Charging method and device based on time-of-use electricity price, equipment, medium and product

Abstract

The invention discloses a charging method, device and equipment based on time-of-use electricity price, a medium and a product, and relates to the field of coordinated charging scheduling. The method comprises the following steps: acquiring information data; the information data comprises charging information data of the target electric vehicle in the schedulable unit; the schedulable unit is extracted based on the charging activity chain; the charging activity chain is an activity sequence which is obtained by sorting operation data of all electric vehicles in the target area according to timestamps and is used for representing starting and ending of a driving event; determining a charging scheme based on the mixed integer linear programming model; the mixed integer linear programming model is a mathematical model obtained by adopting a charging control simulation method, performing charging mode identification based on a charging activity chain and performing statistical analysis on charging characteristic data under each charging mode after identification. The invention aims to realize flexible scheduling of charging so as to transfer the charging peak load and realize peak clipping and valley filling.

Description

Technical Field

[0001] The present application relates to the field of coordinated charging scheduling, and in particular to a charging method, device, equipment, medium and product based on time-of-use electricity prices. Background Art

[0002] Driven by the coordinated development of transportation electrification and decarbonized power generation, the large-scale adoption of electric vehicles provides a promising solution to the problems of fossil fuel depletion and greenhouse gas emissions. As an important demand-side flexible resource, electric vehicles can alleviate peak loads through smart charging, achieve efficient integration of renewable energy, and achieve vehicle-grid integration (EGI). Existing smart charging schemes have shown great potential in reshaping the power load profile by shifting electric vehicle charging to off-peak hours or adopting low-rate charging during peak hours.

[0003] Taking full advantage of the flexibility of EV charging to achieve efficient EGI depends not only on the development of enabling control algorithms, but also on accurate analysis of the travel activities of EV users. Therefore, it is necessary to fully understand the inherent flexibility of EV charging behavior and understand how to effectively influence these behaviors to maximize EGI benefits. In summary, how to achieve flexible charging scheduling to shift charging peak loads and achieve peak shaving and valley filling is crucial. Summary of the invention

[0004] The purpose of this application is to provide a charging method, device, equipment, medium and product based on time-of-use electricity prices, which can realize flexible scheduling of charging to shift charging peak loads and achieve peak shaving and valley filling.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a charging method based on time-of-use electricity prices, comprising:

[0007] Acquire information data; the information data includes charging information data of the target electric vehicle in the dispatchable unit; the dispatchable unit is extracted based on the charging activity chain; the dispatchable unit includes: a charging stage and an idle parking stage; the charging activity chain is an activity sequence that is obtained by sorting the operating data of all electric vehicles in the target area according to timestamps and is used to characterize the start and end of a driving event;

[0008] According to the information data, a charging plan is determined based on a mixed integer linear programming model; the charging plan is determined based on time-of-use electricity prices to minimize the charging coefficient of each charging period; the charging plan is used to transfer the charging peak load to achieve peak shaving and valley filling; the mixed integer linear programming model adopts a charging control simulation method, identifies the charging mode based on the charging activity chain, and statistically analyzes the charging characteristic data under each identified charging mode to obtain a mathematical model; the charging characteristic data includes: charging power, charging time and battery SOC.

[0009] In a second aspect, the present application provides a charging device based on time-of-use electricity prices, comprising:

[0010] An information data acquisition module is used to acquire information data; the information data includes charging information data of the target electric vehicle in the dispatchable unit; the dispatchable unit is extracted based on the charging activity chain; the dispatchable unit includes: a charging stage and an idle parking stage; the charging activity chain is an activity sequence that is obtained by sorting the operating data of all electric vehicles in the target area according to timestamps and is used to characterize the start and end of a driving event;

[0011] A charging scheme determination module is used to determine a charging scheme based on the information data and a mixed integer linear programming model; the charging scheme is determined based on time-of-use electricity prices to minimize the charging coefficient of each charging period; the charging scheme is used to transfer the charging peak load to achieve peak shaving and valley filling; the mixed integer linear programming model is a mathematical model obtained by using a charging control simulation method, identifying the charging mode based on the charging activity chain, and statistically analyzing the charging characteristic data under each identified charging mode; the charging characteristic data includes: charging power, charging time and battery SOC.

[0012] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned charging method based on time-of-use electricity prices.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned charging method based on time-of-use electricity prices.

[0014] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned charging method based on time-of-use electricity prices.

[0015] According to the specific embodiments provided in this application, this application has the following technical effects:

[0016] The present application provides a charging method, device, equipment, medium and product based on time-of-use electricity price. By acquiring information data, a charging scheme is determined based on the time-of-use electricity price to minimize the charging coefficient of each charging period; wherein the information data includes the charging information data of the target electric vehicle in the dispatchable unit; the dispatchable unit is extracted based on the charging activity chain; the charging activity chain is an activity sequence for characterizing the start and end of the driving event based on the operation data of all electric vehicles in the target area, sorted according to the timestamp; the mixed integer linear programming model is a mathematical model obtained by using a charging control simulation method, identifying the charging mode based on the charging activity chain, and statistically analyzing the charging characteristic data under each charging mode after identification. The present application evaluates the flexibility of electric vehicles under different charging control strategies through rule-based charging mode recognition, and extracts the key categories of the electric vehicle charging activity chain characterized by the sequence of parking and charging activities between adjacent trips from the actual electric vehicle operation data. By switching the charging mode for simulation, the time-of-use electricity price intelligent charging is designed, thereby realizing flexible charging scheduling to transfer the charging peak load and realize peak shaving and valley filling. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 is a flow chart of a charging method based on time-of-use electricity prices;

