Bms-based coolant flow and temperature collaborative control method and system

Abstract

This invention provides a method and system for coordinated control of coolant flow and temperature based on a Battery Management System (BMS), belonging to the field of power battery thermal management and battery management control technology. This invention decouples thermal management control requirements into temperature regulation requirements and flow regulation requirements, and establishes a state-related coordinated mapping relationship based on the thermal management requirement state to generate target coolant temperature and target coolant flow rate. Then, it performs coordinated control on one or more actuators among the variable-speed pump, proportional valve, three-way valve, heat exchanger, refrigeration unit, heating unit, and branch switching components. Based on the battery temperature feedback, coolant temperature feedback, and coolant flow rate feedback after execution, the coordinated control relationship between coolant temperature and flow rate is updated online. This method can balance average temperature control, local hot spot suppression, temperature difference equalization, low-temperature preheating, and execution stability, improving the adaptive capability and comprehensive control performance of power battery liquid cooling thermal management.

Description

Technical Field

[0001] This invention relates to the field of power battery thermal management and battery management control technology, specifically to a method and system for coordinated control of coolant flow and temperature based on a battery management system (BMS). Background Technology

[0002] Power battery systems generate significant heat accumulation under conditions of high-rate charging and discharging, fast charging, low-temperature start-up, and continuous high-power output. Furthermore, non-uniform heat generation often exists between different cells or modules within the battery pack. If the average temperature of the battery pack is too high, it can easily lead to accelerated capacity decay, increased internal resistance, and reduced lifespan. If the maximum temperature of a single cell is too high, it will increase the risk of localized thermal runaway. Excessive temperature differences will affect cell consistency, power output capability, and charge / discharge balance. Therefore, power battery thermal management must not only focus on overall temperature control but also consider suppressing localized hot spots and regulating temperature uniformity.

[0003] Among existing liquid cooling thermal management control methods, one approach directly derives the coolant flow rate based on the battery pack temperature difference; another approach determines the target coolant temperature based on the battery area temperature; and a third approach uses predictive control to regulate water pumps, valves, or heat exchange components. While these approaches address some issues in flow control, liquid temperature control, or actuator linkage, they generally suffer from problems such as a single control objective, lack of decoupling of thermal management requirements, fixed coordination between target liquid temperature and target flow rate, and insufficient priority switching capability under different operating conditions.

[0004] Especially when alternating operating conditions such as high-temperature heat dissipation, temperature difference equalization, low-temperature preheating, and transient high-power response occur, using the same control logic to simultaneously drive both flow and temperature regulation channels often leads to over-regulation in one channel, insufficient response in the other, frequent switching of actuators, and high overall energy consumption. Meanwhile, battery degradation, changes in environmental boundaries, and performance drift of liquid-cooled actuators also alter the actual impact of flow and temperature regulation on the thermal state, making it difficult to maintain a stable and effective working relationship in the long term. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for coordinated control of coolant flow and temperature based on a BMS, so as to solve at least one of the technical problems existing in the background art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a method for coordinated control of coolant flow and temperature based on a BMS, comprising:

[0008] The thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system are collected. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack, and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge / discharge current, charge / discharge rate, power demand, or state of health. The environmental state parameters include at least one or more of the following: ambient temperature or external heat transfer boundary conditions.

[0009] Based on the thermal state parameters, electrical state parameters and environmental state parameters, the current thermal management requirement state of the power battery system is identified. The thermal management requirement state includes at least one of the following: high temperature heat dissipation state, temperature difference equalization state, low temperature preheating state, transient power response state or composite regulation state.

[0010] Based on the thermal management requirement status, the thermal management control requirements of the power battery system are decoupled into temperature regulation requirements and flow regulation requirements. The temperature regulation requirements are used to characterize the target supply temperature or heat exchange capacity requirement of the coolant, and the flow regulation requirements are used to characterize the circulation intensity or flow path distribution requirement of the coolant.

[0011] Based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement status, combined with preset control rules, thermal management mapping models, and / or optimized allocation models, the target coolant temperature and target coolant flow rate are determined, and the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate are dynamically adjusted under different thermal management requirement statuses.

[0012] Based on the target coolant temperature and the target coolant flow rate, at least one actuator in the liquid cooling system is linked for control to adjust the coolant circulation flow rate, supply temperature, flow path distribution, and heat exchange capacity.

[0013] Based on the battery temperature feedback, coolant temperature feedback, coolant flow feedback, and battery operating condition change information after the execution control, the target coolant temperature and the target coolant flow are rolled over and corrected. The thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow are updated online to form a closed-loop adaptive collaborative thermal management control based on BMS.

[0014] As a further limitation of the first aspect of the present invention, when the thermal management demand state is a high-temperature heat dissipation state, the target coolant temperature is preferentially reduced; when the highest temperature of a single cell and / or the rate of increase of the highest temperature of a single cell exceeds a preset hot spot threshold, the priority of the target coolant flow rate in the coordinated control is increased to suppress the rapid development of local hot spots; when the thermal management demand state is a temperature difference balance state, the target coolant flow rate is preferentially increased and / or the flow path distribution is adjusted to enhance the heat exchange uniformity between different battery modules and limit the rapid drop in the target coolant temperature to avoid local overcooling.

[0015] As a further limitation of the first aspect of the present invention, when the thermal management demand state is a low-temperature preheating state, the target coolant temperature is preferentially increased and the target coolant flow rate is limited to a preset low flow range to reduce heat dispersion caused by excessive circulation flow under low-temperature conditions; when the thermal management demand state is a transient power response state, the target coolant flow rate is feedforward increased according to the power demand change rate, charge / discharge rate change rate and / or real-time heat generation intensity change rate, and the target coolant temperature is pre-adjusted simultaneously to improve the thermal response speed under high-power transient conditions.

[0016] As a further limitation of the first aspect of the present invention, the target coolant temperature and the target coolant flow rate are generated through a state-related coupling mapping relationship corresponding to the thermal management demand state. The state-related coupling mapping relationship includes at least temperature channel weight, flow channel weight and cross-coupling coefficient to characterize the synergistic influence relationship between the temperature regulation demand and the flow regulation demand.

[0017] As a further limitation of the first aspect of the present invention, the optimized allocation model takes at least two of the following as optimization targets: average temperature deviation of battery pack, maximum temperature deviation of individual cells, temperature difference, energy consumption of liquid cooling system, power consumption of cooling or heating, and switching cost of actuator. It also solves the cooperative control quantity by combining target coolant temperature constraint, target coolant flow constraint, and actuator working boundary. The change rate constraint, minimum holding time constraint, and upper and lower limit constraints are set for the target coolant temperature and the target coolant flow to limit abrupt adjustment of actuator and improve the stability of cooperative control.

[0018] As a further limitation of the first aspect of the present invention, based on the response relationship between the highest temperature of the single cell, the average temperature of the battery pack, and the temperature difference before and after the execution of control and the change in the target coolant temperature and the change in the target coolant flow rate, a temperature-flow coordinated response sensitivity matrix is ​​constructed or updated, and the coordinated control relationship is corrected online based on the temperature-flow coordinated response sensitivity matrix; when the deviation between the coolant flow rate feedback and the target coolant flow rate, the deviation between the coolant temperature feedback and the target coolant temperature, or the deviation between the predicted temperature response and the actual temperature response exceeds a preset threshold for multiple consecutive control cycles, a conservative coordinated control mode is triggered, wherein the conservative coordinated control mode includes one or more of the following: increasing the safety margin, narrowing the search range of control variables, and invoking a fault protection control strategy.

