Collaborative planning method and system for charging station and 5G communication base station

Abstract

The invention discloses a cooperative planning method and system for a charging station and a 5G communication base station, and relates to the technical field of communication, the cooperative planning method for the charging station and the 5G communication base station comprises the following steps: collecting real-time parameter data of the 5G communication base station and the charging station; establishing a dynamic planning model for collaborative planning of the charging station and the 5G communication base station; solving the dynamic programming model by using an analysis algorithm; and optimizing the preliminary layout and operation strategies of the charging station and the 5G communication base station based on a vector ordinal optimization algorithm. Through the dynamic planning model and the improved suburb wolf optimization algorithm, the layout and operation strategy of the charging station and the 5G communication base station can be effectively optimized, and the charging fault can be predicted in advance, so that a manager has enough time to take countermeasures, equipment is protected from being damaged, the reliability of the system is improved, and the service life of the system is prolonged. And normal operation of the charging station and the 5G communication base station is ensured.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular to a method and system for collaborative planning of charging stations and 5G communication base stations. Background Art

[0002] In recent years, issues such as traffic congestion and traffic safety have received widespread attention. The rapid development of intelligent transportation has placed higher demands on the security and services of the Internet of Vehicles (IoV). Within the IoV, users can communicate with each other and with infrastructure (such as base stations and wireless access points), enabling information transmission and sharing. This, however, has led to an explosive growth in communication demand for IoV, and wireless spectrum resources are insufficient to meet existing application requirements. Existing wireless networks also suffer from uneven deployment, poor communication quality during high-speed vehicle movement, and vehicles moving out of wireless network coverage during driving. For EV (Electric Vehicle) charging stations to participate in V2G (Vehicle to Grid) services such as market frequency regulation, they need to use communication systems to obtain real-time information about EVs docked at charging stations and EVs scheduled for charging on the road. This information is then transmitted to distribution system operators (DSOs), enabling the distribution network control center to formulate effective EV charging strategies that meet EV user charging needs while reducing charging costs and alleviating grid power supply pressure during peak load periods. At the same time, the distribution network center needs to send information such as electricity prices and dispatch instructions to EV charging stations based on the real-time operating status of the distribution network to obtain corresponding auxiliary services.

[0003] In the distribution network, EV charging stations integrate fast-charging equipment and decentralized charging piles, and are connected to the distribution network via 10kV. This connection not only requires the construction of new power supply lines, but also generates significant line losses in some heavily loaded areas. In terms of communication networks, although 5G networks can meet the communication requirements of seamless coverage, high capacity, low latency, and high reliability, they consume a large amount of electricity in ultra-dense networking scenarios of small base stations, resulting in more power losses in the distribution network and even potentially causing overloads on upstream incoming lines of the same voltage level, affecting the safe and stable operation of the distribution network. However, existing technologies cannot effectively optimize the layout and operation strategies of charging stations and 5G communication base stations, making it difficult to achieve optimal resource allocation, reducing overall efficiency, and making it difficult to predict charging failures in advance, leaving managers with insufficient time to take countermeasures and protecting equipment from damage.

[0004] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention proposes a collaborative planning method and system for charging stations and 5G communication base stations, which solves the problems proposed in the above background technology that the existing layout and operation strategies of charging stations and 5G communication base stations cannot be effectively optimized, which is not convenient for achieving the optimal allocation of resources and reduces the overall efficiency. It is also not convenient to predict charging failures in advance, which makes it difficult for managers to take countermeasures and protect equipment from damage.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0007] According to one aspect of the present invention, a method for collaborative planning of charging stations and 5G communication base stations is provided, and the method for collaborative planning of charging stations and 5G communication base stations comprises the following steps:

[0008] S1. Collect real-time parameter data of 5G communication base stations and charging stations, and pre-process the real-time parameter data;

[0009] S2. Establish a dynamic planning model for coordinated planning of charging stations and 5G communication base stations based on the preprocessing results;

[0010] S3. Use the improved coyote optimization algorithm to solve the dynamic programming model and obtain the layout and operation strategy of the preliminary charging stations and 5G communication base stations;

[0011] S4. Optimize the layout and operation strategy of the preliminary charging stations and 5G communication base stations based on the vector order optimization algorithm to obtain the optimal layout and operation strategy;

[0012] S5. Apply the optimal layout and operation strategy in the field, establish a fault prediction model, and predict the occurrence of charging faults at the next moment;

[0013] S6. Based on the prediction results, take countermeasures to ensure the continuity and reliability of the synergy between the charging station and the 5G communication base station.

[0014] Furthermore, collecting real-time parameter data of 5G communication base stations and charging stations and preprocessing the real-time parameter data includes the following steps:

[0015] S11. Deploy data collection nodes in designated areas and obtain real-time parameter data of charging stations and 5G communication base stations through communication protocols;

[0016] S12, obtaining raw data from the data acquisition node, and performing wavelet transform on the raw data, converting the data from the time domain to the frequency domain, and screening out wave field records of each frequency band containing interference waves;

[0017] S13, tracking the interference waves one by one in the wave field records of each frequency band, and obtaining the interference wave direction at the calculation sample point;

[0018] S14, selecting a plurality of windows centered at the calculation sample point to perform median filtering, and recovering the interference signal at the calculation sample point;

[0019] S15, recovering the interference signals in the wave field records of each frequency band one by one, and performing inverse wavelet transform to obtain the entire wave field of the interference wave;

[0020] S16. Subtract the interference wave field from the original data to obtain the effective signal wave field, and generate denoised real-time parameter data based on the effective signal wave field.

