Early warning protection method of V2G charging pile in work

Abstract

The invention discloses an early warning protection method for a V2G charging pile in work, relates to the technical field of charging piles, and aims to perform accurate evaluation and early warning protection on the working state of the charging pile, reduce the downtime caused by faults, improve the overall reliability of the V2G charging pile and guarantee the normal charging requirement of an electric vehicle. The charging strategy is dynamically adjusted according to the power grid state and the electric vehicle charging demand, stable operation and load balance of the power grid are facilitated, and the utilization rate of power grid resources is improved. The service life of equipment in the charging pile can be prolonged through effective protection, and the equipment maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of charging piles, and particularly relates to a warning protection method for a V2G charging pile during operation. Background Art

[0002] With the popularization of electric vehicles (EVs) and the development of vehicle-to-grid (V2G) technology, as a key device connecting electric vehicles and the power grid, the safety and reliability of V2G charging piles face many challenges. During V2G charging, the charging pile needs to cope with a complex power grid environment, the charging requirements of different electric vehicles, and various possible fault conditions. For example, problems such as power grid voltage fluctuations, current overloads, and abnormal electric vehicle batteries can all affect the safety and stability of the charging process, and even cause serious consequences such as equipment damage and fires. Some existing charging pile protection technologies often focus on the detection and protection of single faults, lacking the ability to comprehensively analyze and early warn of multiple complex working conditions, and it is difficult to meet the requirements of efficient and safe operation of V2G charging piles. Summary of the Invention

[0003] The present invention provides a warning protection method for a V2G charging pile during operation, which improves the safety of the charging pile, enhances the reliability of the charging pile, and prolongs the service life of the charging pile equipment.

[0004] The technical solution adopted by the present invention to overcome its technical problems is as follows:

[0005] A warning protection method for a V2G charging pile during operation includes:

[0006] S1. The data acquisition module collects the parameters in the charging pile at time intervals of Δt to obtain the grid voltage U(t) at time t, the grid current I(t) at time t, the voltage Ub(t) of the electric vehicle battery at time t, the current Ib(t) of the electric vehicle battery at time t, the temperature Tb(t) of the electric vehicle battery at time t, and the temperature Tp(t) of the internal power device of the charging pile at time t;

[0007] S2. Filter the grid voltage U(t) at time t, the grid current I(t) at time t, the voltage Ub(t) of the electric vehicle battery at time t, the current Ib(t) of the electric vehicle battery at time t, the temperature Tb(t) of the electric vehicle battery at time t, and the temperature Tp(t) of the internal power device of the charging pile at time t to obtain the filtered grid voltage U′(t) at time t, the filtered grid current I′(t) at time t, the filtered voltage Ub′(t) of the electric vehicle battery at time t, the filtered current Ib′(t) of the electric vehicle battery at time t, the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the filtered temperature Tp′(t) of the internal power device of the charging pile at time t;

[0008] S3. Use the data analysis and processing module to calculate the grid voltage volatility ΔU′%, the grid current volatility ΔI′%, the voltage volatility ΔUb′% of the electric vehicle battery, the current volatility ΔIb′% of the electric vehicle battery, the temperature volatility ΔTb′% of the electric vehicle battery, and the temperature volatility ΔTp′% of the internal power device of the charging pile based on the filtered grid voltage U′(t) at time t, the filtered grid current I′(t) at time t, the filtered voltage Ub′(t) of the electric vehicle battery at time t, the filtered current Ib′(t) of the electric vehicle battery at time t, the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the filtered temperature Tp′(t) of the internal power device of the charging pile at time t;

[0009] S4. Input the grid voltage volatility ΔU′%, the grid current volatility ΔI′%, the voltage volatility ΔUb′% of the electric vehicle battery, the current volatility ΔIb′% of the electric vehicle battery, the temperature volatility ΔTb′% of the electric vehicle battery, and the temperature volatility ΔTp′% of the internal power device of the charging pile into the neural network evaluation model, and output the output layer neuron y l , and determine the current state of the charging pile according to the output layer neuron y l . Preferably, in step S1, the value of Δt is 10 milliseconds.