[0019] Secondly, the present invention provides a coolant flow and temperature coordinated control system based on a BMS, comprising:

[0020] The acquisition module is used to collect thermal state parameters, electrical state parameters and environmental state parameters of the power battery system. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge and discharge current, charge and discharge rate, power demand or health state. The environmental state parameters include at least one or more of the following: ambient temperature or external heat exchange boundary conditions.

[0021] The identification module is used to identify the current thermal management requirement state of the power battery system based on the thermal state parameters, electrical state parameters and environmental state parameters. The thermal management requirement state includes at least one of the following: high temperature heat dissipation state, temperature difference equalization state, low temperature preheating state, transient power response state or composite regulation state.

[0022] The decoupling module is used to decouple the thermal management control requirements of the power battery system into temperature regulation requirements and flow regulation requirements according to the thermal management requirement status. The temperature regulation requirements are used to characterize the target supply temperature or heat exchange capacity requirements of the coolant, and the flow regulation requirements are used to characterize the circulation intensity or flow path distribution requirements of the coolant.

[0023] The adjustment module is used to determine the target coolant temperature and target coolant flow rate based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement state, combined with preset control rules, thermal management mapping model, and / or optimized allocation model, and dynamically adjust the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate under different thermal management requirement states;

[0024] The linkage control module is used to perform linkage control on at least one actuator in the liquid cooling system based on the target coolant temperature and the target coolant flow rate, so as to adjust the coolant circulation flow rate, supply temperature, flow path distribution and heat exchange capacity.

[0025] The correction module is used to perform rolling corrections on the target coolant temperature and the target coolant flow rate based on the battery temperature feedback, coolant temperature feedback, coolant flow rate feedback and battery operating condition change information after the execution control, and to update the thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow rate online, so as to form a closed-loop adaptive collaborative thermal management control based on BMS.

[0026] Thirdly, the present invention provides a non-transitory computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the BMS-based coordinated control method for coolant flow and temperature as described in the first aspect.

[0027] Fourthly, the present invention provides a computer device including a memory and a processor, wherein the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the BMS-based coolant flow and temperature coordinated control method as described in the first aspect.

[0028] Fifthly, the present invention provides an electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to cause the electronic device to execute instructions for implementing the BMS-based coolant flow and temperature coordinated control method as described in the first aspect.

[0029] The beneficial effects of this invention are as follows: This invention does not directly map temperature difference to target flow rate, nor does it directly map zone temperature to target liquid temperature. Instead, it first decouples thermal management control requirements into temperature regulation requirements and flow rate regulation requirements, and then recouples and allocates these two types of requirements according to the state of thermal management requirements. Therefore, it can avoid mutual interference between flow rate channels and liquid temperature channels under different control objectives. The introduction of composite regulation states and state-related collaborative mapping matrices allows high-temperature heat dissipation, temperature difference equalization, low-temperature preheating, and transient power response to share a unified control framework while maintaining differentiated collaborative mechanisms, thus better adapting to complex operating conditions. The introduction of real-time heat generation intensity estimated by the BMS and the temperature-flow collaborative response sensitivity matrix enables the control process to simultaneously possess heat source sensing capabilities and online correction capabilities, thereby improving control foresight, robustness, and engineering feasibility.

[0030] The advantages of additional aspects of the invention will be set forth more clearly in the following description or will be learned by practice of the invention. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the overall logic principle of the BMS-based coolant flow and temperature coordinated control method described in this embodiment of the invention.

[0033] Figure 2 This is a flowchart of the BMS-based coolant flow and temperature coordinated control method according to an embodiment of the present invention.

[0034] Figure 3 This is a schematic diagram of the thermal management requirement identification, target decoupling, and state-related recoupling control framework described in an embodiment of the present invention.

[0035] Figure 4 This is a schematic diagram illustrating the relationship between actuator linkage control and closed-loop online correction as described in an embodiment of the present invention. Detailed Implementation

[0036] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0037] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0038] It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as here.

[0039] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, and / or groups thereof.

[0040] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0041] To facilitate understanding of the present invention, the present invention will be further explained and described below with reference to the accompanying drawings and specific embodiments. However, the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0042] Those skilled in the art should understand that the accompanying drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily essential for implementing the present invention.

[0043] Example 1

[0044] In this embodiment 1, a coolant flow and temperature coordinated control system based on a Battery Management System (BMS) is first provided, comprising: an acquisition module for acquiring thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system, wherein the thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack, and the temperature difference; the electrical state parameters include at least one or more of the following: state of charge, charge / discharge current, charge / discharge rate, power demand, or health state; and the environmental state parameters include at least one or more of the following: ambient temperature or external heat exchange boundary conditions. An identification module is used to identify the current thermal management demand state of the power battery system based on the thermal state parameters, electrical state parameters, and environmental state parameters, wherein the thermal management demand state includes at least one of the following: high-temperature heat dissipation state, temperature difference equalization state, low-temperature preheating state, transient power response state, or composite regulation state. A decoupling module is used to decouple the thermal management control demand of the power battery system into temperature regulation demand and flow regulation demand according to the thermal management demand state, wherein the temperature regulation demand characterizes the target supply temperature or heat exchange capacity demand of the coolant, and the flow regulation demand characterizes the coolant circulation intensity or flow path allocation demand. An adjustment module is used to determine the target coolant temperature and target coolant flow rate based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement status, combined with preset control rules, a thermal management mapping model, and / or an optimized allocation model. It then dynamically adjusts the control priority and coupling relationship between the target coolant temperature and the target coolant flow rate under different thermal management requirement statuses. A linkage control module is used to perform linkage control on at least one actuator in the liquid cooling system based on the target coolant temperature and the target coolant flow rate to adjust the coolant circulation flow rate, supply temperature, flow path allocation, and heat exchange capacity. A correction module is used to perform rolling corrections on the target coolant temperature and target coolant flow rate based on battery temperature feedback, coolant temperature feedback, coolant flow rate feedback, and battery operating condition change information after control execution. It also updates the thermal management requirement status and the collaborative control relationship between the target coolant temperature and the target coolant flow rate online to form a closed-loop adaptive collaborative thermal management control based on the BMS.

[0045] In this embodiment, the above-described system is used to implement a BMS-based method for coordinated control of coolant flow and temperature, characterized by the following steps:

[0046] S1. Operation Status Acquisition Steps: The Battery Management System (BMS) acquires the thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack, and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge / discharge current, charge / discharge rate, power demand, and health status. The environmental state parameters include at least one or more of the following: ambient temperature and external heat transfer boundary conditions.

[0047] S2. Thermal management status identification step: Based on the thermal status parameters, electrical status parameters and environmental status parameters, identify the current thermal management requirement status of the power battery system. The thermal management requirement status includes at least one of the following: high temperature heat dissipation status, temperature difference equalization status, low temperature preheating status, transient power response status and composite regulation status.

[0048] In this embodiment, the identification of thermal management demand states adopts a deep learning classification model based on TabTransformer. The model is deployed and runs online by the BMS, taking multi-dimensional operating condition data of the current control cycle and the previous N historical cycles as input, outputting the probability distribution of each thermal management demand state, and taking the state corresponding to the highest probability value as the current identification result.

[0049] Specifically, the training of the model in this embodiment includes the following:

[0050] (a) Input characteristics

[0051] The model's input feature vector x(k) contains the following numerical and categorical features:

[0052] Numerical features (24 dimensions in total): 8 dimensions of thermal state: maximum temperature of individual cells, minimum temperature of individual cells, average temperature of battery pack, temperature difference, maximum temperature rise rate, average temperature rise rate, supply liquid temperature, return liquid temperature; 6 dimensions of electrical state: state of charge, charge / discharge current, charge / discharge rate, power demand, health status, real-time heat generation intensity; 2 dimensions of environmental state: ambient temperature, external heat exchange capacity; 8 dimensions of feedback: historical sequence of numerical features from the previous N steps.

