A SOM-based hierarchical model to short-term load forecasting

Carpinteiro, O.A.S.; Reis, Agnaldo J. Rocha

doi:10.1109/ptc.2005.4524693

Cited by 7 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lately, the efforts concentrate on unsupervised learning neural networks such as Self Organizing Maps (SOMs) [15], [16], on Neural Networks [17] or on hybrid systems that combine both Self Organizing maps (SOMs) and algorithms, such as support vector machines (SVMs) [18] or k-Nearest Neighbors (kNN) [5]. In case of microgrids, however, we need to focus on short and very short-term forecasting.…”

Section: Background and Literature Review On Machine Learning Formentioning

confidence: 99%

Machine Learning Platform for Profiling and Forecasting at Microgrid Level

Mele

Elias

Ktena

2019

Electrical, Control and Communication Engineering

View full text Add to dashboard Cite

The shift towards distributed generation and microgrids has renewed the interest in forecasting algorithms and methods, which need to take into account the advances in information, metering and control technologies in order to address the challenges of forecasting problems. Technologies such as machine learning have been proven useful for short-term electricity load forecasting, especially for microgrids, as they can also take into account several types of historical data and can adapt to changes often encountered in small-scale systems and on a short time scale. In this paper, we present a flexible and easily customized modular toolbox, called Divinus, for electricity use profiling and forecasting in microgrids. Divinus may support a variety of machine learning algorithms for forecasting and profiling that can be used independently or combined. For demonstration purposes, we have implemented Self-Organizing Maps for profiling and k-Neighbors for forecasting. The testing of the platform was based on electricity consumption data of the Euripus campus of the National and Kapodistrian University of Athens in Evia, Greece, from January 2010 till March 2018. The tests that have been carried out so far show that the platform can be easily customized and the algorithms examined yield high accuracy and acceptable mean errors for the case of a university campus energy profile.

show abstract

Section: Background and Literature Review On Machine Learning Formentioning

confidence: 99%

Machine Learning Platform for Profiling and Forecasting at Microgrid Level

Mele

Elias

Ktena

2019

Electrical, Control and Communication Engineering

View full text Add to dashboard Cite

show abstract

“…In the Brazilian context, in [5] a semi-clustering approach based on SOM is used to forecast short-term load patterns (hourly granularity) in a not-mentioned Brazilian electric utility. While in [10] a step by step approach is presented to clustering load curves for an electric utility in the state of Maranhão.…”

Section: Related Workmentioning

confidence: 99%

Discovering and Labelling of Temporal Granularity Patterns in Electric Power Demand With a Brazilian Case Study

Servidone

Conti

2016

Pesqui. Oper.

View full text Add to dashboard Cite

Clustering is commonly used to group data in order to represent the behaviour of a system as accurately as possible by obtaining patterns and profiles. In this paper, clustering is applied with partitioning-clustering techniques, specifically, Partitioning around Medoids (PAM) to analyse load curves from a city of Southeastern Brazil in São Paulo state. A top-down approach in time granularity is performed to detect and to label profiles which could be affected by seasonal trends and daily/hourly time blocks. Timegranularity patterns are useful to support the improvement of activities related to distribution, transmission and scheduling of energy supply. Results indicated four main patterns which were post-processed in hourly blocks by using shades of grey to help final-user to understand demand thresholds according to the meaning of dark grey, light grey and white colours. A particular and different behaviour of load curve was identified for the studied city if it is compared to the classical behaviour of urban cities.

show abstract

“…Self-Organizing Maps (SOM) have also been used in different ways in the past; for example, Marín et al [21] apply SOM for the classification of historical data; Joya et al [22] use SOM for forecasting the load and analysis of contingencies; Mori and Itagaki [23] use SOM for the classification of data and combines it with Radial Basis Function Network (RBFN); Carpinteiro and Reis [24] apply two SOM in cascade; Wang [25] uses SOM and fuzzy for forecasting; Fan and Chen [26] show two different stages in the forecasting process: a first stage with SOM is used to classify and in the second, 24 Support Vector Machine (SVM) are applied to every group. In Farhadi and Tafreshi [27], SOM is used to classify normal and abnormal days, and a MultiLayer Perceptron (MLP) to manage temperature data.…”

Section: Related Workmentioning

confidence: 99%

Artificial neural networks for short-term load forecasting in microgrids environment

Hernández

Baladrón

Aguiar

et al. 2014

Energy

194

View full text Add to dashboard Cite

The adaptation of energy production to demand has been traditionally very important for utilities in order to optimize resource consumption. This is especially true also in microgrids where many intelligent elements have to adapt their behaviour depending on the future generation and consumption conditions. However, traditional forecasting has been performed only for extremely large areas, such as nations and regions. This work aims at presenting a solution for Short-Term Load Forecasting (STLF) in microgrids, based on a three-stage architecture which starts with pattern recognition by a Self-Organizing Map (SOM), a clustering of the previous partition via kmeans algorithm, and finally demand forecasting for each cluster with a MultiLayer Perceptron.Model validation was performed with data from a microgrid-sized environment provided by the Spanish company Iberdrola.

show abstract

A SOM-based hierarchical model to short-term load forecasting

Cited by 7 publications

References 12 publications

Machine Learning Platform for Profiling and Forecasting at Microgrid Level

Machine Learning Platform for Profiling and Forecasting at Microgrid Level

Discovering and Labelling of Temporal Granularity Patterns in Electric Power Demand With a Brazilian Case Study

Artificial neural networks for short-term load forecasting in microgrids environment

Contact Info

Product

Resources

About