Olga Diamante

A soft computing approach to fuzzy sky-hook control of semiactive suspension

This paper describes a soft computing approach to control a semiactive sky-hook suspension system. The fuzzy controller has been optimized by means of a genetic algorithm and implemented using a dedicated hardware device. After a preliminary study on a quarter car, a seven degree-of-freedom (7-DOF) model was implemented and used for controller simulation and optimization. The proposed control system takes the vehicle heave into account, improving the opposing requirements of comfort and drivability. The following conditions were optimized: drivability over random test roads, sine-wave holes, and sleeper-plates at different speeds. Fuzzy control makes it possible to change the characteristic parameters of the sky-hook suspension in an optimal way according to the different road conditions. The number of fuzzy controller inputs is kept as small as possible, avoiding the use of expensive sensors that increase the overall cost of the system. The proposed system is also easily applicable to existing commercial suspension systems. In addition to the comfort features, the fuzzy controller allows a reduction in load fluctuation, abnormal behavior and tire consumption.

Non-integer order integration by using neural networks

In this paper a new method for non-integer order integration is presented. It is based on a Multi-layer Perceptron modelization that allows an easy circuit design and a wide range of integration order. The neural network approach allows one to overcome some problems related to practical implementation of the circuit. An example of application regarding the control of a chaotic system is also reported in order to validate the proposed neural network modeling technique.