[0053] Category-based characteristics (3 dimensions in total): Current operating mode: normal discharge, normal charging, fast charging, parking insulation, and attenuation compensation; Current thermal management status: the status identified in the previous cycle; Battery system: ternary lithium, lithium iron phosphate, and others.

[0054] (II) TabTransformer Model Architecture:

[0055] The model consists of four parts: (1) Numerical feature embedding layer: The 24-dimensional numerical features are mapped to the d_model=64-dimensional embedding space through independent MLPs to obtain 24 64-dimensional numerical embedding vectors. (2) Categorical feature embedding layer: The three categorical features are mapped to 64-dimensional embedding vectors through independent embedding tables. (3) Transformer encoder: The 29 embedding vectors (24 numerical embeddings + 5 categorical embeddings) are concatenated into sequence X and fed into the L-layer Transformer encoder. (4) Classification head: The [CLS] label output by the Transformer encoder (or global average pooling of the output) is activated by two layers of MLP and Softmax to output the probability distribution of five thermal management demand states: .

[0056] In the Transformer encoder, each layer contains a multi-head self-attention network and a feedforward network:

[0057] ;

[0058] in, The multi-head self-attention mechanism enables the model to automatically learn the cross-coupling relationships between different sensor quantities.

[0059] (III) Loss Function and Training: The model is trained using labeled actual BMS running data. The training loss function is cross-entropy loss: ;

[0060] in, Let S be the number of training samples, and S be the set of five thermal management requirement states. Let i be the true label of the i-th sample. Here are the predicted probabilities of state s from the model. Training is performed using the AdamW optimizer with an initial learning rate of 1×10⁻⁻⁻⁶. 4 Weight decay 1×10⁻ 5 The batch size is 256, the training rounds are 500, and an early stopping strategy is adopted (training stops when the validation set loss does not decrease for 10 consecutive rounds).

[0061] (iv) Parameter meaning: = Scaling factor in attention calculation, set to 64; L = Number of Transformer encoder layers, configured to 4; Number of attention heads set to 4, each head dimension The hidden layer dimension of the feedforward network is 256.

[0062] (V) Dataset Construction: The dataset is constructed based on actual BMS running data. See section S4 for a detailed description of the dataset. The state labels are obtained as follows: First, the rule-based state scoring method described in Example 2 is used to automatically generate initial labels for each time series segment. Then, domain experts review and correct any ambiguous boundary segments to form the final labeled dataset. The dataset is divided into a training set, a validation set, and a test set in a 7:2:1 ratio.

[0063] S3. Thermal management target decoupling step: According to the thermal management demand state, the thermal management control demand of the power battery system is decoupled into temperature regulation demand and flow regulation demand. The temperature regulation demand is used to characterize the target supply temperature or heat exchange capacity demand of the coolant, and the flow regulation demand is used to characterize the coolant circulation intensity or flow path distribution demand.

[0064] In this embodiment, the decoupling of thermal management control requirements is achieved through the multi-task learning architecture of the TabTransformer model, rather than relying on an explicit weighting formula. The decoupling process is embedded in the model's output layer design: after the shared front-end TabTransformer encoder extracts a unified multi-dimensional operating condition feature representation, it connects to two independent output branches:

[0065] (1) Temperature regulation demand branch: The shared features are activated through MLP and Sigmoid to output the normalized temperature regulation demand intensity. Then, by inverse normalization to the actual range, we obtain... : ;

[0066] (2) Flow regulation demand branch: The shared features are activated through MLP and Sigmoid to output the normalized flow regulation demand intensity. ∈ [0,1], then inverse normalization yields : The two branches share encoder parameters but have independent fully connected layer parameters. During training, they automatically learn how to separate the regulation signals required for temperature control and flow control from a unified multidimensional state representation through backpropagation.

[0067] S4. Cooperative control quantity generation step: Based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement state, combined with preset control rules and thermal management mapping model, determine the target coolant temperature and the target coolant flow rate, and dynamically adjust the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate under different thermal management requirement states.

[0068] In this embodiment, both the thermal management mapping model and the optimal allocation model employ a deep reinforcement learning model based on Soft Actor-Critic (SAC) to realize the transformation from multi-dimensional operating conditions to collaborative control variables. , The model performs an end-to-end mapping of the system. It simultaneously undertakes the dual functions of mapping relationships and optimizing allocation. During training, it automatically learns the optimal cooperative control strategy through interaction with the environment (or batch offline training with historical data).

[0069] (I) Construction of the dataset

[0070] The dataset is built based on actual BMS runtime data and includes the following processing steps:

[0071] Step 1: Raw Data Acquisition: Deploy a data acquisition terminal on a power battery test bench equipped with a liquid-cooled thermal management system, and continuously record the following fields at a sampling frequency of 1Hz:

[0072] 1. Timestamps, all thermal state variables, electrical state variables, and environmental state variables reported by the BMS;

[0073] 2. Target coolant temperature actually output by the BMS And historical N-step values;

[0074] 3. Target coolant flow rate actually output by the BMS And historical N-step values;

[0075] 4. Actuator feedback: actual pump speed, valve opening, cooling / heating power;

[0076] 5. Temperature feedback after execution: The measured value;

[0077] 6. Corresponding thermal management requirement status label;

[0078] 7. The operation mode of the annotation.

[0079] Accumulate no less than 10,000 hours of operational data, covering all four seasons and encompassing various vehicle models and battery systems, with a total sample size of no less than 3.6 × 10⁻⁶. 7 strip.

[0080] Step 2: Data Preprocessing

[0081] 1. Missing value handling: Short-term missing values ​​are filled using linear interpolation; segments with consecutive missing values ​​exceeding 10 seconds are directly removed;

[0082] 2. Outlier detection: Identify abnormal sensor jumps based on the 3σ principle and perform smooth corrections;

[0083] 3. Time alignment: The delay of each sensor is calibrated to a unified time reference;

[0084] 4. Normalization: Standardize all numerical features to a mean of 0 and a variance of 1;

[0085] 5. Sample construction: A sample is formed by taking N=30 consecutive time steps and sliding the sample with a step size of 1.

[0086] Step 3: Dataset partitioning

[0087] In chronological order, the first 70% is the training set, the middle 15% is the validation set, and the last 15% is the test set.

[0088] (II) SAC Reinforcement Learning Model

[0089] State space: The 29 embedding vectors described in S2 (embedded representations of 24 numerical features + 5 categorical features) are merged into a state vector s∈R. 29×64 After average pooling, s ∈ R 64 .

[0090] Action space: continuous two-dimensional action After being normalized to the [0,1] interval, they are inversely normalized to the range of physical quantities.

[0091] Reward function: The training reward function r(s,a) of SAC balances thermal safety, temperature uniformity, energy consumption, and smoothness.

[0092]

[0093] Where: f(·) represents a piecewise function: when the temperature is within the safe range, it uses a squared penalty; when it exceeds the safe threshold, it switches to an exponential penalty (significantly increasing the penalty intensity). Normalized average temperature deviation; This represents the normalized maximum temperature deviation of the monomer. Normalized temperature difference; Normalized total power consumption; ||a-a_prev||² is the square of the change in motion, suppressing frequent switching of the actuator; To constrain the penalty for violating instructions, it is applied when the action exceeds the constraint boundary; w1~w6 are the reward weights, which are determined through hyperparameter tuning.

