ACQUIRING A QUALITATIVE UNDERSTANDING OF SYSTEM BEHAVIOR THROUGH AI INDUCTIVE INFERENCE.
Planning and Operation of Electric Energy Systems, Proceedings of the IFAC Symposium., Rio de Janeiro, Brezilya, ss.35-41, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Basıldığı Şehir: Rio de Janeiro
- Basıldığı Ülke: Brezilya
- Sayfa Sayıları: ss.35-41
- Boğaziçi Üniversitesi Adresli: Evet
Özet
Algorithmic solutions exist for many power system control problems and at first glance it would seem that Artificial Intelligence methodology might not have a role in that problem domain. We argue instead that a plethora of lengthy numerical solutions does not result in useful qualitative understanding of system behavior and indicate how inductive inference can accomplish that. In that process, combined use of symbolic and numerical computing is facilitated through dynamic adaptive quantization of numerical signal levels.