New energy vehicle wireless charging and ETC non-inductive payment device

Abstract

The invention relates to the technical field of Internet of Things intelligent equipment, in particular to a new energy vehicle wireless charging and ETC non-inductive payment device. The system comprises an information identification module, a power grid monitoring module, a charging control module and a bill generation module. Identification and authentication of the new energy vehicle are completed through the information identification module, wireless charging is carried out on the new energy vehicle through the electromagnetic resonance technology, after charging is finished, a bill is generated through the bill generation module, the vehicle-mounted OBU is verified again through ETC, matching with the bill is carried out, payment and fee deduction are completed, non-sensitiveness of charging and payment of the new energy vehicle is achieved, and meanwhile the charging efficiency is improved. A neural network model for judging the power supply capacity of the power grid is established by using the power grid monitoring module, the output power of the power grid is dynamically adjusted according to a proportion adjustment strategy between the load rate and the output power of the power grid, the charging demand of a user is considered, an optimization strategy of the wireless charging power of the new energy vehicle is formulated, and the wireless charging efficiency of the new energy vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things intelligent devices, and in particular to a new energy vehicle wireless charging and ETC contactless payment device. Background Art

[0002] The charging of new energy vehicles is mainly done by wired charging with charging piles. While the charging piles take up a certain space, users are required to manually insert the gun, scan the code to turn it on, remove the gun after completion, and make mobile payments. The overall operation process is relatively cumbersome. The integration of wireless charging technology and ETC payment technology can effectively solve this problem.

[0003] Wireless charging technology is mainly based on electromagnetic resonance, which transmits energy through the resonance phenomenon of the electromagnetic field. Electromagnetic resonance wireless charging technology can achieve higher charging efficiency and longer transmission distance. It is mainly composed of a power supply, a transmitting panel, a vehicle-mounted receiving panel and a control unit. When the resonant frequency of the transmitting end of the power supply is consistent with the resonant frequency of the receiving end of the car, the battery can be charged through the resonance effect.

[0004] At present, when new energy vehicles are wirelessly charged, there are too many power-consuming devices, which leads to a decrease in the power supply capacity of the power grid, and is unable to meet the electricity needs of nearby residents and industrial and commercial enterprises. In order to dynamically adjust the wireless charging power of new energy vehicles according to the power supply capacity of the power grid, improve the wireless charging efficiency of new energy vehicles, give priority to ensuring the electricity needs of residents and industrial and commercial enterprises, and at the same time, use ETC to realize contactless payment and reduce the operation process. Therefore, we propose a new energy vehicle wireless charging and ETC contactless payment device. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that when new energy vehicles are wirelessly charged, the charging power of the new energy vehicles cannot be dynamically adjusted according to the power supply capacity of the power grid, and to comprehensively consider user needs, improve wireless charging efficiency, and reasonably distribute power grid energy.

[0006] To achieve the above-mentioned purpose, the present invention provides a new energy vehicle wireless charging and ETC non-contact payment device, including an information recognition module, a power grid monitoring module, a charging control module and a bill generation module;

[0007] The information identification module uses ETC to turn on the radio frequency scanning vehicle OBU to identify vehicle information and complete vehicle authentication. The user sets the charging amount or charging time through the mobile phone APP, and transmits the information and vehicle battery status information to the control charging module;

[0008] The power grid monitoring module uses a smart meter to monitor the voltage data, current data and frequency data in the power grid in real time, and uses the voltage, current and frequency in the power grid as input layer nodes and the power supply capacity of the power grid as output layer nodes to construct a neural network model for judging the power supply capacity of the power grid, uses historical data to train the neural network model, and transmits the power supply capacity data of the power grid to the control charging module based on the power supply capacity of the power grid judged by the voltage data, current data and frequency data in the power grid monitored in real time;

[0009] The control charging module receives the power supply capacity data of the power grid transmitted by the power grid monitoring module, takes the power grid load rate as an indicator, and dynamically adjusts the output power according to the ratio adjustment strategy. At the same time, it receives the charging amount or charging time data set by the user and the vehicle battery status data transmitted by the information identification module, and comprehensively analyzes the received data. Using the genetic algorithm, under the power supply capacity of the power grid, considering the charging needs of the user, an optimization strategy for the wireless charging power of the new energy vehicle is formulated. According to the optimization strategy, the wireless charging power and time are counted, and transmitted to the bill generation module;

[0010] The bill generation module generates a charging bill based on the statistical wireless charging power and duration, and sends the bill to the information identification module, verifies the on-board OBU again through ETC, matches it with the bill and completes the payment deduction, and finally pushes the payment information to the mobile terminal for the user to view.

