New energy automobile Internet-of-Things intelligent charging equipment with wind, light, storage and charging functions

Abstract

The invention belongs to the field of intelligent charging, and particularly discloses a wind, light, storage and charging new energy automobile internet-of-things intelligent charging device which comprises a wind energy and light energy collection module, an energy storage module, a charging module, a user terminal module and a safety guarantee module. A deep neural network method based on a residual error network structure is used for monitoring the temperature of a battery pack of an automobile battery and an energy storage module in real time, automatic power off is achieved when abnormal high temperature occurs, misjudgment is reduced, the safety problem caused by overheating is effectively prevented, and personnel and property safety is guaranteed; the RGB color algorithm model and the YCbCr color algorithm model are used to segment the fire area and detect the center position of the fire, so that the fire area can be identified and segmented more accurately, the judgment accuracy is improved, and accurate target guidance is provided for subsequent actions such as fire extinguishment.

Description

Technical Field

[0001] The present invention relates to the field of intelligent charging, and specifically refers to an Internet of Things intelligent charging device for new energy vehicles that combines wind, light, storage and charging. Background Art

[0002] New energy vehicle IoT smart charging equipment is an intelligent device that provides charging services for new energy vehicles. It is connected to the Internet of Things to achieve remote monitoring, control and management. It has fast charging technology and can replenish more power for vehicles in a short time. However, existing new energy vehicle IoT smart charging equipment rarely detects abnormally high battery temperatures during charging. If abnormally high battery temperatures cannot be detected and handled in time, it may bring safety hazards, such as causing thermal runaway of the battery or even fire. Current fire extinguishing equipment mainly faces all charging equipment. Unnecessary fire extinguishing operations in non-fire areas may cause unnecessary damage and losses to new energy vehicles and charging equipment that are not on fire but are affected by fire extinguishing measures. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a wind, light, storage, and charging new energy vehicle IoT intelligent charging device. In view of the fact that the existing new energy vehicle IoT intelligent charging devices rarely detect abnormally high temperatures of batteries during charging, if the abnormally high temperatures of batteries cannot be detected and processed in time, it may bring safety hazards, such as causing thermal runaway of batteries or even fires. The present invention uses a deep neural network method based on a residual network structure to monitor the temperature of the battery pack of the vehicle battery and the energy storage module in real time, automatically cuts off the power when abnormally high temperatures occur, reduces misjudgment, effectively prevents safety problems caused by overheating, and ensures the safety of personnel and property; in view of the technical problem that the current fire extinguishing equipment mainly faces all charging equipment, performs unnecessary fire extinguishing operations in non-fire areas, and may cause unnecessary damage and losses to new energy vehicles and charging equipment that are not on fire but affected by fire extinguishing measures, the present invention uses RGB color algorithm model and YCbCr color algorithm model to segment the fire area and detect the center position of the fire, which can more accurately identify and segment the fire area, improve the accuracy of judgment, and provide accurate target guidance for subsequent fire extinguishing and other actions.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides a wind, light, storage, and charging new energy vehicle Internet of Things intelligent charging device, the wind, light, storage, and charging new energy vehicle Internet of Things intelligent charging device comprises a wind energy and light energy collection module, an energy storage module, a charging module, a user terminal module, and a safety assurance module;

[0005] The wind energy and light energy collection module is equipped with a wind power generation device and a solar panel, and converts wind energy and light energy into electrical energy through a conversion device;

[0006] The energy storage module is equipped with a battery pack for storing the electric energy generated by the wind power generation device and the solar panel, and providing electric energy when the wind energy and light energy are insufficient;

[0007] The charging module is installed with a charging interface adapted to new energy vehicles to charge new energy vehicles. When the vehicle battery is overheated, the current is too large, or the battery has abnormal fluctuations, the power is automatically cut off, and the user is notified in the user terminal module;

[0008] The user terminal module uses mobile network communication technology to connect to the charging module to achieve remote monitoring, data transmission and user interaction. The user can view the energy storage status of the energy storage module and the charging status of the car on the user terminal, and make reservations for charging, start charging and pause charging.

