Mostrando ítems 1-5 de 5

A time-series forecasting performance comparison for neural networks with state space and ARIMA models

Soto-Ferrari, Milton; Chams-Anturi, Odette; Escorcia-Caballero, Juan P. (IEOM Society InternationalUnited States, 2020)

This research focuses on the development of an automated forecasting procedure that implement State Space (SS), Auto Regressive Integrated Moving Average (ARIMA), and Neural Networks (NN) to identify the best forecasting ...
Analyzing and predicting power consumption profiles using big data

amelec, viloria; Prieto Pulido, Ronald Antonio; García Guiliany, Jesús; Martínez Ventura, Jairo; Hernández Palma, Hugo; Jinete Torres, José; REDONDO BILBAO, OSMAN ENRIQUE; Pineda Lezam, Omar Bonerge (Communications in Computer and Information Science, 2019)

The Euclidean distance (ED), the mean absolute error (MAE), the mean absolute percentage error (MAPE) and the root of the mean quadratic error (RMQE) are used to evaluate the predictive capability of the models supported ...
Caracterización del estado actual de la generación de seguridad en la región caribe colombiana

Figueredo Barrios, Kineereth Juliana; Simmonds Mendoza, Armando Javier (Corporación Universidad de la CostaIngeniería Eléctrica, 2020)

Security Generation (SG) is defined by CREG 015 of 2018 as the imposed generation required to supply the restrictions of the National Interconnected System (SIN) and operational areas, these restrictions are associated to ...
Evaluation of bottom-up and top-down strategies for aggregated forecasts: state space models and arima applications

Soto-Ferrari, Milton; Chams-Anturi, Odette; Escorcia-Caballero, Juan P.; Hussain, Namra; Khan, Muhammad (Universidad de la Costa, 2019-09-20)

Abstract. In this research, we consider monthly series from the M4 competition to study the relative performance of top-down and bottom-up strategies by means of implementing forecast automation of state space and ARIMA ...
Wind speed prediction based on univariate ARIMA and OLS on the Colombian Caribbean Coast

Palomino, Kevin; Reyes, Fabiola; Núñez, José; Valencia, Guillermo; Herrera Acosta, Roberto (Journal of Engineering Science and Technology Review, 2020)

Greater incorporation of wind energy into power systems has necessitated the development of accurate and reliable techniques for wind speed forecasting. However, although there are multiple studies, none are set up for the ...

Envíos recientes

A time-series forecasting performance comparison for neural networks with state space and ARIMA models
...
Soto-Ferrari, Milton | 2020

This research focuses on the development of an automated forecasting procedure that implement State Space (SS), Auto Regressive Integrated Moving Average (ARIMA), and Neural Networks (NN) to identify the best forecasting strategy for time series with numerous patterns. The proposed approach is applied on multiple time series exhibiting different series patterns from the M4 competition. Based on our study, the performance of ARIMA models showed superior results when compared to the ETS performance for seasonal data. In addition, NN and ARIMA showed a higher performance for cyclical and non-stationary data. NN performance was competitive in all types of data patterns. ARIMA stepwise selection procedure proved to be the most accurate in general for all the series. This delimited development is part of a comprehensive application that will encompass a dashboard tool designed to automatize forecasting procedures of different types of time series presented in the industry Keywords

LEER
Analyzing and predicting power consumption profiles using big data
...
amelec, viloria | 2019

The Euclidean distance (ED), the mean absolute error (MAE), the mean absolute percentage error (MAPE) and the root of the mean quadratic error (RMQE) are used to evaluate the predictive capability of the models supported by each statistical method, asserting, according to the assessment, that the best predictions come from the ARIMA method. This paper presents a prediction study for two buildings located at the University of Mumbai in India, in order to determine a method that fits the forecasts of organization expenses

LEER
Caracterización del estado actual de la generación de seguridad en la región caribe colombiana
...
Figueredo Barrios, Kineereth Juliana | 2020

