Shared charging pile recommendation method based on dynamic Huffman model and bilateral matching

Through the dynamic Huffman model and bilateral matching algorithm, the frequency model and evaluation coefficient of charging piles are built, and the recommendation of shared charging piles is optimized, which solves the problems of insufficient matching accuracy and user experience in the existing system, and achieves more accurate and personalized charging Pile recommendation improves charging experience and safety.

CN118917607AInactive Publication Date: 2024-11-08国网甘肃省电力公司甘南供电公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411001978.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-25
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing shared charging pile recommendation system has limited matching accuracy, failing to fully consider user needs and comprehensive performance of charging piles, and lacking dynamic optimization, resulting in inaccurate or outdated recommendation results and insufficient user experience.

Method used

A shared charging pile recommendation method based on dynamic Huffman model and bilateral matching is adopted. The usage frequency model of charging piles is constructed through dynamic Huffman trees, and the charging pile evaluation coefficient and user needs are optimized to match to generate a personalized recommendation list.

Benefits of technology

It improves the utilization rate and user experience of charging piles, ensures the safety of the charging process, updates the status of the charging piles in real time, provides more accurate and timely recommendation results, and adapts to different user needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917607A_ABST
    Figure CN118917607A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of charging pile recommendation, and particularly relates to a shared charging pile recommendation method based on a dynamic Huffman model and bilateral matching. Comprising the steps of dynamic Huffman model construction, user demand analysis and charging pile bilateral matching, charging pile evaluation coefficient quantification, dynamic Huffman coding, bilateral matching optimization and recommendation result generation, summarizing the calculated optimal matching degree of each charging pile, sorting all candidate charging piles according to the optimal matching degree, and performing recommendation on the candidate charging piles. And a recommendation list is created, the recommendation list is sorted according to the optimal matching degree of the charging piles from high to low, the sorted recommendation list is displayed to the user, and the user can select the most suitable charging pile according to own requirements.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of charging pile recommendation, and in particular relates to a shared charging pile recommendation method based on a dynamic Huffman model and bilateral matching. Background Art

[0002] With the popularity of electric vehicles, shared charging piles are an important infrastructure, and their intelligent recommendation system is crucial to improving charging convenience and efficiency. The field of intelligent charging pile recommendation technology focuses on how to achieve efficient allocation of charging pile resources and meet users' personalized needs through algorithm optimization.

[0003] Existing shared charging pile recommendation systems often have the following shortcomings: limited matching accuracy. Many systems only consider basic matching conditions, such as the idle status and compatibility of the charging piles, and fail to fully consider the specific needs of users and the comprehensive performance of the charging piles; lack of dynamic optimization. Some systems fail to update the usage status and performance evaluation of the charging piles in real time, resulting in inaccurate or outdated recommendation results; insufficient user experience. Existing technologies may fail to provide personalized recommendation lists. Users still need to manually screen among multiple options, which increases the user's operational burden.

[0004] In view of the above-mentioned shortcomings, the prior art has proposed some corresponding solutions and the prior art may have taken the following measures:

[0005] A basic matching algorithm is used to achieve basic matching of charging piles with user needs, such as idle time and compatibility check.

[0006] A simple scoring system is used to score charging piles based on simple indicators such as the frequency of use of charging piles, and charging piles with high scores are recommended.

[0007] However, these technologies still have some shortcomings. The basic matching algorithm ignores important factors such as charging efficiency and safety. The simple scoring system may fail to comprehensively consider multiple evaluation coefficients and lacks a comprehensive evaluation of the performance of charging piles.

[0008] To this end, we propose a shared charging pile recommendation method based on dynamic Huffman model and bilateral matching. Summary of the invention

[0009] In view of the above deficiencies in the prior art, the technical problem to be solved by the present invention is to provide a shared charging pile recommendation method and system based on a dynamic Huffman model and bilateral matching, aiming to improve the utilization rate of charging piles and the user charging experience while ensuring the safety of the charging process.