[0011] As a further improvement of the technical solution, the power grid monitoring module includes a model creation unit and a model training unit;

[0012] The model creation unit establishes an input layer with the voltage, current and frequency in the power grid as nodes, determines a hidden layer according to the number of nodes in the input layer, contains 2 to 4 neurons, uses the ReLU function as the activation function between neurons, and uses the softmax function as the output layer activation function to convert the output into a category probability distribution;

[0013] The model training unit divides the sorted data into a training set, a validation set and a test set according to a certain ratio, uses a multi-classification cross entropy loss function to measure the difference between the model's predicted probability distribution and the true probability distribution, and trains the neural network model with the minimum loss function as the goal.

[0014] As a further improvement of the present technical solution, the model training unit uses the Adam optimization algorithm to calculate the first-order moment estimate and the second-order moment estimate of the gradient to adjust the weight of the neural network.

[0015] As a further improvement of the technical solution, the control charging module includes a power adjustment unit and an efficiency optimization unit;

[0016] The power adjustment unit trains historical data with a polynomial regression model to obtain a proportional adjustment strategy between the grid load rate and the output power, and adjusts the output power according to the proportion;

[0017] The efficiency optimization unit uses a genetic algorithm to adjust the wireless charging power of each new energy vehicle according to the user's charging needs and the output power of the power grid, and formulates an optimization strategy for the wireless charging power of the new energy vehicles.

[0018] As a further improvement of the technical solution, when the power adjustment unit trains historical data with a polynomial regression model, a mean square error loss function is used to measure the difference between the output power adjustment ratio predicted by the model and the actual adjustment ratio. The formula of the mean square error loss function is:

[0019]

[0020] Among them, M is the loss function, m is the number of training set samples, To predict the output power adjustment ratio, y i Adjust the ratio to the actual output power.

[0021] As a further improvement of the technical solution, the optimization efficiency unit defines a fitness function based on grid stability and user satisfaction, which is used to measure the quality of the power adjustment strategy. The fitness function formula for grid stability evaluation is:

[0022] F 1 =1 / (1+α×f+β×u+γ×p)

[0023] Among them, F 1 is the fitness function value of power grid stability, α is the load factor weight, f is the load factor, β is the voltage fluctuation weight, u is the voltage fluctuation, γ is the frequency fluctuation weight, and p is the frequency fluctuation;

[0024] The adaptability function formula of user satisfaction is:

[0025]

[0026] Among them, F 2 is the user satisfaction fitness function value, δ is the weight coefficient, n is the number of evaluations, p i is the power adjustment factor, p o is the initial power factor.

[0027] As a further improvement of the present technical solution, the efficiency optimization unit selects the optimal power adjustment strategy based on the highest fitness function value, copies it to the next generation with an elite retention strategy, and continues to participate in evolution. At the same time, it uses multi-point crossover for gene exchange and determines the mutation probability based on the accuracy and value range of the power adjustment coefficient.

[0028] As a further improvement of the present technical solution, the bill generation module generates a charging bill by calculating and accumulating the fees according to the charging time period and the power consumption stage based on the charging power strategy optimized by the control charging module and adopting a dynamic unit price billing method.

[0029] As a further improvement of the technical solution, the information identification module includes an ETC identification unit and an authentication and deduction unit;

[0030] The ETC identification unit uses ETC to turn on the radio frequency to scan the on-board OBU to identify the vehicle information and complete the vehicle authentication;

[0031] The authentication and deduction unit receives the charging bill information, verifies the on-board OBU again through ETC, matches it with the bill and completes the payment deduction, and finally pushes the payment information to the mobile terminal for the user to view.

[0032] As a further improvement of the present technical solution, the ETC identification unit networks the wireless charging device and the ETC control module and distributes them in a local area network controlled by the same routing, supports Gigabit Ethernet, and uses Category 6a network cables to ensure the stability of data transmission.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. The wireless charging and ETC senseless payment device for new energy vehicles controls the ETC radio frequency scanning on-board OBU to identify vehicle information and complete vehicle authentication through the ETC identification unit in the information identification module. Then, the charging power of the new energy vehicle is controlled by the control charging module, and the wireless charging is performed by electromagnetic resonance technology. After charging, the bill generation module generates a bill, and the on-board OBU is verified again through ETC, matched with the bill and payment is completed. Finally, the payment information is pushed to the mobile terminal for the user to view. The whole process does not require the user to adjust the charging power personally, and the overall closed loop of charging and payment is automatically completed, realizing the senseless charging and payment of new energy vehicles.