[0021] Furthermore, the optimization algorithm is used to solve the dynamic programming model to obtain a preliminary layout and operation strategy for charging stations and 5G communication base stations, which includes the following steps:

[0022] S31. Import the original layout and operation strategy, perform initialization processing, and use the Floyd algorithm to calculate the shortest distance between any two 5G communication base stations;

[0023] S32. Each candidate solution represents a layout and operation strategy. The first N numbers represent the numbers of the selected charging stations and 5G base stations, and the last N numbers represent the set levels, which correspond one-to-one with the positions of the first N numbers.

[0024] S33. Find the global optimal candidate solution according to the upper-level decision-making goal, and add environmental influencing factors to each group of local optimal candidate solutions and group trends to jointly influence the growth of candidate solutions;

[0025] S34, substituting the candidate solution after growth into the Kent mapping, and comparing the Kent candidate solution with the original candidate solution before growth using the lower-level decision objective formula as the evaluation criterion;

[0026] S35. Eliminate poorly performing layouts and operational strategies by comparing lower-level decision-making objectives;

[0027] S36. Use the elite retention strategy to replace poorly performing layout and operation strategies with global suboptimal candidate solutions based on the upper-level decision-making goals;

[0028] S37. Loop through steps S33 to S36. If the number of iterations exceeds the threshold, terminate the iteration and output the optimal candidate solution as the preliminary layout and operation strategy for charging stations and 5G base stations.

[0029] Furthermore, the method of finding the global optimal candidate solution according to the upper-level decision-making goal and adding environmental influencing factors to each group of local optimal candidate solutions and group trends to jointly influence the growth of candidate solutions includes the following steps:

[0030] S331. Define an upper-level decision objective function to evaluate the fitness of candidate solutions;

[0031] S332, initializing a set of candidate solutions and randomly generating the position of each candidate solution;

[0032] S333. Calculate the fitness of each candidate solution according to the objective function;

[0033] S334. Find the global optimal candidate solution as the best solution;

[0034] S335. For each group of optimal candidate solutions, randomly generate an environmental impact factor, and calculate the average position of each group of local optimal candidate solutions as the group trend;

[0035] S336: Substitute the original candidate solution position into the growth formula to calculate the new candidate solution position, and update the position of each candidate solution;

[0036] S337. Repeat steps S333 to S336 until the termination condition is met.

[0037] Furthermore, the step of substituting the grown candidate solution into the Kent mapping and comparing the Kent candidate solution with the original candidate solution before the growth using the lower-level decision objective formula as the evaluation criterion includes the following steps:

[0038] S341. Train and iterate the candidate solution set to obtain a grown candidate solution set;

[0039] S342, performing data normalization processing on the grown candidate solution set to make it in a completely chaotic state;

[0040] S343, randomly generate parameters of the Kent mapping and preset a value range;

[0041] S344. Use the Kent mapping formula to map each candidate solution after growth to generate a new candidate solution individual;

[0042] S345. Compare the fitness of each candidate solution with its Kent mapping result, and retain the candidate solution with the highest fitness as the Kent candidate solution;

[0043] S346, compare the fitness of the Kent candidate solution with the original candidate solution set, and retain the candidate solution set individual with the highest fitness to enter the next generation set;

[0044] S347. Repeat steps S342 to S346 until the iteration termination condition is reached.

[0045] Furthermore, the growth formula is:

[0046]

[0047] in, Indicates the position of the new candidate solution after growth;

[0048] represents the original candidate solution position;

[0049] r1 represents the weight of the environmental impact factor after normalization;

[0050] r2 represents the normalized weight of the random weight influencing factor of growth factor δ1;

[0051] r3 represents the normalized weight of the random weight influencing factor of growth factor δ2;

[0052] p represents the pth candidate solution;

[0053] t represents the tth iteration;

[0054] c represents the position coordinate of the cth dimension;

[0055] x represents the position vector of the candidate solution.

[0056] Furthermore, the layout and operation strategy of the preliminary charging stations and 5G communication base stations are optimized based on the vector order optimization algorithm. The optimal layout and operation strategy includes the following steps:

[0057] S41. Collect the preliminary layout and operation strategies of charging stations and 5G communication base stations as the initial solution set;

[0058] S42. Establish a multi-objective planning model for charging stations and 5G communication base stations, including objective functions and constraints;

[0059] S43. Generate a set of feasible solutions through a heuristic algorithm based on the constraints of the model;

[0060] S44, constructing a rough evaluation function and performing a rough evaluation on the feasible solution;

[0061] S45. Obtaining a sequence curve of feasible solutions based on the rough evaluation results;

[0062] S46, selecting the solution of the front section of the sequence curve as the selected set and evaluating it;

[0063] S47. Apply the vector order optimization algorithm to the selected set to obtain the Pareto optimal frontier;

[0064] S48. Select the solution with the best comprehensive indicators from the Pareto optimal frontier as the optimal solution for the layout and operation strategy of charging stations and 5G communication base stations.

[0065] Furthermore, applying the optimal layout and operation strategy in the field, establishing a fault prediction model, and predicting the occurrence of charging faults at the next moment include the following steps:

[0066] S51. In actual scenarios, deploy charging stations and 5G base stations according to the optimal layout and operation strategy;

[0067] S52. Collect real-time operation data and status information of the charging station and the 5G communication base station as sample data;

[0068] S53. Extracting time domain features and frequency domain features of the charging station and the 5G communication base station based on the sample data;

[0069] S54, labeling the feature data into two categories: normal and fault, and using a decision tree algorithm to build an initial fault prediction model;

[0070] S55. Use the training set to train the initial model and obtain the accuracy of each decision tree;

[0071] S56. Calculate the differences between decision trees and optimize the model using an optimization algorithm to obtain a random forest model;

[0072] S57. Input the time domain features and frequency domain features of the charging station and the 5G communication base station into the random forest model, and predict the probability of a charging station failure at the next moment based on the feature data.

