A queuing theory-based electric vehicle charging and discharging scheduling method for a V2G network

Abstract

The application belongs to the field of V2G, and particularly relates to a kind of electric vehicle charging and discharging scheduling method of V2G network based on queuing theory, comprising: constructing V2G electric vehicle charging and discharging network;EV user authentication and sharing key information;In the background of TOUP, generate charging and discharging queuing model according to PLAS scheme;Construct CS recommendation algorithm based on queuing theory model with cut-off parameter CO;According to the constructed CS recommendation algorithm, generate charging and discharging scheduling response for EV user;According to network service feedback information, V2G network real-time updates cut-off parameter CO and user summary information to update optimal scheduling strategy in real time;The method can effectively coordinate the charging and discharging process of electric vehicle, obtain profit from unused storage power, and maximize the reduction of high power demand generated by a large number of electric vehicle users charging at the same time, to balance the load of power grid and the load peak-valley ratio at peak time of power grid.

Description

Technical Field

[0001] This invention belongs to the field of V2G, and specifically relates to a method for scheduling the charging and discharging of electric vehicles in a V2G network based on queuing theory. Background Technology

[0002] With increasing concern about fossil fuel consumption and global warming, new energy electric vehicles have become an important mode of transportation. Inevitably, large-scale deployment of electric vehicles may lead to grid overload, degraded power quality, and reduced energy efficiency. Vehicle-to-Grid (V2G) technology has emerged to address this need, aiming to enable interaction between electric vehicles and the power grid. This allows electric vehicles to absorb electricity when grid load is low and release it when grid load is high, earning the difference in revenue. In the V2G model, electric vehicles can adjust charging time and power according to grid demand, and discharge electricity through V2G terminals when the vehicle is not in use. This energy interaction method not only meets the energy replenishment needs of new energy vehicles but also participates in grid ancillary services, driving a profound transformation in energy development.

[0003] Electric vehicles (EVs) have a significant impact on the stable operation and planning of charging stations (CS) and the power system. Due to the high concurrency of EV charging time, it is necessary to conduct in-depth research on how to guide EV charging and discharging, especially when a large number of drivers choose fast charging mode, which will make the CS load more concentrated and large-scale.

[0004] Therefore, how to effectively coordinate the charging and discharging process of electric vehicles, reduce the peak-to-valley load ratio during peak grid periods, and balance user satisfaction and grid energy costs are urgent problems to be solved. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a queuing-based V2G network electric vehicle charging and discharging scheduling method. This method employs a queuing-based electric vehicle charging and discharging scheduling strategy, using a peak load adjustment scheme to divide the EV queues into request charging queues and request discharging queues. Then, a CS recommendation algorithm based on a queuing-based model with a cutoff parameter CO is used to schedule charging and discharging users. The aim is to reduce the waiting time of queued users and flexibly coordinate the power supply from discharging EV users to the grid during peak hours, thereby reducing peak load and improving grid stability. Specifically, this includes:

[0006] S1: Construct the V2G system network and initialize system parameters; the V2G system network includes electric vehicles, charging stations, and the power grid center.

[0007] S2: Electric vehicle users send an authentication request to the power grid center and establish connections with surrounding entities via a wireless network;

[0008] S3: After successful identity authentication, the trolley user uploads their information to the V2G system network;

[0009] S4: Construct vehicle users' waiting time preferences and time-of-use pricing strategies, and set electric vehicle charging and discharging scheduling strategies based on vehicle users' waiting time preferences, time-of-use pricing strategies, and the information of electric vehicle users themselves.

[0010] S5: Generate two types of queuing queues based on the electric vehicle charging and discharging scheduling strategy: the request charging queue and the request discharging queue.

[0011] S6: The CS recommendation algorithm is used to process the request charging queue and request discharging queue to generate the charging and discharging scheduling response for electric vehicle users; the parameters of the CS recommendation algorithm are updated according to the charging and discharging scheduling response to obtain the optimal scheduling strategy.

[0012] Preferably, the process by which an electric vehicle user sends an identity authentication request to the power grid center includes: the electric vehicle pre-registering in the UDC system and receiving an ID; and the UDC assigning a pseudo-identity to the registered EV.

[0013] Preferably, the information of the tram user includes: current location, current state of charge, travel plan, and the maximum acceptable waiting time when receiving service.

[0014] Preferably, constructing vehicle user waiting time preferences and time-of-use pricing strategies includes: collecting vehicle information data, including vehicle user waiting times; establishing a vehicle user waiting time preference model based on the waiting times; obtaining the charging station usage status for the current time period; and formulating corresponding time-of-use pricing strategies based on the user waiting time preference model and the charging station usage status for the current time period. The time-of-use pricing strategy includes setting a charging station usage threshold; if the usage threshold is exceeded, a high-price strategy is implemented; otherwise, a low-price strategy is specified.