[0035] 2. When wirelessly charging new energy vehicles, a neural network model for judging the power supply capacity of the power grid is established using the power grid monitoring module. The power supply capacity of the power grid is judged based on the voltage data, current data and frequency data in the real-time monitoring of the power grid. According to the proportional adjustment strategy between the power grid load rate and the output power, the output power of the power grid is dynamically adjusted. When the power grid load is too large, the output power is automatically reduced to give priority to ensuring the power supply of residents and industrial and commercial enterprises; when the power grid load is low, the output power is increased and automatically stopped after full charging;

[0036] 3. Based on the charging amount or charging time data set by the user, as well as the vehicle battery status data, using genetic algorithms, under the power supply capacity of the power grid, considering the user's charging needs, formulate an optimization strategy for the wireless charging power of new energy vehicles to improve the wireless charging efficiency of new energy vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the overall process of the present invention;

[0038] Figure 2 It is a schematic diagram of the overall details of the present invention;

[0039] Figure 3 It is a schematic diagram of the control charging module flow of the present invention.

[0040] The meaning of each number in the figure is:

[0041] 100. Information identification module; 110. ETC identification unit; 120. Authentication and deduction unit; 200. Power grid monitoring module; 210. Model creation unit; 220. Model training unit; 300. Control charging module; 310. Power adjustment unit; 320. Efficiency optimization unit; 400. Bill generation module. DETAILED DESCRIPTION

[0042] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] At present, when new energy vehicles are wirelessly charging, it is impossible to dynamically adjust the charging power of new energy vehicles according to the power supply capacity of the power grid, and comprehensively consider user needs, improve wireless charging efficiency, and reasonably allocate power grid energy. In order to be able to dynamically adjust the output power of the power grid while ensuring residential and industrial and commercial electricity use, improve the wireless charging efficiency of new energy vehicles, and use ETC to realize contactless payment and reduce operating procedures.

[0044] Therefore, the present invention proposes to establish a neural network model for judging the power supply capacity of the power grid through the power grid monitoring module, dynamically adjust the output power of the power grid according to the proportional adjustment strategy between the power grid load rate and the output power, and use the control charging module to use a genetic algorithm. Under the power supply capacity of the power grid, the charging needs of the user are considered to formulate an optimization strategy for the wireless charging power of new energy vehicles, thereby improving the wireless charging efficiency of new energy vehicles.

[0045] The details are as follows:

[0046] See also Figure 1 As shown, the present invention provides a new energy vehicle wireless charging and ETC non-contact payment device, including an information identification module 100, a power grid monitoring module 200, a charging control module 300 and a bill generation module 400;

[0047] The information identification module 100 uses ETC to turn on the radio frequency scanning vehicle OBU to identify the vehicle information and complete the vehicle authentication. The user sets the charging amount or charging time through the mobile phone APP, and transmits the information and the vehicle battery status information to the control charging module 300;

[0048] The power grid monitoring module 200 uses a smart meter to monitor the voltage data, current data and frequency data in the power grid in real time, and uses the voltage, current and frequency in the power grid as input layer nodes and the power supply capacity of the power grid as output layer nodes to construct a neural network model for judging the power supply capacity of the power grid, uses historical data to train the neural network model, and transmits the power supply capacity data of the power grid to the control charging module 300 based on the power supply capacity of the power grid judged by the voltage data, current data and frequency data in the power grid monitored in real time;

[0049] The control charging module 300 receives the power supply capacity data of the power grid transmitted by the power grid monitoring module 200, and dynamically adjusts the output power according to the ratio adjustment strategy with the power grid load rate as an indicator. At the same time, the charging amount or charging time data set by the user and the vehicle battery status data transmitted by the receiving information identification module 100 are comprehensively analyzed. The received data is formulated by using a genetic algorithm under the power supply capacity of the power grid and considering the charging needs of the user. According to the optimization strategy, the wireless charging power and time are counted and transmitted to the bill generation module 400;

[0050] The bill generation module 400 generates a charging bill based on the statistical wireless charging power and duration, and sends the bill to the information identification module 100, verifies the on-board OBU again through ETC, matches it with the bill and completes the payment deduction, and finally pushes the payment information to the mobile terminal for the user to view.

