CN114156567B - Power battery thermal management system based on machine learning - Google Patents
Power battery thermal management system based on machine learning Download PDFInfo
- Publication number
- CN114156567B CN114156567B CN202111391780.3A CN202111391780A CN114156567B CN 114156567 B CN114156567 B CN 114156567B CN 202111391780 A CN202111391780 A CN 202111391780A CN 114156567 B CN114156567 B CN 114156567B
- Authority
- CN
- China
- Prior art keywords
- temperature
- battery
- heat
- power battery
- battery pack
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 19
- 238000013528 artificial neural network Methods 0.000 claims abstract description 23
- 238000000034 method Methods 0.000 claims abstract description 19
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 15
- 238000005485 electric heating Methods 0.000 claims abstract description 10
- 230000001105 regulatory effect Effects 0.000 claims abstract description 9
- 239000007788 liquid Substances 0.000 claims description 14
- 238000009826 distribution Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 13
- 238000005057 refrigeration Methods 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 9
- 238000001816 cooling Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 7
- 239000000463 material Substances 0.000 description 7
- 238000010438 heat treatment Methods 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000017525 heat dissipation Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 239000011162 core material Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 239000000110 cooling liquid Substances 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 238000011423 initialization method Methods 0.000 description 1
- 229910001416 lithium ion Inorganic materials 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/60—Heating or cooling; Temperature control
- H01M10/63—Control systems
- H01M10/633—Control systems characterised by algorithms, flow charts, software details or the like
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/48—Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte
- H01M10/486—Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte for measuring temperature
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/60—Heating or cooling; Temperature control
- H01M10/62—Heating or cooling; Temperature control specially adapted for specific applications
- H01M10/625—Vehicles
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/60—Heating or cooling; Temperature control
- H01M10/63—Control systems
- H01M10/635—Control systems based on ambient temperature
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Abstract
The invention relates to a power battery thermal management system based on machine learning, which comprises a battery pack, an electronic water pump, an electric heating PTC and a refrigerating heat exchanger which are sequentially connected and form a loop, wherein the battery pack, the electronic water pump, the electric heating PTC and the refrigerating heat exchanger are respectively connected with a controller, the battery pack comprises a battery module, a power battery heat-carrying pipe and a temperature sensor, a machine learning algorithm is integrated in the controller, and the temperature and the flow of a heat carrier required at the current moment can be calculated according to the battery temperature and the battery state, so that the battery temperature is regulated to be kept in a proper range. Compared with the prior art, the invention provides the neural network method for realizing the temperature control of the battery pack, which can automatically learn the experience of manual expert and realize accurate and stable temperature control of the battery pack.
Description
Technical Field
The invention relates to the technical field of intelligent electric automobile power batteries, in particular to a power battery thermal management system based on machine learning.
Background
The power battery is an intelligent electric vehicle energy storage unit which releases electric energy through internal chemical reaction, thereby providing sufficient power for the electric vehicle. The power battery system generally consists of a battery module, a battery management system BMS, a thermal management system, some electrical and mechanical systems, and the like. The safety of intelligent electric automobiles is always valued by the industry. The power lithium ion battery is easy to cause thermal runaway under the conditions of overcharge, needling and collision, so that accidents such as smoking, fire and explosion are caused. Meanwhile, the high temperature influences the performance of the power battery, including the energy density, the service life and other parameters, so that the thermal management system of the power battery is one of the core subsystems of the vehicle-mounted battery.
The power battery on the electric automobile comprises a plurality of power battery single battery cells, and under different driving conditions of the automobile, the single battery cells can generate certain heat when outputting electric energy due to certain internal resistance, so that the temperature of the power battery system becomes high, a large amount of heat is generated in the working process and is accumulated in a narrow battery box body, if the heat cannot be rapidly dissipated in time, the performance and the service life of the battery can be influenced when the temperature of the power battery system exceeds the normal working temperature range, and the service life of the power battery can be influenced even in thermal runaway at high temperature, so that fire explosion and the like are caused.