[0094] In this embodiment, the core formula of the SAC algorithm is: SAC introduces an entropy maximization term into the standard objective of maximizing cumulative reward to promote the exploration of the strategy:

[0095] (1) Strategy optimization objective:

[0096] ;

[0097] That is, while maximizing the action value function Q, we also maximize the policy entropy H, where α is the temperature coefficient that controls the balance between exploration and exploitation.

[0098] (2) Q-function update:

[0099] , where γ is the discount factor and θ' is the parameter of the Q target network.

[0100] (3) Adaptive update of temperature coefficient α: ;in Let the target entropy be -dim(Action), which is -2.

[0101] Core hyperparameter configuration: Discount factor γ = 0.99; SAC temperature coefficient α initial value = 0.2 (adaptively adjusted online); Actor network: 64→256→256→2, hidden layer ReLU, output layer Tanh; Critic network (double Q network): 64→256→256→1, hidden layer ReLU; Experience replay pool capacity R = 1×10^6; Target network soft update coefficient τ = 0.005; Batch size = 256; Learning rate for both Actor and Critic is 3×10⁻ 4 .

[0102] (III) Model training process;

[0103] Phase 1 – Offline Pre-training (based on historical datasets):

[0104] 1. Randomly sample batch transfer samples from dataset D ( ) ;

[0105] 2. Update both Critic networks (minimize) );

[0106] 3. Update the Actor network at regular intervals (minimize) );

[0107] 4. Update the temperature coefficient α;

[0108] 5. Softly update the target network parameters;

[0109] 6. Repeat until convergence (the validation set reward will no longer be increased).

[0110] Phase Two – Online Fine-Tuning: After model deployment, new (s,a,r,s') transfer samples are continuously collected during the actual operation of BMS, using a low learning rate (1×10⁻⁶). -5The strategy network is updated online to enable the control strategy to adapt to battery aging, actuator performance drift and environmental changes.

[0111] (iv) Inference Phase: In each control cycle of the BMS, the current operating condition features are input into the TabTransformer encoder to obtain the shared representation s, which is then input into the trained SAC Actor network to output [ After inverse normalization and the addition of constraint limiting, the final target coolant temperature is obtained. and target coolant flow rate .

[0112] S5. Actuator linkage control steps: Based on the target coolant temperature and the target coolant flow rate, at least one actuator in the liquid cooling system is linked for control. The actuator includes one or more of the following: variable speed liquid pump, proportional valve, three-way valve, heat exchanger, refrigeration unit, heating unit, and branch switching component, to adjust the coolant circulation flow rate, supply temperature, flow path distribution, and heat exchange capacity.

[0113] S6. Closed-loop feedback correction step: Based on the battery temperature feedback, coolant temperature feedback, coolant flow feedback and battery operating condition change information after execution control, the target coolant temperature and the target coolant flow are rolled over and corrected, and the thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow are updated online to form a closed-loop adaptive collaborative thermal management control based on BMS.

[0114] In this embodiment, rolling correction is implemented through the online fine-tuning mechanism of the SAC strategy. The correction mechanism is as follows: After the execution of the k-th control cycle is completed, the BMS obtains the following feedback: Actual battery temperature feedback; Feedback from the actual implementing agency; Actual energy consumption feedback; calculate the immediate reward r(k) and transfer the sample. Store in the online experience replay buffer.

[0115] Rolling correction steps: After each control cycle k is completed:

[0116] 1. Constructing transfer samples Stored in the online experience replay buffer;

[0117] 2. If the cumulative number of samples in the online experience replay buffer is greater than or equal to the batch size:

[0118] a. Sample batches of data from the online experience replay buffer;

[0119] b. Using the learning rate =1×10-5 Update the Critic network;

[0120] c. Using the learning rate Update the Actor network;

[0121] d. Soft update the target network;

[0122] 3. If the reward r(k) is lower than the preset threshold for three consecutive control cycles: enter conservative collaborative control mode: temporarily increase the learning rate to [a certain threshold]. Increase the reward function weights w1 and w2 by 20% (to enhance the penalty for temperature deviation); Add a -5% offset to the action output (safety margin).

[0123] 4. Once the reward r(k) returns to the normal range and remains so for 10 cycles: exit conservative mode and restore normal learning rate and weights.

[0124] In step S1, the thermal state parameters include one or more of the following: battery module zone temperature, coolant supply temperature, coolant return temperature, and coolant supply-return temperature difference; the external heat exchange boundary conditions include at least one or more of the following: radiator air-side inlet temperature, air-side flow rate, available heat exchange capacity of the vehicle thermal management system, available power of the refrigeration unit, and available power of the heating unit.

[0125] After step S1, the BMS estimates the real-time heat generation intensity of the power battery system based on the state of charge, charge / discharge current, state of health, cell open-circuit voltage and / or equivalent internal resistance, and uses the real-time heat generation intensity as an auxiliary input for steps S2 and S4.

[0126] In step S2, a corresponding state score or state confidence level is calculated for at least two types of thermal management demand states (the softmax probability value output by TabTransformer is directly used as the score and confidence level index for each state). When the state scores of two or more types of thermal management demand states are simultaneously higher than their respective preset thresholds, the current state is identified as the composite regulation state, and the weight of each state in the composite regulation state is determined. In step S2, the thermal management demand state identification adopts a combination of hysteresis threshold and minimum hold time to suppress frequent switching between high-temperature heat dissipation state, temperature difference equalization state, low-temperature preheating state, transient power response state, and composite regulation state.

[0127] In step S3, the temperature regulation requirement is mainly determined by the average temperature deviation of the battery pack, the minimum temperature deviation of individual cells, the ambient temperature deviation, and the target operating temperature zone requirement; the flow regulation requirement is mainly determined by the maximum temperature of individual cells, temperature difference, charge / discharge rate, power demand change, and real-time heat generation intensity (automatically learned and determined by the multi-task learning branch of TabTransformer).

[0128] When the thermal management requirement is a high-temperature heat dissipation state, the target coolant temperature is reduced first; when the highest temperature of a single unit and / or the rate of increase of the highest temperature of a single unit exceeds a preset hot spot threshold, the priority of the target coolant flow rate in the coordinated control is increased to suppress the rapid development of local hot spots.

[0129] When the thermal management requirement is in a temperature difference equilibrium state, priority is given to increasing the target coolant flow rate and / or adjusting the flow path distribution to enhance the heat exchange uniformity between different battery modules and limit the rapid drop in target coolant temperature to avoid local overcooling.

[0130] When the thermal management requirement is in a low-temperature preheating state, the target coolant temperature is increased first, and the target coolant flow rate is limited to a preset low flow rate range to reduce heat dissipation caused by excessive circulation flow rate under low-temperature conditions.

[0131] When the thermal management demand state is a transient power response state, the target coolant flow rate is fed forward and increased according to the power demand change rate, charge / discharge rate change rate and / or real-time heat generation intensity change rate, and the target coolant temperature is pre-adjusted simultaneously to improve the thermal response speed under high power transient conditions.

[0132] In step S4, the target coolant temperature and the target coolant flow rate are generated through a state-related coupling mapping relationship corresponding to the thermal management demand state. The state-related coupling mapping relationship includes at least temperature channel weight, flow channel weight, and cross-coupling coefficient (dynamically adjusted by the SAC policy network according to the relative priority of the final reward) to characterize the synergistic influence relationship between the temperature regulation demand and the flow regulation demand.

