Efficient battery heat management system

Abstract

The invention discloses an efficient battery heat management system, which comprises the following steps that the perimeter of a battery is obtained based on big data, a test slope is sequentially divided into a detection area, a verification area and a test area in the inclination direction, the test slope is provided with a plurality of air outlet holes, the length of the detection area and the length of the test area are the perimeter of a single battery, and the length of the verification area is smaller than that of the test area; the length of the verification area is n battery perimeters; enabling the wind tunnel power in the detection area to be the same, and testing the dynamic balance of the battery in the detection area; according to the method, the battery with battery heat on the surface can be rapidly analyzed without manual work, and the sunken type and the sunken part of the battery can be obtained; according to the method, the batteries are verified for multiple times, so that the analysis accuracy is high, the batteries can be automatically classified according to the analysis condition after the analysis is completed, the battery heat management efficiency is greatly improved, and the labor cost is reduced.

Description

Technical Field

[0001] This application relates to the field of battery thermal management, and in particular to an efficient battery thermal management system. Background Art

[0002] With the development of society and the progress of battery technology, high-power batteries are increasingly widely used as the power source for electric vehicles, unmanned aerial vehicles, and high-performance equipment. However, power batteries have relatively high requirements for the working environment temperature sensitivity and service life. If the working environment temperature cannot be controlled within a reasonable range, resulting in too high or too low internal temperature of the battery module, or uneven internal temperature distribution, it will not only affect the performance and service life of the battery, but also pose a safety hazard.

[0003] Currently, with the development of big data technology, big data-based battery thermal management methods have been gradually introduced, which have higher accuracy and reliability. By analyzing and processing a large amount of battery data, more comprehensive battery information and performance can be obtained, improving the ability of battery fault detection and prediction.

[0004] In the related art, the detection of the appearance defects of cylindrical batteries mainly obtains conventional data based on big data, and then measures and experiments are carried out manually. When manually detecting cylindrical batteries, only convex-shaped defects can be detected well. Since some defects are in a concave shape, they are not easily found.

[0005] How to quickly analyze the batteries with battery heat on the surface and improve the efficiency of the entire detection process is an urgent problem to be solved and optimized in battery thermal management. Summary of the Invention

[0006] In order to quickly analyze the batteries with battery heat on the surface and improve the efficiency of the entire detection process, this application provides an efficient battery thermal management system.

[0007] The efficient battery thermal management system provided by this application adopts the following technical solutions:

[0008] In the first aspect, a big data-based battery thermal management method includes the following steps:

[0009] Obtain the battery circumference based on big data, and sequentially divide the test inclined plane into a detection area, a verification area, and a test area in the inclined direction. Among them, several air outlet holes are provided on the test inclined plane, the lengths of the inspection area and the test area are both the circumference of a single battery, and the length of the verification area is n times the circumference of a single battery;

[0010] Make the power of the air outlet holes in the detection area the same, and test the dynamic balance of the battery in the detection area;

[0011] If the dynamic imbalance of the battery in the detection area is measured, the concave defect information and the corresponding air outlet tunnel array are determined according to the degree of battery imbalance;

[0012] The verification area is divided into n sub-verification areas according to the battery circumference, and the air outlet tunnel matching power of the n sub-verification areas is determined according to the air outlet tunnel array determined in the detection area;

[0013] Taking the concave defect information as the initial verification object, testing the dynamic balance of the battery in the n sub-verification areas, obtaining a dynamic balance result set, and analyzing the dynamic balance result set to generate the verified battery heat information;

[0014] Allocate the air outlet tunnel power in the test area according to the verified battery heat information, and predict the battery pre-test stop point area;

[0015] Measure the actual battery stop area. If it is the same as the pre-test stop point area, output the verified battery heat information and collect the batteries in each stop point area.

[0016] Preferably in any of the above solutions, making the air outlet tunnel power in the detection area the same and testing the dynamic balance of the battery in the detection area includes the following steps:

[0017] Encode the air outlet tunnels in the detection area according to the battery surface and assign the same air pressure to the air outlet tunnels;

[0018] Real-time collect the motion images of the battery in the detection area, and preprocess the motion images and extract the battery contour through related technologies;

[0019] Select each pixel in two consecutive motion images, find the corresponding pixel in the next frame image, and calculate the displacement of the battery in two adjacent motion images through pixel point translation;

[0020] Model the displacement variables of consecutive frames to obtain the motion model of the current transformer.

