Neuro-fuzzy control of industrial systems with actuator by F. L. Lewis, J. Campos, R. Selmic

By F. L. Lewis, J. Campos, R. Selmic

Neural networks and fuzzy platforms are version unfastened keep watch over layout ways that signify a bonus over classical regulate whilst facing advanced nonlinear actuator dynamics. Neuro-Fuzzy keep watch over of commercial platforms with Actuator Nonlinearities brings neural networks and fuzzy common sense including dynamical regulate structures. every one bankruptcy provides strong keep an eye on techniques for the layout of clever controllers to make amends for actuator nonlinearities similar to time hold up, friction, deadzone, and backlash that may be present in all business movement platforms, plus a radical improvement, rigorous balance proofs, and simulation examples for every layout. within the ultimate bankruptcy, the authors enhance a framework to enforce clever regulate schemes on genuine structures.

Rigorous balance proofs are extra proven through machine simulations, and appendices include the pc code had to construct clever controllers for real-time purposes. Neural networks seize the parallel processing and studying services of organic apprehensive platforms, and fuzzy common sense captures the decision-making functions of human linguistics and cognitive platforms.

Audience This e-book is written for college kids in a school curriculum, for training engineers in undefined, and for college researchers.

Show description

Read or Download Neuro-fuzzy control of industrial systems with actuator nonlinearities PDF

Similar control systems books

Modeling and Control of Complex Physical Systems: The Port-Hamiltonian Approach

Strength trade is an incredible origin of the dynamics of actual structures, and, therefore, within the examine of advanced multi-domain platforms, methodologies that explicitly describe the topology of power exchanges are instrumental in structuring the modeling and the computation of the system's dynamics and its keep watch over.

Intelligent Diagnosis and Prognosis of Industrial Networked Systems (Automation and Control Engineering)

In an period of in depth pageant the place plant working efficiencies has to be maximized, downtime as a result of equipment failure has develop into extra high priced. to chop working expenditures and bring up sales, industries have an pressing have to expect fault development and closing lifespan of business machines, methods, and platforms.

Fault Detection and Diagnosis in Engineering Systems

That includes a model-based method of fault detection and prognosis in engineering structures, this ebook includes up to date, functional info on combating product deterioration, functionality degradation and significant equipment harm. ;College or college bookstores might order 5 or extra copies at a different scholar rate.

Additional resources for Neuro-fuzzy control of industrial systems with actuator nonlinearities

Example text

However, sampling of nonlinear systems is not an easy topic. In fact, the exact discretization of nonlinear continuous dynamics is based on the Lie derivatives and leads to an infinite series representation. Various approximate discretization techniques use truncated versions of the exact series. , a shift register. Each delay element stores information and requires an initial condition. 2. Discrete-time single-input Brunovsky form. 2 Mathematics and system properties Mathematical background We denote any suitable norm by .

Neurocontroller for friction compensation 47 Let the NN weight tuning be provided by with any constant matrices S = ST > 0, T = TT > 0, E = ET > 0, and k > 0 a small scalar design parameter. 38). 25). Proof. 28), one has Using Property 3 and the tuning rules yields Define Proceeding as in Lewis, Yesildirek, and Liu (1996), one obtains that L is negative as long as either or Therefore, L is negative outside a compact set and \\r\\ and ||Z||f are UUB as long as the control remains valid within this set.

In practical situations, we do not care whether the NN weights converge to the actual ideal weights that offer good approximation of the unknown robot function f ( x ) . The theorem shows that in a very practical sense, the tracking error r(t) is nevertheless bounded to a small value. The tuning parameter k offers a design tradeoff between the relative eventual magnitudes of ||r|| and \\Z\\ F. , size L) of the NN. , more hidden units), the NN estimation error e# decreases, so the bounding constant Co will decrease, resulting in smaller tracking errors.

Download PDF sample

Rated 4.14 of 5 – based on 4 votes