A method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on transportation electrification

Abstract

The present invention discloses a method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification, including: predicting the demand for battery replacement according to the travel and battery replacement behavior of electric taxis in stations through a well-constructed "vehicle-battery-road-station" interconnected system framework based on traffic electrification and battery replacement mode; obtaining the peak energy storage utilization rate and the comprehensive utilization rate of battery replacement stations based on the predicted battery replacement demand; establishing a battery replacement station battery status time series analysis model; and proposing a two-stage day-ahead charging optimization strategy for battery replacement stations that considers smoothing load fluctuations and improving the comprehensive utilization rate of battery replacement stations based on the battery replacement demand and battery status time series analysis model. The present invention predicts the demand for battery replacement of electric taxis and optimizes charging at battery replacement stations based on traffic electrification, predicts the day-ahead battery replacement demand of electric taxis from the perspective of traffic electrification, effectively gives play to the controllable and energy storage characteristics of battery replacement stations, and plays a role in smoothing load fluctuations and shaving peaks and filling valleys.

Description

Technical Field

[0001] The present invention belongs to the field of battery swap station charging, and relates to the battery swap demand of electric taxis, energy storage utilization rate and the formulation of charging strategies for each battery swap station, and specifically to a method for predicting the battery swap demand of electric taxis and optimizing charging at battery swap stations based on transportation electrification. Background Art

[0002] As one of the main ways to quickly replenish electric energy, the battery swap mode has the advantages of fast battery swap speed and strong battery controllability, and can serve vehicles that are in urgent need of electric energy replenishment. From the perspective of unified battery specifications, electric taxis are highly compatible with the battery swap mode. Therefore, it is of great significance to study the battery swap demand prediction and battery swap station optimization strategy of electric taxis under the battery swap mode. Since most of the existing research in the field of battery swapping is still separated from the road-electricity coupling network, the battery swap demand prediction of electric taxis and the energy storage analysis methods of battery swap stations still have defects, such as unclear research objects, loose connection with electrified transportation, and lack of relevant battery swap station energy storage utilization evaluation indicators, making it difficult to achieve ideal results.

[0003] Therefore, a new technical solution is needed to solve this problem. Summary of the invention

[0004] Purpose of the invention: In order to overcome the deficiencies in the prior art, a method for predicting the battery replacement demand of electric taxis and optimizing charging at battery replacement stations based on transportation electrification is provided, which fully considers the impact of transportation, electricity and information and communication systems on regular battery replacement demand and charging load, simulates the day-ahead battery replacement demand of electric taxis from the perspective of electrification of transportation, proposes the concept of peak energy storage utilization, and proposes a two-stage day-ahead optimization strategy that considers smoothing load fluctuations and improving energy storage utilization, which can effectively give play to the controllable and energy storage characteristics of batteries in battery replacement stations, and play a role in smoothing load fluctuations and shaving peaks and filling valleys.

[0005] Technical solution: To achieve the above purpose, the present invention provides a method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification, comprising the following steps:

[0006] S1: Through the established “vehicle-battery-road-station” interconnected system framework based on transportation electrification and battery swapping mode, the demand for battery swapping is predicted according to the travel of electric taxis and the battery swapping behavior in stations;

[0007] S2: Based on the predicted battery swapping demand, the peak energy storage utilization rate is obtained, and the comprehensive utilization rate of the battery swapping station is obtained;

[0008] S3: Establish a battery status timing analysis model for battery swap stations;

[0009] S4: Based on the battery swapping demand and battery status timing analysis model, a two-stage battery swap station day-ahead charging optimization strategy is proposed to consider smoothing load fluctuations and improving the comprehensive utilization rate of battery swap stations.

[0010] Furthermore, the "vehicle-battery-road-station" interconnected system framework based on traffic electrification and battery swapping mode in step S1 includes four parts, namely: interaction between electric taxis and road networks, interaction between electric taxis and battery swapping stations, interaction between electric taxis and batteries, and interaction between batteries and battery swapping stations.

[0011] Furthermore, the method for predicting the demand for battery swapping in step S1 is as follows: based on the travel and battery swapping behavior of electric taxis, the driving path, operating status and battery SOC of electric taxis under multi-information interconnection are simulated, and the number of battery packs provided by the battery swapping station at the beginning of each dispatching period to meet the normal operation of taxis is obtained by counting, which is the battery swapping demand under the "vehicle-battery-road-station" interconnection. Here, based on the arrival information of electric taxis in each period, the number of vehicles that need to be swapped in each period is counted, which is the battery swapping demand.