[0073] Furthermore, the differences between decision trees are calculated and the model is optimized using an optimization algorithm. The random forest model is obtained by the following steps:

[0074] S561, initialize the optimization algorithm, set the initial temperature and initial layout strategy;

[0075] S562, repeating steps S563 to S566 according to the current temperature;

[0076] S563: Randomly perturb the current layout strategy to generate a new layout strategy;

[0077] S564, calculating the difference between the prediction accuracy of the new layout strategy and the prediction accuracy of the current layout strategy;

[0078] S565. If the prediction accuracy of the new layout strategy is higher than that of the current layout strategy, the new layout strategy is accepted as the current layout strategy. Otherwise, the acceptance probability of the new layout strategy is calculated according to the Metropolis rule. If the acceptance probability is greater than the random number, the new layout strategy is accepted as the current layout strategy. Otherwise, the current layout strategy is retained.

[0079] S566: If the set termination condition is met, the current layout strategy is output as the optimal layout strategy. The termination condition is that the new layout strategy is not accepted in several consecutive Metropolis chains or the end temperature is reached. Otherwise, the temperature is decayed by the decay function and the process returns to step S562.

[0080] S567. Repeat steps S563 to S566 to select N prediction models that meet the preset prediction accuracy from the initial prediction models and combine them into a new random forest model as the fault prediction model.

[0081] According to another aspect of the present invention, a charging station and 5G communication base station collaborative planning system is provided, which includes: an information acquisition and processing module, a charging facility planning and construction module, a model solving module, a multi-objective planning optimization module, a fault prediction module, and a fault handling and maintenance module;

[0082] Among them, the information acquisition and processing module is used to collect real-time parameter data of 5G communication base stations and charging stations, and process the real-time parameter data;

[0083] The charging facility planning and construction module is used to establish a dynamic planning model for the coordinated planning of charging stations and 5G communication base stations based on the preprocessing results;

[0084] The model solving module is used to solve the dynamic programming model using analytical algorithms to obtain preliminary layout and operation strategies for charging stations and 5G communication base stations;

[0085] A multi-objective planning optimization module is used to optimize the layout and operation strategy of the preliminary charging stations and 5G communication base stations based on the vector order optimization algorithm to obtain the optimal layout and operation strategy;

[0086] The fault prediction module is used to apply the optimal layout and operation strategy in the field, establish a fault prediction model, and predict the occurrence of charging faults at the next moment;

[0087] The fault handling and maintenance module is used to take countermeasures based on the prediction results to ensure the continuity and reliability of the synergy between the charging station and the 5G communication base station.

[0088] The beneficial effects of the present invention are:

[0089] 1. Through the dynamic programming model and the improved coyote optimization algorithm, the present invention can effectively optimize the layout and operation strategy of charging stations and 5G communication base stations, achieve optimal resource allocation, improve overall efficiency, and predict charging failures in advance, so that managers have enough time to take countermeasures and protect equipment from damage. As a result, the optimized layout and operation strategy of charging stations and 5G communication base stations can better cope with various emergencies, improve the reliability of the system, and ensure the normal operation of charging stations and 5G communication base stations.

[0090] 2. The present invention uses the improved coyote optimization algorithm to solve the problem, so that various factors and constraints of the system can be better considered, rather than just considering the optimization of a certain aspect, so as to obtain a comprehensive and reasonable layout and operation strategy for charging stations and 5G communication base stations. The preliminary layout and operation strategy obtained can effectively achieve the precise positioning of charging stations and 5G communication base stations, improve resource utilization efficiency, and save construction and operation costs.

[0091] 3. By applying the vector order optimization algorithm, the present invention can further optimize the layout and operation strategy of charging stations and 5G communication base stations, further achieve optimal resource allocation, save resources, and improve efficiency. In addition, by using the vector order optimization algorithm to adjust and optimize the layout and operation strategy, the pressure on a single base station or charging station can be reduced, the network load can be balanced, and the stability and operation efficiency of the network can be increased, thereby promoting the development of collaborative planning technology for charging stations and 5G communication base stations.

[0092] 4. By establishing a fault prediction model, the present invention can predict potential charging faults in advance, so that countermeasures can be formulated before the fault occurs, greatly reducing the risk of fault occurrence and the cost of handling the fault. In addition, by predicting potential charging faults, operation and maintenance personnel can promptly inspect and maintain the equipment, thereby preventing damage to the equipment and extending the service life of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0094] Figure 1 This is a flowchart of a method for collaborative planning of charging stations and 5G communication base stations according to an embodiment of the present invention. DETAILED DESCRIPTION

[0095] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0096] In the description of the present invention, unless otherwise specified, "plurality" means two or more. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0097] According to an embodiment of the present invention, a method and system for collaborative planning of charging stations and 5G communication base stations are provided.

[0098] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the charging station and 5G communication base station collaborative planning method of an embodiment of the present invention, the charging station and 5G communication base station collaborative planning method includes the following steps:

[0099] S1. Collect real-time parameter data of 5G communication base stations and charging stations, and pre-process the real-time parameter data;

[0100] Specifically, the parameter information includes the location, capacity, connector type, load curve of the charging station, the location, coverage range, channel parameters, user distribution, transportation network within the area, power load distribution information, user distribution and mobility patterns within the area, and related constraints on candidate construction sites for charging stations and 5G base stations.

