An orderly charging system guided by an interactive real-time pricing strategy for electric vehicle charging load
By constructing an orderly charging model through electricity price guidance and genetic algorithms, the problem of grid load fluctuations caused by disorderly charging of electric vehicles was solved, and load peak shaving and valley filling and user willingness adjustment were realized, thereby improving grid stability and user satisfaction.
Patent Information
- Application Number
- CN202111142207.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-09-28
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2041-09-28
AI Technical Summary
Disorderly charging of electric vehicles causes fluctuations in the regional power grid load, resulting in peak loads being compounded, affecting power supply reliability and power quality, and also causing a waste of power supply resources.
By adopting an electricity price-guided strategy, an ordered charging model is constructed through a genetic algorithm. Combined with predictions of user charging intentions, the electricity price is dynamically adjusted to guide EV charging behavior and achieve load shaving and valley filling.
It effectively smooths out grid load fluctuations, reduces peak-valley differences, lowers user charging costs, and improves grid stability and user satisfaction.
Smart Images

Figure CN114004450B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an electric vehicle charging pricing model based on real-time adjustment of regional power grid load and an orderly charging model guided by real-time electricity prices. The present invention is applicable to a real-time electricity price guidance strategy adopted to reduce the peak-to-valley difference of the regional power grid and control the large-scale access of electric vehicle loads, and belongs to the field of power marketing technology. Background Art
[0002] Under the current environmental protection situation, electric vehicles have developed rapidly, and EV loads, as a new form of load, account for an increasingly important proportion in the power system. As a link between the power generation and transmission system and end users, the distribution network is an important link to ensure power supply reliability and improve the operation of the power grid. With the increasing penetration rate of flexible loads represented by EVs, their randomness makes the operating state of the system more complicated, thus bringing new challenges to the safe and stable operation of the distribution network.
[0003] At present, EV owners are not subject to the allocation of charging piles. Disorderly charging is mainly based on the behavior of car owners. Car owners only consider the convenience and efficiency of charging. However, large-scale EV load access to the power grid may cause regional power grid load fluctuations. Charging during the peak period of the original load of the power grid will cause load "peak on peak", exceeding the power supply capacity and bearing capacity of the regional power grid, causing transformer overload, affecting power supply reliability and power quality. In turn, it affects the normal electricity consumption of residents, limits the charging load, prolongs the charging time, and leads to user complaints.
[0004] In order to solve the current situation of tight power supply capacity in regional distribution networks, power grid companies often increase investment to meet the growing demand for electric vehicles. The charging of electric vehicles is highly intermittent, which causes the peak-to-valley difference to widen. When EVs are not charging, it causes a waste of power supply resources.
[0005] Therefore, understanding the temporal distribution characteristics of EV load and using electricity price guidance strategies to smooth load fluctuations are important measures to maintain regional power grid load stability while improving user satisfaction, but there are currently no related reports in the existing technology. Summary of the invention
[0006] The purpose of the present invention is to solve the above-mentioned problems and provide an orderly charging system guided by an interactive real-time pricing strategy for electric vehicle charging loads. The system fully analyzes the impact of electricity price guidance strategies on the characteristics of EV load charging and discharging behaviors, fully considers residents' charging willingness, studies the disorderly charging model of EV charging in residential areas from the perspective of the power grid, considers the charging needs of EVs, takes grid load peak shaving and valley filling as the first goal, and proposes time-of-use electricity price pricing rules for achieving orderly charging control in residential areas.
[0007] The orderly charging model guided by the interactive real-time pricing strategy of electric vehicle charging load of the present invention is special in that it includes the following steps:
[0008] 1) Construction conditions of regional power grid electric vehicle charging model
[0009] The user's charging habits will directly affect the time distribution characteristics of the charging load. The three key factors of starting charging time, mileage, and SOC are analyzed, and the Monte Carlo random simulation of the charging behavior of a large number of EVs is used;
[0010] 2) Build an orderly charging load model
[0011] The genetic algorithm is used to solve the optimal time for EV charging under the guidance of electricity prices. The calculation process is as follows:
[0012] (1) Initialization: Set the evolutionary generation counter t = 0, set the maximum evolutionary generation T, and randomly generate M individuals as the initial population P(0);
[0013] (2) Individual evaluation: Calculate the fitness of each individual in the population P(t);
[0014] (3) Selection operation: The selection operator is applied to the population. The purpose of selection is to directly pass on the optimized individuals to the next generation or to generate new individuals through pairing and crossover and then pass on to the next generation. The selection operation is based on the fitness evaluation of individuals in the population.