[0051] like Figure 2As shown, the power grid monitoring module 200 includes a model creation unit 210 and a model training unit 220;

[0052] The model creation unit 210 establishes an input layer with the voltage, current and frequency in the power grid as nodes, determines a hidden layer containing 2 to 4 neurons according to the number of nodes in the input layer, uses the ReLU function as the activation function between neurons, and uses the softmax function as the output layer activation function to convert the output into a category probability distribution;

[0053] The model training unit 220 divides the sorted data into a training set, a validation set, and a test set according to a certain ratio, uses a multi-classification cross entropy loss function to measure the difference between the model prediction probability distribution and the true probability distribution, and trains the neural network model with the minimum loss function as the goal;

[0054] Among them, the function expression of the ReLU function is: f(x)=max(0,x). When the input is less than 0, the function output is 0; when the input is greater than or equal to 0, the function output is x itself.

[0055] The ReLU function is mainly a comparison and selection operation, and the calculation speed is faster. In the training process of neural networks, especially when processing large-scale data, this simple calculation method can greatly reduce the calculation time, thereby accelerating the training process, effectively avoiding the gradient disappearance problem, and enabling the weights to be effectively updated, thereby improving the learning ability of the neural network.

[0056] When dividing the sorted data, first randomly shuffle all the sorted data to avoid the influence of certain order rules that may exist in the original data collection or sorting process on the division results, ensure that each data subset can evenly contain various types of data samples, and calculate the index position corresponding to 70% of the total data. For example, if there are 1,000 data in total, then the size of the training set should be 700 data. Through the index operation, the first 700 data are divided into the training set. These data will be used in the main training process of the model to let the model learn the relationship between the input features (voltage, current and frequency) and the output label (power supply capacity of the power grid);

[0057] Next, from the remaining data (i.e., data with index 701 and later), select 20% of the total data as the validation set. Continuing with the above example, there are 300 data items remaining, so the size of the validation set should be 200 data items. The validation set is mainly used to adjust and optimize the model's hyperparameters (such as the number of hidden layer neurons, learning rate, etc.) during the training process to prevent the model from overfitting. By continuously evaluating the performance of the model on the validation set, the hyperparameters are adjusted according to the changes in performance indicators to find the best model configuration.

[0058] Finally, the remaining data constitutes the test set. The remaining 100 data are the test set. After the model training is completed, the test set is used to finally evaluate the performance of the model on unseen data. It can truly reflect the generalization ability of the model, that is, the model's prediction accuracy for new data.

[0059] In order to better calculate the weight of the neural network, the model training unit 220 uses the Adam optimization algorithm to calculate the first-order moment estimation and the second-order moment estimation of the gradient to adjust the weight of the neural network;

[0060] The core idea of ​​the Adam optimization algorithm is to maintain an adaptive learning rate for each parameter and use the first-order and second-order moment information of the gradient to accelerate convergence and reduce oscillations during training.

[0061] When the voltage data, current data and frequency data of the smart meter in the real-time monitoring grid are input into the neural network for forward propagation and loss calculation, the gradient of the loss function with respect to the parameters is calculated through back propagation. Assuming that the current calculated gradient is g t , update the first-order moment estimate m according to the following formula:

[0062] m t =β 1 ×m t-1 +(1-β 1 )×g t

[0063] Among them, m t is the first-order moment estimate of the current step, m t-1 is the first-order moment estimate of the previous step, g t is the gradient, β 1 is the weighted average.

[0064] Update the second-order moment estimate v according to the following formula:

[0065]

[0066] Among them, v t is the second-order moment estimate of the current step, v t-1 is the second-order moment estimate of the previous step, is the square of the gradient, β 2 is the weighted average.

[0067] The second-order moment estimation is used to measure the variance of the gradient. The change of the gradient is captured by the weighted average of the square of the gradient. In the training, the voltage, current and frequency in the power grid are used as the input layer nodes, and the power supply capacity of the power grid is used as the output layer node to build a neural network model for judging the power supply capacity of the power grid. After each input and back propagation of the training sample (or small batch sample), the Adam optimization algorithm is used to update the parameters of the model according to the above steps. After multiple training cycles, the parameters of the model will gradually converge to a better value, thereby improving the model's judgment on the power supply capacity of the power grid.