[0015] Preferably, a charging and discharging scheduling strategy for electric vehicles is set: at the beginning of each time period T, the EV sends its own information to the UDC system; based on its own information, the vehicle users are divided into sailing users and parked users; based on the trip plan and current state of charge in its own information, the minimum SOC required by the sailing users is calculated, and it is checked whether the remaining SOC meets the charging and discharging requirements, and the charging and discharging command is executed based on the check result.

[0016] Furthermore, calculating the minimum SOC required by the user of the navigation system includes:

[0017]

[0018] Where ε is the driving efficiency coefficient, R max The maximum driving distance of an EV when fully charged, where D is the driving distance.

[0019] Furthermore, the formula for charging and discharging is:

[0020] SOC ch =SOC ini +K ch *D ch *T EV_schedule_ch

[0021]

[0022] Among them, SOC ini This represents the initial state of charge of the EV; K ch and K disch These are the charging efficiency and discharging efficiency, respectively; D ch and D disch T represents the charge / discharge rate; EV_schedule_ch ,T EV_schedule_disch These represent the time for the EV charging and discharging processes, respectively.

[0023] Preferably, the CS recommendation algorithm is used to process the request charging queue and request discharging queue, including: assuming n is the number of busy sockets and C_O is the queue cutoff parameter, when n < C_O, all arriving electric vehicle users of both charging and discharging categories are served according to the first-come, first-served principle; when C_O ≤ n < S, electric vehicle users discharging are given priority in service, and electric vehicle users requesting charging services are added to the queuing queue.

[0024] Preferably, updating the CS recommendation algorithm parameters based on the charging and discharging scheduling response includes: for discharging services, if the average waiting time of the queue in the discharging scheduling response is greater than the user's waiting time preference, then the CO cutoff parameter is updated by decreasing; for charging services, if the charging scheduling response is a response that meets the charging needs of electric vehicle users, then the electric vehicle users join the charging queue of the charging station; if the charging scheduling response is a response that does not meet the charging needs of electric vehicle users, then the profile information parameters are updated, wherein the profile information includes the current location, current state of charge, travel plan, and maximum acceptable waiting time.

[0025] The beneficial effects of this invention are as follows:

[0026] This invention uses the peak load adjustment scheme PLAS to divide the EV queues requesting charging and the EV queues requesting discharging. It employs a CS recommendation algorithm based on a queuing theory model with a cutoff parameter CO to schedule charging and discharging users. The aim is to reduce the waiting time of queue users and flexibly coordinate the power supply of discharging EV users to the grid during peak hours, thereby reducing peak load and improving grid stability. Attached Figure Description

[0027] Figure 1 This is a model of the V2G network charging and discharging scheduling system in this invention;

[0028] Figure 2 This is a flowchart of the peak load adjustment scheme in this invention;

[0029] Figure 3 This is a diagram of the charge / discharge queuing theory model with a cutoff parameter in this invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Figure 1 The proposed V2G network charging and discharging scheduling system model of this invention mainly consists of three core entities: EV, CS, and UDC. EVs with bidirectional energy trading capabilities can perform charging and discharging operations and can be scheduled to trade within specific CSs, thereby regulating the load within the area. CSs can provide corresponding rewards to EVs that actively participate in scheduling activities, and are equipped with smart meters to record information such as the amount of energy traded and the electricity price. UDC, as a central data agency, is responsible for registering the legal identities of EVs and CSs, and also processes all transaction data created between them.

[0032] A method for scheduling the charging and discharging of electric vehicles in a V2G network based on queuing theory, comprising:

[0033] S1: Construct the V2G system network and initialize system parameters; the V2G system network includes electric vehicles, charging stations, and the power grid center.

[0034] S2: Electric vehicle users send an authentication request to the power grid center and establish connections with surrounding entities via a wireless network;

[0035] S3: After successful identity authentication, the trolley user uploads their information to the V2G system network;

[0036] S4: Construct vehicle users' waiting visual preferences and time-sharing pricing strategies, and set electric vehicle charging and discharging scheduling strategies based on vehicle users' waiting visual preferences, time-sharing pricing strategies, and electric vehicle users' own information.

[0037] S5: Generate two types of queuing queues based on the electric vehicle charging and discharging scheduling strategy: the request charging queue and the request discharging queue.