[0012] The operation status of the electric taxi in step S1 includes carrying passengers, looking for passengers without load, and changing batteries without load. When carrying passengers, the route planning is carried out according to the passenger preference with the goal of minimizing the taxi fare or the shortest time; when changing batteries without load, the route planning is carried out by comprehensively considering the sum of the time to the battery swap station T1, the queuing time T2, and the time from the battery swap station to the next best passenger-seeking point T3 to minimize the sum; when looking for passengers without load, the supply and demand balance relationship between taxi-hailing and passenger-pickup in the region is considered, and a probability selection model for looking for passengers without load is proposed based on the OD matrix, which takes into account the proportion of taxis in each region:

[0013]

[0014] Among them, P(i,j,t) is the probability that an electric taxi chooses area j as the location to find customers at time t, s(j,t) is the proportion of taxis in area j, and P od (i,j,t) is the OD probability from region i to region j at time t, and M is the total number of regions.

[0015] In step S1, the battery replacement behavior of the electric taxi is analyzed. Based on time discretization, the specific analysis method is as follows:

[0016] Assume that there are three adjacent time periods: t-1, t, and t+1. Taking time period t as an example, t (0) and t (1) The starting and ending times of the t period are respectively divided into the following three scenarios for analysis:

[0017] Scenario 1: t (0) ~t (1) can be reached within t (0)~t (1) The battery swap is successfully completed within t+1. However, if there is a queue in the battery swap station at this time, the electric taxi will enter the queue and wait for t+1 (0) Then decide whether to arrange for battery replacement;

[0018] Scenario 2: The trip start time of the electric taxi is at t-1 (0) Previously, the ET that successfully replaced the battery will be at t+1 (0) Start taxi operation after leaving the station;

[0019] Scenario 3: The red straight line represents t (0) ~t (1) Arrived but unable to (0) ~t (1) The vehicles that have completed the battery replacement will enter the queue and (0) Arrange battery swaps and calculate the number of vehicles that successfully complete battery swaps in period t, which is the battery swap demand U(t) of the battery swap station in period t. Here, the battery swap demand U(t) does not refer to the total number of electric taxis with insufficient battery power in and outside the station in period t, but the total number of vehicles that the battery swap station should serve in period t.

[0020] Furthermore, the peak energy storage utilization rate η in step S2 peak (t) is:

[0021]

[0022] Among them, N sdd (t) and N sod (t) is the number of excess reserve batteries in disordered and regulated states, N sd (t) and N so (t) is the number of reserve batteries at the beginning of each period under disorder and regulation conditions, U(t) is the battery swap demand in period t; when the battery swap station is in a disordered state, N sod (t) = N sdd (t), at this time η peak (t) = 0, the utilization rate is 0. N sdd (t) and N sod (t) reflects the remaining energy storage capacity in the battery swap station. This part of the energy storage is not used. In the regulation state, N sod The smaller (t) is, the higher the utilization rate is.

[0023] The comprehensive utilization rate of battery swap stations is:

[0024]

[0025] Among them, t1 is the peak start time period, and t2 is the peak end time period.

[0026] Furthermore, the method for establishing the battery status timing analysis model of the battery swap station in step S3 is:

[0027] The batteries are divided into five categories: batteries to be charged, batteries to be put into use, batteries to be charged, batteries to be produced and batteries to be stored;

[0028] At the beginning of period t, the number of reserve batteries that the battery swap station can provide is:

[0029] N s (t)=max{0,N s (t-1)+y(t-1)-U(t-1)}

[0030] Among them, N s (t) is the reserve battery quantity at the beginning of period t, y(t-1) is the battery quantity produced in period t-1. This part of the battery is charged at the end of period t-1 and can provide battery replacement service at the beginning of period t. U(t-1) is the battery replacement demand in period t-1.

[0031] To simplify the analysis, it is assumed that a certain amount of batteries are regularly put into the control area, and the maximum amount of batteries put into the control area at the beginning of period t is x max (t) The number of charging piles that are not connected to the battery at the end of time period t-1 N r (t-1) and the battery level N in the waiting area at the end of the t-1 period w Impact of (t-1):

[0032] x max (t) = min{N r (t-1),N w (t-1)}

[0033] The number of charging piles that are not connected to the battery at the end of time period t N r (t) The total number of charging piles N all , t period of time in the battery capacity N c The impact of battery y(t) and battery output at the end of period t:

[0034] N r (t)=max{0,N all -N c (t)+y(t)}

[0035] N c (t) = N c (t-1)-y(t-1)+x(t)

[0036] The output y(t) and input x(t) have a certain constraint relationship, which can be expressed as:

[0037] y(t)=y(x(t))

[0038] All replaced batteries need to enter the waiting area, and then wait for regular scheduling to decide whether to put them into the control area for charging:

[0039] N w (t)=max{0,N w (t-1)+U(t)-x(t)}

[0040] This paper believes that the initial reserve batteries should be fully charged before the start of a new day of operation of the battery swap station, and the number is recorded as N B At the beginning of time period t, the batteries in the battery swap station should satisfy the following balance relationship:

[0041] N s (t+1)+N c (t)+N w (t)-y(t)=N B .