[0021] Preferably in any of the above solutions, the step of selecting each pixel in two consecutive motion images, finding the corresponding pixel in the next frame image, and calculating the displacement of the battery in two adjacent motion images through pixel point translation includes the following steps:

[0022] Extract the battery feature points through related technologies. Let the pixel coordinates of the feature points in the t-th frame image be (x, y), and the displacement of the feature points (x, y) from the t-th frame image to the t + 1-th frame image be characterized by the optical flow vector Δv = (Δx, Δy);

[0023] Solve the optical flow vector Δv through the formula: where A is the coefficient matrix in matrix form, (u, v) are the pixel coordinates of the feature points, I x and I yThey are the gradients at the pixel positions of the feature points respectively.

[0024] In any of the above solutions, preferably, the motion model of the current device is obtained by modeling the displacement variables of consecutive frames, including the following steps:

[0025] For each moment t, let the position of the battery in the camera coordinate system be (x t , y t , z t ), and the motion model of the battery between two adjacent frames is obtained: where Δx t , Δy t and Δz t are the displacement vectors of the battery in the camera coordinate system in two adjacent frames of images;

[0026] The motion model between two adjacent frames is extended to the entire time series to obtain the motion model of the entire time series:

[0027] Among them, the matrix in the t-th row is the motion model between two adjacent frames, is the sum of the displacement vectors of the battery in the camera coordinate system in the first t frames of images;

[0028] Initialize the position of the battery in the camera coordinate system as (x 0 , y 0 , z 0 ), set the initial timestamp as t 0 , and for each subsequent moment t i , let the optical flow vector between the previous moment t i-1 be;

[0029] According to the motion model of the entire time series, calculate the displacement vector of the battery in the camera coordinate system; and use Δx i , Δy i , Δz i as the feature vector, use t i - t 0 as the eigenvalue, and form all of them into the feature set D;

[0030] Perform regression on the feature set D through a regression algorithm to obtain the motion model of the battery in the camera coordinate system, where is the predicted position of the battery at time t i , and f(t i ) is the motion model of the battery at time t i .

[0031] In any of the above solutions, preferably, if it is determined that the battery is dynamically unbalanced in the detection area, the concave defect information and the corresponding air outlet tunnel array are determined according to the degree of battery imbalance, including the following steps:

[0032] Obtain the predicted position and actual position of the battery at the same moment, and calculate the difference e through the formula: t , where ||·|| 2 is the Euclidean distance;

[0033] If the difference e t is greater than the difference threshold e, it is determined that the battery has dynamic imbalance at time t;

[0034] Output the battery part corresponding to the air outlet hole group column experienced by the battery at time t as the concave part, and judge the difference e according to the concave type difference interval t The concave type to which it belongs, where the battery heat information includes the concave type and the concave part.

[0035] Preferably, in any of the above solutions, the verification area is divided into n groups of sub-verification areas according to the battery perimeter, and the air outlet holes of the n groups of sub-verification areas are matched with the power according to the air outlet hole group column determined by the detection area, including the following steps:

[0036] Pair the air outlet holes corresponding to the concave part with the air outlet holes in the n groups of sub-verification areas to map the concave part of the battery onto the air outlet holes of each group of sub-verification areas;

[0037] Control the air pressure of the corresponding air outlet holes in the n groups of sub-verification areas to be the same as the air pressure of the air outlet holes in the detection area to verify the concave type of the battery n times.

[0038] Preferably, in any of the above solutions, using the concave defect information as the initial verification object, testing the dynamic balance of the battery in the n groups of sub-verification areas, obtaining the dynamic balance result set, and analyzing the dynamic balance result set to generate the verified battery heat information, including the following steps:

[0039] Taking the concave type as the verification object, the air pressure of the air outlet hole corresponding to the concave part as the verification parameter, and constructing the motion model of the battery in each group of sub-verification areas;

[0040] Through the motion model of each sub-verification area, calculate the difference between the actual position and the predicted position of the battery at the corresponding air outlet hole;

[0041] Statistically analyze the differences of the n verification sub-areas, calculate the average difference of the verification area, and take the concave type to which the average difference of the verification area belongs as the verified concave type.

