Control method, device and equipment for electric vehicle parking lot

Abstract

The invention provides a control method, device and equipment for an electric vehicle parking lot, and belongs to the field of power grid cooperative control. A preset vehicle number prediction model is adopted to obtain a parking number prediction value; adopting a preset parking time length prediction model to obtain a parking time length prediction value; and according to the optimization objective function and a preset constraint condition, solving an optimization control model, and based on the output of the optimization control model, carrying out optimization control on charging and discharging of the electric vehicle. According to the invention, in the target parking lot control process, the state of the electric vehicle, such as the parking time and the number of the electric vehicle, is considered, the intelligent degree of parking lot control is improved, and the control efficiency of the parking lot on charging and discharging of the electric vehicle is improved.

Description

Technical Field

[0001] The present invention application relates to the field of grid collaborative control, and particularly to a control method, device and equipment for an electric vehicle parking lot. Background Art

[0002] The current vehicle-to-grid (V2G, referring to electric vehicles feeding power to the grid) technology mainly relies on a centralized management control strategy to coordinate the charging and discharging behaviors of electric vehicles and adjust the load in combination with the grid demand. However, the use of vehicles is unpredictable. For example, the time when a vehicle enters the parking lot, the time when it leaves, the power consumption, the charging frequency, etc. are difficult to predict. In addition, the access of a large number of electric vehicles increases the complexity and volatility of the grid load. Therefore, the existing centralized parking lot control method, due to the lack of sufficient consideration of the state of electric vehicles (such as factors like parking duration, the number of electric vehicles, etc.), is not conducive to the intelligent control of the parking lot. Summary of the Invention

[0003] The present invention application provides a control method, device and equipment for an electric vehicle parking lot to solve the technical problem of how to improve the intelligent degree of parking lot control.

[0004] To solve the above technical problem, the present invention application provides a control method for an electric vehicle parking lot, which is applied to the power system of a target parking lot. The control method includes:

[0005] Obtain the parking historical data of the target parking lot;

[0006] According to the parking historical data, use a preset vehicle quantity prediction model to obtain a predicted parking quantity value; according to the parking historical data, use a preset parking duration prediction model to obtain a predicted parking duration value;

[0007] According to the predicted parking quantity value and the predicted parking duration value, construct an optimization control model and an optimization objective function;

[0008] According to the optimization objective function and preset constraint conditions, solve the optimization control model, and based on the output of the optimization control model, optimize the control of the charging and discharging of electric vehicles.

[0009] As a preferred solution, the optimization objective function includes:

[0010]

[0011] Where J ei (P ei ) is the optimization objective function, SOC i (tk+1 ) represents the state of charge of the \(i\)-th electric vehicle at time \(t\), k+1 where \(\overline{SOC}\) is the expected average state of charge sequence of the electric vehicle fleet output by the day-ahead energy plan, \(\gamma_1\) is the weight coefficient of the system state, \(\gamma_2\) is the weight coefficient of the control input, and \(P\) ... ei (t k ) represents the charging and discharging power of the \(i\)-th vehicle at time \(t\), k where \(P_{ij}\) refers to the iteration point of the \(j\)-th iteration of the charging and discharging power of the \(i\)-th electric vehicle, and \(E\) ... cost (t k ) is the predicted electricity price at time \(t\), and \(\gamma_3\) is the weight coefficient of the time-of-use electricity price. k ...

[0012] As a preferred solution, the optimization control model includes:

[0013]

[0014] where \(SOC_i(t)\) is the state of charge of the \(i\)-th electric vehicle at time \(t\), \(n_i(t)\) represents the vehicle state coefficient at time \(t\), \(E_i\) is the battery charge capacity of the \(i\)-th electric vehicle, \(\delta_i(t)\) represents the energy loss at time \(t\), \(\Delta t\) is the time variation, and \(\alpha(t)\) is the parking duration coefficient. i (t k )... k ... t (t k )... k ... i ... t (t k )... k ... k )...

[0015] As a preferred solution, the preset constraint conditions include a first constraint condition and a second constraint condition;

[0016] The first constraint condition is:

[0017]

[0018] The second constraint condition is:

[0019]

[0020] where \(SOC_{imin}\) represents the minimum value allowed for the state of charge of the \(i\)-th electric vehicle, \(SOC_{imax}\) represents the maximum value allowed for the state of charge of the \(i\)-th electric vehicle, \(P_{imin}\) represents the minimum value allowed for the charging and discharging power of the \(i\)-th electric vehicle, and \(P_{imax}\) i，min ... i，max ... ei，min ... ei，maxRepresents the maximum value allowed for the charging and discharging power of the \(i\)-th electric vehicle. Refers to the grid power at time \(t\) obtained by optimizing and solving through the objective function \(R(P g (t k ), u(t k ))), where \(u k (t * ) represents the control variable at time \(t\) obtained by optimizing and solving through the objective function \(R(P k (t g ), u(t k )), and \(T\) represents the transpose. k ))), and \(T\) represents the transpose. k As an optimal solution, the first constraint condition is solved by the sequential quadratic programming algorithm, and the solution process is as follows:

[0021] As an optimal solution, the first constraint condition is solved by the sequential quadratic programming algorithm, and the solution process is as follows:

[0022]

[0023] where \(\Delta P ei (t k ) represents the change in \(P ei (t k ), and \(H\), \(C\), \(A\), \(B\), and \(g\) are all parameter variables.