[0019] Figure 2 Schematic diagram of the constructed electric vehicle charging activity chain;

[0020] Figure 3 A schematic diagram of the share of identified electric vehicle charging modes;

[0021] Figure 4 A schematic diagram showing the proportion of different charging modes among individual electric vehicle users;

[0022] Figure 5 is a schematic diagram of charging power distribution; Figure 5 (a) is a schematic diagram of charging power distribution in mode A; Figure 5 (b) is a schematic diagram of charging power distribution in mode B; Figure 5 (c) is a schematic diagram of charging power distribution in mode C; Figure 5 (d) is a schematic diagram of charging power distribution in mode D;

[0023] Figure 6 Schematic diagram of the distribution of different charging power ranges for four main charging modes;

[0024] Figure 7 It is a schematic diagram of charging power level analysis; Figure 7 (a) is the charging power distribution diagram using GMM; Figure 7 (b) is a graph showing the percentage of slow and fast charging times in each mode;

[0025] Figure 8 is a schematic diagram of charging duration distribution; where: Figure 8 (a) is the duration distribution diagram under slow charging; Figure 8 (b) is the duration distribution diagram under fast charging; Figure 8 (c) is the parking duration distribution diagram during slow charging; Figure 8 (d) is the parking duration distribution diagram during fast charging;

[0026] Fig. 9 is a schematic diagram of the time distribution of charging activities; Fig. 9 (a) is the time distribution diagram of slow charging activity; Fig. 9 (b) in the figure is the time distribution diagram of the fast charging activity;

[0027] Fig.10 is a schematic diagram of slow time distribution; Fig.10 (a) is the slow charging time distribution diagram of mode A; Fig.10 (b) is the slow charging time distribution diagram of mode B; Fig.10 (c) is the slow charging time distribution diagram of mode B; Fig.10 (d) is the slow charging time distribution diagram of mode B;

[0028] Fig.11 It is a schematic diagram of fast time distribution; Fig.11 (a) is the fast charging time distribution diagram of mode A; Fig.11 (b) is the fast charging time distribution diagram of mode B; Fig.11 (c) is the fast charging time distribution diagram of mode B; Fig.11 (d) is the fast charging time distribution diagram of mode B;

[0029] Fig.12 Schematic diagram of battery SOC distribution at the beginning and end of charging activity; Fig.12 (a) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode A under slow charging; Fig.12(b) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode B under slow charging; Fig.12 (c) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode C under slow charging; Fig.12 (d) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode D under slow charging; Fig.12 (e) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode A under fast charging; Fig.12 (f) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode B under fast charging; Fig.12 (g) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode C under fast charging; Fig.12 (h) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode D under fast charging;

[0030] Fig.13 It is a schematic diagram of the average daily charging load curve; Fig.13 (a) is the charging load curve of all electric vehicles; Fig.13 (b) is a charging load curve corresponding to slow charging and fast charging; Fig.13 (c) is the charging load curve corresponding to mode A, mode B, mode C and mode D under slow charging; Fig.13 (d) is a charging load curve diagram corresponding to mode A, mode B, mode C and mode D under fast charging;

[0031] Fig.14 It is a schematic diagram of power load distribution; Fig.14 (a) is a schematic diagram of the change of charging load of electric vehicles in the network during peak and valley periods before and after coordinated charging control in the MinFee-S scenario; Fig.14 (b) is a schematic diagram of the change of charging load of electric vehicles in the network during peak and valley periods before and after coordinated charging control in the MinFee-F scenario; Fig.14 (c) is a schematic diagram of the total load distribution of slow charging; Fig.14 (d) is a schematic diagram of the total load distribution of fast charging. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0033] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0034] In an exemplary embodiment, Figure 1 As shown, a charging method based on time-of-use electricity price is provided. The charging method based on time-of-use electricity price includes:

[0035] Step 100: Acquire information data. The information data includes charging information data of the target electric vehicle in the dispatchable unit; the dispatchable unit is extracted based on the charging activity chain; the dispatchable unit includes: a charging stage and an idle parking stage; the charging activity chain is based on the operation data of all electric vehicles in the target area, and is an activity sequence obtained by sorting according to timestamps to characterize the start and end of the driving event.

[0036] Step 200: Determine the charging plan based on the information data and the mixed integer linear programming model. The charging plan is determined based on the time-of-use electricity price to minimize the charging coefficient of each charging period; the charging plan is used to transfer the charging peak load to achieve peak shaving and valley filling; the mixed integer linear programming model is a mathematical model obtained by using a charging control simulation method, identifying the charging mode based on the charging activity chain, and statistically analyzing the charging characteristic data under each charging mode after identification; the charging characteristic data includes: charging power, charging time and battery SOC.