[0015] (4) Crossover operation: The crossover operator is applied to the population. The crossover operator plays a core role in the genetic algorithm.
[0016] (5) Mutation operation: Apply the mutation operator to the population. That is, the gene values at certain loci of the individual strings in the population are changed. After the population P(t) undergoes selection, crossover, and mutation operations, the next generation population P(t+1) is obtained.
[0017] (6) Termination condition judgment: If t = T, the individual with the maximum fitness obtained in the evolution process is output as the optimal solution and the calculation is terminated;
[0018] The following are the optimization conditions set during the solution process:
[0019] (1) To ensure the travel needs of EV owners, the total charging load of EVs in T time periods should remain unchanged, that is, the orderly charging load and the disorderly charging load are equal, as shown in formula (1):
[0020]
[0021] In the formula, x oi is the disordered charging load of EVs in period i, and the genetic algorithm is used to find xi Optimal solution; x i The orderly charging load of EVs in period i is calculated by controlling the orderly charging of EVs to solve the optimal charging load at time i;
[0022] In addition, in order to avoid causing a greater load burden on the power grid, resulting in a higher load peak or the phenomenon of "peak-to-peak inversion", it is required that after the optimization, the peak value of the power grid load curve does not exceed the peak value of the original load curve, as shown in formula (2):
[0023] Q max ≤Q 0max (2)
[0024] In the formula, Q max is the peak value of the power grid load curve after dispatch; Q 0max The peak value of the total load curve of residential load and EV before dispatch;
[0025] (2) On the basis of reducing the peak-valley difference, the genetic algorithm is used for further optimization to achieve the goal of reducing user charging costs, and the EV charging load is transferred to the time period with relatively low electricity costs. At the same time, it should be considered that if a large number of EVs start charging at the same time at a time point with low electricity costs, a new load peak will be generated, resulting in a "peak-valley inversion" phenomenon. To avoid this phenomenon, during the time period with low charging electricity costs, the start charging time of the EV after scheduling should also take into account the load changes and then determine the number of EVs connected to the grid at a certain time;
[0026] 3) Prediction of user charging willingness
[0027] The user's charging willingness is nonlinear and unpredictable. The fuzzy intelligent algorithm is used to simplify the user's charging willingness into a linear system, as follows:
[0028] Household EVs are parked for 95% of the day. The parking time, time-of-use electricity price, and EV SOC are used as the evaluation criteria. These three quantities are used as the input of the fuzzy inference system model, and the input membership function is set. The output is the charging probability of the EV. The joint Gaussian membership function is shown in formula (3):
[0029]
[0030] Where x is the time-of-use electricity price, δ is the SOC of the EV, and c is the parking duration.
[0031] Through the above calculations, we get the EV charging probability values affected by these three factors. The membership function simplifies the nonlinear variable calculation of user charging willingness into several intuitive data, the value of probability f. By changing the values of x, δ and c, the EV charging probability f will tend to these values.
[0032] 4) Disordered charging model considering user willingness
[0033] By calculating the charging probability of each vehicle and judging whether the vehicle is charging during this period, a disorderly charging model considering the user's charging willingness is simulated, as follows:
[0034] (1) First, the SOC, starting charging time, and parking duration of the first EV are simulated, and the charging probability of the vehicle is calculated using a fuzzy algorithm based on the current time-of-use electricity price.