[0024] As an optimal solution, before constructing the optimal control model, the control method further includes:

[0025] Obtaining the real-time number of electric vehicles in the target parking lot, the original value of the charging demand at the current moment, and the remaining charging facility capacity;

[0026] Normalizing the real-time number of electric vehicles, the original value of the charging demand, and the remaining charging facility capacity to obtain the normalized number of electric vehicles, the normalized charging demand, and the normalized remaining charging facility capacity;

[0027] Using the normalization result, solving the following formula to obtain the electricity price prediction value:

[0028] P = \(\beta_0+\beta_1\times N norm +\beta_2\times D norm +\beta_3\times C norm +\varepsilon\);

[0029] where \(\beta_0\) is the intercept term, \(\beta_1\) is the coefficient of the normalized number of electric vehicles \(N norm ), \(\beta_2\) is the coefficient of the normalized charging demand \(D norm ), \(\beta_3\) is the coefficient of the normalized remaining charging facility capacity \(C norm ), \(\varepsilon\) is the error term, and \(P\) is the electricity price prediction value.

[0030] As a preferred solution, the parking historical data includes the actual value of the parking quantity. According to the parking historical data, using a preset vehicle quantity prediction model to obtain the predicted value of the parking quantity, including:

[0031] Obtaining the predicted value of the parking quantity according to the following formula:

[0032]

[0033] where F t is the predicted value of the parking quantity in the t-th period, V t-n+i is the actual value of the parking quantity in the (t - n + i)-th period, n is the total number of periods of the actual value, and ω i is the weight in the i-th period.

[0034] As a preferred solution, according to the parking historical data, using a preset parking duration prediction model to obtain the predicted value of the parking duration, including:

[0035] Obtaining the predicted value of the parking duration according to the following formula:

[0036] T = β0 + β1×θ1 + β2×θ2 + β3×θ3 + β4×θ4 + ε;

[0037] where T is the predicted value of the parking duration, θ1 is the time feature parameter, θ2 is the vehicle type feature parameter, θ3 is the parking lot layout feature parameter, θ4 is the environmental factor feature parameter, ε is the error value, and β0, β1, β2, β3, and β4 are the parameters to be estimated by the least squares method.

[0038] Correspondingly, the present invention application also provides a control device for an electric vehicle parking lot, which is applied to the power system of a target parking lot. The control device includes an acquisition module, a parking information prediction module, a construction module, and an optimization control module; wherein,

[0039] The acquisition module is used to acquire the parking historical data of the target parking lot;

[0040] The parking information prediction module, according to the parking historical data, uses a preset vehicle quantity prediction model to obtain the predicted value of the parking quantity; according to the parking historical data, uses a preset parking duration prediction model to obtain the predicted value of the parking duration;

[0041] The construction module is used to construct an optimization control model and an optimization objective function according to the predicted value of the parking quantity and the predicted value of the parking duration;

[0042] The optimization control module is used to solve the optimization control model according to the optimization objective function and preset constraints, and to optimize the charging and discharging of the electric vehicle based on the output of the optimization control model.

[0043] As a preferred solution, the optimization objective function includes:

[0044]

[0045]

[0046] Among them, J ei (P ei ) is the optimization objective function, SOC i (t k+1 ) represents the number of electric vehicles i at t k+1 The state of charge at the moment, is the average state of charge sequence of the electric vehicle group expected by the day-ahead energy planning output, γ1 is the weight coefficient of the system state, γ2 is the weight coefficient of the control input, P ei (t k ) indicates that the i-th vehicle is at t k The charge and discharge power at each moment, Refers to the jth iteration point of the charging and discharging power of the i-th electric vehicle, E cost (t k ) is t k The electricity price forecast value at the moment, γ3 is the weight coefficient of the time-of-use electricity price.

[0047] As a preferred solution, the optimization control model includes:

[0048]

[0049] Among them, SOC i (t k ) is the number of electric vehicles i at t k The state of charge at the moment, n t (t k ) indicates that at t k The state coefficient of the vehicle at the moment, E i is the battery charge capacity of the i-th electric vehicle, β t (t k ) represents t k The energy loss at the moment, Δt is the time change, α(t k ) is the parking time coefficient.

[0050] As a preferred solution, the preset constraint conditions include a first constraint condition and a second constraint condition;

[0051] The first constraint condition is:

[0052]

[0053] The second constraint condition is:

[0054]

[0055] Among them, SOC i，min represents the minimum value allowed for the state of charge of the i-th electric vehicle, and SOC i，max represents the maximum value allowed for the state of charge of the i-th electric vehicle. P ei，min represents the minimum value allowed for the charging and discharging power of the i-th electric vehicle, and P ei，max represents the maximum value allowed for the charging and discharging power of the i-th electric vehicle. refers to the grid power at time t g (t k ) obtained by optimizing and solving through the objective function R(P k )(t k ), u(t * (t k ))), and u g (t k ) refers to the control variable at time t k obtained by optimizing and solving through the objective function R(P k )(t

[0056] As a preferred solution, the first constraint condition is solved by the sequential quadratic programming algorithm, and the solution process is:

[0057]

[0058]

[0059] Among them, ΔP ei (t k ) represents the change in P ei (t k ), and H, C, A, B, and g are all parameter variables.

[0060] As a preferred solution, the control device further includes an electricity price prediction module, and the electricity price prediction module is used to:

[0061] Obtain the real-time number of electric vehicles in the target parking lot, the original value of the charging demand at the current moment, and the remaining charging facility capacity;

[0062] Normalize the real-time quantity of the electric vehicles, the original value of the charging demand, and the remaining capacity of the charging facilities to obtain the normalized quantity of the electric vehicles, the normalized charging demand, and the normalized remaining capacity of the charging facilities;

[0063] Use the normalization result to solve the following formula to obtain the predicted electricity price value:

[0064] P = β0 + β1×N norm + β2×D norm + β3×C norm + ε;

[0065] where β0 is the intercept term, β1 is the coefficient of the normalized quantity of the electric vehicles N norm , β2 is the coefficient of the normalized charging demand D norm , β3 is the coefficient of the normalized remaining capacity of the charging facilities C norm , ε is the error term, and P is the predicted electricity price value.