The thermal management system of the power battery mainly strengthens the heating and heat dissipation capacity of the battery through the temperature sensing and controlling device, ensures that the battery works in a proper temperature range and keeps reasonable temperature distribution in the battery box. Aiming at the characteristic of temperature control of a power battery, the current temperature control method mainly comprises the following steps:
(1) The internal structure and materials of the power battery are optimized, special parts such as the anode, the cathode and the like are made of low-resistance materials, the temperature rise in the working process of the power battery is reduced, but the battery materials are changed, and the battery cost is fully increased when precious material parts are selected, so that the power battery is relatively expensive;
(2) And a material with better heat conductivity is selected outside the battery, and the temperature rise in the working process of the battery is reduced through an original air cooling heat dissipation mode. The natural air cooling heat dissipation form is difficult to control the diffused heat, so that the temperature of the power battery cannot be accurately controlled in a smaller proper range. In addition, the natural air cooling mode cannot enable the temperature of the battery to rise to a proper working range in a cold low-temperature environment.
(3) And a temperature sensor is arranged in the battery, and the liquid temperature of the heat-carrying liquid is controlled by a traditional PID method, so that the battery is stabilized in a proper range. However, the conventional PID method has a large number of super parameters, each group of parameter effects need to be tested continuously, and the optimal solution is difficult to obtain.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a power battery thermal management system based on machine learning.
The aim of the invention can be achieved by the following technical scheme:
the utility model provides a power battery thermal management system based on machine learning, includes battery package, electronic water pump, electrical heating PTC and the refrigeration heat exchanger that connects gradually and constitute the return circuit, battery package, electronic water pump, electrical heating PTC and the refrigeration heat exchanger connect the controller respectively, battery package inside includes battery module, power battery heat carrier tube and temperature sensor, the inside integrated machine learning algorithm that is used for calculating the temperature and the flow of the required heat carrier of current moment of having of controller.
The power battery heat-carrying pipe comprises a heat-carrying pipe inlet, a heat-carrying pipe outlet and heat-carrying branch pipes, wherein the heat-carrying pipe inlet and the heat-carrying pipe outlet are positioned on one side, the whole flow passage is provided with two primary branches which are respectively used for cooling modules and electric devices on two sides, each primary branch is provided with a plurality of heat-carrying branch pipes, each heat-carrying branch pipe is internally provided with a self-adaptive flow regulating valve, and when the temperature of the power battery is lower than a lower threshold limit or higher than an upper threshold limit, the self-adaptive flow regulating valve automatically expands.
The temperature sensor is arranged on the surface of each battery module inside the battery pack.
Further, five temperature sensors are respectively adhered to the upper surface and the lower surface of each battery module inside the battery pack, and are respectively arranged at the four vertex angles and the center position of each battery module, and the temperature of each battery module is estimated through a temperature sensor fusion algorithm.
The expression of the temperature estimation of each battery module through the temperature sensor fusion algorithm is as follows:
wherein: t is t f The temperature value is the fused temperature value; t is t k For the measurement value of the kth temperature sensor, n is the total number of temperature sensors; omega k Is the weight;is a standard normal distribution density function; mu is the average value of the measured temperature; sigma is the standard deviation of the measured temperature.
The temperature of each battery module is represented by constructing an mxn temperature distribution grid pattern in the plane of the battery pack, and the temperature in each grid of the temperature distribution grid pattern is obtained by a linear interpolation method. In each grid, the temperature value and the state of the power battery are configured into a characteristic vector with fixed length, and the state of the power battery comprises the voltage, the total current, the battery SOC and the working mode of each battery module.
The temperature value of each grid and the state of the power battery are used as input, and the temperature and the flow required by the heat carrier in the heat carrier pipe are output through the neural network, so that the refrigeration heat exchanger, the electric heating PTC and the electronic water pump are controlled.