[0010] In order to achieve the above objectives, the present invention adopts the following technical solutions:

[0011] The shared charging pile recommendation method based on the dynamic Huffman model and bilateral matching includes the following steps:

[0012] Step S1. Dynamic Huffman model construction, collecting status information of charging piles, calculating the usage frequency of each charging pile, and using the calculated usage frequency to construct a dynamic Huffman model. In the dynamic Huffman model, the charging pile is used as a leaf node and the usage frequency is used as a weight;

[0013] Step S2. Analyze user demand and match charging piles bilaterally, obtain the user's charging request, and use the charging pile status information collected in step S1 and the obtained user charging request to perform preliminary matching using a bilateral matching algorithm;

[0014] Step S3. Quantify the charging pile evaluation coefficients, define the charging pile evaluation coefficients, and quantify each coefficient, and use the quantified coefficients to assign a comprehensive score to each charging pile. The charging pile evaluation coefficients include usage frequency, charging efficiency, and safety factor;

[0015] Step S4. Dynamic Huffman coding, using the dynamic Huffman tree constructed in step S1 to encode the charging pile, converting the code into a comparable numerical value to indicate the priority of the charging pile;

[0016] Step S5. Bilateral matching optimization, combining the evaluation coefficients in step S3 and the encoding results in step S4 to optimize the bilateral matching;

[0017] Step S6. Generate recommendation results, summarize the optimization matching degree of each charging pile calculated in step S5, use the obtained optimization matching degree to sort all candidate charging piles, create a recommendation list, and sort the recommendation list from high to low according to the optimization matching degree of the charging piles. The sorted recommendation list is displayed to the user, and the user can choose the most suitable charging pile according to his or her needs.

[0018] Furthermore, in step S1, the method for constructing a dynamic Huffman tree includes:

[0019] Step S11. Regularly collect status information of each charging pile, including charging power, current occupancy status (whether idle or occupied), historical usage data, etc.;

[0020] Step S12: Calculate the usage frequency. For each charging pile i, calculate the usage frequency of the charging pile i during a specific observation period. The calculation formula for the usage frequency is:

[0021]

[0022] Among them, f i is the frequency of use; N iis the number of charging times, which is the total number of times charging pile i is used during the observation period; T is the total observation time, which is the time span of data collection;

[0023] Step S13. Dynamic Huffman tree construction: using the usage frequency of each charging pile i as a weight, a dynamic Huffman tree is constructed. Each leaf node in the dynamic Huffman tree represents a charging pile, and the internal nodes represent a combination of charging piles. During the construction process, the charging piles (or charging pile combinations) with the lowest usage frequency are gradually merged to form new nodes. This process continues until a single root node is formed.

[0024] Step S14: Priority coding. In the Huffman tree constructed in the above steps, the coding of each charging pile i is determined according to the position of each charging pile i in the tree. A shorter coding path indicates a higher frequency of use and priority.

[0025] Furthermore, in step S2, the step of preliminary matching by the bilateral matching algorithm includes:

[0026] Step S21. Demand acquisition, extracting the real-time status information of the charging pile from the dynamic Huffman tree model constructed in step S1;

[0027] Step S22. bilateral matching algorithm design, using the bilateral matching algorithm to match user needs with the status information of the charging pile, and considering the availability, compatibility and time requirements of the charging pile during the matching process;

[0028] Among them, availability refers to the idle state of the charging pile within the time window required by the user, compatibility refers to whether the charging pile supports the battery type of the user's electric vehicle, and time requirement refers to the time window specified in the user's charging request;

[0029] The matching algorithm logic is;

[0030] For each charging station, first check whether it is compatible with the battery type of the user's electric vehicle;

[0031] If the charging pile is compatible, check the occupancy status of the charging pile between the charging start time and end time specified by the user;

[0032] If the charging station is available within the time window required by the user, the time window matching degree is calculated;

[0033] Battery type compatibility is indicated as:

[0034]

[0035] Step S23. Matching degree calculation: Calculate the preliminary matching degree according to user needs and charging pile status, and sort according to the matching degree.

[0036] Furthermore, the method of calculating the preliminary matching degree and sorting according to the matching degree includes:

[0037] Step S231: Calculate the degree of satisfaction of the charging capacity requirement, the formula is as follows:

[0038]

[0039] In the formula, M i P is the charging capacity demand satisfaction; chi is the charging power of charging pile i; T e T is the charging end time; s D is the charging start time; req is the charging amount requested by the user, in kilowatt-hours;

[0040] Step S232: Calculate the time window matching degree, the formula is as follows:

[0041]

[0042] In the formula, T i is the time window matching degree, T e T is the charging end time; s T is the charging start time; l The length of the intersection of the charging pile idle time and the user time window;

[0043] Step S233. Calculate the preliminary matching degree, the formula is as follows:

[0044] M 初步,i =ω1×B i +ω2×M i +ω3×T i ;

[0045] In the formula, M 初步,i is the initial matching degree of charging pile i to user needs; T i is the time window matching degree; M i B is the charging capacity demand satisfaction; i Battery compatibility; ω1 is the weight of battery compatibility; ω2 is the weight of charging capacity demand satisfaction, which reflects the impact of charging speed on user satisfaction. If charging speed is a key factor, a higher weight is given; ω3 is the weight of time window matching, which reflects the user's emphasis on charging time flexibility. If the user has strict requirements on charging time, a higher weight is given;

[0046] Step S234. Candidate list generation, based on the preliminary matching degree M 初步,i , filter out the charging piles that meet the basic conditions, and add all the charging piles that meet the conditions to the candidate list;

[0047] Step S235. Sorting and selection, according to M 初步,i Sort the candidate charging piles and select the charging pile with the highest matching degree as the preliminary recommendation.