[0066] As a preferred solution, the parking historical data includes the actual value of the parking quantity. The parking information prediction module uses a preset vehicle quantity prediction model based on the parking historical data to obtain the predicted parking quantity value, including:

[0067] The parking information prediction module obtains the predicted parking quantity value according to the following formula:

[0068]

[0069] where F t is the predicted parking quantity value in the t-th period, V t-n+i is the actual parking quantity value in the (t - n + i)-th period, n is the total number of periods of the actual value, and ω i is the weight in the i-th period.

[0070] As a preferred solution, the parking information prediction module uses a preset parking duration prediction model based on the parking historical data to obtain the predicted parking duration value, including:

[0071] The parking information prediction module obtains the predicted parking duration value according to the following formula:

[0072] T = β0 + β1×θ1 + β2×θ2 + β3×θ3 + β4×θ4 + ε;

[0073] where T is the predicted parking duration value, θ1 is the time feature parameter, θ2 is the vehicle type feature parameter, θ3 is the parking lot layout feature parameter, θ4 is the environmental factor feature parameter, ε is the error value, and β0, β1, β2, β3, and β4 are the parameters to be estimated by the least squares method.

[0074] Correspondingly, the present invention application further provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the control method of the electric vehicle parking lot is implemented.

[0075] Correspondingly, the present invention application further provides a computer-readable storage medium. The computer-readable storage medium includes a stored computer program. When the computer program runs, the device where the computer-readable storage medium is located is controlled to execute the control method of the electric vehicle parking lot.

[0076] Compared with the prior art, the present invention application has the following beneficial effects:

[0077] The present invention application provides a control method, device, equipment, and medium for an electric vehicle parking lot, which is applied to the power system of a target parking lot. The control method includes: obtaining the parking historical data of the target parking lot; according to the parking historical data, using a preset vehicle quantity prediction model to obtain a predicted parking quantity value; according to the parking historical data, using a preset parking duration prediction model to obtain a predicted parking duration value; according to the predicted parking quantity value and the predicted parking duration value, constructing an optimization control model and an optimization objective function; according to the optimization objective function and preset constraint conditions, solving the optimization control model, and based on the output of the optimization control model, optimizing the charging and discharging of electric vehicles. The present invention application predicts the vehicle quantity and parking duration based on the parking historical data of the target parking lot, constructs an optimization model and an optimization objective function; and then solves the optimization control model based on the optimization objective function and preset constraint conditions, which can consider relevant factors such as the state of electric vehicles, such as parking duration and the number of electric vehicles, during the control process of the target parking lot, improve the intelligent level of parking lot control, and improve the control efficiency of the charging and discharging of electric vehicles in the parking lot. Description of the Drawings

[0078] Figure 1 It is a flowchart of an embodiment of the control method for an electric vehicle parking lot provided by the present invention application.

[0079] Figure 2 It is an architecture diagram of an embodiment of the intelligent Internet of Things parking lot management system provided by the present invention application.

[0080] Figure 3 It is a structural diagram of an embodiment of the control device for an electric vehicle parking lot provided by the present invention application. Detailed Embodiment

[0081] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0082] Embodiment 1

[0083] Please refer to Figure 1 , Figure 1 which is a control method for an electric vehicle parking lot provided by the present invention application, including steps S101 to S104. Each step is described in detail as follows:

[0084] Step S101, obtaining the parking historical data of the target parking lot.

[0085] The control method of the electric vehicle parking lot in this embodiment is applied to the power system of the target parking lot. The power system includes charging facilities of the target parking lot, etc., and the charging facilities are used for charging electric vehicles.

[0086] In addition, the power system of the above-mentioned target parking lot can be connected to a data management system for obtaining the results of data processing. The data management system can be a cloud or a server, which has sufficient performance required for data processing.

[0087] Furthermore, the control method of the electric vehicle parking lot in this embodiment is applied to the central control module of the power system of the target parking lot. The central control module can be set in a computer device, and the computer device includes but is not limited to smart phones, laptop computers, tablet computers, desktop computers, as well as physical servers and cloud servers, etc.

[0088] Furthermore, the target parking lot in this embodiment can adopt a smart Internet of Things parking lot management system, and its architecture is as Figure 2 shown. The management system can include an electric vehicle charging pile system and a wireless data base station. The above-mentioned electric vehicle charging pile system includes a charging pile controller, external devices and sensors, aiming to achieve precise control of the charging process, monitoring of the operating state, and real-time collection of environmental parameters.

[0089] After preprocessing the collected data, the system transmits it to the Internet of Things parking lot management system through the wireless data base station. As a communication bridge between the charging pile and the parking lot management system, the wireless data base station adopts NB-IoT (NarrowBand Internet of Things) high-speed communication technology to ensure the efficiency and anti-interference ability of data transmission.

[0090] In this embodiment, the Internet of Things parking lot management system is the center of the entire intelligent charging pile control system, responsible for receiving, storing, managing, and analyzing data information from the charging piles. Through data mining and analysis technologies, historical data such as the parking duration, charging time, battery capacity, and state of charge of electric vehicles are collected, and data such as the number of electric vehicles in the parking lot, parking duration, and charging price are predicted.

[0091] The target parking lot in this embodiment includes intelligent ground lock sensors for parking spaces, which are used to detect whether the parking space is occupied and the entry and exit times of vehicles. When a vehicle enters or leaves a parking space, the intelligent ground lock sensor will send corresponding signals to the central control module.

[0092] The target parking lot also includes environmental monitoring sensors, such as temperature sensors, humidity sensors, etc., which are used to monitor the environmental conditions in the parking lot because environmental factors may affect the battery performance of electric vehicles.