[0042] Preferably, in any of the above solutions, distributing the air outlet hole power in the test area according to the verified battery heat information, and pre-testing the battery stay point area, including the following steps:

[0043] Preset the battery stay area and mark the battery stay area according to the concave type category;

[0044] Build a motion model of the battery in the test area when there is no air pressure in the air outlet tunnel of the test area, and predict the final position of the battery passing through the test area;

[0045] Allocate the air pressure in the air outlet tunnel in the test area to change the predicted final position of the battery passing through the test area to the battery staying area that meets the verified depression types.

[0046] Preferably, in any of the above solutions, measure the actual staying area of the battery. If it is the same as the pre-tested staying point area, output the verified battery heat information, and collect the batteries in each staying point area, including the following steps:

[0047] Obtain the final position of the battery actually passing through the test area, and determine whether it belongs to the battery staying area of the verified depression type;

[0048] If it belongs, output the verified depression type and the depression part, and collect the batteries of the same depression type.

[0049] In a second aspect, a battery heat management system based on big data, the system includes:

[0050] A division module, used to obtain the battery perimeter based on big data, and sequentially divide the test inclined plane into a detection area, a verification area, and a test area in the inclined direction. Among them, there are several air outlet tunnels on the test inclined plane. The lengths of the inspection area and the test area are both the perimeter of a single battery, and the length of the verification area is n times the perimeter of the battery;

[0051] A detection module, used to make the power of the air outlet tunnels in the detection area the same, and test the dynamic balance of the battery in the detection area;

[0052] A determination module, used to determine the concave defect information and the corresponding air outlet tunnel array according to the battery imbalance degree if the battery dynamic balance in the detection area is measured to be unbalanced;

[0053] A power distribution module, used to divide the verification area into n sub-verification areas according to the battery perimeter, and match the power of the air outlet tunnels in the n sub-verification areas according to the air outlet tunnel array determined in the detection area;

[0054] A verification module, used to use the concave defect information as the initial verification object, test the dynamic balance of the battery in the n sub-verification areas, obtain a dynamic balance result set, and analyze the dynamic balance result set to generate the verified battery heat information;

[0055] A test module, used to allocate the power of the air outlet tunnels in the test area according to the verified battery heat information, and pre-test the staying point area of the battery;

[0056] A classification module is used to measure the actual staying area of the battery. If it is the same as the pre-tested staying point area, it outputs the verified battery heat information and collects the batteries in each staying point area.

[0057] In summary, the present application includes at least one of the following beneficial technical effects:

[0058] The battery heat management method based on big data provided by the present application can quickly analyze the battery with battery heat on the surface without manual labor, and can obtain its depression types and depression positions; through multiple verifications of the battery, it has high analysis accuracy and can automatically classify the battery according to the analysis situation after the analysis is completed, greatly improving the efficiency of battery heat management and reducing the labor cost. Description of the Drawings

[0059] Figure 1 It is a block diagram showing the steps of the battery heat management method based on big data mainly embodied in this embodiment.

[0060] Figure 2 It is a block diagram showing the sub-steps of S200 mainly embodied in this embodiment;

[0061] Figure 3 It is a block diagram showing the sub-steps of S300 mainly embodied in this embodiment;

[0062] Figure 4 It is a block diagram showing the sub-steps of S400 mainly embodied in this embodiment;

[0063] Figure 5 It is a block diagram showing the sub-steps of S500 mainly embodied in this embodiment;

[0064] Figure 6 It is a block diagram showing the sub-steps of S600 mainly embodied in this embodiment;

[0065] Figure 7 It is a block diagram showing the sub-steps of S700 mainly embodied in this embodiment;

[0066] Figure 8 It is a block diagram showing the battery heat management system based on big data mainly embodied in this embodiment;

[0067] Figure 9 It is a schematic diagram showing each area of the test inclined plane mainly embodied in this embodiment;

[0068] Figure 10 It is a schematic diagram showing the position of the test inclined plane and the battery mainly embodied in this embodiment.

[0069] Reference numerals: 1, division module; 2, detection module; 3, determination module; 4, power distribution module; 5, verification module; 6, test module; 7, classification module. Detailed implementation manners

[0070] In order to make the technical solutions and advantages of this application clearer, the following further details this application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0071] To better understand the above technical solutions, the following further details this application in combination with the accompanying Figures 1-10 drawings.

[0072] This application provides a battery heat management method based on big data, including the following steps:

[0073] S100. Obtain the battery perimeter based on big data, and sequentially divide the test inclined plane into a detection area, a verification area, and a test area in the inclined direction. Among them, several air outlet holes are provided on the test inclined plane. The lengths of the inspection area and the test area are both the perimeter of a single battery, and the length of the verification area is n battery perimeters;

[0074] S200. Make the power of the air outlet holes in the detection area the same, and test the dynamic balance of the battery in the detection area;

[0075] S300. If it is determined that the battery in the detection area is dynamically unbalanced, determine the concave defect information and the corresponding air outlet hole array according to the degree of battery imbalance;

[0076] S400. Divide the verification area into n sub-verification areas according to the battery perimeter, and match the power of the air outlet holes of the n sub-verification areas according to the air outlet hole array determined in the detection area;

[0077] S500. Take the concave defect information as the initial verification object, test the dynamic balance of the battery in the n sub-verification areas, obtain a dynamic balance result set, and analyze the dynamic balance result set to generate the verified battery heat information;

[0078] S600. Allocate the power of the air outlet holes in the test area according to the verified battery heat information, and pre-test the battery stay point area;

[0079] S700. Measure the actual battery stay area. If it is the same as the pre-tested stay point area, output the verified battery heat information, and collect the batteries in each stay point area.

[0080] In the battery heat management method based on big data described in the embodiments of this application, the battery heat includes defects such as grooves and cracks that are recessed in the normal surface.

[0081] It should be noted that the above steps are only the preferred implementation order. In the specific implementation process, without affecting the overall implementation effect, some steps can be interchanged.

[0082] In S100, the perimeter of the cylindrical battery to be analyzed can be obtained in advance through big data, that is, the perimeter of the circular shape on the side of the cylindrical battery. This perimeter can represent the displacement length of the battery rotating 360 degrees at a point. Furthermore, by dividing the test inclined plane into multiple regions with the perimeter as the segmentation length, it is possible to meet the identification of the battery heat on the battery surface, avoiding lack of data and generating redundant data. Among them, to prevent the movement speed of the battery on the test inclined plane from being too large, the inclination angle of the test inclined plane should not be too large. The width of the test inclined plane should be slightly larger than the width of the battery, and the air outlet mode of the air outlet tunnel is set to trickle air outlet.

[0083] In the battery heat management method based on big data described in the embodiments of the present application, in step S100, obtaining the battery perimeter through the big data-based method can improve the accuracy and reliability of battery heat management. By dividing the test inclined plane into three regions, relevant tests and analyses can be carried out in each region, enabling a more detailed and comprehensive understanding of the battery situation, and improving the depth and breadth of the analysis.

[0084] In step S200, by making the power of the air outlet tunnel the same, the influence of the power factor on the test results can be eliminated, improving the reliability of the test results and providing a basis for subsequent analysis and processing.

[0085] In step S300, if the battery is dynamically unbalanced in the detection area, it is possible to determine whether there is battery heat information, and determine the specific attributes of the defect information through the degree of imbalance. At the same time, corresponding air outlet tunnel arrays are determined for different defect information, preparing for the next verification.

[0086] In step S400, by dividing the verification area into multiple sub-regions, the state of the battery in different regions can be understood more detailedly. According to the air outlet tunnel array determined in the detection area, the power of the air outlet tunnel for each sub-region is matched, facilitating subsequent targeted verification of the battery.

[0087] In step S500, using the concave defect information as the initial verification object can perform more targeted verification. And through n groups of sub-verification regions, the battery can be verified n times, greatly improving the reliability of the test results.

[0088] In step S600, according to the verified concave defect information, the power of the air outlet tunnel is distributed, enabling secondary verification and improving the reliability of the test results.

[0089] In step S700, by measuring the actual residence area of the battery, the state of the battery can be understood more accurately, it can be determined whether the battery residence position is the same as the pre-tested area, and the verified battery heat information is output, realizing the secondary verification of the battery heat of the battery, and at the same time, batteries of different defect types can be classified and collected.

[0090] Specifically, in S200, the air outlet tunnels in the detection area have the same power, and the dynamic balance of the battery in the detection area is tested, including the following steps:

[0091] S210, code the air outlet tunnels in the detection area according to the battery surface, and assign the same air pressure to the air outlet tunnels;

[0092] S220, collect the motion images of the battery in the detection area in real time, and preprocess the motion images through relevant technologies and extract the battery contour;

[0093] S230, select each pixel in two frames of motion images, find the corresponding pixel in the next frame of image, and calculate the displacement of the battery in adjacent two frames of motion images through pixel point translation;

[0094] S240, model according to the displacement variables of consecutive frames to obtain the motion model of the current transformer.

[0095] In S210, since the side circumference of the battery is used as the segmentation length, each area on the battery surface can be matched with the air outlet tunnel, so that each area on the battery is matched with a unique air outlet tunnel within the interval with the circumference as the segmentation length.

[0096] Further, in S230, each pixel in two frames of motion images is selected, the corresponding pixel is found in the next frame of image, and the displacement of the battery in adjacent two frames of motion images is calculated through pixel point translation, including the following steps:

[0097] S231, extract the battery feature points through relevant technologies, set the pixel coordinates of the feature points in the t-th frame of image as (x, y), and characterize the displacement of the feature point (x, y) from the t-th frame of image to the (t + 1)-th frame of image through the optical flow vector Δv = (Δx, Δy);

[0098] S232, through the formula: Solve the optical flow vector Δv, where A is the coefficient matrix in matrix form, (u, v) is the pixel coordinates of the feature point, I x and I y are the gradients at the pixel positions of the feature points respectively.

[0099] Further, in S240, model according to the displacement variables of consecutive frames to obtain the motion model of the current transformer, including the following steps:

[0100] S241. For each moment \(t\), let the position of the battery in the camera coordinate system be \((x\) t , \(y\) t , \(z\) t ). The motion model of the battery for two adjacent frames is obtained: where \(\Delta x\) t , \(\Delta y\) t and \(\Delta z\) t are the displacement vectors of the battery in the camera coordinate system in two adjacent frames of images;

[0101] S242. Generalize the motion model of two adjacent frames to the entire time series to obtain the motion model of the entire time series:

[0102] Among them, the matrix in the \(t\)-th row is the motion model between two adjacent frames, is the sum of the displacement vectors of the battery in the camera coordinate system in the first \(t\) frames of images;

[0103] S243. Initialize the position of the battery in the camera coordinate system as \((x\) 0 , \(y\) 0 , \(z\) 0 ), set the initial time stamp as \(t\) 0 , and for each subsequent moment \(t\) i , let the optical flow vector between the previous moment \(t\) i-1 be \(\Delta v\) i-1,i = \((u\) i-1,i , \(v\) i-1,i );

[0104] S244. Calculate the displacement vector \(\Delta X\) of the battery in the camera coordinate system according to the motion model of the entire time series i = \((\Delta x\) i , \(\Delta y\) i , \(\Delta z\) i ); and take \(\Delta x\) i , \(\Delta y\) i , \(\Delta z\) i as the feature vector, take \(t\) i - \(t\) 0 as the eigenvalue, and form the feature set \(D\) with all \((\Delta x\) i , \(\Delta y\) i , \(\Delta z\) i , \(t\) i - \(t\) 0 );

[0105] S245. Perform regression on the feature set \(D\) through a regression algorithm to obtain the motion model of the battery in the camera coordinate system. Among them, is the predicted position of the battery at time \(t\) i , and \(f(t\) i ) is the motion model of the battery at time \(t\) i .

[0106] Specifically, if the S300 measures the dynamic imbalance of the battery in the detection area, it determines the concave defect information and the corresponding air outlet tunnel array according to the degree of battery imbalance, including the following steps:

[0107] S310, obtain the predicted position of the battery at the same moment and the actual position and calculate the difference e through the formula: t , where ||·|| 2 is the Euclidean distance;

[0108] S320, if the difference e t is greater than the difference threshold e, it is determined that the battery has a dynamic imbalance at time t;

[0109] S330, output the battery part matching the air outlet tunnel array experienced by the battery at time t as the concave part, and judge the difference e according to the concave type difference interval t belonging to the concave type, where the battery heat information includes the concave type and the concave part.

[0110] In S330, multiple groups of determination intervals can be preset in advance, and each group of determination differences corresponds to a concave type respectively. Then, by calculating the difference size and determining the determination interval to which the difference belongs, the corresponding concave type can be obtained. Among them, since the generation of the difference is the result of the gas acting on the defective and non-defective areas, therefore, the horizontal part where the battery defect is located can be located through the corresponding air outlet tunnel row and column.