[0010] Further, in step S1, the grid voltage U(t) at time t and the grid current I(t) at time t are measured by using a voltage and current sensor installed at the incoming line end where the charging pile is connected to the grid.

[0011] Further, in step S1, the Hall effect sensor installed at the charging pile and the electric vehicle charging interface is used to measure the voltage Ub(t) of the electric vehicle battery at time t and the current Ib(t) of the electric vehicle battery at time t. Further, in step S1, the thermistor sensor installed at the battery management system BMS is used to measure the temperature Tb(t) of the electric vehicle battery at time t, and the thermistor sensor installed on the heat sink of the charging pile power module is used to measure the temperature Tp(t) of the internal power device of the charging pile at time t.

[0012] Further, in step S2, the Butterworth filter is used to filter the voltage U(t) of the power grid at time t to obtain the filtered voltage U′(t) of the power grid at time t, the Butterworth filter is used to filter the current I(t) of the power grid at time t to obtain the filtered current I′(t) of the power grid at time t, the Butterworth filter is used to filter the voltage Ub(t) of the electric vehicle battery at time t to obtain the filtered voltage Ub′(t) of the electric vehicle battery at time t, the Butterworth filter is used to filter the current Ib(t) of the electric vehicle battery at time t to obtain the filtered current Ib′(t) of the electric vehicle battery at time t, the Butterworth filter is used to filter the temperature Tb(t) of the electric vehicle battery at time t to obtain the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the Butterworth filter is used to filter the temperature Tp(t) of the internal power device of the charging pile at time t to obtain the filtered temperature Tp′(t) of the internal power device of the charging pile at time t. Further, step S3 includes the following steps:

[0013] S31. Calculate the power grid voltage volatility ΔU′% through the formula where U′ max is the maximum value of the voltage of the power grid, and U′ min is the minimum value of the voltage of the power grid;

[0014] S32. Calculate the voltage volatility ΔUb′% of the electric vehicle battery through the formula where Ub′ max is the maximum value of the voltage of the electric vehicle battery, and Ub′ min is the minimum value of the voltage of the electric vehicle battery;

[0015] S33. Calculate the power grid current volatility ΔI′% through the formula where I′(t - Δt) is the filtered current of the power grid at the previous moment of time t;

[0016] S34. Calculate through the formula The current volatility ΔIb′% of the electric vehicle battery is calculated, where Ib′(t - Δt) is the current of the filtered electric vehicle battery at the previous moment of time t;

[0017] S35. Through the formula The temperature volatility ΔTb′% of the electric vehicle battery is calculated, where Tb′(t - Δt) is the temperature of the filtered electric vehicle battery at the previous moment of time t;

[0018] S36. Through the formula The volatility ΔTp′% of the temperature of the internal power device of the charging pile is calculated, where Tp′(t - Δt) is the temperature of the internal power device of the charging pile at the previous moment of time t.

[0019] Further, step S4 includes the following steps:

[0020] S41. Take the grid voltage volatility ΔU′% as the input vector x1 of the neural network evaluation model, take the grid current volatility ΔI′% as the input vector x2 of the neural network evaluation model, take the voltage volatility ΔUb′% of the electric vehicle battery as the input vector x3 of the neural network evaluation model, take the current volatility ΔIb′% of the electric vehicle battery as the input vector x4 of the neural network evaluation model, take the temperature volatility ΔTb′% of the electric vehicle battery as the input vector x5 of the neural network evaluation model, and take the volatility ΔTp′% of the temperature of the internal power device of the charging pile as the input vector x6 of the neural network evaluation model;