[0093] In addition, the electric vehicles in this embodiment can be equipped with intelligent in-vehicle terminals, which have the function of communicating with the parking lot management system and can upload relevant vehicle information in real time, including the number of vehicles, charging time, state of charge of the battery, etc.

[0094] In addition, when the electric vehicle connects to the charging pile and starts charging, the intelligent in-vehicle terminal will send a charging start signal to the parking lot management system and record the charging start time. During the charging process, the intelligent in-vehicle terminal will monitor the charging status in real time. When the charging is completed, it will send a charging end signal to the management system again and record the charging end time. By calculating the difference between the charging end time and the charging start time, the charging time of the vehicle can be obtained.

[0095] The electric vehicle described in this embodiment includes a Battery Management System (BMS for short), which can obtain the remaining power information of the vehicle battery in real time. When the vehicle enters the parking lot, the intelligent in-vehicle terminal will upload the current remaining battery power information obtained from the battery management system to the parking lot management system. At the same time, during the charging process, the intelligent in-vehicle terminal will continuously monitor the change of the battery power and upload the updated battery capacity data to the management system in real time.

[0096] The above state of charge mainly refers to the remaining power of the electric vehicle battery, or the relationship between the remaining power and the rated capacity. The intelligent in-vehicle terminal can communicate with the vehicle's battery management system to obtain the SOC value (State of Charge, which refers to the ratio of the remaining battery power to the rated capacity in this embodiment) of the battery in real time and upload it to the parking lot management system. The common formula for the state of charge (SOC) is as follows:

[0097] SOC=(QS / Q N ) × 100%;

[0098] Among them, Q S refers to the remaining power of the electric vehicle battery, and Q N refers to the rated capacity of the electric vehicle battery.

[0099] In this embodiment, the parking historical data includes, but is not limited to, parking quantity historical data, parking duration historical data, charging demand historical data, charging facility historical data, etc.

[0100] The parking historical data is used to analyze and predict the prediction data of the target parking lot, such as the parking duration, parking quantity, etc. at a certain future moment.

[0101] Step S102, according to the parking historical data, using a preset vehicle quantity prediction model, obtain a parking quantity prediction value; according to the parking historical data, using a preset parking duration prediction model, obtain a parking duration prediction value.

[0102] In a preferred implementation manner, the above vehicle quantity prediction model can be implemented by the following formula:

[0103] Obtain the parking quantity prediction value according to the following formula:

[0104]

[0105] Among them, F t is the parking quantity prediction value in the t-th period, V t-n+i is the actual value of the parking quantity in the t - n + i-th period, n is the total number of periods of the actual value, and ω i is the weight in the i-th period.

[0106] In this embodiment, according to the number of electric vehicles entering the parking lot in different time periods in the historical data, the weighted moving average method is used to predict the number of electric vehicles in the parking lot. This method gives different weights to the actual values in different periods, and the weights of the more recent data are larger, which can better reflect the recent change trend. For example, during the morning rush hour on weekdays, the weight of the electric vehicle quantity data in the same period of the previous day can be set to 0.7, and the weight of the data in the same period two days ago can be set to 0.3.

[0107] The division basis for the first period, second period, third period, …, i-th period of this example can be chronological order. Specifically, it can be divided according to the historical data of the number of electric vehicles entering the parking lot in different time periods. Taking the morning rush hour on weekdays as an example, the same period of the previous day is set as the relatively recent period (with a larger weight, such as 0.7), which can be regarded as a certain period relatively close to the current prediction period; the same period of the day before yesterday is set as the second recent period (with a smaller weight, such as 0.3), which can be regarded as a period farther from the same period of the previous day. Therefore, the division of the number of periods can be carried out in chronological order according to the distance from the current prediction period, and in the weighted moving average method, the closer the data period is to the current prediction period, the greater the weight assigned.

[0108] In a preferred implementation manner, obtaining the predicted parking duration value by using a preset parking duration prediction model according to the parking historical data includes:

[0109] Obtaining the predicted parking duration value according to the following formula:

[0110] T = β0 + β1×θ1 + β2×θ2 + β3×θ3 + β4×θ4 + ε;

[0111] Wherein, T is the predicted parking duration value, θ1 is the time feature parameter, θ2 is the vehicle type feature parameter, θ3 is the parking lot layout feature parameter, θ4 is the environmental factor feature parameter, ε is the error value, and β0, β1, β2, βx and β4 are the parameters to be estimated by the least squares method.

[0112] This implementation manner can analyze the average parking duration of different types of vehicles (such as private cars, taxis, etc.) in different time periods and different regions based on historical parking data statistics. At the same time, consider the impact of the charging demand of the vehicle on the parking duration. Use a linear regression model to predict the parking duration of electric vehicles, assuming that there is a linear relationship between the parking duration and each feature, and estimate the parameters of the model by methods such as the least squares method mentioned above.

[0113] This implementation manner collects a large amount of historical parking data, including parking duration data under different time periods, different vehicle types (private cars, taxis, etc.), different parking lot layouts (such as the number of parking spaces, distribution of charging piles, etc.) and different environmental factors (such as weather, surrounding facilities, etc.). Use methods such as the least squares method to estimate the parameters β0, β1, β2, β3, β4 in the formula. Specifically, the goal of the least squares method is to find a set of parameter values that minimize the sum of the squares of the errors between the parking duration predicted by the model and the actual parking duration.

[0114] After estimating the parameter values, for newly emerging situations, substitute the corresponding time feature parameter θ1, vehicle type feature parameter θ2, parking lot layout feature parameter θ3, and environmental factor feature parameter θ4 into formula T. The calculated T value is the predicted parking duration. Since ε is the error value, in actual prediction, it is generally ignored or its influence range is estimated through statistical methods.