[0037] The mixed integer linear programming model includes: an objective function and constraints; the constraints include: a charging demand constraint and a time constraint; wherein the time constraint is determined based on the start time and the end time of each of the dispatchable units.

[0038] The objective functions include:

[0039]

[0040] Where F is the objective function; T i，end is the end time of the ith schedulable unit; T i,start is the start time of the ith schedulable unit; P i，avg is the average charging power in the i-th dispatchable unit; x(t) is the binary variable at the t-th moment; C(t) is the electricity price at the t-th moment.

[0041] The charging demand constraints include:

[0042]

[0043] Among them, T i，end is the end time of the ith schedulable unit; T i，startis the start time of the ith schedulable unit; P i (t) is the charging power at the tth time in the i-th dispatchable unit; E i is the total energy billed in the i-th dispatchable unit.

[0044] In one embodiment, according to the information data, based on a mixed integer linear programming model, a charging scheme is determined, specifically including:

[0045] According to the information data, the objective function is solved based on the constraint conditions to obtain the charging plan; the information data includes: the battery capacity of the target electric vehicle, the starting SOC, the ending SOC, the charging start time and the charging end time.

[0046] As an optional implementation, the charging method based on time-of-use electricity price also includes:

[0047] According to the charging plan, the setting cycle and pricing mechanism of time-of-use electricity prices are adjusted to encourage off-peak charging.

[0048] This application evaluates the flexibility of electric vehicles under different charging control strategies based on rule-based charging mode recognition, and extracts key categories of electric vehicle charging activity chains characterized by the sequence of parking and charging activities between adjacent trips from actual electric vehicle operation data. By switching charging modes for simulation, time-of-use electricity price (ToU) smart charging is designed. Among them, the charging mode recognition method can serve as the basis for other subsequent studies, and the designed time-of-use electricity price smart charging method can shift the charging peak load to a certain extent, achieving the effect of peak shaving and valley filling.

[0049] The design concept corresponding to the method provided in this application is divided into the following steps.

[0050] 1. Build a charging activity chain.

[0051] The dataset used in this application contains real electric vehicle operation data of 45,093 light electric vehicles operating in a certain city in September 2021, including 326,998 charging times. The original data was collected with a time resolution of 30 seconds in accordance with the Chinese national standard GB / T32960, and further processed into data segments including parking, charging and driving periods. The information provided mainly includes vehicle model, battery capacity, timestamp, vehicle status, battery state of charge (SOC), cumulative mileage, etc.

[0052] In order to reduce the impact of data anomalies on further analysis, the collected raw data were decoded, abnormal sessions were eliminated, erroneous parameters were corrected, and missing session data were supplemented.

[0053] First, data cleaning is performed to remove charging sessions with missing information, unreasonable energy or power values, or duration less than 1 minute.

[0054] In addition, charging periods longer than 24 hours were eliminated to ensure data integrity for accurate analysis. The data for each EV was then organized chronologically and timestamps were checked. Adjacent periods were then identified and added to the dataset based on time and SOC differences between them.

[0055] Finally, the processed session data is arranged in chronological order, and sessions showing the same vehicle status are merged into one. This data cleaning and reconstruction process is repeated until all sessions are sorted with consecutive timestamps and the status of two adjacent segments is obviously different. The processed session data is then used to construct the electric vehicle chain.

[0056] In this application, the charging activity chain of an electric vehicle user is defined as a sequence of activities that starts and ends with a driving event, with no intervening driving activities in between. The acquired continuous electric vehicle operation data is segmented into sessions marked as “driving” events, such as Figure 2 As shown. Charging activity chains, also called charging patterns, are extracted from activity chains that contain at least one charging event. These charging patterns clearly show the correlation between parking and charging activities between two adjacent trips, revealing the charging flexibility of electric vehicles in the time dimension. Each period within a specific charging pattern is considered as a unified "dispatchable unit". Each dispatchable unit contains a specific sequence of activities, including charging and idle parking. A dispatchable unit refers to the time from the arrival to the departure of an electric vehicle, which is essentially the interval between the end of one driving phase and the beginning of the next driving phase.

[0057] 2. Identify the charging patterns of electric vehicle users.

[0058] The identification of charging modes helps to quantify the potential for large-scale changes in EV behavior under coordinated charging strategies. The charging control simulation is based on switching charging modes by modifying charging time and power. The identified charging modes are shown in Table 1 and Figure 3 As shown. Figure 3 As shown, four main charging activity chain modes account for 97% of the total charging activities.

[0059] Table 1 Classification of different charging modes

[0060]

[0061] Specifically, the charging mode is identified as the following modes:

[0062] Pattern A is the most common pattern, which defines a charging pattern in which the electric vehicle is continuously charged during the entire parking period between two trips. It accounts for about 46.6% of the total charging activity, indicating that nearly half of the charging activity occurs without idle parking time. This may be due to the preferences of electric vehicle users or parking restrictions (such as high parking fees).

[0063] Pattern B accounts for 7.9%, including charging to departure, with a period of idle time before charging.

[0064] Pattern C accounts for 9.4%, which means that there is idle parking time for charging before and after two trips.

[0065] Pattern D accounts for 36%, which is defined as charging immediately after arrival, followed by an extended parking period until departure.