[0035] (2) Use a random function to determine whether the vehicle is charging and calculate its charging power based on the selected fast charging or conventional charging method;
[0036] (3) Calculate the charging load of all EVs charged during the hour;
[0037] (4) After calculating the EV charging load within 24 hours, it is superimposed on the residential load curve to obtain the load curve of the regional power grid under the condition of disorderly EV charging;
[0038] 5) Pricing strategy considering electric vehicle load characteristics
[0039] Assuming that the electricity price used by the current power grid is a real-time electricity price, the change in the load rate of the community transformer is collected every 1 hour, and the electricity price is adjusted accordingly based on the collected power change information. The total charging cost that users in the area need to pay in one day is shown in formula (4):
[0040]
[0041] Where pj is the charging electricity price in period j; Pi is the power of the i-th vehicle; N is the total number of private electric vehicles in the residential area; t is the continuous charging time of EV; x ij There are only two values: 0 or 1. The value 0 or 1 represents whether the i-th tram studied has charging behavior in the j-th period; 1 represents that the tram is charging, and 0 represents that the tram is idle, as shown in formula (5):
[0042]
[0043] By changing the current charging price, the charging willingness of some users can be increased or decreased, thereby changing the starting charging time of these users, avoiding large-scale charging load access to the power grid during peak hours of residential electricity consumption, which would affect the stability of the power supply of the power grid.
[0044] 6) Orderly charging model under time-of-use electricity price considering user charging willingness
[0045] Take the sum of the disordered charging load and the residential load in the 24th hour of the previous day, and compare it with the average total load to get the load multiple n1. The electricity price at the next moment (i.e., 1 o'clock) is obtained through the load multiple. This electricity price will guide the load at points 1-2. Then calculate the load at points 1-2, and compare it with the average total load to get the load multiple n2, and get the electricity price at the next moment. By analogy, the electricity price and load changes at each moment of the day are calculated. By controlling the electricity price, the user's charging willingness can be changed, that is, the charging probability can be changed. By reducing or increasing the charging probability, the size of the EV charging load at a certain moment can be adjusted as needed.
[0046] The present invention proposes a method for calculating the "peak and valley" time-of-use electricity price based on the dynamic adjustment of the real-time load of the regional power grid; proposes an idea and method for building an orderly charging model for EV charging using a genetic algorithm; proposes a simplified calculation method for the charging probability of EV users; by collecting the real-time load value of the regional power grid, dynamically formulating the EV charging electricity price at the current moment, changing the charging willingness of EV users, and guiding EV charging behavior, the uninterrupted real-time interactive mutual influence between the regional power grid load and the EV charging electricity price is avoided, thereby avoiding the widening of the peak-valley difference of the power grid caused by the uncontrollable charging behavior of a large number of EVs, and further achieving the win-win goal of reducing the charging cost of EV users and reducing the peak-valley difference of the regional power grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 :A flow chart of building an orderly charging load model in an orderly charging model guided by an interactive real-time pricing strategy for electric vehicle charging loads of the present invention;
[0048] Figure 2 :A flow chart of predicting user charging willingness in an orderly charging model guided by an interactive real-time pricing strategy for electric vehicle charging loads in the present invention;
[0049] Figure 3 :A flow chart of a disordered charging model taking into account user willingness in an ordered charging model guided by an interactive real-time pricing strategy for electric vehicle charging loads according to the present invention;
[0050] Figure 4 :A flowchart of a pricing strategy taking into account electric vehicle load characteristics in an orderly charging model guided by an interactive real-time pricing strategy for electric vehicle charging load of the present invention;
[0051] Figure 5 : A flow chart of an orderly charging model under time-of-use electricity price considering user charging willingness in an orderly charging model guided by an interactive real-time pricing strategy for electric vehicle charging loads of the present invention;
[0052] Figure 6 :Schematic diagram of peak and valley electricity prices in Shandong power grid. DETAILED DESCRIPTION
[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0054] Example 1
[0055] This embodiment is an orderly charging model guided by an interactive real-time pricing strategy for electric vehicle charging loads. Figure 1-5 , including the following structures and steps:
[0056] 1) Construction conditions of regional power grid electric vehicle charging model
[0057] User charging habits (such as the choice of charging start time) will directly affect the time distribution characteristics of charging load. In the absence of regulatory measures, users tend to choose "charging after driving demand ends" or "charging when driving demand cannot be met". The time distribution characteristics of EV charging load are extremely important for regional loads. If the user's charging load peak coincides with the basic power load peak of the regional power grid, it may cause insufficient power supply capacity. In the study of the time distribution characteristics of charging load, we focus on the analysis of key factors such as the starting charging time, mileage, and SOC; we use a Monte Carlo (MC) random simulation model based on probability statistics, which is suitable for analyzing the random charging behavior of a large number of EVs;
[0058] EV charging behavior in different charging modes has different impacts on regional loads. Slow charging requires a longer charging time each time. Fast charging uses high current to charge EVs, which can complete the replenishment of vehicle energy in a shorter time, but this charging mode has a greater impact on the power grid and will also reduce the life of the battery. The user's choice of charging mode directly affects the charging power level and duration at a certain moment. The user's choice of charging mode is random and limited by the charging facilities installed at the location. Therefore, this feature needs to be considered and analyzed when modeling the charging load, and the real-time output power of the charging facility can be used as a decision variable in the orderly charging study.