[0115] It can be understood that although the number of electric vehicles can also be deduced through the vehicle quantity prediction model, the charging requirements and parking purposes of each electric vehicle are different. For example, some vehicle owners may only park temporarily for a few minutes and do not need to charge; while some vehicle owners may need to park for a long time and charge, and the charging duration will also vary due to factors such as the remaining battery power of the vehicle and the power of the charging pile. Therefore, even if the vehicle quantity is determined, the parking duration of each vehicle has great uncertainty. For the parking lot manager, accurately predicting the parking duration helps to better plan the parking lot resources, such as reasonably arranging the use of charging piles and planning the allocation of parking spaces. Therefore, through the above vehicle duration prediction model in this embodiment, combined with the prediction of the parking duration, it is beneficial for the target parking lot to make more reasonable decisions.

[0116] In a preferred embodiment, before constructing the optimization control model in step S102, the control method further includes:

[0117] Obtain the real-time number of electric vehicles in the target parking lot, the original value of the charging demand at the current moment, and the remaining capacity of the charging facilities;

[0118] Perform normalization processing on the real-time number of electric vehicles, the original value of the charging demand, and the remaining capacity of the charging facilities to obtain the normalized number of electric vehicles, the normalized charging demand, and the normalized remaining capacity of the charging facilities;

[0119] Use the normalization result to solve the following formula to obtain the predicted electricity price value:

[0120] P = β0 + β1×N norm +β2×D norm +β3×C norm +ε;

[0121] where β0 is the intercept term, β1 is the coefficient of the normalized number of electric vehicles N norm β2 is the coefficient of the normalized charging demand D norm β3 is the coefficient of the normalized remaining capacity of the charging facilities C norm ε is the error term, and P is the predicted electricity price value.

[0122] In this embodiment, the predicted electricity price value, which is the target variable to be predicted in this implementation manner, is P. In this implementation manner, the real-time number N of electric vehicles in the target parking lot is normalized. For example, the normalized number of electric vehicles after processing has a value range between zero and one. N min is the minimum value of the number of electric vehicles in the parking lot in historical data, and N max is the maximum value of the number of electric vehicles in the parking lot in historical data.

[0123] Let the original value of the charging demand at the current moment be D, which can be measured by counting the proportion of vehicles with charging demand in the total number of vehicles within a certain period of time. Similarly, it can also be normalized so that its value is between zero and one. For example:

[0124] The normalized charging demand where D min and D max are respectively the minimum and maximum values of the number of vehicles with charging demand in the target parking lot or a certain parking lot every day in the statistics of the past year.

[0125] Let the remaining charging facility capacity be C, which can be represented by the ratio of the number of remaining available charging piles to the total number of charging piles, and it is also normalized so that its value is between zero and one. For example:

[0126] The normalized remaining charging facility capacity is where C min and C max are respectively the minimum and maximum values of the remaining charging facility capacity during the statistical period.

[0127] Then, the electricity price prediction value can be obtained by solving according to the above formula P = β0 + β1×N norm + β2×D norm + β3×C norm + ε.

[0128] Among them, β0 is the intercept term, representing the basic charging price when the number of electric vehicles, the charging demand, and the remaining charging facility capacity are all in a certain benchmark state (for example, the normalized values are all 0). It reflects other cost factors besides these three main factors, such as site rental costs, equipment maintenance costs, etc.

[0129] β1 is the electric vehicle quantity variable N normThe coefficient reflects the impact degree of the number of electric vehicles on the charging price. If β1>0, it indicates that as the number of electric vehicles in the parking lot increases (the value increases after normalization), the charging price will rise; conversely, if β1<0, it means that the price will decrease when the number increases (but according to the actual situation, generally β1>0 is more reasonable).

[0130] β2 is the coefficient of the charging demand D norm which reflects the impact degree of the charging demand on the charging price. When β2>0, as the charging demand increases (the value increases after normalization), the charging price will rise, which conforms to the market rule that the greater the demand, the higher the price.

[0131] β3 is the coefficient of the remaining charging facility capacity variable C norm which reflects the impact degree of the remaining charging facility capacity on the charging price. Generally, β3<0, that is, the greater the remaining charging facility capacity (the value increases after normalization), the lower the charging price, because there are more resources available and the competition is relatively small.

[0132] ε is the error term, representing the impact of other factors not considered in the model on the charging price, such as sudden policy adjustments, price changes of surrounding competitors, etc. In practical applications, by collecting a large amount of historical data and using methods such as the least squares method, the values of β0, β1, β2, and β3 can be estimated to minimize the sum of the squares of the errors between the charging price predicted by the model and the actual price, thereby improving the accuracy of the model.

[0133] Step S103: Construct an optimization control model and an optimization objective function according to the predicted parking quantity value and the predicted parking duration value.

[0134] In a preferred embodiment, the optimization objective function includes:

[0135]

[0136] where J ei (P ei ) is the optimization objective function, SOC i (t k+1 ) represents the state of charge of the i-th electric vehicle at time t k+1 , is the expected average state of charge sequence of the electric vehicle fleet output by the day-ahead energy plan, γ1 is the weight coefficient of the system state, γ2 is the weight coefficient of the control input, P ei (t k ) represents the charging and discharging power of the i-th vehicle at time t k , refers to the iteration point of the j-th iteration of the charging and discharging power of the i-th electric vehicle, E cost(t k ) is t k The electricity price forecast value at the moment, γ3 is the weight coefficient of the time-of-use electricity price.