[0066] As can be seen from the definition, mode B represents delayed charging, while mode D represents immediate charging to meet the user's urgent charging needs for electric vehicles, such as mode A. Delayed charging (mode B) or a combination of immediate and delayed charging (mode C) is generally because the charging point is not available upon arrival. For idle parking periods, modes B, C, and D allow for a certain charging flexibility in terms of charging time and power. The existence of these three modes also shows that existing electric vehicle behavior models that assume immediate charging upon arrival to predict charging needs may produce significant errors. Less common modes, such as intermittent mode E (less than 0.1%), are ignored in the charging mode of this application.

[0067] The charging behavior of electric vehicle (EV) users can consist of a single mode or multiple different modes. Figure 4 As shown, different EV user groups are distinguished based on the complexity of their charging patterns. Figure 4The categories of pattern composition, such as single mode, dual mode, tri-mode and quad-mode, are shown in different colors. The proportion reflects the number of EV users whose charging behavior belongs to a specific mode composition category. Only categories with a proportion greater than 0.1% are shown to emphasize the most relevant information. These categories account for 98.6% of the total number of users, of which the largest part is dual-mode users of mode A and mode D, accounting for about 23% of all studied EV users. About 16% of users only exhibit mode A, followed by multi-mode users of mode A, mode C and mode D, accounting for about 12.5%. The identification and analysis of charging modes provides a basis for flexibly quantifying a certain type of user and the number of users, and incorporating them into coordinated charging plans by changing their behavior. This will help decision makers identify target EV users for smart charging (EV-Grid Integration, EGI).

[0068] 3. Analysis of electric vehicle charging characteristics.

[0069] Statistical analysis of the charging power, charging time, and battery SOC of EV charging activities and the provision of a charging load profile under each EV charging mode can better explore the distribution of EV charging time and energy usage patterns, infer the scenarios in which each charging mode occurs, and provide a statistical basis for evaluating the potential flexibility of charging behavior.

[0070] Charging power is a key factor in characterizing the charging scenario of electric vehicle users, as it determines the duration of the charging cycle and indicates the urgency of the user to charge. Charging power is calculated as the average power obtained by multiplying the product of the battery capacity and the SOC difference divided by the charging time. Figure 5 and Figure 6 The charging power distributions of four primary charging modes with similar distribution shapes are shown. Figure 5 Part (a) is a schematic diagram of charging power distribution in mode A; Figure 5 Part (b) is a schematic diagram of charging power distribution in mode B; Figure 5 Part (c) is a schematic diagram of charging power distribution in mode C; Figure 5 Part (d) is a schematic diagram of charging power distribution in mode D.

[0071] To further classify these patterns, a Gaussian mixture model (GMM) is used to identify different charging power levels based on the entire dataset, such as Figure 7 The results show that the charging power is symmetrically distributed around 10.2kW. Therefore, slow charging and fast charging are defined as charging with an average power of less than 10.2kW and greater than 10.2kW, respectively.

[0072] in, Figure 7Part (a) is the charging power distribution diagram using GMM; Figure 7 Part (b) shows the percentage of slow and fast charging times in each mode.

[0073] like Figure 8 As shown, slow charging is more common among most EV users compared to fast charging. Modes B, C, and D overwhelmingly favor slow charging, while Mode A shows a more balanced distribution, with a ratio of 4:6 for fast and slow charging. The higher prevalence of slow charging in Modes B, C, and D, especially at power levels of 3kW and 7kW, indicates that these power levels correspond to the most common charging facilities, especially in residential areas. This also suggests that users with a charging power of 7kW are more likely to have private charging piles or easy access to charging facilities, as 7kW is the mainstream charging power of slow charging piles in residential areas. Therefore, in the dataset used, the higher incidence of slow charging may not necessarily indicate the preference of EV users. The charging power distribution of fast charging is more dispersed, with an average power of around 30 kW.

[0074] Charging patterns with sufficient idle parking time can provide considerable flexibility for coordinated charging control. Figure 8 The charging and total parking duration for each charging mode are shown. The total parking duration includes charging and potential idle parking. The corresponding statistics are shown in Table 2.

[0075] in, Figure 8 (a) is the duration distribution diagram under slow charging; Figure 8 (b) is the duration distribution diagram under fast charging; Figure 8 (c) is the parking duration distribution diagram during slow charging; Figure 8 (d) in the figure is the parking duration distribution diagram during fast charging.

[0076] The analysis of the idle-to-charge time ratio under different charging modes emphasizes the need to develop intelligent charging strategies tailored to the actual usage patterns of electric vehicles. By quantifying the idle time relative to the charging time, the charging mode that provides the most flexibility can be identified.

[0077] Table 2 Parking, charging and idle schedules under different charging modes

[0078]

[0079] In order to infer the scenarios in which these charging modes occur, the binary distribution depicted by the heat map can intuitively show the charging activities occurring at different times and charging powers, combined with the charging power and the start and end time as input features. Fig. 9 As shown, the timeline extends from 4:00 a.m. of the current day to 4:00 a.m. of the next day, where Fig. 9 (a) is the time distribution diagram of slow charging activity; Fig. 9 (b) in the figure is the time distribution diagram of fast charging activity.