[0042] Furthermore, the two-stage day-ahead charging optimization strategy of the battery swap station in step S4 is specifically as follows:

[0043] S4-1: In the first stage, the goal is to smooth out load fluctuations:

[0044]

[0045]

[0046] Among them, P l (t) is the regional load, P c is the charging power of a single charging pile, T is the discrete optimization period, N c (t) is the number of charged batteries in period t, P av To optimize the average value of the load during the cycle;

[0047] S4-2: Based on the optimization in the first stage, the optimization results in the first stage are used as the constraints in the second stage, with the goal of maximizing the comprehensive utilization rate of the battery swap station:

[0048] maxF2=η.

[0049] Furthermore, the constraints in step S4-2 include the following constraints (1) to (4):

[0050] At the beginning of time period t, the number of batteries put into charging x(t) should be greater than or equal to 0, that is,

[0051] x(t)≥0,t=1,...,T (1)

[0052] At the beginning of time period t, the number of batteries put into charging x(t) should not be greater than the number of batteries to be charged N w (t-1) and the number of charging piles not connected to the battery Nr (t-1), that is:

[0053] x(t)≤min{N r (t-1),N w (t-1)} (2)

[0054] Constraint (2) is the upper limit constraint of the input amount x(t), which describes the maximum number of batteries that can be put into the battery swap station at the beginning of time period t;

[0055] At the beginning of period t, the reserve battery capacity N should be guaranteed s (t) is not less than the battery replacement demand U(t), that is:

[0056] N s (t)≥U(t) (3)

[0057] By introducing the relevant conditions, we can get:

[0058]

[0059] The charging optimization of the battery swap station is to reasonably distribute the charging amount to each time period on a time scale. The total charging amount remains unchanged before and after the optimization, that is:

[0060]

[0061] Among them, E t To optimize the charge capacity, E sum is the total charge capacity.

[0062] The "vehicle-battery-road-station" interconnected system framework based on transportation electrification and battery swapping mode is the basis of this invention; it fully considers the impact of transportation, power and information and communication systems on regular battery swapping needs and charging loads, emphasizes the close connection between electric taxis, batteries, road networks and battery swapping stations, and gives full play to the controllable and energy storage characteristics of battery swapping stations.

[0063] The present invention is applicable to the forecast of the battery replacement demand of electric taxis and the charging method at battery replacement stations under the background of traffic electrification, and the main research object is battery replacement taxis.

[0064] The following is a comprehensive description of the technical points involved in the present invention:

[0065] The "vehicle-battery-road-station" interconnected system framework based on transportation electrification and battery swapping mode: describes the interactive relationship between the four main entities of battery swapping electric taxis, batteries, road networks and battery swapping stations, forming a "power consumption-battery swapping-charging" closed-loop system based on information flow-energy flow-traffic flow.

[0066] The probability model of empty-load passenger-seeking considering the proportion of taxis in each region: It not only takes into account the attractiveness of each region to the current location of electric taxis, but also takes into account the balance of supply and demand of taxis in the region, providing a basis for the destination selection of empty-load passenger-seeking.

[0067] Electric taxi battery replacement demand forecasting method: Based on the electrification of transportation, full consideration is given to taxi travel, battery replacement characteristics, and charging arrangements at battery replacement stations, and the number of batteries that the battery replacement stations should provide in each period is calculated.

[0068] Battery status analysis model of battery swap stations: Batteries in battery swap stations have different states. Batteries are divided into five categories: batteries to be charged, input batteries, charging batteries, output batteries and reserve batteries. The mathematical relationship between batteries in different states is described.

[0069] Peak energy storage utilization rate of battery swap stations: Quantitatively analyzes the utilization of energy storage batteries in battery swap stations under a certain control strategy, provides an evaluation indicator for the effectiveness analysis of the control strategy, and proves the feasibility of using rapid electric energy to supplement vehicles as energy storage resources under the battery swap mode.

[0070] The day-ahead charging optimization strategy for battery swap stations: A two-stage optimization method is adopted. The first stage considers minimizing load fluctuations, and the second stage considers maximizing energy storage utilization, providing a charging optimization solution for battery swap stations.

[0071] Beneficial effects: Compared with the prior art, the present invention makes a reasonable prediction of the demand for electric taxi replacement points based on traffic electrification, and its replacement point demand has a "wave-like" change trend, giving full play to the role of rapid power replenishment of electric taxis with strong travel characteristics in peak shaving and valley filling and smoothing load fluctuations, while improving the energy storage utilization rate and giving full play to the buffering capacity of energy storage for battery replacement demand; the proposed energy storage utilization rate provides an evaluation indicator for the battery replacement station regulation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a framework diagram showing the "car-electricity-road-station" interconnection system;

[0073] Figure 2 Display diagram of remaining energy storage capacity in the battery swap station;

[0074] Figure 3 Displays the road topology.

