Electric vehicle peak-shaving charging and discharging method based on micro-metering system

Abstract

The application relates to a micro-cluster metering system-based electric vehicle peak-shaving charging and discharging method, which comprises the following steps: 1) dividing community power consumption into public infrastructure power consumption P, resident life power consumption R and charging pile power consumption C; 2) obtaining an upper limit T of a community transformer; 3) obtaining resident life power consumption historical data, and training a resident life power consumption prediction network model M by using the resident life power consumption historical data; 4) obtaining resident life power consumption real-time data, and predicting resident life power consumption R in a short future period by using the resident life power consumption real-time data and the resident life power consumption prediction network model M prediction ; 6) calculating a charging pile charging upper limit threshold C threshold =T-P-R prediction ; 7) using the charging pile charging upper limit threshold C threshold as a constraint to schedule electric vehicle charging and discharging. The method is favorable to scientifically and reasonably scheduling electric vehicle charging and discharging in a community, and further improves community power grid power consumption safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric vehicle charging, in particular to an electric vehicle off-peak charging and discharging method based on a micro-set metering system. BACKGROUND

[0002] Electric vehicle charging has time and space randomness, and a large number of electric vehicles accessing in a short time will cause the phenomenon of "peak on peak" of the community power grid load, and there is a certain safety hazard. The current electric vehicle charging and discharging scheduling method has some problems in software and hardware and data, and it is necessary to comprehensively consider the combination of charging pile software and hardware and data sharing, and to develop a more perfect and intelligent charging and discharging scheduling system.

[0003] The existing scheduling method is often based on the current data of the charging pile, and the cloud-software-hardware combination is poor, which leads to the problems of lack of estimation of future power consumption, inability to guarantee power safety under high load, and difficulty in data sharing in the scheduling process. SUMMARY

[0004] The purpose of the present application is to provide an electric vehicle off-peak charging and discharging method based on a micro-set metering system, which is beneficial to scientifically and reasonably scheduling the charging and discharging of electric vehicles in the community, and thus improves the power safety of the community power grid.

[0005] In order to achieve the above purpose, the technical scheme adopted by the present application is: an electric vehicle off-peak charging and discharging method based on a micro-set metering system, comprising the following steps:

[0006] 1) dividing the community power consumption into public infrastructure power consumption P, residential life power consumption R and charging pile power consumption C;

[0007] 2) obtaining the upper limit T of the community transformer;

[0008] 3) obtaining the historical data of residential life power consumption, and training a residential life power consumption prediction network model M using the historical data of residential life power consumption;

[0009] 4) obtaining real-time data of residential life power consumption, and predicting the residential life power consumption R in the near future using the real-time data of residential life power consumption and the residential life power consumption prediction network model M prediction ;

[0010] 6) calculating the charging pile charging upper limit threshold C threshold = T-P-R prediction ;

[0011] 7) the micro-set metering system schedules the charging and discharging of electric vehicles with the charging pile charging upper limit threshold C threshold as a constraint.

[0012] Furthermore, in step 1), the electricity consumption of the community is divided into unchangeable public infrastructure electricity consumption P and variable residential electricity consumption, and the residential electricity consumption is further divided into residential living electricity consumption R and charging pile electricity consumption C.

[0013] Furthermore, in the cloud, a residential electricity consumption prediction network model M is trained using historical residential electricity consumption data and then deployed. Real-time residential electricity consumption data is then input into the deployed model M to predict the residential electricity consumption R in the near future. prediction Furthermore, by obtaining the future short-term residential electricity consumption R... prediction The upper limit threshold for charging piles is calculated based on the electricity consumption P of public infrastructure and the upper limit T of the community transformer: C threshold =TPR prediction Finally, the calculated charging upper limit threshold C of the charging pile will be used. threshold The micro-aggregate metering system sent to the execution end;

[0014] At the execution end, the micro-collection metering system uses the charging pile's upper limit threshold C. threshold As a constraint, peak-shaving charging and discharging of electric vehicles is controlled.