[0038] S6: The CS recommendation algorithm is used to process the request charging queue and request discharging queue to generate the charging and discharging scheduling response for electric vehicle users; the parameters of the CS recommendation algorithm are updated according to the charging and discharging scheduling response to obtain the optimal scheduling strategy.

[0039] During the initialization phase, each EV communicates with entities in the V2G network for identity authentication, establishing a secure connection with the UDC (Utility Data Center). EV users need to share their profile information with the V2G network to obtain optimal CS (Customer Service) information. This EV profile information includes their current location. EV Current State of Charge (SOC) and Trip Schedule EV The longest acceptable waiting time ω when receiving service. EV .

[0040] like Figure 2 As shown, the PLAS peak load adjustment scheme divides the EV queue into a charging queue and a discharging queue. The steps are as follows:

[0041] Step 1: At the beginning of each time period T, EV sends its own profile information to the UDC system;

[0042] Step 2: For the vehicle trip plan in the incoming EV summary information, divide the vehicle into two types of user entities: navigation users and parking users;

[0043] Step 3: Obtain the current state of charge from the EV summary information;

[0044] Step 4: For users on the road, calculate the minimum required SOC based on the mileage, and check whether their remaining SOC meets the charging and discharging requirements. This is categorized into users requiring scheduled charging and users requiring scheduled discharging. Assume the maximum distance an EV can travel on a full charge is R. max .

[0045] The SOC required for EV to complete distance D is represented by the following relationship:

[0046]

[0047] Among them, SOC tripR represents the percentage of electrical energy required to travel a certain distance, ε represents the driving efficiency coefficient, D represents the distance the vehicle needs to travel, and R represents the distance the vehicle needs to travel. max This indicates the maximum driving distance an EV can travel on a full charge.

[0048] Step 5: For parked users, based on the current SOC status, they are divided into users who need to be scheduled for charging and users who need to be scheduled for discharging.

[0049] For the proposed PLAS scheme, the decision of whether an EV participates in the charging or discharging process is based on the EV's profile information and preferences. The charging and discharging process is represented by the following formula:

[0050] SOC ch =SOC ini +K ch *D ch *T EV_schedule_ch

[0051]

[0052] Among them, SOC ini This represents the initial state of charge of the EV; K ch and K disch These are the charging efficiency and discharging efficiency, respectively; D ch and D disch T represents the charge / discharge rate; EV_schedule_ch ,T EV_schedule_disch These represent the time for the EV charging and discharging processes, respectively.

[0053] Charging efficiency and discharging efficiency describe the battery efficiency and performance of an electric vehicle, reflecting the battery degradation effects caused by the charging and discharging modes of the electric vehicle and incorporating battery life cost considerations. They can be affected by factors such as charging and discharging frequency, charging and discharging rate, and operating conditions.

[0054] Based on the PLAS scheme, two types of electric vehicles were generated: EVs requesting charging services and EVs requesting discharging services. A priority-based system with separate buffers was used to model the CS (Cyclic Components). The model's input data consisted of the electric vehicles scheduled for charging and discharging, and the output data consisted of the electric vehicles with the required SOC (State of Charge).

[0055] Each charging station (CS) is considered a multi-server queue with S identical servers. Assume the arrival process of electric vehicles (EVs) follows a Poisson distribution with an arrival rate of λ. disch and λ chTherefore, EVs will only receive discharge service when the relevant queue is empty and there is a free outlet. Furthermore, when no outlet is available, newly arriving EVs requesting discharge service will not interrupt other EVs that are already charging, and service is provided to each type of electric vehicle based on a first-come, first-served (FCFS) principle.

[0056] Figure 3 This is a diagram of the charging / discharging queuing theory model with a cutoff parameter in this invention. The cutoff parameter CO is defined for the charging process of electric vehicles according to the following principles: 1) When the number of busy outlets is less than CO, service is provided to any arriving electric vehicle. 2) When a busy outlet with a cutoff value of CO becomes idle, charging service is provided to electric vehicles scheduled for charging. 3) EV users scheduled for discharging are immediately served when an outlet becomes idle; otherwise, they can queue.

[0057] The proposed queuing model prioritizes discharging over charging services, allowing EV users to benefit from unused electricity and assisting the grid in meeting its peak load demands. Therefore, arriving EV users requesting charging services will wait indefinitely if the number of busy outlets equals or exceeds a specified cutoff level (CO). The operational parameters are as follows:

[0058] λ=λ ch +λ disch

[0059] μ = μ ch =μ disch

[0060] ρ ch =λ ch / μ

[0061] ρ disch =λ disch / μ

[0062] ρ=ρ disch +ρ ch

[0063] Where λ is the vehicle arrival rate, λ ch For the arrival rate of charging vehicles, λ disch Let μ be the vehicle arrival rate, and μ be the total service rate. ch For charging service rate, μ disch ρ represents the discharge service rate, and ρ represents the system service strength. ch and ρ disch Service strength for charging and discharging respectively

[0064] In the queuing model, let n be the number of busy outlets. When n < C_O, all arriving EV users of both charging and discharging types can immediately access the service. When C_O ≤ n < S, only EV users discharging are served, and arriving EV users requesting charging services will be queued.

