WO2022258079A1 - 基于深度学习的电传动推土机电池热管理控制方法及系统 - Google Patents

基于深度学习的电传动推土机电池热管理控制方法及系统 Download PDF

Info

Publication number
WO2022258079A1
WO2022258079A1 PCT/CN2022/106154 CN2022106154W WO2022258079A1 WO 2022258079 A1 WO2022258079 A1 WO 2022258079A1 CN 2022106154 W CN2022106154 W CN 2022106154W WO 2022258079 A1 WO2022258079 A1 WO 2022258079A1
Authority
WO
WIPO (PCT)
Prior art keywords
load
battery
thermal management
electric drive
micro
Prior art date
Application number
PCT/CN2022/106154
Other languages
English (en)
French (fr)
Inventor
闫伟
李国祥
梅娜
胡滨
万庆江
李荣忠
刘荫荫
纪嘉树
李嘉颀
Original Assignee
山东大学
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 山东大学 filed Critical 山东大学
Priority to US18/014,448 priority Critical patent/US20230273593A1/en
Publication of WO2022258079A1 publication Critical patent/WO2022258079A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
    • G05B19/4155—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by programme execution, i.e. part programme or machine function execution, e.g. selection of a programme
    • E—FIXED CONSTRUCTIONS
    • E02—HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
    • E02F—DREDGING; SOIL-SHIFTING
    • E02F9/00—Component parts of dredgers or soil-shifting machines, not restricted to one of the kinds covered by groups E02F3/00 - E02F7/00
    • E02F9/20—Drives; Control devices
    • E02F9/2058—Electric or electro-mechanical or mechanical control devices of vehicle sub-units
    • E02F9/2091—Control of energy storage means for electrical energy, e.g. battery or capacitors
    • 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
    • E—FIXED CONSTRUCTIONS
    • E02—HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
    • E02F—DREDGING; SOIL-SHIFTING
    • E02F3/00—Dredgers; Soil-shifting machines
    • E02F3/04—Dredgers; Soil-shifting machines mechanically-driven
    • E02F3/76—Graders, bulldozers, or the like with scraper plates or ploughshare-like elements; Levelling scarifying devices
    • E02F3/80—Component parts
    • E02F3/84—Drives or control devices therefor, e.g. hydraulic drive systems
    • E—FIXED CONSTRUCTIONS
    • E02—HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
    • E02F—DREDGING; SOIL-SHIFTING
    • E02F9/00—Component parts of dredgers or soil-shifting machines, not restricted to one of the kinds covered by groups E02F3/00 - E02F7/00
    • E02F9/20—Drives; Control devices
    • E02F9/2058—Electric or electro-mechanical or mechanical control devices of vehicle sub-units
    • E02F9/2062—Control of propulsion units
    • E02F9/207—Control of propulsion units of the type electric propulsion units, e.g. electric motors or generators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/29—Graphical models, e.g. Bayesian networks
    • G06F18/295—Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/50—Machine tool, machine tool null till machine tool work handling
    • G05B2219/50333—Temperature
    • 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
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/60—Other road transportation technologies with climate change mitigation effect
    • Y02T10/70—Energy storage systems for electromobility, e.g. batteries

