CN112818594A - Multi-objective optimization battery pack structure method based on neural network - Google Patents
Multi-objective optimization battery pack structure method based on neural network Download PDFInfo
- Publication number
- CN112818594A CN112818594A CN202110116697.9A CN202110116697A CN112818594A CN 112818594 A CN112818594 A CN 112818594A CN 202110116697 A CN202110116697 A CN 202110116697A CN 112818594 A CN112818594 A CN 112818594A
- Authority
- CN
- China
- Prior art keywords
- battery pack
- neural network
- optimization
- battery
- pack structure
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Physiology (AREA)
- Secondary Cells (AREA)
Abstract
The invention discloses a multi-objective optimization battery pack structure method based on a neural network, which comprises the following steps: s1: converting the three-dimensional model of the battery pack into a two-dimensional model, and then simulating in COMSOL to generate a plurality of groups of structural parameters influencing the performance value of the battery pack; s2, taking a plurality of groups of structural parameters as input of the neural network, taking the performance value of the battery pack as output of the neural network, and training the neural network by adopting a Bayesian regularization algorithm; and S3, performing structural parameter optimization on the neural network trained in the step S2 by adopting an NSGA2 algorithm to obtain multiple groups of pareto optimal solutions, and selecting one group with the lowest highest temperature in the pareto optimal solutions as a final scheme of the battery pack. The invention greatly shortens the period of structure optimization of the battery pack and improves the performance of the battery pack.
Description
Technical Field
The invention relates to the technical field of battery systems, in particular to a multi-objective optimization battery pack structure method based on a neural network.
Background
With the increasing exacerbation of the problems of air pollution and resource shortage, more and more countries begin to vigorously push the development of electric vehicles. The core of the electric automobile is a power battery, and the performance of the power battery affects the quality and the user experience of the electric automobile. The influence of temperature on the performance of the power battery is obvious, and when heavy current is discharged, if the heat generated by the battery is not dissipated in time, the battery has the problems of life attenuation, combustion, even explosion and the like. Therefore, in order to maintain the temperature within the optimal range and improve the endurance mileage of the electric vehicle, a battery pack with a reasonable design is of great significance for improving the performance of the power battery.
At present, three-dimensional battery packs are widely adopted for simulating the performance of the battery packs, but a lot of time is consumed for simulating the three-dimensional battery packs once. When the big data is adopted to train the neural network, a large amount of simulation is needed to obtain the performance data of the battery pack, so that the three-dimensional simulation of the battery pack consumes a large amount of time, and even several months are needed to obtain the trained neural network. Also, previous battery pack optimization only targets the highest temperature, and does not take into account the power consumption required to dissipate the heat from the battery pack. Therefore, the structural arrangement of the battery pack can be unreasonable due to the problems of long process time and poor consideration of factors in the current battery pack structural simulation.
Disclosure of Invention
The invention aims to provide a multi-objective optimization battery pack structure method based on a neural network. The invention greatly shortens the period of structure optimization of the battery pack and improves the performance of the battery pack.
The technical scheme of the invention is as follows: a multi-objective optimization battery pack structure method based on a neural network comprises the following steps:
s1: converting the three-dimensional model of the battery pack into a two-dimensional model, and then simulating in COMSOL to generate a plurality of groups of structural parameters influencing the performance value of the battery pack;
s2, taking a plurality of groups of structural parameters as input of the neural network, taking the performance value of the battery pack as output of the neural network, and training the neural network by adopting a Bayesian regularization algorithm;
and S3, performing structural parameter optimization on the neural network trained in the step S2 by adopting an NSGA2 algorithm to obtain multiple groups of pareto optimal solutions, and selecting one group with the lowest highest temperature in the pareto optimal solutions as a final scheme of the battery pack.
According to the multi-objective optimization battery pack structure method based on the neural network, the structure parameters comprise an inlet position, an outlet position, an inlet width, an outlet width and a battery interval.