[0137] In a preferred embodiment, the optimization control model includes:

[0138]

[0139] Among them, SOC i (t k ) is the number of electric vehicles i at t k The state of charge at the moment, n t (t k ) indicates that at t k The state coefficient of the vehicle at the moment (for example, the identification coefficient of whether the vehicle is connected to the grid, when the vehicle is connected to the grid and ready to charge and discharge, n t (t k )=1, when not connected n t (t k )=0), E i is the battery charge capacity of the i-th electric vehicle, δ t (t k ) represents t k The energy loss at the moment, Δt is the time change, α(t k ) is the parking time coefficient.

[0140] In a preferred embodiment, the preset constraint condition includes a first constraint condition and a second constraint condition;

[0141] The first constraint condition is:

[0142]

[0143] The second constraint is:

[0144]

[0145] Among them, SOC i，min Indicates the minimum value allowed for the state of charge of the i-th electric vehicle, SOC i，max P represents the maximum value allowed by the state of charge of the i-th electric vehicle. ei，min It represents the minimum value allowed for the charging and discharging power of the i-th electric vehicle, P ei，max represents the maximum value allowed for the charging and discharging power of the i-th electric vehicle, Refers to the objective function R(P g (t k ),u(t k )) is optimized and solved to obtain the kThe optimal value of the grid power at a certain moment, which is determined in the day-ahead electricity trading plan and is used to guide the charging and discharging control of individual vehicles in the subsequent process to achieve the optimal operation of the entire system, u * (t k ) refers to the optimal value of the control variable at time t obtained by optimizing and solving through the objective function R(P g (t k ), u(t k ). It, together with k , constitutes the optimal operation parameter combination determined in the day-ahead planning stage, which plays a role in restricting and guiding the real-time charging and discharging control of individual vehicles. T represents the transpose, and the function argmaxR(x, y) represents the values of variables x and y when R(x, y) takes the maximum value.

[0146] Step S104: Solve the optimization control model according to the optimization objective function and preset constraint conditions, and optimize the charging and discharging of the electric vehicle based on the output of the optimization control model.

[0147] It can be understood that the above first constraint condition is a nonlinear real-time optimization control problem with constraints, and the solution of the above problem can be achieved by adopting the sequential quadratic programming algorithm. Sequential quadratic programming is one of the effective methods for solving nonlinear optimization problems with constraints, and its specific solution process is described as follows:

[0148]

[0149]

[0150] Among them, ΔP ei (t k ) represents the change in P ei (t k ), and H, C, A, B, and g are all parameter variables. g(P ei ) is a vector function.

[0151] Among them, j represents the number of iterations. The sequential quadratic programming algorithm gradually approaches the optimal solution through continuous iteration. represents the iteration point of the charging and discharging power of the i-th vehicle at time t in the j-th iteration. As the number of iterations j increases, k is continuously updated and gradually approaches the optimal solution that satisfies the constraint conditions of the optimization problem.

[0152] is the second derivative of the objective function J ei (P ei The second derivative of with respect to P ei (at the iteration point ​​evaluated at a certain point), which reflects the curvature information of the objective function and is used to determine the coefficient of the quadratic term in quadratic programming problems; is the objective function J ei (P ei ) with respect to P ei of the first-order derivative (evaluated at the iteration point ), which represents the gradient direction of the objective function at that point; is the constraint function g(P ei ) with respect to P ei of the first-order derivative (evaluated at the iteration point ), which is used to describe the rate of change of the constraint condition; is to take the negative value of the constraint function g(P ei ) (evaluated at the iteration point ), which plays a certain role in the constraint conditions of quadratic programming. Through these definitions and operations, the original real-time optimization control problem is transformed into a quadratic programming problem, and then solved iteratively by the sequential quadratic programming algorithm.

[0153] In this embodiment, j can be initialized to 0, and an initial value is selected. Secondly, the parameters and are calculated respectively. Solve the quadratic programming problem to obtain Set Finally, judge the termination. If , the iteration ends, where ε>0 represents a preset numerical value indicating the precision. Otherwise, take j = j + 1 and return to the step of calculating the parameters.

[0154] This embodiment solves the first constraint condition by applying the Sequential Quadratic Programming (abbreviated as SQP) algorithm, and then combines the second constraint condition and the optimization objective function to solve the optimization control model, and obtains the control scheme for the electric vehicle based on the output of the optimization control model. It effectively solves the real-time optimization control problem during the charging and discharging process of electric vehicles, enables the intelligent Internet of Things parking lot management system to achieve orderly charging and discharging of vehicles in the distributed charging and discharging control parking lot scenario, and significantly improves the charging efficiency and grid stability.

[0155] Correspondingly, as Figure 3 shown, the present invention application also provides a control device 300 for an electric vehicle parking lot, which is applied to the power system of the target parking lot. The control device 300 includes an acquisition module 301, a parking information prediction module 302, a construction module 303, and an optimization control module 304; wherein,

[0156] The acquisition module 301 is configured to acquire the parking historical data of the target parking lot;

[0157] The parking information prediction module 302 obtains a predicted parking quantity value by using a preset vehicle quantity prediction model according to the parking historical data; and obtains a predicted parking duration value by using a preset parking duration prediction model according to the parking historical data;

[0158] The construction module 303 is configured to construct an optimal control model and an optimal objective function according to the predicted parking quantity value and the predicted parking duration value;

[0159] The optimal control module 304 is configured to solve the optimal control model according to the optimal objective function and preset constraint conditions, and perform optimal control on the charging and discharging of the electric vehicle based on the output of the optimal control model.

[0160] As a preferred solution, the optimal objective function includes:

[0161]

[0162] Where J ei (P ei ) is the optimal objective function, SOC i (t k+1 ) represents the state of charge of the i-th electric vehicle at time t k+1 , is the expected average state of charge sequence of the electric vehicle fleet output by the day-ahead energy plan, γ1 is the weight coefficient of the system state, γ2 is the weight coefficient of the control input, P ei (t k ) represents the charging and discharging power of the i-th vehicle at time t k , refers to the iteration point of the j-th iteration of the charging and discharging power of the i-th electric vehicle, E cost (t k ) is the predicted electricity price value at time t k , and γ3 is the weight coefficient of the time-of-use electricity price.