Definitions

  • the invention belongs to the field of battery thermal management control, and in particular relates to a method and system for battery thermal management control of an electric drive bulldozer based on deep learning.
  • the electric drive bulldozer has the advantage of zero emissions, and its application is becoming more and more widely.
  • the normal operation of the power battery is the basis for ensuring its safe operation, and it is also the core of the power system of the electric drive bulldozer, and a complete and efficient battery thermal management system is to ensure that the power battery works in a suitable temperature range key.
  • the temperature is too low, the ion activity of the electrolyte inside the battery will decrease, and the battery discharge rate will decrease, which will seriously affect the power performance of the vehicle; if the temperature is too high, it will cause adverse effects on battery polarization, and even cause fire and explosion. security incident.
  • the electric compressor consumes more power as the speed increases.
  • the control strategy of the battery thermal management system of an electric drive bulldozer generally sets the compressor speed to several gears, which cannot meet the needs of specific working conditions. Therefore, how to match the reasonable speed of the electric compressor in real time according to the heat dissipation requirements of the battery pack has great application value for the effective work of the battery thermal management system of the electric drive bulldozer and energy saving and emission reduction.
  • the present invention provides a battery thermal management control method and system for electric drive bulldozers based on deep learning.
  • the control strategy combined with the first-order Markov chain model to construct a weighted working condition that can more accurately reflect the actual heat dissipation demand of the battery, improve the control strategy of the battery thermal management system, so that the system can meet the cooling demand of the battery at the same time, the maximum Minimize energy consumption.
  • the first aspect of the present invention provides a deep learning-based battery thermal management control method for an electric drive bulldozer.
  • a battery thermal management control method for an electric drive bulldozer based on deep learning including:
  • the real-time battery outlet water temperature, passenger compartment temperature, passenger compartment target temperature and ambient temperature, as well as the motor speed, motor torque and battery state of charge under weighted working conditions are used as the input of the electric compressor speed prediction model to predict the electric compressor Speed, in order to obtain the thermal management control strategy of electric drive bulldozer battery.
  • the second aspect of the present invention provides a battery thermal management control system for an electric drive bulldozer based on deep learning.
  • a battery thermal management control system for an electric drive bulldozer based on deep learning including:
  • a micro-load fragment prediction module which is used to obtain the micro-load fragment of the current load spectrum, and use the Markov chain model to predict the next micro-load fragment;
  • a weighted working condition parameter calculation module which is used to calculate the motor speed, motor torque and battery state of charge under the weighted working condition based on the weight of the current micro-load segment and the predicted micro-load segment;
  • Electric compressor speed prediction module which is used to use real-time battery outlet water temperature, passenger compartment temperature, passenger compartment target temperature and ambient temperature, and motor speed, motor torque and battery state of charge under weighted conditions as electric compressor speed prediction
  • the input quantity of the model predicts the speed of the electric compressor to obtain the thermal management control strategy of the electric drive bulldozer battery.
  • a third aspect of the present invention provides a computer readable storage medium.
  • a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps in the above-mentioned deep learning-based battery thermal management control method for an electric drive bulldozer are implemented.
  • a fourth aspect of the present invention provides a computer device.
  • a computer device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the program, the above-mentioned deep learning-based electric drive bulldozer battery thermal Steps in a management control method.
  • the present invention provides a battery thermal management control method for electric drive bulldozers based on deep learning.
  • the target temperature of the passenger compartment is the training sample
  • the support vector machine algorithm in deep learning is used to train the sample data to obtain the compressor speed prediction model
  • the Markov chain model is used to predict the next micro-load segment according to the current load spectrum.
  • Forecasting using a specific formula to calculate the weight of the existing segment and the predicted segment, introducing the weighted load spectrum into the prediction model to obtain the speed of the electric compressor, forming a battery thermal management control strategy, combining the evaporator, condenser, and battery Micro Channel heat exchangers, electronic expansion valves and other components form a battery thermal management system for electric drive bulldozers.
  • It uses the self-defined objective weighting method formula to build a weighted working condition that can more accurately reflect the actual heat dissipation demand of the battery on the basis of the Markov chain model for the control of the battery thermal management system of the electric drive bulldozer, so that the compressor speed prediction model more precise.
  • the present invention uses the Markov chain model to predict future working conditions and weight them with the current working conditions, which can more accurately reflect the actual heat dissipation requirements of the battery.
  • the battery thermal management control strategy of the electric drive bulldozer is thus combined with the improved support vector
  • the computer algorithm and the first-order Markov chain model predict the compressor speed of the battery thermal management system of the electric drive bulldozer, and the error is smaller. While effectively improving the heat dissipation effect, the energy consumption can be minimized, which provides a basis for the control strategy of the thermal management system. Research and development provides a basis and is of great significance to energy conservation and emission reduction.
  • Fig. 1 is a schematic flow chart of a battery thermal management control method for an electric drive bulldozer based on deep learning according to an embodiment of the present invention
  • Fig. 2 is the training result of the support vector machine model of the compressor rotational speed control strategy according to the embodiment of the present invention.
  • this embodiment provides a deep learning-based battery thermal management control method for an electric drive bulldozer, which specifically includes the following steps:
  • Step S101 Obtain the micro-load segment of the current load spectrum, and use the Markov chain model to predict the next micro-load segment.
  • the Markov chain model before using the Markov chain model to predict the next micro-load segment, it also includes:
  • the Markov chain model is used to predict the load spectrum working conditions of electric drive bulldozers, and the probability transition matrix of various working conditions is established based on the existing data.
  • the state at time t+1 The load is S t+1
  • P ij is the transition probability that the current load s i reaches the next state load s j
  • N ij is the number of events that the current load s i reaches the next state load s j
  • the working condition with the maximum transition probability is selected as the prediction result of the next state load.
  • Step S102 Based on the weights of the current micro-load segment and the predicted micro-load segment, calculate the motor speed, motor torque and battery state of charge under weighted conditions.
  • the correlation coefficient between time t and time load weight is -1, and the objective weight is determined according to the difference between the load spectra of the two fragments.
  • the greater the difference the more information provided at time t+1, which can play a role in the comprehensive evaluation.
  • the weight at time t+1 is:
  • x t is the load of the electric drive bulldozer at time t
  • x t+1 is the load of the electric drive bulldozer at time t+1
  • W t+1 is the weight at time t+1
  • the correlation coefficient between time t and load weight at time t+1 is -1
  • the weighted load is x t ⁇ (1-W t+1 )+x t+1 ⁇ W t+1
  • ⁇ , ⁇ , ⁇ , and M are all constants.
  • Step S103 Taking the real-time battery outlet water temperature, passenger compartment temperature, target temperature of the passenger compartment and ambient temperature, motor speed, motor torque and battery state of charge under weighted working conditions as the input of the electric compressor speed prediction model, predicting Electric compressor speed to get thermal management control strategy for electric drive bulldozer battery.
  • the rotation speed prediction model of the electric compressor is obtained through training and learning by using a support vector machine algorithm improved by a double population genetic algorithm.
  • the double-population genetic algorithm divides the population into two sub-populations, among which the population with low fitness is subjected to adaptive Cauchy mutation, and the population with high fitness is subjected to adaptive Gaussian mutation to complete the optimization process.
  • Individual i is updated as:
  • x i and x i ' are respectively the i-th chromosome before and after mutation
  • range is the range of individual movement
  • N i (0,1) is a Gaussian distribution random number
  • C i (0,1) is a Cauchy distribution random number.
  • the improved support vector machine algorithm based on the dual population adaptive genetic algorithm randomly generates the penalty factor and path by setting the initial parameters such as the number of populations, the maximum number of iterations, the crossover probability, and the generation gap.
  • the combination of variance parameters of the radial basis kernel function is used as the initial population, and each generation undergoes selection, crossover and double population adaptive mutation operations to find the combination of the penalty factor and the variance parameter of the radial basis kernel function that minimizes the error of the support vector machine prediction model.
  • the training samples of the electric compressor speed prediction model are obtained through the joint calculation of one-dimensional and three-dimensional thermal management software simulation.
  • the battery thermal management subsystem set different ambient temperature, vehicle speed, battery heat dissipation (equivalent to motor speed, motor torque and battery output power), passenger compartment temperature, and target temperature of the passenger compartment as the input of training samples, and the simulation results are
  • the compressor speed and corresponding duty cycle that meet the battery safety temperature requirements are used as sample output, and the above data are used as training samples for the support vector machine prediction model.
  • the support vector machine algorithm improved by the double population genetic algorithm is trained and learned, and the prediction model of the compressor speed under different working conditions is obtained, and the control strategy of the battery thermal management system of the electric drive bulldozer is formed.
  • This embodiment provides a battery thermal management control system for an electric drive bulldozer based on deep learning, which specifically includes the following modules:
  • a micro-load fragment prediction module which is used to obtain the micro-load fragment of the current load spectrum, and use the Markov chain model to predict the next micro-load fragment;
  • a weighted working condition parameter calculation module which is used to calculate the motor speed, motor torque and battery state of charge under the weighted working condition based on the weight of the current micro-load segment and the predicted micro-load segment;
  • Electric compressor speed prediction module which is used to use real-time battery outlet water temperature, passenger compartment temperature, passenger compartment target temperature and ambient temperature, and motor speed, motor torque and battery state of charge under weighted conditions as electric compressor speed prediction
  • the input quantity of the model predicts the speed of the electric compressor to obtain the thermal management control strategy of the electric drive bulldozer battery.
  • each module in the deep learning-based electric drive bulldozer battery thermal management control system in this embodiment is the same as each step in the deep learning-based electric drive bulldozer battery thermal management control method in Embodiment 1
  • the specific implementation process is the same, and will not be repeated here.
  • This embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in the above-mentioned deep learning-based battery thermal management control method for an electric drive bulldozer are implemented.
  • This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and operable on the processor.
  • the processor executes the program, the above-mentioned deep learning-based Steps in a battery thermal management control method for an electric drive bulldozer.
  • the embodiments of the present invention may be provided as methods, systems, or computer program products. Accordingly, the present invention can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) having computer-usable program code embodied therein.
  • a computer-usable storage media including but not limited to disk storage and optical storage, etc.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions
  • the device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM) or a random access memory (Random AccessMemory, RAM), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Biophysics (AREA)
  • Mining & Mineral Resources (AREA)
  • Civil Engineering (AREA)
  • Structural Engineering (AREA)
  • Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Human Computer Interaction (AREA)
  • Automation & Control Theory (AREA)
  • Manufacturing & Machinery (AREA)
  • Physiology (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Power Engineering (AREA)
  • Mechanical Engineering (AREA)
  • Genetics & Genomics (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Control Of Positive-Displacement Pumps (AREA)
  • Operation Control Of Excavators (AREA)