In the multi-objective optimization battery pack structure method based on the neural network, the inlet position and the outlet position are adjusted and increased once every 20mm, and the range is [20100] mm; the width of the inlet and the width of the outlet are adjusted and increased once every 5mm, and the range is [1530] mm; the battery distance is adjusted and increased every 0.4mm, and the range is [23.6] mm; there were 2000 battery pack configuration parameter combinations.
In the multi-objective optimization battery pack structure method based on the neural network, 70% of 2000 battery pack structure parameters are used for training the neural network, 15% of the 2000 battery pack structure parameters are used for prediction, and 15% of the 2000 battery pack structure parameters are used for verification.
In the foregoing method for optimizing a battery pack structure based on multiple objectives of a neural network, the battery pack performance values include the maximumHigh temperature, maximum temperature differential, and power consumed within the battery pack for cooling the battery; wherein the power W consumed for cooling the batterypIs represented as follows:
Wp=(Pin-Pout)×Q0;
in the formula: pinIs the pressure at the inlet location; poutIs the pressure at the outlet location; q0Is the inlet air flow rate.
In the multi-objective optimization battery pack structure method based on the neural network, the number of the hidden layers and the number of the output layers of the neural network are respectively set to be 20 and 3.
In the aforementioned multi-objective optimization battery pack structure method based on the neural network, in step S3, during the structure parameter optimization, the genetic algebra is set to be 500, and the population number is set to be 200.
Compared with the prior art, the invention considers that the three-dimensional battery pack needs to consume much time each time of simulation, when the neural network is trained by adopting big data, the performance data of the battery pack can be obtained only by carrying out large-scale simulation, therefore, the three-dimensional simulation of the battery pack can consume much time, even the trained neural network can be obtained for a plurality of months, therefore, the invention converts the battery model into the two-dimensional model to carry out COMSOL simulation, reduces the simulation period, improves the design efficiency, then uses a plurality of groups of structural parameters which influence the performance value of the battery pack and are obtained by the two-dimensional model simulation as the input of the neural network, uses the performance value as the output, trains the neural network by utilizing a Bayesian regularization algorithm aiming at a complex strong nonlinear system of the battery pack, effectively solves the overfitting problem, then optimizes the structural parameters by utilizing an NSGA2 algorithm to obtain a plurality of groups of pareto optimal, the final scheme of the battery pack obtained by the method is tested, the maximum temperature difference is reduced by 67.4%, and the power consumption is reduced by 26%. The performance of the battery pack is greatly improved. In addition, the invention further optimizes the training, input and output step parameters of the neural network and the genetic algebra and population quantity optimized by the structural parameters, and further improves the prediction precision of the neural network.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic three-dimensional model of a battery pack;
FIG. 3 is a schematic diagram of a two-dimensional model of a battery pack;
FIG. 4 is a graph of a maximum temperature performance parameter profile obtained for different battery pack combinations;
FIG. 5 is a graph of maximum temperature difference performance parameter distribution obtained for different battery pack combinations;
FIG. 6 is a graph of power consumption performance parameter distribution obtained for different battery pack combinations;
FIG. 7 is a training set regression graph;
FIG. 8 is a test set regression graph;
FIG. 9 is a neural network regression graph;
FIG. 10 is a pareto optimal solution set diagram;
FIG. 11 is a schematic diagram of the temperature of the batteries in the battery pack before optimization;
fig. 12 is a schematic diagram of the temperature of the batteries in the optimized battery pack.
Detailed Description
The invention is further illustrated by the following figures and examples, which are not to be construed as limiting the invention.