[0068] like Figure 3 As shown, the control charging module 300 includes a power adjustment unit 310 and an efficiency optimization unit 320;

[0069] The power adjustment unit 310 trains historical data with a polynomial regression model to obtain a proportional adjustment strategy between the grid load rate and the output power, and adjusts the output power according to the proportion;

[0070] The efficiency optimization unit 320 uses a genetic algorithm to adjust the wireless charging power of each new energy vehicle according to the user's charging demand and the output power of the power grid, and formulates an optimization strategy for the wireless charging power of the new energy vehicle;

[0071] Among them, the training formula of the polynomial regression model is:

[0072] y=β 0 +β 1 x+β 2 x 2 +…+β n x n

[0073] Among them, y is the actual output power adjustment ratio, x is the grid load rate, and β is the corresponding coefficient.

[0074] For example, relevant data from the past year was obtained from a certain region's power grid operation and management system, recording the power grid load rate and the corresponding output power adjustment ratio every 15 minutes. Some of the data are as follows:

[0075] Serial number Grid load rate (%) Output power adjustment ratio (%) 1 20 100 2 40 90 3 50 80 4 70 60 5 90 20

[0076] At the beginning, we were not sure about the specific polynomial relationship between the grid load rate and the output power adjustment ratio, so we tried to start with a lower degree and first selected a quadratic polynomial for modeling. Later, we decided whether to increase the degree of the polynomial based on the performance of the model on the validation set.

[0077] The formula for a quadratic polynomial is:

[0078] y=β 0 +β1 x+β 2 x 2

[0079] Among them, y is the actual output power adjustment ratio, x is the grid load rate, and β is the corresponding coefficient.

[0080] First, set β 0 , β 1 , β 2 The coefficients are initialized to small random numbers in the interval [-0.1, 0.1].

[0081] In order to better measure the difference between the output power adjustment ratio predicted by the model and the actual adjustment ratio, when the power adjustment unit 310 trains historical data with a polynomial regression model, a mean square error loss function is used to measure the difference between the output power adjustment ratio predicted by the model and the actual adjustment ratio. The formula of the mean square error loss function is:

[0082]

[0083] Among them, M is the loss function, m is the number of training set samples, To predict the output power adjustment ratio, y i Adjust the ratio to the actual output power.

[0084] Input the grid load rate data in the training set into the quadratic polynomial model to obtain the predicted output power adjustment ratio, then calculate the loss function, and then use the gradient descent algorithm to update the coefficients. Repeat this process, for example, for 1,000 iterations. At the same time, evaluate the performance of the model on the validation set every 100 iterations (calculate the loss function on the validation set), observe the changes in the validation set loss, and prevent overfitting;

[0085] If the model performance is not ideal, try increasing the polynomial degree to three times, retrain, validate, and test to see if the error index can be reduced. The formula for the cubic polynomial is:

[0086] y=β 0 +β 1 x+β 2 x 2 +β 3 x 3

[0087] Among them, y is the actual output power adjustment ratio, x is the grid load rate, and β is the corresponding coefficient.

[0088] Assume that, after optimization, a cubic polynomial model is finally determined:

[0089] y=1.2-0.03x+0.0002x 2 -0.000001x 3

[0090] The cubic polynomial model is used as a proportional adjustment strategy between the grid load rate and the output power. When the grid load rate is 50%, substituting x=50 into the formula yields:

[0091] y=1.2-0.03×50+0.0002×50 2 +0.000001×50 3 =75

[0092] That is, the recommended output power adjustment ratio is 75% at this time, which means that the output power should be adjusted accordingly according to this ratio to adapt to the grid load conditions and ensure the stable operation of the grid.

[0093] In order to comprehensively analyze the grid stability and user satisfaction as the standard for measuring the quality of the power adjustment strategy, the optimization efficiency unit 320 defines the fitness function based on the grid stability and user satisfaction as the standard for measuring the quality of the power adjustment strategy. The fitness function formula for grid stability evaluation is:

[0094] F 1 =1 / (1+α×f+β×u+γ×p)

[0095] Among them, F 1 is the fitness function value of power grid stability, α is the load factor weight, f is the load factor, β is the voltage fluctuation weight, u is the voltage fluctuation, γ is the frequency fluctuation weight, and p is the frequency fluctuation;

[0096] The adaptability function formula of user satisfaction is:

[0097]

[0098] Among them, F 2 is the user satisfaction fitness function value, δ is the weight coefficient, n is the number of evaluations, p i is the power adjustment factor, p o is the initial power factor.