[0075] Figure 4 Display diagram for base load of regional power grid;

[0076] Figure 5 Display diagram of battery swapping demand for each battery swapping station;

[0077] Figure 6 The total load of the region is displayed after superimposing the disordered and ordered charging loads;

[0078] Figure 7 It is a graph showing the number of excess reserve batteries in the orderly state of the first stage;

[0079] Figure 8 It is a graph showing the number of excess reserve batteries in the orderly state of the second stage;

[0080] Fig. 9 It is a graph showing the number of excess reserve batteries in a disordered state;

[0081] Fig.10 Optimize the energy storage utilization display diagram for phase 1;

[0082] Fig.11 Optimized energy storage utilization displayed for phase 2. DETAILED DESCRIPTION

[0083] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0084] The present invention provides a method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification, comprising the following steps:

[0085] S1: Build a "vehicle-battery-road-station" interconnected system framework based on transportation electrification and battery swapping mode:

[0086] The framework of the "vehicle-power-road-station" interconnection system is as follows Figure 1 As shown in the figure, it mainly includes four parts: 1) Interaction between "electric taxi" and "road network": electric taxis without battery swap requests can continuously publish demands on the order platform and accept system dispatch orders, and constantly change their states between carrying passengers and looking for passengers without loads. Vehicles with battery swap requests need to exit the taxi operation to select stations and plan routes. This interaction process is reflected in the energy flow as the continuous attenuation of battery power; 2) Interaction between "electric taxi" and "battery swap station": determine whether to start battery swap or wait for battery swap based on the arrival time of the electric taxi, the queue in the station, and the number of reserve batteries. The vehicle that has completed the battery swap will start taxi operation immediately after leaving the station; 3) Interaction between "electric taxi" and "battery": the electric taxi unloads the battery and loads the fully charged battery stored in the battery swap station. The user's vehicle and battery are separated, and the ownership of the battery charge is returned to the battery swap station; 4) Interaction between "battery" and "battery swap station": orderly charging is carried out in combination with the operation of the power grid on the premise of meeting the subsequent battery swap needs.

[0087] S2: Based on the interactive framework, the demand for battery swapping is predicted according to the travel and battery swapping behaviors of electric taxis:

[0088] Based on the travel and battery-swapping behavior of electric taxis, the driving path, operating status and battery SOC of electric taxis under multi-information interconnection are simulated, and the number of battery packs provided by the battery swap station at the beginning of each scheduling period to meet the normal operation of taxis is calculated, which is the battery swapping demand under the "vehicle-battery-road-station" interconnection.

[0089] Electric taxis have continuous driving behavior. In the daily operation process, there are three states: carrying passengers, looking for passengers without load, and swapping batteries without load. When carrying passengers, the route planning is carried out according to the passenger preference with the goal of minimizing the taxi fare or the shortest time; when swapping batteries without load, the route planning is carried out by comprehensively considering the sum of the time to the battery swap station T1, the queuing time T2, and the time from the battery swap station to the next best passenger search point T3; when looking for passengers without load, the supply and demand balance relationship between taxi hailing and passenger pickup in the region is considered, and a probability selection model for looking for passengers without load is proposed based on the OD matrix, which takes into account the proportion of taxis in each region:

[0090]

[0091] Among them, P(i,j,t) is the probability that an electric taxi chooses area j as the location to find customers at time t, s(j,t) is the proportion of taxis in area j, and P od (i,j,t) is the OD probability from region i to region j at time t, and M is the total number of regions.

[0092] Based on the time discretization, the analysis method of the battery swapping behavior of electric taxis in the battery swapping station is as follows: Assuming that there are three adjacent time periods t-1, t, and t+1, taking time period t as an example, t (0) and t (1) are the start and end time of period t respectively. It is divided into the following three scenarios:

[0093] Scenario 1: t (0) ~t (1) can be reached within t (0) ~t (1) The battery swap is successfully completed within t+1. However, if there is a queue in the battery swap station at this time, the electric taxi will enter the queue and wait for t+1 (0) Then decide whether to arrange for battery replacement;

[0094] Scenario 2: The trip start time of the electric taxi is at t-1 (0) Previously, the ET that successfully replaced the battery will be at t+1 (0) Start taxi operation after leaving the station;

[0095] Scene 3 The red straight line represents t (0) ~t (1) Arrived but unable to (0) ~t (1)The vehicles that have completed the battery replacement will enter the queue and (0) Arrange battery swapping. Calculate the number of vehicles that successfully complete battery swapping in period t, which is the battery swapping demand U(t) of the battery swapping station in period t. Here, the battery swapping demand U(t) does not refer to the total number of electric taxis with insufficient battery power in and outside the station in period t, but the total number of vehicles that the battery swapping station should serve in period t.

[0096] S3: Based on the interactive framework, the concept of peak energy storage utilization is proposed:

[0097] Peak energy storage utilization factor η peak (t) is:

[0098]

[0099] Among them, N sdd (t) and N sod (t) is the number of excess reserve batteries in disordered and regulated states, N sd (t) and N so (t) is the number of reserve batteries at the beginning of each period under disorder and regulation conditions, and U(t) is the battery swap demand in period t. When the battery swap station is in a disordered state, N sod (t) = N sdd (t), at this time η peak (t) = 0, the utilization rate is 0. N sdd (t) and N sod (t) reflects the remaining energy storage capacity in the battery swap station, such as Figure 2 As shown, this part of the energy storage is not used. In the regulation state, N sod The smaller (t) is, the higher the utilization rate is.