Example (b): a multi-objective optimization battery pack structure method based on a neural network is shown in figure 1 and comprises the following steps:
s1: converting the three-dimensional model of the battery pack into a two-dimensional model, and then simulating in COMSOL to generate a plurality of groups of structural parameters influencing the performance value of the battery pack; because the calculation of the three-dimensional battery pack needs hours to obtain a numerical result, and the calculation of the two-dimensional battery pack only needs minutes, the two-dimensional battery pack can be used for simulating instead of the three-dimensional battery pack so as to simplify the structure of the battery pack, the three-dimensional model of the battery pack is axisymmetric, the small details in the battery pack are omitted, and the schematic diagram of the air-cooled three-dimensional battery pack is shown in fig. 2. The left lower side of the battery pack is provided with an air inlet for cooling air, the right upper side of the battery pack is provided with an air outlet for air, and the air enters the battery pack from the inlet, cools the battery and then flows out of the battery pack from the outlet; the two-dimensional module of the battery packThe pattern is shown in figure 3. Therefore, after the three-dimensional model of the battery pack is converted into the two-dimensional model, simulation is carried out in COMSOL to generate a plurality of groups of structural parameters influencing the performance value of the battery pack, wherein the structural parameters comprise an import position IpoOutlet position JpoInlet width IwiOutlet width JwiAnd the cell spacing d; the battery pack performance values include a maximum temperature, a maximum temperature differential, and a power consumed within the battery pack to cool the battery; in order to ensure safe operation of the battery thermal management system and to improve performance output of the battery pack, the smaller the maximum temperature difference, the maximum temperature and the power consumption within the battery pack, wherein the power W consumed for cooling the battery is betterpIs represented as follows:
Wp=(Pin-Pout)×Q0;
in the formula: pinIs the pressure at the inlet location; poutIs the pressure at the outlet location; q0Is the inlet air flow rate.
The battery pack structure requires multiple changes in optimization and requires the reconstruction of the neural network once the structure changes. Therefore, to improve the simulation efficiency, the model was parameterized in COMSOL, with inlet and outlet positions adjusted every 20mm to increase in the range [20100]]mm; the inlet width and outlet width are adjusted and increased every 5mm, the range is [1530]mm; the cell pitch is increased every 0.4mm, and the range is [23.6]]mm; the total number of the structure parameter combinations of the battery pack is 2000; COMSOL inlet flow of 0.014m3The outlet employs the boundary conditions of free flow, thereby obtaining the maximum temperature, maximum temperature differential, and power consumption performance parameters for the 2000 pack cell package, as shown in fig. 4, 5, and 6, respectively.
S2, taking a plurality of groups of structure parameters as input of a neural network, taking a performance value of a battery pack as output of the neural network, adopting the BP neural network in the embodiment, establishing an approximate functional relation between the structure parameters of the battery pack and the performance value of the battery pack based on the advantages of strong mapping capability, strong self-adaptability, fault-tolerant capability and the like of the BP neural network, wherein 70% of 2000 structure parameters of the battery pack are used for training the neural network, 15% of the 2000 structure parameters of the battery pack are used for prediction, and 15% of the 2000 structure parameters of the battery pack are used for verification; the number of hidden layers and output layers of the neural network is set to 20 and 3 respectively; in the neural network, aiming at a complex strong nonlinear system of a battery pack, a Bayesian regularization algorithm is adopted to train the neural network in order to effectively solve the problem of overfitting; after training, the BP neural network regression graphs shown in fig. 7-9 are obtained, fig. 7 is a schematic diagram describing the regression capability of a training set, fig. 8 is a schematic diagram describing the regression capability of a test set, it can be seen from fig. 7 and 8 that the target value and the output result are basically on the same straight line, the closer to 1, the higher the prediction accuracy is, fig. 9 is a neural network regression graph of the total data, and it can be seen from fig. 9 that the prediction accuracy of the trained lifting network is very high.
And S3, performing structural parameter optimization on the neural network trained in the step S2 by adopting an NSGA2 algorithm, wherein during the structural parameter optimization, the genetic algebra is set to be 500, and the population number is set to be 200, so that an ideal result can be better obtained. Because multiple targets do not exist, multiple targets are optimal simultaneously, but a group of optimal solution sets exist, and the solution sets are called Pareto solution sets. The elements in the pareto solution set are referred to as non-dominant solutions. The Pareto solution set distribution of power consumption, maximum temperature and maximum temperature difference after optimization by the NSGA2 algorithm (i.e., a fast non-dominated multi-objective optimization algorithm with elite retention policy, which is a multi-objective optimization algorithm based on Pareto optimal solution) is shown in fig. 10, the Pareto solution with the minimum maximum temperature among multiple solutions meeting requirements is taken as the final solution after algorithm optimization, and finally, the corresponding battery pack design scheme in the embodiment is: the inlet position is 86.2mm, the outlet position is 76.6mm, the inlet width is 26.4mm, the outlet width is 30.0mm, and the cell spacing is 2.6 mm.