[0021] S42. The neural network evaluation model has m neurons. Through the formula Calculate the j-th hidden layer neuron h j , j ∈ {1, 2,..., m}, f(·) is the ReLu function, x i is the i-th input vector of the neural network evaluation model, i ∈ {1, 2,..., 6}, w ij is the connection weight between the input vector and the hidden layer, b j is the bias of the j-th hidden layer neuron;

[0022] S43. Through the formula Calculate the output layer neuron y output by the neural network evaluation model l , where g(·) is the Softmax function, v jl is the connection weight between the j-th hidden layer neuron and the l-th output layer neuron, c lis the bias of the l-th output layer neuron, where l = 1, 2, 3; S44. Compare the magnitudes among the output layer neurons y1, y2, and y3. If the value of the output layer neuron y1 is the largest, it is determined that the charging pile is currently in a normal working state. If the value of the output layer neuron y2 is the largest, it is determined that the charging pile is currently in a slightly abnormal state. If the value of the output layer neuron y3 is the largest, it is determined that the charging pile is currently in a severely abnormal state.

[0023] Further, when the value of the output layer neuron y3 is the largest, a warning signal is sent through the warning module. The warning signal is real-time prompted by the voice and light alarm device, and the warning signal is sent to the remote monitoring center using the wireless communication module.

[0024] Further, it also includes a protection execution module disposed in the charging pile. When the warning module sends a warning signal, if the grid voltage volatility ΔU′% exceeds 10% and the duration exceeds 10 seconds, the protection execution module adjusts the power factor correction circuit of the charging pile to reduce the input current. If the temperature Tb(t) of the electric vehicle battery at time t is greater than 55°C and the current volatility ΔIb′% of the electric vehicle battery is greater than 20%, the protection execution module reduces the charging current of the charging pile and starts the cooling fan of the charging pile.

[0025] The beneficial effects of the present invention are:

[0026] 1. Improve safety: Discover potential fault risks in advance and give warnings, and take protection measures in a timely manner to effectively avoid safety accidents caused by grid abnormalities, battery failures, etc.

[0027] 2. Enhance reliability: Accurately evaluate and give warning protection to the working state of the charging pile, reduce the downtime caused by faults, improve the overall reliability of the V2G charging pile, and ensure the normal charging requirements of electric vehicles.

[0028] 3. Optimize vehicle-grid interaction: Dynamically adjust the charging strategy according to the grid state and the charging requirements of electric vehicles, which is beneficial to the stable operation of the grid and load balancing, and improves the utilization rate of grid resources.

[0029] 4. Extend the equipment life: Effective protection can extend the service life of the internal equipment of the charging pile and reduce the equipment maintenance cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is the method flow chart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0031] The following will further describe the present invention in conjunction with the attached Figure 1 to further illustrate the present invention.

[0032] An early warning protection method for a V2G charging pile during operation, including:

[0033] S1. The data acquisition module collects the parameters in the charging pile at time intervals of Δt to obtain the grid voltage U(t) at time t, the grid current I(t) at time t, the voltage Ub(t) of the electric vehicle battery at time t, the current Ib(t) of the electric vehicle battery at time t, the temperature Tb(t) of the electric vehicle battery at time t, and the temperature Tp(t) of the internal power device of the charging pile at time t.

[0034] S2. Filter the grid voltage U(t) at time t, the grid current I(t) at time t, the voltage Ub(t) of the electric vehicle battery at time t, the current Ib(t) of the electric vehicle battery at time t, the temperature Tb(t) of the electric vehicle battery at time t, and the temperature Tp(t) of the internal power device of the charging pile at time t respectively to obtain the filtered grid voltage U′(t) at time t, the filtered grid current I′(t) at time t, the filtered voltage Ub′(t) of the electric vehicle battery at time t, the filtered current Ib′(t) of the electric vehicle battery at time t, the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the filtered temperature Tp′(t) of the internal power device of the charging pile at time t.