[0133] In step S4, the optimization allocation model takes at least two of the following as optimization objectives: average temperature deviation of battery pack, maximum temperature deviation of single cell, temperature difference, energy consumption of liquid cooling system, power consumption of cooling or heating, and switching cost of actuator. It also solves the cooperative control quantity by combining the target coolant temperature constraint, the target coolant flow constraint, and the working boundary of actuator.

[0134] The optimized model is a SAC policy network. The training objective of SAC is equivalent to solving the following optimization problem:

[0135] This objective function is equivalent to the multinomial optimal control objective of the optimization model in Example 2:

[0136] : From the reward function reflect;

[0137] :Depend on reflect;

[0138] min ΔT_pack: by reflect;

[0139] min energy consumption: by reflect;

[0140] min Actuator switching: by reflect;

[0141] Constraints satisfied: by reflect.

[0142] In steps S4 and S5, constraints are set on the rate of change, minimum holding time, and upper and lower limits for the target coolant temperature and the target coolant flow rate to limit abrupt adjustments by the actuator and improve the stability of coordinated control.

[0143] In step S5, when the liquid cooling system includes multiple parallel branches or multiple liquid cooling plate areas, the branch allocation coefficient is determined according to the battery module temperature sorting results and temperature difference distribution results in each area, and the coolant flow distribution of different branches is adjusted through the proportional valve, three-way valve and / or branch switching component.

[0144] In step S6, based on the response relationship between the highest temperature of the single cell, the average temperature of the battery pack, and the temperature difference before and after the execution of control, and the change in the target coolant temperature and the change in the target coolant flow rate, a temperature-flow rate coordinated response sensitivity matrix is ​​constructed or updated, and the coordinated control relationship is corrected online based on the temperature-flow rate coordinated response sensitivity matrix.

[0145] When the deviation between the coolant flow feedback and the target coolant flow, the deviation between the coolant temperature feedback and the target coolant temperature, or the deviation between the predicted temperature response and the actual temperature response exceeds a preset threshold for multiple consecutive control cycles, a conservative collaborative control mode is triggered. The conservative collaborative control mode includes one or more of the following: increasing the safety margin, narrowing the search range of control variables, and invoking a fault protection control strategy.

[0146] In this embodiment, the method further includes switching the corresponding thermal management parameter set, control priority parameter set, and actuator constraint boundary set according to the different operating modes of the power battery system in the normal discharge mode, normal charging mode, fast charging mode, parking insulation mode, and attenuation compensation mode.

[0147] Example 2

[0148] In this embodiment 2, a method for coordinated control of coolant flow and temperature based on BMS is provided to solve the problems in the prior art where flow control and liquid temperature control are independent or mechanically bound, control priority is difficult to dynamically adjust under operating conditions, and execution feedback is difficult to correct the coordinated relationship in reverse.

[0149] like Figures 1 to 4 As shown, a dual-channel collaborative control framework for liquid cooling thermal management of power batteries is proposed. This framework uses the Battery Management System (BMS) as the core for state perception and decision-making. Instead of directly mapping all thermal management requirements to a single control variable, it first identifies the current thermal management requirements from the battery's thermal state, electrical state, and environmental state. Then, it decouples the thermal management requirements into temperature regulation requirements and flow regulation requirements. Subsequently, based on the current requirement state, it recouples the two types of requirements according to state correlation, ultimately generating the target coolant temperature and target coolant flow rate, and achieving closed-loop regulation through actuator linkage.

[0150] The core of this embodiment lies not in simply increasing the coolant flow rate, nor in simply changing the coolant supply temperature, but in establishing a control link of "state identification—target decoupling—priority reconfiguration—cooperative execution—online update". Through this link, the target coolant temperature and target coolant flow rate can assume different primary and secondary control functions under different thermal management demand states, thereby enabling differentiated cooperative adjustment mechanisms for high-temperature heat dissipation, temperature difference equalization, low-temperature preheating, and transient power response.

[0151] At the state perception level, this embodiment uses the BMS to uniformly collect the thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system. Preferably, it further combines information such as coolant supply temperature, coolant return temperature, supply-return temperature difference, and external heat exchange capacity to form multi-dimensional operating condition state variables. Unlike schemes that control based solely on a single temperature threshold, this approach incorporates average temperature, maximum single-cell temperature, minimum single-cell temperature, temperature difference, state of charge, charge / discharge rate, power demand, health status, and environmental boundaries into the decision-making basis to improve the completeness of thermal management requirement identification.

[0152] In this embodiment, the real-time heat generation intensity of the battery system can be estimated by the BMS, and this real-time heat generation intensity can be incorporated into the state identification and cooperative control quantity generation process. The real-time heat generation intensity can be expressed as:

[0153] ;

[0154] in, I(k) represents the real-time heat generation intensity during the k-th control cycle, and I(k) represents the charging and discharging current. This represents the equivalent internal resistance estimated by BMS based on SOC, SOH, and temperature. Indicates the average temperature of the battery pack. This represents the rate of change of open-circuit voltage with respect to temperature. By incorporating the real-time heat generation intensity, control decisions can be made not only based on the thermal output quantity but also simultaneously sensing changes in heat source intensity.

[0155] At the thermal management demand identification level, instead of a simple binary classification of hot and cold, the current thermal management demand is divided into at least four states: high-temperature heat dissipation, temperature difference equilibrium, low-temperature preheating, transient power response, and composite regulation. Among them, the composite regulation state is used to describe the operating condition where two or more types of thermal management demands coexist significantly, so as to avoid forcibly adopting a single mode of control when multiple objectives conflict.

[0156] A status score is calculated for different thermal management requirements, and the status is confirmed by combining the hysteresis threshold and the minimum hold time. For example, the high-temperature heat dissipation status score can be mainly determined by the highest temperature of the individual cell, the average temperature deviation, and the real-time heat generation intensity; the temperature difference balance status score can be mainly determined by the temperature difference, the temperature dispersion of the zones, and the supply and return liquid temperature difference; the low-temperature preheating status score can be mainly determined by the lowest temperature of the individual cell, the ambient temperature, and the low-temperature charging requirement; and the transient power response status score can be mainly determined by the power demand change rate and the rate change rate.

[0157] At the decoupling level of control requirements, the thermal management requirements of the power battery are decomposed into two control channels: temperature regulation requirements and flow regulation requirements. Temperature regulation requirements primarily determine the coolant supply temperature or heat exchange capacity level, while flow regulation requirements primarily determine the coolant circulation intensity, branch distribution, and heat exchange range. This decoupling mechanism allows the temperature and flow channels to correspond to different thermal management objectives, rather than being driven synchronously by the same error.

[0158] Specifically, the temperature regulation requirement and the flow rate regulation requirement can be expressed as follows:

[0159] ;

[0160] ;

[0161] in, Indicates temperature regulation requirements. This indicates the flow regulation demand, and s represents the current thermal management demand status. Indicates the average temperature deviation. Indicates the minimum temperature deviation. This represents the normalized real-time heat production intensity. Indicates environmental boundary deviation, Indicates the temperature deviation of hot spots. Indicates the temperature difference of the battery pack. Indicates the charge / discharge rate. The power demand change rate is represented by a1(s) to a4(s) and b1(s) to b4(s), which are weighting coefficients related to the thermal management demand state. Through this method, average temperature control, low-temperature preheating, and local hotspot suppression are treated differently in the control structure.

[0162] At the target generation level, this embodiment does not... and Instead of independently converting them into target coolant temperature and target coolant flow rate, a state-dependent coupling mapping relationship is further established to recouple and allocate the two types of demands. Target coolant temperature and target coolant flow rate can be expressed as:

[0163] ;in, Indicates the target coolant temperature. Indicates the target coolant flow rate. This represents the collaborative mapping matrix corresponding to the current thermal management demand state. The collaborative mapping matrix includes at least temperature channel weights, flow channel weights, and cross-coupling coefficients, used to characterize the collaborative relationship between the liquid temperature channel and the flow channel.