[0111] Specifically, the S400 divides the verification area into n groups of sub-verification areas according to the battery perimeter, and matches the air outlet tunnel power of the n groups of sub-verification areas according to the air outlet tunnel array determined in the detection area, including the following steps:

[0112] S410, pair the air outlet tunnel corresponding to the concave part with the air outlet tunnels in the n groups of sub-verification areas to map the concave part of the battery onto the air outlet tunnels of each group of sub-verification areas;

[0113] S420, control the air pressure of the corresponding air outlet tunnels in the n groups of sub-verification areas to be the same as the air pressure of the air outlet tunnels in the detection area to verify the concave type of the battery n times.

[0114] Specifically, the S500 uses the concave defect information as the initial verification object, tests the dynamic balance of the battery in the n groups of sub-verification areas, obtains the dynamic balance result set, and analyzes the dynamic balance result set to generate the verified battery heat information, including the following steps:

[0115] S510, use the concave type as the verification object, the air pressure of the air outlet tunnel corresponding to the concave part as the verification parameter, and construct the motion model of the battery in each group of sub-verification areas;

[0116] S520. Calculate the difference between the actual position and the predicted position of the battery at the corresponding air outlet of the wind tunnel through the motion model of each sub-verification area.

[0117] S530. Statistically analyze the differences in the n verification sub-areas, calculate the average difference in the verification area, and use the depression type to which the average difference in the verification area belongs as the verified depression type.

[0118] In S510, the motion model of the battery in the verification area can be constructed by the above-mentioned method for constructing the motion model in the detection area. Similarly, the motion model of the battery in the test area is constructed below.

[0119] Specifically, the S600 allocates the power of the air outlet of the wind tunnel in the test area according to the verified battery heat information, and for the battery pre-test stop point area, it includes the following steps:

[0120] S610. Preset the battery stop area and mark the battery stop area according to the depression type category.

[0121] S620. Construct the motion model of the battery in the test area under the condition of no air pressure at the air outlet of the wind tunnel in the test area, and predict the final position of the battery passing through the test area.

[0122] S630. Allocate the air pressure at the air outlet of the wind tunnel in the test area to change the predicted final position of the battery passing through the test area to the battery stop area that meets the verified depression type.

[0123] In S610, the battery stop area is preset at the exit of the test area for collecting the batteries that have completed the test, and it also includes a battery stop area for collecting normal batteries.

[0124] Specifically, the S700 measures the actual stop area of the battery. If it is the same as the pre-test stop point area, it outputs the verified battery heat information and collects the batteries in each stop point area, including the following steps:

[0125] S710. Obtain the final position of the battery actually passing through the test area and determine whether it belongs to the battery stop area of the verified depression type.

[0126] S720. If it belongs, output the verified depression type and the depression location, and collect the batteries of the same depression type.

[0127] In S720, if the final position of the battery actually passing through the test area does not belong to the battery stop area of the verified depression type, the battery can be re-operated from S100 to S700.

[0128] The present application also provides a battery heat management system based on big data, and the system includes:

[0129] A division module 1, configured to obtain the battery perimeter based on big data, and sequentially divide the test inclined plane into a detection area, a verification area, and a test area in the inclined direction, wherein a plurality of air outlet holes are provided on the test inclined plane, the lengths of the inspection area and the test area are both the perimeter of a single battery, and the length of the verification area is the perimeter of n batteries;

[0130] A detection module 2, configured to make the power of the air outlet holes in the detection area the same, and test the dynamic balance of the battery in the detection area;

[0131] A determination module 3, configured to, if it is determined that the battery is dynamically unbalanced in the detection area, determine the concave defect information and the corresponding air outlet hole column according to the degree of battery imbalance;

[0132] A power distribution module 4, configured to divide the verification area into n sub-verification areas according to the battery perimeter, and match the power of the air outlet holes of the n sub-verification areas according to the air outlet hole column determined in the detection area;

[0133] A verification module 5, configured to use the concave defect information as the initial verification object, test the dynamic balance of the battery in the n sub-verification areas, obtain a dynamic balance result set, and analyze the dynamic balance result set to generate verified battery heat information;

[0134] A test module 6, configured to allocate the power of the air outlet holes in the test area according to the verified battery heat information, and pre-test the battery stay point area;

[0135] A classification module 7, configured to measure the actual stay area of the battery, and if it is the same as the pre-tested stay point area, output the verified battery heat information, and collect the batteries in each stay point area.