[0099] For example, collect the grid load rate data for each hour within 24 hours, Lf = [20%, 22%, 25%, ..., 30%], first calculate the load rate mean:

[0100]

[0101] Then calculate the standard deviation of the load factor:

[0102]

[0103] Similarly, find the corresponding voltage fluctuation u and frequency fluctuation p, set the voltage fluctuation weight β = 0.3, the frequency fluctuation weight γ = 0.2, and the weight coefficient α = 0.5, and use θ lf is the load rate f, which is substituted into the adaptability function formula of the power grid stability assessment to obtain the adaptability function value F of the power grid stability assessment 1 .

[0104] Assume that there are 4 electrical devices, and the data before and after power adjustment are as follows:

[0105] serial number Original power(KW) Adjusted power(KW) Power adjustment range (KW) 1 10 9.5 0.5 2 8 7.8 0.2 3 6 6.3 0.3 4 5 4.5 0.5

[0106] The average power adjustment range is:

[0107]

[0108] Then the power adjustment factor p i Adjustment range with average power Equal, substitute it into the fitness function formula of user satisfaction, and get the fitness function value F of user satisfaction 2 , comprehensively considering the grid stability and user satisfaction, the fitness function F = F 1 ×F 2 The value of measures the quality of the power adjustment strategy. The larger the value, the better the power adjustment strategy performs in ensuring the stable operation of the power grid and meeting user needs. In the optimization algorithm, this fitness function can be used to evaluate and select a better power adjustment strategy.

[0109] In order to select the optimal power adjustment strategy based on the highest fitness function value, the optimization efficiency unit 320 selects the optimal power adjustment strategy based on the highest fitness function value, copies it to the next generation with the elite retention strategy, and continues to participate in the evolution. At the same time, multi-point crossover is used for gene exchange, and the mutation probability is determined according to the accuracy and value range of the power adjustment coefficient;

[0110] Assume that we have a chromosome representation of a power adjustment strategy, which represents the power adjustment coefficients of five power-consuming devices. Each chromosome (individual) can be represented as a gene sequence, for example:

[0111] Individual 1: [0.7, 0.6, 0.5, 0.4, 0.3], Individual 2: [0.8, 0.7, 0.6, 0.5, 0.4];

[0112] First, determine the number of crossover points, for example, select 2 crossover points, and then randomly determine the positions of the crossover points within the chromosome length. Assuming that the chromosome length is 5 (corresponding to 5 genes), the randomly selected crossover points are the 2nd and 4th positions. For individual 1 and individual 2, the gene fragments between the crossover points are exchanged, and two new offspring individuals are obtained after the exchange.

[0113] Offspring 1: [0.7, 0.7, 0.6, 0.4, 0.3], Offspring 2: [0.8, 0.6, 0.5, 0.5, 0.4];

[0114] In complex grid power adjustment scenarios, there may be complex relationships between the power adjustment coefficients of different power-consuming devices. Multi-point crossover can better explore these relationships and thus find a better combination of power adjustment strategies.

[0115] Suppose we have an individual (chromosome): [0.7, 0.6, 0.5, 0.4, 0.3], the mutation probability is 10%, for each gene, generate a random number r, if r is less than the mutation probability, then mutate, assuming the first gene meets the mutation conditions, since the precision is 0.01, the mutation step length can be set to ±0.01, randomly choose whether to increase or decrease the gene value, assuming it is increased, the first gene after mutation becomes 0.71, and the new individual is obtained: [0.71, 0.6, 0.5, 0.4, 0.3].

[0116] In order to generate a charging bill more accurately, the bill generation module 400 calculates the fees according to the charging period and the power consumption stage respectively and accumulates them according to the charging power strategy optimized by the control charging module 300 to generate a charging bill;

[0117] Get the charging start and end time information from the wireless charging device or charging management system, and divide the day into different time periods, such as peak hours (such as 5pm-8pm on weekdays), flat periods (such as 9am-5pm, 8pm-10pm), and valley periods (such as 10pm-6am). Determine the charging duration of each charging process in each time period, record the charging power at different times during the charging process, calculate the power consumption in each time period by integrating the power over time (or simply multiplying the power by the time interval and accumulating), and at the same time, track the cumulative power consumption during the entire charging process, generate the bill as an electronic document (such as PDF) or display it to the user in a clear interface in the user-side charging application.

