A novel method is presented capable of constructing rule-bases via self-learning for the use of fuzzy controllers. The controlled process is assumed to be a multivariable system with strong interaction within variable...
详细信息
A novel method is presented capable of constructing rule-bases via self-learning for the use of fuzzy controllers. The controlled process is assumed to be a multivariable system with strong interaction within variables and with pure time delays in control. The objective of the proposed system is to build, in the case of two-input two-output systems, two separated and decoupled rule-bases for two control loops with some design requirements. The paper is divided into two parts. In the first part, a system structure comprising four functional modules is proposed. Then, the paper focuses on the issues concerning the learning algorithm. By introducing learning errors, three learning update laws are suggested. Furthermore, the convergence property of the learning algorithms is analysed in the sense of some defined norms. In addition, some comments and remarks about the proposed algorithms are given. The second part of the paper deals mainly with the issues of the methodology for rule-base formation and the application to the problem of multivariable control of blood pressure.
A general orthogonal parameter estimation algorithm is derived to estimate both the structure and the parameters for a wide range of stochastic nonlinear systems which can be described by a nonlinear rational model. S...
A general orthogonal parameter estimation algorithm is derived to estimate both the structure and the parameters for a wide range of stochastic nonlinear systems which can be described by a nonlinear rational model. Simulation studies are included to demonstrate the performance of the algorithm.
Both continuous and discrete time transfer functions of non-linear systems are analysed and interpreted in the frequency domain by investigating the properties and graphical representation of these functions. The cont...
Both continuous and discrete time transfer functions of non-linear systems are analysed and interpreted in the frequency domain by investigating the properties and graphical representation of these functions. The contributions that some typical terms from non-linear time domain models make to the transfer functions is illustrated to provide a better understanding of the frequency response behaviour of complex non-linear dynamic systems.
An approximate solution of the problem of linear-quadratic optimal control of the ground-induced vibrations of a vehicle travelling with variable velocity is presented. A new model in the form of a retarded functional...
详细信息
An approximate solution of the problem of linear-quadratic optimal control of the ground-induced vibrations of a vehicle travelling with variable velocity is presented. A new model in the form of a retarded functional differential equation is established which represents both vehicle dynamics and surface properties. By formulating the model in the spatial domain a number of technical difficulties are removed, and it is seen how constant spatial delays can be accommodated by using rational approximations. Equivalence between the space- and time-domain formulations is established, and an example demonstrating the viability of the approach is discussed.
A good model is essential in simulating and analyzing a system satisfactorily. This paper proposes a new computer-aided modelling methodology to construct parsimonious models for dynamic physical systems. A set of dom...
详细信息
A good model is essential in simulating and analyzing a system satisfactorily. This paper proposes a new computer-aided modelling methodology to construct parsimonious models for dynamic physical systems. A set of domain-independent modelling principles and domain-dependent modelling constraints has been formulated to guide parsimonious model construction. Models and the knowledge base for the enhancement of models have been described by bond graphs. After a parsimonious model is generated, two types of qualitative analysis are conducted;one is to derive unknown states from known states and the other is to predict the effects of parameter changes on the whole system. This methodology is only applicable when the structure of a dynamic system is known beforehand. A detailed case study is presented to demonstrate this approach.
This paper describes an example of how fuzzy systems can intergrate with neural networks and what benefits can be obtained from the combination. In particular, by drawing some equivalence between a simplified fuzzy co...
详细信息
In 1939, the Taylor Instrument Companies introduced a completely redesigned version of its Fulscope pneumatic controller. In addition to proportional and reset control actions, this new instrument provided an action w...
详细信息
In 1939, the Taylor Instrument Companies introduced a completely redesigned version of its Fulscope pneumatic controller. In addition to proportional and reset control actions, this new instrument provided an action which the Taylor Instrument Companies called pre-act. In the same year the Foxboro Instrument Company added Hyper-reset to the proportional and reset control actions provided by their Stabilog pneumatic controller. The pre-act and Hyper-reset actions each provide a control action proportional to the derivative of the error signal. Reset provides a control action proportional to the integral of the error signal and hence both controllers offered PID control. The historical development of these controllers is discussed.< >
Multiobjective genetic algorithms (MOGAs) are introduced as a modification of the standard genetic algorithm at the selection level. Rank-based fitness assignment and the implementation of sharing in the objective val...
详细信息
Multiobjective genetic algorithms (MOGAs) are introduced as a modification of the standard genetic algorithm at the selection level. Rank-based fitness assignment and the implementation of sharing in the objective value domain are two of the important aspects of this class of algorithms. The ability of the decision maker (DM) to progressively articulate its preferences while learning about the problem under consideration is one of their most attractive features. Illustrative results of how the DM can interact with the genetic algorithm are presented. They also show the ability of the MOGA to uniformly sample regions of the trade-off surface.< >
An example of how fuzzy systems can integrate with neural networks and what benefits can be obtained from the combination is described. By drawing some equivalence between a simplified fuzzy control algorithm (SFCA) a...
详细信息
An example of how fuzzy systems can integrate with neural networks and what benefits can be obtained from the combination is described. By drawing some equivalence between a simplified fuzzy control algorithm (SFCA) and radial basis functions (RBF) networks it is concluded that the RBF network can be interpreted in the context of fuzzy systems and can be naturally fuzzified into a class of more general networks, referred to as FBFN. The FBFN is used as multivariable rule-based controller with the ability of self-constructing its own rule-base by incorporating an iterative learning control algorithm into the system. The approach is applied to a problem of multivariable blood pressure control with a FBFN-based controller having six inputs and two outputs, representing a complicated control structure.< >
The traditional approach to multiple parameter optimization in genetic algorithm (GA) practice is to combine the coding of the parameters into a single compound bit-string; the so-called concatenated binary mapping. T...
详细信息
The traditional approach to multiple parameter optimization in genetic algorithm (GA) practice is to combine the coding of the parameters into a single compound bit-string; the so-called concatenated binary mapping. This approach has some shortcomings; the GA is a competition-based technique that has a natural tendency to evolve one winner which in complex problems yields a solution that is better on some parameters than the others. An extension to the simple GA, called vector evaluated genetic algorithm (VEGA), has been used in multiobjective optimization where one is not interested in a single solution, but a family of optimal solutions. In VEGA each member of the population is evaluated and assigned a weighted fitness value dependent on how it relates to each objective criteria. The reproduction plan then develops groupings within the populations for each of the objectives to be optimized, ensuring that the improvement of one objective does not adversely affect the others. This, however, requires large population sizes and can be quite inefficient. In cases where the complex task is divisible into simpler optimization problems, a better solution set may be obtained using parallel genetic algorithms to search for the optimal solution to each sub-problem.< >
暂无评论