РАЗРАБОТКА НЕЧЕТКО-ЛОГИЧЕСКОГО КЛАССИФИКАТОРА ФАЗОВОГО СОСТОЯНИЯ ГИДРОМЕТЕОРОВ ДЛЯ АЛГОРИТМА КЛАССИФИКАЦИИ ЗОН ВЕРОЯТНОГО ОБЛЕДЕНЕНИЯ ВОЗДУШНЫХ СУДОВ В МЕТЕОЛОКАТОРАХ X-ДИАПАЗОНА

Abstract

Icing is a dangerous meteorological phenomenon for aircraft and is a process of ice deposition on the external elements of the aircraft structure. In real conditions, the aircraft crew is not always able to quickly recognize the gradual deterioration of controllability, which usually occurs over a long period of time due to icing. This fact leads to emergencies. To prevent such situations, it is necessary to have timely and reliable information about the existing supercooled hydrometeors on the aircraft flight path, which will allow the crew to take preventive measures in a timely manner. The zones of probable icing are characterized by the presence of a large volume of super-cooled water in the liquid state. In the radar range, the liquid volume is characterized by the reflectivity of the meteorological phenomenon, and the phase state of water is characterized by the reflectivity and additional characteristics available only for radar systems with polarization processing of reflected signals. At the moment, there is no functionality for detecting supercooled liquid in the meteorological radars of the near airfield zone. The decision on the presence of icing can be made only on the basis of indirect signs and predictive models that do not have the necessary degree of reliability. This article proposes to use a radar that provides a full polarization scattering matrix of hydrometeors, which makes it possible to relate the characteristics of reflected signals to such parameters of hydrometeors as their size, shape, spatial orientation and dielectric state, from which one can proceed to practically important meteorological characteristics, including such, which cannot be obtained without taking polarization into account. Hence, the problem of detecting supercooled liquid water can be reduced to the problem of classifying hydrometeors. To solve this problem, the article pro poses a fuzzy-logical algorithm for the classification of hydrometeors specific to the phenomenon of aircraft icing and performed its initial training using simulation data. The effectiveness of the proposed classification algorithm has been analyzed. It is shown that the developed algorithm ensures the justification of the classification when observing hydrometeors in the case of dangerous icing of the aircraft of the order of 98%.