[0009] The safety guarantee module includes a high-definition camera and a linkage fire unit. The high-definition camera is connected to the user terminal module, and the user can view the monitoring video in real time. The linkage fire unit includes a fire smoke sensor, a fire alarm camera, an intelligent fire ejector and a fire curtain. The fire alarm camera can remotely capture the fire hazards of new energy vehicles and battery packs and convey instructions to the intelligent fire ejector and fire curtain for fire extinguishing and isolation;

[0010] The charging module is provided with an infrared radiation camera, and uses a deep neural network method based on a residual network structure to check the temperature of the battery pack of the vehicle battery and the energy storage module, and automatically cuts off the power when an abnormally high temperature occurs. The deep neural network method based on a residual network structure specifically includes the following steps:

[0011] Step A1: collecting a battery infrared image set, wherein the battery infrared image set includes infrared images of a vehicle battery and a battery pack and corresponding status labels, wherein the status labels include normal and abnormal categories;

[0012] Step A2: Create and initialize a convolutional neural network as a high temperature anomaly classification model, with the ResNet structure as the backbone of the high temperature anomaly classification model. The input is an infrared image with a pixel size of 40×24 and a channel number of 1. The infrared image is enhanced and fed into the convolution layer, batch normalization layer, and ReLu activation layer. The infrared image is downsampled to make the pixel size of the infrared image 19×22, and the number of channels is increased to F, which is a multiple of 32.

[0013] Step A3: design two residual block strings, each of which has three residual blocks, and the step sizes of the residual blocks are 2, 1, and 1 respectively. Apply the two residual block strings to the infrared image, and each residual block string pauses the sampling of the image, and increases the number of channels to 2F. The two residual block strings output an infrared image with 5×6 pixels and 2F channels.

[0014] Step A4: The spatial resolution of the infrared image is merged through the global average pooling layer, and the 2F feature is output. The probability of each state label of the infrared image is calculated through the Softmax layer, and the state label with the highest probability is output. The weight parameters of all convolutional layers are L2-regularized with a weight factor of 0.01.

[0015] Step A5: Assume that the loss function of the high temperature anomaly classification model is the classification cross entropy function, and increase the loss function weight of the minority class and reduce the loss function weight of the majority class to reduce overfitting. The formula used is as follows: ;

[0016] In the formula, is the loss function, is the total number of infrared images, is the number of categories of state labels, It is The number of infrared images with the same state labels, is the state label corresponding to the infrared image, is the state label predicted by the high temperature anomaly classification model, It is The weight coefficient of the state label;

[0017] Step A6: The infrared radiation camera takes high-temperature infrared photos of the car battery and battery pack in real time, and the high-temperature anomaly classification model detects and classifies the high-temperature infrared photos. When the high-temperature infrared photo output result is an abnormal category, the charging module automatically cuts off the power.

[0018] The fire camera in the safety assurance module uses a wavelet analysis method based on image processing to detect the fire area of ​​the new energy vehicle and the battery pack. The wavelet analysis method based on image processing specifically includes the following steps:

[0019] Step B1: A fire camera is installed above the battery pack and each charging port. The fire camera captures an original image, and the original image is denoised using a two-dimensional wavelet transform. The original image is enhanced into an RGB color image using an RGB model color algorithm, and the chroma and brightness of the RGB color image are enhanced at the same time. The color space of the RGB color image is converted from RGB to YCbCr, and each RGB component is transformed into a Y component, a Cb component, and a Cr component.

[0020] Step B2: Set the fire area segmentation rules to segment the fire area. The formula used is as follows: ; ;

[0021] In the formula, and They are fire area segmentation rule 1 and fire area segmentation rule 2. They are the horizontal and vertical coordinates of the pixel points of the RGB color image, is an RGB color image, is the Y component, is the Cb component, is the Cr component, is the mean value of the Y component, is the average value of the Cr component;

[0022] Step B3: Set the fire center detection rule to detect the fire center. The formula used is as follows: ; ;

[0023] In the formula, and They are Fire Center Detection Rule 3 and Fire Center Detection Rule 4, is the standard deviation of the Cr component;

[0024] Step B4: If the fire center is detected, all charging ports are automatically powered off. At the same time, the fire curtains on both sides of the parking space where the fire center is located are opened to isolate the fire, and the intelligent fire sprayer above the parking space is opened to extinguish the fire.

[0025] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0026] (1) In view of the fact that existing new energy vehicle IoT smart charging devices rarely detect abnormally high battery temperatures during the charging process, if abnormally high battery temperatures cannot be detected and handled in a timely manner, they may pose safety hazards, such as causing thermal runaway or even fires. The present invention uses a deep neural network method based on a residual network structure to monitor the temperature of the battery pack of the vehicle battery and the energy storage module in real time. Based on the deep neural network, the temperature judgment is more accurate. When abnormally high temperatures occur, the power is automatically cut off to reduce misjudgment, effectively prevent safety problems caused by overheating, and ensure the safety of personnel and property.

[0027] (2) In view of the technical problem that current fire extinguishing equipment mainly faces all charging equipment and performs unnecessary fire extinguishing operations in non-fire areas, which may cause unnecessary damage and losses to new energy vehicles and charging equipment that are not on fire but are affected by fire extinguishing measures, the present invention uses RGB color algorithm model and YCbCr color algorithm model to segment the fire area and detect the center position of the fire, which can more accurately identify and segment the fire area, improve the accuracy of judgment, and provide accurate target guidance for subsequent fire extinguishing and other actions. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A module connection diagram of a wind, light, storage and charging new energy vehicle IoT intelligent charging device provided by the present invention.

[0029] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0031] Example 1: See Figure 1 This embodiment provides a wind, light, storage, and charging new energy vehicle IoT smart charging device, which includes a wind, light, storage, and charging new energy vehicle IoT smart charging device, including a wind and light energy collection module, an energy storage module, a charging module, a user terminal module, and a safety assurance module;

[0032] The wind energy and light energy collection module is equipped with a wind power generation device and a solar panel, and converts wind energy and light energy into electrical energy through a conversion device;

[0033] The energy storage module is equipped with a battery pack for storing the electric energy generated by the wind power generation device and the solar panel, and providing electric energy when the wind energy and light energy are insufficient;

[0034] The charging module is installed with a charging interface adapted to new energy vehicles to charge new energy vehicles. When the vehicle battery is overheated, the current is too large, or the battery has abnormal fluctuations, the power is automatically cut off, and the user is notified in the user terminal module;

[0035] The user terminal module uses mobile network communication technology to connect to the charging module to achieve remote monitoring, data transmission and user interaction. The user can view the energy storage status of the energy storage module and the charging status of the car on the user terminal, and make reservations for charging, start charging and pause charging.

[0036] The security module includes a high-definition camera and a linkage fire-fighting unit. The high-definition camera is connected to the user terminal module, and the user can view the monitoring video in real time. The linkage fire-fighting unit includes a fire smoke sensor, a fire alarm camera, an intelligent fire-fighting ejector and a fire-fighting fire-proof curtain. The fire alarm camera can remotely capture fire hazards of new energy vehicles and battery packs and convey instructions to the intelligent fire-fighting ejector and the fire-fighting fire-proof curtain for fire extinguishing and isolation.

[0037] Example 2: See Figure 1 This embodiment is based on the above embodiment. The charging module is provided with an infrared radiation camera. A deep neural network method based on a residual network structure is used to check the temperature of the battery pack of the vehicle battery and the energy storage module. When an abnormally high temperature occurs, the power is automatically cut off. The deep neural network method based on a residual network structure specifically includes the following steps:

[0038] Step A1: collecting a battery infrared image set, wherein the battery infrared image set includes infrared images of a vehicle battery and a battery pack and corresponding status labels, wherein the status labels include normal and abnormal categories;

[0039] Step A2: Create and initialize a convolutional neural network as a high temperature anomaly classification model, with the ResNet structure as the backbone of the high temperature anomaly classification model. The input is an infrared image with a pixel size of 40×24 and a channel number of 1. The infrared image is enhanced and fed into the convolution layer, batch normalization layer, and ReLu activation layer. The infrared image is downsampled to make the pixel size of the infrared image 19×22, and the number of channels is increased to F, which is a multiple of 32.

[0040] Step A3: design two residual block strings, each of which has three residual blocks, and the step sizes of the residual blocks are 2, 1, and 1 respectively. Apply the two residual block strings to the infrared image, and each residual block string pauses the sampling of the image, and increases the number of channels to 2F. The two residual block strings output an infrared image with 5×6 pixels and 2F channels.

[0041] Step A4: The spatial resolution of the infrared image is merged through the global average pooling layer, and the 2F feature is output. The probability of each state label of the infrared image is calculated through the Softmax layer, and the state label with the highest probability is output. The weight parameters of all convolutional layers are L2-regularized with a weight factor of 0.01.

[0042] Step A5: Assume that the loss function of the high temperature anomaly classification model is the classification cross entropy function, and increase the loss function weight of the minority class and reduce the loss function weight of the majority class to reduce overfitting. The formula used is as follows: ;

[0043] In the formula, is the loss function, is the total number of infrared images, is the number of categories of state labels, It is The number of infrared images with the same state labels, is the state label corresponding to the infrared image, is the state label predicted by the high temperature anomaly classification model, It is The weight coefficient of the state label;

[0044] Step A6: The infrared radiation camera takes high-temperature infrared photos of the car battery and battery pack in real time, and the high-temperature anomaly classification model detects and classifies the high-temperature infrared photos. When the high-temperature infrared photo output result is an abnormal category, the charging module automatically cuts off the power.

[0045] Through the above operations, the existing new energy vehicle IoT smart charging equipment rarely detects abnormally high temperatures of batteries during the charging process. If the abnormally high temperatures of batteries cannot be detected and handled in time, it may bring safety hazards, such as causing thermal runaway of batteries or even fires and other dangerous situations. The present invention uses a deep neural network method based on a residual network structure to monitor the temperature of the battery pack of the vehicle battery and the energy storage module in real time, and based on the deep neural network, the temperature judgment is more accurate. When abnormally high temperatures occur, the power is automatically cut off, which reduces misjudgment, effectively prevents safety problems caused by overheating, and ensures the safety of people and property.

[0046] Example 3: See Figure 1 This embodiment is based on the fire camera in the safety assurance module described in the above embodiment using a wavelet analysis method based on image processing to detect the fire area of ​​the new energy vehicle and the battery pack. The wavelet analysis method based on image processing specifically includes the following steps:

[0047] Step B1: A fire camera is installed above the battery pack and each charging port. The fire camera captures an original image, and the original image is denoised using a two-dimensional wavelet transform. The original image is enhanced into an RGB color image using an RGB model color algorithm, and the chroma and brightness of the RGB color image are enhanced at the same time. The color space of the RGB color image is converted from RGB to YCbCr, and each RGB component is transformed into a Y component, a Cb component, and a Cr component.

[0048] Step B2: Set the fire area segmentation rules to segment the fire area. The formula used is as follows: ; ;

[0049] In the formula, and They are fire area segmentation rule 1 and fire area segmentation rule 2. They are the horizontal and vertical coordinates of the pixel points of the RGB color image, is an RGB color image, is the Y component, is the Cb component, is the Cr component, is the mean value of the Y component, is the average value of the Cr component;

[0050] Step B3: Set the fire center detection rule to detect the fire center. The formula used is as follows: ; ;

[0051] In the formula, and They are Fire Center Detection Rule 3 and Fire Center Detection Rule 4, is the standard deviation of the Cr component;

[0052] Step B4: If the fire center is detected, all charging ports are automatically powered off. At the same time, the fire curtains on both sides of the parking space where the fire center is located are opened to isolate the fire, and the intelligent fire sprayer above the parking space is opened to extinguish the fire.

[0053] Through the above operations, in view of the technical problem that the current fire extinguishing equipment mainly faces all charging equipment and performs unnecessary fire extinguishing operations in non-fire areas, which may cause unnecessary damage and losses to new energy vehicles and charging equipment that are not on fire but are affected by fire extinguishing measures, the present invention uses RGB color algorithm model and YCbCr color algorithm model to segment the fire area and detect the center position of the fire, which can more accurately identify and segment the fire area, improve the accuracy of judgment, and provide accurate target guidance for subsequent fire extinguishing and other actions.

[0054] Embodiment 4: This embodiment is based on the above embodiment. The fire camera in the security module in the above embodiment 1 is of Raspberry Pi camera Module V2, and the lens pixel is 8 million.

[0055] Embodiment 5: This embodiment is based on the above embodiment. The main structure of the residual block in the above embodiment 2 has two layer sequences (Norm, ReLu, Conv). Each convolution layer has the same number of filters as the input infrared image. The Stride (S=1, 2) in the convolution layer determines the output spatial samples: if S=1, it is the same as the input space, and if S=2, it is half of the input space.

[0056] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0057] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0058] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A wind, light, storage and charging new energy vehicle IoT intelligent charging device, characterized in that: It includes wind energy and light energy collection module, energy storage module, charging module, user terminal module and safety assurance module; The wind energy and light energy collection module is equipped with a wind power generation device and a solar panel, and converts wind energy and light energy into electrical energy through a conversion device; The energy storage module is equipped with a battery pack for storing the electric energy generated by the wind power generation device and the solar panel, and providing electric energy when the wind energy and light energy are insufficient; The charging module is installed with a charging interface adapted to new energy vehicles to charge new energy vehicles. When the vehicle battery is overheated, the current is too large, or the battery has abnormal fluctuations, the power is automatically cut off, and the user is notified in the user terminal module; The user terminal module uses mobile network communication technology to connect to the charging module to achieve remote monitoring, data transmission and user interaction. The user can view the energy storage status of the energy storage module and the charging status of the car on the user terminal, and make reservations for charging, start charging and pause charging. The security module includes a high-definition camera and a linkage fire-fighting unit. The high-definition camera is connected to the user terminal module, and the user can view the monitoring video in real time. The linkage fire-fighting unit includes a fire smoke sensor, a fire alarm camera, an intelligent fire-fighting ejector and a fire-fighting fire-proof curtain. The fire alarm camera can remotely capture fire hazards of new energy vehicles and battery packs and convey instructions to the intelligent fire-fighting ejector and the fire-fighting fire-proof curtain for fire extinguishing and isolation.