[0100] N sdd (t) represents the total controllable energy storage capacity in the disordered state, N sod (t) represents the total controllable energy storage capacity under a certain control strategy; N sdd (t)-N sod (t) represents the energy storage capacity that has been called under a certain regulation strategy. It can be found that under a specific regulation state, energy storage still has the ability to be further developed and can provide other auxiliary services for the power grid.

[0101] The comprehensive utilization rate of battery swap stations is:

[0102]

[0103] Among them, t1 is the peak start time period, and t2 is the peak end time period.

[0104] S4: Based on the interactive framework, establish a battery status timing analysis model for battery swap stations:

[0105] The battery status timing analysis model of the battery swap station divides batteries into five categories: batteries to be charged, batteries in use, batteries being charged, batteries in production, and batteries in reserve.

[0106] At the beginning of period t, the number of reserve batteries that the battery swap station can provide is:

[0107] N s (t)=max{0,N s (t-1)+y(t-1)-U(t-1)}

[0108] Among them, N s (t) is the reserve battery quantity at the beginning of period t, y(t-1) is the battery quantity produced in period t-1. This part of the battery finishes charging in period t-1 and can provide battery replacement service at the beginning of period t. U(t-1) is the battery replacement demand in period t-1.

[0109] To simplify the analysis, it is assumed that a certain amount of batteries are regularly put into the control area, and the maximum amount of batteries put into the control area at the beginning of period t is x max (t) The number of charging piles that are not connected to the battery at the end of time period t-1 N r (t-1) and the battery level N in the waiting area at the end of the t-1 period w Impact of (t-1):

[0110] x max (t) = min{N r (t-1),N w (t-1)}

[0111] The number of charging piles that are not connected to the battery at the end of time period t N r (t) The total number of charging piles N all , t period of time in the battery capacity N c The impact of battery y(t) and battery output at the end of period t:

[0112] N r (t)=max{0,N all -N c (t)+y(t)}

[0113] N c (t) = N c (t-1)-y(t-1)+x(t)

[0114] The output y(t) and input x(t) have a certain constraint relationship, which can be expressed as:

[0115] y(t)=y(x(t))

[0116] All replaced batteries need to enter the waiting area, and then wait for regular scheduling to decide whether to put them into the control area for charging:

[0117] N w (t)=max{0,N w (t-1)+U(t)-x(t)}

[0118] This paper believes that the initial reserve batteries should be fully charged before the start of a new day of operation of the battery swap station, and the number is recorded as N B At the beginning of time period t, the batteries in the battery swap station should satisfy the following balance relationship:

[0119] N s (t+1)+N c (t)+N w (t)-y(t)=N B

[0120] S5: Based on the battery swap demand and battery status timing analysis model, a two-stage battery swap station day-ahead charging optimization strategy is proposed that takes into account load fluctuations and improving energy storage utilization:

[0121] 1) In the first stage, the goal is to smooth out load fluctuations:

[0122]

[0123]

[0124] Among them, P l (t) is the regional load, P c is the charging power of a single charging pile, T is the discrete optimization period, N c (t) is the number of charged batteries in period t, P av is the average value of the load during the optimization period.

[0125] 2) Based on the optimization in the first stage, the optimization results in the first stage are used as the constraints in the second stage, with the maximum peak utilization as the goal:

[0126] maxF2=η

[0127] The relevant constraints are:

[0128] 1) At the beginning of time period t, the number of batteries put into charging x(t) should be greater than or equal to 0, that is,

[0129] x(t)≥0,t=1,...,T

[0130] 2) At the beginning of time period t, the number of batteries put into charging x(t) should not be greater than the number of batteries to be charged N w (t-1) and the number of charging piles not connected to the battery N r (t-1), that is:

[0131] x(t)≤min{N r (t-1),N w (t-1)}

[0132] Constraint 2) is the upper limit constraint of the input amount x(t), which describes the maximum number of batteries to be charged that can be put into the battery swap station at the beginning of time period t.

[0133] 3) At the beginning of period t, the reserve battery capacity N should be guaranteed s (t) is not less than the battery replacement demand U(t), that is:

[0134] N s (t)≥U(t)

[0135] By introducing the relevant conditions, we can get:

[0136]

[0137] 4) Charging optimization of battery swap stations is to reasonably distribute the charging amount to each time period on a time scale, and the total charging amount remains unchanged before and after optimization, that is:

[0138]

[0139] Among them, E t To optimize the charge capacity, E sum is the total charge capacity.

