Remember me
 | Home  | Contact us

One of four PNRA scholarships available in 2008/09:

PhD research project at the University of Manchester - application of artificial immune system algorithm to distribution networks

Biological immune systems are highly parallel, distributed, and adaptive systems, which use learning, memory, and associative retrieval to solve recognition and classification tasks. In particular, they learn to recognize relevant patterns, remember previously seen patterns, and use combinatorics to construct pattern detectors efficiently. This remarkable information processing abilities of immune systems provide important aspects in the field of computation. This emerging field is referred to as Immunological Computation, Immunocomputing, or Artificial Immune Systems (AIS).

AIS are capable of constructing and maintaining a dynamical and structural identity, capable of learning to identify previously unseen invaders and remembering what it has learnt. Numerous immune algorithms now exist, based on processes identified within human immune systems. These computational techniques have many potential applications; distributed and adaptive control, machine learning, pattern recognition, fault detection, computer security, optimization, and distributed system design.

A vast array of unexplored application areas exists for AIS, including many problem areas for which traditional artificial intelligence research techniques (neural networks, genetic algorithms, expert systems) were not appropriate.

The complexity of electricity distribution networks, large number of parameters, unavailability or/and uncertainty of many parameters and phenomena involved require application of advanced non-deterministic modeling and computational tools. This project therefore, aims to address the issue of identification of failure prone areas of distribution network through application of AIS and develop an algorithm (and software) for identifying these areas. The objective is to develop a methodology for deciding on optimal design, operation and maintenance of distribution networks in order to reduce network interruptions.