基于DWT的汽油机爆震实时控制系统研究
摘要
火花点火式发动机广泛应用于各种汽车。爆震是汽油机的一种不正常燃烧状态,严重爆震会使汽油机各项性能全面下降,甚至损坏发动机。但是汽油机工作在爆震边缘时将获得最佳动力性和经济性,因此检测汽油机爆震并控制汽油机的点火提前角使其工作在爆震边缘有重要意义。
本文研究Mallat小波分解及重构算法,提出了消除Mallat单子带重构算法过程中产生频率折叠的方法,并应用该算法从汽油机振动信号中提取出爆震特征。分析d2子带信号特征,提出了基于DWT(Discrete Wavelet Transform,离散小波变换)的特征域信号平均能量的爆震强度评价方法,研究表明该方法精度较高,运算量小,同时方便离散小波变换应用于爆震实时控制。
基于DWT分析汽油机振动信号特征,提出适于实时应用的爆震控制策略。该策略根据发动机的转速和负荷将发动机的工作状态划分为爆震区和非爆震区,两者采用不同的控制策略。发动机在非爆震区域工况下,其最佳点火提前角为初始点火提前角、基本点火提前角和修正点火提前角之和。爆震区最佳点火提前角为基本点火提前角、爆震修正和其他修正之和。
应用Matlab/Altera Dsp Builder建立基于离散小波变换的汽油机爆震实时控制系统。在Simulink环境下对该模型进行功能仿真,应用SignalCompiler将生成的模型转化为VHDL代码。使用ModelSim对VHDL代码进行RTL级仿真,应用Quartus II对VHDL代码进行编译适配、静态时序仿真和管脚绑定。选择合适的FPGA(Field Programmable Gate Array,现场可编程门阵列)开发板完成工程文件的下载。开发FPGA与ECU(Electronic Control Unit,电子控制单元)之间的通信接口,方便ECU对爆震实时控制系统反馈以及本控制系统的扩展应用。
本文将离散小波变换应用于爆震特征提取,实现了对汽油机爆震的实时精确控制。
SI (Spark Ignition) engine has been widely used in various automobiles. Knock is a kind of abnormal combustion. Heavy knock worsens the performance of gasoline engine or even damage it, but weak knock improves the dynamic performance and the fuel economy of it. Therefore measuring knock and controlling the advance ignition angle to make the engine work at the state of weak knock are very important.
According to the characteristics of the d2 sub-band signals, the knock intensity determination is raised based on the average energy characteristics of the knock signal, it is of high accuracy, with a small amount of computing ,and is convenient to the application of discrete wavelet transform for real-time control.
The characteristics of the vibration signals of the gasoline engine is analyzed based on DWT and put forward the strategies for real-time control.The engine opration is divided into state of non-knock and knock according to engine speed and load, which use different control strategies. When worked in non-knock conditions, the most optimum spark angle is the adder of the initial spark angle, the basic ignition angle and the amendment spark angle. In knock regions, the most optimum spark angle is the adder of the basic spark angle, the knock amendment angle and other amendments.
The real-time knock control system of the gasoline engine is built based on discrete wavelet transform with the application of Matlab/Altera Dsp Builder. The functional simulation of the model is done in Simulink environment, the model is generated into VHDL code with the application of SignalCompiler. Use ModelSim to accomplish RTL-level simulation of the VHDL code, compile VHDL code with the application of Quartus II, static timing simulation and pin binding. Choose a suitable FPGA development board and download the document to complete the works. Communication interface between the ECU and FPGA is developed, which is convenient for the ECU to feedback the system and the expendation of the system.
Discrete wavelet transoform is used in this paper to extract the character of engine knock, realize the accuract reall-time control of the gasoline engine.
本文研究Mallat小波分解及重构算法,提出了消除Mallat单子带重构算法过程中产生频率折叠的方法,并应用该算法从汽油机振动信号中提取出爆震特征。分析d2子带信号特征,提出了基于DWT(Discrete Wavelet Transform,离散小波变换)的特征域信号平均能量的爆震强度评价方法,研究表明该方法精度较高,运算量小,同时方便离散小波变换应用于爆震实时控制。
基于DWT分析汽油机振动信号特征,提出适于实时应用的爆震控制策略。该策略根据发动机的转速和负荷将发动机的工作状态划分为爆震区和非爆震区,两者采用不同的控制策略。发动机在非爆震区域工况下,其最佳点火提前角为初始点火提前角、基本点火提前角和修正点火提前角之和。爆震区最佳点火提前角为基本点火提前角、爆震修正和其他修正之和。
应用Matlab/Altera Dsp Builder建立基于离散小波变换的汽油机爆震实时控制系统。在Simulink环境下对该模型进行功能仿真,应用SignalCompiler将生成的模型转化为VHDL代码。使用ModelSim对VHDL代码进行RTL级仿真,应用Quartus II对VHDL代码进行编译适配、静态时序仿真和管脚绑定。选择合适的FPGA(Field Programmable Gate Array,现场可编程门阵列)开发板完成工程文件的下载。开发FPGA与ECU(Electronic Control Unit,电子控制单元)之间的通信接口,方便ECU对爆震实时控制系统反馈以及本控制系统的扩展应用。
本文将离散小波变换应用于爆震特征提取,实现了对汽油机爆震的实时精确控制。
SI (Spark Ignition) engine has been widely used in various automobiles. Knock is a kind of abnormal combustion. Heavy knock worsens the performance of gasoline engine or even damage it, but weak knock improves the dynamic performance and the fuel economy of it. Therefore measuring knock and controlling the advance ignition angle to make the engine work at the state of weak knock are very important.
According to the characteristics of the d2 sub-band signals, the knock intensity determination is raised based on the average energy characteristics of the knock signal, it is of high accuracy, with a small amount of computing ,and is convenient to the application of discrete wavelet transform for real-time control.
The characteristics of the vibration signals of the gasoline engine is analyzed based on DWT and put forward the strategies for real-time control.The engine opration is divided into state of non-knock and knock according to engine speed and load, which use different control strategies. When worked in non-knock conditions, the most optimum spark angle is the adder of the initial spark angle, the basic ignition angle and the amendment spark angle. In knock regions, the most optimum spark angle is the adder of the basic spark angle, the knock amendment angle and other amendments.
The real-time knock control system of the gasoline engine is built based on discrete wavelet transform with the application of Matlab/Altera Dsp Builder. The functional simulation of the model is done in Simulink environment, the model is generated into VHDL code with the application of SignalCompiler. Use ModelSim to accomplish RTL-level simulation of the VHDL code, compile VHDL code with the application of Quartus II, static timing simulation and pin binding. Choose a suitable FPGA development board and download the document to complete the works. Communication interface between the ECU and FPGA is developed, which is convenient for the ECU to feedback the system and the expendation of the system.
Discrete wavelet transoform is used in this paper to extract the character of engine knock, realize the accuract reall-time control of the gasoline engine.
引文
1李建权.火花点火发动机爆震燃烧的研究.天津大学博士学位论文. 1998: 58~70
2 Taylor CF著,张胜瑕,程未云等译.内燃机(下册).人民交通出版,1981: 230~235
3李继军.火花点火发动机挤气射流-紊流型燃烧室及基于缸内压力波动检测爆震控制系统研究.西安交通大学博士学位论文. 1995: 36~58
4陈家哗等编.内燃机燃烧.哈尔滨船舶工程学院出版社,1988: 158~162
5 Knock Measurement for Fuel Evaluation in Spark Ignition Engines. Mechanical Systems and Signal Processing. 2001, 13(3): 395~407
6 Giorgio Rizzoni, Xing Cindy Chen. Detection of Internal Combustion Engine Knock Using Time-Frequency Distributions. IEEE Transaction on Instrumentation an Measurements. 1993, 19(3): 360~363
7 A.M.Zoubir. A Comparative Study of Multiple Test Based Techniques for Optimal Sensor Location for Knock Detection. Mechanical Systems and Signal Processing. 2000, 14(3): 353~369
8 M.D.Boland, A.M.Zoubir. Identification of Time-Varying Non-Linear Systems with Application to Knock Detection in Combustion Engines.IE EE Transaction on Instrumentation and Measurements. 1997, 15(3): 799~802
9 C.Hudson, X.Gao, R.Stone. Knock Measurement for Fuel Evaluation in Spark Ignition Engines. Mechanical Systems and Signal Porcessing. 2001, 13(3): 395~407
10 Konig. Application of Time-Frequency Analysis for Optimum Non-Equidistant Sampling of Automotive Signals Captured at Knock. IEEE Transaction on Instrumentation and Measurements. 1996, 15(3): 2746~2749
11 M.Kancyasu, N.Kurihara, K.Katogi, K.Tabuchi. An Advanced Engine Knock Detection Module Performs Higher Accurate MBT Control and Fuel Consumption Improvement. IEEE Transaction on Instrumentation and Measurements. 1995, 20(3): 11~14
12蒋德明主编.内燃机原理中国农业机械出版社,1981: 238~242
13张振东,王伯年,王祝炜.汽车发动机爆震检测与量化方法,上海理工大学学报. 1998, 12(4): 273~279
14于秀敏,林学东,钱耀义.汽油机爆震信号测量与应用分析.吉林工业大学学报. 2000, 3(2): 31~35
15 Heywood J B, Internal Combustion Engine fundamentals,McGraw-Hill book Company. 1988
16何正嘉,赵纪元,孟庆丰,刘新明.机械检测诊断中非平稳信号的研究西安交通大学学报. 1996, 30(7): 36~42
17杨建国,刘晓峰,林波.基于DWT的汽油机爆震特征提取及爆震强度判定.内燃机学报. 2003, 21(3): 233~238
18虞和济.设备故障诊断技术的现状及发展.基础自动化. 1996: 1~5
19孙中圣.变速箱故障建模及特征提取的研究.哈尔滨工业大学硕士学位论文. 2002: 22~33
20 GT.Zheng, WJ.Wang. A New Campestral Analysis Procedure of Recovering Excitations for Transient Components of Vibration Signals and Applications to Rotating Machinery Condition Monitoring. Journal o f Vibration and Acoustics. 2001, 123(4): 222~228
21 S.Gade, K.Gram-Hansen. The Analysis of Non-stationary Signals. Sound and Vibration. 1997: 40~46
22朱利民,钟秉林,贾民平.振动信号短时分析方法及在机械故障诊断中的应用.机械工程学报. 2005, 13(3): 400~405
23 M .WTrethewey. Window and Overlap Processing Effects on Power Estimates from Spectra. Mechanical System and Signal Processing. 2000, 14(2): 267~278
24崔锦泰.小波分析导论.西安交通大学出版社,1995: 217~219
25陈涛,屈梁生,耿中行.小波分析及其在机械故障诊断中的应用.机械工程学报. 1997, 33(3): 76~79
26张建峰.汽油机爆震特征提取与诊断研究.哈尔滨工业大学硕士学位论文. 2003: 6~23
27李建权.火花点火发动机爆震燃烧的研究.天津大学博士学位文. 1998: 58~70
28王永光.汽油机点火控制的研究.哈尔滨工业大学硕士学位论文. 2004: 13~17
29孙杰.神经网络在汽油机爆震诊断中应用的研究.哈尔滨工业大学硕士学位论文. 2005: 25~35
30杨建国.小波分析及其工程应用.机械工业出版社. 2005: 27~37
31程建康.汽油机点火控制策略分析与研究.哈尔滨工业大学学士学位论文. 2007: 15~20
32潘松,黄继业,王国栋.现代DSP技术. 2003: 163~186
33 Ueda, Antonio Moran. Nonlinear Design Approach to Four Wheel Steering. System Using Neural Network. Vehicle System Network. 1995(24): 156~160
34 ChrisM A.Virtual Sensing: A Neural Network-based Intelligent Performance and Emission Prediction System for On-Board Diagnostics and Engine Control. SAE2005: 98~100
35 A.de Lucas, A.Duran, M.Carmona, M.Lapuerta. Modeling diesel particulate emission with neural networks. Fuel 2003, (80): 539~548
36 S.A.Grossberg. Nonlinear Neural Networks. Principles, and Architectures. Neural Network, 2005, (1): 47~61
37 Vemuri A T, Polycarpou M M. Neural-network-based robust fault diagnosis in robotic systems. I EEE Transactions on Neural Network, 1997, 8(6): 14~17
38贺晓萍. FPGA与单片机串行通信的实现.湖南理工学院学报. 2005: 18~25
39林京.基于连续小波变换的信号检测技术与故障诊断.机械工程学报. 2000, 36(12): 95~100
40 O.Rioul, M.Vetterli. Wavelet and Signal Processing. IEEE Signal Processing Magazine, 1991: 14~38
41 Y.WU, R. DU. Feature Extraction and Assessment Using Wavelet Packets for Monitoring of Machining Processes. Mechanical Systems and Signal Processing. 2006, 10(1): 29~53
42 S.G .Mallat. A Theory of Multi-resolution Signal Decomposition: The Wavelet Representation. IEEE Transaction on Patten Analysis and Machine Intelligence. 1989, 11(7): 674~693
2 Taylor CF著,张胜瑕,程未云等译.内燃机(下册).人民交通出版,1981: 230~235
3李继军.火花点火发动机挤气射流-紊流型燃烧室及基于缸内压力波动检测爆震控制系统研究.西安交通大学博士学位论文. 1995: 36~58
4陈家哗等编.内燃机燃烧.哈尔滨船舶工程学院出版社,1988: 158~162
5 Knock Measurement for Fuel Evaluation in Spark Ignition Engines. Mechanical Systems and Signal Processing. 2001, 13(3): 395~407
6 Giorgio Rizzoni, Xing Cindy Chen. Detection of Internal Combustion Engine Knock Using Time-Frequency Distributions. IEEE Transaction on Instrumentation an Measurements. 1993, 19(3): 360~363
7 A.M.Zoubir. A Comparative Study of Multiple Test Based Techniques for Optimal Sensor Location for Knock Detection. Mechanical Systems and Signal Processing. 2000, 14(3): 353~369
8 M.D.Boland, A.M.Zoubir. Identification of Time-Varying Non-Linear Systems with Application to Knock Detection in Combustion Engines.IE EE Transaction on Instrumentation and Measurements. 1997, 15(3): 799~802
9 C.Hudson, X.Gao, R.Stone. Knock Measurement for Fuel Evaluation in Spark Ignition Engines. Mechanical Systems and Signal Porcessing. 2001, 13(3): 395~407
10 Konig. Application of Time-Frequency Analysis for Optimum Non-Equidistant Sampling of Automotive Signals Captured at Knock. IEEE Transaction on Instrumentation and Measurements. 1996, 15(3): 2746~2749
11 M.Kancyasu, N.Kurihara, K.Katogi, K.Tabuchi. An Advanced Engine Knock Detection Module Performs Higher Accurate MBT Control and Fuel Consumption Improvement. IEEE Transaction on Instrumentation and Measurements. 1995, 20(3): 11~14
12蒋德明主编.内燃机原理中国农业机械出版社,1981: 238~242
13张振东,王伯年,王祝炜.汽车发动机爆震检测与量化方法,上海理工大学学报. 1998, 12(4): 273~279
14于秀敏,林学东,钱耀义.汽油机爆震信号测量与应用分析.吉林工业大学学报. 2000, 3(2): 31~35
15 Heywood J B, Internal Combustion Engine fundamentals,McGraw-Hill book Company. 1988
16何正嘉,赵纪元,孟庆丰,刘新明.机械检测诊断中非平稳信号的研究西安交通大学学报. 1996, 30(7): 36~42
17杨建国,刘晓峰,林波.基于DWT的汽油机爆震特征提取及爆震强度判定.内燃机学报. 2003, 21(3): 233~238
18虞和济.设备故障诊断技术的现状及发展.基础自动化. 1996: 1~5
19孙中圣.变速箱故障建模及特征提取的研究.哈尔滨工业大学硕士学位论文. 2002: 22~33
20 GT.Zheng, WJ.Wang. A New Campestral Analysis Procedure of Recovering Excitations for Transient Components of Vibration Signals and Applications to Rotating Machinery Condition Monitoring. Journal o f Vibration and Acoustics. 2001, 123(4): 222~228
21 S.Gade, K.Gram-Hansen. The Analysis of Non-stationary Signals. Sound and Vibration. 1997: 40~46
22朱利民,钟秉林,贾民平.振动信号短时分析方法及在机械故障诊断中的应用.机械工程学报. 2005, 13(3): 400~405
23 M .WTrethewey. Window and Overlap Processing Effects on Power Estimates from Spectra. Mechanical System and Signal Processing. 2000, 14(2): 267~278
24崔锦泰.小波分析导论.西安交通大学出版社,1995: 217~219
25陈涛,屈梁生,耿中行.小波分析及其在机械故障诊断中的应用.机械工程学报. 1997, 33(3): 76~79
26张建峰.汽油机爆震特征提取与诊断研究.哈尔滨工业大学硕士学位论文. 2003: 6~23
27李建权.火花点火发动机爆震燃烧的研究.天津大学博士学位文. 1998: 58~70
28王永光.汽油机点火控制的研究.哈尔滨工业大学硕士学位论文. 2004: 13~17
29孙杰.神经网络在汽油机爆震诊断中应用的研究.哈尔滨工业大学硕士学位论文. 2005: 25~35
30杨建国.小波分析及其工程应用.机械工业出版社. 2005: 27~37
31程建康.汽油机点火控制策略分析与研究.哈尔滨工业大学学士学位论文. 2007: 15~20
32潘松,黄继业,王国栋.现代DSP技术. 2003: 163~186
33 Ueda, Antonio Moran. Nonlinear Design Approach to Four Wheel Steering. System Using Neural Network. Vehicle System Network. 1995(24): 156~160
34 ChrisM A.Virtual Sensing: A Neural Network-based Intelligent Performance and Emission Prediction System for On-Board Diagnostics and Engine Control. SAE2005: 98~100
35 A.de Lucas, A.Duran, M.Carmona, M.Lapuerta. Modeling diesel particulate emission with neural networks. Fuel 2003, (80): 539~548
36 S.A.Grossberg. Nonlinear Neural Networks. Principles, and Architectures. Neural Network, 2005, (1): 47~61
37 Vemuri A T, Polycarpou M M. Neural-network-based robust fault diagnosis in robotic systems. I EEE Transactions on Neural Network, 1997, 8(6): 14~17
38贺晓萍. FPGA与单片机串行通信的实现.湖南理工学院学报. 2005: 18~25
39林京.基于连续小波变换的信号检测技术与故障诊断.机械工程学报. 2000, 36(12): 95~100
40 O.Rioul, M.Vetterli. Wavelet and Signal Processing. IEEE Signal Processing Magazine, 1991: 14~38
41 Y.WU, R. DU. Feature Extraction and Assessment Using Wavelet Packets for Monitoring of Machining Processes. Mechanical Systems and Signal Processing. 2006, 10(1): 29~53
42 S.G .Mallat. A Theory of Multi-resolution Signal Decomposition: The Wavelet Representation. IEEE Transaction on Patten Analysis and Machine Intelligence. 1989, 11(7): 674~693