Analysis and design for a class of complex control systems part II: fuzzy controller design

Analysis and design for a class of complex control systems part I: fuzzy modelling and identification

This paper proposes a systematic methodology of fuzzy logic modeling for complex system modeling. It has a unified parameterized reasoning formulation, an improved fuzzy clustering algorithm, and an efficient strategy of selecting significant system inputs and their membership functions. The reasoning mechanism introduces 4 parameters whose variation provides a continuous range of inference operation. As a result, we are no longer restricted to standard extremes in any step of reasoning. The fuzzy model itself can then adjust the reasoning process by optimizing the inference parameters based on input-output data. The fuzzy rules are generated through fuzzy c-means (FCM) clustering. Major bottlenecks are addressed and analytical solutions are suggested. We also address the classification process to extend the derived fuzzy partition to the entire output space. In order to select suitable input variables among a finite number of candidates (unlike traditional approaches) we suggest a new strategy through which dominant input parameters are assigned in one step and no iteration process is required. Furthermore, a clustering technique called fuzzy fine clustering is introduced to assign the input membership functions. In order to evaluate the proposed methodology, two examples-a nonlinear function and a gas furnace dynamic procedure-are investigated in detail. The significant improvement of the model is concluded compared to other fuzzy modeling approaches.

Development of a systematic methodology of fuzzy logic modeling

This paper presents a design method for a class of fuzzy control systems. The class of fuzzy systems considered can be represented by the Takagi-Sugeno fuzzy model which is a type of dynamic fuzzy model. A constructive algorithm is developed to obtain the stabilizing feedback control law for the system. The main contribution of this paper is the development of an equivalent principle; that is, the design of a fuzzy control system is equivalent to the design of a set of linear time-invariant ‘extreme’ systems. Thus any design method in linear control system theory can be used to design a fuzzy control system. An example is given to illustrate the application of the method.

Stability analysis and design for a class of continuous-time fuzzy control systems

A function-based evaluation approach is proposed for a systematic study of fuzzy proportional-integral-derivative (PID)-like controllers. This approach is applied for deriving process-independent design guidelines from addressing two issues: simplicity and nonlinearity. To examine the simplicity of fuzzy PID controllers, we conclude that direct-action controllers exhibit simpler design properties than gain-scheduling controllers. Then, we evaluate the inference structures of direct-action controllers in five criteria: control-action composition, input coupling, gain dependency, gain-role change, and rule/parameter growth. Three types of fuzzy PID controllers, using one-, two- and three-input inference structures, are analyzed. The results, according to the criteria, demonstrate some shortcomings in Mamdani's two-input controllers. For keeping the simplicity feature like a linear PID controller, a one-input fuzzy PID controller with "one-to-three" mapping inference engine is recommended. We discuss three evaluation approaches in a nonlinear approximation study: function-estimation-based, generalization-capability-based and nonlinearity-variation-based approximations. The study focuses on the last approach. A nonlinearity evaluation is then performed for several one-input fuzzy PID controllers based on two measures: nonlinearity variation index and linearity approximation index. Using these quantitative indices, one can make a reasonable selection of fuzzy reasoning mechanisms and membership functions without requiring any process information. From the study we observed that the Zadeh-Mamdani's "max-min-gravity" scheme produces the highest score in terms of nonlinearity variations, which is superior to other schemes, such as Mizumoto's "product-sum-gravity" and "Takagi-Sugeno-Kang" schemes.

H. Nakanishi

Papers

A review and comparison of six reasoning methods