[0015] Furthermore, a residential electricity consumption prediction network model M is trained using historical residential electricity consumption data. The implementation method is as follows:

[0016] Construct a residential electricity consumption prediction network model M based on LSTM encoding and GRU decoding;

[0017] Historical residential electricity consumption data is used to generate time series data in chronological order. Then, the generated time series data is input into the residential electricity consumption prediction network model M to train the residential electricity consumption prediction network model M.

[0018] For the input time series data, the residential electricity consumption prediction network model M first uses an LSTM network to encode the time series data to generate encoded features E; then it uses a GRU network to decode the encoded features E to generate decoded features D; and finally it uses a fully connected network to predict the decoded features D to generate prediction results.

[0019] During model training, mean squared error is used as the loss function, and the network is optimized by the SGD optimizer with momentum. The mean squared error (MSE), root mean square error (RMSE), and mean absolute error (MAE) are used for evaluation. Finally, the trained residential electricity consumption prediction network model M is obtained.

[0020] Deploy a pre-trained residential electricity consumption prediction network model M to predict residential electricity consumption R in the near future. prediction .

[0021] Further, the micro-collective metering system sets a charging pile charging upper threshold C threshold As a constraint, the electric vehicle charging and discharging is scheduled, and the implementation method is:

[0022] 701) The micro-collective metering system monitors the electric vehicles being charged in the charging pile, and judges whether the electric vehicles are fully charged. If yes, the data CAR of the electric vehicles fully charged is removed from the database of the micro-collective metering system data , and the database data is updated, and then step 702) is executed, otherwise step 702) is directly executed;

[0023] 702) It is judged whether there is a new electric vehicle accessing. If yes, step 703) is executed, otherwise step 704) is executed;

[0024] 703) The data CAR of the new accessing electric vehicle is stored in the database new ;

[0025] 704) The minimum charging power CAR min is selected from the database, that is, the minimum charging power CAR min is selected from all electric vehicles in the waiting state;

[0026] 705) The sum of the current charging pile total power C current and the minimum charging power CAR min is calculated, that is, C current +CAR min ;

[0027] 706) It is judged whether the sum of the current charging pile total power and the minimum charging power is less than the charging pile charging upper threshold, that is, whether C current +CAR min <C threshold is satisfied, if yes, step 707) is executed, otherwise the next scheduling time is waited to continue the next scheduling;

[0028] 707) The minimum charging power CAR min corresponding electric vehicle is charged, and the current charging pile total power is updated by the sum of the current charging pile total power and the minimum charging power, that is, C current =C current +CAR min ;

[0029] 708) The database data is updated;

[0030] 709) It is judged whether there is still electric vehicle data in the waiting state in the database. If yes, step 4) is returned to execute, otherwise the next scheduling time is waited to continue the next scheduling.

[0031] Compared with the prior art, the present application has the following beneficial effects: a micro-cluster metering system-based electric vehicle off-peak charging and discharging method is provided, which predicts the residential life electricity of a community in a future period of time based on the electricity data of the community, and scientifically and reasonably schedules the charging and discharging of electric vehicles in the community under the premise of ensuring the residential electricity, thereby avoiding the safety hazards of the community electricity in the high-load condition, and improving the safety of the community power grid electricity. Therefore, the present application has strong practicability and broad application prospect. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 is a method implementation flowchart of an embodiment of the present application;

[0033] Figure 2 is an implementation principle diagram of training a residential life electricity prediction network model M in an embodiment of the present application;

[0034] Figure 3 is an implementation principle block diagram of a micro-cluster metering system in an embodiment of the present application. DETAILED DESCRIPTION

[0035] The present application will be further described below in combination with the drawings and embodiments.

[0036] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0037] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and furthermore, it should be understood that when the terms "comprise" and / or "include" are used in the specification, there is a presence of a feature, step, operation, device, component and / or combinations thereof.

[0038] As shown in Figure 1 , the present embodiment provides a micro-cluster metering system-based electric vehicle off-peak charging and discharging method, which comprises the following steps:

[0039] 1) The electricity of a community is divided into public infrastructure electricity P, residential life electricity R and charging pile electricity C.

[0040] 2) Obtain the upper limit T of the community transformer.

