Electric vehicle intelligent rapid charging device based on Internet +

By establishing a safety warning model in the charging system of electric vehicles, analyzing the physical data of the battery using neural network models, predicting and warning of charging failures in advance, the problem of inability to early warning in the existing technology is solved, and the safety protection and real-time battery life are achieved monitor.

CN118977611AInactive Publication Date: 2024-11-19YANCHENG INST OF TECH +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410970726.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing electric vehicle charging system only monitors the current during charging, and cannot be warning in advance and prevents charging failures, resulting in battery damage.

Method used

Using an intelligent fast charging device based on the Internet, a security warning model is established on the charging end and the cloud server, and a neural network model is used to analyze the physical data of electric vehicle batteries in real time, predict charging failures in advance, and stop when the warning model judges abnormalities. Charge.

Benefits of technology

It realizes early failure warning of electric vehicle batteries, avoids battery damage, ensures charging safety, and provides real-time monitoring during charging and driving.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118977611A_ABST
    Figure CN118977611A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of electric vehicle charging, and particularly relates to an electric vehicle intelligent quick charging device based on Internet +, which comprises a charging pile and a charging gun, the charging gun is plugged with an electric vehicle charging interface for charging, and an intelligent quick charging early warning system runs on the charging pile. The intelligent fast charging early warning system comprises a charging end and a cloud service end, and the charging end collects physical data in the process that a charging gun charges an electric vehicle and sends the physical data to the cloud service end through a wireless network. And the cloud server side judges whether the charging is abnormal or not in the subsequent charging process according to the physical data through a safety early warning model, and if the charging is abnormal, charging is stopped, and an early warning is given out. According to the invention, charging faults after a certain period of time can be pre-judged in advance, early warning can be carried out in time, early warning is carried out in advance, and charging safety of the electric vehicle is guaranteed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle charging, and in particular relates to an Internet+-based intelligent fast charging device for electric vehicles. Background Art

[0002] With the vigorous development of new energy technologies around the world, electric vehicles and their supporting charging facilities are becoming increasingly popular; charging safety of electric vehicles is particularly important. A Chinese patent with patent number CN206914155U discloses an intelligent fast charging system for electric vehicles, which comprises: an input end connected to the power grid for accessing alternating current; a power module arranged between the input end and the output end for converting the accessed alternating current into direct current; and an output end connected to the power module for outputting the direct current converted by the power module, wherein the power module comprises: an EMI filter for filtering current; a full-bridge rectifier circuit for rectifying current; an active PFC circuit for preventing current surges; a full-bridge power conversion circuit for full-wave power conversion; a high-frequency transformer for voltage conversion; a high-frequency rectifier for rectifying current; and a filter connected to the high-frequency rectifier for filtering current.

[0003] In the prior art, only the current during the charging process is monitored, and charging early warning protection is completed by monitoring the current. Usually, when the monitoring shows that the charging current is too large, the charging current is already too large and has caused damage to the new energy battery. For this reason, we propose an intelligent fast charging device for electric vehicles based on Internet+. Summary of the invention

[0004] The purpose of the present invention is to provide an Internet+-based intelligent fast charging device for electric vehicles, which can obtain the physical data of electric vehicle batteries, establish a safety warning model based on the physical data, analyze and process the physical data through the safety warning model, predict in advance that charging failures will occur after a certain period of time, and issue timely warnings and early warnings to ensure the safety of electric vehicle charging.

[0005] The technical solution adopted by the present invention is as follows:

[0006] An Internet+-based intelligent fast charging device for electric vehicles comprises a charging pile and a charging gun. The charging gun is charged after being plugged into a charging interface of an electric vehicle. An intelligent fast charging early warning system runs on the charging pile. The intelligent fast charging early warning system comprises a charging terminal and a cloud service terminal. The charging terminal collects physical data of the charging gun in the process of charging the electric vehicle, and sends the physical data to the cloud service terminal through a wireless network. A safety early warning model is established inside the cloud service terminal. The cloud service terminal determines whether an abnormality will occur in the subsequent charging process according to the physical data through the safety early warning model. If an abnormality occurs, charging is stopped and an early warning is issued. If no abnormality occurs, the physical data of the charging terminal continues to be obtained.

[0007] Preferably, the car charging terminal is provided with a first data interface, and the charging gun is provided with a second data interface, and the second data interface is adapted to the first data interface.

[0008] Preferably, the charging end is provided with a data acquisition module, and the data acquisition module collects physical data of the electric vehicle battery during the charging process through the second data interface of the charging gun and the first data interface of the car charging end. The battery physical data includes battery power data, battery temperature data, battery internal resistance data, battery voltage data and battery pack pressure data.