Security Generation (SG) is defined by CREG 015 of 2018 as the imposed generation required to supply the restrictions of the National Interconnected System (SIN) and operational areas, these restrictions are associated to the assets that make use of the SIN, regional interconnection and local distribution, which are classified as technical and operational restrictions (Resolución CREG 015 del 2018, 2018). Currently, different methods are used to forecast the SG that may be presented in the future taking into account the restrictions presented by the distribution networks with the purpose of creating alternatives to reduce the charges caused by the same. Based on the above, the objective of this work is to propose a prioritization scheme for the projects of the national expansion plan 2019 - 2032 to have an impact on the reduction of SG in the Caribbean Region of Colombia. Initially, a sweep was made of the technical and operational guidelines of Security Generation, the regulations in force to date were consulted, the generation forecast models were identified, as well as the studies aimed at the Security Generation forecast taking into account the restrictions in the network. Then, data was downloaded from XM's website from 2018 to 2020 in order to carry out a Security Generation forecast using ARIMA models As a result, a new linear model is obtained. The Colombian Caribbean Region was considered as a case study. Finally, the generation and transmission projects proposed by the UPME were identified, with the purpose of prioritizing those that can contribute the most to the reduction of Safety Generation in the Colombian Caribbean Region.

LEER
Evaluation of bottom-up and top-down strategies for aggregated forecasts: state space models and arima applications
...
Soto-Ferrari, Milton | 2019-09-20

Abstract. In this research, we consider monthly series from the M4 competition to study the relative performance of top-down and bottom-up strategies by means of implementing forecast automation of state space and ARIMA models. For the bottomup strategy, the forecast for each series is developed individually and then these are combined to produce a cumulative forecast of the aggregated series. For the top-down strategy, the series or components values are first combined and then a single forecast is determined for the aggregated series. Based on our implementation, state space models showed a higher forecast performance when a top-down strategy is applied. ARIMA models had a higher forecast performance for the bottom-up strategy. For state space models the top-down strategy reduced the overall error significantly. ARIMA models showed to be more accurate when forecasts are first determined individually. As part of the development we also proposed an approach to improve the forecasting procedure of aggregation strategies.

LEER
Wind speed prediction based on univariate ARIMA and OLS on the Colombian Caribbean Coast
...
Palomino, Kevin | 2020

Greater incorporation of wind energy into power systems has necessitated the development of accurate and reliable techniques for wind speed forecasting. However, although there are multiple studies, none are set up for the Colombia Caribbean coast. This is a disadvantage because the potential of wind resources in this region is greater than the hydroelectric potential of the whole country, but all this potential has yet to be developed. In this paper, based on time series, Autoregressive Integrated Moving Average (ARIMA), and Multiple Regression with Ordinary Least Squares (OLS) in the study, two models are proposed and their performance for wind speed prediction is compared. The data were collected in the meteorological station located in the experimental farm of the Atlantic University, in Barranquilla, Colombia, and variables analyzed included wind speed, wind direction, temperature, relative humidity, solar radiation, and pressure. The results of the two approaches indicated that among all the involved models, the ARIMA model has the best predicting performance. Also, it is essential to highlight that through this work, decision-makers would explore the local wind potential, allowing for the possibility of predicting future wind speed, and thus giving them the ability to plan the production and the interaction of other sources of energy.

LEER

Listar por tema "ARIMA"

A time-series forecasting performance comparison for neural networks with state space and ARIMA models ﻿

Analyzing and predicting power consumption profiles using big data ﻿

Caracterización del estado actual de la generación de seguridad en la región caribe colombiana ﻿

Evaluation of bottom-up and top-down strategies for aggregated forecasts: state space models and arima applications ﻿

Wind speed prediction based on univariate ARIMA and OLS on the Colombian Caribbean Coast ﻿

Envíos recientes

A time-series forecasting performance comparison for neural networks with state space and ARIMA models

Analyzing and predicting power consumption profiles using big data

Caracterización del estado actual de la generación de seguridad en la región caribe colombiana

Evaluation of bottom-up and top-down strategies for aggregated forecasts: state space models and arima applications

Wind speed prediction based on univariate ARIMA and OLS on the Colombian Caribbean Coast

A time-series forecasting performance comparison for neural networks with state space and ARIMA models

Analyzing and predicting power consumption profiles using big data

Caracterización del estado actual de la generación de seguridad en la región caribe colombiana

Evaluation of bottom-up and top-down strategies for aggregated forecasts: state space models and arima applications

Wind speed prediction based on univariate ARIMA and OLS on the Colombian Caribbean Coast