[0101] S2. Establish a dynamic planning model for coordinated planning of charging stations and 5G communication base stations based on the preprocessing results;

[0102] Specifically, the dynamic programming model includes:

[0103] Determine status: The status may include the number, location, and operating status of charging stations, as well as the number, location, and operating status of 5G communication base stations.

[0104] Determine decisions: Decisions may include where to build a new charging station or 5G base station, whether to shut down an operating charging station or 5G base station, etc.

[0105] Determine the state transition function: The state transition function describes how to get to the new state based on the current state and the decision. For example, if the decision is to build a new charging station somewhere, then the number and location of charging stations will change.

[0106] Determine the reward function: The reward function describes the reward (or benefit) obtained by making a certain decision in a certain state. For example, if the newly built charging station can improve the service coverage, then the reward may be the increased number of people served.

[0107] S3. Use the improved coyote optimization algorithm to solve the dynamic programming model and obtain the layout and operation strategy of the preliminary charging stations and 5G communication base stations;

[0108] S4. Optimize the layout and operation strategy of the preliminary charging stations and 5G communication base stations based on the vector order optimization algorithm to obtain the optimal layout and operation strategy;

[0109] S5. Apply the optimal layout and operation strategy in the field, establish a fault prediction model, and predict the occurrence of charging faults at the next moment;

[0110] S6. Based on the prediction results, take countermeasures to ensure the continuity and reliability of the synergy between the charging station and the 5G communication base station.

[0111] In one embodiment, collecting real-time parameter data of 5G communication base stations and charging stations and preprocessing the real-time parameter data includes the following steps:

[0112] S11. Deploy data collection nodes in designated areas and obtain real-time parameter data of charging stations and 5G communication base stations through communication protocols;

[0113] S12, obtaining raw data from the data acquisition node, and performing wavelet transform on the raw data, converting the data from the time domain to the frequency domain, and screening out wave field records of each frequency band containing interference waves;

[0114] S13, tracking the interference waves one by one in the wave field records of each frequency band, and obtaining the interference wave direction at the calculation sample point;

[0115] S14, selecting a plurality of windows centered at the calculation sample point to perform median filtering, and recovering the interference signal at the calculation sample point;

[0116] S15, recovering the interference signals in the wave field records of each frequency band one by one, and performing inverse wavelet transform to obtain the entire wave field of the interference wave;

[0117] S16. Subtract the interference wave field from the original data to obtain the effective signal wave field, and generate denoised real-time parameter data based on the effective signal wave field.

[0118] In one embodiment, solving the dynamic programming model using an analytical algorithm to obtain a preliminary layout and operation strategy for charging stations and 5G communication base stations includes the following steps:

[0119] S31. Import the original layout and operation strategy, perform initialization processing, and use the Floyd algorithm to calculate the shortest distance between any two 5G communication base stations;

[0120] Specifically, the Floyd algorithm is a classic algorithm for searching for the shortest path between any two points in a weighted graph. It uses dynamic programming ideas to gradually improve the path to obtain the shortest path length, splits the path through intermediate nodes, and compares and selects the better indirect path.

[0121] Specifically, the goal of selecting optimal indirect paths in Floyd's algorithm is to find the shortest path between two points. Initially, only the distances between directly connected vertices are known, without any knowledge of possible shorter paths through intermediate vertices. Therefore, selecting optimal indirect paths isn't about directly using them, but rather about using them to update and refine our understanding of direct paths, ultimately achieving the shortest path.

[0122] S32. Each candidate solution represents a layout and operation strategy. The first N numbers represent the numbers of the selected charging stations and 5G base stations, and the last N numbers represent the set levels, which correspond one-to-one with the positions of the first N numbers.

[0123] S33. Find the global optimal candidate solution according to the upper-level decision-making goal, and add environmental influencing factors to each group of local optimal candidate solutions and group trends to jointly influence the growth of candidate solutions;

[0124] S34, substituting the candidate solution after growth into the Kent mapping, and comparing the Kent candidate solution with the original candidate solution before growth using the lower-level decision objective formula as the evaluation criterion;

[0125] S35. Eliminate poorly performing layouts and operational strategies by comparing lower-level decision-making objectives;

[0126] S36. Use the elite retention strategy to replace poorly performing layout and operation strategies with global suboptimal candidate solutions based on the upper-level decision-making goals;

[0127] Specifically, the elite retention strategy itself is to replace the worst solution with the optimal solution, avoiding the elimination of excellent solutions during the optimization process, and achieving the purpose of retaining the genes of excellent solutions. Therefore, the elite selection strategy is adjusted to name the suboptimal wolf before the entire population grows as Beta wolf, and use Beta wolf to replace the wolf with the worst social adaptability in the grown wolf pack, thereby improving search efficiency while retaining the genes of excellent solutions in the population.

[0128] S37. Loop through steps S33 to S36. If the number of iterations exceeds the threshold, terminate the iteration and output the optimal candidate solution as the preliminary layout and operation strategy for charging stations and 5G base stations.

[0129] Specifically, the analysis algorithm aims to improve the coyote optimization algorithm. The traditional coyote optimization algorithm uses coyote growth as the primary means of obtaining new solutions during optimization. The coyote optimization algorithm groups the entire coyote population into groups for growth. During this growth process, the group's alpha wolves and group cultural trends (cults) guide the growth of coyotes within each group. This significantly limits the diversity of the COA coyote population and information exchange within the population. This results in low global traversal during the coyote growth process. Furthermore, due to the low birth rate and low population diversity of COA coyotes, the algorithm's exploration capabilities are poor, making it prone to falling into local optima. This present invention primarily improves the coyote growth model (i.e., the update formula in this invention) within the coyote optimization algorithm.

