By Steven X. Ding
Data-driven layout of Fault prognosis and Fault-tolerant regulate platforms offers easy statistical procedure tracking, fault prognosis, and keep an eye on tools and introduces complicated data-driven schemes for the layout of fault prognosis and fault-tolerant keep an eye on structures catering to the wishes of dynamic commercial approaches. With ever expanding calls for for reliability, availability and security in technical techniques and resources, approach tracking and fault-tolerance became very important concerns surrounding the layout of computerized keep an eye on platforms. this article indicates the reader how, due to the swift improvement of data know-how, key suggestions of data-driven and statistical technique tracking and regulate can now develop into regular in commercial perform to handle those concerns. to permit for self-contained research and facilitate implementation in actual functions, very important mathematical and regulate theoretical wisdom and instruments are incorporated during this ebook. significant schemes are offered in set of rules shape and tested on commercial case structures. Data-driven layout of Fault prognosis and Fault-tolerant keep an eye on platforms should be of curiosity to approach and regulate engineers, engineering scholars and researchers with a keep an eye on engineering background.
Read or Download Data-driven Design of Fault Diagnosis and Fault-tolerant Control Systems PDF
Best products books
The Newnes are aware of it All sequence takes the simplest of what our authors have written over the last few years and creates a one-stop reference for engineers thinking about markets from communications to embedded platforms and all over the place in among. PIC layout and improvement a average healthy for this reference sequence because it is without doubt one of the preferred microcontrollers on the planet and we now have numerous fantastically authored books at the topic.
SCM is among the most popular themes in production and distribution, and prefer JIT and TQC it calls for a company dedication. This booklet offers either primary rules of SCM in addition to a suite of directions to help in functional software of SCM. it is going to be one of many first books out there that bargains solely with SCM and its program.
- Finnie's Notes on Fracture Mechanics: Fundamental and Practical Lessons
- Introduction to Digital Electronics
- Tables for ascertaining the Density of American Mineral-Oils and their products by means of the Thermo-Areometer
- Advanced Design Techniques for RF Power Amplifiers (Analog Circuits and Signal Processing)
- System-on-Chip Test Architectures. Nanometer Design for Testability
Extra info for Data-driven Design of Fault Diagnosis and Fault-tolerant Control Systems
1 Some Elementary Concepts 25 hypothesis Ho is then tested against an alternative hypothesis H1 . In this context, the fault detection problem described in Sect. 5) f ⇒= 0, faulty. 6) The purpose of applying hypothesis testing is to determine whether the sampling (the online process data) supports the rejection of the null hypothesis. In other words, it is to detect if f ⇒= 0. In order to make a reliable decision for or against the rejection of the null hypothesis, that means a decision for fault-free or faulty, a function of the samples will be defined, which is also a random variable.
For this purpose, 22 training data sets have been generated, which include 21 faulty and the normal operating conditions. These data sets have been available from the Internet and widely adopted by the control and fault diagnosis communities, in particular as a benchmark process for the comparison studied, for instance the work reported in . Ricker has developed a Simulink TEP simulator and presented a control scheme applied to the TEP in . The simulator can be downloaded at his website. As mentioned previously, a further motivation for introducing these three application cases is that they will serve as application examples for illustrating and demonstrating the results of our study in the forthcoming chapters.
26) i=1 are sample mean vector and covariance matrix of y and for N → ◦ lim y¯ N = E(y) = μ and lim S N −1 = V ar (y) = Σ. 2 for Jth,T 2 • Online fault detection using the algorithm given below. 3 Online fault detection algorithm S1: Collect (online) measurement data yk+i , i = 1, . . , n, and calculate y¯ = 1 n n yk+i , λy¯ = y¯ − y¯ N i=1 S2: (Online) calculate test statistic ⎦ J = n λy¯ T S N−1−1 λy¯ S3: Make a decision ⎤ J ≤ Jth,T 2 =∈ fault-free . 9) with unknown E(y), V ar (y) = Σ. Assume that there are two groups of process data available: training data (recorded for the offline computation) yi , i = 1, .
Data-driven Design of Fault Diagnosis and Fault-tolerant Control Systems by Steven X. Ding