Visualization method for multidimensional random processes

Abstract

The article proposes a method for visualizing multidimensional random process realizations using the example of the concentrations of harmful gases emitted into the atmosphere from a thermal power plant. The method is based on the transformation of gas concentration values in one point of multidimensional space at the same time into a two-dimensional curve, which is described by the sum of products of normalized concentrations by orthogonal Legendre functions of the corresponding order. The combination of such curves on a two-dimensional plane at discrete times creates a characteristic image that can be used to visually detect features of gas concentrations over time by a human operator.