[0048] Furthermore, in step S3, the quantitative formula of the comprehensive score is:

[0049] C i =ω F ×F+ω E ×E+ω S ×S;

[0050] Among them, the usage frequency F reflects the frequency of use of the charging pile, which is quantified by the number of times the charging pile is used in a specific time period;

[0051] Charging efficiency E, which reflects the amount of electricity that a charging pile can provide to an electric vehicle per unit time, is quantified by the number of times the charging pile is used within a specific time period;

[0052] Safety factor S reflects the safety and reliability of the charging pile, which is quantified by counting the failure rate, maintenance records or other safety-related indicators of the charging pile;

[0053] C i For the comprehensive score of each charging pile, ω F is the weight corresponding to the usage frequency F, ω E is the weight corresponding to the charging efficiency E, ω S is the weight corresponding to the safety factor S;

[0054] The ω F ,ω E and ω S Need to meet the normalization condition: ω F +ω E +ω S =1.

[0055] Step S4. Dynamic Huffman coding, using the dynamic Huffman tree constructed in step S1 to encode the charging pile, converting the code into a comparable numerical value to indicate the priority of the charging pile;

[0056] Further, the method of converting the code into a comparable numerical value includes:

[0057] Step S41. Determine a weighting factor. According to the comprehensive score of each charging pile calculated in step S3, determine a weighting factor δ. The weighting factor δ is the comprehensive score C. i The function is expressed as: δ = 1 + log (C i ), log is the natural logarithm;

[0058] Step S42. Calculate the weighted length. For each charging pile, multiply the length of its Huffman code by the corresponding weighting factor to obtain the weighted length. After the weighted length is converted into a numerical value, it is used to compare the priorities of different charging piles. The formula is:

[0059] WL i =L i ×δ;

[0060] WL i is the weighted length of charging pile i, L i is the Huffman code length of charging pile i, and δ is based on the comprehensive score C of charging pile i. i Weighting factor of WL i The smaller it is, the more frequently the charging station is used and the better the overall score is, so it has a higher priority.

[0061] Step S5. Bilateral matching optimization, combining the evaluation coefficients in step S3 and the encoding results in step S4 to optimize the bilateral matching;

[0062] Furthermore, in step S5, the method for optimizing bilateral matching in combination with the evaluation coefficient in step S3 and the encoding result in step S4 includes:

[0063] Step S51. Re-evaluate the matching degree, using the preliminary matching degree in step S2 as a basis, taking into account the comprehensive score obtained in step S3 and the weighted length value obtained in step S4, and adjusting the preliminary matching degree. The adjustment formula is:

[0064]

[0065] M 初步,i =ω1×B i +ω2×M i +ω3×T i ;

[0066] C i =ω F ×F+ω E ×E+ω S ×S;

[0067] WL i =L i ×δ;

[0068] A di is the matching degree of charging pile i after adjustment, M 初步,i is the initial matching degree of charging pile i to user needs, T i is the time window matching degree; M i B is the charging capacity demand satisfaction; iBattery compatibility; ω1 is the weight of battery compatibility; ω2 is the weight of charging capacity requirement satisfaction; ω3 is the weight of time window matching; WL i is the weighted length of charging pile i; L i is the Huffman code length of charging pile i; δ is based on the comprehensive score C of charging pile i i ; ε is a small positive number used to avoid division by zero.

