Multi-module battery heat dissipation cooperative control method based on swarm intelligence algorithm

Abstract

The invention discloses a multi-module battery heat dissipation cooperative control method based on a swarm intelligence algorithm. The method comprises the following steps: S1, generating a temperature data set; s2, preprocessing the temperature data set of each battery module; s3, forming a brightness value set; s4, based on a brightness value similarity principle in a firefly algorithm, automatically identifying battery modules with similar brightness values in the brightness value set; s5, generating an initial heat dissipation path set; s6, forming a new dynamic heat dissipation path; s7, when the cooling unit of a certain battery module breaks down or fails, the control system allocates part of cooling resources from the heat dissipation resources of other normal battery modules by starting a redundant heat dissipation path; and S8, realizing the balance between the cooling power and the temperature control efficiency. The application of the firefly algorithm in the battery field not only improves the heat dissipation efficiency, but also significantly improves the adaptability of the system to temperature fluctuation.

Description

Technical Field

[0001] The present invention relates to the technical field of batteries, and particularly to a multi-module battery heat dissipation cooperative control method based on a swarm intelligence algorithm. Background Art

[0002] With the rapid development of new energy vehicles and the wide application of high-efficiency energy storage technologies, multi-module battery systems have gradually become the mainstream of battery pack design. However, the temperature management of battery modules remains a complex challenge, directly affecting the performance, lifespan, and safety of the battery system. During the operation of the battery system, due to differences in the working states, heat dissipation efficiencies of different modules, and changes in environmental conditions, temperature non-uniformity between modules often occurs. When the temperature of some modules is too high, it will increase the risk of thermal runaway, leading to a reduction in the stability and safety of the entire battery system, and even overheating and fire accidents.

[0003] Currently, the temperature management technologies for multi-module battery systems mainly rely on traditional fixed cooling paths and centralized control methods, and uniformly allocate cooling resources through the feedback of simple temperature sensors. Although the traditional methods can basically regulate the battery temperature, in complex systems with a large number of modules and diverse heat dissipation requirements, there are obvious defects. First of all, traditional methods often cannot dynamically adjust according to the real-time temperature differences of each module, resulting in high-temperature modules not being cooled in time, and the temperature is prone to accumulation, triggering potential safety hazards. Secondly, traditional methods lack distributed cooperative control and can only allocate cooling resources according to fixed cooling paths, making it difficult to adapt to sudden temperature changes. Once the cooling unit of a certain module fails, the system lacks an emergency mechanism, which will cause the temperature of that module to continue to rise. Moreover, traditional cooling control technologies consume high energy and cannot be reasonably allocated according to the actual load status and temperature requirements, resulting in energy waste.

[0004] In addition, existing temperature management technologies have not fully utilized the advantages of swarm intelligence algorithms and have not effectively realized the intelligent cooperative management of cooling resources in multi-module systems. Traditional methods usually only rely on simple control strategies, lacking flexibility and adaptability, and cannot flexibly adjust the cooling path and resource allocation when the temperatures of different modules fluctuate. Therefore, the existing technologies are unable to cope with the problems of temperature imbalance, sudden temperature changes, and cooling unit failures in multi-module battery systems, and are difficult to meet the requirements of new energy battery systems for efficient heat dissipation and energy consumption optimization. Summary of the Invention

[0005] An object of the present invention is to propose a multi-module battery heat dissipation cooperative control method based on a swarm intelligence algorithm. The application of the firefly algorithm of the present invention in the battery field not only improves the heat dissipation efficiency but also significantly enhances the adaptability of the system to temperature fluctuations.

[0006] A multi-module battery heat dissipation collaborative control method based on a swarm intelligence algorithm according to an embodiment of the present invention includes the following steps:

[0007] S1. Collect real-time temperature data of each battery module in the multi-module battery system through multiple temperature sensors in the thermal management system to generate a temperature data set;

[0008] S2. Preprocess the temperature data set of each battery module;

[0009] S3. Map the temperature level of each battery module in the preprocessed temperature data set to a corresponding brightness value to form a brightness value set;

[0010] S4. Based on the principle of similar brightness values in the firefly algorithm, automatically identify battery modules with similar brightness values in the brightness value set, and divide the battery modules with similar brightness values into the same heat dissipation group;

[0011] S5. The control system preliminarily plans the heat dissipation paths of the battery modules in each heat dissipation group according to the composition of the heat dissipation group, concentrates the cooling resources and allocates them to the battery modules with higher brightness values to generate an initial heat dissipation path set;

[0012] S6. Import the initial heat dissipation path set into the multi-module battery system. During the operation, the control system monitors the change of the temperature data of each battery module in real time, and dynamically adjusts the composition of the heat dissipation group according to the latest temperature data and the change of the brightness value; when it is detected that the temperature of a certain battery module rises abnormally, the system immediately triggers a heat dissipation network reconstruction mechanism, reallocates the cooling resources, and quickly adjusts the cooling resources of the adjacent battery modules to the abnormally heated battery module to form a new dynamic heat dissipation path;

[0013] S7. When a cooling unit of a certain battery module fails or malfunctions, the control system allocates part of the cooling resources from the heat dissipation resources of other normal battery modules by enabling redundant heat dissipation paths and updates the heat dissipation path set;

[0014] S8. The control system automatically adjusts the cooling power and cooling paths of each heat dissipation group according to the ambient temperature, load change and real-time temperature data of the battery module; under high-load working conditions, the system preferentially increases the cooling resource allocation of the high-temperature battery module, and reduces the cooling power under low-load conditions to achieve the balance between the cooling power and the temperature control efficiency.

[0015] Optionally, the S1 specifically includes:

[0016] S11. In the multi-module battery system, set a temperature sensor layout matrix;

[0017] S12: At a preset time interval Δt, the battery module temperature data T collected by each temperature sensor is i (t) Temperature data set of a multi-module battery system:

[0018] T={T 1 (t),T 2 (t),,T n (t)};

[0019] Among them, T i (t) represents the temperature of the i-th battery module at time t;

[0020] S13, preliminarily store and mark the temperature data set T, and form a time series temperature data set T according to the serial number sequence of each battery module seq :

[0021] T seq ={T i (t 1 ),T i (t 2 ),,T i (t m )};

[0022] Among them, t 1 ,t 2 ,...,t m Indicates the continuous time nodes of temperature collection.