[0080] Charging activities are categorized into seven groups (four for slow charging and three for fast charging):

[0081] Morning-Slow (MS): Slow charging that starts and ends in the morning.

[0082] Noon-Slow (NS): Slow charging that starts and ends at noon.

[0083] Afternoon-Slow (AS): Slow charging starts in the afternoon and ends in the evening.

[0084] Evening-Slow (ES): Slow charging that starts in the late afternoon or evening and ends in the early morning of the next day.

[0085] Morning-Fast (MF): fast charging that starts and ends in the morning.

[0086] Noon-Fast (NF): fast charging that starts and ends at noon.

[0087] Afternoon-Fast (AF): Fast charging that starts and ends in the afternoon.

[0088] Data used in constructing the charging activity chain Most electric vehicles are registered as private passenger cars. They are used by different groups of individuals, both for personal and business purposes. They may include daily commuters, ride-hailing drivers, freelancers, teleworkers, and other non-working individuals with flexible work and rest schedules. The specific charging patterns observed in the data indicate different user activities. For example, the MS and MF charging patterns are more likely to be associated with daily commuters charging at their workplace, while the NS and NF patterns may be associated with ride-hailing drivers who tend to resort to public charging stations during their lunch breaks. The AS charging pattern, which starts in the afternoon and ends in the evening, is common among daily commuters who charge in residential areas after get off work. ES charging, which starts charging late in the day and continues until the evening, can be applicable to both commuters and ride-hailing drivers, while AF charging represents random fast charging needs during the day.

[0089] Fig.10 and Fig.11 The distribution of slow and fast charging time on different modes is shown. Fig.10 (a) is the slow charging time distribution diagram of mode A; Fig.10(b) is the slow charging time distribution diagram of mode B; Fig.10 (c) is the slow charging time distribution diagram of mode B; Fig.10 (d) in the figure is the slow charging time distribution diagram of mode B. Fig.11 (a) is the fast charging time distribution diagram of mode A; Fig.11 (b) is the fast charging time distribution diagram of mode B; Fig.11 (c) is the fast charging time distribution diagram of mode B; Fig.11 (d) in the figure is the fast charging time distribution diagram of mode B.

[0090] It can be seen that mode A generally occurs in the afternoon and evening, including AF and all slow charging categories, among which AS and ES are the most frequent. Fig.10 (a) in Figure 2 reveals that many EV users who follow Pattern A start charging around 8am and 7pm and leave after about 2 hours without idle parking. Since Pattern A is the most common pattern, it is assumed to reflect a variety of user types. Although the charging times in Pattern A are consistent with commuting patterns, the pattern may occasionally reflect random or unique schedules, such as users with flexible work schedules or non-commuting EV owners. These users may plan their charging around other daily activities, such as shopping or attending evening events, leading to the charging behavior observed in Pattern A. For example, a user may charge their EV while shopping after get off work and then leave immediately after shopping. Users with flexible schedules such as freelancers, telecommuters, shift workers, or part-time employees may also start their day in the morning, which leads to a charging peak around 8am. After charging, these people may leave immediately and continue their day, whether it is work or other activities. For example, ride-hailing drivers may take a break after the busy evening rush hour, quickly charge, and then continue their shift, which may last all night. Other user groups, such as retirees or stay-at-home people, may also show charging behaviors that differ from traditional commuting patterns. These users may use their EVs for short trips or errands throughout the day, explaining the charging activity around 10 a.m. and 9 p.m. There were no idle stops after charging, suggesting that these users only charged when necessary and left shortly afterwards to continue their daily lives.

[0091] The availability of charging infrastructure may also influence these behaviors, as EV users often take advantage of available charging points, even for short charging sessions. Similar analyses can be applied to other modes. Mode B, including AF, NS, and ES, represents charging behaviors that may be performed by ride-hailing drivers who may face delays due to limited charging points. Mode D, which mainly includes MF, MS, AS, and ES, tends to occur after peak hours in the morning and afternoon, possibly indicating commuter charging patterns. Mode C is an intermediate mode between modes B and D, including NF, MS, AS, and ES. Its slow charging pattern may reflect the charging habits of commuters, while its fast charging pattern suggests that it may be adopted by ride-hailing drivers seeking to charge quickly during lunch hours. Although these inferences have not yet been verified, they provide a preliminary understanding of the billing behaviors associated with different user types, which may help identify potential target users for coordinated billing control.

[0092] The battery SOC at the beginning and end of the charging activity reflects the size and urgency of the user's charging needs. For slow charging, the median SOC at the beginning is always between 40% and 50% in all modes, see Fig.12 .in, Fig.12 (a) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode A under slow charging; Fig.12 (b) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode B under slow charging; Fig.12 (c) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode C under slow charging; Fig.12 (d) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode D under slow charging; Fig.12 (e) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode A under fast charging; Fig.12 (f) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode B under fast charging; Fig.12 (g) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode C under fast charging; Fig.12 (h) is a schematic diagram of battery SOC distribution at the beginning and end of charging in mode D under fast charging.

