0.54 0.59 11 0.43 0.45 5 0.55 0.56 12 0.61 0.63 6 0 / 62 0.59 13 0.57 0.58 7 0.55 0.53 14 0.43 0.46

[0096] Example 2:

[0097] like Figure 4As shown, Embodiment 2 of the present invention provides a safety situation awareness system for the frequency conversion load of a charging pile, comprising:

[0098] The data acquisition module is configured to acquire voltage data from the electricity meter connected to the charging pile.

[0099] The situation awareness module is configured to obtain the safety situation awareness results of the charging pile's frequency conversion load based on the acquired electricity meter voltage data and the pre-trained BP neural network model.

[0100] Among them, the particle swarm optimization algorithm is used to optimize the weights of the BP neural network model. The inertia factor of the particle swarm optimization algorithm is obtained based on the maximum and minimum values ​​of the inertia factor, the number of iterations, the maximum number of iterations, and the transformation coefficients.

[0101] The working method of the system is the same as the safety situation perception method of the charging pile frequency conversion load provided in Example 1, and will not be repeated here.

[0102] Example 3:

[0103] Embodiment 3 of the present invention provides a computer-readable storage medium storing a program thereon, which, when executed by a processor, implements the steps in the safety situation perception method for the frequency conversion load of a charging pile as described in Embodiment 1 of the present invention.

[0104] Example 4:

[0105] Embodiment 4 of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the safety situation perception method for the frequency conversion load of the charging pile as described in Embodiment 1 of the present invention.

[0106] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for safety status perception of frequency conversion load in charging piles, characterized in that, Includes the following processes: Obtain voltage data from the electricity meter connected to the charging pile; Based on the acquired electricity meter voltage data and the pre-trained BP neural network model, the safety situation perception results of the charging pile frequency conversion load are obtained. In this study, the particle swarm optimization algorithm was used to optimize the weights of the BP neural network model. The inertia factor of the particle swarm optimization algorithm was obtained based on the maximum and minimum values ​​of the inertia factor, the number of iterations, the maximum number of iterations, and the transformation coefficients. w ,include: in, and These are the maximum and minimum values ​​of the inertia factor, respectively. For the number of iterations, The maximum number of iterations, The transformation coefficients for the inertia factor; In the particle swarm optimization algorithm, a threshold value is set. When the current iteration number is less than the threshold value, the position of each particle is summed with the global extremum and the average is taken. When the current iteration number is greater than or equal to the threshold value, a random perturbation term is added to disturb the global extremum. The particle swarm optimization algorithm includes: population initialization; calculating the fitness value of each particle using an objective function to obtain individual extreme values ​​and a global extreme value; determining whether the current iteration count is less than a threshold; if so, updating the particle position based on the global extreme value; otherwise, adding a perturbation term to the global extreme value; determining whether the termination condition is met; if so, the iteration stops; otherwise, returning to the step of calculating the fitness value of each particle using the objective function; wherein, updating the particle position based on the global extreme value includes: in, The particle position is updated at the t-th iteration. The position of the particle before the update at the t-th iteration. This is the global extremum at the t-th iteration.

2. The safety status perception method for the frequency conversion load of charging piles as described in claim 1, characterized in that, in, k For random disturbance terms, rand It is a random number.

3. A safety situation awareness system for the frequency conversion load of a charging pile, characterized in that, include: The data acquisition module is configured to acquire voltage data from the electricity meter connected to the charging pile. The situation awareness module is configured to obtain the safety situation awareness results of the charging pile's frequency conversion load based on the acquired electricity meter voltage data and the pre-trained BP neural network model. In this study, the particle swarm optimization algorithm was used to optimize the weights of the BP neural network model. The inertia factor of the particle swarm optimization algorithm was obtained based on the maximum and minimum values ​​of the inertia factor, the number of iterations, the maximum number of iterations, and the transformation coefficients. w ,include: in, and These are the maximum and minimum values ​​of the inertia factor, respectively. For the number of iterations, The maximum number of iterations, The transformation coefficients for the inertia factor; In the particle swarm optimization algorithm, a threshold value is set. When the current iteration number is less than the threshold value, the position of each particle is summed with the global extremum and the average is taken. When the current iteration number is greater than or equal to the threshold value, a random perturbation term is added to disturb the global extremum. The particle swarm optimization algorithm includes: population initialization; calculating the fitness value of each particle using an objective function to obtain individual extreme values ​​and a global extreme value; determining whether the current iteration count is less than a threshold; if so, updating the particle position based on the global extreme value; otherwise, adding a perturbation term to the global extreme value; determining whether the termination condition is met; if so, the iteration stops; otherwise, returning to the step of calculating the fitness value of each particle using the objective function; wherein, updating the particle position based on the global extreme value includes: in, The particle position is updated at the t-th iteration. The position of the particle before the update at the t-th iteration. This is the global extremum at the t-th iteration.

4. The safety situation awareness system for the frequency conversion load of charging piles as described in claim 3, characterized in that, in, k For random disturbance terms, rand It is a random number.

5. A computer-readable storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the safety situation awareness method for the frequency conversion load of the charging pile as described in any one of claims 1-2.

6. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the safety situation perception method for the frequency conversion load of the charging pile as described in any one of claims 1-2.

Images