The neural network comprises a multi-layer perceptron network and a convolution network, wherein the multi-layer perceptron network consists of two fully connected layers, high-dimensional characteristics with the length of C in each grid are deeply extracted, and the distribution structure of the high-dimensional characteristics in M multiplied by N grids is a characteristic graph C multiplied by M multiplied by N; the convolution network comprises a convolution layer and a full connection layer, the convolution layer extracts the high-dimensional characteristics of the characteristic map of C multiplied by M multiplied by N, and the obtained characteristic map is unfolded into characteristic vectors, and then the full connection layer is used for regressing two parameters of temperature and flow.
Before the neural network is used, parameters of the neural network are obtained through training, and training data are obtained after the temperature and flow of the heat carrier are actually regulated by a manual expert.
Compared with the prior art, the power battery thermal management system based on machine learning provided by the invention at least has the following beneficial effects:
1) The invention does not need to change the parts and core materials of the power battery system, and can select a low-cost power battery system according to actual conditions, thereby reducing the cost of the whole power battery system.
2) According to the invention, based on the machine learning method, the manual expert parameter adjustment experience is automatically learned, when the power battery thermal management system is determined, the manual expert is only required to manually adjust for a period of time, the system can automatically learn the expert experience, and the temperature control of the power battery is completed, so that the method is simple, convenient and practical.
3) According to the invention, manual adjustment of control parameters is not needed, all parameters of the system are automatically corrected by back propagation according to expert experience, so that the burden of a control person is reduced, the development efficiency of the power battery thermal management system is improved, and the development period of the thermal management system is shortened.
Drawings
FIG. 1 is a schematic diagram of a power battery thermal management system based on machine learning in an embodiment;
FIG. 2 is a schematic diagram of a heat carrier tube in a power battery according to an embodiment;
FIG. 3 is a graph showing an internal temperature grid map of a power cell according to an embodiment;
FIG. 4 is a schematic diagram of a temperature control method of a power battery thermal management system based on machine learning according to an embodiment;
FIG. 5 is a schematic diagram of a neural network model employed in the embodiments;
fig. 6 is a schematic diagram of neural network training set fabrication in an embodiment.
Detailed Description
The invention will now be described in detail with reference to the drawings and specific examples. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Examples
The invention relates to a power battery thermal management system based on machine learning, which is characterized in that corresponding parts are arranged and installed at proper positions on an intelligent electric automobile, and the parts are connected through a heat carrier pipe to form a loop.
As shown in fig. 1, the system includes a battery pack, an electronic water pump, an electrically heated PTC, a refrigeration heat exchanger, and a controller. The battery pack, the electronic water pump, the electric heating PTC and the refrigeration heat exchanger are sequentially connected, and the output end of the refrigeration heat exchanger is connected with the battery pack. The battery pack, the electronic water pump, the electric heating PTC and the refrigeration heat exchanger are respectively connected with the controller. The battery pack comprises a module, a power battery heat-carrying pipe and a temperature sensor. The controller is internally integrated with a machine learning algorithm, and the temperature and the flow of the heat carrier required at the current moment can be calculated according to the battery temperature and the battery state, so that the battery temperature is regulated and kept in a proper range. The machine learning algorithm adopts a neural network, and training data of the neural network is manufactured by actual regulation data of a human expert.
In the power battery thermal management system, the electric heating PTC is mainly used for heating the battery pack, so that the temperature of the battery pack is increased; the refrigeration heat exchanger mainly takes away heat in the power battery thermal management loop by utilizing cooling liquid in the air conditioner cooling loop, so that the heat carrier liquid is cooled, and the temperature of the battery pack is further reduced; the electronic water pump mainly enables heat-carrying liquid in the heat-carrying pipe inside the battery pack to flow, and heat transfer is quickened.