Abstract

本发明属于电池热管理控制领域，提供了一种基于深度学习的电传动推土机电池热管理控制方法及系统。其中，该方法包括获取当前载荷谱的微载荷片段，利用马尔可夫链模型对下一个微载荷片段进行预测；基于当前微载荷片段和预测微载荷片段的权重，计算加权工况下的电机转速、电机转矩和电池荷电状态；将实时电池出水温度、乘员舱温度、乘员舱目标温度和环境温度以及加权工况下的电机转速、电机转矩和电池荷电状态作为电动压缩机转速预测模型的输入量，预测出电动压缩机转速，以得到电传动推土机电池的热管理控制策略。

Description

基于深度学习的电传动推土机电池热管理控制方法及系统 技术领域
本发明属于电池热管理控制领域，尤其涉及一种基于深度学习的电传动推土机电池热管理控制方法及系统。
背景技术
本部分的陈述仅仅是提供了与本发明相关的背景技术信息，不必然构成在先技术。
推土机因其复杂多变的施工现场，如果处于垃圾场、大型温室等环境中作业产生的尾气难以快速排出，给人们的生存环境带来威胁，电传动推土机具有零排放的优势，应用越来越广泛。对电传动推土机而言，动力电池的正常工作是保证其安全运行的基础，也是电传动推土机动力系统的核心所在，而一套完善且高效的电池热管理系统是保证动力电池工作在适合温度范围的关键。首先，在电池技术一直得不到突破性进展的技术背景下，如何最大限度地使电池工作在最合理温度范围成为研究的热点。若温度过低，则电池内部电解液离子活性降低，电池放电速度降低，将严重影响整车的动力性；若温度过高，则会造成电池极化的不良影响，甚至会发产生起火爆炸等安全事故。
电动压缩机作为热管理系统必不可少的一环，其消耗的功率随转速的提高而增大。电传动推土机电池热管理系统控制策略一般将压缩机转速设置为几个档位，无法满足具体工况需求。因此，如何根据电池组散热需求实时地匹配电动压缩机的合理转速，对于电传动推土机电池热管理系统的有效工作以及节能 减排具有较大的应用价值。发明人发现，由于电池自身的比热容较大，其冷却过程是一个随着时间连续变化的过程，仅考虑当前时刻推土机及电池的状态，难以准确地反映出电池实际散热需求，从而导致电池散热效果不佳。
发明内容
为了解决上述背景技术中存在的技术问题，本发明提供一种基于深度学习的电传动推土机电池热管理控制方法及系统，其在以双种群自适应遗传算法改进的支持向量机预测模型构建压缩机控制策略的基础上，结合一阶马尔可夫链模型构建能更加准确地反映电池实际散热需求的加权工况，改进该电池热管理系统的控制策略，使该系统满足电池制冷需求的同时，最大限度地降低能耗。
为了实现上述目的，本发明采用如下技术方案：
本发明的第一个方面提供一种基于深度学习的电传动推土机电池热管理控制方法。
一种基于深度学习的电传动推土机电池热管理控制方法，包括：
获取当前载荷谱的微载荷片段，利用马尔可夫链模型对下一个微载荷片段进行预测；
基于当前微载荷片段和预测微载荷片段的权重，计算加权工况下的电机转速、电机转矩和电池荷电状态；
将实时电池出水温度、乘员舱温度、乘员舱目标温度和环境温度以及加权工况下的电机转速、电机转矩和电池荷电状态作为电动压缩机转速预测模型的输入量，预测出电动压缩机转速，以得到电传动推土机电池的热管理控制策略。
本发明的第二个方面提供一种基于深度学习的电传动推土机电池热管理控制系统。
一种基于深度学习的电传动推土机电池热管理控制系统，包括：
微载荷片段预测模块，其用于获取当前载荷谱的微载荷片段，利用马尔可夫链模型对下一个微载荷片段进行预测；
加权工况参数计算模块，其用于基于当前微载荷片段和预测微载荷片段的权重，计算加权工况下的电机转速、电机转矩和电池荷电状态；
电动压缩机转速预测模块，其用于将实时电池出水温度、乘员舱温度、乘员舱目标温度和环境温度以及加权工况下的电机转速、电机转矩和电池荷电状态作为电动压缩机转速预测模型的输入量，预测出电动压缩机转速，以得到电传动推土机电池的热管理控制策略。
本发明的第三个方面提供一种计算机可读存储介质。
一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如上述所述的基于深度学习的电传动推土机电池热管理控制方法中的步骤。
本发明的第四个方面提供一种计算机设备。
一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现如上述所述的基于深度学习的电传动推土机电池热管理控制方法中的步骤。
与现有技术相比，本发明的有益效果是：
