针对传统PID轨压控制技术参数整定繁琐、时间成本较高等问题，提出基于NI PXI的智能PID参数自整定技术。利用NI PXI现场可编程、数据高速传输、并行处理的特点，通过改进的分级BP神经网络进行PID参数自整定。在无需进行大量标定实验的前提下，完成轨压控制，降低时间成本。实验结果表明：该方案在自行设计的高压共轨系统上较好地实现了轨压控制，稳态轨压实验与连续喷油实验的控制效果与成熟的基于MAP图的PID技术差异不大，2种技术稳态轨压实验最大偏差为2.1 MPa，连续喷油实验最大偏差为2 MPa；改进的分级BP-PID技术的响应速度较快，低轨压响应速度为0.424 s，而基于MAP图的PID技术响应速度为2.815 s。基于NI PXI的轨压智能控制降低实验时间成本，保证了良好的轨压控制效果，轨压控制的响应速度较快。
Aiming at the characteristics of cumbersome tuning work and high time cost of traditional PID rail pressure control technical parameters and so on, the intelligent PID parameter self-tuning technology based on NI PXI was studied in this paper. Utilizing the characteristics of NI PXI field programmable, high-speed data transmission and parallel processing, PID parameter self-tuning was carried out through the improved hierarchical BP neural network, and the rail pressure control was completed without a large number of calibration experiments that reducing time cost. The experimental results showed that the scheme achieved a good rail pressure control on the self-designed high-pressure common rail system. The control effect of the steady-state rail pressure experiment and the continuous fuel injection experiment were not much different from that of the mature MAP-based PID technology. The maximum deviation of the steady-state rail pressure experiment of the two technologies was 2.1 MPa and the maximum deviation of the continuous fuel injection experiment was 2 MPa; but the response speed of the improved hierarchical BP-PID technology was fast. The response speed of low rail pressure was 0.424 s while the response speed of MAP-based PID technology was 2.815 s. By means of using rail pressure intelligent control based on NI PXI, the experiment time cost is reduced, rail pressure is controlled well and the response speed of the rail pressure control is fast.
2022,44(9): 102-108 收稿日期：2021-06-07
DOI：10.3404/j.issn.1672-7649.2022.09.021
分类号：TK423
基金项目：国家自然科学基金重点国际（地区）合作研究项目（52020105009）
作者简介：王欣然(1997-)，男，硕士研究生，研究方向为内燃机建模控制
参考文献：
[1] 许越, 倪培永, 邓红喜. 不同供油提前角对船用柴油机排放的影响[J]. 舰船科学技术, 2021, 43(1): 117–121
XU Yue, NI Pei-yong, DENG Hong-xi. Effect of different fuel advance angles on marine diesel engine emissions[J]. Ship Science and Technology, 2021, 43(1): 117–121
[2] 张韩西子, 刘瀛昊, 戴钰杰, 等. 船舶柴油机颗粒物排放控制技术研究进展[J]. 舰船科学技术, 2021, 43(7): 118–122
ZHANG Han-xi-zi, LIU Ying-hao, DAI Yu-jie, et al. Research progress of marine diesel engine’s PM emission control technology[J]. Ship Science and Technology, 2021, 43(7): 118–122
[3] 田丙奇. 柴油机高压共轨喷油系统动态特性研究[D]. 哈尔滨: 哈尔滨工程大学, 2014.
[4] 杨强, 杨建国. 船用中速柴油机高压共轨系统的现状与发展趋势[J]. 船海工程, 2019, 48(3): 142–146
YANG Qiang, YANG Jian-guo. Present situation and development trend of high pressure common rail system of marine medium-speed diesel engine[J]. Ship & Ocean Engineering, 2019, 48(3): 142–146
[5] 宋国民, 刘学瑜. 基于参数自调整PID控制的高压共轨数字调压系统[J]. 柴油机, 2002(5): 25–27
SONG Guo-min, LIU Xue-yu. Digital pressure regulating system for high pressure common-rail based on parameters self-adjusting pid control[J]. Diesel Engine, 2002(5): 25–27
[6] SU Hai-feng, HAO Gang, LI Peng-zhi, et al. Feed forward fuzzy PID controller for common-rail pressure control of diesel engine[C]//2010 International Conference on Measuring Technology and Mechatronics Automation, 2010(2): 264-267.
[7] 史卫全, 李铁. 基于NI CompactRIO的压力反馈式喷油控制系统设计与开发[J]. 车用发动机, 2017(6): 48–51
SHI Wei-quan, LI Tie. Design and development of pressure feedback injection control system based on NI CompactRIO[J]. Vehicle Engine, 2017(6): 48–51
[8] 杨阳. 高压共轨柴油机电控系统分析与研究[D]. 上海: 上海交通大学, 2007.
[9] 李悟早, 郭术义, 任思杰. 模糊控制理论综述[J]. 河南科技, 2019(11): 12–15
LI Wu-zao, GUO Shu-yi, REN Si-jie. Summary of fuzzy control theory[J]. Henan Science and Technology, 2019(11): 12–15
[10] PAOLO L, BRUNO M, ALESSANDRO R. A control-oriented model of a Common rail injection system for diesel engines[C]//2005 IEEE Conference on Emerging Technologies and Factory Automation, 2005(1): 557-563.
[11] PAOLO L, BRUNO M, ALESSANDRO R. Nonlinear modelling and control of a common rail injection system for diesel engines[J]. Applied Mathematical Modelling, 2007, 31: 1770–1784
[12] SEUNGWOO H, JAEWOOK S, MYOUNGHO S. Common rail pressure controller for diesel engines using an empirical model[C]//2012 IEEE Vehicle Power and Propulsion Conference, 2012: 887-892.
[13] SEUNGWOO H, JAEWOOK S, JEONGWON S, et al. Coordinated control strategy for the common-rail pressure using a metering unit and a pressure control valve in diesel engines[J]. Proceedings of the Institution of Mechanical Engineers, Part D:Journal of Automobile Engineering 229, 2014: 898–911
[14] 刘益民. 基于改进BP神经网络的PID控制方法的研究[D]. 西安: 中国科学院研究生院(西安光学精密机械研究所), 2007.
[15] 王勇. 高压共轨电控柴油机神经网络轨压控制研究[D]. 大连: 大连理工大学, 2008.
[16] 李航, 杜璠, 胡晓兵, 等. 改进的BP神经网络PID控制器在气体浓度控制中的研究[J]. 四川大学学报(自然科学版), 2020, 57(6): 1103–1109
LI Hang, DU Fan, HU Xiao-bing, et al. Research on improved BP neural network PID controller in gas concentration control[J]. Journal of Sichuan University(Natural Science Edition), 2020, 57(6): 1103–1109
[17] 陈桢皓. 柴油微引燃乙醇发动机的研究与仿真优化[D]. 上海: 上海交通大学, 2018.
[18] 苏瑜, 荆文芳, 卢晓春, 等. 基于BP-PID控制的载波频率准确度提高算法[J]. 系统工程与电子技术, 2021, 43(7): 1894-1903.
SU Yu, JING Wen-fang, LU Xiao-chun, et al. Carrier frequency accuracy improvement algorithm based on BP-PID control[J]. Systems Engineering and Electronics: 1894-1903.
[19] 赵志强, 范伯元. 模糊PID数字调速器的设计[J]. 内燃机工程, 1999(2): 21–28
ZHAO Zhi-qiang, FAN Bo-yuan. Fuzzy PID design of digital governor[J]. Chinese Internal Combustion Engine Engineering, 1999(2): 21–28