[0118] The information identification module 100 includes an ETC identification unit 110 and an authentication and deduction unit 120;

[0119] The ETC identification unit 110 uses ETC to turn on the radio frequency to scan the onboard OBU to identify the vehicle information and complete the vehicle authentication;

[0120] The authentication and deduction unit 120 receives the charging bill information, verifies the vehicle-mounted OBU again through ETC, matches it with the bill and completes the payment deduction, and finally pushes the payment information to the mobile terminal for the user to view;

[0121] In order to ensure the stability of data transmission, the ETC identification unit 110 distributes the wireless charging device and the ETC control module in the same routing-controlled local area network, supports Gigabit Ethernet, and uses Category 6 network cables to ensure the stability of data transmission;

[0122] The star topology is adopted, the router is placed in the center, the wireless charging device and the ETC control module are used as access devices, and are connected to each port of the router through a Category 6 network cable. This structure is convenient for centralized management and maintenance, and when a device or network cable fails, it will not affect the normal communication of other devices;

[0123] After the wireless charging device and the ETC control module are networked, data interaction and collaborative work between the two are facilitated. For example, the ETC system can transmit the vehicle's identity information (such as license plate number, owner information, etc.) to the wireless charging device to achieve automatic authentication and billing association for charging. When the vehicle is finished charging, the wireless charging device can feed back the charging information (such as charging amount, cost, etc.) to the ETC system so that the cost can be settled or charging-related information can be recorded in the ETC account, providing users with a more convenient one-stop service.

[0124] In summary, the working principle of this solution is as follows:

[0125] The wireless charging and ETC senseless payment device for new energy vehicles controls the ETC radio frequency scanning on-board OBU to identify vehicle information and complete vehicle authentication through the ETC identification unit 110 in the information identification module 100, and then controls the charging power of the new energy vehicle by controlling the charging module 300, and wirelessly charges the new energy vehicle by using the electromagnetic resonance technology. After charging is completed, the bill generation module generates a bill, and the on-board OBU is verified again by ETC, matched with the bill and payment is completed, and finally the payment information is pushed to the mobile terminal for the user to view. The whole process does not require personal adjustment of the charging power, and the overall closed loop of charging and payment is automatically completed, realizing the senseless charging and payment of new energy vehicles. When wirelessly charging the new energy vehicle, the electric The network monitoring module 200 establishes a neural network model for judging the power supply capacity of the power grid. The power supply capacity of the power grid is judged according to the real-time monitoring of the voltage data, current data and frequency data in the power grid. According to the proportional adjustment strategy between the power grid load rate and the output power, the output power of the power grid is dynamically adjusted. When the power grid load is too large, the output power is automatically reduced to give priority to residential and industrial and commercial electricity use; when the power grid load is low, the output power is increased and automatically stops after being fully charged. According to the charging amount or charging time data set by the user, as well as the vehicle battery status data, a genetic algorithm is used to consider the user's charging needs under the power supply capacity of the power grid, and an optimization strategy for the wireless charging power of new energy vehicles is formulated to improve the wireless charging efficiency of new energy vehicles.

[0126] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A wireless charging and ETC non-contact payment device for new energy vehicles, characterized by: It comprises an information identification module (100), a power grid monitoring module (200), a charging control module (300) and a bill generation module (400); The information identification module (100) uses ETC to start radio frequency scanning of the vehicle-mounted OBU to identify vehicle information and complete vehicle authentication. The user sets the charging amount or charging time through a mobile phone APP, and transmits the information and vehicle battery status information to the control charging module (300); The power grid monitoring module (200) uses a smart meter to monitor voltage data, current data and frequency data in the power grid in real time, and uses the voltage, current and frequency in the power grid as input layer nodes and the power supply capacity of the power grid as output layer nodes to construct a neural network model for judging the power supply capacity of the power grid, uses historical data to train the neural network model, and transmits the power supply capacity data of the power grid to the control charging module (300) based on the power supply capacity of the power grid judged by monitoring the voltage data, current data and frequency data in the power grid in real time; The control charging module (300) receives the power supply capacity data of the power grid transmitted by the power grid monitoring module (200), and dynamically adjusts the output power according to the ratio adjustment strategy with the power grid load rate as an indicator. At the same time, it receives the charging amount or charging time data set by the user and the vehicle battery status data transmitted by the information identification module (100), and comprehensively analyzes the received data. Using a genetic algorithm, it considers the charging needs of the user under the power supply capacity of the power grid and formulates an optimization strategy for the wireless charging power of the new energy vehicle. According to the optimization strategy, the wireless charging power and time are counted and transmitted to the bill generation module (400); The bill generation module (400) generates a charging bill based on the statistical wireless charging power and duration, and sends the bill to the information identification module (100), verifies the vehicle-mounted OBU again through ETC, matches it with the bill and completes the payment deduction, and finally pushes the payment information to the mobile terminal for viewing by the user.