(1)本发明提供了基于深度学习的电传动推土机电池热管理控制方法在该控制策略中，以环境温度、电机转速、电机转矩、电池包出水温度、电池荷电状态、乘员舱温度、乘员舱目标温度为训练样本，采用深度学习中的支持向量机算法对样本数据进行训练，得到压缩机转速预测模型，利用马尔可夫链模型 依据当前载荷谱的微载荷片段对下一个微载荷片段进行预测，采用特定公式计算现有片段和预测片段的权重，将加权后的载荷谱引入到预测模型中得到电动压缩机转速，形成电池热管理控制策略，结合蒸发器、冷凝器、电池的微通道热交换器、电子膨胀阀等部件，形成用于电传动推土机的电池热管理系统。其利用自定义客观赋权法公式，在马尔可夫链模型的基础上构建能更加准确地反映电池实际散热需求的加权工况用于电传动推土机电池热管理系统控制，使得压缩机转速预测模型更加准确。
(2)本发明由马尔可夫链模型预测未来工况并与当前时刻工况加权，能够更加准确地反映电池实际散热需求，由此得到的电传动推土机电池热管理控制策略，结合改进支持向量机算法和一阶马尔可夫链模型预测电传动推土机电池热管理系统的压缩机转速，误差更小，在有效改善散热效果的同时，可以最大限度地降低能耗，为热管理系统控制策略的研发提供了依据，对节能减排具有重要意义。
本发明附加方面的优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本发明的实践了解到。
附图说明
构成本发明的一部分的说明书附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。
图1是本发明实施例的基于深度学习的电传动推土机电池热管理控制方法流程示意图；
图2是本发明实施例的压缩机转速控制策略支持向量机模型训练结果。
具体实施方式
下面结合附图与实施例对本发明作进一步说明。
应该指出，以下详细说明都是例示性的，旨在对本发明提供进一步的说明。除非另有指明，本文使用的所有技术和科学术语具有与本发明所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本发明的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合。
实施例一
如图1所示，本实施例提供了一种基于深度学习的电传动推土机电池热管理控制方法，其具体包括如下步骤：
步骤S101：获取当前载荷谱的微载荷片段，利用马尔可夫链模型对下一个微载荷片段进行预测。
具体地，在利用马尔可夫链模型对下一个微载荷片段进行预测之前还包括：
利用历史微载荷片段，建立各类工况的概率转移矩阵。
采用马尔可夫链模型对电传动推土机载荷谱工况进行预测，依据已有的数据，建立各类工况的概率转移矩阵。设当前状态载荷为s i(i＝1,.2,...,p)，下一状态载荷为s j(j＝1,.2,...,q)，令t+1时刻状态载荷为S t+1，则从当前状态S t＝s i转移到下一个状态S t+1＝s j的转移概率可以表示为：
Figure PCTCN2022106154-appb-000001
其中P ij为当前载荷s i达到下一状态载荷s j的转移概率,N ij为当前载荷s i达到下一状态载荷 s j的事件发生次数，
Figure PCTCN2022106154-appb-000002
为当前载荷s i达到任意下一状态载荷的总事件发生次数。
在具体实施中，在利用马尔可夫链模型对下一个微载荷片段进行预测的过程中，基于当前载荷谱的微载荷片段，选取具有最大转移概率的工况作为下一状态载荷的预测结果。
具体地，基于当前载荷s i，选取具有最大转移概率的工况作为下一状态载荷s j的预测结果
Figure PCTCN2022106154-appb-000003
最终得到一阶马尔可夫链预测模型。
步骤S102：基于当前微载荷片段和预测微载荷片段的权重，计算加权工况下的电机转速、电机转矩和电池荷电状态。
t时刻与时刻载荷权重相关系数为-1，根据两个片段载荷谱差异性大小来确定客观权重，差异越大，t+1时刻提供的信息量越多，在综合评价中所能起到的作用也越大，其权重也就越大。t+1时刻的权重为：
Figure PCTCN2022106154-appb-000004
其中
Figure PCTCN2022106154-appb-000005
且
Figure PCTCN2022106154-appb-000006
x t为t时刻电传动推土机载荷，x t+1为t+1时刻电传动推土机载荷，W t+1为t+1时刻的权重，t时刻与t+1时刻载荷权重相关系数为-1，加权后的载荷为x t×(1-W t+1)+x t+1×W t+1；α,β,γ,M均为常数。