The power battery heat-carrying pipe comprises a heat-carrying pipe inlet, a heat-carrying pipe outlet and a heat-carrying branch pipe. As shown in fig. 2, the specific structure of the heat carrier tube in the battery pack is symmetrical. The water inlet and the water outlet are positioned at one side, the whole flow channel is divided into two branches (primary branches) firstly, and the modules and the electric devices at the two sides are cooled respectively. Each branch is divided into a plurality of heat-carrying branch pipes (secondary branches). The heat carrying branch pipe adjusts the flow resistance of the flow channel at the bottom of each module through the self-adaptive flow adjusting valve, when the temperature is higher than the upper limit threshold or lower than the lower limit threshold, the internal channel of the throttle valve expands to increase the flow, so that the local temperature of the battery can be quickly changed. The self-adaptive flow regulating valve can be divided into two sections, one section is composed of a thermal expansion material, the other section is composed of a cold expansion material, the two sections are spliced with each other, the front-back relation of the positions is unlimited, and the flow of the heat-carrying liquid can be self-adaptively regulated between the branch pipes according to the temperature of the nearby modules. The heat-carrying pipe is mainly designed by taking the temperature of the battery cells of the soft package module into consideration, wherein the temperature of the battery cells is gradually decreased from the middle to the two sides, and the temperature of the battery cells in the middle is highest, so that the heat-carrying pipe is designed to carry heat-carrying liquid to flow in from the middle, and the temperature difference of the battery cells in the module can be reduced to the greatest extent.
As shown in fig. 3, in order to correctly detect the temperature in the battery pack, the temperature in the battery pack is controlled to be maintained in a fixed range in real time, and temperature sensors are attached to the surfaces of the battery cells of each module, and are respectively located at the four top corners and the center of the module. As a preferable scheme, the temperature sensors are closely attached to the surface of a single module of the battery pack, namely, 5 temperature sensors are respectively attached to the upper surface and the lower surface of each module and are respectively positioned at the four vertex angles and the center.
Considering the failure and error of the temperature sensor, the temperature of each module is estimated by a sensor fusion algorithm, and the formula is as follows:
wherein: t is t f The temperature value is the fused temperature value; t is t k Measurement for the kth temperature sensorThe magnitude, n, is the total number of temperature sensors; omega k Is the weight;is a standard normal distribution density function; mu is the average value of the measured temperature; sigma is the standard deviation of the measured temperature.
The temperature values of the modules are subjected to a linear interpolation method to generate a power battery temperature distribution grid diagram, namely: the plane of the battery pack is constructed with an M x N temperature distribution grid pattern, and the temperature in each grid is calculated from the temperature of each module by a linear interpolation method. In the grid, the temperature value and the state of the power battery are configured as a fixed length feature vector. The state of the power battery includes the voltage, total current, battery SOC, and operating mode of each module. In order to expand the characteristic information in each grid, after fitting is carried out by adopting a cubic equation through historical frame information, the current temperature change speed and the temperature change acceleration can be calculated, and the battery pack temperature at the future moment is further predicted through an equal acceleration temperature change model. Each grid is composed of battery temperature and battery state parameters, and is thus constructed as an input to the neural network algorithm.
The interpolation algorithm formula is:
wherein: t is the temperature value of the current grid; t is t i And t j Is an intermediate variable; t is t a 、t b 、t c And t d Temperature values of adjacent sensors at the upper left corner, the upper right corner, the lower left corner and the lower right corner respectively; (x, y) is the current grid coordinates; (x) 1 ，y 1 ) Sensor coordinates for the upper left corner; (x) 2 ，y 2 ) Is the lower right corner sensor coordinate.
As shown in fig. 4, the invention mainly adopts a neural network method to complete the temperature control of the power battery, the neural network inputs the battery temperature and the battery state parameters, and the neural network outputs the temperature of the heat carrier and the flow of the heat carrier. By constructing the input and output training samples, the neural network automatically adjusts the network weight, so that the network can automatically complete the temperature control of the power battery. The neural network is integrated in the controller, obtains the temperature and flow required by the heat carrier in the heat carrier pipe according to the input quantity, and further controls the refrigeration heat exchanger, the electric heating PTC and the electronic water pump, thereby completing the temperature control of the power battery pack.