[0163] As a preferred solution, the optimal control model includes:

[0164]

[0165] Where SOC i (t k ) is the state of charge of the i-th electric vehicle at time t k , n t (t k ) represents the vehicle state coefficient at time t k , Ei is the battery charge capacity of the i-th electric vehicle, δ t (t k ) represents the energy loss at time t, Δt is the time variation, α(t k ) is the parking duration coefficient. k ) is the parking duration coefficient.

[0166] As a preferred solution, the preset constraint conditions include a first constraint condition and a second constraint condition;

[0167] The first constraint condition is:

[0168]

[0169] The second constraint condition is:

[0170]

[0171] Among them, SOC i，min represents the minimum value allowed for the state of charge of the i-th electric vehicle, SOC i，max represents the maximum value allowed for the state of charge of the i-th electric vehicle, P ei，min represents the minimum value allowed for the charging and discharging power of the i-th electric vehicle, P ei，max represents the maximum value allowed for the charging and discharging power of the i-th electric vehicle, refers to the grid power at time t obtained by optimizing and solving through the objective function R(P g (t k ), u(t k ))), u k is the control variable at time t obtained by optimizing and solving through the objective function R(P * (t k ), u(t g (t k ))), and T represents the transpose. k ))), and T represents the transpose. k ))), and T represents the transpose.

[0172] As a preferred solution, the first constraint condition is solved by the sequential quadratic programming algorithm, and the solution process is:

[0173]

[0174] Among them, ΔP ei (t k ) represents the change in P ei (t k ), and H, C, A, B, and g are all parameter variables.

[0175] As a preferred solution, the control device 300 further includes an electricity price prediction module, and the electricity price prediction module is configured to, before the construction module 303 constructs an optimal control model:

[0176] Obtain the real-time number of electric vehicles in the target parking lot, the original value of the charging demand at the current moment, and the remaining capacity of the charging facilities;

[0177] Perform normalization processing on the real-time number of electric vehicles, the original value of the charging demand, and the remaining capacity of the charging facilities to obtain the normalized number of electric vehicles, the normalized charging demand, and the normalized remaining capacity of the charging facilities;

[0178] Use the normalization result to solve the following formula to obtain the electricity price prediction value:

[0179] P = β0 + β1 × N norm + β2 × D norm + β3 × C norm + ε;

[0180] where β0 is the intercept term, β1 is the coefficient of the normalized number of electric vehicles N norm β2 is the coefficient of the normalized charging demand D norm β3 is the coefficient of the normalized remaining capacity of the charging facilities C norm ε is the error term, and P is the electricity price prediction value.

[0181] As a preferred solution, the parking historical data includes the actual value of the parking quantity. The parking information prediction module 302 obtains a parking quantity prediction value according to the parking historical data by using a preset vehicle quantity prediction model, including:

[0182] The parking information prediction module 302 obtains the parking quantity prediction value according to the following formula:

[0183]

[0184] where F t is the parking quantity prediction value in the t-th period, V t-n+i is the actual value of the parking quantity in the t - n + i-th period, n is the total number of periods of the actual value, and ω i is the weight in the i-th period.

[0185] As a preferred solution, the parking information prediction module 302 obtains a parking duration prediction value according to the parking historical data by using a preset parking duration prediction model, including:

[0186] The parking information prediction module 302 obtains the parking duration prediction value according to the following formula:

[0187] T = β0 + β1×θ1 + β2×θ2 + β3×θ3 + β4×θ4 + ε;

[0188] Wherein, T is the predicted value of the parking duration, θ1 is the time feature parameter, θ2 is the vehicle type feature parameter, θ3 is the parking lot layout feature parameter, θ4 is the environmental factor feature parameter, ε is the error value, and β0, β1, β2, β3, and β4 are the parameters to be estimated by the least squares method.

[0189] Correspondingly, the present invention application further provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the control method of the electric vehicle parking lot is implemented.

[0190] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the terminal, connecting various parts of the entire terminal through various interfaces and lines.

[0191] The memory can be used to store the computer program. By running or executing the computer program stored in the memory and calling the data stored in the memory, the processor realizes various functions of the terminal. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0192] Correspondingly, the present invention application further provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the control method of the electric vehicle parking lot.

[0193] Among them, if the modules integrated in the control device / terminal of the electric vehicle parking lot are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0194] Compared with the prior art, the present invention application has the following beneficial effects:

[0195] The present invention application provides a control method, device, equipment and medium for an electric vehicle parking lot, which is applied to the power system of a target parking lot. The control method includes: obtaining the parking history data of the target parking lot; according to the parking history data, using a preset vehicle quantity prediction model to obtain a predicted parking quantity value; according to the parking history data, using a preset parking duration prediction model to obtain a predicted parking duration value; according to the predicted parking quantity value and the predicted parking duration value, constructing an optimization control model and an optimization objective function; according to the optimization objective function and preset constraint conditions, solving the optimization control model, and based on the output of the optimization control model, optimizing the charging and discharging of electric vehicles. The present invention application predicts the vehicle quantity and parking duration based on the parking history data of the target parking lot, constructs an optimization model and an optimization objective function; and then solves the optimization control model based on the optimization objective function and preset constraint conditions, which can consider relevant factors such as the state of electric vehicles, such as parking duration and the number of electric vehicles, during the control process of the target parking lot, improve the intelligent level of parking lot control, and improve the control efficiency of the charging and discharging of electric vehicles in the parking lot.