[0065] Once the charging or discharging service for the current EV user is completed, and the busy socket is less than CO, EV users requiring charging service will proceed according to the FCFS order. n, N disch N ch Three parameters can be used to describe the cutoff parameters in the M / M / S queuing model. The designed model is a three-dimensional Markov chain. The steady-state probability is defined as:

[0066] π(N disch N ch ,n)=Pr(N disch N ch ,n)

[0067] Where, N disch ≥0, N ch ≥0, 0≤n≤S, π(N) disch N ch ,n)=0,N disch This represents the number of EVs requesting discharge services that are currently in a waiting state; N ch is the number of EVs requesting charging service that are in a waiting state; n is the number of busy outlets. Let the initial conditions be:

[0068]

[0069] The steady-state equation of the model is:

[0070] (λ+nμ)π(0,0,n)=λπ(0,0,n-1)+(n+1)μπ(0,0,n+1)

[0071] Where 0 ≤ n ≤ CO-1.

[0072] When N ch When >0, the steady-state equation of the model is:

[0073] (λ+COμ)π(0,N ch CO)=λ ch π(0,N ch -1,CO)+

[0074] (CO+1)μπ(0,N ch ,CO+1)+COμπ(0,N ch +1,CO)

[0075] When CO+1≤n<S-1, it can be derived that:

[0076] (λ+nμ)π(0,N ch ,n)=λ disch π(0,N ch ,n-1)+λ ch π(0,N ch -1,n)+

[0077] (n+1)λπ(0,N ch (n+1)

[0078] (λ+Sμ)π(0,N ch ,S)=λ disch π(0,N ch ,S-1)+λ ch π(0,N ch -1,S)+

[0079] Sμπ(1,N ch ,S)

[0080] When N disch When >0, the steady-state equation of the model is:

[0081] (λ+Sμ)π(N disch N ch ,S)=λ disch π(N disch -1,N ch ,S)+

[0082] λ ch π(N disch N ch -1,S)+Sμπ(N disch +1,N ch ,S)

[0083] When 0 ≤ n ≤ CO, the steady-state equation of the model is:

[0084]

[0085] When CO ≤ n < S, we can assume for:

[0086]

[0087] For EV users queuing for discharge services, the average waiting time for discharge is {W}. disch If a character follows an M / M / S queue, the probability of waiting is non-zero; therefore:

[0088]

[0089] The average length of the discharge user queue is:

[0090] {Loq disch}=λ disch {W disch}

[0091] The average waiting time in the EV charging user queue is:

[0092]

[0093]

[0094] The average length of the charging queue for EV users is:

[0095] {Loq ch}=λ ch {W ch}

[0096] The CS recommendation algorithm based on a queuing theory model with a cutoff parameter CO takes the peak time of the scheduling cycle as input parameters, and the EV charge / discharge scheduling plan is the output of the PLAS scheme. For EV users planning to receive charge / discharge services, the algorithm can convey two responses: a positive response of a suitable CS address, and a negative response of notifying them of their waiting time preference ω. EV There is no charging / discharging service available, meaning CS cannot provide charging / discharging services to EV users.

[0097] For discharge services, when the minimum waiting time satisfies {W disch}>ω EV The CRA algorithm updates the cutoff parameter CO by decreasing it to give electric vehicles more opportunities to sell their electrical energy, thus satisfying the waiting time preference ω initially expressed by EV users. EV The best CS is selected again in this way.

[0098] This invention employs a CS recommendation algorithm to process the request charging queue and request discharging queue, including: Let n be the number of busy sockets and C_O be the queue cutoff parameter. When n < C_O, all arriving electric vehicle users of both charging and discharging categories are served according to the first-come, first-served principle; when C_O ≤ n < S, electric vehicle users discharging are given priority in service, and electric vehicle users requesting charging services are added to the queuing queue.