The structure schematic diagram of the neural network model is shown in fig. 5, and the neural network mainly comprises a multi-layer perceptron network and a convolution network, wherein the multi-layer perceptron network mainly comprises 2 full-connection layers, and the high-dimensional characteristics with the length of C in each grid are deeply extracted. The distribution of high-dimensional features in an mxn grid can be constructed as a feature map c×m×n. The convolution network mainly comprises a convolution layer and a full connection layer, the convolution layer mainly extracts the high-dimensional characteristics of the characteristic images of C multiplied by M multiplied by N, the obtained characteristic images are unfolded into characteristic vectors, and two parameters of regression temperature and flow of the full connection layer are used.
Preferably, the convolution network may be a mainstream image classification network, and the convolution part may be a network structure such as Resnet50, resnet101, FPN, and the like, and use pre-training weights of the network. The full connection layer of the convolutional network can initialize parameters by adopting a Kaiming Initialization method, thereby being beneficial to the rapid convergence of the whole network.
Based on the neural network model, firstly, feature vectors in each temperature grid graph are constructed, wherein the feature vectors comprise temperature related features (current temperature value, temperature values of previous frames, current temperature change speed, temperature change acceleration and temperature values of next frames) and battery state features (voltage, total current, battery SOC and working mode of each module), and more feature information is more beneficial to the neural network learning. The network maps the input to a high dimensional space to accomplish the corresponding temperature control task. The input feature vector is firstly subjected to feature extraction on the features in each grating through a multi-layer perceptron network MLP, and after the high-dimensional features are extracted, the multi-channel pseudo image is constructed, so that a convolutional neural network CNN can be adopted to further regress to obtain temperature and flow values. In the training process, a smoothL1 loss function is adopted, and the formula is as follows:
l total ＝SmoothL1(T pred -T gt )+SmoothL1(Q pred -Q gt )
wherein: t (T) pred The temperature of the heat carrier liquid is output by a network; t (T) gt The temperature required by the actual heat carrier liquid; q (Q) pred The flow of the heat carrier liquid is output by a network; q (Q) gt Is the flow required by the actual heat carrier liquid.
As shown in fig. 6, in the process of producing the neural network training set data, the current actual battery temperature and battery state parameters are actually displayed and recorded, and the flow and temperature data of the heat carrier in the heat carrier tube are adjusted according to experience by an artificial expert through observing the actual condition of the power battery, so that the flow and temperature data of the heat carrier are also stored in real time. Thereby creating the data set required for training by means of the manual expert experience data.
The method combines the advantages of self-learning of the neural network, constructs the training set required by the neural network in a simple mode, can be completed by an experienced artificial expert in the process of acquiring the training set data, can obtain a network model of related experience only by manually adjusting a plurality of groups of data, and has the advantages of simplicity, practicability and the like.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions may be made without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (3)
1. The power battery thermal management system based on machine learning is characterized by comprising a battery pack, an electronic water pump, an electric heating PTC and a refrigerating heat exchanger which are sequentially connected and form a loop, wherein the battery pack, the electronic water pump, the electric heating PTC and the refrigerating heat exchanger are respectively connected with a controller, the battery pack comprises a battery module, a power battery heat carrying pipe and a temperature sensor, and a machine learning algorithm for calculating the temperature and the flow of a heat carrier required at the current moment is integrated in the controller;
the temperature sensor is arranged on the surface of each battery module inside the battery pack; five temperature sensors are respectively stuck to the upper surface and the lower surface of each battery module in the battery pack, and are respectively arranged at the four vertex angles and the center position of each battery module, and the temperature of each battery module is estimated through a temperature sensor fusion algorithm;
the expression of the temperature estimation of each battery module through the temperature sensor fusion algorithm is as follows:
，
，
，
，
wherein:t f the temperature value is the fused temperature value;t k is the firstkThe measured value of the individual temperature sensor(s),nis the total number of temperature sensors;ω k for the rightWeighing;is a standard normal distribution density function;μfor measuring the temperature average;σfor measuring the standard deviation of temperature;
the temperature of each battery module is represented by constructing an M x N temperature distribution grid graph on the plane of the battery pack, and the temperature in each grid of the temperature distribution grid graph is obtained by a linear interpolation method; in each grid, the temperature value and the state of the power battery are configured into a characteristic vector with fixed length, and the state of the power battery comprises the voltage, the total current, the battery SOC and the working mode of each battery module; the temperature value of each grid and the state of the power battery are used as input, and the temperature and the flow required by the heat carrier in the heat carrier pipe are output through a neural network, so that the refrigeration heat exchanger, the electric heating PTC and the electronic water pump are controlled;
the neural network comprises a multi-layer perceptron network and a convolution network, wherein the multi-layer perceptron network consists of two fully connected layers, high-dimensional characteristics with the length of C in each grid are deeply extracted, and the distribution structure of the high-dimensional characteristics in M multiplied by N grids is a characteristic graph C multiplied by M multiplied by N; the convolution network comprises a convolution layer and a full connection layer, the convolution layer carries out high-dimensional feature extraction on the feature map of C multiplied by M multiplied by N, and after the obtained feature map is unfolded into feature vectors, two parameters of regression temperature and flow of the full connection layer are used;
the neural network model adopts SmoothL1 as a loss function in the training process:
，
，
wherein:T pred the temperature of the heat carrier liquid is output by a network;T gt the temperature required by the actual heat carrier liquid;Q pred the flow of the heat carrier liquid is output by a network;Q gt is the flow required by the actual heat carrier liquid.