[0130] In one embodiment, the method of finding the global optimal candidate solution according to the upper-level decision-making goal and adding environmental influencing factors to each group of local optimal candidate solutions and group trends to jointly influence the growth of candidate solutions includes the following steps:

[0131] S331. Define an upper-level decision objective function to evaluate the fitness of candidate solutions;

[0132] S332, initializing a set of candidate solutions and randomly generating the position of each candidate solution;

[0133] S333. Calculate the fitness of each candidate solution according to the objective function;

[0134] S334. Find the global optimal candidate solution as the best solution;

[0135] S335. For each group of optimal candidate solutions, randomly generate an environmental impact factor, and calculate the average position of each group of local optimal candidate solutions as the group trend;

[0136] S336: Substitute the original candidate solution position into the growth formula to calculate the new candidate solution position, and update the position of each candidate solution;

[0137] S337. Repeat steps S333 to S336 until the termination condition is met.

[0138] In one embodiment, the growth formula is:

[0139]

[0140] in, Indicates the position of the new candidate solution after growth;

[0141] represents the original candidate solution position;

[0142] r1 represents the weight of the environmental impact factor after normalization;

[0143] r2 represents the normalized weight of the random weight influencing factor of growth factor δ1;

[0144] r3 represents the normalized weight of the random weight influencing factor of growth factor δ2;

[0145] p represents the pth candidate solution;

[0146] t represents the tth iteration;

[0147] c represents the position coordinate of the cth dimension;

[0148] x represents the position vector of the candidate solution.

[0149] Substituting the grown candidate solution into the Kent mapping and comparing the Kent candidate solution with the original candidate solution before growing using the lower-level decision objective formula as the evaluation criterion includes the following steps:

[0150] S341. Train and iterate the candidate solution set to obtain a grown candidate solution set;

[0151] S342, performing data normalization processing on the grown candidate solution set to make it in a completely chaotic state;

[0152] Specifically, normalization: Normalization is to unify the dimensions of the data to make different data comparable. Normalizing the prediction results helps to eliminate the impact of the data dimensions and facilitates subsequent weighted processing and summation calculations.

[0153] S343, randomly generate parameters of the Kent mapping and preset the value range (0, 1);

[0154] Specifically, the Kent map is a chaotic map, and when the value is 0.5, it is a symmetric chaotic map.

[0155] S344. Use the Kent mapping formula to map each candidate solution after growth to generate a new candidate solution individual;

[0156] S345. Compare the fitness of each candidate solution with its Kent mapping result, and retain the candidate solution with the highest fitness as the Kent candidate solution;

[0157] S346, compare the fitness of the Kent candidate solution with the original candidate solution set, and retain the candidate solution set individual with the highest fitness to enter the next generation set;

[0158] S347. Repeat steps S342 to S346 until the iteration termination condition is reached.

[0159] In one embodiment, optimizing the layout and operation strategy of the preliminary charging stations and 5G communication base stations based on the vector order optimization algorithm to obtain the optimal layout and operation strategy includes the following steps:

[0160] S41. Collect the preliminary layout and operation strategies of charging stations and 5G communication base stations as the initial solution set;

[0161] S42. Establish a multi-objective planning model for charging stations and 5G communication base stations, including objective functions and constraints;

[0162] Specifically, the objective function takes into account costs, benefits, environmental impacts, etc.

[0163] Specifically, the constraints include: geographical and environmental constraints (including geographical location restrictions of charging stations and 5G base stations, such as the minimum distance from residential areas and industrial areas, avoiding environmentally sensitive areas, etc.), technical and equipment constraints (technical specifications of charging stations and 5G base stations, such as power capacity, coverage, signal strength, etc.), energy supply constraints (energy supply issues of charging stations, such as grid access capabilities, renewable energy utilization, etc.), economic and cost constraints (including economic factors such as construction costs, operating costs, and maintenance costs of charging stations and base stations), and social and market demand constraints (based on market research, considering social factors such as user needs, acceptance, and frequency of use).

[0164] S43. Generate a set of feasible solutions through a heuristic algorithm based on the constraints of the model;

[0165] S44, constructing a rough evaluation function and performing a rough evaluation on the feasible solution;

[0166] S45. Obtaining a sequence curve of feasible solutions based on the rough evaluation results;

[0167] S46, selecting the solution of the front section of the sequence curve as the selected set and evaluating it;

[0168] S47. Apply the vector order optimization algorithm to the selected set to obtain the Pareto optimal frontier;

[0169] Specifically, the Pareto optimal front is a mapping of the set of all feasible Pareto optimal solutions in a multi-objective optimization problem in the objective space.

[0170] S48. Select the solution with the best comprehensive indicators from the Pareto optimal frontier as the optimal solution for the layout and operation strategy of charging stations and 5G communication base stations.

[0171] In one embodiment, applying the optimal layout and operation strategy in the field, establishing a fault prediction model, and predicting the occurrence of a charging fault at the next moment includes the following steps:

[0172] S51. In actual scenarios, deploy charging stations and 5G base stations according to the optimal layout and operation strategy;

[0173] S52. Collect real-time operation data and status information of the charging station and the 5G communication base station as sample data;

[0174] S53. Extracting time domain features and frequency domain features of the charging station and the 5G communication base station based on the sample data;

[0175] Specifically, time-domain features are features extracted from the real-time operating data of the charging station according to the time series, reflecting the changes in the charging station status over time. For example, the statistical characteristics of data such as voltage, current, and temperature.