[0196] The specific embodiments described above further elaborate on the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is only for the specific embodiments of the present invention and is not intended to limit the protection scope of the present invention. In particular, it is pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A control method for an electric vehicle parking lot, characterized in that, A power system applied to a target parking lot, the control method includes: Obtain the parking historical data of the target parking lot; According to the parking historical data, use a preset vehicle quantity prediction model to obtain a predicted parking quantity value; according to the parking historical data, use a preset parking duration prediction model to obtain a predicted parking duration value; According to the predicted parking quantity value and the predicted parking duration value, construct an optimal control model and an optimal objective function; According to the optimal objective function and preset constraint conditions, solve the optimal control model, and based on the output of the optimal control model, optimize the charging and discharging of electric vehicles.

2. The control method of an electric vehicle parking lot according to claim 1, characterized in that, The optimal objective function includes: Among them, J ei (P ei ) is the said optimization objective function, SOC i (t k+1 ) represents the state of charge of the i-th electric vehicle at time t k+1 . is the expected average state of charge sequence of the electric vehicle fleet output by the day-ahead energy plan, γ1 is the weight coefficient of the system state, γ2 is the weight coefficient of the control input, P ei (t k ) represents the charging and discharging power of the i-th vehicle at time t k . refers to the iteration point of the j-th iteration of the charging and discharging power of the i-th electric vehicle, E cost (t k ) is the predicted electricity price value at time t k , and γ3 is the weight coefficient of the time-of-use electricity price.

3. The control method of an electric vehicle parking lot according to claim 2, characterized in that, The optimal control model includes: Among them, SOC i (t k ) is the state of charge of the i-th electric vehicle at time t k , n t (t k ) represents the vehicle state coefficient at time t k , E i is the battery charge capacity of the i-th electric vehicle, δ t (t k ) represents the energy loss at time t k , Δt is the time change, and α(t k ) is the parking duration coefficient.

4. The control method of an electric vehicle parking lot according to claim 3, characterized in that, The preset constraint conditions include a first constraint condition and a second constraint condition; The first constraint condition is: The second constraint condition is: Among them, SOC i,min represents the minimum value allowed for the state of charge of the i-th electric vehicle, SOC i,max represents the maximum value allowed for the state of charge of the i-th electric vehicle, P ei,min represents the minimum value allowed for the charging and discharging power of the i-th electric vehicle, P ei,max represents the maximum value allowed for the charging and discharging power of the i-th electric vehicle, refers to the grid power at time t g (t k ), u(t k )) obtained by optimizing and solving through the objective function R(P k ), u * (t k ) refers to the control variable at time t g (t k ), u(t k )) obtained by optimizing and solving through the objective function R(P k ), and T represents the transpose.

5. The control method of an electric vehicle parking lot according to claim 4, characterized in that, The first constraint condition is solved by a sequential quadratic programming algorithm, and the solution process is: where, ΔP ei (t k ) represents the change in P ei (t k ), and H, C, A, B, and g are all parameter variables.

6. The control method of an electric vehicle parking lot according to claim 2, characterized in that, Before constructing the optimal control model, the control method further includes: Obtain the real-time quantity of electric vehicles in the target parking lot, the original value of the charging demand at the current moment, and the remaining charging facility capacity; Perform normalization processing on the real-time quantity of electric vehicles, the original value of the charging demand, and the remaining charging facility capacity to obtain the normalized quantity of electric vehicles, the normalized charging demand, and the normalized remaining charging facility capacity; Use the normalization result to solve the following formula to obtain the predicted electricity price value: P = β0 + β1×N norm + β2×D norm + β3×C norm + ε; Among them, β0 is the intercept term, β1 is the coefficient of the number of electric vehicles N after normalization processing norm β2 is the degree of the charging demand D after normalization processing norm β3 is the coefficient of the remaining charging facility capacity C after normalization processing norm ε is the error term, and P is the predicted electricity price value 7. The control method of an electric vehicle parking lot according to claim 1, characterized in that, The parking historical data includes the actual parking quantity value. According to the parking historical data, using a preset vehicle quantity prediction model to obtain a predicted parking quantity value includes: Obtain the predicted parking quantity value according to the following formula: Among them, F t is the predicted value of the parking quantity in the t-th period, V t-n+i is the actual value of the parking quantity in the (t - n + i)-th period, n is the total number of periods of the actual value, ω i is the weight in the i-th period.

8. The control method of an electric vehicle parking lot according to claim 1, characterized in that According to the parking historical data, using a preset parking duration prediction model to obtain a predicted parking duration value includes: Obtain the predicted parking duration value according to the following formula: T = β0 + β1×θ1 + β2×θ2 + β3×θ3 + β4×θ4 + ε; Wherein, T is the predicted parking duration value, θ1 is a time feature parameter, θ2 is a vehicle type feature parameter, θ3 is a parking lot layout feature parameter, θ4 is an environmental factor feature parameter, ε is an error value, and β0, β1, β2, β3, and β4 are parameters to be estimated by the least squares method.

9. A control device for an electric vehicle parking lot, characterized in that, A control device applied to a power system of a target parking lot, the control device includes an acquisition module, a parking information prediction module, a construction module, and an optimal control module; wherein, The acquisition module is used to obtain the parking historical data of the target parking lot; The parking information prediction module, according to the parking historical data, uses a preset vehicle quantity prediction model to obtain a predicted parking quantity value; according to the parking historical data, uses a preset parking duration prediction model to obtain a predicted parking duration value; The construction module is used to construct an optimal control model and an optimal objective function according to the predicted parking quantity value and the predicted parking duration value; The optimization control module is used to solve the optimization control model according to the optimization objective function and preset constraint conditions, and optimize the charging and discharging of the electric vehicle based on the output of the optimization control model.

10. A terminal device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the control method of the electric vehicle parking lot according to any one of claims 1 to 8.

Images