[0164] By introducing the aforementioned collaborative mapping matrix, the control priority between the target liquid temperature and the target flow rate can be dynamically changed under different conditions.

[0165] For example, under high-temperature heat dissipation conditions, the cooperative mapping matrix preferably increases the weight of the temperature channel and increases the flow cross-coupling coefficient when the highest temperature of a single unit or the temperature rise rate of a hot spot exceeds a threshold; under temperature difference equilibrium conditions, it is preferable to increase the weight of the flow channel and the branch allocation weight to enhance temperature uniformity; under low-temperature preheating conditions, it is preferable to increase the weight of the temperature channel and limit the upper limit of the flow rate to reduce the heat carried away by the low temperature; under transient power response conditions, it is preferable to introduce a feedforward boost term to the flow channel to improve the response speed under short-term thermal shock.

[0166] At the level of solving the collaborative control variables, this embodiment can employ one or more of the following: preset control rules, thermal management mapping model, and optimized allocation model. The optimized allocation model uses the average temperature deviation, the highest temperature deviation of a single unit, the temperature difference, the energy consumption of the liquid cooling system, the power consumption of cooling or heating, and the switching cost of the actuator as comprehensive optimization objectives, and combines the target coolant temperature constraint, the target coolant flow constraint, the working boundary of the actuator, and the change rate constraint for solution.

[0167] The overall optimization objective can be expressed as:

[0168]

[0169] in, Indicates the power consumption of the liquid pump. The power consumption of the cooling and / or heating units is represented by u(k), the control vector of the actuator is represented by w1 to w6, and the weighting coefficients are represented by w1 to w6. This method ensures thermal safety while balancing temperature uniformity, operational stability, and system energy consumption.

[0170] At the closed-loop feedback correction level, the collaborative mapping matrix or its key parameters are updated online using battery temperature feedback, coolant temperature feedback, coolant flow feedback and operating condition change information after execution control, so that the collaborative relationship between target liquid temperature and target flow can be adaptively adjusted with changes in environment, actuator performance drift and battery degradation state.

[0171] The online update process can be based on a temperature-flow coordinated response sensitivity matrix. This matrix describes the response relationship between the highest temperature of a single cell, the average temperature of the battery pack, and the temperature difference to changes in coolant temperature and flow rate. Using this matrix, closed-loop error information can be applied inversely to the coordinated mapping matrix, thereby improving control consistency over long-term operation.

[0172] In this embodiment, the power battery liquid cooling thermal management objects include the power battery system, BMS, liquid cooling circuit, and actuators connected to the liquid cooling circuit. The actuators include at least a variable-speed liquid pump and may further include a proportional valve, a three-way valve, a heat exchanger, a refrigeration unit, a heating unit, and a branch switching component. The BMS is used to collect battery thermal state, electrical state, and environmental state, and to identify and make decisions regarding thermal management requirements.

[0173] During the k-th control cycle, the BMS collects the highest temperature of the single cell. Minimum temperature of monomer Average battery pack temperature Temperature difference State of charge (SOC) (k), charge / discharge current (I) (k), charge / discharge rate Power requirements Health status (SOH(k)) and ambient temperature Coolant supply temperature Coolant return temperature Information such as coolant flow rate Q(k).

[0174] In this embodiment, the real-time heat generation intensity can be further estimated by the BMS. Preferably, R_eq(k) is estimated by BMS based on SOC, SOH, and temperature. Compared to methods that rely solely on apparent temperature for decision-making, this heat generation intensity estimation can reflect the trend of heat load changes earlier.

[0175] In this embodiment, a status score is calculated for different thermal management requirement states. The high-temperature heat dissipation status score can be expressed as:

[0176]

[0177] The temperature difference equilibrium state score can be expressed as:

[0178]

[0179] The low-temperature preheating status score can be expressed as:

[0180]

[0181] The transient power response state score can be expressed as:

[0182] ;

[0183] in, This indicates the temperature dispersion of the battery module. Indicates the charging rate. This indicates the high temperature discrimination threshold. c1 represents the low-temperature discrimination threshold, and c1 to f3 represent the scoring weights. After normalizing each scoring result, the state confidence level of each thermal management requirement state is obtained.

[0184] In this embodiment, when the confidence level of at least two types of thermal management demand states is higher than their respective thresholds, the current state is identified as a composite adjustment state.

[0185] The weights of each substate under the composite adjustment state can be expressed as:

[0186]

[0187] To avoid frequent switching, a hysteresis threshold and a minimum hold duration are introduced into the state transition judgment. When the confidence level of a candidate state first exceeds the entry threshold, entry is only permitted if the duration of that state reaches a certain threshold. The switch is confirmed only after the confidence level of the state drops below the exit threshold; exiting the state is only allowed when the confidence level of the state drops below the exit threshold.

[0188] After identifying the thermal management requirements, the process moves to the target decoupling phase. Based on the current state, temperature regulation requirements are then constructed. and flow regulation needs :

[0189] .

[0190] in, , This represents the environmental boundary deviation. Through the above decoupling, the average temperature control, hot spot control, low-temperature preheating, and temperature difference equalization are no longer driven by the same error quantity.

[0191] Under high-temperature heat dissipation conditions, the temperature channel plays a primary control role, mainly reducing the target coolant temperature. When the highest unit temperature or temperature rise rate consistently exceeds the hot spot threshold, the flow channel weight is increased to accelerate the removal of heat from the hot spot. Under temperature difference equilibrium conditions, the flow channel plays a primary control role, adjusting the flow rate in different areas in conjunction with the branch distribution coefficient. Under low-temperature preheating conditions, the primary focus is on increasing the target coolant temperature, while limiting the flow rate to a low range to improve heat utilization. Under transient power response conditions, a [missing information - likely a function or mechanism] is introduced... The feedforward flow enhancement is triggered to prepare for future short-term thermal shocks in advance.

[0192] In this embodiment, the state-related cooperative mapping matrix is ​​used. Generate the target coolant temperature and target coolant flow rate, i.e.:

[0193]

[0194] For a composite regulation state, a composite cooperative mapping matrix can be formed by weighting multiple single-state cooperative mapping matrices, i.e.:

[0195] During the coordinated control phase of the implementing agencies, according to and Generate the actuator control vector u(k). Preferably, Where n_p(k) represents the pump speed, Indicates the opening degree of the proportional valve. Indicates the opening degree of the three-way valve. Indicates the target output of the cooling unit. Indicates the target output of the heating unit. This represents the branch allocation coefficient.

[0196] When the liquid cooling system includes multiple parallel branches or multiple liquid cooling plate regions, branch allocation coefficients can be generated based on the temperature ranking and temperature difference distribution of the battery modules in each region. For example, for the r-th branch, its branch allocation coefficient can be expressed as:

[0197]

[0198] in, It can be obtained by weighting the highest temperature deviation, average temperature deviation, and temperature difference contribution of the corresponding area of ​​the branch. By adjusting the branch allocation coefficient, hot spots and highly discrete areas can be given priority in obtaining heat exchange resources.

[0199] To balance thermal safety and energy consumption, this implementation method can also construct an optimized allocation model.

[0200] Preferably, the comprehensive objective function is:

[0201] The solution is obtained by combining the working boundary of the actuator, the target liquid temperature boundary, the target flow rate boundary, the rate of change constraint, and the minimum holding time constraint.