[0009] Preferably, the cloud server first obtains the charging order of the charging pile, then normalizes the order, then obtains the physical data under each charging order, and divides the physical data into real-time data and historical data, and then feeds the real-time data and historical data back to the safety warning model.

[0010] Preferably, the safety warning model first sets up a neural network model, first uses the charging physical data in the historical data as the initial neuron data for the neural network model to learn, and then the neural network model data judges the result, and compares the judgment result with the actual result of the historical data. If they are different, the initial parameters of the neural network model are adjusted until the judgment result output by the neural network model is the same as the actual result in the historical data, and finally the neural network model is output.

[0011] Preferably, the real-time data is fed back to the safety warning model, and the safety warning model uses a neural network model to predict the real-time data. If the prediction result is that the electric vehicle battery fails after time t, a fault warning is generated. If the prediction result is that the electric vehicle battery is normal, monitoring continues.

[0012] Preferably, each prediction process of real-time data will be used as historical data to optimize the parameters of the neural network model.

[0013] Preferably, the cloud server is communicatively connected to the electric vehicle control system via a wireless network. When the electric vehicle is not charging, the cloud server can still obtain the physical data of the electric vehicle, and also use the safety warning model to issue a warning for the physical data of the electric vehicle when the electric vehicle is not charging.

[0014] Preferably, the cloud service end requires manual authorization from the electric vehicle end when acquiring the physical data of the electric vehicle in a non-charging state.

[0015] The technical effects achieved by the present invention are:

[0016] In the present invention, by setting up a safety warning model, real-time warning can be given for charging data, and early warning can be achieved. Since the prior art gives a warning when a fault occurs, the present invention gives an early warning of an impending fault before it occurs, thereby avoiding damage caused by battery failure and ensuring the safety of battery power use.

[0017] In the present invention, real-time data is predicted by the temperature-time relationship function, battery internal resistance-time relationship function, battery voltage-time relationship function and battery pack pressure-time relationship function in the neural network model, and a prediction is made every minute. When any one of the above functions predicts that a fault will occur after time t, the safety warning model directly issues a warning, and at the same time the cloud reading end controls the charging pile to turn off the power and stop charging. In actual use, it is possible to predict that a fault will occur 10 minutes in advance. For example, the real-time data reflects an abnormal temperature rise rate, and according to the temperature-time relationship function, it is concluded that the temperature will reach the warning temperature after 10 minutes, so it is possible to predict that a fault will occur 10 minutes in advance.

[0018] In the present invention, after obtaining authorization, the cloud service end can communicate data with the control system of the electric vehicle, and can also obtain the physical data of the electric vehicle battery. After obtaining the physical data, the workflow is the same as that in the charging process. The electric vehicle battery is given a real-time warning through a safety warning model, so that the present invention can not only give a real-time warning to the electric vehicle battery during the charging process, but also give a real-time warning during the driving of the electric vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a system block diagram of the intelligent fast charging device for electric vehicles in the present invention;

[0020] Figure 2 It is a system block diagram of the safety early warning model in the present invention;

[0021] Figure 3 It is a flow chart of the intelligent fast charging device for electric vehicles in the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose and advantages of the present invention more clearly understood, the present invention is specifically described below in conjunction with embodiments. It should be understood that the following text is only used to describe one or several specific embodiments of the present invention, and does not strictly limit the scope of protection of the specific claims of the present invention.

[0023] like Figure 1 As shown, an intelligent fast charging device for electric vehicles based on Internet+ includes a charging pile and a charging gun. The charging gun is charged after being plugged into the charging interface of the electric vehicle. An intelligent fast charging early warning system runs on the charging pile. The so-called intelligent fast charging early warning system includes a charging end and a cloud service end. The charging end collects physical data during the charging process of the charging gun to the electric vehicle, and sends the physical data to the cloud service end through a wireless network. A safety early warning model is established inside the cloud service end. The cloud service end uses the safety early warning model to determine whether there will be abnormalities in the subsequent charging process based on the physical data. If there is an abnormality, the charging is stopped and an early warning is issued. If there is no abnormality, the physical data of the charging end continues to be obtained.

[0024] In the present invention, by setting up a safety warning model, real-time warning can be given for charging data, and early warning can be achieved. Since the prior art gives a warning when a fault occurs, the present invention gives an early warning of an impending fault before it occurs, thereby avoiding damage caused by battery failure and ensuring the safety of battery power use.

[0025] The car charging terminal is provided with a first data interface, and the charging gun is provided with a second data interface, and the second data interface and the first data interface are adapted to each other.

[0026] In actual use of the present invention, when charging an electric vehicle, the second data interface on the charging gun is inserted into the first data interface of the charging end of the electric vehicle, so that during the charging process, the charging pile obtains the physical data of the electric vehicle battery in real time.