[0069] Step S52: Weight adjustment: adjust the weights of each evaluation coefficient according to the different importance of user needs, determine the user's preferences for factors such as charging speed, safety and accessibility, redistribute the weights of frequency of use, charging efficiency and safety factor, and update the calculation formula of the comprehensive score:

[0070]

[0071] A dw is the updated comprehensive score, F is the frequency of use, E is the charging efficiency, S is the safety factor, is the weight coefficient adjusted according to user needs, reflecting the importance of usage frequency, charging efficiency and safety factor in user needs. and The normalization conditions need to be met:

[0072] Step S53. Combine the adjusted comprehensive score and the weighted length value in step S4 to calculate the optimal matching degree of each charging pile. The calculation formula of the optimal matching degree is:

[0073]

[0074] Q i To optimize the matching, A dw is the updated comprehensive score, C i For the comprehensive score, ML i is the weighted length of charging pile i;

[0075] Step S6. Generate recommendation results, summarize the optimization matching degree of each charging pile calculated in step S5, use the obtained optimization matching degree to sort all candidate charging piles, and create a recommendation list. The recommendation list will be sorted from high to low according to the optimization matching degree of the charging piles, and the sorted recommendation list will be displayed to the user, so that the user can choose the most suitable charging pile according to his or her needs.

[0076] In summary, due to the adoption of the above technical solution, the beneficial technical effects of the invention are:

[0077] The shared charging pile recommendation method based on the dynamic Huffman model and bilateral matching comprehensively quantifies the performance of charging piles by introducing evaluation factors such as frequency of use, charging efficiency and safety factor. This multi-dimensional evaluation method not only improves the accuracy of the evaluation, but also makes the evaluation results more comprehensive, which can more comprehensively reflect the actual performance of charging piles and user satisfaction.

[0078] The shared charging pile recommendation method based on dynamic Huffman model and bilateral matching uses the dynamic Huffman model to collect and process the status information of the charging pile in real time, including charging power, occupancy status and historical usage data. The real-time update mechanism ensures that the recommendation system can reflect the latest status of the charging pile, improving the timeliness and reliability of the recommendation results.

[0079] The shared charging pile recommendation method based on the dynamic Huffman model and bilateral matching generates a ranked recommendation list based on the user's specific needs and the optimized matching degree of the charging pile. This personalized recommendation method enables users to quickly find the most suitable charging pile according to their preferences and needs, greatly improving user experience and satisfaction.

[0080] The shared charging pile recommendation method based on the dynamic Huffman model and bilateral matching allows the weight of each evaluation coefficient to be adjusted according to the different needs of users for factors such as charging speed, safety and accessibility. The flexible weight adjustment mechanism enables the recommendation system to adapt to the specific needs of different user groups, improving the adaptability and flexibility of the system.

[0081] Instruction Manual

[0082] Figure 1 This is a flow chart of the recommended method for shared charging piles based on dynamic Huffman model and bilateral matching.

[0083] Figure 2 Flowchart of the method for constructing a dynamic Huffman tree.

[0084] Figure 3 The flowchart is a method for performing preliminary matching using a bilateral matching algorithm.

[0085] Figure 4 Flowchart of a method for optimizing bilateral matching. DETAILED DESCRIPTION

[0086] In order to make the purpose, technical solution and advantages of the invention more clear, the invention is further described in detail below in combination with the embodiments. It should be understood that the specific embodiments described here are only used to explain the invention and are not used to limit the invention.

[0087] like Figure 1 As shown in FIG. 1 , the shared charging pile recommendation method based on the dynamic Huffman model and bilateral matching includes the following steps:

[0088] Step S1. Dynamic Huffman model construction, collect the status information of charging piles, including charging power, occupancy status, historical usage data, etc., calculate the usage frequency of each charging pile as the weight for constructing the Huffman tree, construct a dynamic Huffman tree, take the charging pile as the leaf node, and use the frequency as the priority;

[0089] The steps to construct a dynamic Huffman tree include:

[0090] Step S11. Regularly collect status information of each charging pile, which may include charging power, current occupancy status (whether idle or occupied), historical usage data, etc.;

[0091] Step S12: Calculate the usage frequency. For each charging pile i, calculate the usage frequency of the charging pile during a specific observation period. The calculation formula for the usage frequency is:

[0092]

[0093] Among them, f i is the frequency of use; N i is the number of charging times, which is the total number of times charging pile i is used during the observation period; T is the total observation time, which is the time span of data collection;

[0094] Step S13. Dynamic Huffman tree construction: Use the usage frequency of each charging pile i as a weight to construct a dynamic Huffman tree. Each leaf node in the dynamic Huffman tree represents a charging pile, and the internal nodes represent a combination of charging piles. During the construction process, the charging piles (or charging pile combinations) with the lowest usage frequency are gradually merged to form new nodes. This process continues until a single root node is formed.

[0095] Step S14: Priority coding. In the Huffman tree constructed in the above steps, the coding of each charging pile is determined according to its position in the tree. A shorter coding path indicates a higher frequency of use and priority.