[0202] During the closed-loop feedback correction phase, the BMS acquires the execution control... , , , , Using feedback quantities such as Q(k+1), calculate the deviation between the target execution quantity and the actual execution quantity, and apply the state-related collaborative mapping matrix. Perform online correction. Preferably, a temperature-flow coordinated response sensitivity matrix can be constructed:

[0203]

[0204] Based on closed-loop error The co-mapping matrix is ​​updated recursively, i.e.:

[0205]

[0206] in, This indicates updating the step size matrix. This represents the error vector composed of temperature feedback deviation, flow rate feedback deviation, and liquid temperature feedback deviation. In this way, the synergistic relationship can be dynamically adjusted as the cooling circuit ages, pump efficiency changes, battery degradation occurs, and environmental boundaries change.

[0207] When the deviations between the coolant flow feedback and the target coolant flow rate, the coolant temperature feedback and the target coolant temperature, or the predicted temperature response and the actual temperature response exceed a threshold for multiple consecutive control cycles, a conservative cooperative control mode can be triggered. The conservative cooperative control mode includes one or more of the following: increasing the safety margin, narrowing the control variable search range, reducing the actuator switching frequency, and invoking fault protection strategies.

[0208] If the BMS detects during fast charging... Rapidly rising and When the temperature approaches the hot spot threshold, the flow channel weight is increased in the transient power response state to increase the pump speed in advance. If the average temperature continues to rise, the system switches to high-temperature heat dissipation state and further reduces the target coolant temperature. If the temperature difference between modules increases at the same time, the system enters a composite adjustment state, which simultaneously takes into account both hot spot suppression and temperature uniformity targets through a composite collaborative mapping matrix.

[0209] For example, under low-temperature charging conditions, if If the temperature is below the low-temperature preheating threshold, the target coolant temperature is increased first and the pump is controlled to operate in a low-flow range to achieve preheating. Once the minimum temperature recovers to the allowable charging range and the average temperature is close to the target operating temperature range, the flow rate is gradually increased to improve the temperature uniformity during the charging process.

[0210] In summary, this embodiment achieves coordinated control of coolant temperature and flow rate through state recognition, target decoupling, state-related recoupling, actuator linkage, and closed-loop online correction mechanism centered on BMS, which can improve the foresight, balance, and adaptability of power battery liquid cooling thermal management.

[0211] Example 3

[0212] This embodiment 3 provides a non-transitory computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, they implement the BMS-based coolant flow and temperature coordinated control method described above. The method includes:

[0213] The thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system are collected. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack, and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge / discharge current, charge / discharge rate, power demand, or state of health. The environmental state parameters include at least one or more of the following: ambient temperature or external heat transfer boundary conditions.

[0214] Based on the thermal state parameters, electrical state parameters and environmental state parameters, the current thermal management requirement state of the power battery system is identified. The thermal management requirement state includes at least one of the following: high temperature heat dissipation state, temperature difference equalization state, low temperature preheating state, transient power response state or composite regulation state.

[0215] Based on the thermal management requirement status, the thermal management control requirements of the power battery system are decoupled into temperature regulation requirements and flow regulation requirements. The temperature regulation requirements are used to characterize the target supply temperature or heat exchange capacity requirement of the coolant, and the flow regulation requirements are used to characterize the circulation intensity or flow path distribution requirement of the coolant.

[0216] Based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement status, combined with preset control rules, thermal management mapping models, and / or optimized allocation models, the target coolant temperature and target coolant flow rate are determined, and the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate are dynamically adjusted under different thermal management requirement statuses.

[0217] Based on the target coolant temperature and the target coolant flow rate, at least one actuator in the liquid cooling system is linked for control to adjust the coolant circulation flow rate, supply temperature, flow path distribution, and heat exchange capacity.

[0218] Based on the battery temperature feedback, coolant temperature feedback, coolant flow feedback, and battery operating condition change information after the execution control, the target coolant temperature and the target coolant flow are rolled over and corrected. The thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow are updated online to form a closed-loop adaptive collaborative thermal management control based on BMS.

[0219] Example 4

[0220] This embodiment 4 provides a computer device, including a memory and a processor. The processor and the memory communicate with each other. The memory stores program instructions that can be executed by the processor. The processor calls the program instructions to execute the BMS-based coolant flow and temperature coordinated control method described above. The method includes:

[0221] The thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system are collected. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack, and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge / discharge current, charge / discharge rate, power demand, or state of health. The environmental state parameters include at least one or more of the following: ambient temperature or external heat transfer boundary conditions.

[0222] Based on the thermal state parameters, electrical state parameters and environmental state parameters, the current thermal management requirement state of the power battery system is identified. The thermal management requirement state includes at least one of the following: high temperature heat dissipation state, temperature difference equalization state, low temperature preheating state, transient power response state or composite regulation state.

[0223] Based on the thermal management requirement status, the thermal management control requirements of the power battery system are decoupled into temperature regulation requirements and flow regulation requirements. The temperature regulation requirements are used to characterize the target supply temperature or heat exchange capacity requirement of the coolant, and the flow regulation requirements are used to characterize the circulation intensity or flow path distribution requirement of the coolant.

[0224] Based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement status, combined with preset control rules, thermal management mapping models, and / or optimized allocation models, the target coolant temperature and target coolant flow rate are determined, and the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate are dynamically adjusted under different thermal management requirement statuses.

[0225] Based on the target coolant temperature and the target coolant flow rate, at least one actuator in the liquid cooling system is linked for control to adjust the coolant circulation flow rate, supply temperature, flow path distribution, and heat exchange capacity.

[0226] Based on the battery temperature feedback, coolant temperature feedback, coolant flow feedback, and battery operating condition change information after the execution control, the target coolant temperature and the target coolant flow are rolled over and corrected. The thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow are updated online to form a closed-loop adaptive collaborative thermal management control based on BMS.

[0227] Example 5

[0228] This embodiment 5 provides an electronic device, including: a processor, a memory, and a computer program; wherein, the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory to cause the electronic device to execute instructions for implementing the BMS-based coolant flow and temperature coordinated control method as described above, the method including:

[0229] The thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system are collected. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack, and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge / discharge current, charge / discharge rate, power demand, or state of health. The environmental state parameters include at least one or more of the following: ambient temperature or external heat transfer boundary conditions.

[0230] Based on the thermal state parameters, electrical state parameters and environmental state parameters, the current thermal management requirement state of the power battery system is identified. The thermal management requirement state includes at least one of the following: high temperature heat dissipation state, temperature difference equalization state, low temperature preheating state, transient power response state or composite regulation state.

[0231] Based on the thermal management requirement status, the thermal management control requirements of the power battery system are decoupled into temperature regulation requirements and flow regulation requirements. The temperature regulation requirements are used to characterize the target supply temperature or heat exchange capacity requirement of the coolant, and the flow regulation requirements are used to characterize the circulation intensity or flow path distribution requirement of the coolant.

[0232] Based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement status, combined with preset control rules, thermal management mapping models, and / or optimized allocation models, the target coolant temperature and target coolant flow rate are determined, and the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate are dynamically adjusted under different thermal management requirement statuses.

[0233] Based on the target coolant temperature and the target coolant flow rate, at least one actuator in the liquid cooling system is linked for control to adjust the coolant circulation flow rate, supply temperature, flow path distribution, and heat exchange capacity.

[0234] Based on the battery temperature feedback, coolant temperature feedback, coolant flow feedback, and battery operating condition change information after the execution control, the target coolant temperature and the target coolant flow are rolled over and corrected. The thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow are updated online to form a closed-loop adaptive collaborative thermal management control based on BMS.