例如：
Figure PCTCN2022106154-appb-000007
M＝80。
需要说明的是，α,β,γ,M的具体数值，本领域技术人员可根据实际情况来具体设置，此处不再累述。
步骤S103：将实时电池出水温度、乘员舱温度、乘员舱目标温度和环境温度以及加权工况下的电机转速、电机转矩和电池荷电状态作为电动压缩机转速预测模型的输入量，预测出电动压缩机转速，以得到电传动推土机电池的热管 理控制策略。
在本实施例中，电动压缩机转速预测模型是采用双种群遗传算法改进的支持向量机算法进行训练学习而得到的。
双种群遗传算法在执行变异操作时，将种群划分为两个子种群，其中适应度较低的种群进行自适应柯西变异，适应度较高的种群进行自适应高斯变异，完成寻优过程。个体i更新为：
Figure PCTCN2022106154-appb-000008
其中，
Figure PCTCN2022106154-appb-000009
为比例变换函数，f(x i)为个体x i的适应度函数值，f min和f max分别为当次迭代种群中各个体的适应度函数的最小值和最大值(以求最小值为例，适应度函数值越小，个体越优)。x i和x i'分别为变异前后第i个染色体，range为个体移动范围，N i(0,1)为高斯分布随机数，C i(0,1)为柯西分布随机数。
在训练电动压缩机转速预测模型的过程中，基于双种群自适应遗传算法改进的支持向量机算法通过设定种群个数、最大迭代次数、交叉概率、代沟等初始参数，随机生成惩罚因子和径向基核函数的方差参数组合作为初始种群，每代经过选择、交叉和双种群自适应变异操作，寻找使支持向量机预测模型误差最小的惩罚因子和径向基核函数的方差参数组合。
其中，电动压缩机转速预测模型的训练样本是通过一维、三维热管理软件仿真联合运算得到的。对于电池热管理子系统，设置不同的环境温度、车速、电池散热量(相当于电机转速、电机转矩及电池输出功率)、乘员舱温度、乘员舱目标温度，作为训练样本输入量，仿真得到满足电池安全温度要求的压缩机转速及对应的占空比作为样本输出量，并将上述数据作为该支持向量机预测模 型的训练样本。
对采用双种群遗传算法改进的支持向量机算法进行训练学习，得到不同工况下压缩机转速的预测模型，形成电传动推土机电池热管理系统控制策略。
基于得到的电池热管理控制策略，结合蒸发器、冷凝器、电池的微通道热交换器、电子膨胀阀等部件，最终形成用于电传动推土机的电池热管理系统。
实施例二
本实施例提供了一种基于深度学习的电传动推土机电池热管理控制系统，其具体包括如下模块：
微载荷片段预测模块，其用于获取当前载荷谱的微载荷片段，利用马尔可夫链模型对下一个微载荷片段进行预测；
加权工况参数计算模块，其用于基于当前微载荷片段和预测微载荷片段的权重，计算加权工况下的电机转速、电机转矩和电池荷电状态；
电动压缩机转速预测模块，其用于将实时电池出水温度、乘员舱温度、乘员舱目标温度和环境温度以及加权工况下的电机转速、电机转矩和电池荷电状态作为电动压缩机转速预测模型的输入量，预测出电动压缩机转速，以得到电传动推土机电池的热管理控制策略。
此处需要说明的是，本实施例的基于深度学习的电传动推土机电池热管理控制系统中的各个模块，与实施例一中的基于深度学习的电传动推土机电池热管理控制方法中的各个步骤一一对应，其具体实施过程相同，此处不再累述。
实施例三
本实施例提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如上述所述的基于深度学习的电传动推土机电池热管理 控制方法中的步骤。
实施例四
本实施例提供了一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现如上述所述的基于深度学习的电传动推土机电池热管理控制方法中的步骤。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用硬件实施例、软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使 得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random AccessMemory，RAM)等。
以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (10)