2. According to claim 1, a wind, light, storage, and charging new energy vehicle IoT intelligent charging device is characterized in that: The charging module is provided with an infrared radiation camera, and uses a deep neural network method based on a residual network structure to check the temperature of the battery pack of the vehicle battery and the energy storage module, and automatically cuts off the power when an abnormally high temperature occurs. The deep neural network method based on a residual network structure specifically includes the following steps: Step A1: collecting a battery infrared image set, wherein the battery infrared image set includes infrared images of a vehicle battery and a battery pack and corresponding status labels, wherein the status labels include normal and abnormal categories; Step A2: Create and initialize a convolutional neural network as a high temperature anomaly classification model, with the ResNet structure as the backbone of the high temperature anomaly classification model. The input is an infrared image with a pixel size of 40×24 and a channel number of 1. The infrared image is enhanced and fed into the convolution layer, batch normalization layer, and ReLu activation layer. The infrared image is downsampled to make the pixel size of the infrared image 19×22, and the number of channels is increased to F, which is a multiple of 32. Step A3: design two residual block strings, each of which has three residual blocks, and the step sizes of the residual blocks are 2, 1, and 1 respectively. Apply the two residual block strings to the infrared image, and each residual block string pauses the sampling of the image, and increases the number of channels to 2F. The two residual block strings output an infrared image with 5×6 pixels and 2F channels. Step A4: The spatial resolution of the infrared image is merged through the global average pooling layer, and the 2F feature is output. The probability of each state label of the infrared image is calculated through the Softmax layer, and the state label with the highest probability is output. The weight parameters of all convolutional layers are L2-regularized with a weight factor of 0.

01. Step A5: Assume that the loss function of the high temperature anomaly classification model is the classification cross entropy function, and increase the loss function weight of the minority class and reduce the loss function weight of the majority class to reduce overfitting. The formula used is as follows: ； In the formula, is the loss function, is the total number of infrared images, is the number of categories of state labels, It is The number of infrared images with the state labels, is the state label corresponding to the infrared image, is the state label predicted by the high temperature anomaly classification model, It is The weight coefficient of the state label; Step A6: The infrared radiation camera takes high-temperature infrared photos of the car battery and battery pack in real time, and the high-temperature anomaly classification model detects and classifies the high-temperature infrared photos. When the high-temperature infrared photo output result is an abnormal category, the charging module automatically cuts off the power.

3. According to claim 2, a wind, light, storage, and charging new energy vehicle IoT intelligent charging device is characterized in that: The fire camera in the safety assurance module uses a wavelet analysis method based on image processing to detect the fire area of ​​the new energy vehicle and the battery pack. The wavelet analysis method based on image processing specifically includes the following steps: Step B1: A fire camera is installed above the battery pack and each charging port. The fire camera captures an original image, and the original image is denoised using a two-dimensional wavelet transform. The original image is enhanced into an RGB color image using an RGB model color algorithm, and the chroma and brightness of the RGB color image are enhanced at the same time. The color space of the RGB color image is converted from RGB to YCbCr, and each RGB component is transformed into a Y component, a Cb component, and a Cr component. Step B2: Set the fire area segmentation rules to segment the fire area. The formula used is as follows: ； ； In the formula, and They are fire area segmentation rule 1 and fire area segmentation rule 2. They are the horizontal and vertical coordinates of the pixel points of the RGB color image, is an RGB color image, is the Y component, is the Cb component, is the Cr component, is the mean value of the Y component, is the average value of the Cr component; Step B3: Set the fire center detection rule to detect the fire center. The formula used is as follows: ； ； In the formula, and They are Fire Center Detection Rule 3 and Fire Center Detection Rule 4, is the standard deviation of the Cr component; Step B4: If the fire center is detected, all charging ports are automatically powered off. At the same time, the fire curtains on both sides of the parking space where the fire center is located are opened to isolate the fire, and the intelligent fire sprayer above the parking space is opened to extinguish the fire.

Images