[0093] The median starting SOC for fast charging for each mode ranged from 30% to 40%, indicating that users who chose fast charging tended to start at lower energy levels compared to users who used slow charging. This variation may indicate that fast charging users are more comfortable operating their EVs at lower SOC levels, perhaps due to the availability of fast charging facilities, which increases their confidence in the reliability of their EVs. Alternatively, the decision to choose fast charging may be driven by lower SOC levels, reflecting greater urgency.

[0094] Regardless of slow or fast charging, Mode C and Mode D are more likely to have high end-of-charge SOC levels, ranging from 90% to 100%. In contrast, the distribution of end-of-charge SOC for Modes A and B is lower and more diverse. This suggests that Modes C and D may have sufficient parking time to support near-full or full charging, while charging activities in Modes A and B may be constrained by the driver's personal schedule or charging facility limitations. Mode B tends to start with a lower SOC but end with a medium SOC, which means that users with frequent Mode B behavior may face severe limitations in charging availability. In contrast, EV users with a high percentage of Modes C and D can participate in coordinated charging control with high flexibility.

[0095] In addition, the application also counts the total charging load for each charging mode to provide a baseline for inflexible charging load profiles. The total charging load represents the accumulation of energy charged during each charging session. Fig.13 shows the average daily total EV charging load over the course of a month, and Fig.13 The slow and fast charging cases of each charging mode are shown separately along with their 95% envelopes over 30 days. The proportion of charging load contributed by each mode is also indicated in each sub-figure. Fig.13 (a) is the charging load curve of all electric vehicles; Fig.13 (b) is a charging load curve corresponding to slow charging and fast charging; Fig.13 (c) is the charging load curve corresponding to mode A, mode B, mode C and mode D under slow charging; Fig.13 (d) in the figure is the charging load curve corresponding to mode A, mode B, mode C and mode D under fast charging.

[0096] In general, slow charging accounts for the majority of the total private EV load in this application, with a variance of 3100kW 2, with a peak-to-valley difference of 10532kW. The peak in the morning is relatively smooth, with the main peak consistent with the evening peak of residential electricity consumption, while the lowest point is around 7 to 8 am. Mode A and Mode D contribute the most to the total charging load distribution, with Mode A accounting for approximately 41% of the daily slow charging load and Mode D accounting for approximately 40% of the daily slow charging load. For the fast charging load, Mode A and Mode D contribute approximately 68% and 22%, respectively. The slow charging load curves of all four modes exhibit peak periods that overlap with the evening peak of electricity demand, and have secondary peaks during the day, especially for Mode D. The overall fast charging load curve is generally smoother, with less variation throughout the day, with a variance of 1728kW 2 , the peak-to-valley difference is 5735kW. However, it should be noted that during the daytime period from 7am to 6pm, its variance is greater than that of the slow charging load, and the fast charging load during the day exceeds the fast charging load at night and in the early morning. This indicates that fast charging, characterized by short duration and high power level, is more frequent during the day.

[0097] It can be seen from the above that different charging modes have different degrees of impact on power load. Some modes (such as mode D) may have a negative load impact on the power grid, but also provide great flexibility for coordinated charging control.

[0098] 4. Coordinated charging scheduling for switching charging modes.

[0099] In order to quantify the effectiveness of the proposed changes in charging behavior under different coordinated charging strategies, simulations are performed by performing switching between charging modes based on EV charging data. In order to evaluate the flexibility of electric vehicle charging, this application redefines the flexibility definition of demand-side resources. A shifting load refers to a load that can be moved in a timely manner, but must be moved as a whole without interrupting the entire charging cycle. Transferable (shifting) loads allow charging interruptions and redistribute the charging load within the available scheduling time. Reducible loads are achieved by reducing the charging power. It can be inferred from the definition of the charging mode that the charging load of mode A is more rigid and less controllable. In contrast, the charging loads of modes B, C, and D exhibit greater scheduling flexibility due to idle parking time, allowing shifting, reducible, or switchable load adjustments. The detailed definitions are as follows:

[0100] Shiftable: Due to stops before or after charging, the charging load can be shifted overall while maintaining the same average charging rate and duration of the charging session.

[0101] Scalable: By spreading the charging process over the entire parking period, the charging power can be reduced to a lower level.

[0102] Switchable: Charging can be rescheduled during parking windows to take advantage of lower electricity rates, allowing charging to be paused throughout the charging process.

[0103] Modifying charging behavior to switch from one mode to another (e.g., Mode C or Mode D to Mode B, or Mode B, Mode C, Mode D to Mode A) allows emulating different grid management strategies. This is achieved by adjusting the raw charging session data. Specifically:

[0104] Going from Mode C or Mode D to Mode B simulates a delayed charging scenario where charging is shifted to the last possible minute.

[0105] Evaluate the impact of reducing power during a charging session from Mode B, Mode C, or Mode D to Mode A, or evaluate the impact of smart charging using Time-of-Use tariffs, ToU rates that strategically reallocate charging sessions to maximize cost-effectiveness.