The pre-optimization and post-optimization battery pack design schemes are then simulated in COMSOL to obtain the pre-optimization and post-optimization battery pack performance outputs shown in fig. 11 and 12. As can be seen from fig. 11 and 12, the maximum temperature of the pack before optimization is 320.4K, the maximum temperature difference is 14.1K, and the power consumption is 0.9W, and the maximum temperature of the pack after optimization according to the neural network and the NSGA2 algorithm is 312.9K, the maximum temperature difference is 4.6K, and the power consumption is 0.67W. The maximum temperature of the battery pack optimized by the neural network and the NSGA2 algorithm is reduced by 7.5K, the maximum temperature difference is reduced by 67.4%, the power consumption is reduced by 26%, the three-dimensional model is converted into the two-dimensional model, the design efficiency can be greatly improved, and a large amount of time is saved.
In conclusion, aiming at the unsatisfactory result of the traditional neural network prediction established by a small amount of data obtained by a three-dimensional model, the neural network established by a large amount of battery pack performance data obtained by two-dimensional battery pack simulation can predict the performance of the battery pack more accurately, and the design cycle of the battery pack is improved. The neural network established according to the two-dimensional battery pack model and the NSGA2 algorithm are of great guiding significance for the structural design of the battery pack. The invention solves the contradiction between the highest temperature, the maximum temperature difference and the power consumption of the battery pack, and can obtain the battery pack structure with low highest temperature, small maximum temperature difference and small power consumption. Therefore, the temperature in the battery pack is ensured to be in a proper range, and the endurance mileage of the electric automobile is improved.
Claims (7)
1. A multi-objective optimization battery pack structure method based on a neural network is characterized in that: the method comprises the following steps:
s1: converting the three-dimensional model of the battery pack into a two-dimensional model, and then simulating in COMSOL to generate a plurality of groups of structural parameters influencing the performance value of the battery pack;
s2, taking a plurality of groups of structural parameters as input of the neural network, taking the performance value of the battery pack as output of the neural network, and training the neural network by adopting a Bayesian regularization algorithm;
and S3, performing structural parameter optimization on the neural network trained in the step S2 by adopting an NSGA2 algorithm to obtain multiple groups of pareto optimal solutions, and selecting one group with the lowest highest temperature in the pareto optimal solutions as a final scheme of the battery pack.
2. The multi-objective neural network-based battery pack structure optimization method according to claim 1, wherein: the structural parameters include inlet position, outlet position, inlet width, outlet width, and cell spacing.
3. The multi-objective optimization battery pack structure method based on the neural network as claimed in claim 2, wherein: wherein the inlet position and the outlet position are adjusted and increased once every 20mm, and the range is [20100] mm; the width of the inlet and the width of the outlet are adjusted and increased once every 5mm, and the range is [1530] mm; the battery distance is adjusted and increased every 0.4mm, and the range is [23.6] mm; there were 2000 battery pack configuration parameter combinations.
4. The multi-objective optimization battery pack structure method based on the neural network as claimed in claim 3, wherein: 70% of the 2000 battery pack structure parameters were used for training the neural network, 15% for prediction, and 15% for validation.
5. The multi-objective optimization battery pack structure method based on the neural network as claimed in claim 2, wherein: the battery pack performance values include a maximum temperature, a maximum temperature differential, and a power consumed within the battery pack to cool the battery; wherein the power W consumed for cooling the batterypIs represented as follows:
Wp=(Pin-Pout)×Q0;
in the formula: pinIs the pressure at the inlet location; poutIs the pressure at the outlet location; q0Is the inlet air flow rate.
6. The multi-objective neural network-based battery pack structure optimization method according to claim 1, wherein: the number of hidden layers and output layers of the neural network is set to 20 and 3, respectively.
7. The multi-objective neural network-based battery pack structure optimization method according to claim 1, wherein: in step S3, in the structural parameter optimization, the number of genetic generations is set to 500, and the population number is set to 200.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110116697.9A CN112818594B (en) | 2021-01-28 | 2021-01-28 | Multi-objective battery pack structure optimizing method based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110116697.9A CN112818594B (en) | 2021-01-28 | 2021-01-28 | Multi-objective battery pack structure optimizing method based on neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112818594A true CN112818594A (en) | 2021-05-18 |
CN112818594B CN112818594B (en) | 2023-05-30 |
Family
ID=75859791
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110116697.9A Active CN112818594B (en) | 2021-01-28 | 2021-01-28 | Multi-objective battery pack structure optimizing method based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112818594B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113435016A (en) * | 2021-06-10 | 2021-09-24 | 同济大学 | Multi-objective optimization design method of hybrid thermal management system based on regression model algorithm |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20060110832A (en) * | 2005-04-20 | 2006-10-25 | 가부시키가이샤 덴소 | Method and apparatus for detecting charged state of secondary battery based on neural network calculation |
US20130260357A1 (en) * | 2012-03-27 | 2013-10-03 | Lauren Reinerman-Jones | Skill Screening |
US20160018472A1 (en) * | 2014-07-18 | 2016-01-21 | Samsung Electronics Co., Ltd. | Method and apparatus for estimating state of battery |
CN107145628A (en) * | 2017-03-31 | 2017-09-08 | 中南大学 | The method of prediction lithium battery cycle life based on electrochemical heat coupling model |
CN107301266A (en) * | 2017-05-15 | 2017-10-27 | 中山职业技术学院 | A kind of ferric phosphate lithium cell LOC evaluation methods and system |
US20180026454A1 (en) * | 2016-07-22 | 2018-01-25 | Eos Energy Storage, Llc | Battery Management System |
CN109507598A (en) * | 2017-09-11 | 2019-03-22 | 安徽师范大学 | The lithium battery SOC prediction technique of the LM-BP neural network of Bayesian regularization |
WO2020106894A1 (en) * | 2018-11-20 | 2020-05-28 | Georgia Tech Research Corporation | Systems and methods for enhanced engineering design and optimization |
CN111695301A (en) * | 2020-06-16 | 2020-09-22 | 中国科学院深圳先进技术研究院 | Method and device for predicting battery charge state, storage medium and equipment |
CN112016237A (en) * | 2019-05-13 | 2020-12-01 | 北京航空航天大学 | Deep learning method, device and system for lithium battery life prediction |
US20200394278A1 (en) * | 2019-06-14 | 2020-12-17 | Abraham Varon-Weinryb | Hybrid Finite Element and Artificial Neural Network Method and System for Safety Optimization of Vehicles |
-
2021
- 2021-01-28 CN CN202110116697.9A patent/CN112818594B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20060110832A (en) * | 2005-04-20 | 2006-10-25 | 가부시키가이샤 덴소 | Method and apparatus for detecting charged state of secondary battery based on neural network calculation |
US20130260357A1 (en) * | 2012-03-27 | 2013-10-03 | Lauren Reinerman-Jones | Skill Screening |
US20160018472A1 (en) * | 2014-07-18 | 2016-01-21 | Samsung Electronics Co., Ltd. | Method and apparatus for estimating state of battery |
US20180026454A1 (en) * | 2016-07-22 | 2018-01-25 | Eos Energy Storage, Llc | Battery Management System |
CN109769400A (en) * | 2016-07-22 | 2019-05-17 | Eos能源储存有限责任公司 | Battery management system |
CN107145628A (en) * | 2017-03-31 | 2017-09-08 | 中南大学 | The method of prediction lithium battery cycle life based on electrochemical heat coupling model |
CN107301266A (en) * | 2017-05-15 | 2017-10-27 | 中山职业技术学院 | A kind of ferric phosphate lithium cell LOC evaluation methods and system |
CN109507598A (en) * | 2017-09-11 | 2019-03-22 | 安徽师范大学 | The lithium battery SOC prediction technique of the LM-BP neural network of Bayesian regularization |