[0096] Step S2. Analyze user needs and bilaterally match charging piles to obtain the user's charging request, which includes the battery type of the electric vehicle, the required charging amount, the charging time window, etc., and use the charging pile status information collected in step S1 and the user's charging request to perform preliminary matching using the bilateral matching algorithm;

[0097] Step S21. Demand acquisition, extracting the real-time status information of the charging pile from the dynamic Huffman tree model constructed in step S1;

[0098] Step S22. bilateral matching algorithm design, using the bilateral matching algorithm to match user needs with the status information of the charging pile, and considering the availability, compatibility and time requirements of the charging pile during the matching process;

[0099] Among them, availability refers to the idle state of the charging pile within the time window required by the user, compatibility refers to whether the charging pile supports the battery type of the user's electric vehicle, and time requirement refers to the time window specified in the user's charging request;

[0100] The matching algorithm logic is: for each charging pile, check the battery type compatibility. If compatible, check the occupancy status between the charging start time and the end time. If the time window match is available, calculate the battery type compatibility as follows:

[0101]

[0102] Step S23. Matching degree calculation: Calculate the preliminary matching degree according to user needs and charging pile status, and sort according to the matching degree.

[0103] Methods for calculating preliminary matching scores and sorting according to matching scores include:

[0104] Step S231: Calculate the degree of satisfaction of the charging capacity requirement, the formula is as follows:

[0105]

[0106] In the formula, M i P is the charging capacity demand satisfaction; chi is the charging power of charging pile i; T e T is the charging end time; s D is the charging start time; req is the charging amount requested by the user, in kilowatt-hours;

[0107] The purpose of the formula is to calculate whether the charging pile l can provide at least D within the time window given by the user. req The charging amount in kilowatt-hours, if the power and time window of the charging pile allow, then the charging amount satisfaction will be 1, indicating that the demand is fully met; if the power or time window of the charging pile is not enough to provide the required amount of electricity, then the charging amount satisfaction will be less than 1, indicating that the demand is not fully met;

[0108] Step S232: Calculate the time window matching degree, the formula is as follows:

[0109]

[0110] In the formula, T i is the time window matching degree, T e T is the charging end time;s T is the charging start time; s is the intersection length of the charging pile idle time and the user time window, and the intersection length T of the charging pile idle time and the user time window s It refers to the length of the time period that the charging pile is actually available between the start and end time specified by the user. If the idle time of the charging pile falls completely within the user time window and the length is equal to the length of the user time window, then the time matching degree is 1, indicating a complete match; if only part of the idle time of the charging pile falls within the user time window, or the length is less than the length of the user time window, then the time matching degree will be less than 1;

[0111] Step S233. Calculate the preliminary matching degree, the formula is as follows:

[0112] M 初步,i =ω1×B i +ω2×M i +ω3×T i ;

[0113] In the formula, M 初步,i is the initial matching degree of charging pile i to user needs; T i is the time window matching degree; M i B is the charging capacity demand satisfaction; i Battery compatibility; ω1 is the weight of battery compatibility, which is usually set to 1, because incompatible charging piles do not meet the basic needs of users; ω2 is the weight of charging capacity demand satisfaction, which reflects the impact of charging speed on user satisfaction. If charging speed is a key factor, a higher weight is given; ω3 is the weight of time window matching, which reflects the user's emphasis on charging time flexibility. If the user has strict requirements on charging time, a higher weight is given;

[0114] Step S234. Candidate list generation, based on the preliminary matching degree M 初步,i , filter out the charging piles that meet the basic conditions, and add all the charging piles that meet the conditions to the candidate list;

[0115] Step S235. Sorting and selection, according to M 初步,i Sort the candidate charging piles and select the charging pile with the highest matching degree as the preliminary recommendation.

[0116] Step S3. quantify the charging pile evaluation coefficient, define the charging pile evaluation coefficient, and quantify each coefficient, and use the quantified coefficient to assign a comprehensive score to each charging pile. The charging pile evaluation coefficient includes the frequency of use, charging efficiency and safety factor;

[0117] Frequency of use F, which reflects how frequently the charging pile is used, is quantified by the number of times the charging pile is used within a specific time period;

[0118] Charging efficiency E, which reflects the amount of electricity that a charging pile can provide to an electric vehicle per unit time, is quantified by the number of times the charging pile is used within a specific time period;

[0119] Safety factor S reflects the safety and reliability of the charging pile, which is quantified by counting the failure rate, maintenance records or other safety-related indicators of the charging pile;

[0120] The formula for calculating the comprehensive score is:

[0121] C i =ω F ×F+ω E ×E+ω S ×S;

[0122] Among them, C i For the comprehensive score of each charging pile, ω F is the weight corresponding to the usage frequency F, ω E is the weight corresponding to the charging efficiency E, ω S is the weight corresponding to the safety factor S;

[0123] The ω F ,ω E and ω S Need to meet the normalization condition: ω F +ω E +ω S =1;

[0124] Step S4. Dynamic Huffman coding, using the dynamic Huffman tree constructed in step S1 to encode the charging pile, converting the code into a comparable numerical value to indicate the priority of the charging pile;

[0125] The dynamic Huffman tree constructed in step S1 enables each charging pile to have a priority code based on the frequency of use. These codes are usually variable-length codes, where shorter codes represent higher frequency of use and priority. In order to convert the codes into comparable numerical values, these variable-length codes need to be converted into fixed-length numerical values ​​for comparison and sorting. The method of converting the codes into comparable numerical values ​​includes:

[0126] Step S41. Determine a weighting factor. According to the comprehensive score of each charging pile calculated in step S3, determine a weighting factor δ. The weighting factor δ is the comprehensive score C. i The function is expressed as: δ = 1 + log (C i ), log is the natural logarithm;

[0127] Step S42. Calculate the weighted length. For each charging pile, multiply the length of its Huffman code by the corresponding weighting factor to obtain the weighted length. After the weighted length is converted into a numerical value, it is used to compare the priorities of different charging piles. The formula is:

[0128] WL i =L i ×δ;

[0129] WL i is the weighted length of charging pile i, L i is the Huffman code length of charging pile i, and δ is based on the comprehensive score C of charging pile i. i Weighting factor of WL i The smaller it is, the more frequently the charging station is used and the better the overall score is, so it has a higher priority.

[0130] Step S5. bilateral matching optimization, combining the evaluation coefficient in step S3 and the encoding result in step S4 to optimize the bilateral matching, and using the obtained optimized matching degree;

[0131] Methods for optimizing bilateral matching include:

[0132] Step S51. Re-evaluate the matching degree, using the preliminary matching degree in step S2 as a basis, taking into account the comprehensive score obtained in step S3 and the weighted length value obtained in step S4, and adjusting the preliminary matching degree. The adjustment formula is:

[0133]

[0134] M 初步,i =ω i ×B i +ω2×M i +ω3×T i ;

[0135] C i =ω F ×F+ω E ×E+ω S ×S;

[0136] WL i =L i ×δ;

[0137] A di is the matching degree of charging pile i after adjustment, M 初步,i is the initial matching degree of charging pile i to user needs, T i is the time window matching degree; M i B is the charging capacity demand satisfaction; iBattery compatibility; ω1 is the weight of battery compatibility; ω2 is the weight of charging capacity requirement satisfaction; ω3 is the weight of time window matching; WL i is the weighted length of charging pile i; L i is the Huffman code length of charging pile i; δ is based on the comprehensive score C of charging pile i i ; ε is a small positive number used to avoid division by zero.

[0138] Step S52: Weight adjustment: adjust the weights of each evaluation coefficient according to the different importance of user needs, determine the user's preferences for factors such as charging speed, safety and accessibility, redistribute the weights of frequency of use, charging efficiency and safety factor, and update the calculation formula of the comprehensive score:

[0139]

[0140] A dw is the updated comprehensive score, F is the frequency of use, E is the charging efficiency, S is the safety factor, is the weight coefficient adjusted according to user needs, reflecting the importance of usage frequency, charging efficiency and safety factor in user needs. and The normalization conditions need to be met:

[0141] Step S53. Combine the adjusted comprehensive score and the weighted length value in step S4 to calculate the optimal matching degree of each charging pile. The calculation formula of the optimal matching degree is:

[0142]

[0143] Q i To optimize the matching, A dw is the updated comprehensive score, C i For the comprehensive score, WL i is the weighted length of charging pile i;

[0144] Step S6. Generate recommendation results, summarize the optimization matching degree of each charging pile calculated in step S5, sort all candidate charging piles according to the optimization matching degree, create a recommendation list, and sort the recommendation list from high to low according to the optimization matching degree of the charging pile. The sorted recommendation list is displayed to the user, and the user can choose the most suitable charging pile according to his or her needs.

[0145] The above description is a preferred embodiment of the invention and is not intended to limit the invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the invention should be included in the protection scope of the invention.