  1. 一种基于深度学习的电传动推土机电池热管理控制方法，其特征在于，包括：
    获取当前载荷谱的微载荷片段，利用马尔可夫链模型对下一个微载荷片段进行预测；
    基于当前微载荷片段和预测微载荷片段的权重，计算加权工况下的电机转速、电机转矩和电池荷电状态；
    将实时电池出水温度、乘员舱温度、乘员舱目标温度和环境温度以及加权工况下的电机转速、电机转矩和电池荷电状态作为电动压缩机转速预测模型的输入量，预测出电动压缩机转速，以得到电传动推土机电池的热管理控制策略。
  2. 如权利要求1所述的基于深度学习的电传动推土机电池热管理控制方法，其特征在于，在利用马尔可夫链模型对下一个微载荷片段进行预测之前还包括：
    利用历史微载荷片段，建立各类工况的概率转移矩阵。
  3. 如权利要求2所述的基于深度学习的电传动推土机电池热管理控制方法，其特征在于，在利用马尔可夫链模型对下一个微载荷片段进行预测的过程中，基于当前载荷谱的微载荷片段，选取具有最大转移概率的工况作为下一状态载荷的预测结果。
  4. 如权利要求1所述的基于深度学习的电传动推土机电池热管理控制方法，其特征在于，当前微载荷片段和预测微载荷片段的权重的计算公式为：
    Figure PCTCN2022106154-appb-100001
    其中
    Figure PCTCN2022106154-appb-100002
    且
    Figure PCTCN2022106154-appb-100003
    x t为t时刻电传动推土机载荷，x t+1为t+1时刻电传动推土机载荷，W t+1为t+1时刻的权重，t时刻与t+1时刻载荷权重相关系数为-1，加权后的载荷为x t×(1-W t+1)+x t+1×W t+1；α,β,γ,M均为常数。
  5. 如权利要求1所述的基于深度学习的电传动推土机电池热管理控制方法，其特征在于，电动压缩机转速预测模型是采用双种群遗传算法改进的支持向量机算法进行训练学习而得到的。
  6. 如权利要求1所述的基于深度学习的电传动推土机电池热管理控制方法，其特征在于，在训练电动压缩机转速预测模型的过程中，基于双种群自适应遗传算法改进的支持向量机算法通过设定初始参数，随机生成惩罚因子和径向基核函数的方差参数组合作为初始种群，每代经过选择、交叉和双种群自适应变异操作，寻找使支持向量机预测模型误差最小的惩罚因子和径向基核函数的方差参数组合。
  7. 如权利要求1所述的基于深度学习的电传动推土机电池热管理控制方法，其特征在于，电动压缩机转速预测模型的训练样本是通过一维、三维热管理软件仿真联合运算得到的。
  8. 一种基于深度学习的电传动推土机电池热管理控制系统，其特征在于，包括：
    微载荷片段预测模块，其用于获取当前载荷谱的微载荷片段，利用马尔可夫链模型对下一个微载荷片段进行预测；
    加权工况参数计算模块，其用于基于当前微载荷片段和预测微载荷片段的权重，计算加权工况下的电机转速、电机转矩和电池荷电状态；
    电动压缩机转速预测模块，其用于将实时电池出水温度、乘员舱温度、乘员舱目标温度和环境温度以及加权工况下的电机转速、电机转矩和电池荷电状态作为电动压缩机转速预测模型的输入量，预测出电动压缩机转速，以得到电传动推土机电池的热管理控制策略。
  9. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现如权利要求1-7中任一项所述的基于深度学习的电传动推土机电池热管理控制方法中的步骤。
  10. 一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述程序时实现如权利要求1-7中任一项所述的基于深度学习的电传动推土机电池热管理控制方法中的步骤。
PCT/CN2022/106154 2021-06-09 2022-07-18 基于深度学习的电传动推土机电池热管理控制方法及系统 WO2022258079A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/014,448 US20230273593A1 (en) 2021-06-09 2022-07-18 Method and system for control of battery thermal management of electric drive bulldozers based on deep learning