[0136] The beneficial effects provided by the present application are:

[0137] The battery heat management method based on big data provided by the present application can quickly analyze the battery with battery heat on the surface without manual intervention, and can obtain the types and locations of its depressions; through multiple verifications of the battery, it has high analysis accuracy and can automatically classify the battery according to the analysis situation after the analysis is completed, greatly improving the efficiency of battery heat management and reducing the labor cost.

[0138] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. An efficient battery thermal management system, including a battery thermal management method based on big data, characterized in that: The following steps are involved: The battery circumference is obtained based on big data, and the test slope is divided into the inspection area, verification area and test area in the tilt direction. The test slope is provided with several wind tunnels. The length of the inspection area and the test area is the circumference of a single battery, and the length of the verification area is n battery circumferences. Make the wind tunnel power in the test area the same and test the dynamic balance of the battery in the test area; If the dynamic imbalance of the battery in the detection area is measured, the concave defect information and the corresponding wind tunnel group are determined according to the degree of battery imbalance; The verification area is divided into n groups of sub-verification areas according to the battery circumference, and the wind tunnels of the n groups of sub-verification areas are matched with power according to the wind tunnel groups determined in the test area; Taking the concave defect information as the initial verification object, the dynamic balance of the battery in n groups of sub-verification areas is tested to obtain a dynamic balance result set, and the dynamic balance result set is analyzed to generate the verified battery thermal information; The wind tunnel power is distributed in the test area based on the verified battery heat information, and the battery pre-test stop point area is set; The actual battery stay area is measured. If it is the same as the pre-test stay area, the verified battery heat information is output and the batteries in each stay area are collected.

2. The battery thermal management method based on big data according to claim 1, characterized in that: The method of making the wind tunnel power in the test area the same and testing the dynamic balance of the battery in the test area includes the following steps: The wind tunnels in the test area are coded according to the battery surface and the same air pressure is assigned to the wind tunnels; Collect the moving images of the battery in the detection area in real time, and pre-process the moving images and extract the battery contour through relevant technologies; Select each pixel in two frames of motion images, find the corresponding pixel in the next frame, and calculate the displacement of the battery in two adjacent frames of motion images by pixel translation; By modeling the displacement variables of continuous frames, the motion model of the current transmitter is obtained.

3. The battery thermal management method based on big data according to claim 2, characterized in that: The method of selecting each pixel in two frames of motion images, finding the corresponding pixel in the next frame of image, and calculating the displacement of the battery in two adjacent frames of motion images by pixel translation includes the following steps: The battery feature points are extracted by using relevant technologies. The pixel coordinates of the feature points in the t-th frame image are assumed to be (x, y), and the displacement of the feature points (x, y) from the t-frame image to the t+1-frame image is represented by the optical flow vector Δv=(Δx, Δy). By formula: Solve the optical flow vector Δv, where A is the coefficient matrix in matrix form, (u, v) is the pixel coordinate of the feature point, and I x and I y are the gradients at the pixel positions of the feature points.

4. The battery thermal management method based on big data according to claim 3, characterized in that: The method of modeling the displacement variables of the continuous frames to obtain the motion model of the current generator comprises the following steps: For each moment t, let the position of the battery in the camera coordinate system be (x t ,y t ,z t ), and obtain the battery motion model of two adjacent frames: where Δx t , Δy t and Δz t is the displacement vector of the battery in the camera coordinate system in two adjacent needle images; The motion model of two adjacent frames is extended to the entire time series to obtain the motion model of the entire time series: Among them, the matrix in the tth row is the motion model between two adjacent frames. is the sum of the displacement vectors of the battery in the camera coordinate system in the previous t frames; Initialize the position of the battery in the camera coordinate system to (x0, y0, z0), set the initial timestamp to t0, and for each subsequent time t i , assuming that the previous time t i-1 The optical flow vector between is Δv i-1,i =(u i-1,i ,v i-1,i ); According to the motion model of the entire time series, the displacement vector ΔX of the battery is calculated in the camera coordinate system i =(Δx i ,Δy i ,Δz i );and Δx i ,Δy i ,Δz i As the feature vector, t i -t0 as the eigenvalue, all (Δx i ,Δy i ,Δz i ,t i -t0) constitutes a feature set D; The feature set D is regressed through the regression algorithm to obtain the motion model of the battery in the camera coordinate system. in, The battery at time t i The predicted position at time f(t i ) is the battery at time t i motion model.