[0176] Specifically, frequency domain features: By performing Fourier transforms on the charging station operation data, the frequency domain information extracted reflects the intensity distribution of different frequency components in the signal. This can reflect the periodic pattern of the charging station.

[0177] S54, labeling the feature data into two categories: normal and fault, and using a decision tree algorithm to build an initial fault prediction model;

[0178] Specifically, the characteristics of normal feature data are as follows: the time domain feature values ​​are stable, without large fluctuations or deviations. The frequency domain features do not contain abnormal frequency components. All feature parameters are within the normal design range and historical statistical range.

[0179] Specifically, the characteristics of fault-related feature data include: drastic fluctuations in time domain characteristics, deviating from the normal range; abnormal high-frequency or low-frequency components in the frequency domain; and key feature parameters exceeding the normal design range and historical statistical range.

[0180] S55. Use the training set to train the initial fault prediction model and obtain the accuracy of each decision tree;

[0181] S56. Calculate the differences between decision trees and optimize the model using an optimization algorithm to obtain a random forest model;

[0182] S57. Input the time domain features and frequency domain features of the charging station and the 5G communication base station into the random forest model, and predict the probability of a charging station failure at the next moment based on the feature data.

[0183] Specifically, the optimization algorithm uses simulated annealing: The fitness value of an initial solution tree is randomly selected as the initial solution. Using a series of Markov chains generated by a decreasing control parameter t, a new solution generation device and acceptance rules are used. The process of "generating a new solution, calculating the objective function difference, determining whether to accept the new solution, and then accepting or rejecting the new solution" is repeated, continuously iterating the current solution. When the algorithm concludes, the optimal solution tree is output. This cycle is repeated, and a solution tree with high detection performance is selected from the initial forest to form a solution forest.

[0184] In one embodiment, calculating the differences between decision trees and optimizing the model using an optimization algorithm to obtain a random forest model includes the following steps:

[0185] S561, initialize the optimization algorithm, set the initial temperature and initial layout strategy;

[0186] S562, repeating steps S563 to S566 according to the current temperature;

[0187] S563: Randomly perturb the current layout strategy (i.e., the current decision tree combination) to generate a new layout strategy (new decision tree combination);

[0188] S564, calculating the difference between the prediction accuracy of the new layout strategy and the prediction accuracy of the current layout strategy;

[0189] S565. If the prediction accuracy of the new layout strategy is higher than that of the current layout strategy (i.e., the difference is less than 0), the new layout strategy is accepted as the current layout strategy. Otherwise, the acceptance probability of the new layout strategy is calculated according to the Metropolis rule. If the acceptance probability is greater than the random number, the new layout strategy is accepted as the current layout strategy. Otherwise, the current layout strategy is retained.

[0190] Specifically, the Metropolis rule is an acceptance-rejection criterion used in Monte Carlo simulation. The basic idea of ​​this rule is: if the new state has lower energy (or higher probability) than the current state, then accept the new state; otherwise, accept the new state with a probability that is the negative exponent of the energy difference (or probability ratio) between the new state and the current state.

[0191] S566: If the set termination condition is met, the current layout strategy is output as the optimal layout strategy. The termination condition is that the new layout strategy is not accepted in several consecutive Metropolis chains or the end temperature is reached. Otherwise, the temperature is decayed by the decay function and the process returns to step S562.

[0192] Specifically, the formula for attenuation of the attenuation function is:

[0193]

[0194] Among them, m s Represents the layout strategy at the nth iteration step;

[0195] m0 represents the initial layout strategy.

[0196] S567. Repeat steps S563 to S566 to select N prediction models that meet the preset prediction accuracy from the initial prediction models and combine them into a new random forest model as the fault prediction model.

[0197] According to another embodiment of the present invention, a charging station and 5G communication base station collaborative planning system is also provided, the charging station and 5G communication base station collaborative planning system comprising: an information acquisition and processing module, a charging facility planning and construction module, a model solving module, a multi-objective planning optimization module, a fault prediction module, and a fault handling and maintenance module;

[0198] Among them, the information acquisition and processing module is used to collect real-time parameter data of 5G communication base stations and charging stations, and process the real-time parameter data;

[0199] The charging facility planning and construction module is used to establish a dynamic planning model for the coordinated planning of charging stations and 5G communication base stations based on the preprocessing results;

[0200] The model solving module is used to solve the dynamic programming model using analytical algorithms to obtain preliminary layout and operation strategies for charging stations and 5G communication base stations;

[0201] A multi-objective planning optimization module is used to optimize the layout and operation strategy of the preliminary charging stations and 5G communication base stations based on the vector order optimization algorithm to obtain the optimal layout and operation strategy;

[0202] The fault prediction module is used to apply the optimal layout and operation strategy in the field, establish a fault prediction model, and predict the occurrence of charging faults at the next moment;

[0203] The fault handling and maintenance module is used to take countermeasures based on the prediction results to ensure the continuity and reliability of the synergy between the charging station and the 5G communication base station.

[0204] In summary, with the help of the above technical solutions of the present invention, the present invention uses the improved coyote optimization algorithm for solution, so that it can better consider the various factors and constraints of the system, rather than just considering the optimization of a certain aspect, thereby obtaining a comprehensive and reasonable layout and operation strategy for charging stations and 5G communication base stations, and the preliminary layout and operation strategy obtained can effectively achieve the precise positioning of charging stations and 5G communication base stations, improve resource utilization efficiency, and save construction and operation costs. The present invention applies the vector order optimization algorithm, so that the layout and operation strategy of charging stations and 5G communication base stations can be further optimized, further achieving the optimal allocation of resources, saving resources, and improving efficiency. The vector order optimization algorithm is used to adjust and optimize the layout and operation strategy, which can reduce the pressure on a single base station or charging station, balance the network load, and increase the stability and operation efficiency of the network, thereby promoting the development of collaborative planning technology for charging stations and 5G communication base stations. By establishing a fault prediction model, the present invention can predict potential charging faults in advance, so that countermeasures can be formulated before the fault occurs, greatly reducing the risk of fault occurrence and the cost of handling the fault. By predicting potential charging faults, operation and maintenance personnel can promptly inspect and maintain the equipment, thereby preventing damage to the equipment and extending the service life of the equipment.

[0205] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for collaborative planning of charging stations and 5G communication base stations, characterized in that: The collaborative planning method of the charging station and the 5G communication base station includes the following steps: S1. Collect real-time parameter data of 5G communication base stations and charging stations, and pre-process the real-time parameter data; S2. Establish a dynamic planning model for coordinated planning of charging stations and 5G communication base stations based on the preprocessing results; S3. Solve the dynamic programming model using analytical algorithms to obtain a preliminary layout and operation strategy for charging stations and 5G communication base stations; S4. Optimize the layout and operation strategy of the preliminary charging stations and 5G communication base stations based on the vector order optimization algorithm to obtain the optimal layout and operation strategy; S5. Apply the optimal layout and operation strategy in the field, establish a fault prediction model, and predict the occurrence of charging faults at the next moment; S6. Based on the prediction results, take countermeasures to ensure the continuity and reliability of the synergy between the charging station and the 5G communication base station.

2. A method for collaborative planning of charging stations and 5G communication base stations according to claim 1, characterized in that: The collecting of real-time parameter data of 5G communication base stations and charging stations and preprocessing of the real-time parameter data includes the following steps: S11. Deploy data collection nodes in designated areas and obtain real-time parameter data of charging stations and 5G communication base stations through communication protocols; S12, obtaining raw data from the data acquisition node, and performing wavelet transform on the raw data, converting the data from the time domain to the frequency domain, and screening out wave field records of each frequency band containing interference waves; S13, tracking the interference waves one by one in the wave field records of each frequency band, and obtaining the interference wave direction at the calculation sample point; S14, selecting a plurality of windows centered at the calculation sample point to perform median filtering, and recovering the interference signal at the calculation sample point; S15, recovering the interference signals in the wave field records of each frequency band one by one, and performing inverse wavelet transform to obtain the entire wave field of the interference wave; S16. Subtract the interference wave field from the original data to obtain the effective signal wave field, and generate denoised real-time parameter data based on the effective signal wave field.

3. The method for collaborative planning of charging stations and 5G communication base stations according to claim 1, characterized in that: Solving the dynamic programming model using an optimization algorithm to obtain a preliminary layout and operation strategy for charging stations and 5G communication base stations includes the following steps: S31. Import the original layout and operation strategy, perform initialization processing, and use the Floyd algorithm to calculate the shortest distance between any two 5G communication base stations; S32. Each candidate solution represents a layout and operation strategy. The first N numbers represent the numbers of the selected charging stations and 5G base stations, and the last N numbers represent the set levels, which correspond one-to-one with the positions of the first N numbers. S33. Find the global optimal candidate solution according to the upper-level decision-making goal, and add environmental influencing factors to each group of local optimal candidate solutions and group trends to jointly influence the growth of candidate solutions; S34, substituting the candidate solution after growth into the Kent mapping, and comparing the Kent candidate solution with the original candidate solution before growth using the lower-level decision objective formula as the evaluation criterion; S35. Eliminate poorly performing layouts and operational strategies by comparing lower-level decision-making objectives; S36. Use the elite retention strategy to replace poorly performing layout and operation strategies with global suboptimal candidate solutions based on the upper-level decision-making goals; S37. Loop through steps S33 to S36. If the number of iterations exceeds the threshold, terminate the iteration and output the optimal candidate solution as the preliminary layout and operation strategy for charging stations and 5G base stations.

4. A charging station and 5G communication base station collaborative planning method according to claim 3, characterized in that: The method of finding the global optimal candidate solution according to the upper-level decision-making goal and adding environmental influencing factors to each group of local optimal candidate solutions and group trends to jointly influence the growth of candidate solutions includes the following steps: S331. Define an upper-level decision objective function to evaluate the fitness of candidate solutions; S332, initializing a set of candidate solutions and randomly generating the position of each candidate solution; S333. Calculate the fitness of each candidate solution according to the objective function; S334. Find the global optimal candidate solution as the best solution; S335. For each group of optimal candidate solutions, randomly generate environmental impact factors, and calculate the average position of each group of local optimal candidate solutions as the group trend; S336: Substitute the original candidate solution position into the growth formula to calculate the new candidate solution position, and update the position of each candidate solution; S337. Repeat steps S333 to S336 until the termination condition is met.

5. A method for collaborative planning of charging stations and 5G communication base stations according to claim 4, characterized in that: Substituting the grown candidate solution into the Kent mapping and comparing the Kent candidate solution with the original candidate solution before growing using the lower-level decision objective formula as the evaluation criterion includes the following steps: S341. Train and iterate the candidate solution set to obtain a grown candidate solution set; S342, performing data normalization processing on the grown candidate solution set to make it in a completely chaotic state; S343, randomly generate parameters of the Kent mapping and preset a value range; S344. Use the Kent mapping formula to map each candidate solution after growth to generate a new candidate solution individual; S345. Compare the fitness of each candidate solution with its Kent mapping result, and retain the candidate solution with the highest fitness as the Kent candidate solution; S346, compare the fitness of the Kent candidate solution with the original candidate solution set, and retain the candidate solution set individuals with the highest fitness to enter the next generation set; S347. Repeat steps S342 to S346 until the iteration termination condition is reached.