[0035] S3. The data analysis and processing module calculates the grid voltage volatility ΔU′%, the grid current volatility ΔI′%, the voltage volatility ΔUb′% of the electric vehicle battery, the current volatility ΔIb′% of the electric vehicle battery, the temperature volatility ΔTb′% of the electric vehicle battery, and the temperature volatility ΔTp′% of the internal power device of the charging pile according to the filtered grid voltage U′(t) at time t, the filtered grid current I′(t) at time t, the filtered voltage Ub′(t) of the electric vehicle battery at time t, the filtered current Ib′(t) of the electric vehicle battery at time t, the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the filtered temperature Tp′(t) of the internal power device of the charging pile at time t.

[0036] S4. Input the grid voltage volatility ΔU′%, the grid current volatility ΔI′%, the voltage volatility ΔUb′% of the electric vehicle battery, the current volatility ΔIb′% of the electric vehicle battery, the temperature volatility ΔTb′% of the electric vehicle battery, and the temperature volatility ΔTp′% of the internal power device of the charging pile into the neural network evaluation model, and output the output layer neuron y l , according to the output layer neuron y lDetermine the current state of the charging pile. It can monitor the working state and related parameters of the charging pile in real time. By comprehensively analyzing various data, potential fault risks can be discovered in advance, and effective protection measures can be taken in a timely manner, thereby improving the safety, reliability and operating efficiency of the V2G charging pile, and ensuring the safe and stable interaction between the power grid and electric vehicles.

[0037] In one embodiment of the present invention, in step S1, the value of Δt is 10 milliseconds.

[0038] In one embodiment of the present invention, in step S1, the voltage U(t) of the power grid at time t and the current I(t) of the power grid at time t are measured by using voltage and current sensors installed at the incoming line end where the charging pile is connected to the power grid.

[0039] In one embodiment of the present invention, in step S1, the voltage Ub(t) of the electric vehicle battery at time t and the current Ib(t) of the electric vehicle battery at time t are measured by using Hall effect sensors installed at the charging interface between the charging pile and the electric vehicle.

[0040] In one embodiment of the present invention, in step S1, the temperature Tb(t) of the electric vehicle battery at time t is measured by using a thermistor sensor installed at the battery management system BMS, and the temperature Tp(t) of the internal power device of the charging pile at time t is measured by using a thermistor sensor installed on the heat sink of the charging pile power module.

[0041] In one embodiment of the present invention, in step S2, the Butterworth filter is used to filter the voltage U(t) of the power grid at time t to obtain the filtered voltage U′(t) of the power grid at time t, the Butterworth filter is used to filter the current I(t) of the power grid at time t to obtain the filtered current I′(t) of the power grid at time t, the Butterworth filter is used to filter the voltage Ub(t) of the electric vehicle battery at time t to obtain the filtered voltage Ub′(t) of the electric vehicle battery at time t, the Butterworth filter is used to filter the current Ib(t) of the electric vehicle battery at time t to obtain the filtered current Ib′(t) of the electric vehicle battery at time t, the Butterworth filter is used to filter the temperature Tb(t) of the electric vehicle battery at time t to obtain the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the Butterworth filter is used to filter the temperature Tp(t) of the internal power device of the charging pile at time t to obtain the filtered temperature Tp′(t) of the internal power device of the charging pile at time t.

[0042] In one embodiment of the present invention, step S3 includes the following steps:

[0043] S31. By the formula Calculate to obtain the power grid voltage volatility ΔU′%, where U′max is the maximum value of the grid voltage, U′ min is the minimum value of the grid voltage;

[0044] S32. Through the formula calculate the voltage volatility ΔUb′% of the electric vehicle battery. In the formula, Ub′ max is the maximum voltage of the electric vehicle battery, Ub′ min is the minimum voltage of the electric vehicle battery;

[0045] S33. Through the formula calculate the grid current volatility ΔI′%. In the formula, I′(t - Δt) is the filtered grid current at the previous moment of time t;