5. The battery thermal management method based on big data according to claim 4, characterized in that: If the dynamic imbalance of the battery in the detection area is measured, the concave defect information and the corresponding wind tunnel group are determined according to the degree of battery imbalance, including the following steps: Get the predicted position of the battery at the same time and actual location And calculate the difference e by the formula: t , where ||·||2 is the Euclidean distance; If the difference e t If it is greater than the difference threshold e, it is judged that the battery has a dynamic imbalance at time t; The battery part matched by the wind tunnel group experienced by the battery at time t is output as the concave part, and the difference value e is determined according to the difference value interval of the concave type. t The battery thermal information includes the type of depression and the location of the depression.

6. The battery thermal management method based on big data according to claim 5, characterized in that: The verification area is divided into n groups of sub-verification areas according to the battery circumference, and the wind tunnels of the n groups of sub-verification areas are matched with power according to the wind tunnel groups determined in the detection area, including the following steps: Pair the air outlet tunnel corresponding to the concave part with the air outlet tunnels in n groups of sub-verification areas, so as to map the concave part of the battery to the air outlet tunnel in each group of sub-verification areas; The corresponding wind tunnel pressures in the n groups of sub-verification areas are controlled to be the same as the wind tunnel pressure in the detection area, so as to verify the types of battery depressions n times.

7. The battery thermal management method based on big data according to claim 6, characterized in that: The method of taking the concave defect information as the initial verification object, testing the dynamic balance of the battery in n groups of sub-verification areas, obtaining a dynamic balance result set, and analyzing the dynamic balance result set to generate verified battery heat information includes the following steps: The type of depression is used as the verification object, the corresponding wind tunnel pressure of the depression is used as the verification parameter, and the movement model of the battery in each group of sub-verification areas is constructed; The difference between the actual position and the predicted position of the battery at the corresponding wind tunnel exit is calculated using the motion model of each sub-verification area. The difference values ​​of the n verification sub-areas are counted, the average value of the difference values ​​of the verification area is calculated, and the depression type to which the average value of the difference values ​​of the verification area belongs is taken as the verified depression type.

8. The battery thermal management method based on big data according to claim 7, characterized in that: The method of distributing the wind tunnel power in the test area according to the verified battery heat information and pre-testing the battery stop point area includes the following steps: Pre-set battery retention areas and mark the battery retention areas according to the types of depressions; Construct a battery motion model in the test area without air pressure at the exit wind tunnel of the test area, and predict the final position of the battery after passing through the test area; The air pressure at the wind tunnel outlet in the test area is distributed so that the predicted final position of the battery after passing through the test area is changed to a battery retention area that satisfies the verified depression type.

9. The battery thermal management method based on big data according to claim 8, characterized in that: The actual stay area of ​​the measured battery is the same as the pre-test stay area, and the verified battery heat information is output, and the batteries in each stay area are collected, including the following steps: Obtain the final position of the battery after it actually passes through the test area, and determine whether it belongs to the battery retention area of ​​the verified depression type; If it does, the verified dent type and dent location are output, and batteries with the same dent type are collected.

10. An efficient battery thermal management system, further comprising a battery thermal management system based on big data, characterized in that: The system comprises: A division module (1) is used to obtain the battery circumference based on big data, and divide the test slope into a detection area, a verification area and a test area in the direction of inclination, wherein the test slope is provided with a plurality of wind tunnels, the length of the inspection area and the test area are both the circumference of a single battery, and the length of the verification area is n battery circumferences; A detection module (2) is used to make the wind tunnel power in the detection area the same and test the dynamic balance of the battery in the detection area; A determination module (3) is used to determine the concave defect information and the corresponding wind tunnel group according to the degree of battery imbalance if the battery in the detection area is dynamically unbalanced; A power distribution module (4) is used to divide the verification area into n groups of sub-verification areas according to the battery circumference, and match the power of the wind tunnels of the n groups of sub-verification areas according to the wind tunnel groups determined in the detection area; A verification module (5) is used to use the concave defect information as an initial verification object, test the dynamic balance of the battery in n groups of sub-verification areas, obtain a dynamic balance result set, and analyze the dynamic balance result set to generate verified battery thermal information; A test module (6) is used to distribute the wind tunnel power in the test area according to the verified battery heat information, and to pre-test the battery stop point area; The classification module (7) is used to measure the actual battery stay area, and if it is the same as the pre-test stay point area, output the verified battery heat information and collect the batteries in each stay point area.

Images