[0106] In order to make full use of the variable charging behavior, a smart charging strategy based on a mixed integer linear programming (MILP) model is designed. The model optimizes the charging schedule in Mode B, Mode C, and Mode D to redistribute the transferable load using time-of-use electricity prices. The algorithm is shown below:

[0107]

[0108]

[0109] Where C(t) is the electricity price at time t; T i，start is the start time of the ith schedulable unit; T i，end is the end time of the ith schedulable unit; P i,avg is the average charging power in the ith dispatchable unit; x(t) is a binary variable at the tth time, which is 1 if charging occurs at time t, otherwise 0; P i (t) is the charging power at the tth time in the i-th dispatchable unit; E i is the total energy billed in the ith dispatchable unit; T i,cs is the charging start time in the ith dispatchable unit; T i,ce is the charging end time in the ith dispatchable unit; SOC i,start is the starting SOC of the ith dispatchable unit; SOC i,end End SOC for the i-th dispatchable unit; Capacity i is the battery capacity of the electric vehicle; N i is the charging factor (electricity fee) charged by ToU without intelligent charging of dispatchable units; S i is the electricity cost (charging coefficient) under ToU with dispatchable unit intelligent charging function; CS i,save Provide cost savings through smart charging of dispatchable units.

[0110] Based on the above analysis, a smart charging MinFee scenario with ToU rate is designed to make full use of the parking time between adjacent trips while meeting the charging needs of electric vehicle users. It corresponds to the evaluation of the translation, reduction and transferability potential of EV charging load.

[0111] In the scenario of smart charging with time-of-use electricity price, it is assumed that the total parking time and charging power remain unchanged. The MILP-based smart charging method is applied to mode B, mode C, and mode D to minimize the charging cost of each charging period under the time-of-use electricity price. Its slow charging and fast charging scenarios are denoted as MinFee-S and MinFee-F, respectively. The MILP model is constrained by the charging demand and the start and end time of each dispatchable unit, and the objective function is used to minimize the single charging cost. The maximum charging power is set to the average power in the original charging session data, and the minimum charging power is set to 0. The electricity price scheme refers to the time-of-use electricity price for industrial and commercial users in Beijing.

[0112] In order to investigate whether the change in EV charging behavior can effectively shift the peak overlapping charging load and contribute to valley filling, a visual net load-time flow model is constructed. The model illustrates the daily shift of EV charging load during peak and valley periods through coordination. First, the daily curve of EV charging load and the load difference before and after coordination are performed. Then, the amount of offset charging load is calculated by analyzing whether the increased or decreased load in each period flows into or out of the adjacent period and how much flows into or out of the adjacent period. This process is iteratively performed for each scenario until the start and end points of the net load for each period are determined.

[0113] To flatten the electricity load distribution, the ToU pricing mechanism uses the price difference between peak and off-peak periods to encourage EV users to charge their vehicles during low-price periods. Fig.14 Figure 1 is a schematic diagram of power load distribution. Fig.14 As shown in (a), in the MinFee-S scenario, most of the charging load during the night peak and night flat periods is transferred to the valley period with the lowest electricity price. Some high-priced morning peak loads are transferred to the adjacent afternoon flat period and even to the valley period of the next day. This transferred load leads to a significant reduction in EV charging load during peak hours, with the morning peak load reduced from 9.4% to 2.8%, and the evening peak load reduced from 28% to 3.4%, respectively. At the same time, the load in the valley period increased sharply from 37% to 74.1%. This price-driven smart charging allows electric vehicle users to avoid peak charging to reduce their charging costs.

[0114] Fig.14(b) In the MinFee-F scenario, part of the charging load during the morning-flat, evening-peak, and evening-flat periods is shifted to the valley period of the next day, while more than one-third of the morning-peak load is shifted to the later afternoon-flat period. This results in a reduction in EV charging load from 17.2% to 9.6% during the morning peak and from 25.9% to 16.7% during the evening peak. It can be concluded that the smart charging approach with ToU tariffs is more effective for slow charging scenarios than for fast charging scenarios because fast charging activities mainly occur during the day when pricing incentives are less effective.

[0115] like Fig.14 (c)- Fig.14 As shown in (d) in Figure 2, a sharp load peak of EV charging load is found around 11 pm in both MinFee-S and MinFee-F scenarios. This sudden surge may lead to overload of the local grid because EV users usually start charging during the only low-price period of the day. In the MinFee-F scenario, a new load peak is observed between 1 pm and 3 pm. The new peak may coincide with the peak of PV generation, indicating that ToU pricing has a positive impact on promoting the integration of renewable energy. Therefore, decision makers must design suitable time-of-use electricity price cycles and effective dynamic pricing mechanisms to incentivize off-peak charging while ensuring compatibility with renewable energy production patterns and avoiding the formation of new peaks.

[0116] Based on the charging data samples collected by the electric vehicle big data platform and combined with the charging behavior of electric vehicles, this application proposes a time-of-use electricity price intelligent charging method based on charging pattern recognition. The importance of identifying idle parking time and its ratio to the required charging time is emphasized, which is a key indicator for identifying potential electric vehicle users who choose intelligent charging control. By examining the sequence and timing of parking and charging events, it is able to quantify the potential potential of electric vehicles to change the charging load. In the MinFee scheme, using one hour of idle parking time, up to 0.45 kWh of net charging load can be transferred from peak hours, and 0.54 kWh can be transferred to the valley period for slow charging. For fast charging, the transferred loads during peak and valley periods reach 0.55 kWh and 0.38 kWh, respectively.