[0041] 3) Obtain the historical data of residential life electricity, and train a residential life electricity prediction network model M using the historical data of residential life electricity.

[0042] 4) Obtain real-time data of resident life electricity, and predict resident life electricity R in the near future by using real-time data of resident life electricity and resident life electricity prediction network model M prediction .

[0043] 6) Calculate the charging pile charging upper threshold C threshold = T-P-R prediction .

[0044] 7) The micro-metering system schedules the charging and discharging of electric vehicles by taking the charging pile charging upper threshold C threshold as a constraint.

[0045] In this embodiment, the electricity of the cell is divided into unchangeable public infrastructure electricity P and variable resident electricity, which is further divided into resident life electricity R and charging pile electricity C.

[0046] In this embodiment, the computing resources of the cloud and the execution end are fully combined to achieve the best scheduling efficiency and implementation effect.

[0047] In the cloud, resident life electricity prediction network model M is trained and deployed by using resident life electricity historical data; then resident life electricity real-time data is input into the deployed resident life electricity prediction network model M to predict resident life electricity R in the near future prediction ; and then the charging pile charging upper threshold C threshold = T-P-R prediction is calculated by using the obtained resident life electricity R in the near future prediction , public infrastructure electricity P and cell transformer upper limit T; finally, the calculated charging pile charging upper threshold C threshold is sent to the micro-metering system of the execution end.

[0048] In the execution end, the micro-metering system controls the peak-shaving charging and discharging of electric vehicles by taking the charging pile charging upper threshold C threshold as a constraint.

[0049] Figure 2 The implementation principle of training resident life electricity prediction network model M in this embodiment is shown. As shown in Figure 2 , the implementation method of training resident life electricity prediction network model M by using resident life electricity historical data is as follows:

[0050] A) Construct resident life electricity prediction network model M based on LSTM encoding-GRU decoding.

[0051] B) The historical data of residential life electricity is generated into time series data in chronological order, and then the generated time series data is input into the residential life electricity prediction network model M, and the residential life electricity prediction network model M is trained.

[0052] C) For the input time series data, the residential life electricity prediction network model M first encodes the time series data using the LSTM network to generate the encoding feature E; then decodes the encoding feature E using the GRU network to generate the decoding feature D; and then predicts the decoding feature D using the full connection network to generate the prediction result.

[0053] D) During the model training process, the mean square error is used as the loss function, the network is optimized by the SGD optimizer with momentum, and the mean square error MSE, the root mean square error RMSE and the mean absolute error MAE are used for evaluation; Finally, the trained residential life electricity prediction network model M is obtained.

[0054] E) Deploy the trained residential life electricity prediction network model M to predict the residential life electricity R prediction .

[0055] In step 7), the micro-set metering system takes the charging pile charging upper threshold C threshold As a constraint, the charging and discharging of electric vehicles are scheduled, and the specific implementation process is as follows:

[0056] 701) The micro-set metering system monitors the electric vehicles being charged in the charging pile, judges whether the electric vehicles are fully charged, and if so, removes the fully charged electric vehicle data CAR data from the database of the micro-set metering system, updates the database data, and then executes step 702), otherwise directly executes step 702);

[0057] 702) Determine whether there is a new electric vehicle accessing, if yes, execute step 703), otherwise execute step 704);

[0058] 703) Store the new access electric vehicle data CAR new in the database;

[0059] 704) Select the minimum charging power CAR min in the database, that is, select the minimum charging power CAR min corresponding to the electric vehicle from all electric vehicles in the waiting state;

[0060] 705) Calculate the sum C current +CAR min of the current charging pile total power C current and the minimum charging power CAR min;

[0061] 706) judge whether the sum of the current charging pile total power and the minimum charging power is less than the charging pile charging upper threshold, that is, whether C current + CAR min < C threshold is, execute step 707), otherwise wait for the next scheduling time to continue the next scheduling;

[0062] 707) update the minimum charging power CAR min corresponding to the electric vehicle charging, and update the current charging pile total power with the sum of the current charging pile total power and the minimum charging power, that is, let C current = C current + CAR min ;

[0063] 708) update the database data;

[0064] 709) judge whether there is still electric vehicle data in the database in a waiting state, yes, return to execute step 4), otherwise wait for the next scheduling time to continue the next scheduling.