[0140] Based on the above electric vehicle charging guidance and pricing method, a simulation analysis is performed in this embodiment, as follows:

[0141] The network topology of this embodiment is as follows Figure 3 As shown in the figure, the road length, carrying capacity and functional area zoning are based on the reference

[13] . A total of 5 battery swap stations are set up, 1# to 5# are located at the 6th, 7th, 13th, 17th and 26th nodes of the road network respectively, and each station is equipped with 10 battery swap devices. Assuming that the number of battery swap electric taxis in the region is 1,000, the unit energy consumption is 0.23 kwh / km, according to the reference

[19] , the ratio of reserve batteries to vehicle battery swap demand is 0.6:1, and the number of reserve batteries should be 600. According to the reference

[11] , the ratio of reserve batteries to charging piles is 3:2, and the total number of charging equipment is set to about 400. The reserve batteries are 150, 150, 100, 50, and 150 respectively, and the number of charging equipment is 100, 100, 70, 35, and 100 respectively. Taxis are gradually introduced into the interconnection system from 6:00-9:00 and gradually withdraw from the interconnection system from 21:00-24:00. The initial SOC is evenly distributed between 0.5 and 0.8. The basic load curve of the regional power grid in this embodiment is shown in the attached figure. Figure 4 shown.

[0142] 1. Analysis of battery replacement demand

[0143] The demand for battery replacement at each station is as follows Figure 5 As shown in the figure, in the spatial dimension, the size and trend of battery swapping demand are different, indicating that the geographical distribution of battery swapping stations will also have a certain impact on the battery swapping demand. In the temporal dimension, whether it is the total regional battery swapping demand or the battery swapping demand at a single battery swapping station, the overall trend is roughly "wave-shaped" and has a cyclical characteristic, which is mainly affected by the continuous "operation-battery swapping" process of electric taxis.

[0144] At the same time, for a single battery swap station, the peak value of its battery swap demand also varies in the time dimension. The most significant are 3# and 5# battery swap stations. The peak value of 3# battery swap station has the characteristics of "large on both sides and small in the middle". The peak value of 5# battery swap station is the highest from 15:00 to 18:00, which coincides with the commercial area gradually entering the peak business period. Electric taxi drivers are more likely to operate in the commercial area and swap batteries nearby.

[0145] 2. Analysis of orderly and disorderly charging at battery swap stations

[0146] In this embodiment, the "charge and replace" charging mode is considered as disordered charging. Figure 6 The total regional load after superimposing disordered and ordered charging loads is shown in the figure. The peak-to-valley differences after disordered charging, first-stage optimization, and second-stage optimization are 50.7243 MW, 41.3726 MW, and 41.2126 MW, respectively. The objective function values ​​F for smoothing load fluctuations are 29.340×10 3 , 24.042×10 3 and 24.039×10 3 After the first and second stage optimization, the peak-to-valley difference decreased by 9.3517MW and 9.5117MW, respectively, which was 18.44% and 18.75% better than that under disorder. After the two-stage optimization, F improved by 18% compared with disorder. Orderly charging load can give full play to the buffering effect of reserve batteries, reduce the amount of battery input during peak hours, reduce the load in the two peak periods of 9:00-12:00 and 17:00-21:00, and the peak reduction from 9:00 to 12:00 is obvious. The maximum peak reduction of the first and second stage optimization is 6.4MW and 6.44MW compared with disordered charging. At the same time, the battery input is increased during the low period of the day, and the load during the period of 12:00-16:00 is improved. After the electric taxis cease operation at 24:00, there will be no more demand for battery replacement at the battery swap station. At this time, the reserve batteries will be maintained at a lower level, while the batteries in the charging area will be maintained at a higher level. These batteries can be arranged to be charged during the low hour period from 24:00 to 6:00 the next morning, thereby achieving the effect of peak shaving and valley filling and smoothing load fluctuations.

[0147] 3. Analysis of energy storage utilization during peak hours

[0148] In the first and second stage optimization and disordered state, the change curve of the number of excess reserve batteries (number of reserve batteries - battery replacement demand) is as follows Figure 7 —As shown in 9.

[0149] During the two peak load periods of 9:00-12:00 and 17:00-21:00, the excess reserve batteries in the orderly state are reduced, indicating that the reserve batteries have a buffering effect during these two periods, absorbing part of the battery replacement demand while ensuring that the battery charging is not too much. Regardless of the orderly or disordered conditions, the number of excess reserve batteries at the 5# battery swap station is less than 0 from 17:00 to 18:00, indicating that the reserve batteries of the 5# battery swap station cannot meet the battery replacement demand at this time, affecting the normal operation of the battery swap station. Electric taxis need to wait for extra time to replace batteries. To eliminate this situation, the number of reserve batteries or charging piles needs to be increased. In the orderly state, the number of excess reserve batteries after about 20:00 is 0, indicating that the reserve batteries at the beginning of each period are exactly equal to the battery replacement demand in that period. The speed at which the control warehouse produces fully charged batteries can effectively follow the battery replacement demand, ensuring that more batteries to be charged can be charged during the valley filling period. The control ability of the battery swap station is affected by the total number of batteries in the battery swap station. The larger the total number of batteries, the stronger its control ability and the better the effect of peak shaving and valley filling.

[0150] During the period from 10:00 to 21:00, the utilization rate after optimization in the first and second stages is as follows: Fig.10 —As shown in 11.

[0151] During the period from 10:00 to 21:00, the average utilization rates of the battery swap stations after the first phase of optimization were 44.39%, 29.41%, 45.41%, 30.48%, and 41.95%, and the average utilization rates of the battery swap stations after the second phase of optimization were 49.04%, 33.21%, 53.55%, 30.48%, and 41.95%. The second phase of optimization improved the utilization rates of the 1# to 3# battery swap stations by 10.48%, 12.92%, and 17.92%, respectively. The utilization rates of the 1# to 3# battery swap stations in the first and second phases are as follows: Fig.10 and Fig.11As shown. The 4# and 5# battery swap stations are mainly affected by the demand for battery swapping and the total amount of reserve batteries. The energy storage utilization rate has not been improved without affecting the load leveling and peak-to-valley difference in the first stage. The reserve batteries of the battery swap stations after the two-stage optimization have played a certain energy storage role and reduced the load peak. It can be found that after 19:00, the utilization rate of the battery swap station is maintained at a high level, and even the utilization rate is 100%. One charging cycle in advance, according to the minimum amount to meet the demand for battery swapping, these batteries are put into the control area for charging, so that the number of batteries in the waiting area is maintained at a high level. The batteries in the waiting area wait for the low load of the day to be put into charging, which plays a role in increasing the valley load.

[0152] In summary, the electric taxi battery replacement demand prediction and battery replacement station charging optimization method based on transportation electrification proposed in the present invention combines the spatiotemporal and behavioral characteristics of the transportation network, user travel characteristics, etc. to predict the battery replacement demand of electric taxis on the day before, revealing the characteristics of its "wave-like" change trend; analyze the state changes of the battery in the battery replacement station and the changes in vehicle behavior, arrange the charging plan of the battery in the battery replacement station with the goal of smoothing load fluctuations, give full play to the energy storage characteristics of the battery replacement station and the controllability of the battery after the separation of the vehicle and the battery, and as one of the fast energy supplement methods, effectively solve the uncontrollability of fast charging on a time scale. The proposed method plays a role in smoothing load fluctuations, shaving peaks and filling valleys, and has a certain energy storage utilization effect.

Claims

1. A method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification, characterized in that: The steps include: S1: Through the established "vehicle-battery-road-station" interconnected system framework based on transportation electrification and battery swapping mode, the demand for battery swapping is predicted according to the travel of electric taxis and the battery swapping behavior in stations; S2: Based on the predicted battery swapping demand, the peak energy storage utilization rate is obtained, and the comprehensive utilization rate of the battery swapping station is obtained; S3: Establish a battery status timing analysis model for battery swap stations; S4: Based on the battery swap demand and battery status timing analysis model, a two-stage battery swap station day-ahead charging optimization strategy is proposed that takes into account the smoothing of load fluctuations and the improvement of the comprehensive utilization rate of battery swap stations; The method for predicting the battery swap demand in step S1 is as follows: based on the travel and battery swap behavior of electric taxis, the driving path, operating status and battery SOC of electric taxis under multi-information interconnection are simulated, and based on the arrival information of electric taxis, the number of battery packs provided by the battery swap station at the beginning of each dispatching period to meet the normal operation of taxis is obtained by statistics, that is, the battery swap demand under the "vehicle-battery-road-station" interconnection; The peak energy storage utilization rate η in step S2 peak (t) is: Among them, N sdd (t) and N sod (t) is the number of excess reserve batteries in disordered and regulated states, N sd (t) and N so (t) is the number of reserve batteries at the beginning of each period under disorder and regulation conditions, and U(t) is the battery replacement demand in period t; The comprehensive utilization rate of battery swap stations is: Among them, t1 is the peak start time, and t2 is the peak end time; The method for establishing the battery status timing analysis model of the battery swap station in step S3 is: The batteries are divided into five categories: batteries to be charged, batteries to be put into use, batteries to be charged, batteries to be produced and batteries to be stored; At the beginning of period t, the number of reserve batteries that the battery swap station can provide is: N s (t)=max{0,N s (t-1)+y(t-1)-U(t-1)} Among them, N s (t) is the reserve battery quantity at the beginning of period t, y(t-1) is the battery quantity produced in period t-1. This part of the battery is charged at the end of period t-1 and can provide battery replacement service at the beginning of period t. U(t-1) is the battery replacement demand in period t-1. Assuming that batteries are regularly added to the control area, the maximum amount of batteries added at the beginning of period t is x max (t) The number of charging piles that are not connected to the battery at the end of time period t-1 N r (t-1) and the battery level N in the waiting area at the end of the t-1 period w Impact of (t-1): x max (t)=min{N r (t-1),N w (t-1)} The number of charging piles that are not connected to the battery at the end of time period t N r (t) The total number of charging piles N all , t period of time in the battery capacity N c The impact of battery y(t) and battery output at the end of period t: N r (t)=max{0,N all -N c (t)+y(t)} N c (t)=N c (t-1)-y(t-1)+x(t) The output y(t) and input x(t) have a certain constraint relationship, which can be expressed as: y(t)=y(x(t)) All replaced batteries need to enter the waiting area, and then wait for regular scheduling to decide whether to put them into the control area for charging: N w (t)=max{0,N w (t-1)+U(t)-x(t)} Before the new day of operation of the battery swap station begins, the initial reserve batteries should be fully charged, and the number is recorded as N. B At the beginning of time period t, the batteries in the battery swap station should satisfy the following balance relationship: N s (t+1)+N c (t)+N w (t)-y(t)=N B 。 2. The method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification according to claim 1 is characterized in that: The "vehicle-battery-road-station" interconnected system framework based on traffic electrification and battery swapping mode in step S1 includes four parts, namely: the interaction between electric taxis and road networks, the interaction between electric taxis and battery swapping stations, the interaction between electric taxis and batteries, and the interaction between batteries and battery swapping stations.