[0046] S34. Through the formula calculate the current volatility ΔIb′% of the electric vehicle battery. In the formula, Ib′(t - Δt) is the filtered current of the electric vehicle battery at the previous moment of time t;

[0047] S35. Through the formula calculate the temperature volatility ΔTb′% of the electric vehicle battery. In the formula, Tb′(t - Δt) is the filtered temperature of the electric vehicle battery at the previous moment of time t;

[0048] S36. Through the formula calculate the volatility ΔTp′% of the temperature of the power device inside the charging pile. In the formula, Tp′(t - Δt) is the temperature of the power device inside the charging pile at the previous moment of time t.

[0049] In an embodiment of the present invention, step S4 includes the following steps:

[0050] S41. Take the grid voltage volatility ΔU′% as the input vector x1 of the neural network evaluation model, take the grid current volatility ΔI′% as the input vector x2 of the neural network evaluation model, take the voltage volatility ΔUb′% of the electric vehicle battery as the input vector x3 of the neural network evaluation model, take the current volatility ΔIb′% of the electric vehicle battery as the input vector x4 of the neural network evaluation model, take the temperature volatility ΔTb′% of the electric vehicle battery as the input vector x5 of the neural network evaluation model, and take the volatility ΔTp′% of the temperature of the power device inside the charging pile as the input vector x6 of the neural network evaluation model;

[0051] S42. The neural network evaluation model has m neurons. Through the formula calculate the jth hidden layer neuron h j , j ∈ {1, 2,..., m}. Preferably, the value of m can be set to 100. f(·) is the ReLu function, xi is the i-th input vector of the neural network evaluation model, i∈{1,2,...,6}, w ij is the connection weight between the input vector and the hidden layer, b j is the bias of the jth hidden layer neuron;

[0052] S43. Through the formula Calculate the output layer neuron y of the neural network evaluation model output l , where g(·) is the Softmax function, v jl is the connection weight between the jth hidden layer neuron and the lth output layer neuron, c l is the bias of the lth output layer neuron, l = 1, 2, 3; S44. Compare the sizes of the output layer neuron y1, the output layer neuron y2, and the output layer neuron y3. If the value of the output layer neuron y1 is the largest, it is determined that the charging pile is currently in a normal working state. If the value of the output layer neuron y2 is the largest, it is determined that the charging pile is currently in a mild abnormal state. If the value of the output layer neuron y3 is the largest, it is determined that the charging pile is currently in a serious abnormal state.

[0053] In one embodiment of the present invention, when the value of the output layer neuron y3 is the largest, an early warning signal is issued through the early warning module. The early warning signal is prompted in real time through voice and light alarm devices, and the early warning signal is sent to the remote monitoring center using the wireless communication module.

[0054] One embodiment of the present invention further includes a protection execution module within the charging pile. When the early warning module issues a warning signal, if the grid voltage fluctuation rate ΔU′% exceeds 10% and lasts for more than 10 seconds, the protection execution module adjusts the charging pile's power factor correction circuit to reduce input current. If the electric vehicle battery temperature Tb(t) at time t is greater than 55°C and the electric vehicle battery current fluctuation rate ΔIb′% is greater than 20%, the protection execution module reduces the charging pile's charging current and activates the charging pile's cooling fan. Similarly, if the temperature of the power devices within the charging pile exceeds a set value, the protection execution module reduces output power to prevent damage to the power devices.