[0065] In this embodiment, the implementation principle of the micro integrated metering system is as shown in Figure 3 The micro integrated metering system includes a switch pole, a switch detection module, a power module, a metering and communication module, and a display and Bluetooth module. The switch pole is provided with a current transformer and a leakage current transformer. The input end of the switch pole is connected with 220V mains and supplies power to the power module. The power module converts AC to DC and outputs DC to the switch detection module and the metering and communication module. The metering and communication module is electrically connected with the current transformer and the leakage current transformer on the switch pole to sample the load current and the leakage current. The metering and communication module is electrically connected with the switch detection module to receive the switch position detection signal and send the opening and closing control signal. The metering and communication module is electrically connected with the display and Bluetooth module to interact data.

[0066] The input end of the switch pole is connected with 220V mains, that is, connected with L and N lines, and then supplies power to the power module. The power module converts AC to DC and outputs 9V and 15V DC voltage to supply power to the switch detection module, the metering and communication module, and the display and Bluetooth module.

[0067] The metering and communication module collects and monitors the leakage current by collecting the leakage current signal input by the leakage current transformer on the switch pole. The metering module samples the load current by collecting the load current signal input by the current transformer on the switch pole.

[0068] The switch detection module detects the rotating position of the switch pole gear and sends a switch position detection signal to the metering and communication module. The metering module issues an opening or closing control signal to the switch detection module by judging the switch state. The switch detection module realizes the opening and closing of the switch through the opening and closing transmission.

[0069] In the embodiment, the switch pole is further provided with an operating mechanism, a contact, a protection device (a tripping device), an arc extinguishing system, a terminal temperature measurement module, and an incoming and outgoing line voltage detection module. The metering and communication module mainly comprises an MCU module, a storage module, an ESAM safety module, an RS485 communication module, a weak current output interface, and an indicator light module. The display and Bluetooth module mainly comprises a liquid crystal display module, a key module, and a Bluetooth communication module.

[0070] The switch detection module is provided with a motor and a driving circuit, an electric leakage opening coil and a driving circuit, and an opening and closing position detection circuit. The motor and the driving circuit control the action of the opening and closing motor through a special driving chip to ensure the output power and improve the driving stability. The electric leakage opening coil and the driving circuit use a unidirectional thyristor to control the action time of the opening coil. The opening and closing position detection circuit detects the position of the opening and closing gears and the real position information of the lever to ensure the reliability of the opening and closing.

[0071] The micro integrated metering system integrates functions of electric energy metering, communication, temperature monitoring, control, remote protection setting, short circuit protection, and reverse power online monitoring. It has a 4-terminal temperature measurement function, supports electric leakage detection and incoming and outgoing line voltage detection. The metering and switch control functions and the electric leakage protection function are suitable for residential power application scenarios. The incoming and outgoing line voltage detection and frequency detection functions are suitable for photovoltaic grid-connected application scenarios. The metering and control functions are suitable for charging pile orderly power consumption scenarios. The 4-terminal temperature measurement, electric leakage current detection, reverse power online monitoring, incoming and outgoing line voltage detection, and short circuit protection functions make the micro integrated metering system safer and more reliable.