[0117] Based on the same inventive concept, the embodiment of the present application also provides a charging device based on time-sharing electricity price for implementing the above-mentioned charging method based on time-sharing electricity price. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the charging device based on time-sharing electricity price provided below can refer to the limitations of the charging method based on time-sharing electricity price above, and will not be repeated here.

[0118] In an exemplary embodiment, a charging device based on time-of-use electricity price is provided, comprising:

[0119] The information data acquisition module is used to acquire information data; the information data includes charging information data of the target electric vehicle in the dispatchable unit; the dispatchable unit is extracted based on the charging activity chain; the dispatchable unit includes: a charging stage and an idle parking stage; the charging activity chain is based on the operating data of all electric vehicles in the target area, and is sorted according to timestamps to obtain an activity sequence used to characterize the start and end of a driving event.

[0120] The charging scheme determination module is used to determine the charging scheme according to the information data based on the mixed integer linear programming model; the charging scheme is determined based on the time-of-use electricity price to minimize the charging coefficient of each charging period; the charging scheme is used to transfer the charging peak load to achieve peak shaving and valley filling; the mixed integer linear programming model adopts the charging control simulation method, identifies the charging mode based on the charging activity chain, and statistically analyzes the charging characteristic data under each charging mode after identification, to obtain a mathematical model; the charging characteristic data includes: charging power, charging time and battery SOC.

[0121] In an exemplary embodiment, a computer device is provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a charging method based on a time-of-use electricity price.

[0122] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0123] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0124] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization of the corresponding device owner. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0125] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A charging method based on time-of-use electricity price, characterized in that: The charging method based on time-of-use electricity price includes: Acquire information data; the information data includes charging information data of the target electric vehicle in the dispatchable unit; the dispatchable unit is extracted based on the charging activity chain; the dispatchable unit includes: a charging stage and an idle parking stage; the charging activity chain is an activity sequence that is obtained by sorting the operating data of all electric vehicles in the target area according to timestamps and is used to characterize the start and end of a driving event; According to the information data, a charging plan is determined based on a mixed integer linear programming model; the charging plan is determined based on time-of-use electricity prices to minimize the charging coefficient of each charging period; the charging plan is used to transfer the charging peak load to achieve peak shaving and valley filling; the mixed integer linear programming model adopts a charging control simulation method, identifies the charging mode based on the charging activity chain, and statistically analyzes the charging characteristic data under each identified charging mode to obtain a mathematical model; the charging characteristic data includes: charging power, charging time and battery SOC.

2. The charging method based on time-of-use electricity price according to claim 1, characterized in that: The mixed integer linear programming model includes: an objective function and constraints; the constraints include: a charging demand constraint and a time constraint; wherein the time constraint is determined based on the start time and the end time of each of the dispatchable units.

3. The charging method based on time-of-use electricity price according to claim 2, characterized in that: The objective function includes: Where F is the objective function; T i，end is the end time of the ith schedulable unit; T i，start is the start time of the ith schedulable unit; P i，avg is the average charging power in the i-th dispatchable unit; x(t) is the binary variable at the t-th moment; C(t) is the electricity price at the t-th moment.

4. The charging method based on time-of-use electricity price according to claim 2, characterized in that: The charging demand constraints specifically include: Among them, T i，end is the end time of the ith schedulable unit; T i，start is the start time of the ith schedulable unit; P i (t) is the charging power at the tth time in the i-th dispatchable unit; E i is the total energy billed in the i-th dispatchable unit.

5. The charging method based on time-of-use electricity price according to claim 2, characterized in that: According to the information data, a charging plan is determined based on a mixed integer linear programming model, specifically including: According to the information data, the objective function is solved based on the constraint conditions to obtain a charging plan; the information data includes: the battery capacity, starting SOC, ending SOC, charging start time and charging end time of the target electric vehicle.

6. The charging method based on time-of-use electricity price according to claim 1, characterized in that: The charging method based on time-of-use electricity price also includes: According to the charging scheme, the setting period and pricing mechanism of the time-of-use electricity price are adjusted to encourage off-peak charging.

7. A charging device based on time-of-use electricity price, characterized in that: The charging device based on time-of-use electricity price includes: An information data acquisition module is used to acquire information data; the information data includes charging information data of the target electric vehicle in the dispatchable unit; the dispatchable unit is extracted based on the charging activity chain; the dispatchable unit includes: a charging stage and an idle parking stage; the charging activity chain is an activity sequence that is obtained by sorting the operating data of all electric vehicles in the target area according to timestamps and is used to characterize the start and end of a driving event; A charging scheme determination module is used to determine a charging scheme based on the information data and a mixed integer linear programming model; the charging scheme is determined based on time-of-use electricity prices to minimize the charging coefficient of each charging period; the charging scheme is used to transfer the charging peak load to achieve peak shaving and valley filling; the mixed integer linear programming model is a mathematical model obtained by using a charging control simulation method, identifying the charging mode based on the charging activity chain, and statistically analyzing the charging characteristic data under each identified charging mode; the charging characteristic data includes: charging power, charging time and battery SOC.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the charging method based on time-of-use electricity price as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the charging method based on time-of-use electricity price described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the charging method based on time-of-use electricity price described in any one of claims 1 to 6 is implemented.

Images