[0023] Optionally, the S2 specifically includes:

[0024] S21, for the temperature data set T seq Perform noise filtering and set the noise filtering threshold ∈ for the temperature data T of each battery module. i (t), when |T i (t)-T i,avg |>∈, the corresponding temperature data is marked as noise data and removed, where T i,avg represents the average temperature value of the i-th battery module;

[0025] S22, the temperature data set T after noise filtering clean Perform data screening and set the temperature range to [T min ,T max ], the filter condition is T min ≤T cleani ≤T max , T min and T max are the minimum and maximum safe temperature values ​​of the battery module, respectively, to generate the filtered temperature data set T sel ;

[0026] S23. Detect abnormal temperature points for each data point in the filtered temperature data set T sel Set an abnormal detection threshold δ. If the temperature data point T of a certain battery module seli satisfies |T seli - T i,avg | > δ, then mark it as an abnormal temperature point and generate an abnormal temperature point set T abn ;

[0027] S24. Finally, obtain the processed normal temperature data set:

[0028] T norm = T sel - T abn .

[0029] Optionally, the specific steps of S3 include:

[0030] S31. Map the temperature value T of each battery module in the preprocessed normal temperature data set T norm to the corresponding brightness value L normi , where the brightness value L i represents the temperature level of the battery module and has a positive correlation with the temperature value. The higher the temperature of the battery module, the higher its brightness value: i where α

[0031]

[0032] is a proportionality coefficient used to adjust the mapping relationship between temperature and brightness, and β 1 is the brightness reference value; 1

[0033] S32. Combine the brightness values L of all battery modules i to form a brightness value set L = {L 1 , L 2 ,..., L n}, and sort them in descending order of brightness value.

[0034] Optionally, the specific steps of S4 include:

[0035] S41. Set the spatial coordinates X i = (x i , y i , z i ) in the battery system for each battery module i, forming a position set X;

[0036] S42. Calculate the comprehensive similarity S ij between each pair of battery modules i and j based on the brightness value set L and the position set X: ​

[0037]

[0038] Among them, S ij represents the comprehensive similarity between battery module i and battery module j, which is used to measure the similarity degree of the two battery modules in terms of brightness value and spatial position. The smaller the comprehensive similarity, the closer the temperature levels and spatial positions of the two battery modules. L max and L min are respectively the maximum and minimum values in the brightness value set L, which are used to normalize the brightness difference. λ 1 is the spatial attenuation coefficient, which is used to adjust the influence degree of the spatial distance on the comprehensive similarity S ij , reflecting the restriction of the spatial distance on heat dissipation cooperation. r ij is the Euclidean distance between battery module i and battery module j, and the calculation formula is:

[0039]

[0040] S43. Set the comprehensive similarity threshold S th . When S ij ≤S th , it is considered that battery module i and battery module j have a high similarity, and they are classified into the same heat dissipation group G k . Using the clustering method based on the firefly algorithm, all battery modules are classified into m heat dissipation groups to achieve cooperative heat dissipation control of battery modules with similar brightness values and close spatial positions within the same heat dissipation group.

[0041] Optionally, the S43 specifically includes:

[0042] S431. Set the comprehensive similarity threshold S th as the judgment criterion for heat dissipation group division. When the comprehensive similarity S ij of any two battery modules i and j ≤S th , it is considered that the temperature levels and spatial positions of the two battery modules are similar and can be classified into the same heat dissipation group;

[0043] S432. Initialize the brightness value L i and spatial position X i of each battery module based on the clustering mechanism of the firefly algorithm as the initial state of the firefly, and construct a firefly population:

[0044] F = {f 1 , f 2 ,..., f n};

[0045] Among them, f i = (L i , X i) represents the brightness value and spatial position of the i-th battery module;

[0046] S433. During the clustering process, set the attractiveness function A ij representing the attraction force of firefly i to firefly j:

[0047]

[0048] where A ij is the attractiveness of firefly i to firefly j, representing the aggregation trend between battery modules, and γ 1 is the attractiveness reference value, used to control the aggregation intensity between fireflies, and η 1 is the clustering sensitivity parameter, used to adjust the influence degree of the comprehensive similarity S ij on the attractiveness;

[0049] S434. When firefly j is attracted by firefly i, that is, A ij > S th at this time, attract battery module j to the adjacent position of battery module i and divide it into the same heat dissipation group G k The updated set of heat dissipation groups is:

[0050]

[0051] S435. Loop through the attractiveness calculation and firefly clustering process until each battery module belongs to one of the m heat dissipation groups, realizing the collaborative heat dissipation control of battery modules with similar brightness values and close spatial positions.

[0052] Optionally, the S5 specifically includes:

[0053] S51. Set the cooling priority weight W according to the set of brightness values of the battery modules in each heat dissipation group : i :

[0054]

[0055] where W i represents the cooling priority weight of the i-th battery module in the heat dissipation group indicating the proportion of the cooling resource allocation of the i-th battery module in the entire group;

[0056] S52. Allocate the total amount of cooling resources R k to each battery module in the heat dissipation group According to the cooling priority weight W i the amount of cooling resources R i allocated to each battery module i is:

[0057] R i = R k ·W i ;

[0058] Among them, R i represents the amount of cooling resources allocated to the i-th battery module, so that the battery module with a higher brightness value preferentially obtains more cooling resources;

[0059] S53. Set the set of heat dissipation path parameters used to describe the flow direction and path strength of the cooling resources. Each heat dissipation path parameter P i is:

[0060]

[0061] Among them, P i represents the heat dissipation path strength allocated to the i-th battery module, and β 2 is the proportionality coefficient for adjusting the heat dissipation path strength, and γ 2 is the path reference value;

[0062] S54. Output the set of heat dissipation path parameters to the control system to generate the initial heat dissipation path set P.