[0072] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0073] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0074] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0075] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0076] The above description is only preferred embodiments of the present application, not intended to limit other forms of the application. Any person familiar with the art can make changes or modifications to the above-mentioned technical content of the disclosure, or equivalent embodiments with equivalent changes. However, any simple modification, equivalent change and modification of the above-mentioned embodiments without departing from the technical solution of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. A method for peak-shaving charging and discharging of electric vehicles based on a micro- metering system, characterized in that, The method comprises the following steps: 1) dividing the electricity consumption of a community into public infrastructure electricity P, residential electricity R and charging pile electricity C; 2) obtaining an upper limit T of a transformer in the community; 3) obtaining historical data of residential electricity consumption and training a residential electricity consumption prediction network model M using the historical data; 4) Obtain real-time data of resident life electricity, and predict resident life electricity R in the future short term by using real-time data of resident life electricity and resident life electricity prediction network model M prediction ; 6) Calculate the charging pile charging upper threshold: C threshold = T - P - R prediction ; 7) micro-metering system with upper threshold C for charging at a charging station threshold As a constraint, the charging and discharging of the electric vehicle is scheduled; In the cloud, the resident life electricity prediction network model M is trained and deployed by using the resident life electricity historical data; then the resident life electricity real-time data is input into the deployed resident life electricity prediction network model M to predict the resident life electricity R in the future short term prediction ; further, the charging pile charging upper limit threshold C threshold is calculated by the obtained resident life electricity R prediction , public infrastructure electricity P and cell transformer upper limit T prediction ; finally, the calculated charging pile charging upper limit threshold C threshold is sent to the micro-metering system of the execution end At the execution end, the micro-metering system sets a charging upper threshold C threshold As a constraint, the control of electric vehicles off-peak charging and discharging; The method for training the residential electricity consumption prediction network model M using the historical data of residential electricity consumption is as follows: A residential electricity consumption prediction network model M based on LSTM encoding-GRU decoding is constructed; Time series data is generated from the historical data of residential electricity consumption in chronological order, and then the generated time series data is input into the residential electricity consumption prediction network model M to train the model; For the input time series data, the residential electricity consumption prediction network model M first encodes the time series data using an LSTM network to generate encoding features E, then decodes the encoding features E using a GRU network to generate decoding features D, and finally predicts the decoding features D using a fully connected network to generate a prediction result; During the model training process, the mean square error is used as the loss function, the network is optimized by the SGD optimizer with momentum, and the mean square error MSE, the root mean square error RMSE and the mean absolute error MAE are used for evaluation; and finally the trained residential electricity consumption prediction network model M is obtained; The trained resident life electricity prediction network model M is deployed to predict resident life electricity R in the future short term prediction ; The micro-metering system takes the upper threshold C of the charging pile charging as a constraint threshold As a constraint, the charging and discharging of the electric vehicle is scheduled, and the implementation method is: 701) The micro integrated metering system monitors the electric vehicles being charged in the charging piles, judges whether the electric vehicles are fully charged, if yes, removes the data CAR of the electric vehicles fully charged from the database of the micro integrated metering system, updates the database data, and then executes step 702), otherwise, directly executes step 702); data 701) The micro integrated metering system monitors the electric vehicles being charged in the charging piles, judges whether the electric vehicles are fully charged, if yes, removes the data CAR of the electric vehicles fully charged from the database of the micro integrated metering system, updates the database data, and then executes step 702), otherwise, directly executes step 702); data 701) The micro integrated metering system monitors the electric vehicles being charged in the charging piles, judges 702) determining whether there is a new electric vehicle accessing, if yes, performing step 703), otherwise performing step 704); 703) storing the new access electric car data CAR in the database new ; 704) Selecting the smallest charging power CAR in the database min i.e. the smallest charging power CAR from all electric vehicles in the waiting state min the corresponding electric vehicle; 705) Calculate the current total power C of the charging pile. current With minimum charging power CAR min The sum of C current +CAR min ; 706) whether the sum of the current charging pile total power and the minimum charging power is less than the charging pile charging upper threshold, i.e., whether C current + CAR min < C threshold is, step 707) is executed, otherwise, the next scheduling time is waited for to continue the next scheduling. 707)CAR for minimum charging power min Corresponding electric vehicle charging, and updating the current charging pile total power with the sum of the current charging pile total power and the minimum charging power, i.e. let C current = C current + CAR min ; 708) updating the database data; 709) determining whether there is still electric vehicle data in the waiting state in the database, if yes, returning to perform step 4), otherwise waiting for the next scheduling time to continue the next scheduling.

2. The method of claim 1, wherein the micro- metering system is based on a micro- metering system. In step 1), the electricity consumption of a community is divided into public infrastructure electricity P which cannot be changed and variable residential electricity which is further divided into residential electricity R and charging pile electricity C. ​

Images