Forecasting Values of Chromaticity of Drinking and Source Waters Using Arima Model and Neural Network

Abstract

Aim. In the present investigation artificial neural network (ANN) and ARIMA-model are compared for forecasting of data of colour of water.Methods. Data corresponds to the colour of water of groundwater and drinking water of water intake of south-east region of the Republic of Belarus. The definition of colour was carried out for the period from 2009 to 2017. twice a day, the time series of values included 5215 values. The parameters of the models were estimated by 85% of the time series values, and the remaining 15% of the values (the test period) compared the forecast values with the actual ones. Optimal configurations of ARIMA-models were determined from the results of comparing the averaged values of the root mean squared errors (RMSE); optimal configurations of ANN were determined from the results of comparing the averaged values of RMSE and correlation coefficients (CC) on the test periods.Results. Comparison of forecasting methods was carried out on the basis of the averaged values of mean absolute error and mean relative error on the test periods. It was revealed that ANN allows to obtain the predicted values of colour of water more accurate than ARIMA-model.Main conclusions. Software implementation of ANN in the MATLAB environment empowers with sufficient accuracy get forecast values of groundwater and drinking water for 100 values.

Ensuring the safety of geotechnical objects based on forecasting the parameters of the water regime of rivers using SARIMA models

“…The results of the analysis of the works of foreign and domestic scientists in the field of water level forecasting and modeling of water dynamics revealed that many researchers in their work use mathematical methods of forecasting hydrological characteristics to possibly reduce their negative impact on the territories where geotechnical objects are located [1][2][3][4]. In their research [1], the authors propose a methodological approach to determining flood characteristics (start and end dates, peak flow and volume) based on the results of using mathematical models of peak threshold exceeding and analysis of the decline curve.…”

“…The authors [2] propose a method for searching for parameters of the seasonal autoregressive integrated moving average (SARIMA) model for predicting the behavior of groundwater time series with insufficient length and data gaps. In [3], Russian researchers compared artificial neural network modeling methods and ARIMA models for predicting water color values, during which it was revealed that artificial neural network models allow obtaining predictive color values with slightly greater accuracy. Researchers of the University of Salerno in [4] for the sustainable management of urban water supply networks, the authors propose to use the method of forecasting the relevant parameters (water level in the reservoir, water demand) using ARIMA models, which demonstrate reliable results at the stages of verification and forecasting.…”