scholarly journals Dynamic Water-Level Neural-Network Forecast Model on Non-Stationary Time Series

2002 ◽  
Vol 14 (1) ◽  
pp. 19-24
Author(s):  
XUE Lianqing ◽  
◽  
CUI Guangbai ◽  
CHEN Kaiqi
2020 ◽  
Vol 3 (1) ◽  
pp. 362-372
Author(s):  
Svitlana Antoshchuk ◽  
Oksana Babilunha ◽  
Thanh Tran Kim ◽  
Anatolii Nikolenko ◽  
Tien Nguyen Thi Khanh

Author(s):  
Wei Ming Wong ◽  
Mohamad Yusry Lee ◽  
Amierul Syazrul Azman ◽  
Lew Ai Fen Rose

The aim of this study is to use the Box-Jenkins method to build a flood forecast model by analysing real-time flood parameters for Pengkalan Rama, Melaka river, hereafter known as Sungai Melaka. The time series was tested for stationarity using the Augmented Dickey-Fuller (ADF) and differencing method to render a non-stationary time series stationary from 1 July 2020 at 12:00am to 30th July 2020. A utocorrelation (ACF) and partial autocorrelation (PACF) functions was measured and observed using visual observation to identify the suitable model for water level time series. The parameter Akaike Information Information Criterion (AIC) and the Bayesian Information Criterion (BIC) were used to find the best ARIMA model (BIC). ARIMA (2, 1, 3) was the best ARIMA model for the Pengkalan Rama, with an AIC of 5653.7004 and a BIC of 5695.209. The ARIMA (2, 1, 3) model was used to produce a lead forecast of up to 7 hours for the time series. The model's accuracy was tested by comparing the original and forecast sequences by using Pearson r and R squared. The ARIMA model appears to be adequate for Sungai Melaka, according to the findings of this study. Finally, the ARIMA model provides an appropriate short-term water level forecast with a lead forecast of up to 7 hours. As a result, the ARIMA model is undeniably ideal for river flooding.


Author(s):  
А.И. Епихин ◽  
С.И. Кондратьев ◽  
Е.В. Хекерт

В статье рассмотрены особенности прогнозирования многомерных нестационарных временных рядов с использованием нейромоделирования. Для построения прогнозных моделей используются самые современные концепции математического моделирования: методы фрактального и интеллектуального анализа эволюции систем, инструментарий нечеткой логики и искусственных нейронных сетей. В качестве примера рассмотрен процесс подачи топлива с примесями водорода в быстроходных дизелях на кораблях. Рассмотрен выбор топологии искусственной нейронной сети. Принимая во внимание обозначенную задачу исследования и особенности работы СЭУ, представляется целесообразным использовать инструменты нейромоделирования, которые позволяют реализовать новые подходы к прогнозированию динамики многомерных нестационарных временных рядов, происходящих в двигателе. Рассмотрен принцип формирования многомерной базы знаний для нейронной модели с использованием комбинации ключевых концептов теории нечеткой математики и понятия фрактальной размерности, сформированного в рамках множества Мандельброта. Рассмотрены способы обучения нейросети. The article discusses the features of forecasting multidimensional non-stationary time series using neuromodeling. To build predictive models, the most modern concepts of mathematical modeling are used: methods of fractal and intellectual analysis of the evolution of systems, tools of fuzzy logic and artificial neural networks. As an example, the process of supplying fuel with hydrogen impurities in high-speed diesel engines on ships is considered. The choice of the topology of an artificial neural network is considered. Taking into account the designated research task and the features of the SEP operation, it seems appropriate to use neuro-modeling tools that allow implementing new approaches to predicting the dynamics of multidimensional non-stationary time series occurring in the engine. The principle of formation of a multidimensional knowledge base for a neural model using a combination of key concepts of the theory of fuzzy mathematics and the concept of fractal dimension, formed within the framework of the Mandelbrot set, is considered. Methods of training a neural network are considered.


Sign in / Sign up

Export Citation Format

Share Document