Increasing requirements for a fuel economy, exhaust emissions and the output performance and also the complexity of the automotive engines necessitate the development of a new generation of the engine control functionality. This book offers the solutions of a number of the engine control and estimation problems and consists of ten Chapters grouped in four Parts. Idle speed control and cylinder flow estimation techniques are presented in the first Part of the book; engine torque and friction estimation methods are presented in the second Part.
Engine misfire and Cam Profile Switching diagnostic methods are presented in the third Part; and engine knock detection and control algorithms are discussed in the fourth Part of the book. The algorithms presented in the first Part of the book use a mean value engine model and the techniques described in the rest of the book are based on the cylinder individual engine model. The book provides a sufficiently wide coverage of the engine functionality. The book also offers a tool-kit of new techniques developed by the author which was used for the problems described above.
The techniques can be listed as follows: input estimation, composite adaptation, spline and trigonometric interpolations, a look-up table adaptation and a threshold detection adaptation. These methods can successfully be used for other engine control and estimation applications. These methods are listed in Table 0.1 and Table 0.2 which contain a brief description of the methods, application areas and references providing a reader with the overview and a guidance through the book. One of the key techniques used in this book is the statistical techniques. A periodic nature of the engine rotational dynamics and a cycle-to-cycle variability allows the presentation of the engine signals as statistical signals utilizing such statistical variables as mean values and standard deviations.
The detection of the engine events such as misfire events, knock events and others can be associated with the statistical hypotheses. The statistical hypotheses can in turn be tested via decision making procedures, described in Table 0.3 for example. Two basic types of errors can be made in the statistical tests of the hypotheses called α-risk and β-risk specified by the engineer. The detection performance of the engine events such as misfire events, knock events and others can in turn be associated with these errors, for example with α-risk. The same detection performance (with the same α-risk) can be achieved if the parameters of the signal such as a mean value and a standard deviation involved in the detection of the engine events change due to aging of the engine components, for example.
Many types of the engine event detection problems can be formulated as hypothesis-testing problems aiming to a robust detec tion providing the same detection performance for new and aged engines. A great potential for a robust engine control system design is in a combination of statistical hypotheses and a feedback principle. A control aim can also be associated with the statistical hypothesis and the feedback can be used for either rejection or not rejection of a null hypothesis. A tracking error which a difference between the value of the statistic associated with a hypothesis test and a desired value of the statistic
A desirable hypothesis achieved by a feedback when the tracking error converges to zero, in turn determines desired statistical properties of the closed loop system. For example, the rejection of a null hypothesis in favor of the alternative hypothesis achieved by the engine knock feedback described in Chapter 10 offers a desired statistical separation between a mean value of the maximum amplitude of the knock sensor signal at a given frequency and the threshold value. This in turn, allows the design of a robust engine knock control system with desired α-risk and probability of the knock occurrence.
|If you need to buy the hard copy of this E-Book, Click here|
Download Automotive Engines Control Estimation Statistical Detection By Alexander A Stotsky easily in PDF format for free