[0027] The charging end is provided with a data acquisition module, which collects the physical data of the electric vehicle battery during the charging process through the second data interface of the charging gun and the first data interface of the car charging end. The battery physical data includes battery power data, battery temperature data, battery internal resistance data, battery voltage data and battery pack pressure data.

[0028] In the present invention, temperature monitoring: a temperature sensor is used to timely monitor the temperature change of the battery to monitor the temperature. When the temperature abnormally rises and exceeds the critical temperature, it indicates that the electric vehicle battery has a fault; internal resistance monitoring: usually, when the battery operates within the normal temperature range, the internal resistance of the battery decreases with the increase of temperature, but when it exceeds the normal working range or even thermal runaway occurs, the internal resistance of the battery will increase significantly; therefore, by monitoring the internal resistance change of the battery, especially when the internal resistance increases significantly, it may indicate that the battery has a fault; voltage monitoring: the thermal runaway process of the battery will also cause abnormal changes in the voltage, eventually dropping to 0V; monitoring the voltage change of the battery can detect battery performance degradation or abnormal conditions; internal pressure monitoring: the gas and smoke generated during the thermal runaway of the battery cause internal pressure changes and bulging phenomena; therefore, the thermal runaway situation can be warned by monitoring the pressure changes caused by the generation of gas inside the battery.

[0029] The cloud server first obtains the charging orders of the charging piles, normalizes the orders, obtains the physical data under each charging order, divides the physical data into real-time data and historical data, and then feeds the real-time data and historical data back to the safety warning model.

[0030] like Figure 2 As shown, the safety warning model first sets up a neural network model, first uses the charging physical data in the historical data as the initial neuron data for the neural network model to learn, and then uses the neural network model data to judge the result, and compares the judgment result with the actual result of the historical data. If they are different, the initial parameters of the neural network model are adjusted until the judgment result output by the neural network model is the same as the actual result in the historical data, and finally outputs the neural network model; the real-time data is fed back to the safety warning model, and the safety warning model uses the neural network model to predict the real-time data. If the prediction result is that the electric vehicle battery fails after time t, a fault warning is generated. If the prediction result is that the electric vehicle battery is normal, monitoring continues.

[0031] In actual use, for example, for historical temperature data, a neural network model is used to obtain a temperature-time relationship function through an interpolation algorithm. Through the temperature-time relationship function, it is calculated that the temperature rises to the warning temperature after the corresponding time, and the result is proofread according to the historical real data. If the corresponding time when the temperature rises to the warning temperature is different from the actual result, the parameters of the temperature-time relationship function are adjusted, the temperature-time relationship function is optimized, and finally the temperature-time relationship function is output. Similarly, the battery power-time relationship function; the battery internal resistance-time relationship function, the battery voltage-time relationship function and the battery pack pressure-time relationship function are output, and then a complete neural network model is output.

[0032] like Figure 3As shown, in actual use, the present invention predicts real-time data through the temperature-time relationship function, battery internal resistance-time relationship function, battery voltage-time relationship function and battery pack pressure-time relationship function in the neural network model, and makes a prediction every minute. When any one of the above functions predicts that a fault will occur after time t, the safety warning model directly issues a warning, and at the same time the cloud reading end controls the charging pile to turn off the power and stop charging. In actual use, it can be predicted that a fault will occur 10 minutes in advance. For example, the real-time data reflects an abnormal temperature rise rate. According to the temperature-time relationship function, it is concluded that the temperature will reach the warning temperature after 10 minutes. Therefore, it can be predicted that a fault will occur 10 minutes in advance.

[0033] Each prediction process of real-time data will be used as historical data to optimize the parameters of the neural network model.

[0034] In the present invention, by continuously optimizing the parameters of the neural network model, the neural network model is made more precise and the output prediction results are more accurate.

[0035] The cloud server is connected to the electric vehicle control system through a wireless network. When the electric vehicle is not charging, the cloud server can still obtain the physical data of the electric vehicle, and also use the safety warning model to issue warnings for the physical data of the electric vehicle when the electric vehicle is not charging. The cloud server requires manual authorization from the electric vehicle when obtaining the physical data of the electric vehicle when the electric vehicle is not charging.

[0036] In the present invention, after obtaining authorization, the cloud service end can communicate data with the control system of the electric vehicle, and can also obtain the physical data of the electric vehicle battery. After obtaining the physical data, the workflow is the same as that in the charging process. The electric vehicle battery is given a real-time warning through a safety warning model, so that the present invention can not only give a real-time warning to the electric vehicle battery during the charging process, but also give a real-time warning during the driving of the electric vehicle.

[0037] like Figure 1-3 As shown, the working principle of the present invention is:

[0038] First: when charging an electric vehicle, the second data interface on the charging gun is inserted into the first data interface of the electric vehicle charging end, so that during the charging process, the charging pile can obtain the physical data of the electric vehicle battery in real time. The battery physical data includes battery power data, battery temperature data, battery internal resistance data, battery voltage data and battery pack pressure data; then, the physical data is sent to the cloud service end through the wireless network. A safety warning model is established inside the cloud service end. The safety warning model first sets a neural network model. For example, for historical temperature data, the neural network model is used to obtain the temperature-time relationship function through the interpolation algorithm. Through the temperature-time relationship function, the temperature rises to the warning temperature after the corresponding time, and the result is proofread according to the historical real data. If the corresponding time when the temperature rises to the warning temperature is different from the real result, the parameters of the temperature-time relationship function are adjusted, the temperature-time relationship function is optimized, and finally the temperature-time relationship function is output. Relationship function, similarly outputs battery power-time relationship function; battery internal resistance-time relationship function, battery voltage-time relationship function and battery pack pressure-time relationship function, and then outputs a complete neural network model; the real-time data is predicted through the temperature-time relationship function, battery internal resistance-time relationship function, battery voltage-time relationship function and battery pack pressure-time relationship function in the neural network model, and a prediction is made every minute. When any of the above functions predicts that a fault will occur after time t, the safety warning model directly issues a warning, and the cloud reading end controls the charging pile to turn off the power and stop charging. In actual use, it can predict that a fault will occur 10 minutes in advance. For example, the real-time data reflects an abnormal temperature rise rate. According to the temperature-time relationship function, the temperature will reach the warning temperature after 10 minutes, so the fault can be predicted 10 minutes in advance.

[0039] The above is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art unless otherwise specified and limited.

Claims

1. An intelligent fast charging device for electric vehicles based on Internet+, comprising a charging pile and a charging gun, wherein the charging gun is charged after being plugged into the charging interface of the electric vehicle, and is characterized in that: An intelligent fast charging warning system is running on the charging pile, and the intelligent fast charging warning system includes a charging end and a cloud service end. The charging end collects physical data of the charging gun during the charging of the electric vehicle, and sends the physical data to the cloud service end through a wireless network. A safety warning model is established inside the cloud service end. The cloud service end uses the safety warning model to determine whether there will be any abnormalities in the subsequent charging process based on the physical data. If there will be any abnormalities, the charging will be stopped and a warning will be issued. If there will be no abnormalities, the physical data of the charging end will continue to be obtained.

2. According to claim 1, an Internet+-based intelligent fast charging device for electric vehicles is characterized in that: The car charging terminal is provided with a first data interface, and the charging gun is provided with a second data interface, and the second data interface is adapted to the first data interface.

3. The electric vehicle intelligent fast charging device based on Internet+ according to claim 1 is characterized in that: The charging end is provided with a data acquisition module, and the data acquisition module collects physical data of the electric vehicle battery during the charging process through the second data interface of the charging gun and the first data interface of the vehicle charging end. The battery physical data includes battery power data, battery temperature data, battery internal resistance data, battery voltage data and battery pack pressure data.

4. The electric vehicle intelligent fast charging device based on Internet+ according to claim 1, characterized in that: The cloud server first obtains the charging order of the charging pile, then normalizes the order, obtains the physical data under each charging order, and divides the physical data into real-time data and historical data, and then feeds the real-time data and historical data back to the safety warning model.

5. The electric vehicle intelligent fast charging device based on Internet+ according to claim 1, characterized in that: The safety warning model first sets up a neural network model, first uses the charging physical data in the historical data as the initial neuron data for the neural network model to learn, and then the neural network model data judges the result, and compares the judgment result with the actual result of the historical data. If they are different, the initial parameters of the neural network model are adjusted until the judgment result output by the neural network model is the same as the actual result in the historical data, and finally the neural network model is output.

6. The electric vehicle intelligent fast charging device based on Internet+ according to claim 1, characterized in that: The real-time data is fed back to the safety warning model, and the safety warning model uses a neural network model to predict the real-time data. If the prediction result is that the electric vehicle battery fails after time t, a fault warning is generated. If the prediction result is that the electric vehicle battery is normal, monitoring continues.

7. The electric vehicle intelligent fast charging device based on Internet+ according to claim 1, characterized in that: Each prediction process of real-time data will be used as historical data to optimize the parameters of the neural network model.

8. The electric vehicle intelligent fast charging device based on Internet+ according to claim 1, characterized in that: The cloud service end is connected to the electric vehicle control system through a wireless network. When the electric vehicle is not charging, the cloud service end can still obtain the physical data of the electric vehicle, and also use the safety warning model to issue a warning for the physical data of the electric vehicle when the electric vehicle is not charging.

9. The electric vehicle intelligent fast charging device based on Internet+ according to claim 1, characterized in that: The cloud service end requires manual authorization from the electric vehicle end when acquiring the physical data of the electric vehicle in a non-charging state.