[0099] The CS recommendation algorithm parameters are updated based on the charging and discharging scheduling response, including: for discharging services, if the average waiting time in the queue in the discharging scheduling response is greater than the user's waiting time preference, the CO cutoff parameter is decreased; for charging services, if the charging scheduling response is suitable for EV users to charge, the EV users are added to the charging queue of that charging station; if the charging scheduling response is not suitable for EV users to charge, the profile information parameters are updated, where the profile information includes the current location, current state of charge, travel plan, and maximum acceptable waiting time.

[0100] This method can generate profits from the potential stored electricity of electric vehicles while minimizing the high power demand on the grid caused by a large number of electric vehicle users charging simultaneously. It balances grid load and peak-to-valley ratio by dispatching some dischargeable vehicles to the high-load grid during peak hours.

[0101] The above-described embodiments further illustrate the purpose, technical solution, and advantages of the present invention. It should be understood that the above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made to the present invention within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for scheduling the charging and discharging of electric vehicles based on queuing theory in V2G networks, characterized in that, include: S1: Construct the V2G system network and initialize system parameters; the V2G system network includes electric vehicles, charging stations, and the power grid center. S2: Electric vehicle users send an authentication request to the power grid center and establish connections with surrounding entities via a wireless network; S3: After successful identity authentication, the trolley user uploads their information to the V2G system network; S4: Construct vehicle users' waiting time preferences and time-of-use pricing strategies. Based on vehicle users' waiting time preferences, time-of-use pricing strategies, and electric vehicle users' own information, set up electric vehicle charging and discharging scheduling strategies. Setting up electric vehicle charging and discharging scheduling strategies: At the beginning of each time period T, the EV sends its own information to the UDC system; based on its own information, vehicle users are divided into sailing users and parked users; based on the trip plan and current state of charge in its own information, the minimum SOC required for sailing users is calculated, and it is checked whether the remaining SOC meets the charging and discharging requirements. Based on the check results, the charging and discharging commands are executed. Calculating the minimum SOC required by the user of the navigation system includes: ； in, This is the driving efficiency coefficient. The maximum driving distance of an EV when fully charged, where D is the driving distance; The formula for charging and discharging is: ； ； in, This represents the initial state of charge of the EV. and These are charging efficiency and discharging efficiency, respectively. and This refers to the charge / discharge rate; , These are the times for the EV charging and discharging processes, respectively. S5: Generate two types of queuing queues based on the electric vehicle charging and discharging scheduling strategy: the request charging queue and the request discharging queue. S6: The CS recommendation algorithm is used to process the request charging queue and request discharging queue to generate charging and discharging scheduling responses for electric vehicle users; the parameters of the CS recommendation algorithm are updated based on the charging and discharging scheduling responses to obtain the optimal scheduling strategy. The CS recommendation algorithm is used to process the request charging queue and request discharging queue, including: Let... For the number of busy sockets, This is the queue cutoff parameter, when At that time, all arriving electric vehicle users of both charging and discharging categories will be provided with service on a first-come, first-served basis; when At that time, priority will be given to electric vehicle users who are discharging, and electric vehicle users who request charging services will be added to the queue. Updating the CS recommendation algorithm parameters based on the charge / discharge scheduling response includes: for the discharge service, if the average waiting time of the queue in the discharge scheduling response is greater than the user's waiting time preference, then... The cutoff parameters are updated incrementally. For charging services, when the charging dispatch response is in line with the trolley user's charging needs, the trolley user joins the charging queue of the charging station. When the charging dispatch response is not in line with the trolley user's charging needs, the overview information parameters are updated, including the current location, current state of charge, trip plan, and maximum acceptable waiting time.

2. The electric vehicle charging and discharging scheduling method based on queuing theory in a V2G network according to claim 1, characterized in that, The process of an electric vehicle user sending an identity authentication request to the power grid center includes: the electric vehicle pre-registering in the UDC system and receiving an ID; and the UDC assigning a pseudo-identity to the registered EV.

3. The electric vehicle charging and discharging scheduling method based on queuing theory in a V2G network according to claim 1, characterized in that, The information of tram users includes: current location, current state of charge, travel plan, and the maximum acceptable waiting time when receiving service.

4. The electric vehicle charging and discharging scheduling method based on queuing theory in a V2G network according to claim 1, characterized in that, The process of constructing vehicle user waiting time preferences and time-of-use pricing strategies includes: collecting vehicle information data, including vehicle user waiting times; establishing a vehicle user waiting time preference model based on waiting times; obtaining the charging station usage status for the current time period; and formulating corresponding time-of-use pricing strategies based on the user waiting time preference model and the charging station usage status for the current time period. The time-of-use pricing strategy includes setting a charging station usage threshold; if the usage threshold is exceeded, a high-price strategy is implemented; otherwise, a low-price strategy is implemented.

Images