Claims

1. A shared charging pile recommendation method based on dynamic Huffman model and bilateral matching, characterized in that: The following steps are involved: Step S1. Dynamic Huffman model construction, collecting status information of charging piles, calculating the usage frequency of each charging pile, and using the calculated usage frequency to construct a dynamic Huffman model. In the dynamic Huffman model, the charging pile is used as a leaf node and the usage frequency is used as a weight; Step S2. Analyze user demand and match charging piles bilaterally, obtain the user's charging request, and use the charging pile status information collected in step S1 and the obtained user charging request to perform preliminary matching using a bilateral matching algorithm; Step S3. quantifying the charging pile evaluation coefficients, defining the charging pile evaluation coefficients, and quantifying each coefficient, using the quantified coefficients to assign a comprehensive score to each charging pile; Step S4. Dynamic Huffman coding, using the dynamic Huffman tree constructed in step S1 to encode the charging pile, converting the code into a comparable numerical value to indicate the priority of the charging pile; Step S5. Bilateral matching optimization, combining the evaluation coefficients in step S3 and the encoding results in step S4 to optimize the bilateral matching; Step S6. Generate recommendation results, summarize the optimization matching degree of each charging pile calculated in step S5, use the obtained optimization matching degree to sort all candidate charging piles, create a recommendation list, and sort the recommendation list from high to low according to the optimization matching degree of the charging piles. The sorted recommendation list is displayed to the user, and the user can choose the most suitable charging pile according to his or her needs.

2. The method for recommending shared charging piles based on dynamic Huffman model and bilateral matching according to claim 1, characterized in that: The method for constructing a dynamic Huffman tree in step S1 includes: Step S11. Regularly collect status information of each charging pile, the status information includes charging power, current occupancy status, and historical usage data; Step S12: Use frequency calculation to calculate the charging frequency of each charging pile. l , calculate charging pile l The frequency of use during a specific observation period. The frequency of use is calculated as: Among them, f i is the frequency of use; N i is the number of charging times, which is the total number of times charging pile i is used during the observation period; T is the total observation time, which is the time span of data collection; Step S13. Dynamic Huffman tree construction: using the usage frequency of each charging pile i as a weight, a dynamic Huffman tree is constructed, in which each leaf node represents a charging pile, and the internal nodes represent a combination of charging piles; Step S14: Priority coding. In the Huffman tree constructed in the above steps, the coding of each charging pile i is determined according to the position of each charging pile i in the tree. A shorter coding path indicates a higher frequency of use and priority.

3. The method for recommending shared charging piles based on dynamic Huffman model and bilateral matching according to claim 1, characterized in that: In step S2, the method of performing preliminary matching using the bilateral matching algorithm includes: Step S21. Demand acquisition, extracting the real-time status information of the charging pile from the dynamic Huffman tree model constructed in step S1; Step S22. bilateral matching algorithm design, using the bilateral matching algorithm to match user needs with the status information of the charging pile, and considering the availability, compatibility and time requirements of the charging pile during the matching process; Among them, availability refers to the idle state of the charging pile within the time window required by the user, compatibility refers to whether the charging pile supports the battery type of the user's electric vehicle, and time requirement refers to the time window specified in the user's charging request; The matching algorithm logic is; For each charging station, first check whether it is compatible with the battery type of the user's electric vehicle; If the charging pile is compatible, check the occupancy status of the charging pile between the charging start time and end time specified by the user; If the charging station is available within the time window required by the user, the time window matching degree is calculated; Battery type compatibility is indicated as: Step S23: Calculate the matching degree: calculate the preliminary matching degree according to the user's needs and the status of the charging pile, and sort by the matching degree.

4. The method for recommending shared charging piles based on dynamic Huffman model and bilateral matching according to claim 2 is characterized in that: In step S23, the method of calculating the preliminary matching degree and sorting according to the matching degree includes: Step S231: Calculate the degree of satisfaction of the charging capacity requirement, the formula is as follows: In the formula, M i P is the charging capacity demand satisfaction; chi is the charging power of charging pile i; T e T is the charging end time; s D is the charging start time; req is the charging amount requested by the user, in kilowatt-hours; Step S232: Calculate the time window matching degree, the formula is as follows: In the formula, T i is the time window matching degree, T e T is the charging end time; s T is the charging start time; l The length of the intersection of the charging pile idle time and the user time window; Step S233. Calculate the preliminary matching degree, the formula is as follows: M 初步j =ω1×B i +ω2×M i +ω3×T i ; In the formula, M 初步j is the initial matching degree of charging pile i to user needs; T i is the time window matching degree; M i B is the charging capacity demand satisfaction; i Battery compatibility; ω1 is the weight of battery compatibility; ω2 is the weight of charging capacity demand satisfaction, which reflects the impact of charging speed on user satisfaction. If charging speed is a key factor, a higher weight is given; ω3 is the weight of time window matching, which reflects the user's emphasis on charging time flexibility. If the user has strict requirements on charging time, a higher weight is given; Step S234. Candidate list generation, based on the preliminary matching degree M 初步j , filter out the charging piles that meet the basic conditions, and add all the charging piles that meet the conditions to the candidate list; Step S235. Sorting and selection, according to M 初步j Sort the candidate charging piles and select the charging pile with the highest matching degree as the preliminary recommendation.