[0055] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A warning protection method for a V2G charging pile during operation, characterized in that, Including: S1. The data acquisition module collects the parameters in the charging pile at time intervals of Δt, obtaining the grid voltage U(t) at time t, the grid current I(t) at time t, the voltage Ub(t) of the electric vehicle battery at time t, the current Ib(t) of the electric vehicle battery at time t, the temperature Tb(t) of the electric vehicle battery at time t, and the temperature Tp(t) of the power device inside the charging pile at time t; S2. Perform filtering operations on the grid voltage U(t) at time t, the grid current I(t) at time t, the voltage Ub(t) of the electric vehicle battery at time t, the current Ib(t) of the electric vehicle battery at time t, the temperature Tb(t) of the electric vehicle battery at time t, and the temperature Tp(t) of the power device inside the charging pile at time t respectively, obtaining the filtered grid voltage U′(t) at time t, the filtered grid current I′(t) at time t, the filtered voltage Ub′(t) of the electric vehicle battery at time t, the filtered current Ib′(t) of the electric vehicle battery at time t, the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the filtered temperature Tp′(t) of the power device inside the charging pile at time t; S3. The data analysis and processing module calculates the grid voltage volatility ΔU′%, the grid current volatility ΔI′%, the voltage volatility ΔUb′% of the electric vehicle battery, the current volatility ΔIb′% of the electric vehicle battery, the temperature volatility ΔTb′% of the electric vehicle battery, and the temperature volatility ΔTp′% of the power device inside the charging pile according to the filtered grid voltage U′(t) at time t, the filtered grid current I′(t) at time t, the filtered voltage Ub′(t) of the electric vehicle battery at time t, the filtered current Ib′(t) of the electric vehicle battery at time t, the filtered temperature Tb′(t) of the electric vehicle battery at time t, and the filtered temperature Tp′(t) of the power device inside the charging pile at time t; S4. Input the grid voltage volatility ΔU′%, grid current volatility ΔI′%, voltage volatility ΔUb′% of the electric vehicle battery, current volatility ΔIb′% of the electric vehicle battery, temperature volatility ΔTb′% of the electric vehicle battery, and temperature volatility ΔTp′% of the power device inside the charging pile into the neural network evaluation model, and output the output layer neuron y l , and based on the output layer neuron y l judge the current state of the charging pile.

2. The early warning protection method of the V2G charging pile according to claim 1 during operation, characterized in that: In step S1, the value of Δt is 10 milliseconds.

3. The early warning protection method of the V2G charging pile according to claim 1 during operation is characterized in that: In step S1, the grid voltage U(t) at time t and the grid current I(t) at time t are measured by the voltage and current sensors installed at the incoming line end where the charging pile is connected to the grid.

4. The early warning protection method of the V2G charging pile according to claim 1 during operation is characterized in that: In step S1, the voltage Ub(t) of the electric vehicle battery at time t and the current Ib(t) of the electric vehicle battery at time t are measured by the Hall effect sensors installed at the charging interface between the charging pile and the electric vehicle.

5. The early warning protection method of the V2G charging pile according to claim 1 during operation, characterized in that: In step S1, the temperature Tb(t) of the electric vehicle battery at time t is measured by the thermistor sensor installed at the battery management system BMS, and the temperature Tp(t) of the power device inside the charging pile at time t is measured by the thermistor sensor installed on the heat sink of the charging pile power module.

6. The early warning protection method of the V2G charging pile according to claim 1 during operation is characterized in that: In step S2, the Butterworth filter is used to filter the grid voltage U(t) at time t to obtain the filtered grid voltage U′(t) at time t. The Butterworth filter is used to filter the grid current I(t) at time t to obtain the filtered grid current I′(t) at time t. The Butterworth filter is used to filter the voltage Ub(t) of the electric vehicle battery at time t to obtain the filtered voltage Ub′(t) of the electric vehicle battery at time t. The Butterworth filter is used to filter the current Ib(t) of the electric vehicle battery at time t to obtain the filtered current Ib′(t) of the electric vehicle battery at time t. The Butterworth filter is used to filter the temperature Tb(t) of the electric vehicle battery at time t to obtain the filtered temperature Tb′(t) of the electric vehicle battery at time t. The Butterworth filter is used to filter the temperature Tp(t) of the internal power device of the charging pile at time t to obtain the filtered temperature Tp′(t) of the internal power device of the charging pile at time t.