2. The machine learning-based power battery thermal management system according to claim 1, wherein the power battery heat-carrying pipe comprises a heat-carrying pipe inlet, a heat-carrying pipe outlet and a heat-carrying branch pipe, the heat-carrying pipe inlet and the heat-carrying pipe outlet are positioned at one side, the whole flow passage is provided with two primary branches for cooling modules and electric devices at two sides respectively, each primary branch is provided with a plurality of heat-carrying branch pipes, each heat-carrying branch is internally provided with a self-adaptive flow regulating valve, and when the temperature of the power battery is lower than a lower threshold limit or higher than an upper threshold limit, the self-adaptive flow regulating valve automatically expands.
3. The machine learning based power cell thermal management system of claim 1, wherein the neural network is trained to obtain parameters of the neural network prior to use, and wherein training data is obtained after the actual adjustment of heat carrier temperature and flow by human expert.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111391780.3A CN114156567B (en) | 2021-11-19 | 2021-11-19 | Power battery thermal management system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111391780.3A CN114156567B (en) | 2021-11-19 | 2021-11-19 | Power battery thermal management system based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114156567A CN114156567A (en) | 2022-03-08 |
| CN114156567B true CN114156567B (en) | 2024-03-26 |
Family
ID=80457406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111391780.3A Active CN114156567B (en) | 2021-11-19 | 2021-11-19 | Power battery thermal management system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114156567B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116387692B (en) * | 2023-05-31 | 2023-08-18 | 宁波齐云新材料技术有限公司 | Lithium battery pack intelligent heat dissipation control system and control method thereof |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109193075A (en) * | 2018-09-28 | 2019-01-11 | 合肥工业大学 | Power battery of pure electric automobile method for controlling cooling system based on intensified learning |
| CN110146823A (en) * | 2019-06-06 | 2019-08-20 | 重庆大学 | A kind of series-connected cell group SOC estimation method based on information fusion |
| CN111079905A (en) * | 2019-12-27 | 2020-04-28 | 北京迈格威科技有限公司 | Convolutional neural network processing method, device and electronic system |
| CN211265669U (en) * | 2019-12-25 | 2020-08-14 | 合肥国轩高科动力能源有限公司 | Full-parallel power battery cooling system |
| CN111572407A (en) * | 2020-04-29 | 2020-08-25 | 辽宁工业大学 | Hybrid electric vehicle thermal management system and control method thereof |
| CN113300027A (en) * | 2021-04-19 | 2021-08-24 | 江苏大学 | Power battery thermal management system based on predictive control and control method thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11850970B2 (en) * | 2019-08-18 | 2023-12-26 | Board Of Regents, The University Of Texas System | J-type air-cooled battery thermal management system and method |
| US11709080B2 (en) * | 2019-12-26 | 2023-07-25 | Horiba Stec, Co., Ltd. | Flow rate estimation for mass flow controller |
-
2021
- 2021-11-19 CN CN202111391780.3A patent/CN114156567B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109193075A (en) * | 2018-09-28 | 2019-01-11 | 合肥工业大学 | Power battery of pure electric automobile method for controlling cooling system based on intensified learning |
| CN110146823A (en) * | 2019-06-06 | 2019-08-20 | 重庆大学 | A kind of series-connected cell group SOC estimation method based on information fusion |