[0235] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0236] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0237] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0238] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, whereby a series of operational steps are performed to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0239] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that, based on the technical solutions disclosed in the present invention, various modifications or variations that can be made by those skilled in the art without creative effort should be included within the scope of protection of the present invention.

Claims

1. A BMS-based coolant flow and temperature collaborative control method, characterized in that, include: The thermal state parameters, electrical state parameters, and environmental state parameters of the power battery system are collected. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack, and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge / discharge current, charge / discharge rate, power demand, or state of health. The environmental state parameters include at least one or more of the following: ambient temperature or external heat transfer boundary conditions. Based on the thermal state parameters, electrical state parameters and environmental state parameters, the current thermal management requirement state of the power battery system is identified. The thermal management requirement state includes at least one of the following: high temperature heat dissipation state, temperature difference equalization state, low temperature preheating state, transient power response state or composite regulation state. Based on the thermal management requirement status, the thermal management control requirements of the power battery system are decoupled into temperature regulation requirements and flow regulation requirements. The temperature regulation requirements are used to characterize the target supply temperature or heat exchange capacity requirement of the coolant, and the flow regulation requirements are used to characterize the circulation intensity or flow path distribution requirement of the coolant. Based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement status, combined with preset control rules, thermal management mapping models, and / or optimized allocation models, the target coolant temperature and target coolant flow rate are determined, and the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate are dynamically adjusted under different thermal management requirement statuses. Based on the target coolant temperature and the target coolant flow rate, at least one actuator in the liquid cooling system is controlled in a coordinated manner to adjust the coolant circulation flow rate, supply temperature, flow path distribution, and heat exchange capacity. Based on the battery temperature feedback, coolant temperature feedback, coolant flow feedback, and battery operating condition change information after the execution control, the target coolant temperature and the target coolant flow are rolled over and corrected. The thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow are updated online to form a closed-loop adaptive collaborative thermal management control based on BMS.

2. The BMS-based coolant flow and temperature coordinated control method according to claim 1, characterized in that, When the thermal management requirement is a high-temperature heat dissipation state, the target coolant temperature is reduced first. When the highest temperature of a single cell and / or the rate of increase of the highest temperature of a single cell exceeds a preset hot spot threshold, the priority of the target coolant flow rate in the coordinated control is increased to suppress the rapid development of local hot spots. When the thermal management requirement is a temperature difference balance state, the target coolant flow rate is increased first and / or the flow path distribution is adjusted to enhance the heat exchange uniformity between different battery modules and limit the rapid drop in the target coolant temperature to avoid local overcooling.

3. The BMS-based coolant flow and temperature coordinated control method according to claim 1, characterized in that, When the thermal management demand state is a low-temperature preheating state, the target coolant temperature is increased first, and the target coolant flow rate is limited to a preset low flow range to reduce heat dispersion caused by excessive circulation flow under low-temperature conditions. When the thermal management demand state is a transient power response state, the target coolant flow rate is fed forward and increased according to the power demand change rate, charge / discharge rate change rate and / or real-time heat generation intensity change rate, and the target coolant temperature is pre-adjusted simultaneously to improve the thermal response speed under high-power transient conditions.

4. The BMS-based coolant flow and temperature coordinated control method according to claim 1, characterized in that, The target coolant temperature and the target coolant flow rate are generated through a state-related coupling mapping relationship corresponding to the thermal management demand state. The state-related coupling mapping relationship includes at least temperature channel weights, flow channel weights, and cross-coupling coefficients to characterize the synergistic influence relationship between the temperature regulation demand and the flow regulation demand.

5. The BMS-based method for coordinated control of coolant flow and temperature according to claim 1, characterized in that, The optimized allocation model takes at least two of the following as optimization objectives: average temperature deviation of battery pack, maximum temperature deviation of individual cells, temperature difference, energy consumption of liquid cooling system, power consumption of cooling or heating, and switching cost of actuator. It also combines the target coolant temperature constraint, the target coolant flow constraint, and the working boundary of actuator to solve the cooperative control quantity. Constraints are set on the rate of change, minimum holding time, and upper and lower limits for the target coolant temperature and the target coolant flow rate to limit abrupt adjustments by the actuator and improve the stability of coordinated control.

6. The BMS-based method for coordinated control of coolant flow and temperature according to claim 1, characterized in that, Based on the response relationship between the highest temperature of the single cell, the average temperature of the battery pack, and the temperature difference before and after the execution of control, and the change in the target coolant temperature and the change in the target coolant flow rate, a temperature-flow coordinated response sensitivity matrix is ​​constructed or updated, and the coordinated control relationship is corrected online based on the temperature-flow coordinated response sensitivity matrix. When the deviation between the coolant flow feedback and the target coolant flow, the deviation between the coolant temperature feedback and the target coolant temperature, or the deviation between the predicted temperature response and the actual temperature response exceeds a preset threshold for multiple consecutive control cycles, a conservative collaborative control mode is triggered. The conservative collaborative control mode includes one or more of the following: increasing the safety margin, narrowing the search range of the control variables, and invoking the fault protection control strategy.

7. A BMS-based coolant flow and temperature coordinated control system, characterized in that, include: The acquisition module is used to collect thermal state parameters, electrical state parameters and environmental state parameters of the power battery system. The thermal state parameters include at least the highest temperature of a single cell, the lowest temperature of a single cell, the average temperature of the battery pack and the temperature difference. The electrical state parameters include at least one or more of the following: state of charge, charge and discharge current, charge and discharge rate, power demand or health state. The environmental state parameters include at least one or more of the following: ambient temperature or external heat exchange boundary conditions. The identification module is used to identify the current thermal management requirement state of the power battery system based on the thermal state parameters, electrical state parameters and environmental state parameters. The thermal management requirement state includes at least one of the following: high temperature heat dissipation state, temperature difference equalization state, low temperature preheating state, transient power response state or composite regulation state. The decoupling module is used to decouple the thermal management control requirements of the power battery system into temperature regulation requirements and flow regulation requirements according to the thermal management requirement status. The temperature regulation requirements are used to characterize the target supply temperature or heat exchange capacity requirements of the coolant, and the flow regulation requirements are used to characterize the circulation intensity or flow path distribution requirements of the coolant. The adjustment module is used to determine the target coolant temperature and target coolant flow rate based on the average temperature of the battery pack, the highest temperature of the individual cells, the temperature difference, the state of charge, the charge / discharge rate, the ambient temperature, and the thermal management requirement state, combined with preset control rules, thermal management mapping model, and / or optimized allocation model, and dynamically adjust the control priority and coupling relationship of the target coolant temperature and the target coolant flow rate under different thermal management requirement states; The linkage control module is used to perform linkage control on at least one actuator in the liquid cooling system based on the target coolant temperature and the target coolant flow rate, so as to adjust the coolant circulation flow rate, supply temperature, flow path distribution and heat exchange capacity. The correction module is used to perform rolling corrections on the target coolant temperature and the target coolant flow rate based on the battery temperature feedback, coolant temperature feedback, coolant flow rate feedback and battery operating condition change information after the execution control, and to update the thermal management demand status and the collaborative control relationship between the target coolant temperature and the target coolant flow rate online, so as to form a closed-loop adaptive collaborative thermal management control based on BMS.

8. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the BMS-based coolant flow and temperature coordinated control method as described in any one of claims 1-6.

9. A computer device, characterized in that, The system includes a memory and a processor, which communicate with each other. The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the BMS-based coolant flow and temperature coordinated control method as described in any one of claims 1-6.

10. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to cause the electronic device to execute instructions for implementing the BMS-based coolant flow and temperature coordinated control method as described in any one of claims 1-6.

Images