Integration of symbolic and connectionist AI techniques in the development of Decision Support Systems applied to biochemical processes

摘要: There exist complex systems, such as bio-chemical plants, which require a constant management to be kept in optimal operating conditions. To this end, automation is the only feasible option: tasks involved, ranging from fault detection diagnosis control, can hardly performed by traditional model-based controllers alone, but instead additional application of artificial intelligence-based techniques. AI vast field, covering many technologies roughly classified “Hard” or “Soft” techniques: former typically use symbolic representation data and elaborate it logic reasoning; latter, instead, process information at sub-symbolic level, exploiting interactions simple elaboration units hence they are also known connectionist Since different techniques more suitable for problems, infrastructure likely include than one module. These modules should able interact order exploit mutual potentialities when processing coming managed system. Moreover, imperfect vague and/or uncertain incomplete so deal with appropriately. In dissertation, we claim not that taking imperfection into account necessary feature an “intelligent” module interfaced real-world system, extending formal logical reasoning allows obtain deeper integration computing simply using them together, cascade parallel, obtaining strongly hybrid simpler yet robust their pure counterparts. will shown deployed effectively within combines concepts service, agent event natural way.