2. The new energy vehicle wireless charging and ETC non-contact payment device according to claim 1 is characterized in that: The power grid monitoring module (200) comprises a model creation unit (210) and a model training unit (220); The model creation unit (210) uses the voltage, current and frequency in the power grid as nodes to establish an input layer, determines a hidden layer containing 2 to 4 neurons according to the number of nodes in the input layer, uses the ReLU function as an activation function between neurons, and uses the softmax function as an output layer activation function to convert the output into a category probability distribution; The model training unit (220) divides the sorted data into a training set, a validation set and a test set according to a certain ratio, uses a multi-classification cross entropy loss function to measure the difference between the model predicted probability distribution and the true probability distribution, and trains the neural network model with the minimum loss function as the goal.

3. The wireless charging and ETC non-contact payment device for new energy vehicles according to claim 2 is characterized in that: The model training unit (220) uses the Adam optimization algorithm to calculate the first-order moment estimation and the second-order moment estimation of the gradient to adjust the weight of the neural network.

4. The new energy vehicle wireless charging and ETC non-contact payment device according to claim 1 is characterized in that: The control charging module (300) comprises a power adjustment unit (310) and an efficiency optimization unit (320); The power adjustment unit (310) trains historical data with a polynomial regression model to obtain a ratio adjustment strategy between the grid load rate and the output power, and adjusts the output power according to the ratio; The efficiency optimization unit (320) uses a genetic algorithm to adjust the wireless charging power of each new energy vehicle according to the charging demand of the user and the output power of the power grid, and formulates an optimization strategy for the wireless charging power of the new energy vehicle.

5. The new energy vehicle wireless charging and ETC contactless payment device according to claim 4 is characterized in that: When the power adjustment unit (310) trains historical data with a polynomial regression model, a mean square error loss function is used to measure the difference between the output power adjustment ratio predicted by the model and the actual adjustment ratio. The formula of the mean square error loss function is: Among them, M is the loss function, m is the number of training set samples, To predict the output power adjustment ratio, y i Adjust the ratio to the actual output power.

6. The new energy vehicle wireless charging and ETC non-contact payment device according to claim 4 is characterized in that: The optimization efficiency unit (320) defines a fitness function based on power grid stability and user satisfaction, which is used to measure the quality of the power adjustment strategy. The fitness function formula for power grid stability evaluation is: F1=1 / (1+α×f+β×u+γ×p) Among them, F1 is the fitness function value of the power grid stability, α is the load rate weight, f is the load rate, β is the voltage fluctuation weight, u is the voltage fluctuation, γ is the frequency fluctuation weight, and p is the frequency fluctuation; The adaptability function formula of user satisfaction is: Among them, F2 is the user satisfaction fitness function value, δ is the weight coefficient, n is the number of evaluations, and p i is the power adjustment factor, p o is the initial power factor.

7. The new energy vehicle wireless charging and ETC non-contact payment device according to claim 4 is characterized in that: The efficiency optimization unit (320) selects the optimal power adjustment strategy based on the highest fitness function value, and copies it to the next generation using the elite retention strategy to continue to participate in evolution. At the same time, it uses multi-point crossover to exchange genes and determines the mutation probability based on the accuracy and value range of the power adjustment coefficient.

8. The new energy vehicle wireless charging and ETC non-contact payment device according to claim 1 is characterized in that: The bill generation module (400) calculates the fees according to the charging time period and the power consumption stage respectively and then accumulates them according to the charging power strategy optimized by the charging control module (300) to generate a charging bill.

9. The new energy vehicle wireless charging and ETC non-contact payment device according to claim 1, characterized in that: The information identification module (100) comprises an ETC identification unit (110) and an authentication and fee deduction unit (120); The ETC identification unit (110) uses ETC to turn on the radio frequency to scan the on-board OBU to identify vehicle information and complete vehicle authentication; The authentication and deduction unit (120) receives the charging bill information, verifies the vehicle-mounted OBU again through ETC, matches it with the bill and completes the payment deduction, and finally pushes the payment information to the mobile terminal for viewing by the user.

10. The wireless charging and ETC non-contact payment device for new energy vehicles according to claim 9, characterized in that: The ETC identification unit (110) networks the wireless charging device and the ETC control module and distributes them in a local area network controlled by the same routing, supports Gigabit Ethernet, and uses Category 6a network cables for connection to ensure the stability of data transmission.

Images