5. The method for recommending shared charging piles based on dynamic Huffman model and bilateral matching according to claim 1, characterized in that: In step S3, the calculation formula of the comprehensive score is: C i =ω F ×F+ω E ×E+ω S ×S; Among them, C i For the comprehensive score of each charging pile, ω F is the weight corresponding to the usage frequency F, ω E is the weight corresponding to the charging efficiency E, ω S is the weight corresponding to the safety factor S; The ω F ,ω E and ω S Need to meet the normalization condition: ω F +ω E +ω S =1; F is the frequency of use, which reflects how frequently the charging pile is used and is quantified by the number of times the charging pile is used within a specific time period; E is the charging efficiency, which reflects the amount of electricity that the charging pile can provide to the electric vehicle per unit time and is quantified by the number of times the charging pile is used within a specific time period; S is the safety factor, which reflects the safety and reliability of the charging pile and is quantified by counting the failure rate, maintenance records or other safety-related indicators of the charging pile.

6. The method for recommending shared charging piles based on dynamic Huffman model and bilateral matching according to claim 1, characterized in that: In step S4, the method of converting the code into a comparable value includes: Step S41. Determine a weighting factor. According to the comprehensive score of each charging pile calculated in step S3, determine a weighting factor δ. The weighting factor δ is the comprehensive score C. i The function is expressed as: δ = 1 + log (C i ), log is the natural logarithm; Step S42. Calculate the weighted length. For each charging pile, multiply the length of its Huffman code by the corresponding weighting factor to obtain the weighted length. After the weighted length is converted into a numerical value, it is used to compare the priorities of different charging piles. The weighted length formula is: WL i =L i ×δ; WL i is the weighted length of charging pile i, L i is the Huffman code length of charging pile i, and δ is based on the comprehensive score C of charging pile i. i Weighting factor of WL i The smaller it is, the more frequently the charging station is used and the better the overall score is, so it has a higher priority.

7. The method for recommending shared charging piles based on dynamic Huffman model and bilateral matching according to claim 1, characterized in that: In step S5, the method for optimizing bilateral matching includes: Step S51. Re-evaluate the matching degree, using the preliminary matching degree in step S2 as a basis, taking into account the comprehensive score obtained in step S3 and the weighted length value obtained in step S4, and adjusting the preliminary matching degree. The adjustment formula is: M 初步j =ω1×B i +ω2×M i +ω3×T i ; C i =ω F ×F+ω E ×E+ω S ×S; WL i =L i ×δ; A di is the matching degree of charging pile i after adjustment, M 初步j is the initial matching degree of charging pile i to user needs, T i is the time window matching degree; M i B is the charging capacity demand satisfaction; i Battery compatibility; ω1 is the weight of battery compatibility; ω2 is the weight of charging capacity requirement satisfaction; ω3 is the weight of time window matching; WL i is the weighted length of charging pile i; L i is the Huffman code length of charging pile i; δ is based on the comprehensive score C of charging pile i i The weighting factor of ; ε is a small positive number used to avoid division by zero; Step S52: Weight adjustment: adjust the weights of each evaluation coefficient according to the different importance of user needs, determine the user's preferences for factors such as charging speed, safety and accessibility, redistribute the weights of frequency of use, charging efficiency and safety factor, and update the calculation formula of the comprehensive score: A dw is the updated comprehensive score, F is the frequency of use, E is the charging efficiency, S is the safety factor, is the weight coefficient adjusted according to user needs, reflecting the importance of usage frequency, charging efficiency and safety factor in user needs. and The normalization conditions need to be met: Step S53. Combine the adjusted comprehensive score and the weighted length value in step S4 to calculate the optimal matching degree of each charging pile. The calculation formula of the optimal matching degree is: Q i To optimize the matching, A dw is the updated comprehensive score, C i For the comprehensive score, WL i is the weighted length of charging pile i.