6. A charging station and 5G communication base station collaborative planning method according to claim 4, characterized in that: The growth formula is: in, Indicates the position of the new candidate solution after growth; represents the original candidate solution position; r1 represents the weight of the environmental impact factor after normalization; r2 represents the normalized weight of the random weight influencing factor of growth factor δ1; r3 represents the normalized weight of the random weight influencing factor of growth factor δ2; p represents the pth candidate solution; t represents the tth iteration; c represents the position coordinate of the cth dimension; x represents the position vector of the candidate solution.

7. The method for collaborative planning of charging stations and 5G communication base stations according to claim 1, characterized in that: The vector order optimization algorithm is used to optimize the layout and operation strategy of the preliminary charging stations and 5G communication base stations to obtain the optimal layout and operation strategy, which includes the following steps: S41. Collect the preliminary layout and operation strategies of charging stations and 5G communication base stations as the initial solution set; S42. Establish a multi-objective planning model for charging stations and 5G communication base stations, including objective functions and constraints; S43. Generate a set of feasible solutions through a heuristic algorithm based on the constraints of the model; S44, constructing a rough evaluation function and performing a rough evaluation on the feasible solution; S45. Obtaining a sequence curve of feasible solutions based on the rough evaluation results; S46, selecting the solution of the front section of the sequence curve as the selected set and evaluating it; S47. Apply the vector order optimization algorithm to the selected set to obtain the Pareto optimal frontier; S48. Select the solution with the best comprehensive indicators from the Pareto optimal frontier as the optimal solution for the layout and operation strategy of charging stations and 5G communication base stations.

8. The method for collaborative planning of charging stations and 5G communication base stations according to claim 1, characterized in that: Applying the optimal layout and operation strategy in the field, establishing a fault prediction model, and predicting the occurrence of a charging fault at the next moment includes the following steps: S51. In actual scenarios, deploy charging stations and 5G base stations according to the optimal layout and operation strategy; S52. Collect real-time operation data and status information of the charging station and the 5G communication base station as sample data; S53. Extracting time domain features and frequency domain features of the charging station and the 5G communication base station based on the sample data; S54, labeling the feature data into two categories: normal and fault, and using a decision tree algorithm to build an initial fault prediction model; S55. Use the training set to train the initial model and obtain the accuracy of each decision tree; S56. Calculate the differences between decision trees and optimize the model using an optimization algorithm to obtain a random forest model; S57. Input the time domain features and frequency domain features of the charging station and the 5G communication base station into the random forest model, and predict the probability of a charging station failure at the next moment based on the feature data.

9. A charging station and 5G communication base station collaborative planning method according to claim 8, characterized in that: The calculation of the differences between decision trees and the optimization of the model using an optimization algorithm to obtain a random forest model includes the following steps: S561, initialize the optimization algorithm, set the initial temperature and initial layout strategy; S562, repeating steps S563 to S566 according to the current temperature; S563: Randomly perturb the current layout strategy to generate a new layout strategy; S564, calculating the difference between the prediction accuracy of the new layout strategy and the prediction accuracy of the current layout strategy; S565. If the prediction accuracy of the new layout strategy is higher than that of the current layout strategy, the new layout strategy is accepted as the current layout strategy. Otherwise, the acceptance probability of the new layout strategy is calculated according to the Metropolis rule. If the acceptance probability is greater than the random number, the new layout strategy is accepted as the current layout strategy. Otherwise, the current layout strategy is retained. S566: If the set termination condition is met, the current layout strategy is output as the optimal layout strategy. The termination condition is that the new layout strategy is not accepted in several consecutive Metropolis chains or the end temperature is reached. Otherwise, the temperature is decayed by the decay function and the process returns to step S562. S567. Repeat steps S563 to S566 to select N prediction models that meet the preset prediction accuracy from the initial prediction models and combine them into a new random forest model as the fault prediction model.

10. A charging station and 5G communication base station collaborative planning system, used to implement the charging station and 5G communication base station collaborative planning method according to any one of claims 1 to 9, characterized in that: The charging station and 5G communication base station collaborative planning system includes: information acquisition and processing module, charging facility planning and construction module, model solution module, multi-objective planning optimization module, fault prediction module and fault handling and maintenance module; The information acquisition and processing module is used to collect real-time parameter data of 5G communication base stations and charging stations, and process the real-time parameter data; The charging facility planning and construction module is used to establish a dynamic planning model for the coordinated planning of charging stations and 5G communication base stations based on the preprocessing results; The model solving module is used to solve the dynamic programming model using an analytical algorithm to obtain a preliminary layout and operation strategy for charging stations and 5G communication base stations; The multi-objective planning optimization module is used to optimize the layout and operation strategy of the preliminary charging stations and 5G communication base stations based on the vector order optimization algorithm to obtain the optimal layout and operation strategy; The fault prediction module is used to apply the optimal layout and operation strategy in the field, establish a fault prediction model, and predict the occurrence of charging faults at the next moment; The fault handling and maintenance module is used to take countermeasures based on the prediction results to ensure the continuity and reliability of the synergy between the charging station and the 5G communication base station.

Images