Energy internet park EV charging scheduling system based on dynamic priority and battery life optimization

Abstract

The invention discloses an energy internet park EV charging scheduling system based on dynamic priority and battery life optimization, and the system comprises a dynamic priority distribution module which monitors the SOC of a vehicle battery in real time, and inserts a vehicle into the front end of a charging queue when the SOC is less than or equal to a preset threshold value; the common queues are ranked according to a first-arrival first-service principle, and the common vehicles are continued after the emergency vehicles are jumped; the lithium battery degradation calculation module is used for quantifying the battery capacity degradation rate based on a multi-stress factor model of the charging depth, the average SOC, the temperature and the charging duration; the multi-stress factor model calculates the degradation amount through chemical characteristic parameters calibrated through experiments; the multi-target optimization module is used for integrating optimization algorithms for dynamically switching global exploration and local mining and generating a scheduling strategy by taking the minimum charging cost, the user waiting time and the battery degradation cost as targets; according to the optimization algorithm, the solving efficiency is improved through a population diversity tracking strategy and an elite-random double-search mechanism.

Description

Technical Field

[0001] The present invention relates to the field of energy management technology, and in particular to an energy internet park EV charging scheduling system based on dynamic priority and battery life optimization. The system is suitable for energy management scenarios in energy internet parks that include electric vehicles with different charging requirements and have multiple charging stations. Background Art

[0002] As an innovative model for intelligent energy management, smart charging stations within the Energy Internet Park provide an optimal environment for optimizing clean energy cycles. The daily operation of the charging dispatch system for electric vehicle charging stations within the Energy Internet Park is optimized. By designing dynamic priority allocation mechanisms (such as queue-jumping for emergency vehicles and load balancing for non-emergency vehicles), this ensures rapid response for vehicles with high emergency demands while reducing the park's dependence on the power grid during peak hours.

[0003] Compared with other types of batteries, lithium-ion batteries have high power density and longer cycle life, making them a prime candidate for electric vehicle (EV) applications. EVs equipped with lithium-ion batteries are gaining increasing attention for applications in mobile energy storage, intelligent transportation, and as demand response resources for data centers and energy internet parks. A key factor in EV charging plans is operating costs, much of which stems from the degradation of battery cells. Currently, most studies do not consider battery degradation models or only use simple inference models for estimation, which greatly limits the accuracy of smart charging system operation assessments. Therefore, it is necessary to gain a deeper understanding of the chemical properties of lithium batteries and establish accurate EV lithium battery degradation models to generate accurate charging scheduling strategies.

[0004] The multi-period, multi-station, and multi-EV charging scheduling problem in the energy internet park environment has obvious high-dimensional and complex characteristics. Traditional intelligent algorithm-based solution methods are prone to shortcomings such as slow convergence and falling into local optimality.

[0005] The Chinese patent application number is 201710053957.6 and is titled: A Charging Scheduling Method for Electric Vehicle Charging Stations. First, based on the active network loss and matching degree, a first charging scheduling model is obtained. Then, all possible charging behaviors at the current moment are traversed, and the charging behavior that minimizes the value of the first charging scheduling model is used as the charging strategy at the current moment. This method can achieve the maximum matching between electric vehicle charging and new energy output, while maintaining the network loss in the distribution network area where the two are located within a relatively low range. This method does not take into account the different charging needs of electric vehicles when charging, and does not formulate a reasonable charging strategy for emergency charging situations, which affects the rationality of the charging method; this method does not consider the impact of lithium battery degradation on the design of the charging strategy, resulting in an unreasonable model; this method does not design a fast and efficient intelligent optimization algorithm to solve the optimal strategy, which affects the accuracy and effectiveness of the solution.

[0006] The Chinese patent application number is 202411013447.2 and is titled: A DC Charging Scheduling Method and System Based on Electric Vehicle Charging Priority. Clustering vector processing is used to divide all electric vehicles into charging priorities based on the calculated vector group differentiation information, thereby obtaining a first electric vehicle group and a second electric vehicle group. Since the charging priority of any electric vehicle in the first electric vehicle group is higher than the charging priority of any electric vehicle in the second electric vehicle group, it is equivalent to dividing all electric vehicles with higher charging priorities into the first electric vehicle group, and finally generating a DC charging scheduling strategy based on the charging priority. However, this method does not take into account the real-time scheduling needs of EVs at different charging stations at different times in the energy exchange network campus scenario, and cannot reflect the complex EV scheduling arrangements in the campus environment; this method only divides priorities, but does not define and design strategies for emergency charging scenarios; this method does not consider the impact of lithium battery degradation on charging strategy design, resulting in an unreasonable model; this method does not design a fast and efficient intelligent optimization algorithm to solve the optimal strategy, which affects the accuracy and effectiveness of the solution. Summary of the Invention

[0007] In response to the defects and shortcomings of the existing technology, the present invention provides an energy internet park EV charging scheduling system and method based on dynamic priority and battery life optimization. Its core innovations include:

[0008] Dynamic priority allocation mechanism: By real-time monitoring of vehicle battery SOC (≤ preset threshold triggering emergency queue interruption) and charging station load, emergency queues and ordinary queues are divided, achieving a response time of seconds for vehicles with high emergency needs (response time ≤ 30 minutes). Ordinary vehicles are dynamically postponed on a first-come, first-served basis, balancing system fairness and response efficiency.

[0009] High-precision lithium battery degradation model: Based on the depth of charge (DOD), average SOC, temperature and charging time, a multi-stress factor interaction model is constructed, and chemical characteristic parameters (such as α sei and β sei ) quantifies battery capacity degradation rate with an error of <5%, significantly better than traditional linear models (error >20%);

[0010] Improved optimization algorithm (MSCSO-PDG): This algorithm introduces a population diversity tracking strategy to dynamically switch between global exploration and local exploitation modes. It also combines a dual search mechanism of elite and random solutions (elite individuals lead local optimization, while ordinary individuals perform global search). This algorithm addresses the pain points of slow convergence and the tendency to fall into local optimality in high-dimensional and complex scheduling problems, improving computational efficiency by 40%.

[0011] Multi-objective collaborative optimization framework: Integrates time-of-use electricity prices, renewable energy consumption, user waiting time, and battery life costs, dynamically adjusts target priorities through weight coefficients, reduces overall costs by 15%-20%, and reduces peak grid load by 30%;

[0012] Real-time interaction and elastic expansion capabilities: Based on the MQTT protocol, 5-minute data synchronization is achieved, limiting the length of the general queue (L max ≤10), supports concurrent scheduling of 1000+ EVs, and ensures stability in high-load scenarios.

[0013] The present invention provides an intelligent and highly robust EV charging management solution for energy internet parks, which combines economy, environmental protection and user experience.

[0014] The present invention specifically adopts the following technical solutions:

[0015] An Energy Internet Park EV charging scheduling system based on dynamic priority and battery life optimization includes:

[0016] Dynamic priority allocation module:

[0017] Real-time monitoring of vehicle battery SOC. When SOC ≤ preset threshold, the vehicle is inserted into the front of the charging queue.

[0018] The general queue is arranged on a first-come, first-served basis. When an emergency vehicle cuts in line, the general vehicles will be delayed;

[0019] Lithium battery degradation calculation module:

[0020] Quantify the battery capacity degradation rate based on a multi-stress factor model based on charge depth, average SOC, temperature, and charging time;

[0021] The multi-stress factor model calculates the degradation amount through experimentally calibrated chemical property parameters;

[0022] Multi-objective optimization module:

[0023] An optimization algorithm that dynamically switches between global exploration and local mining generates a scheduling strategy with the goal of minimizing charging costs, user waiting time, and battery degradation costs.

[0024] The optimization algorithm improves solution efficiency through a population diversity tracking strategy and an elite-random dual search mechanism.

[0025] Furthermore, the multi-objective optimization module further includes:

[0026] Population division unit: divide the optimized population into elite individuals, medium individuals and ordinary individuals;

[0027] Diversity monitoring unit: calculates population diversity indicators in real time and dynamically switches between global exploration and local mining modes;

[0028] Search strategy execution unit: elite individuals perform local search based on the current optimal solution, and ordinary individuals perform global search following random solutions.

[0029] Furthermore, the specific implementation of the multi-stress factor model includes:

[0030] Calendar aging calculation unit: Calculates calendar aging capacity loss based on charging duration, average SOC, and battery temperature through experimental calibration parameters;

[0031] Cycle aging calculation unit: Calculates cycle aging capacity loss based on depth of charge, average SOC and battery temperature through experimental calibration parameters;

[0032] Total degradation rate integration unit: Add the calendar aging and cycle aging capacity losses and multiply them by the unit capacity cost to output the total degradation rate.

[0033] Furthermore, it also includes a real-time data interaction module:

[0034] Low-latency communication between vehicles, charging piles and servers is achieved based on the MQTT protocol, with a data update cycle of ≤5 minutes;

[0035] Dynamically synchronize charging station load, time-of-use electricity prices, and renewable energy output data.

[0036] Furthermore, the dynamic priority allocation module further includes:

[0037] Queue length control unit: used to limit the maximum length of ordinary queues to avoid high concurrency congestion;

[0038] Emergency response constraint unit: used to set the maximum response time threshold for emergency vehicles to start charging.

[0039] Furthermore, the multi-objective optimization module integrates time-of-use electricity prices and a renewable energy scheduling unit to prioritize charging demands during off-peak electricity prices and periods of high renewable energy output.

[0040] And, an energy internet park EV charging scheduling method, comprising the following steps:

[0041] Real-time monitoring of vehicle battery SOC. When SOC ≤ preset threshold, the vehicle is inserted into the front of the charging queue, and ordinary vehicles are deferred on a first-come, first-served basis.

[0042] Quantify the battery capacity degradation rate based on a multi-stress factor model based on charge depth, average SOC, temperature, and charging time;

[0043] An optimization algorithm that dynamically switches between global exploration and local mining is used to generate a scheduling strategy that minimizes charging costs, user waiting time, and battery degradation costs.

[0044] Furthermore, the optimization algorithm dynamically switches between global exploration and local mining in the following ways:

[0045] Divide the population into elite individuals, middle individuals and ordinary individuals;

[0046] Real-time monitoring of population diversity, dynamically switching between global exploration and local mining modes;

[0047] Elite individuals perform local search based on the current optimal solution, and ordinary individuals perform global search following random solutions.

[0048] Furthermore, the optimization algorithm specifically includes:

[0049] Population dynamics classification and role allocation:

[0050] The optimized population is divided into three categories: elite individuals, medium individuals and ordinary individuals, corresponding to the roles of roosters, hens and chicks respectively;

[0051] Elite individuals account for 20% and dominate the mining of local optimal solutions;

[0052] Ordinary individuals account for 20% and perform global exploration;

[0053] Medium individuals account for 60%, balancing local and global searches;

[0054] Population diversity tracking and mode switching:

[0055] Real-time calculation of population diversity indicators to reflect the distribution dispersion of individuals in the search space;

[0056] When the diversity index is lower than the historical threshold and the optimal solution has not been improved, the global exploration mode is triggered to expand the search scope;

[0057] When the diversity index is higher than the threshold or the optimal solution is continuously optimized, the local mining mode is triggered to focus on the current optimal solution;

[0058] Differentiated search strategies:

[0059] The elite individual performs a local fine search based on the current optimal solution, and the disturbance range is adaptively adjusted;

[0060] Ordinary individuals follow random solutions to perform global exploration and introduce random perturbations to avoid local optimality;

[0061] Intermediate individuals refer to both elite solutions and random solutions, dynamically balancing exploration and exploitation;

[0062] Dynamic character periodic adjustments:

[0063] Every fixed generation, the population roles are redistributed according to the fitness value to adapt to the changes in the search state.

[0064] Furthermore, the calculation of the battery capacity degradation rate includes the following steps:

[0065] Total degradation rate calculation: multiply the unit capacity cost by the sum of calendar aging capacity loss and cycle aging capacity loss;

[0066] Calendar aging capacity loss: calculated based on the battery chemical characteristic parameters experimentally calibrated based on the charging duration, average state of charge and battery temperature;

[0067] Cycle aging capacity loss: calculated based on the battery chemical characteristic parameters experimentally calibrated based on the depth of charge, average state of charge and battery temperature;

[0068] In the multi-stress factor model, chemical property parameters are calibrated through battery aging experiments to reflect the nonlinear attenuation characteristics of battery materials.

[0069] And, an electronic device includes a memory, a processor, and a computer program stored in the memory and runnable on the processor, characterized in that the processor implements the steps of the above method when executing the program.

[0070] A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0071] Compared with the prior art, the present invention and its preferred embodiments have at least the following beneficial effects:

[0072] Significantly improves the efficiency of emergency charging demand response: Through a dynamic priority allocation mechanism, emergency vehicles are triggered to jump the queue based on real-time SOC thresholds, and the order of ordinary vehicles in the queue is dynamically adjusted. This achieves rapid response to high-demand vehicles and overall system load balancing, balancing response speed and fairness.

[0073] Improved accuracy in battery life assessment and cost control: A multi-stress factor-based lithium battery degradation model comprehensively considers the interactive effects of multiple parameters such as depth of charge, temperature, SOC, and charging duration, significantly improving the accuracy of battery capacity degradation rate calculations and providing a reliable basis for charging strategy formulation.

[0074] Enhanced efficiency in solving high-dimensional, complex scheduling problems: The improved optimization algorithm dynamically switches between global exploration and local mining modes, combining a dual search mechanism of elite solutions and random solutions. This effectively improves the algorithm's convergence speed and global optimization capabilities, making it suitable for complex scheduling scenarios with multiple vehicles, multiple power stations, and multiple time periods.

[0075] Achieve multi-objective collaborative optimization and efficient resource utilization: By integrating time-of-use electricity prices, renewable energy consumption, user waiting time, and battery life costs, a flexible and adjustable multi-objective optimization framework is built to balance economic efficiency, environmental protection, and user experience, reducing dependence on peak loads on the grid.

[0076] Ensure system real-time and scalability: Based on low-latency communication protocols and a dynamic queue length control mechanism, it supports high-concurrency scheduling requirements in large-scale campus environments, avoids congestion, and ensures dynamic policy updates.

[0077] In addition, the optimal solution further enhances the robustness and practicality of the technical solution by experimentally calibrating the chemical characteristic parameters of the battery degradation model, refining the population division rules of the optimization algorithm, and integrating time-of-use electricity prices and renewable energy output data. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0079] Figure 1 This is a flowchart of EV charging scheduling in an energy internet park based on dynamic priority and battery life optimization according to an embodiment of the present invention;

[0080] Figure 2 This is a flow chart of the MSCSO-PDG algorithm according to an embodiment of the present invention;

[0081] Figure 3 This is a flow chart of the dynamic priority allocation and emergency charging queue-jumping mechanism for EVs according to an embodiment of the present invention. DETAILED DESCRIPTION

[0082] Hereinafter, specific embodiments of the present application will be described in detail with reference to the accompanying drawings. Based on these detailed descriptions, those skilled in the art will be able to clearly understand the present application and implement the present application. Without violating the principles of the present application, the features of different embodiments may be combined to obtain new implementations, or certain features of certain embodiments may be substituted to obtain other preferred implementations.

[0083] To make the features and advantages of the present invention more clearly understood, the following embodiments are specifically described in detail with reference to the accompanying drawings.

[0084] An embodiment of the present invention provides an energy internet park EV charging scheduling system based on dynamic priority and battery life optimization: the charging scheduling system obtains real-time EV vehicle status information (including SOC, location) and charging station operating parameters (including electricity price, load) in the park; a charging queue is generated based on a dynamic priority allocation mechanism, low SOC vehicles are prioritized and emergency charging queue-jumping requests are processed; the battery management system collects multi-stress factor data of lithium battery operation, and calculates the real-time capacity decay rate by coupling calendar aging and cycle aging models; the optimization algorithm module integrates charging cost, time cost and battery loss to construct a multi-objective function, and adopts the MSCSO-PDG algorithm with a population diversity tracking strategy for multi-dimensional optimization solution, in which the elite individuals adopt a dual-mode search strategy to guide population evolution; the scheduling terminal receives the optimal solution set and generates the optimal charging plan for the park. The present invention solves the multi-objective coordination problem in the prior art that EV charging scheduling is difficult to take into account user emergency needs, battery health and economic optimization.

[0085] The EV intelligent charging scheduling system in the Energy Internet Park is crucial for the energy management optimization and low-carbon energy-saving operation of the park. The system provided by the embodiment of the present invention is based on real-time vehicle status (such as SOC, location) and charging station information (electricity price, load), and adopts a dynamic priority allocation mechanism to prioritize the scheduling of low-power vehicles and support emergency charging queues to ensure immediate response to vehicles with high emergency needs. Secondly, the present invention deeply studies the influence of multiple stress factors on the battery aging process during the operation of lithium batteries, and then establishes an accurate capacity degradation model that combines lithium battery calendar aging and cycle aging to provide a basis for the subsequent scheduling strategy formulation; finally, by integrating charging cost, waiting time and battery degradation cost to construct a multi-objective function, a novel multi-elite search chicken swarm optimization algorithm MSCSO-PDG based on population diversity tracking is designed to solve the intelligent charging cost model. MSCSO-PDG introduces a population diversity tracking strategy to make real-time judgments on the search status of the population; two search schemes are designed for the elite individuals in the population, and the remaining individuals follow the elite individuals to conduct multi-dimensional search, achieving a balance between algorithm exploration and mining, thereby effectively alleviating the problems of slow convergence and easy falling into local optimality in the traditional CSO algorithm. Based on the above steps, the present invention realizes the formulation of the optimal charging strategy for EVs in the Energy Internet Park, while meeting the diverse charging needs of EV users.

[0086] The present invention proposes an energy internet park EV charging scheduling system based on dynamic priority and battery life optimization, which is implemented in the following steps:

[0087] Step 1: System Model Construction. EV users access the park's intelligent charging scheduling system via wireless communication technology. Based on information such as charging station and EV status data, EV battery parameters, and time-of-use electricity prices, the system outputs each vehicle's charging time slot, charging node allocation, and charging power level, with the goal of minimizing vehicle charging costs and EV user waiting time.

[0088] Step 2: Define the decision variables for each electric car: Used to identify whether electric vehicle i is connected to charging station k during time period t; Represents the charging power of electric vehicle i in time period t. Next, determine the emergency charging identification variable of the Ev charging problem: Represents the emergency charging indicator variable. When the SOC of electric vehicle i in time period t meets hour, is 1, otherwise is 0.

[0089] Step 3: Construct the objective function of the park smart charging system:

[0090]

[0091] Among them, N EV represents the number of EVs, α and β are weight coefficients, and α + β = 1; t and T represent the current time and the total length of the time period, respectively. is the waiting time; Δt represents the duration of charging power conversion into energy. χ is the emergency charging delay penalty coefficient; represents the delay of vehicle i from the time the demand is issued to the actual start of charging time t; is the electricity price of charging station k at time t; γ i represents the capacity degradation rate of the i-th EV battery.

[0092] Step 4: Campus EV charging strategy based on dynamic priority allocation and emergency charging queue-jumping mechanism. First, define the EV queue type, which is divided into two types. Emergency queue Q emergency : Storage satisfaction EV, SOC emergency It can be set according to specific circumstances and needs. normal : Store other vehicles and sort them on a first-come, first-served basis. State variables represents the queue position of vehicle i at charging station k (0 means charging); This is an emergency sign, with the same definition as above.

[0093] (1): Calculation of EV charging demand. req,i It represents the total charging power requirement of the i-th electric vehicle (unit: kWh), that is, the total battery capacity is E bat,i The electric vehicle needs to be upgraded from the current SOC (SOC initial,i ) to the target SOC (SOC targct,i )The total amount of electrical energy required.

[0094] (2): Emergency vehicle queue-jumping rule: When vehicle i triggers emergency charging When , it is immediately inserted into the front of the queue of charging station k, which can be expressed as:

[0095]

[0096] Then the ordinary vehicles in the original queue are moved back one position.

[0097] (3): Update rules for ordinary vehicle queues: Whenever time period t is updated, vehicle i in the ordinary queue is adjusted according to the first-come-first-served principle:

[0098]

[0099] in is the position of vehicle i in the normal queue sequence at charging station k at time t+1.

[0100] (4): Queuing time calculation. The waiting time of vehicle i in time period t and delay penalty The calculation of is as follows:

[0101]

[0102] Among them, t start,i and t req,i are the initial period when vehicle i issues a charging request and the period when charging actually starts, respectively; is an indicator function, which takes the value 1 if the i-th vehicle belongs to the normal queue and 0 otherwise.

[0103] Step 5: Accurately model the battery wear cost. i Used to measure the battery loss cost caused by charging the i-th EV. The mainstream battery of EV is lithium-ion battery, which will age during use, thereby generating aging cost. Lithium-ion batteries are divided into calendar aging and cycle aging. Calendar aging refers to the phenomenon that the capacity of the battery slowly decreases as the shelf time increases during the shelf process. This aging will occur even if the battery is not used and is in a static state. Cycle aging is the irreversible capacity loss that occurs in lithium-ion batteries during the charging cycle. Cycle aging only occurs when the battery is charging. Most current studies have not given an accurate EV lithium battery capacity degradation model. In the present invention, based on the semi-empirical degradation characteristic data of EV lithium batteries, the capacity degradation rate γ is designed. i To accurately describe the battery degradation process, it is calculated as follows:

[0104]

[0105] L deg =L t +L c (7)

[0106]

[0107] f t (Δt,σ,T)=S t (Δt)S σ (σ)S T (T)(10)

[0108] f c (δ,σ,T)=S δ (δ)S σ (σ)S T (T)(11)

[0109]

[0110] S t (Δt)=k t Δt(15)

[0111] Among them, formula (6) is to calculate γ i , where C per is the cost per kWh of battery capacity, L deg The battery capacity degradation caused by EV charging is assumed in this paper to be the same for all EVs, so i is no longer used to distinguish individual EVs. deg Including calendar aging capacity L t and cycle aging capacity L c Two aspects. t and L c Calculated by formula (8)-(9), α sei and β sei These are all chemical characteristic parameters of the battery, f t and f c Represent the calendar aging rate and cycle aging rate of EV batteries respectively. From formulas (10)-(11), we can see that f t and f c They can be expressed as depth of charge δ, average soc(σ), battery temperature T c and the charging duration Δt; and the above four quantities have corresponding stress factor models S δ (δ), S σ (σ), S T (T) and S t (Δt), used to describe the influence of four quantities on battery aging, which are described by formulas (12)-(15) respectively, where σ ref and T ref is the experimental calibration reference value, k δ1 , k δ2 , k δ3 , k σ , k T and k t They are

[0112] The battery experimental parameters are obtained by fitting the actual battery aging test.

[0113] Step 6: The constraints of the system model are:

[0114]

[0115] Among the above constraints, formulas (16)-(17) are the charging power interval constraints and total charging request constraints of each electric vehicle, which prevent the charging power from exceeding the upper and lower bounds of the power; formulas (18)-(19) are the update and boundary constraint formulas of SOC, which define the update rules of SOC and the upper and lower bounds of SOC, η is the charging efficiency, and E bat,i represents the rated capacity of the battery of vehicle i; Formula (20) is the capacity constraint of the charging station, N k is the maximum charging capacity of the kth charging station; Formula (21) is the emergency charging response time constraint, and the emergency vehicle needs to max_response The charging starts within 1 second. Formula (22) constrains the length of the normal queue to not exceed the threshold L max , to avoid excessive congestion.

[0116] Step 7: Solve the optimization problem.

[0117] Aiming at the minimum cost charging optimization problem, the present invention designs a novel multi-elite search chicken swarm optimization algorithm MSCSO-PDG based on population diversity tracking to solve the intelligent charging cost model. MSCSO-PDG first introduces a population diversity tracking strategy to determine the search status of the population in real time; then, based on the algorithm's exploration and mining balance, two sets of search schemes are designed for the elite population in the population based on elite solutions and random solutions. The hen and chick groups follow the elite individuals to dynamically execute the mining and exploration modes, thereby ensuring the algorithm's optimization efficiency while achieving the accuracy and robustness of the optimization solution.

[0118] Its design is based on the Multi-elite search chicken swarm optimization algorithm based on population diversity guidance (MSCSO-PDG) to solve the optimization problem. When solving the optimization problem, each individual in the chicken swarm corresponds to a solution to the optimization problem. The hierarchy of the chicken swarm is determined according to the fitness values of the individuals in the chicken swarm, that is, the individuals with the highest fitness values are set as roosters, the individuals in the middle are set as hens, and the individuals at the bottom are set as chicks. The number of the three types of chickens is represented by the parameter N. R 、N H and N CRepresented. The hierarchical relationship between roosters, hens, and chicks, the mate relationship between roosters and hens, and the mother-child relationship between hens and chicks remain unchanged within G generations and are updated after each G iteration. The search dimension D of the swarm algorithm is defined as the decision variable dimension of the problem. The number of the charging station k selected by vehicle i in time period t is a discrete variable; the charging power of vehicle i in time period t is a continuous variable; the position of each individual in the population is represented by a position matrix. Each row of the matrix represents the vehicle number, each column represents the charging period, and each tuple contains the charging station number k and the vehicle charging power P. i t Information. Generate N flock individuals in the D-dimensional search space, each of which includes complete vehicle scheduling information. Individual position represents the value of the j-th dimension of the i-th individual at the t-th iteration.

[0119] Step 8: Initialize the optimization algorithm. Initialize the relevant parameters of MSCSO-PDG, including the population size N, the number of roosters N R , number of hens N H 、Number of chicks C , the swarm update algebra G, the maximum number of function evaluations MAX_FEs, the search step length FL, etc. Randomly generate N swarm individuals Pi = (P1, P2, …, PN), where the spatial dimension D of the swarm is the number of decision variables to be optimized. The minimized objective function of formula (1) is defined as the fitness function of the swarm algorithm.

[0120] Step 9: The process of MSCSO-PDG solving the optimization problem is as follows:

[0121] (1) Initialize the individual optimum and the global optimum, and set the initial number of iterations t to 1.

[0122] (2) Determine whether the current generation is an integer multiple of G. If so, update the hierarchical structure of the chicken swarm algorithm; otherwise, proceed to the next process.

[0123] (3) Determine the diversity of the population according to the following formula:

[0124]

[0125] According to the above formula, determine whether the population enters the exploration stage or the exploitation stage.

[0126] (4) Sort the fitness values of the chickens and determine the hierarchy, subgroup division, and mother-child relationship of the chickens. If the i-th individual is a rooster and the algorithm determines that it is in the mining stage, the following search equation is used:

[0127]

[0128] Among them, randn(0,σ 2 ) means the mean is 0 and the variance is σ 2 A Gaussian distributed random number; ε is a very small constant to avoid zero division errors; f i and f k Represents the fitness value of the i-th rooster and the k-th rooster, and k≠i. Since the equation is based on the rooster as the elite Search around, so the equation has good mining capabilities.

[0129] (5) If the i-th individual is a rooster and the algorithm determines that it is in the exploration stage, the following search equation is used:

[0130]

[0131] Where q is an individual randomly selected from {1,2,…,N}, satisfying q≠i, is a random number in the range of [-1,1]. Due to the introduction of random numbers, the rooster will be guided to a larger foraging range, so this equation has good exploration ability.

[0132] (6) If the i-th individual is a hen, it follows the rooster with index r1 to forage, and its search equation is:

[0133]

[0134] Where c1 and c2 are learning factors; rand is a random number uniformly distributed on [-1, 1]; r1 represents the rooster in the sub-group where the i-th hen is located; r2 is a rooster or hen randomly selected from the entire chicken flock, and r1≠r2≠i.

[0135] (8) If the i-th individual is a chick, it follows the hen with index m to forage, and its search equation is:

[0136]

[0137] Where m is the hen corresponding to the i-th chick; FL is a random number uniformly distributed on [0,2].

[0138] (9) Calculate the fitness value of each individual in the population after the position is updated, and update the individual's optimal x ib =(x ib,1 ,x ib,2 ,…,x ib,D ) and the global optimal x gb =(x gb,1 ,x gb,2 ,…,x gb,D ).

[0139] (10) The swarm algorithm iterates and updates the j-th dimension position of the i-th individual in the population according to formula (9-14), and records the global optimal individual position.

[0140] (11) If the current number of iterations reaches the set maximum number of stops (or reaches the minimum error requirement), the iteration is stopped and the optimal individual position in the population is output as the optimal scheduling decision variable, which includes the charging period of each vehicle, the selected charging station number and the charging power; otherwise, go to (2) to continue execution.

[0141] In summary, the embodiment of the present invention proposes a dynamic priority scheduling framework for EV charging based on the characteristics of the Energy Internet Park. For the first time, it deeply integrates the park-level renewable energy consumption, grid time-of-use electricity prices and EV charging needs, and realizes the coordinated optimization of efficient utilization of clean energy and carbon emissions. At the same time, a dynamic priority allocation mechanism based on real-time vehicle status and charging station information is proposed, which aims to optimize the charging scheduling of electric vehicles and improve charging efficiency and response speed. Taking into account the vehicle's power status (such as SOC) and location information, as well as the electricity price and load conditions of the charging station, it can intelligently schedule low-power vehicles and give priority to charging requests with high emergency needs, making charging scheduling more intelligent and humane, and ensuring the second-level response capability of vehicles with high emergency needs. The design of the system fully considers the actual usage scenarios and has good adaptability and flexibility.

[0142] Aiming at the key operating cost issue in electric vehicle (EV) charging plans, an accurate lithium battery degradation model is proposed to improve the accuracy of the operation evaluation of the intelligent charging system. Most existing studies have not fully considered the battery degradation model, or only rely on simple inference models for estimation, resulting in limited reliability of the charging scheduling strategy. By deeply studying the chemical characteristics of lithium batteries, the present invention establishes a more accurate battery degradation model that can evaluate the health status of the battery and its impact on the charging cost in real time. The introduction of this model enables the charging scheduling strategy to be optimized not only based on the current performance of the battery, but also to qualitatively and quantitatively analyze the impact of environmental factors on battery performance changes, thereby effectively improving the accuracy of the operation evaluation of the intelligent charging system. Through this innovative method, the intelligent charging system can achieve higher flexibility and adaptability, with broad application prospects and significant socio-economic value.

[0143] This paper proposes a novel multi-elite search swarm optimization algorithm (MSCSO-PDG) based on population diversity tracking, which aims to solve the complex optimization problems brought about by multi-time period, multi-power station and multi-electric vehicle charging scheduling in energy internet parks. Traditional intelligent algorithms often face the challenges of slow convergence and falling into local optimality when dealing with such high-dimensional problems. To this end, MSCSO-PDG introduces a population diversity tracking strategy in its design. By monitoring the search status of the population in real time, it dynamically adjusts the search strategy to ensure the flexibility and adaptability of the algorithm. This method develops a unique search mechanism: by combining elite solutions and random solutions, two sets of highly targeted search schemes are designed, allowing the hen and chick groups to closely follow the elite individuals and execute the mining and exploration modes respectively. This dynamic role switching not only improves the algorithm's optimization efficiency, but also significantly enhances the accuracy and robustness of the optimization solution. MSCSO-PDG can effectively overcome the limitations of traditional algorithms and provide a new solution for electric vehicle charging optimization, with important application value and broad market prospects.

[0144] The following is a further demonstration and introduction of the present invention with reference to a specific test example:

[0145] Step 1: System Model Construction. Each charging station in the park is equipped with a smart charging pile, a power controller, and a communication module. These are connected to the park system via wired connections. EV users access the park's intelligent charging dispatch system via wireless communication and report real-time status (such as SOC and location). A local server is set up within the park, using the MQTT protocol for low-latency data transmission. Vehicle status and charging station information are synchronized every five minutes, ensuring real-time interaction between charging piles, vehicles, and the server. Tables 1-2 provide information on park charging station and EV status, EV battery parameters, and time-of-use electricity prices.

[0146] Table 1: Charging station and time-of-use electricity price parameters

[0147]

[0148]

[0149] Table 2: General parameters of electric vehicles

[0150]

[0151] Step 2: Define the decision variables for each electric car: Used to identify whether electric vehicle i is connected to charging station k during time period t; Represents the charging power of electric vehicle i in time period t. Next, determine the emergency charging identification variable of the Ev charging problem: Represents the emergency charging indicator variable. When the SOC of electric vehicle i in time period t meets hour, is 1, otherwise is 0. Table 3 gives examples of three EVs.

[0152] Table 3: Specific examples of electric vehicles

[0153]

[0154] The current time period is set to 08:00 (normal time period, electricity price 0.8 yuan / kWh), i.e. t = 8. Charging station A: 2 idle charging stations; B: 1 idle charging station; C: 3 idle charging stations. EV2 triggers emergency charging, i.e. EV1 and EV3 and are all equal to 0.

[0155] Step 3: Construct the objective function of the park smart charging system:

[0156]

[0157] Among them, N EV = 3 represents the number of EVs, α = 0.6 and β = 0.4 are weight coefficients, and α + β = 1. At the current time t = 8, the total length of the time period T = 24, meaning there are 24 time periods per day. Δt = 1h represents the duration of charging power conversion to energy. χ = 1000 yuan / hour is the penalty coefficient for emergency charging delay.

[0158] represents the delay from the time when vehicle i sends the demand to the actual start of charging time t, is the waiting time; i represents the capacity degradation rate of the i-th EV battery; the calculation of the above three parameters will be given below.

[0159] Step 4: Campus EV charging strategy based on dynamic priority allocation and emergency charging queue-jumping mechanism. First, define the EV queue type, which is divided into two types. Emergency queue Q emergency : Storage satisfaction EV, SOC emergency It can be set according to specific circumstances and needs. normal : Store other vehicles and sort them on a first-come, first-served basis. State variables represents the queue position of vehicle i at charging station k (0 means charging); This is an emergency sign, with the same definition as above.

[0160] (1): Calculation of EV charging demand. req,iIt represents the total charging power requirement of the i-th electric vehicle (unit: kWh), that is, the total battery capacity is E bat,i The electric vehicle needs to be upgraded from the current SOC (SOC initial,i ) to the target SOC (SOC targct,i ) The total amount of electricity required. According to Table 3, we can calculate that EV1 needs to charge 36kWh, EV2 needs to charge 37.5kWh, and EV3 needs to charge 38.5kWh.

[0161] (2): Emergency vehicle queue-jumping rule: When vehicle i triggers emergency charging When , it is immediately inserted into the front of the queue of charging station k, which can be expressed as:

[0162]

[0163] Then the ordinary vehicles in the original queue are postponed one position. The emergency vehicle EV2 is immediately inserted into the front of the queue of the nearest charging station (B) without waiting.

[0164] (3): Update rules for ordinary vehicle queues: Whenever time period t is updated, vehicle i in the ordinary queue is adjusted according to the first-come-first-served principle:

[0165]

[0166] in is the position of vehicle i in the normal queue sequence at charging station k at time t+1. Assume that there are operating vehicles that have not been included in the analysis and have queued in advance at the charging station. Then For the ordinary car EV1, the A power station can be selected; For ordinary EV3 cars, a C power station is optional.

[0167] (4): Queuing time calculation. The waiting time of vehicle i in time period t and delay penalty The calculation of is as follows:

[0168]

[0169] Among them, t start,i and t req,i They are the initial period when vehicle i issues a charging request and the period when charging actually starts: is an indicator function, which takes 1 if the i-th vehicle belongs to the normal queue, and takes 0 otherwise. The specific calculation results of the three vehicles are shown in Table 4:

[0170] Table 4: Calculation of waiting delay time for electric vehicles

[0171]

[0172] Step 5: Accurately model the battery wear cost. i Used to measure the battery loss cost caused by charging the i-th EV. The mainstream battery of EV is lithium-ion battery, which will age during use, thereby generating aging cost. Lithium-ion batteries are divided into calendar aging and cycle aging. Calendar aging refers to the phenomenon that the capacity of the battery slowly decreases as the shelf time increases during the shelf process. This aging will occur even if the battery is not used and is in a static state. Cycle aging is the irreversible capacity loss that occurs in lithium-ion batteries during the charging cycle. Cycle aging only occurs when the battery is charging. Most current studies have not given an accurate EV lithium battery capacity degradation model. In the present invention, based on the semi-empirical degradation characteristic data of EV lithium batteries, the capacity degradation rate γ is designed. i To accurately describe the battery degradation process, it is calculated as follows:

[0173]

[0174] L deg =L t +L c (7)

[0175]

[0176] f t (Δt,σ,T)=S t (Δt)S σ (σ)S T (T)(10)

[0177] f c (δ,σ,T)=S δ (δ)S σ (σ)S T (T)(11)

[0178]

[0179] S t (Δt)=k t Δt(15)

[0180] Among them, formula (6) is to calculate γ i , where C per =100 yuan / kwh is the cost of battery capacity, L deg The battery capacity degradation caused by EV charging is assumed in this paper to be the same for all EVs, so i is no longer used to distinguish individual EVs. deg Including calendar aging capacity L t and cycle aging capacity L c Two aspects. t and Lc Calculated by formula (8)-(9), α sei and β sei These are the chemical characteristic parameters of the battery, which are 5.75e-2 and 121:f respectively. t and f c Represent the calendar aging rate and cycle aging rate of EV batteries respectively. From formulas (10)-(11), we can see that f t and f c They can be expressed as depth of charge δ, average soc(σ), battery temperature T c and the charging duration Δt; and the above four quantities have corresponding stress factor models S δ (δ), S σ (σ), S T (T) and S t (Δt), used to describe the effects of four quantities on battery aging, which are described by formulas (12)-(15) respectively, where σ ref and T ref For experimental calibration reference values, the fraction is equal to 50% and 298k; k δ1 , k δ2 , k δ3 , k σ , k T and k t are the battery experimental parameters, which are equal to 1.40e-5, -5.01e-1, -1.23e5, 1.04, 6.93e-2 and 4.14e-10 respectively. The depth of charge δ, average soc(σ), and battery temperature T of EV1-EV3 c The specific settings of the charging duration Δt and the charging time calculated according to formulas (6)-(15) within Δt The corresponding total degradation rate of lithium batteries is as follows:

[0181] Table 5: Calculation of waiting delay time for electric vehicles

[0182]

[0183] Step 6: The constraints of the system model are:

[0184]

[0185] In the above constraints, formulas (16)-(17) are the charging power interval constraints and total charging request constraints of each electric vehicle, which prevent the charging power from exceeding the upper and lower bounds of the power; formulas (18)-(19) are the update and boundary constraint formulas of SOC, which define the update rules of SOC and the upper and lower bounds of SOC. η = 90% is the charging efficiency, and E bat,iRepresents the rated capacity of the battery of vehicle i. The three vehicles have 60, 50 and 70 kWh respectively; Formula (20) is the capacity constraint of the charging station, N k =1000kwh is the maximum charging capacity of the kth charging station, which is the same for the three charging stations; Formula (21) is the emergency charging response time constraint, and the emergency vehicle needs to be charged within T max_response =30 and start charging. Formula (22) constrains the length of the normal queue to not exceed the threshold L max =10, avoid excessive congestion.

[0186] Step 7: Solve the optimization problem. Design a multi-elite search chicken swarm optimization algorithm based on populationdiversity guidance (MSCSO-PDG) to solve the optimization problem. When solving the optimization problem, each individual in the chicken swarm corresponds to a solution to the optimization problem. The hierarchy of the chicken swarm is determined according to the fitness value of the individuals in the chicken swarm, that is, the individuals with the highest fitness value are set as roosters, the individuals in the middle are set as hens, and the individuals at the bottom are set as chicks. The number of the three types of chickens is represented by the parameter N R 、N H and N C Represented by. The hierarchical relationship between roosters, hens, and chicks, the mate relationship between roosters and hens, and the mother-child relationship between hens and chicks remain unchanged within G = 10 generations and are updated after every 10 iterations. The search dimension of the swarm algorithm is defined as the decision variable dimension of the problem. The number of the charging station k selected by vehicle i in time period t is a discrete variable; the charging power of vehicle i in time period t is a continuous variable; the position of each individual in the population is represented by a position matrix. Each row of the matrix represents the vehicle number, each column represents the charging period, and each tuple contains the charging station number k and the vehicle charging power P i t Information. Generate N flock individuals in the D-dimensional search space, each of which includes complete vehicle scheduling information. Individual position Represents the value of the j-th dimension of the i-th individual at the t-th iteration. The position of the third individual in the flock is generated as follows:

[0187]

[0188] Step 8: Initialize the optimization algorithm. Initialize the relevant parameters of MSCSO-PDG, including N = 20, the number of roosters N = 0.2N, R , number of hens N H =0.6N, R 、Number of chicks C= 0.2N, update generation G = 10, maximum function evaluation times MAX_Fes = 500, search step length FL = rand(0.4, 0.9), etc. A swarm of 20 individuals Pi = (P1, P2, …, PN) is randomly generated, and the spatial dimension D of the swarm is the number of decision variables to be optimized. The minimized objective function of formula (1) is defined as the fitness function of the swarm algorithm.

[0189] Step 9: The process of MSCSO-PDG solving the optimization problem is as follows:

[0190] (1) Initialize the individual optimum and the global optimum, and set the initial number of iterations t to 1.

[0191] (2) Determine whether the current generation is an integer multiple of 10. If so, update the hierarchical structure of the chicken swarm algorithm; otherwise, proceed to the next process.

[0192] (3) Determine the diversity of the population according to the following formula:

[0193]

[0194]

[0195] According to the above formula, determine whether the population enters the exploration stage or the exploitation stage.

[0196] (4) Sort the fitness values of the chickens and determine the hierarchy, subgroup division, and mother-child relationship of the chickens. If the i-th individual is a rooster and the algorithm determines that it is in the mining stage, the following search equation is used:

[0197]

[0198] Among them, randn(0,σ 2 ) means the mean is 0 and the variance is σ 2 A Gaussian distributed random number; ε = 0.000001 is a very small constant to avoid zero division errors; f i and f k Represents the fitness value of the i-th rooster and the k-th rooster, and k≠i. Since the equation is based on the rooster as the elite Search around, so the equation has good mining capabilities.

[0199] (5) If the i-th individual is a rooster and the algorithm determines that it is in the exploration stage, the following search equation is used:

[0200]

[0201] Where q is an individual randomly selected from {1,2,…,N}, satisfying q≠i, is a random number in the range of [-1,1] and is updated every generation. Due to the introduction of random numbers, the rooster will be guided to a larger foraging range, so this equation has good exploration ability.

[0202] (6) If the i-th individual is a hen, it follows the rooster with index r1 to forage, and its search equation is:

[0203]

[0204] Where c1 and c2 are learning factors; rand is a random number uniformly distributed on [-1, 1]; r1 represents the rooster in the sub-group where the i-th hen is located; r2 is a rooster or hen randomly selected from the entire chicken flock, and r1≠r2≠i.

[0205] (8) If the i-th individual is a chick, it follows the hen with index m to forage, and its search equation is:

[0206]

[0207] Where m is the hen corresponding to the i-th chick; FL is a random number uniformly distributed on [0,2].

[0208] (9) Calculate the fitness value of each individual in the population after the position is updated, and update the optimal x of the current generation individual ib =(x ib,1 ,x ib,2 ,…,x ib,D ) and the global optimal x gb =(x gb,1 ,x gb,2 ,…,x gb,D ).

[0209] (10) The swarm algorithm iterates and updates the j-th dimension position of the i-th individual in the population according to formula (9-14), and records the global optimal individual position.

[0210] (11) If the current number of iterations reaches the set maximum number of stops (or reaches the minimum error requirement), the iteration stops and the optimal individual position in the population is output as the optimal scheduling decision variable, which includes the charging period of each vehicle, the selected charging station number and the charging power; otherwise, go to (2) to continue execution. According to the algorithm, the final optimization result of the three vehicles is and the optimal scheduling cost is:

[0211] Optimal scheduling decision:

[0212] Optimal scheduling cost:

[0213] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0214] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium having a computer program stored thereon, which executes the above method when executed by a processor. The storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component.

[0215] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0216] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present disclosure. Various changes and improvements may be made to the present disclosure without departing from the spirit and scope of the present disclosure, and such changes and improvements shall fall within the scope of the present disclosure.

[0217] The present invention is not limited to the above-mentioned optimal implementation mode. Anyone can derive various other forms of methods for determining rotor design parameters based on plateau environments under the guidance of the present invention. All equal changes and modifications made within the scope of the patent application of the present invention should fall within the scope of the present invention.

Claims

1. An energy internet park EV charging scheduling system based on dynamic priority and battery life optimization, characterized by: include: Dynamic priority allocation module: Real-time monitoring of vehicle battery SOC. When SOC ≤ preset threshold, the vehicle is inserted into the front of the charging queue. The general queue is arranged on a first-come, first-served basis. When an emergency vehicle cuts in line, the general vehicles will be delayed; Lithium battery degradation calculation module: Quantify the battery capacity degradation rate based on a multi-stress factor model based on charge depth, average SOC, temperature, and charging time; The multi-stress factor model calculates the degradation amount through experimentally calibrated chemical property parameters; Multi-objective optimization module: An optimization algorithm that dynamically switches between global exploration and local mining generates a scheduling strategy with the goal of minimizing charging costs, user waiting time, and battery degradation costs. The optimization algorithm improves solution efficiency through a population diversity tracking strategy and an elite-random dual search mechanism.

2. The energy internet park EV charging scheduling system based on dynamic priority and battery life optimization according to claim 1 is characterized by: The multi-objective optimization module also includes: Population division unit: divide the optimized population into elite individuals, medium individuals and ordinary individuals; Diversity monitoring unit: calculates population diversity indicators in real time and dynamically switches between global exploration and local mining modes; Search strategy execution unit: elite individuals perform local search based on the current optimal solution, and ordinary individuals perform global search following random solutions.

3. The energy internet park EV charging scheduling system based on dynamic priority and battery life optimization according to claim 1 is characterized by: The specific implementation of the multi-stress factor model includes: Calendar aging calculation unit: Calculates calendar aging capacity loss based on charging duration, average SOC, and battery temperature through experimental calibration parameters; Cycle aging calculation unit: Calculates cycle aging capacity loss based on depth of charge, average SOC and battery temperature through experimental calibration parameters; Total degradation rate integration unit: Add the calendar aging and cycle aging capacity losses and multiply them by the unit capacity cost to output the total degradation rate.

4. The energy internet park EV charging scheduling system based on dynamic priority and battery life optimization according to claim 1 is characterized by: Also includes real-time data interaction module: Low-latency communication between vehicles, charging piles and servers is achieved based on the MQTT protocol, with a data update cycle of ≤5 minutes; Dynamically synchronize charging station load, time-of-use electricity prices, and renewable energy output data.

5. The energy internet park EV charging scheduling system based on dynamic priority and battery life optimization according to claim 1 is characterized by: The dynamic priority allocation module also includes: Queue length control unit: used to limit the maximum length of ordinary queues to avoid high concurrency congestion; Emergency response constraint unit: used to set the maximum response time threshold for emergency vehicles to start charging.

6. The energy internet park EV charging scheduling system based on dynamic priority and battery life optimization according to claim 1 is characterized by: The multi-objective optimization module integrates time-of-use electricity prices and a renewable energy scheduling unit, and is used to prioritize charging demands during off-peak electricity prices and periods of high renewable energy output.

7. An energy internet park EV charging scheduling method, characterized in that: The following steps are involved: Real-time monitoring of vehicle battery SOC. When SOC ≤ preset threshold, the vehicle is inserted into the front of the charging queue, and ordinary vehicles are deferred on a first-come, first-served basis. Quantify the battery capacity degradation rate based on a multi-stress factor model based on charge depth, average SOC, temperature, and charging time; An optimization algorithm that dynamically switches between global exploration and local mining is used to generate a scheduling strategy that minimizes charging costs, user waiting time, and battery degradation costs.

8. The method for scheduling EV charging in an energy internet park according to claim 7, characterized in that: The optimization algorithm achieves dynamic switching between global exploration and local mining by: Divide the population into elite individuals, middle individuals and ordinary individuals; Real-time monitoring of population diversity, dynamically switching between global exploration and local mining modes; Elite individuals perform local search based on the current optimal solution, and ordinary individuals perform global search following random solutions.

9. The method for scheduling EV charging in an energy internet park according to claim 8, characterized in that: The optimization algorithm specifically includes: Population dynamics classification and role allocation: The optimized population is divided into three categories: elite individuals, medium individuals and ordinary individuals, corresponding to the roles of roosters, hens and chicks respectively; Elite individuals account for 20% and dominate the mining of local optimal solutions; Ordinary individuals account for 20% and perform global exploration; Medium individuals account for 60%, balancing local and global searches; Population diversity tracking and mode switching: Real-time calculation of population diversity indicators to reflect the distribution dispersion of individuals in the search space; When the diversity index is lower than the historical threshold and the optimal solution has not been improved, the global exploration mode is triggered to expand the search scope; When the diversity index is higher than the threshold or the optimal solution is continuously optimized, the local mining mode is triggered to focus on the current optimal solution; Differentiated search strategies: The elite individual performs a local fine search based on the current optimal solution, and the disturbance range is adaptively adjusted; Ordinary individuals follow random solutions to perform global exploration and introduce random perturbations to avoid local optimality; Intermediate individuals refer to both elite solutions and random solutions, dynamically balancing exploration and exploitation; Dynamic character periodic adjustments: Every fixed generation, the population roles are redistributed according to the fitness value to adapt to the changes in the search state.

10. The method for EV charging scheduling in an energy internet park according to claim 7, characterized in that: The calculation of the battery capacity degradation rate includes the following steps: Total degradation rate calculation: multiply the unit capacity cost by the sum of calendar aging capacity loss and cycle aging capacity loss; Calendar aging capacity loss: calculated based on the battery chemical characteristic parameters experimentally calibrated based on the charging duration, average state of charge and battery temperature; Cycle aging capacity loss: calculated based on the battery chemical characteristic parameters experimentally calibrated based on the depth of charge, average state of charge and battery temperature; In the multi-stress factor model, chemical property parameters are calibrated through battery aging experiments to reflect the nonlinear attenuation characteristics of battery materials.

Images