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110643527.6A CN113378463B (zh) 2021-06-09 2021-06-09 基于深度学习的电传动推土机电池热管理控制方法及系统
CN202110643527.6 2021-06-09

Publications (1)

Publication Number Publication Date
WO2022258079A1 true WO2022258079A1 (zh) 2022-12-15

Family

ID=77573198

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/106154 WO2022258079A1 (zh) 2021-06-09 2022-07-18 基于深度学习的电传动推土机电池热管理控制方法及系统

Country Status (3)

Country Link
US (1) US20230273593A1 (zh)
CN (1) CN113378463B (zh)
WO (1) WO2022258079A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115629614A (zh) * 2022-12-20 2023-01-20 南京航空航天大学 一种基于遗传算法的仓储多agv路径规划方法
CN116339151A (zh) * 2023-05-12 2023-06-27 西北工业大学 一种针对电动飞机电推进器热管理系统、控制方法及终端

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113378463B (zh) * 2021-06-09 2023-02-28 山东大学 基于深度学习的电传动推土机电池热管理控制方法及系统
CN113947021A (zh) * 2021-10-20 2022-01-18 重庆金康赛力斯新能源汽车设计院有限公司 一种热管理控制方法和热管理控制装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140356656A1 (en) * 2013-05-31 2014-12-04 Laurence Lujun Chen Intelligent Battery Management System and Method for Optimizing Battery Set to the Best Performance
CN109271700A (zh) * 2018-09-10 2019-01-25 柯瑞林 基于深度学习多层网络建模的电池热管理方法及系统
CN111261973A (zh) * 2020-01-19 2020-06-09 重庆大学 基于模型预测控制的电动汽车整车电池热管理方法
CN111611750A (zh) * 2020-05-28 2020-09-01 北京理工大学 一种应用数字孪生技术的锂离子电池充电及热管理方法
CN113378463A (zh) * 2021-06-09 2021-09-10 山东大学 基于深度学习的电传动推土机电池热管理控制方法及系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107117074A (zh) * 2015-11-30 2017-09-01 赵世龙 永磁电机驱动方法
US20200011932A1 (en) * 2018-07-05 2020-01-09 Nec Laboratories America, Inc. Battery capacity fading model using deep learning
CN111723982A (zh) * 2020-06-11 2020-09-29 电子科技大学 基于灰色-马尔可夫链的中长期电力负荷组合预测方法
CN112800541A (zh) * 2021-01-26 2021-05-14 南京航空航天大学 一种基于支持向量机回归的航空发动机载荷谱任务段建模方法
CN112801410A (zh) * 2021-02-26 2021-05-14 国网山东省电力公司 基于大数据技术的电费流入预测分析方法
CN112801415A (zh) * 2021-03-05 2021-05-14 国网山东省电力公司寿光市供电公司 基于马尔科夫链分布模型的超短期负荷预测方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140356656A1 (en) * 2013-05-31 2014-12-04 Laurence Lujun Chen Intelligent Battery Management System and Method for Optimizing Battery Set to the Best Performance
CN109271700A (zh) * 2018-09-10 2019-01-25 柯瑞林 基于深度学习多层网络建模的电池热管理方法及系统
CN111261973A (zh) * 2020-01-19 2020-06-09 重庆大学 基于模型预测控制的电动汽车整车电池热管理方法
CN111611750A (zh) * 2020-05-28 2020-09-01 北京理工大学 一种应用数字孪生技术的锂离子电池充电及热管理方法
CN113378463A (zh) * 2021-06-09 2021-09-10 山东大学 基于深度学习的电传动推土机电池热管理控制方法及系统

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115629614A (zh) * 2022-12-20 2023-01-20 南京航空航天大学 一种基于遗传算法的仓储多agv路径规划方法
CN116339151A (zh) * 2023-05-12 2023-06-27 西北工业大学 一种针对电动飞机电推进器热管理系统、控制方法及终端
CN116339151B (zh) * 2023-05-12 2023-08-04 西北工业大学 一种针对电动飞机电推进器热管理系统、控制方法及终端

Also Published As

Publication number Publication date
CN113378463A (zh) 2021-09-10
US20230273593A1 (en) 2023-08-31
CN113378463B (zh) 2023-02-28

Similar Documents

Publication Publication Date Title
WO2022258079A1 (zh) 基于深度学习的电传动推土机电池热管理控制方法及系统
Zhang et al. Optimal design of building environment with hybrid genetic algorithm, artificial neural network, multivariate regression analysis and fuzzy logic controller
CN109711080B (zh) 一种冷热电联供系统多时间尺度优化运行方法
Xuemei et al. Building cooling load forecasting model based on LS-SVM
Lo et al. Economic dispatch of chiller plant by improved ripple bee swarm optimization algorithm for saving energy
CN107704875A (zh) 基于改进ihcmac神经网络的建筑负荷预测方法和装置
CN113224412B (zh) 动力电池的控温方法、ampc控制器、热管理系统及介质
CN109242188B (zh) 一种钢铁煤气系统长期区间预测及其结构学习方法
CN113050426B (zh) 一种融合遗传蚁群算法的热管理控制方法及系统
CN114537087B (zh) 一种纯电动汽车集成式热管理系统性能优化方法
CN113405223A (zh) 一种基于grnn的冷机台数控制方法及其控制系统
KR20180138371A (ko) 데이터 기반 모델 평가 및 이를 이용한 캡슐형 빙축열 시스템 예측제어 방법
CN115732810A (zh) 一种电动汽车电池包加温系统控制方法
Tian et al. Calibration method for sensor drifting bias in data center cooling system using Bayesian Inference coupling with Autoencoder
Wang et al. Optimization of the powertrain and energy management control parameters of a hybrid hydraulic vehicle based on improved multi-objective particle swarm optimization
Irshad et al. Novel optimized hybrid neuro-fuzzy approach for analysis of cold thermal storage system-assisted air conditioning system performance
Yan et al. Model predictive control of the air-conditioning system for electric bus
Cai et al. Parallel pump and chiller system optimization method for minimizing energy consumption based on a novel multi-objective gorilla troops optimizer
Adnan et al. Modeling baseline electrical energy use of chiller system by artificial neural network
Xuemei et al. A novel building cooling load prediction based on SVR and SAPSO
CN113594513B (zh) 一种质子交换膜燃料电池模型优化控制方法
Pan et al. Optimization of liquid cooling heat dissipation control strategy for electric vehicle power batteries based on linear time-varying model predictive control
Zhu et al. Application of machine learning and its improvement technology in modeling of total energy consumption of air conditioning water system
He et al. Backpropagation Neural Network‐Based Prediction Model for Fuel Cell Thermal Management System
Nguyen et al. A study on artificial intelligence applied modeling for battery cooling system operation optimization

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22819675

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE