Learning internal representations by error propagation
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
An introduction to fuzzy control
An introduction to fuzzy control
System identification (2nd ed.): theory for the user
System identification (2nd ed.): theory for the user
| Hi-index | 0.00 |
Evolutionary computing is widely used to tune intelligent systems which incorporate expert knowledge with data. The linguistic equation (LE) approach is an efficient technique for developing truly adaptive, yet understandable, systems for highly complex applications. Process insight is maintained, while data-driven tuning relates the measurements to the operating areas. Genetic algorithms are well suited for LE models based on nonlinear scaling and linear interactions. New parameter definitions have been developed for the scaling functions to handle efficiently the parameter constraints of the monotonously increasing second order polynomials. While identification approaches are used to define the model structures of the dynamic models. Cascade models, effective delays and working point models are also represented with LE models, i.e. the whole system is configured with a set of parameters. Results show that the efficiency of the systems improves considerably after the implementation of simultaneous tuning of all parameters.