RELATIONSHIPS BETWEEN FILTER-EXTRAPOLATOR PARAMETERS AND QUALITY OF FILTRATION AND FORECAST

Introduction. In the article I analyzed the relationships between parameters of the adaptive exponential smoothing filter and quality of filtration and forecast. Adaptation principle of the exponential filter is based on use of the method of least squares. Aims. To analyze the quality of the algorithm’s adaptation and define the conditions in which it can workappropriately. To get the information that describes connection between data processing parameters and quality of filtration and forecast. Methodology. I have applied concepts of time series analysis and mathematical simulation in Matlab package. Results. I have obtained approximate values of different parameters that describe conditions of the system or device in which this filtering and forecasting algorithm can be integrated. I assessed the quality of smoothing factor adaptation method establishing relationships between parameters. Originality. For the first time I have defined the relationships between rms-errors (filtration and forecast) and the following parameters: the number of steps which the forecast is made for, the quantity of steps which the data processing algorithm uses for quality estimation of filtering process, the value that defines acceptable error and smoothing factor initial value. I also got the information that describes the connection between some of the tparameters mentioned above. Practical value. I have built a structure of the data processing algorithm that can be integrated into different automated control systems. This research gives an opportunity to choose appropriate parameters of the filtering and forecasting algorithm that will give an ability to filter and make a prediction of a signal in channels of measurement and control channels. Proposed data processing algorithm can be implemented as a program for a microcontroller.