Knowledge-guided Genetic Algorithm for input parameter optimisation in environmental modelling

摘要: Abstract The need for input parameter optimisation in environmental modelling is long known. Real-time constraints of disaster propagation predictions require fast and efficient calibration methods to deliver reliable time avoid tragedy. Lately, evolutionary have become popular solve the problem models. Applying a knowledge-guided Genetic Algorithm (GA) we demonstrate how speed up optimsation consequently prediction disasters. Knowledge, obtained from historical synthetical disasters, stored knowledge base provided GA terms chromosome. Despite increased loads knowledge, its retrieval times can be kept near-constant. During mutation, ranges selected parameters are limited forcing explore promising solution areas. Experiments forest fire spread show time-consuming fitness evaluations could lowered remarkably cope with real-time capabilities maintaining error magnitude.