WO2020106894A1 (en) * | 2018-11-20 | 2020-05-28 | Georgia Tech Research Corporation | Systems and methods for enhanced engineering design and optimization |
CN112016237A (en) * | 2019-05-13 | 2020-12-01 | 北京航空航天大学 | Deep learning method, device and system for lithium battery life prediction |
US20200394278A1 (en) * | 2019-06-14 | 2020-12-17 | Abraham Varon-Weinryb | Hybrid Finite Element and Artificial Neural Network Method and System for Safety Optimization of Vehicles |
CN111695301A (en) * | 2020-06-16 | 2020-09-22 | 中国科学院深圳先进技术研究院 | Method and device for predicting battery charge state, storage medium and equipment |
Non-Patent Citations (1)
Title |
---|
余懿衡;钱祥忠;杨光辉;夏克刚;张佳瑶;: "基于模糊控制磷酸铁锂电池SOC的准确估计", 电子测量技术 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113435016A (en) * | 2021-06-10 | 2021-09-24 | 同济大学 | Multi-objective optimization design method of hybrid thermal management system based on regression model algorithm |
Also Published As
Publication number | Publication date |
---|---|
CN112818594B (en) | 2023-05-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xu et al. | Multi-objective component sizing based on optimal energy management strategy of fuel cell electric vehicles | |
CN111581850B (en) | Full-period power battery management system applying digital twinning technology | |
CN113022385B (en) | Parameter matching method for fuel cell lithium battery hybrid power system | |
CN111723483B (en) | Green optimal design integrated system and method for power battery | |
CN112319462B (en) | Energy management method for plug-in hybrid electric vehicle | |
Hu et al. | Hierarchical energy management strategy for fuel cell/ultracapacitor/battery hybrid vehicle with life balance control | |
CN112818594A (en) | Multi-objective optimization battery pack structure method based on neural network | |
Yang et al. | Research on the energy management strategy of extended range electric vehicles based on a hybrid energy storage system | |
Zou et al. | A real-time energy management approach with fuel cell and battery competition-synergy control for the fuel cell vehicle | |
CN106503450A (en) | A kind of electric motor car big data analysis optimization System and method for | |
Han et al. | Energy Management in Plug-In Hybrid Electric Vehicles: Preheating the Battery Packs in Low-Temperature Driving Scenarios | |
Shoukat et al. | Application of digital twin in smart battery electric vehicle: Industry 4.0 | |
Guo et al. | Clustered energy management strategy of plug-in hybrid electric logistics vehicle based on Gaussian mixture model and stochastic dynamic programming | |
Zhao et al. | Research on personalized charging strategy of electric bus under time-varying constraints | |
CN113393064A (en) | Method for predicting service life of cadmium-nickel storage battery of motor train unit and terminal equipment | |
Lei et al. | Physics-informed data-driven modeling approach for commuting-oriented hybrid powertrain optimization | |
CN112052987B (en) | Comprehensive energy system optimization planning method and system considering wind power | |
CN112590763A (en) | Planet series-parallel hybrid electric vehicle energy optimization method based on battery heat and aging | |
Gao et al. | Control strategy for PHEB based on actual driving cycle with driving style characteristic | |
Hua et al. | A novel power allocation strategy considering multi-objective comprehensive optimization for hybrid electric vehicles | |
Geng et al. | Simulation and experimental research on energy management control strategy for fuel cell heavy-duty truck | |
Tang et al. | Degradation adaptive energy management strategy for FCHEV based on the Rule-DDPG method: tailored to the current SOH of the powertrain | |
CN115809810A (en) | Equipment transformation model selection method based on low-carbon economy and terminal | |
CN110930029B (en) | Method and system for calculating economy of whole life cycle of shunting locomotive | |
CN112434374A (en) | Capacity matching method and system for composite power supply system of pure electric vehicle |
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 |