7. The early warning protection method of the V2G charging pile according to claim 1 during operation is characterized in that, Step S3 includes the following steps: S31. Calculate the grid voltage volatility ΔU′% through the formula where U′ max is the maximum value of the grid voltage, and U′ min is the minimum value of the grid voltage; S32. Calculate the voltage volatility ΔUb′% of the electric vehicle battery through the formula where Ub′ max is the maximum voltage of the electric vehicle battery, and Ub′ min is the minimum voltage of the electric vehicle battery; S33. Calculate the grid current volatility ΔI′% through the formula where I′(t - Δt) is the current of the filtered grid at the previous moment before the moment t; S34. Calculate the current volatility ΔIb′% of the electric vehicle battery through the formula where Ib′(t - Δt) is the current of the filtered electric vehicle battery at the previous moment before the t-th moment; S35. The temperature volatility ΔTb′% of the electric vehicle battery is calculated through the formula where Tb′(t - Δt) is the temperature of the filtered electric vehicle battery at the previous moment before the t-th moment; S36. Calculate the temperature volatility ΔTp′% of the internal power device of the charging pile through the formula where Tp′(t - Δt) is the temperature of the internal power device of the charging pile at the previous moment before the moment t.

8. The early warning protection method of the V2G charging pile according to claim 1 during operation, characterized in that, Step S4 includes the following steps: S41. The grid voltage volatility ΔU′% is used as the input vector x1 of the neural network evaluation model. The grid current volatility ΔI′% is used as the input vector x2 of the neural network evaluation model. The voltage volatility ΔUb′% of the electric vehicle battery is used as the input vector x3 of the neural network evaluation model. The current volatility ΔIb′% of the electric vehicle battery is used as the input vector x4 of the neural network evaluation model. The temperature volatility ΔTb′% of the electric vehicle battery is used as the input vector x5 of the neural network evaluation model. The volatility ΔTp′% of the temperature of the internal power device of the charging pile is used as the input vector x6 of the neural network evaluation model. S42. The neural network evaluation model has m neurons, and through the formula the j-th hidden layer neuron h is calculated j , where j ∈ {1, 2,..., m}, f(·) is the ReLu function, and x i is the i-th input vector of the neural network evaluation model, i ∈ {1, 2,..., 6}, w ij is the connection weight between the input vector and the hidden layer, and b j is the bias of the j-th hidden layer neuron; S43. Through the formula Calculate the output layer neuron y of the neural network evaluation model output l , where g(·) is the Softmax function, v jl is the connection weight between the jth hidden layer neuron and the lth output layer neuron, c l is the bias of the l-th output layer neuron, l = 1, 2, 3; S44. Compare the magnitudes of the output layer neurons y1, y2, and y3. If the value of the output layer neuron y1 is the largest, it is determined that the charging pile is currently in a normal working state. If the value of the output layer neuron y2 is the largest, it is determined that the charging pile is currently in a mildly abnormal state. If the value of the output layer neuron y3 is the largest, it is determined that the charging pile is currently in a severely abnormal state.

9. The early warning protection method of the V2G charging pile according to claim 8 during operation, characterized in that: When the value of the output layer neuron y3 is the largest, a warning signal is sent through the warning module. The warning signal is real-time prompted by the voice and light alarm device, and the warning signal is sent to the remote monitoring center using the wireless communication module.

10. The early warning protection method of the V2G charging pile according to claim 9 during operation is characterized in that: It also includes a protection execution module provided in the charging pile. When the warning module sends a warning signal, if the grid voltage volatility ΔU′% exceeds 10% and the duration exceeds 10 seconds, the protection execution module adjusts the power factor correction circuit of the charging pile to reduce the input current. If the temperature Tb(t) of the electric vehicle battery at time t is greater than 55°C and the current volatility ΔIb′% of the electric vehicle battery is greater than 20%, the protection execution module reduces the charging current of the charging pile and starts the cooling fan of the charging pile.

Images