[0063] Optionally, the S6 specifically includes:

[0064] S61. Import the initial heat dissipation path set P into the control system of the multi-module battery system, set the flow direction of the cooling resources and start the initial cooling allocation;

[0065] S62. The control system monitors the temperature data set T(t) of each battery module in real time according to the preset time interval Δt, and updates the brightness value set L(t) of each battery module according to the current temperature data and the change of the brightness value, reflecting the real-time temperature level of the battery module;

[0066] S63. Dynamically adjust the composition of each heat dissipation group based on the latest brightness value set L(t), and use the comprehensive similarity threshold S th to re-determine the brightness value and the spatial position, and re-allocate the battery modules with similar brightness values and spatial positions to appropriate groups to form a new set of heat dissipation groups

[0067] S64. Set the temperature abnormal rise threshold ΔT abn to detect the battery module T i with abnormal temperature rise. When the temperature change of a certain battery module i satisfies the condition:

[0068] ΔT i = T i(t) - T i,avg > ΔT abn ;

[0069] wherein, ΔT i represents the change in the current temperature of battery module i relative to the average temperature T i,avg If ΔT i > ΔT abn , the system determines that the temperature of this battery module has abnormally increased;

[0070] S65. Trigger the heat dissipation network reconstruction mechanism, and based on the spatial distance and brightness value between battery module i and adjacent battery modules, select a set of adjacent cooling resource source battery modules through comprehensive similarity calculation and reallocate cooling resources to construct a new dynamic heat dissipation path P new where the strength of the new heat dissipation path P i,new :

[0071]

[0072] where, W i represents the cooling priority weight of the abnormally heated battery module, and α 2 and β 3 are respectively the cooling resource allocation coefficients;

[0073] S66. The control system imports the reconstructed dynamic heat dissipation path P new into the multi-module battery system, adjusts the cooling resource allocation in real time, preferentially cools the battery module with abnormal temperature, and restores the temperature balance state of the system based on the heat dissipation path reconstruction.

[0074] Optionally, the specific content of S7 includes:

[0075] S71. Set the fault detection threshold ΔT fail and the cooling unit status monitoring set C = {C 1 , C 2 , …, C n}, where C i represents the cooling unit status of the i-th battery module. When the cooling unit of a certain battery module i fails or malfunctions, and the temperature change satisfies the following conditions:

[0076] ΔT i = T i (t) - T i,avg > ΔT fail ;

[0077] then the system determines that the cooling unit of this battery module has failed and triggers the redundant heat dissipation path enabling mechanism;

[0078] S72. In the set of normal battery modules near the failed battery module i, select the redundant cooling resource source battery module R based on the similarity of spatial position and brightness value. i ={j|C j = normal, j≠i}, and calculate the resource allocation priority Q of the failed battery module i and each redundant battery module j. ij ;

[0079] S73. According to the resource allocation priority Q ij re-plan the redundant heat dissipation path P for the failed battery module i. redundant , and calculate the cooling resource allocation amount R of the redundant heat dissipation path. i,redundant :

[0080]

[0081] where R i,redundant represents the total amount of redundant cooling resources allocated to the failed battery module i, and R j is the available cooling resource amount of the redundant battery module j;

[0082] S74. Merge the redundant heat dissipation path P redundant into the existing heat dissipation path set P to form the updated heat dissipation path set P updated =P∪P redundant , and adjust the flow direction of the cooling resources in real time to keep the temperature of the failed battery module within the safe range.

[0083] The beneficial effects of the present invention are:

[0084] (1) Based on the principle of similar brightness values of the firefly algorithm, the present invention automatically divides each battery module in the multi-module battery system into heat dissipation groups, realizing collaborative heat dissipation control between modules. Through the calculation of the similarity of brightness values and spatial positions, battery modules with similar temperature levels can be automatically identified and dynamically divided into the same heat dissipation group to construct a heat dissipation collaborative network. Compared with the traditional heat dissipation method based on a fixed cooling path, the dynamic group division using the firefly algorithm can flexibly adjust the composition of the heat dissipation group when the temperature distribution changes, realizing the precise allocation of cooling resources, thereby effectively reducing the system temperature difference and preventing local overheating problems. The application of the firefly algorithm in the battery field not only improves the heat dissipation efficiency but also significantly enhances the system's adaptability to temperature fluctuations.

[0085] (2) In the case of a failure or malfunction of the cooling unit in the present invention, an efficient reallocation of cooling resources is achieved through a redundant heat dissipation path enabling mechanism. When it is detected that a cooling unit of a certain battery module fails, the system can automatically select a neighboring module as a redundant cooling source based on comprehensive similarity calculation, and quickly guide the redundant cooling resources to the failed module by real-time adjustment of the heat dissipation path parameters, ensuring that its temperature is maintained within a safe range. This not only enhances the fault tolerance of the system but also effectively reduces the risk of thermal runaway caused by single-point failure. Compared with the traditional centralized control method, it has higher flexibility and reliability.

[0086] (3) The present invention introduces a cooling priority weight mechanism, which adjusts the allocation priority of cooling resources in real time according to the real-time temperature data of each battery module and combines a swarm intelligence algorithm. In the case of high-temperature modules, the system can calculate the cooling priority based on the brightness value distribution, enabling the high-temperature modules to obtain cooling resources first and effectively alleviating the trend of their temperature rise. The system dynamically balances the cooling power and energy consumption according to the ambient temperature and load changes, avoiding waste of cooling resources. Compared with the traditional average distribution mode of cooling resources, the cooling priority mechanism of the present invention significantly improves the heat dissipation efficiency, saves energy consumption, and extends the overall service life of the battery system at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0088] Figure 1 is a flowchart of a multi-module battery heat dissipation cooperative control method based on a swarm intelligence algorithm proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0089] Now, the present invention will be further described in detail with reference to the drawings. These drawings are all simplified schematic diagrams, only showing the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention.