| CN211265669U (en) * | 2019-12-25 | 2020-08-14 | 合肥国轩高科动力能源有限公司 | Full-parallel power battery cooling system |
| CN111079905A (en) * | 2019-12-27 | 2020-04-28 | 北京迈格威科技有限公司 | Convolutional neural network processing method, device and electronic system |
| CN111572407A (en) * | 2020-04-29 | 2020-08-25 | 辽宁工业大学 | Hybrid electric vehicle thermal management system and control method thereof |
| CN113300027A (en) * | 2021-04-19 | 2021-08-24 | 江苏大学 | Power battery thermal management system based on predictive control and control method thereof |
Non-Patent Citations (3)
| Title |
|---|
| 刘国良，杨成慧主编；白旭灿，庄淑君，邵麦顿副主编.《MATLAB程序设计基础教程》.西安：西安电子科技大学出版社,2012,235. * |
| 周浩敏，钱政编著.《智能传感技术与系统》.北京：北京航空航天大学出版社,2008,44-50. * |
| 基于实时出行需求和交通路况的电动汽车充电负荷预测;吴钉捷;李晓露;;电力建设;20200801(08);61-71 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114156567A (en) | 2022-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11522487B2 (en) | Building and building cluster energy management and optimization system and method | |
| CN111261973B (en) | Electric automobile whole battery thermal management method based on model predictive control | |
| CN109962318B (en) | Battery pack cooling liquid flow control system and control method based on dual fuzzy control | |
| CN113300027B (en) | Power battery thermal management system based on predictive control and control method thereof | |
| CN108172930B (en) | Battery pack cooling control method and device and battery pack | |
| US20120048953A1 (en) | Temperature adjusting device and an intelligent temperature control method for a sand and dust environment testing system | |
| CN114156567B (en) | Power battery thermal management system based on machine learning | |
| CN105270186B (en) | The battery thermokinetics of depression of order for control models | |
| CN103682523B (en) | For the system and method for the cooling or the heating that control battery | |
| US20140174707A1 (en) | Method and system for thermal storage in a vehicle | |
| CN109271700B (en) | Battery thermal management method and system based on deep learning multilayer network modeling | |
| CN105742736B (en) | The heat management experimental rig and method of a kind of electric automobile power battery | |
| KR20170124884A (en) | Temperature control apparatus and method for energy storage system | |
| CN114335625B (en) | Fuel cell stack temperature control method, device, model predictive controller and system | |
| CN106960719A (en) | Transformer Cooling control method and apply its transformer cooling system | |
| CN112092676A (en) | Estimation and correction method for battery pack temperature field by using virtual temperature sensor | |
| CN109301396A (en) | Battery heat preservation system and its control method under a kind of low-temperature environment of electric vehicle | |
| CN111572407B (en) | Hybrid electric vehicle thermal management system and control method thereof | |
| CN108417906A (en) | Battery management system with Intelligent heat management system control function | |
| CN115732810A (en) | Control method of electric vehicle battery pack heating system | |
| Zhao et al. | A two-stage eco-cooling control strategy for electric vehicle thermal management system considering multi-source information fusion | |
| CN113715691B (en) | New energy automobile battery box temperature control system | |
| FR3097471A1 (en) | EQUIVALENT TEMPERATURE REGULATION PROCESS IN A COCKPIT OF A MEANS OF TRANSPORT EQUIPPED WITH AN AIR CONDITIONING SYSTEM | |
| Hung et al. | Development of a thermal management system for energy sources of an electric vehicle | |
| CN109599612B (en) | Temperature adjusting method and temperature adjusting system for vehicle-mounted battery |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |