Trend detection based on a fuzzy temporal profile model

A fuzzy temporal profile (FTP) is a model through which we describe the evo lution of a certain physical parameter V over time. Thus we define a set of significant points (X-0, X-1, ..., X-N), and we approximate the evolution curve by way of linear sections between them. Each section is defined by wa y of an imprecise constraint on duration, on increase in value and on slope between the points connected by the section. In this article we show a possible method of matching an FTP with a signal, which will enable the detection of profiles of interest on the trace of a physical parameter over time. (C) 1999 Elsevier Science Ltd. All rights res erved.