[0090] Refer to Figure 1 , a multi-module battery heat dissipation cooperative control method based on a swarm intelligence algorithm, includes the following steps:

[0091] S1. Collect the real-time temperature data of each battery module in the multi-module battery system through multiple temperature sensors in the thermal management system to generate a temperature data set;

[0092] S2. Preprocess the temperature data sets of each battery module;

[0093] S3. Map the temperature level of each battery module in the preprocessed temperature data set to a corresponding brightness value to form a brightness value set;

[0094] S4. Based on the principle of similar brightness values in the firefly algorithm, automatically identify battery modules with similar brightness values in the brightness value set, and divide the battery modules with similar brightness values into the same heat dissipation group;

[0095] S5. According to the composition of the heat dissipation group, the control system initially plans the heat dissipation paths of the battery modules within each heat dissipation group, concentrates the cooling resources and allocates them to the battery modules with higher brightness values, and generates an initial heat dissipation path set;

[0096] S6. Import the initial heat dissipation path set into the multi-module battery system. During the operation, the control system monitors the temperature data changes of each battery module in real time, and dynamically adjusts the composition of the heat dissipation group according to the latest temperature data and brightness value changes; when it detects that the temperature of a certain battery module rises abnormally, the system immediately triggers the heat dissipation network reconstruction mechanism, reallocates the cooling resources, and quickly adjusts the cooling resources of adjacent battery modules to the abnormally heated battery module to form a new dynamic heat dissipation path;

[0097] S7. When a cooling unit of a certain battery module fails or malfunctions, the control system allocates part of the cooling resources from the heat dissipation resources of other normal battery modules by enabling redundant heat dissipation paths, and updates the heat dissipation path set;

[0098] S8. The control system automatically adjusts the cooling power and cooling paths of each heat dissipation group according to the ambient temperature, load changes, and real-time temperature data of the battery modules; under high-load working conditions, the system preferentially increases the cooling resource allocation for high-temperature battery modules, and reduces the cooling power under low-load conditions to achieve the balance between the cooling power and the temperature control efficiency.

[0099] In this embodiment, S1 specifically includes:

[0100] S11. In the multi-module battery system, set the temperature sensor layout matrix;

[0101] S12. At a preset time interval Δt, the battery module temperature data T i (t) collected by each temperature sensor constitutes the temperature data set of the multi-module battery system:

[0102] T = {T 1 (t), T 2 (t),, T n (t)};

[0103] Among them, T i (t) represents the temperature of the i-th battery module at time t;

[0104] S13, preliminarily store and mark the temperature data set T, and form a time series temperature data set T according to the serial number sequence of each battery module seq :

[0105] T seq ={T i (t 1 ),T i (t 2 ),,T i (t m )};

[0106] Among them, t 1 ,t 2 ,...,t m Indicates the continuous time nodes of temperature collection.

[0107] In this implementation, S2 specifically includes:

[0108] S21, for the temperature data set T seq Perform noise filtering and set the noise filtering threshold ∈ for the temperature data T of each battery module. i (t), when |T i (t)-T i,avg |>∈, the corresponding temperature data is marked as noise data and removed, where T i,avg represents the average temperature value of the i-th battery module;

[0109] S22, the temperature data set T after noise filtering clean Perform data screening and set the temperature range to [T min ,T max ], the filter criteria are T min and T max are the minimum and maximum safe temperature values ​​of the battery module, respectively, to generate the filtered temperature data set T sel ;

[0110] S23, the filtered temperature data set T sel Each data point in the abnormal temperature point detection is performed, and the abnormal detection threshold δ is set. If the temperature data point of a battery module satisfy Then mark it as an abnormal temperature point and generate an abnormal temperature point set T abn ;

[0111] S24. Finally, the processed normal temperature data set is obtained:

[0112] T norm =T sel -T abn .

[0113] In this embodiment, S3 specifically includes:

[0114] S31. Map the temperature value of each battery module in the preprocessed normal temperature data set T norm to the corresponding brightness value L such that the brightness value L i represents the temperature level of the battery module and has a positive correlation with the temperature value. The higher the temperature of the battery module, the higher its brightness value: i where α

[0115]

[0116] is a proportionality coefficient for adjusting the mapping relationship between temperature and brightness, and β 1 is the brightness reference value; 1

[0117] S32. Combine the brightness values L of all battery modules i to form a brightness value set L = {L 1 , L 2 ,..., L n} and sort them in descending order of brightness value.

[0118] In this embodiment, S4 specifically includes:

[0119] S41. Set the spatial coordinates X i = (x i , y i , z i ) in the battery system for each battery module i to form a position set X;

[0120] S42. Calculate the comprehensive similarity S ij between each pair of battery modules i and j based on the brightness value set L and the position set X

[0121]

[0122] where S ij represents the comprehensive similarity between battery module i and battery module j, which is used to measure the similarity degree of the two battery modules in terms of brightness value and spatial position. The smaller the comprehensive similarity, the closer the temperature levels and spatial positions of the two battery modules. L max and L min are respectively the maximum and minimum values in the brightness value set L, which are used to normalize the brightness gap. λ 1 is a spatial attenuation coefficient for adjusting the influence degree of the spatial distance on the comprehensive similarity S ij , reflecting the restriction of the spatial distance on heat dissipation cooperation. r ij ​is the Euclidean distance between battery module i and battery module j, and the calculation formula is:

[0123]

[0124] S43. Set the comprehensive similarity threshold S th , when S ij ≤S th , it is considered that battery module i and battery module j have a high similarity, and they are divided into the same heat dissipation group G k . Using the clustering method based on the firefly algorithm, all battery modules are divided into m heat dissipation groups to achieve collaborative heat dissipation control of battery modules with similar brightness values and close spatial positions within the same heat dissipation group.

[0125] In this embodiment, S43 specifically includes:

[0126] S431. Set the comprehensive similarity threshold S th as the judgment criterion for heat dissipation group division. When the comprehensive similarity S ij ≤S th of any two battery modules i and j, it is considered that the temperature levels and spatial positions of the two battery modules are similar and can be divided into the same heat dissipation group;

[0127] S432. Initialize the brightness value L i and spatial position X i of each battery module as the initial state of the firefly, and construct a firefly population:

[0128] F = {f 1 , f 2 ,..., f n};

[0129] where f i = (L i , X i ) represents the brightness value and spatial position of the i-th battery module;

[0130] S433. During the clustering process, set the attraction function A ij to represent the attraction magnitude of firefly i to firefly j:

[0131]

[0132] where A ij is the attraction of firefly i to firefly j, representing the aggregation trend between battery modules, γ 1 is the attraction reference value for controlling the aggregation intensity between fireflies, and η 1 is the clustering sensitivity parameter for adjusting the comprehensive similarity S ijDegree of influence on attractiveness;

[0133] S434. When firefly j is attracted by firefly i, i.e., A ij >S th At this time, attract battery module j to the adjacent position of battery module i and divide it into the same heat dissipation group G k Among them, the updated set of heat dissipation groups is:

[0134]

[0135] S435. Loop through the attractiveness calculation and firefly clustering process until each battery module belongs to one of the m heat dissipation groups, realizing the collaborative heat dissipation control of battery modules with similar brightness values and close spatial positions.