3. The method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification according to claim 1 is characterized in that: The operation states of the electric taxi in step S1 include carrying passengers, looking for passengers without load, and changing batteries without load. When carrying passengers, the route planning is carried out according to the passenger preference with the goal of minimizing the taxi fare or the shortest time; when changing batteries without load, the route planning is carried out by comprehensively considering the sum of the time to the battery swap station T1, the queuing time T2, and the time from the battery swap station to the next best passenger-seeking point T3 to minimize the sum; when looking for passengers without load, the supply and demand balance relationship between taxi-hailing and passenger-pickup in the region is considered, and a probability selection model for looking for passengers without load is proposed based on the OD matrix, which takes into account the proportion of taxis in each region: Among them, P(i,j,t) is the probability that an electric taxi chooses area j as the location to find customers at time t, s(j,t) is the proportion of taxis in area j, and P od (i,j,t) is the OD probability from region i to region j at time t, and M is the total number of regions.

4. The method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification according to claim 1 is characterized in that: In step S1, the battery replacement behavior of the electric taxi is analyzed. Based on time discretization, the specific analysis method is as follows: Assume that there are three adjacent time periods: t-1, t, and t+1. Taking time period t as an example, t (0) and t (1) The starting and ending times of the t period are respectively divided into the following three scenarios for analysis: Scenario 1: t (0) ~t (1) can be reached within t (0) ~t (1) The battery swap is successfully completed within t+1. However, if there is a queue in the battery swap station at this time, the electric taxi will enter the queue and wait for t+1 (0) Then decide whether to arrange for battery replacement; Scenario 2: The trip start time of the electric taxi is at t-1 (0) Previously, the ET that successfully replaced the battery will be at t+1 (0) Start taxi operation after leaving the station; Scenario 3: t (0) ~t (1) Arrived but unable to (0) ~t (1) The vehicles that have completed the battery replacement will enter the queue and (0) Arrange battery swapping and calculate the number of vehicles that successfully complete battery swapping in period t, which is the battery swapping demand U(t) of the battery swapping station in period t.

5. The method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification according to claim 1 is characterized in that: The two-stage day-ahead charging optimization strategy in step S4 is specifically as follows: S4-1: In the first stage, the goal is to smooth out load fluctuations: Among them, P l (t) is the regional load, P c is the charging power of a single charging pile, T is the discrete optimization period, N c (t) is the number of charged batteries in period t, P av To optimize the average value of the load during the cycle; S4-2: Based on the optimization in the first stage, the optimization results in the first stage are used as the constraints in the second stage, with the goal of maximizing the comprehensive utilization rate of the battery swap station: max F2 = η.

6. The method for predicting the demand for battery replacement of electric taxis and optimizing charging at battery replacement stations based on traffic electrification according to claim 5 is characterized in that: The constraints in step S4-2 include the following constraints (1) to (4): At the beginning of time period t, the number of batteries put into charging x(t) should be greater than or equal to 0, that is, x(t)≥0,t=1,...,T(1) At the beginning of time period t, the number of batteries put into charging x(t) should not be greater than the number of batteries to be charged N w (t-1) and the number of charging piles not connected to the battery N r (t-1), that is: x(t)≤min{N r (t-1),N w (t-1)}(2) Constraint (2) is the upper limit constraint of the input amount x(t), which describes the maximum number of batteries that can be put into the battery swap station at the beginning of time period t; At the beginning of period t, the reserve battery capacity N should be guaranteed s (t) is not less than the battery replacement demand U(t), that is: N s (t)≥U(t)(3) By introducing the relevant conditions, we can get: The charging optimization of the battery swap station is to reasonably distribute the charging amount to each time period on a time scale. The total charging amount remains unchanged before and after the optimization, that is: Among them, E t To optimize the charge capacity, E sum is the total charge capacity.

Images