吉林大学学报(工学版) ›› 2021, Vol. 51 ›› Issue (4): 1405-1413.doi: 10.13229/j.cnki.jdxbgxb20200280

• 计算机科学与技术 • 上一篇    

基于BPNN在线学习预测模型的扭矩实时跟踪控制

董延华1(),刘靓葳1,2,赵靖华1,3(),李亮1,解方喜3   

  1. 1.吉林师范大学 计算机学院，吉林 四平 136000
    2.长春金融高等专科学校 信息技术学院，长春 130028
    3.吉林大学 汽车仿真与控制国家重点实验室，长春 130022
  • 收稿日期:2020-04-28 出版日期:2021-07-01 发布日期:2021-07-14
  • 通讯作者: 赵靖华 E-mail:computerdyp@jlnu.edu.cn;zhaojh08@mails.jlu.edu.cn
  • 作者简介:董延华（1971-），男，教授，博士.研究方向：人工智能. E-mail： computerdyp@jlnu.edu.cn
  • 基金资助:
    国家自然科学基金项目(61773009);教育技术研究基金项目(2018A01025);吉林省科技发展计划项目(20190302105GX);吉师研创项目(202023)

Real-time torque tracking control based on BPNN online learning prediction model

Yan-hua DONG1(),Jing-wei LIU1,2,Jing-hua ZHAO1,3(),Liang LI1,Fang-xi XIE3   

  1. 1.College of Computer，Jilin Normal University，Siping 136000，China
    2.College of Information Technology，Changchun Finance College，Changchun 130028，China
    3.State Key Laboratory of Automotive Simulation and Control，Jilin University，Changchun 130022，China
  • Received:2020-04-28 Online:2021-07-01 Published:2021-07-14
  • Contact: Jing-hua ZHAO E-mail:computerdyp@jlnu.edu.cn;zhaojh08@mails.jlu.edu.cn

RICH HTML

 6   

PDF (PC)

82

摘要：

基于大量的台架试验数据，利用反向传播神经网络（BPNN）设计了燃油喷射在线学习预测模型，结合PID反馈完成扭矩跟踪的实时控制。其中，燃油喷射BPNN预测模型采用一种实时的简化离散模型，模型的阈值可以在线学习更新，具有参数自适应性。台架试验表明，相比于固定参数的BPNN模型，提出的阈值在线学习的BPNN模型具有更高的预测精度；提出的可变阈值VTBPNN预测前馈加PID反馈控制器能够满足扭矩跟踪的实时性需求，并且相比于普通可变参数VPPID控制器，在瞬态工况干扰下鲁棒性更强，跟踪误差更小。

关键词: 反向传播神经网络, 人工智能, 参数自适应, 扭矩跟踪控制

Abstract:

An online learning prediction model for fuel injection based on a large amount of experimental data is designed by utilizing back propagation neural network (BPNN), and a real-time torque tracking controller consisting of the BPNN predictive feedforward and a PID feedback is proposed. A simplified and discretized real-time model is adopted in the BPNN prediction model. The threshold value of the BPNN prediction model with parameter adaptability can be learned and updated online. Several experimental results show that the BPNN prediction model with online variable threshold proposed in this paper has higher prediction accuracy comparing with the model with fixed threshold, and that the torque tracking controller proposed in this paper can satisfy the real-time control requirement and has smaller error rate under the transient condition comparing with the PID controller.

Key words: back-propagation neural network （BPNN）, artificial intelligence, parameter adaptation, torque tracking control

中图分类号: 

  • TP399

图1

燃油喷射BPNN预测模型的结构"

图2

转速、扭矩以及燃油喷射量数据集"

图3

隐藏层神经元数目误差敏感性分析"

图4

模型线性回归分析"

图5

FTP75部分瞬态工况条件"

图6

FTP75工况下燃油喷射BPNN模型预测效果"

图7

扭矩阶跃目标跟踪效果对比"

图8

图7的局部放大"

图9

扭矩正弦目标跟踪效果对比"

图10

图9的局部放大"

表1

扭矩跟踪误差分析"

误差指标阶跃扭矩目标工况正弦扭矩目标工况
MADRMSEMADRMSE
PID2.15184.60004.39725.3276
BPNN?PID1.68854.01382.62583.5417

图11

VTBPNN-PID 扭矩跟踪控制系统"

图12

FTP75工况下燃油喷射VTBPNN模型预测效果"

图13

扭矩瞬态目标跟踪效果对比"

图14

图13的局部放大"

图15

扭矩跟踪误差对比"

图16

图15的局部放大"

1 Liu B, Wang R, Zhao G, et al. Prediction of rock mass parameters in the TBM tunnel based on BP neural network integrated simulated annealing algorithm[J]. Tunnelling and Underground Space Technology, 2020,95(c):103103-103103.
2 Moskwa J J, Hedrick J K. Modeling and validation of automotive engines for control algorithm development[J]. Journal of Dynamic Systems Measurement and Control-transactions of The ASME, 1992, 114(2): 278-285.
3 Hirahara H, Yoshida K, Iwase M, et al. Torque estimation based on nonlinear engine model considering crankshaft torsion[C]∥IEEE/ASME International Conference on Advanced Intelligent Mechatronics, Piscataway, USA,2010: 682-687.
4 Anjum R, Yar A, Yousufzai I K, et al. Second order sliding mode based speed tracking control for torque management of gasoline engines[C]∥Asian Control Conference, Kitakyushu-shi, Japan,2019: 555-560.
5 Anjum R, Yar A, Bhatti A I, et al. Dual loop speed tracking control for torque management of gasoline engines[C]∥European Control Conference, Naples, Italy,2019: 3084-3089.
6 陈虹,郭露露,宫洵,等. 智能时代的汽车控制[J]. 自动化学报, 2020, 46(7): 1313-1332.
Chen Hong, Guo Lu-lu, Gong Xun, et al. Automotive control in the age of intelligence[J]. Journal of Automation, 2020, 46(7): 1313-1332.
7 Zhao J, Hu Y, Gao B, et al. Sequential optimization of eco-driving taking into account fuel economy and emissions[J]. IEEE Access,2019,7:130841-130853.
8 Park J, Choi S B, Oh J, et al. Adaptive torque tracking control during slip engagement of a dry clutch in vehicle powertrain[J]. Mechanism and Machine Theory, 2019,134: 249-266.
9 Kang M, Shen T. Modeling and optimal control for torque tracking of spark-ignition engines with low pumping loss[C]∥Chinese Control Conference, Chendu, China, 2016: 8956-8961.
10 Vuelvas J, Archbold G, Ruiz F. MPC Weighted sum approach applied to torque tracking and CO2emission reduction on engines[C]∥ IEEE 3rd Colombian Conference on Automatic Control (CCAC), Cartagena, Colombia, 2017: 1-6.
11 Bouvenot J B,Andlauer B,Stabat P,et al. Gas stirling engine µ CHP boiler experimental data driven model for building energy simulation[J]. Energy and Buildings, 2014, 84: 117-131.
12 尤晓东,苏崇宇,汪毓铎. BP神经网络算法改进综述[J]. 民营科技,2018, 23(4):146-147.
You Xiao-dong,Su Cong-yu,Wang Yu-duo. Summary of improvement of BP neural network algorithm[J]. Private Technology, 2018,23(4):146-147.
13 Ong C K,Yao T K, Alfred R, et al. A Comparison of BPNN,RBF, and ENN in Number Plate Recognition[C]∥ Second International Conference on Soft Computing in Data Science, Kuala Lumpur, Malaysia,2016: 37-47.
14 He J, Yang W, Wang J, et al. Fast HEVC coding unit decision based on BP-neural network[J]. International Journal of Grid and Distributed Computing, 2015, 8(4): 289-300.
15 Venayagamoorthy G K, Rohrig K, Erlich I. One step ahead: short-term wind power forecasting and intelligent predictive control based on data analytics[J]. IEEE Power & Energy Magazine, 2012, 10(5): 70-78.
16 Rahman M A, Hoque M A. On-line adaptive artificial neural net-work based vector control of permanent magnet synchronous motors[J]. IEEE Transactions on Energy Conversion, 1998, 13 (4):311-318.
17 Hu Y, Chen H, Wang P, et al. Nonlinear model predictive controller design based on learning model for turbocharged gasoline engine of passenger vehicle[J]. Mechanical Systems and Signal Processing, 2018,109: 74-88.
[1] 赵亚慧,杨飞扬,张振国,崔荣一. 基于强化学习和注意力机制的朝鲜语文本结构发现[J]. 吉林大学学报(工学版), 2021, 51(4): 1387-1395.
[2] 尚福华,曹茂俊,王才志. 基于人工智能技术的局部离群数据挖掘方法[J]. 吉林大学学报(工学版), 2021, 51(2): 692-696.
[3] 赵海英,周伟,侯小刚,张小利. 基于多任务学习的传统服饰图像双层标注[J]. 吉林大学学报(工学版), 2021, 51(1): 293-302.
[4] 欧阳丹彤,马骢,雷景佩,冯莎莎. 知识图谱嵌入中的自适应筛选[J]. 吉林大学学报(工学版), 2020, 50(2): 685-691.
[5] 李贻斌,郭佳旻,张勤. 人体步态识别方法与技术[J]. 吉林大学学报(工学版), 2020, 50(1): 1-18.
[6] 徐谦,李颖,王刚. 基于深度学习的行人和车辆检测[J]. 吉林大学学报(工学版), 2019, 49(5): 1661-1667.
[7] 高万夫,张平,胡亮. 基于已选特征动态变化的非线性特征选择方法[J]. 吉林大学学报(工学版), 2019, 49(4): 1293-1300.
[8] 欧阳丹彤,肖君,叶育鑫. 基于实体对弱约束的远监督关系抽取[J]. 吉林大学学报(工学版), 2019, 49(3): 912-919.
[9] 顾海军, 田雅倩, 崔莹. 基于行为语言的智能交互代理[J]. 吉林大学学报(工学版), 2018, 48(5): 1578-1585.
[10] 董飒, 刘大有, 欧阳若川, 朱允刚, 李丽娜. 引入二阶马尔可夫假设的逻辑回归异质性网络分类方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1571-1577.
[11] 王旭, 欧阳继红, 陈桂芬. 基于垂直维序列动态时间规整方法的图相似度度量[J]. 吉林大学学报(工学版), 2018, 48(4): 1199-1205.
[12] 张浩, 占萌苹, 郭刘香, 李誌, 刘元宁, 张春鹤, 常浩武, 王志强. 基于高通量数据的人体外源性植物miRNA跨界调控建模[J]. 吉林大学学报(工学版), 2018, 48(4): 1206-1213.
[13] 李雄飞, 冯婷婷, 骆实, 张小利. 基于递归神经网络的自动作曲算法[J]. 吉林大学学报(工学版), 2018, 48(3): 866-873.
[14] 刘杰, 张平, 高万夫. 基于条件相关的特征选择方法[J]. 吉林大学学报(工学版), 2018, 48(3): 874-881.
[15] 黄岚, 纪林影, 姚刚, 翟睿峰, 白天. 面向误诊提示的疾病-症状语义网构建[J]. 吉林大学学报(工学版), 2018, 48(3): 859-865.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 《吉林大学学报(工学版)》编辑部
地址：长春市人民大街5988号　电话：+86-431-85095297　传真：+86-431-85094074　E-mail: xbgxb@jlu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发