[0136] In this embodiment, S5 specifically includes:

[0137] S51. According to the set of brightness values of the battery modules in each heat dissipation group Set the cooling priority weight W i :

[0138]

[0139] Among them, W i Represents the cooling priority weight of the i-th battery module in the heat dissipation group And represents the proportion of the cooling resource allocation of the i-th battery module in the whole group;

[0140] S52. Allocate the total amount of cooling resources R k to each battery module in the heat dissipation group According to the cooling priority weight W i The amount of cooling resources R i allocated to each battery module i is:

[0141] R i = R k ·W i ;

[0142] Among them, R i Represents the amount of cooling resources allocated to the i-th battery module, so that the battery module with a higher brightness value can obtain more cooling resources first;

[0143] S53. Set the set of heat dissipation path parameters To describe the flow direction and path strength of the cooling resources, each heat dissipation path parameter P i Is:

[0144]

[0145] Among them, P i represents the heat dissipation path intensity assigned to the i-th battery module, and β 2 is the proportionality coefficient for adjusting the heat dissipation path intensity, and γ 2 is the path reference value;

[0146] S54. Output the heat dissipation path parameter set to the control system to generate the initial heat dissipation path set P.

[0147] In this embodiment, S6 specifically includes:

[0148] S61. Import the initial heat dissipation path set P into the control system of the multi-module battery system, set the flow direction of the cooling resource and start the initial cooling distribution;

[0149] S62. The control system monitors the temperature data set T(t) of each battery module in real time according to the preset time interval Δt, and updates the brightness value set L(t) of each battery module according to the current temperature data and the change of the brightness value, so as to reflect the real-time temperature level of the battery module;

[0150] S63. Dynamically adjust the composition of each heat dissipation group based on the latest brightness value set L(t), and use the comprehensive similarity threshold S th to re-determine the brightness value and the spatial position, and re-allocate the battery modules with similar brightness values and spatial positions to appropriate groups to form a new heat dissipation group set

[0151] S64. Set the temperature abnormal rise threshold ΔT abn to detect the battery module T i with abnormal temperature rise. When the temperature change of a certain battery module i satisfies the condition:

[0152] ΔT i = T i (t) - T i,avg > ΔT abn ;

[0153] Among them, ΔT i represents the change amount of the current temperature of the battery module i relative to the average temperature T i,avg . If ΔT i > ΔT abn , the system determines that the battery module has an abnormal temperature rise;

[0154] S65. Trigger the heat dissipation network reconstruction mechanism, and select the set of adjacent cooling resource source battery modules through comprehensive similarity calculation according to the spatial distance and brightness value between the battery module i and the adjacent battery modules, and re-allocate the cooling resources to construct a new dynamic heat dissipation path Pnew , where the intensity P of the new heat dissipation path i,new :

[0155]

[0156] where, W i represents the cooling priority weight of the abnormally heated battery module, and α 2 and β 3 are respectively the cooling resource allocation coefficients;

[0157] S66. The control system imports the reconstructed dynamic heat dissipation path P new into the multi-module battery system, adjusts the cooling resource allocation in real time, preferentially cools the battery module with abnormal temperature, and restores the temperature balance state of the system on the basis of heat dissipation path reconstruction.

[0158] In this embodiment, S7 specifically includes:

[0159] S71. Set the fault detection threshold ΔT fail and the cooling unit status monitoring set C = {C 1 , C 2 , …, C n}, where C i represents the cooling unit status of the i-th battery module. When the cooling unit of a certain battery module i fails or malfunctions, and the temperature change satisfies the following conditions:

[0160] ΔT i = T i (t) - T i,avg > ΔT fail ;

[0161] Then the system determines that the cooling unit of this battery module fails, and triggers the redundant heat dissipation path enabling mechanism;

[0162] S72. In the set of normal battery modules near the failed battery module i, select the redundant cooling resource source battery module R i = {j∣C j = normal, j ≠ i} based on the similarity of spatial position and brightness value, and calculate the resource allocation priority Q ij of the failed battery module i and each redundant battery module j;

[0163] S73. Re-plan the redundant heat dissipation path P ij for the failed battery module i according to the resource allocation priority Q redundant , and calculate the cooling resource allocation amount R i,redundant of the redundant heat dissipation path:

[0164]

[0165] Among them, R i,redundant represents the total amount of redundant cooling resources allocated to the failed battery module i, and R j is the available cooling resource amount of the redundant battery module j;

[0166] S74. Merge the redundant heat dissipation path P redundant into the existing set of heat dissipation paths P to form an updated set of heat dissipation paths P updated = P ∪ P redundant , and adjust the flow direction of the cooling resources in real time to keep the temperature of the failed battery module within the safe range.

[0167] Example 1:

[0168] In mid-August 2024, a new energy logistics company in City A installed the multi-module battery heat dissipation collaborative control system of the present invention to monitor and manage 100 electric logistics vehicles in the company's fleet. The vehicles are used frequently at high temperatures every day, and the temperatures of the 100 battery modules in the battery system change frequently and unevenly. When traveling long distances in the high-temperature areas from City A to City B, the battery system will have temperature fluctuations due to high load, and the temperatures of some battery modules will approach the dangerous threshold of 70°C, threatening the battery life and driving safety.

[0169] On the afternoon of August 15, a certain logistics vehicle (numbered A01) loaded goods in City A and headed for City B. The outside air temperature at departure was 35°C. The system immediately started the real-time monitoring mode after departure and collected the temperature data of each battery module of vehicle A01. The fleet control center recorded that about 2 hours after the vehicle started, the temperatures of some modules rose rapidly, especially the modules near the middle of the vehicle bottom reached 60°C, much higher than the average temperature of 47°C in other areas.

[0170] The system automatically starts the firefly algorithm to convert the real-time temperatures of all battery modules into brightness values, identifies the modules with the highest temperatures as the areas with the highest brightness, and dynamically divides the modules in this area into a heat dissipation group according to the principle of proximity of brightness values and module spatial positions. The cooling resource priority of this heat dissipation group is automatically increased, and the control system increases the cooling power distribution to the group, stabilizing the temperature in this area within 55°C and preventing the temperature from rising further.

[0171] At 3:00 am on August 18, vehicle A01 was driving on the outskirts of City B, on a highway section with low traffic flow. At this time, the control system detected that the temperature feedback data of the cooling unit C06 stopped updating, presumably due to a failure of the cooling unit. This cooling unit was originally responsible for cooling 6 modules, and at this time, it was detected that the temperatures of modules numbered 67 and 68 rose to 63°C, exceeding the safety threshold of 55°C. The control system immediately started the fault detection program. After confirming the failure of unit C06, the redundant heat dissipation path mechanism was activated.

[0172] The system marks the normal units C04 and C08 adjacent to the cooling unit C06 as redundant cooling sources, and determines the priority cooling requirements of modules 67 and 68 using comprehensive similarity calculation. The system adjusts the cooling path, allocates cooling resources from units C04 and C08 to modules 67 and 68, reduces the temperature to 58°C within a short time and maintains it within the safe range. The whole process takes about 8 minutes, avoiding local overheating caused by the failure of the cooling unit.

[0173] In the afternoon of August 20th, vehicle A01 returned from City B to City A, passing through a high-temperature area in a certain district of City B, with the external temperature reaching as high as 38°C. After driving for about 3 hours, the control system monitored in real time that the temperature of battery module No. 52 increased rapidly, rising from 55°C to 68°C within 10 minutes. The system determined that this module was in a high-risk state of thermal runaway and immediately triggered the heat dissipation network reconstruction mechanism.

[0174] The control system calculated the distance and brightness value differences between module 52 and surrounding modules, and selected the adjacent modules No. 53, 54, and 55 with the highest cooling priority as the cooling resource allocation sources based on the firefly algorithm. Through adjusting the cooling priority weights, the cooling resource supply to module 52 increased rapidly. The system reduced the temperature of module 52 to 61°C within 3 minutes and stabilized it within 60°C, preventing further temperature rise and potential thermal runaway risks.

[0175] To verify the effectiveness of the present invention, the fleet simultaneously adopted the traditional fixed cooling path method on other vehicles and recorded the corresponding temperature, fault handling time, and energy consumption data. The following are the comparison data:

[0176] The comparison data of the temperature control effects of applying the method of the present invention in vehicle A01 and applying the traditional method in other vehicles are shown in Table 1 below:

[0177] Table 1 Comparison data of the temperature control effects of the method of the present invention and the traditional method applied in other vehicles

[0178]

[0179] The data in Table 1 show that under the same driving conditions, the method of the present invention can effectively control the battery temperature of vehicle A01 below 57°C, while the temperature of the vehicle using the traditional method has risen to 68°C in a similar time, and the heat dissipation effect is significantly inferior to that of the method of the present invention.

[0180] When the failure of cooling unit C06 occurred, the fault handling performances of the method of the present invention and the traditional method are shown in Table 2 below:

[0181] Table 2 Fault handling performances of the method of the present invention and the traditional method

[0182]

[0183]

[0184] As can be seen from the data in Table 2, the method of the present invention restores temperature balance within 8 minutes through redundant heat dissipation paths, with the maximum temperature controlled at 58°C, while the traditional method takes 20 minutes and the maximum temperature reaches 72°C, showing a higher risk of thermal runaway.

[0185] The energy consumption data of applying the method of the present invention and the traditional method are as follows:

[0186] Table 3 Energy consumption data of the method of the present invention and the traditional method

[0187] Running time (hours) Energy consumption of vehicle A01 (kWh) Energy consumption of other vehicles (kWh) 2 3.7 4.5 4 7.3 8.6 6 10.8 12.9 8 14.2 17.0

[0188] The data in Table 3 show that the method of the present invention realizes more energy-saving heat dissipation control by dynamically adjusting the heat dissipation groups and the cooling priority weights, reducing the energy consumption by nearly 15% compared with the traditional method.

[0189] As can be seen from the above specific Embodiment 1, the multi-module battery heat dissipation collaborative control method of the present invention effectively solves the problems of uneven battery module temperatures, coping with cooling unit failures, and preferential cooling of high-temperature modules through the dynamic heat dissipation group division of the firefly algorithm, the redundant heat dissipation path mechanism, and the adjustment of cooling priority weights, significantly improving the heat dissipation efficiency and system fault tolerance, and further verifying the effectiveness of the present invention in practical applications, providing higher guarantees for the safe driving of electric logistics vehicles and the battery life.

[0190] Based on the principle of similar brightness values of the firefly algorithm, the present invention automatically divides each battery module in the multi-module battery system into heat dissipation groups, realizing collaborative heat dissipation control between modules. Through the similarity calculation of brightness values and spatial positions, battery modules with similar temperature levels can be automatically identified and dynamically divided into the same heat dissipation group to construct a heat dissipation collaborative network. Compared with the traditional heat dissipation method based on a fixed cooling path, the dynamic group division using the firefly algorithm can flexibly adjust the composition of the heat dissipation group when the temperature distribution changes, realizing the precise allocation of cooling resources, thereby effectively reducing the system temperature difference and preventing local overheating problems. The application of the firefly algorithm in the battery field not only improves the heat dissipation efficiency but also significantly enhances the system's adaptability to temperature fluctuations.

[0191] In the case of a failure or malfunction of the cooling unit, the present invention enables efficient reallocation of cooling resources through a redundant heat dissipation path enabling mechanism. When a failure of the cooling unit of a certain battery module is detected, the system can automatically select an adjacent module as a redundant cooling source based on comprehensive similarity calculation, and quickly guide the redundant cooling resources to the failed module by real-time adjustment of the heat dissipation path parameters, ensuring that its temperature is maintained within a safe range. This not only enhances the fault tolerance of the system, but also effectively reduces the risk of thermal runaway caused by single-point failure. Compared with the traditional centralized control method, it has higher flexibility and reliability.

[0192] The present invention introduces a cooling priority weight mechanism. According to the real-time temperature data of each battery module and combined with a swarm intelligence algorithm, it adjusts the allocation priority of cooling resources in real time. In the case of the appearance of a high-temperature module, the system can calculate the cooling priority based on the brightness value distribution, enabling the high-temperature module to obtain cooling resources first and effectively alleviating the trend of its temperature rise. The system dynamically balances the cooling power and energy consumption according to the ambient temperature and load changes, avoiding waste of cooling resources. Compared with the traditional average distribution mode of cooling resources, the cooling priority mechanism of the present invention significantly improves the heat dissipation efficiency, saves energy consumption, and prolongs the overall service life of the battery system at the same time.

[0193] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered by the protection scope of the present invention.

Claims

1. A multi-module battery heat dissipation collaborative control method based on swarm intelligence algorithm, characterized in that: The steps include: S1. Collecting real-time temperature data of each battery module in a multi-module battery system through multiple temperature sensors in a thermal management system to generate a temperature data set; S2. Preprocessing the temperature data set of each battery module; S3, mapping the temperature level of each battery module in the preprocessed temperature data set to a corresponding brightness value to form a brightness value set; S4. Based on the principle of similar brightness values ​​in the firefly algorithm, the battery modules with similar brightness values ​​are automatically identified in the brightness value set, and the battery modules with similar brightness values ​​are divided into the same heat dissipation group; S5. The control system preliminarily plans the heat dissipation paths of the battery modules in each heat dissipation group according to the composition of the heat dissipation group, allocates cooling resources to the battery modules with higher brightness values, and generates an initial heat dissipation path set; S6. Import the initial heat dissipation path set into the multi-module battery system. During operation, the control system monitors the temperature data changes of each battery module in real time, and dynamically adjusts the composition of the heat dissipation group according to the latest temperature data and brightness value changes. When it is detected that the temperature of a battery module rises abnormally, the system immediately triggers the heat dissipation network reconstruction mechanism, reallocates cooling resources, and quickly adjusts the cooling resources of the adjacent battery modules to the abnormally heated battery module to form a new dynamic heat dissipation path. S7. When a cooling unit of a battery module fails or fails, the control system enables a redundant cooling path, allocates some cooling resources from the cooling resources of other normal battery modules, and updates the cooling path set; S8. The control system automatically adjusts the cooling power and cooling path of each heat dissipation group according to the ambient temperature, load changes and real-time temperature data of the battery module; under high-load working conditions, the system prioritizes increasing the cooling resource allocation of high-temperature battery modules, and reduces the cooling power under low-load conditions, achieving a balance between cooling power and temperature control efficiency.

2. According to the method of claim 1, a multi-module battery heat dissipation collaborative control method based on swarm intelligence algorithm is characterized in that: The S1 specifically includes: S11. In a multi-module battery system, setting a temperature sensor arrangement matrix; S12: At a preset time interval Δt, the battery module temperature data T collected by each temperature sensor is i (t) Temperature data set of a multi-module battery system: T={T1(t),T2(t),,T n (t)}; Among them, T i (t) represents the temperature of the i-th battery module at time t; S13, preliminarily store and mark the temperature data set T, and form a time series temperature data set T according to the serial number sequence of each battery module seq : T seq ={T i (t1),T i (t2),,T i (t m )}; Among them, t1, t2, ..., t m Indicates the continuous time nodes of temperature collection.

3. According to the method of claim 1, a multi-module battery heat dissipation collaborative control method based on swarm intelligence algorithm is characterized in that: The S2 specifically includes: S21, for the temperature data set T seq Perform noise filtering and set the noise filtering threshold ∈ for the temperature data T of each battery module. i (t), when |T i (t)-T i,avg |>∈, the corresponding temperature data is marked as noise data and removed, where T i,avg represents the average temperature value of the i-th battery module; S22, the temperature data set T after noise filtering clean Perform data screening and set the temperature range to [T min ,T max ], the filter criteria are T min and T max are the minimum and maximum safe temperature values ​​of the battery module, respectively, to generate the filtered temperature data set T sel ; S23, the filtered temperature data set T sel Each data point in the abnormal temperature point detection is performed, and the abnormal detection threshold δ is set. If the temperature data point of a battery module satisfy Then mark it as an abnormal temperature point and generate an abnormal temperature point set T abn ; S24. Finally, the processed normal temperature data set is obtained: T norm =T sel -T abn 。 4. According to the method of claim 1, a multi-module battery heat dissipation collaborative control method based on swarm intelligence algorithm is characterized in that: The S3 specifically includes: S31, the pre-processed normal temperature data set T norm The temperature value of each battery module in Mapped to the corresponding brightness value L i , brightness value L i Indicates the temperature level of the battery module and is positively correlated with the temperature value. The higher the temperature of the battery module, the higher the brightness value: Among them, α1 is the proportional coefficient, which is used to adjust the mapping relationship between temperature and brightness, and β1 is the brightness reference value; S32, set the brightness value L of all battery modules i The brightness value set L = {L1, L2,, L n } and sort them by brightness value from high to low.

5. The method for collaboratively controlling heat dissipation of multi-module batteries based on swarm intelligence algorithm according to claim 1 is characterized in that: The S4 specifically includes: S41, set the spatial coordinate X of each battery module i in the battery system i =(x i ,y i ,z i ), forming a position set X; S42, calculating the comprehensive similarity S of each pair of battery modules i and j based on the brightness value set L and the position set X ij : Among them, S ij It represents the comprehensive similarity between battery module i and battery module j, which is used to measure the similarity between the two battery modules in terms of brightness value and spatial position. The smaller the comprehensive similarity, the closer the temperature level and spatial position of the two battery modules are. max and L min are the maximum and minimum values ​​in the brightness value set L, respectively, which are used to normalize the brightness difference. λ1 is the spatial attenuation coefficient, which is used to adjust the spatial distance to the comprehensive similarity S. ij The degree of influence reflects the restriction of spatial distance on heat dissipation coordination. ij is the Euclidean distance between battery module i and battery module j, and the calculation formula is: S43, setting comprehensive similarity threshold S th , when S ij ≤S th When , it is considered that battery module i and battery module j have high similarity and are divided into the same heat dissipation group G k , all battery modules are divided into m heat dissipation groups using a clustering method based on the firefly algorithm, so as to realize the coordinated heat dissipation control of battery modules with similar brightness values ​​and close spatial positions in the same heat dissipation group.

6. The method for collaboratively controlling heat dissipation of multi-module batteries based on swarm intelligence algorithm according to claim 5 is characterized in that: The S43 specifically includes: S431, setting comprehensive similarity threshold S th As the judgment standard for heat dissipation group division, when the comprehensive similarity S of any two battery modules i and j is ij ≤S th When , it is considered that the temperature levels and spatial positions of the two battery modules are similar and can be divided into the same heat dissipation group; S432: Initialize the brightness value L of each battery module based on the clustering mechanism of the firefly algorithm i and spatial position X i As the initial state of fireflies, build a firefly population: F={f1,f2,...,f n }; Among them, f i =(L i ,X i ) represents the brightness value and spatial position of the i-th battery module; S433. In the clustering process, set the attraction function A ij Indicates the attraction of firefly i to firefly j: Among them, A ij is the attraction of firefly i to firefly j, indicating the clustering trend between battery modules, γ1 is the attraction reference value, which is used to control the clustering intensity between fireflies, and η1 is the clustering sensitivity parameter, which is used to adjust the comprehensive similarity S ij The degree of influence on attractiveness; S434, when firefly j is attracted by firefly i, that is, A ij >S th When the battery module j is attracted to the adjacent position of the battery module i and divided into the same heat dissipation group G k In the update, the cooling group set is: S435, cyclically executing the attraction calculation and firefly clustering process until each battery module belongs to one of the m heat dissipation groups, thereby achieving coordinated heat dissipation control of battery modules with similar brightness values ​​and close spatial positions.

7. The method for collaboratively controlling heat dissipation of multi-module batteries based on swarm intelligence algorithm according to claim 1 is characterized in that: The S5 specifically includes: S51, according to each heat dissipation group The battery module brightness value set in Set cooling priority weight W i : Among them, W i Heat Dissipation Team The cooling priority weight of the i-th battery module in the group represents the cooling resource allocation ratio of the i-th battery module in the entire group; S52, the total amount of cooling resources R k Assigned to Cooling Team Each battery module in the i The amount of cooling resources R allocated to each battery module i i : R i =R k ·W i ; Among them, R i Indicates the amount of cooling resources allocated to the i-th battery module, so that the battery modules with higher brightness values ​​are given priority to obtain more cooling resources; S53, setting the heat dissipation path parameter set Used to describe the flow direction and path strength of cooling resources. Each heat dissipation path parameter P i for: Among them, P i represents the heat dissipation path strength assigned to the i-th battery module, β2 is the proportional coefficient for adjusting the heat dissipation path strength, and γ2 is the path reference value; S54, the heat dissipation path parameter set Output to the control system to generate an initial heat dissipation path set P.

8. The method for collaboratively controlling heat dissipation of multi-module batteries based on swarm intelligence algorithm according to claim 1 is characterized in that: The S6 specifically includes: S61, importing the initial heat dissipation path set P into the control system of the multi-module battery system, setting the cooling resource flow direction and starting the initial cooling allocation; S62, the control system monitors the temperature data set T(t) of each battery module in real time according to a preset time interval Δt, and updates the brightness value set L(t) of each battery module according to the current temperature data and brightness value changes to reflect the real-time temperature level of the battery module; S63, dynamically adjust each heat dissipation group based on the latest brightness value set L(t) The composition of th Re-determine the brightness value and spatial position, and reallocate battery modules with similar brightness values ​​and spatial positions to appropriate groups to form a new set of heat dissipation groups. S64, set the temperature abnormal increase threshold ΔT abn To detect abnormally high temperature battery module T i , when the temperature change of a battery module i meets the condition: ΔT i =T i (t)-T i,avg >ΔT abn ; Where, ΔT i Indicates the current temperature of battery module i relative to the average temperature T i,avg The change in ΔT i >ΔT abn , the system determines that the temperature of the battery module has abnormally increased; S65, trigger the heat dissipation network reconstruction mechanism, select the adjacent cooling resource source battery module set through comprehensive similarity calculation according to the spatial distance and brightness value between battery module i and the adjacent battery module, and reallocate the cooling resources to construct a new dynamic heat dissipation path P new , where the new heat dissipation path strength P i,new : Among them, W i represents the cooling priority weight of the abnormally heated battery module, α2 and β3 are cooling resource allocation coefficients respectively; S66, the control system reconstructs the dynamic heat dissipation path P new Import a multi-module battery system, adjust the cooling resource allocation in real time, give priority to cooling the battery modules with abnormal temperatures, and restore the system's temperature balance based on the reconstruction of the heat dissipation path.

9. The method for collaboratively controlling heat dissipation of multi-module batteries based on swarm intelligence algorithm according to claim 1 is characterized in that: The S7 specifically includes: S71, setting fault detection threshold ΔT fail and cooling unit status monitoring set C = {C1, C2, ..., C n }, where C i Indicates the cooling unit status of the i-th battery module. When the cooling unit of a battery module i fails or fails, and the temperature change meets the following conditions: ΔT i =T i (t)-T i,avg >ΔT fail ; The system determines that the battery module cooling unit has failed, triggering the redundant heat dissipation path activation mechanism; S72: Select redundant cooling resource source battery module R from a set of normal battery modules near the failed battery module i based on the similarity of spatial position and brightness value. i ={j|C j = normal, j≠i}, and calculate the resource allocation priority Q between the failed battery module i and each redundant battery module j ij ; S73, according to resource allocation priority Q ij Re-plan redundant heat dissipation path P for failed battery module i redundant , and calculate the cooling resource allocation R of the redundant cooling path i,redundant : Among them, R i,redundant represents the total amount of redundant cooling resources allocated to failed battery module i, R j is the available cooling resource of redundant battery module j; S74, the redundant heat dissipation path P redundant Merge into the existing heat dissipation path set P to form an updated heat dissipation path set P updated =P∪P redundant , and adjust the flow of cooling resources in real time to keep the temperature of the failed battery module within a safe range.

Images