[0059] In addition, different charging facilities have different charging efficiencies (generally greater than 90%). When the actual power requirements of the vehicles are the same, the higher the charging efficiency, the closer the demand for regional power grid load is to the actual demand of the EV. Therefore, 90% is used as the charging efficiency of the equipment during simulation.
[0060] EV charging behavior in different charging modes has different impacts on regional loads. The user's choice of "fast charging" or "slow charging" at a certain moment will directly affect the charging power level and duration at a certain moment. Through market research, it can be seen that EV load also has obvious peak and valley characteristics. This article takes electric private cars as an example. After get off work, 19:00 to 21:00 is the peak time for charging, and 5:00 to 7:00 in the morning before work and 12:00 to 14:00 in the afternoon are small peak times for charging.
[0061] 2) Peak and valley time-of-use electricity price theory
[0062] The present invention borrows the peak-valley time-of-use electricity price theory and further proposes a time-of-use electricity price pricing method based on the real-time dynamic changes of the power grid load "peak", "flat" and "valley".
[0063] In order to reduce the difference between peak and valley loads in the power grid and encourage users to reasonably allocate electricity usage time, the power grid company divides the 24 hours of the day into "peak time", "normal time" and "valley time", and the fees charged in different time periods are also different. In the context of smart grids, power grid companies can guide users' charging behavior by formulating peak and valley time-of-use electricity prices. The model of peak and valley time-of-use electricity prices is shown in formula (1):
[0064]
[0065] Where: C p , C f and C v are peak electricity price, flat electricity price and valley electricity price respectively, C p >C f >C v ; T 1 , T 2 , T 3 The peak, flat and valley electricity price periods are respectively. The peak and valley electricity prices are determined by the local basic electricity price level and the scale of EV development. After the electricity price is determined, electric car owners can choose the charging and discharging time period independently, achieving the electricity price guidance effect in the smart grid environment.
[0066] Many areas in my country are implementing the peak-valley electricity price system, and the peak-valley electricity price is even close to 3 times. Taking Shandong Power Grid as an example, the peak-valley electricity price is Figure 6 shown.
[0067] The current peak-valley time-of-use electricity price cannot guide the charging behavior of electric vehicles well, and there is no special charging electricity price for electric vehicles. This paper collects the real-time load of the regional power grid in real time and formulates a real-time electricity price pricing strategy, which can guide the charging behavior of electric vehicles well.
[0068] 3) Build an orderly charging load model
[0069] Genetic algorithm is used to solve the optimal time for EV charging under the guidance of electricity price. Genetic algorithm is a method to search for the optimal solution by simulating the natural evolution process. This algorithm uses mathematical methods and computer simulation operations to convert the problem-solving process into a process similar to the crossover and mutation of chromosome genes in biological evolution. When solving more complex combinatorial optimization problems, it can usually obtain better optimization results faster than some conventional optimization algorithms. The basic operation process is as follows:
[0070] (1) Initialization: Set the evolutionary generation counter t = 0, set the maximum evolutionary generation T, and randomly generate M individuals as the initial population P(0).
[0071] (2) Individual evaluation: Calculate the fitness of each individual in the population P(t).
[0072] (3) Selection operation: Apply the selection operator to the population. The purpose of selection is to directly pass on the optimized individuals to the next generation or to generate new individuals through pairing and crossover and then pass them on to the next generation. The selection operation is based on the fitness evaluation of individuals in the population.
[0073] (4) Crossover operation: Apply the crossover operator to the population. The crossover operator plays a core role in the genetic algorithm.
[0074] (5) Mutation operation: Apply the mutation operator to the population. That is, change the gene values at certain loci of the individual strings in the population. After the population P(t) undergoes selection, crossover, and mutation operations, the next generation population P(t+1) is obtained.
[0075] (6) Termination condition judgment: If t = T, the individual with the maximum fitness obtained in the evolution process is output as the optimal solution and the calculation is terminated.
[0076] The following are the optimization conditions set during the solution process:
[0077] (1) To ensure the travel needs of EV owners, the total charging load of EVs in T time periods should remain unchanged, that is, the orderly charging load and the disorderly charging load are equal, as shown in formula (2):
[0078]
[0079] In the formula, x oi is the disordered charging load of EVs in period i, and the genetic algorithm is used to find x i Optimal solution; x i is the orderly charging load of EV in period i, that is, by controlling the orderly charging of EV, the optimal charging load at time i is solved.
[0080] In addition, in order to avoid causing a greater load burden on the power grid, resulting in a higher load peak or the phenomenon of "peak-to-peak inversion", it is required that after the optimization, the peak value of the power grid load curve does not exceed the peak value of the original load curve, as shown in formula (3):
[0081] Q max ≤Q 0max (3)
[0082] In the formula, Q max is the peak value of the power grid load curve after dispatch; Q 0max The peak value of the total load curve of residential load and EV before dispatch;
[0083] (2) On the basis of reducing the peak-valley difference, the genetic algorithm is used for further optimization to achieve the goal of reducing users' charging costs and shifting the EV charging load to the time period with relatively low electricity rates. At the same time, it should be considered that if a large number of EVs start charging at the same time at a time point with low electricity rates, a new load peak will be generated, resulting in a "peak-valley inversion" phenomenon. To avoid this phenomenon, during the time period with low charging electricity rates, the starting charging time of the EV after scheduling should also take into account the load changes and then determine the number of EVs connected to the grid at a certain moment.
[0084] 4) Prediction of user charging willingness
[0085] The user's charging willingness is nonlinear and unpredictable. The fuzzy intelligent algorithm is used to simplify the user's charging willingness into a linear system, as follows:
[0086] Household EVs are parked for 95% of the day. The parking time, time-of-use electricity price, and EV SOC are used as the evaluation criteria. These three quantities are used as the input of the fuzzy inference system model, and the input membership function is set. The output is the charging probability of the EV. The joint Gaussian membership function is shown in formula (4):
[0087]
[0088] In the formula, x is the time-of-use electricity price, δ is the SOC of the EV, and c is the parking time. The EV charging probability value affected by these three factors can be obtained. The membership function simplifies the nonlinear variable calculation of the user's charging willingness into several intuitive data, the value of the probability f (for example, 10%, 30%, 50%, 70%, 90%). By changing the values of x, δ and c, the EV charging probability f will tend to these values.
[0089] For the fuzzy inference system, the present invention establishes a set of 27 charging decision rules, as shown in Table 1, with five fuzzy subsets of charging probability output membership functions:
[0090] Table 1 Fuzzy rules table
[0091]
[0092]
[0093] This embodiment divides the user's charging probability into five categories, namely "small, small, medium, large, and large". After defuzzification, the program generates the user's charging probability under different circumstances. By changing the values of electricity price, SOC, and parking time, the corresponding charging probability will also change.
[0094] 5) Disordered charging model considering user willingness
[0095] By calculating the charging probability of each vehicle and judging whether the vehicle is charging during this period, a disorderly charging model considering the user's charging willingness is simulated, as follows:
[0096] (1) Users who choose fast charging have short parking time. Such users need to consider their willingness to charge, that is, the charging start time of such users is affected by SOC, electricity price, and parking time;
[0097] (2) For users who choose regular charging, their willingness to charge also needs to be considered. Since these users park for a long time, the starting time of charging is greatly affected by electricity prices. The charging habits of these users can be guided through electricity price strategies.
[0098] (1) First, the SOC (charging time), starting charging time, and parking time of the first EV are simulated. Combined with the current time-of-use electricity price, the fuzzy algorithm is used to calculate the charging probability of the vehicle.
[0099] (2) Use a random function to determine whether the vehicle is charging and calculate its charging power based on the selected fast charging or conventional charging method.
[0100] (3) Calculate the charging load of all EVs charged in that hour
[0101] (4) After calculating the EV charging load within 24 hours, it is superimposed on the residential load curve to obtain the load curve of the regional power grid under the condition of disorderly EV charging. EV charging increases the peak-to-valley difference of the regional power grid.
[0102] 6) Pricing strategy considering electric vehicle load characteristics
[0103] Taking into account the interests of users and power grids, changes in electricity prices are used to guide and encourage EV users to adjust their charging times, so as to minimize user charging costs while avoiding peaks on peaks. The transfer of charging loads is used to improve the load curve during valley periods, reduce the peak-to-valley difference in the power grid during the current period, and further improve the stability of the power grid.
[0104] Assuming that the electricity price used by the current power grid is a real-time electricity price, the change in the load rate of the community transformer is collected every 1 hour, and the electricity price is adjusted accordingly based on the collected power change information. The total charging cost that users in the area need to pay in one day is shown in formula (5):
[0105]
[0106] Where pj is the charging electricity price in period j; Pi is the power of the i-th vehicle; N is the total number of private electric vehicles in the residential area; t is the continuous charging time of EV; x ij There are only two values: 0 or 1. The value 0 or 1 represents whether the i-th tram studied has charging behavior in the j-th period; 1 represents that the tram is charging, and 0 represents that the tram is idle, as shown in formula (6):
[0107]
[0108] By changing the current charging price, the charging willingness of some users can be increased or decreased, thereby changing the starting charging time of these users, avoiding large-scale charging load access to the power grid during peak hours of residential electricity consumption, which would affect the stability of the power supply of the power grid.
[0109] Working process:
[0110] (1) Calculate the total regional load based on historical operating data and divide it by 24 hours to get the average load a per hour. Randomly select a starting time j, collect the real-time load aj of the regional transformer, and calculate the coefficient nj = aj / a.
[0111] (2) If nj>1, the real-time load in the area at this moment is higher than the average load. According to the value of nj, the charging price is increased according to a certain calculation method; if nj<1, the real-time load in the area at this moment is lower than the average load. According to the value of nj, the charging price is reduced according to a certain calculation method. The real-time dynamic change of electricity prices guides the EV charging situation in the next time period.
[0112] (3) Calculate the real-time load changes in each time period within 24 hours and how they affect the formulation of electricity price strategies.
[0113] By changing the current charging price, the charging willingness of some users can be increased or decreased, thereby changing the starting charging time of these users, avoiding large-scale charging load access to the power grid during peak hours of residential electricity consumption, which would affect the stability of the power supply of the power grid.
[0114] Every hour, the power grid makes a decision on the charging status for the next time period based on real-time electricity prices, current load, average load, etc., and issues the charging electricity price for the next moment as a guiding signal. In this embodiment, a "moderate" electricity price is selected as the per-unit value Price, and a pricing model considering the load characteristics of electric vehicles is formulated as shown in Table 2:
[0115] Table 2. Pricing Strategy Considering the Load Characteristics of Electric Vehicles
[0116] nj Electricity Prices 0<nj≤0.8 (nj-k2*m)*Price 0.8<nj≤0.9 (nj-k1*m)*Price 0.9<nj≤1 (nj-m)*Price 1<nj≤1.1 (nj+m)*Price 1.1<nj≤1.2 (nj+k1*m)*Price 1.2<nj≤1.3 (nj+k2*m)*Price 1.3<nj≤1.4 (nj+k3*m)*Price 1.4<nj (nj+k4*m)*Price
[0117] In the table, nj is the ratio of the total load of the regional power grid to the average load a at time j; k1 to k4 are constants, where 0 < k1 < k2 < k3 < k4; m is a variable parameter determined according to the local number of EVs and their proportion in the total load. Considering the revenue of the power grid and the acceptance degree of electricity prices by residents, reasonable upper and lower limits of electricity prices are selected.
[0118] 7) Ordered Charging Model under Time-of-Use Electricity Price Considering Users' Charging Willingness
[0119] Take the sum of the unordered charging load and the residential load at the 24th hour of the previous day, and compare it with the average total load to obtain the load multiple n1. Through the load multiple, obtain the electricity price for the next moment (i.e., 1 o'clock), which will guide the load from 1 to 2 o'clock. Then calculate the load from 1 to 2 o'clock, compare it with the average total load to obtain the load multiple n2, and obtain the electricity price for the next moment. And so on, calculate the electricity price and load changes at each moment of the day. The working process is described in words as follows:
[0120] (1) According to the aforementioned real-time electricity price pricing calculation method, obtain the real-time electricity price Pj based on the real-time load at time j. Calculate the charging probability of the first vehicle during this time period.
[0121] (2) Use a random function to determine whether the vehicle charges, and calculate its charging power according to the selected fast charging or conventional charging method, and sequentially calculate the charging power of all EVs within this hour.
[0122] (3) Calculate the EV charging load within this hour, and after superimposing the regional residential load, obtain the real-time load for the next time period.
[0123] (4) Until all calculations are completed within 24 hours.
[0124] The present invention analyzes the influence of electricity price guidance strategy on the characteristics of EV load charging and discharging behavior, takes a certain residential community as the research object, fully considers the charging willingness of residents, studies the disorderly charging model of EV charging in residential areas from the perspective of the power grid, considers the charging demand of EV, takes peak shaving and valley filling of power grid load as the first goal, proposes a time-of-use electricity price pricing rule to achieve orderly charging control in residential areas, establishes a pricing model taking into account EV load characteristics, verifies the regulating ability of electricity price guidance strategy on EV load behavior based on this model, and establishes an orderly charging model taking into account residents' charging willingness.
[0125] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. An orderly charging system guided by an interactive real-time pricing strategy for electric vehicle charging loads, characterized in that The following steps are involved: 1) Construction conditions of regional power grid electric vehicle charging model The user's charging habits will directly affect the time distribution characteristics of the charging load. The three key factors of starting charging time, mileage, and SOC are analyzed, and the Monte Carlo random simulation of the charging behavior of a large number of EVs is used; 2) Build an orderly charging load model The genetic algorithm is used to solve the optimal EV charging time under the guidance of electricity prices. The following are the optimization conditions set in the solution process: (1) To ensure the travel needs of EV owners, the total charging load of EVs in T time periods should remain unchanged, that is, the orderly charging load and the disorderly charging load are equal, as shown in formula (1): In the formula, xoi is the disordered charging load of EV in period i, and the genetic algorithm is used to find x i Optimal solution; xi is the orderly charging load of EV in period i, that is, by controlling the orderly charging of EV, the optimal charging load at time i is solved; In addition, in order to avoid causing a greater load burden on the power grid, resulting in a higher load peak or the phenomenon of "peak-to-peak inversion", it is required that after the optimization, the peak value of the power grid load curve does not exceed the peak value of the original load curve, as shown in formula (2): Qmax≤Q0max (2) In the formula, Q max is the peak value of the power grid load curve after dispatch; Q 0max The peak value of the total load curve of residential load and EV before dispatch; (2) On the basis of reducing the peak-valley difference, the genetic algorithm is used for further optimization to achieve the goal of reducing user charging costs, and the EV charging load is transferred to the time period with relatively low electricity costs. At the same time, it should be considered that if a large number of EVs start charging at the same time at a time point with low electricity costs, a new load peak will be generated, resulting in a "peak-valley inversion" phenomenon. To avoid this phenomenon, during the time period with low charging electricity costs, the start charging time of the EV after scheduling should also take into account the load changes and then determine the number of EVs connected to the grid at a certain time; 3) Prediction of user charging willingness The user's willingness to charge is nonlinear and unpredictable. The fuzzy intelligent algorithm is used to simplify the user's willingness to charge into a linear system, as follows: Household EVs are parked for 95% of the day. The parking time, time-of-use electricity price, and EV SOC are used as the evaluation criteria. These three quantities are used as the input of the fuzzy inference system model, and the input membership function is set. The output is the charging probability of the EV. The joint Gaussian membership function is shown in formula (3): In the formula, x is the time-of-use electricity price, δ is the SOC of EV; c is the parking time; Through the above calculations, we get the EV charging probability values affected by these three factors. The membership function simplifies the nonlinear variable calculation of user charging willingness into several intuitive data, the value of probability f. By changing the values of x, δ and c, the EV charging probability f will tend to these values. 4) Disordered charging model considering user willingness By calculating the charging probability of each vehicle and judging whether the vehicle is charging during this period, a disorderly charging model considering the user's charging willingness is simulated, as follows: (1) First, the SOC, starting charging time, and parking duration of the first EV are simulated, and the charging probability of the vehicle is calculated using a fuzzy algorithm in combination with the current time-of-use electricity price. (2) Use a random function to determine whether the vehicle is charging and calculate its charging power based on the selected fast charging or conventional charging method; (3) Calculate the charging load of all EVs charged during the starting charging time; (4) After calculating the EV charging load within 24 hours, it is superimposed on the residential load curve to obtain the load curve of the regional power grid under the condition of disorderly EV charging; 5) Pricing strategy considering electric vehicle load characteristics Assuming that the electricity price used by the current power grid is a real-time electricity price, the change in the load rate of the community transformer is collected every 1 hour, and the electricity price is adjusted accordingly based on the collected power change information. The total charging cost that users in the area need to pay in one day is shown in formula (4): Where pj is the charging electricity price in period j; Pi is the power of the i-th vehicle; N is the total number of private electric vehicles in the residential area; t is the continuous charging time of EV; x ij There are only two values: 0 or 1. The value 0 or 1 represents whether the i-th studied electric vehicle has charging behavior in the j-th period; 1 represents that the electric car is charging, and 0 represents that the electric car is idle, as shown in formula (5): By changing the current charging price, the charging willingness of some users can be increased or decreased, thereby changing the starting charging time of these users, thus avoiding the large-scale charging load being connected to the power grid during the peak period of residential electricity consumption, which would affect the stability of the power supply of the power grid; 6) Orderly charging model under time-of-use electricity price considering user charging willingness Take the sum of the disordered charging load and the residential load in the 24th hour of the previous day, and compare it with the average total load to get the load multiple n1. The electricity price at the next moment (i.e., 1 o'clock) is obtained through the load multiple. This electricity price will guide the load at points 1-2. Then calculate the load at points 1-2 and compare it with the average total load to get the load multiple n2 to get the electricity price at the next moment. By analogy, the electricity price and load changes at each moment of the day are calculated. By controlling the electricity price, the user's willingness to charge can be changed, that is, the charging probability can be changed. By reducing or increasing the charging probability, the size of the EV charging load at a certain moment can be adjusted as needed; An orderly charging system guided by an interactive real-time pricing strategy for electric vehicle charging load, characterized in that the calculation process of step 2) building an orderly charging load model is as follows: (1) Initialization: Set the evolutionary generation counter t = 0, set the maximum evolutionary generation T, and randomly generate M individuals as the initial population P(0); (2) Individual evaluation: Calculate the fitness of each individual in the population P(t); (3) Selection operation: The selection operator is applied to the population. The purpose of selection is to directly pass on the optimized individuals to the next generation or to generate new individuals through pairing and crossover and then pass on to the next generation. The selection operation is based on the fitness evaluation of individuals in the population. (4) Crossover operation: The crossover operator is applied to the population. The crossover operator plays a core role in the genetic algorithm. (5) Mutation operation: The mutation operator is applied to the population, that is, the gene values at certain loci of the individual strings in the population are changed. After the population P(t) undergoes selection, crossover, and mutation operations, the next generation population P(t+1) is obtained. (6) Termination condition judgment: If t = T, the individual with the maximum fitness obtained in the evolution process is output as the optimal solution and the calculation is terminated.
Citation Information
Patent Citations
Ordered charging method of electric cars connected to power grid
CN105515030A
Electric vehicle group orderly charging method and distribution network power optimization method
CN109103878A
Combined optimization charging and discharging method for electric vehicle group with deepening peak regulation
CN110509788A
Orderly charging subsidy scheduling method for electric vehicles
CN112003314A
