Electricity Price and Demand Forecasting in Smart Grids

Motamedi, A.; Zareipour, Hamidreza; Rosehart, William

doi:10.1109/tsg.2011.2171046

Cited by 141 publications

(70 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although high precision, as low as 1.97% mean absolute percentage error (MAPE), has been achieved on large scale (for example national and municipal level (Amjady, 2007;Beccali et al, 2004;Motamedi et al, 2012;Taylor and Mcsharry, 2008)), microgrid, VPP and transformer level forecasting has only recently emerged as a research interest (Amjady and Keynia, 2010;Fatimie et al, 2010;Hernandez et al, 2014;Llanos et al, 2012;Lloret and Valencia, 2013). The results are not very encouraging, with errors ranging from and 5.15% MAPE at university campus level (Fatimie et al, 2010), where power demand peaks at 8 MW during the day, and 7.92% MAPE at university building level (Borges et al, 2011) -up to 13.8% MAPE at village level, where power demand peaks at 15 kW (Llanos et al, 2012).…”

Section: Power Demand Forecasting In Smart Gridsmentioning

confidence: 99%

Prediction-Based Multi-Agent Reinforcement Learning in Inherently Non-Stationary Environments

Marinescu

Dusparić

Clarke

2017

ACM Trans. Auton. Adapt. Syst.

View full text Add to dashboard Cite

Multi-agent reinforcement learning (MARL) is a widely researched technique for decentralised control in complex large-scale autonomous systems. Such systems often operate in environments that are continuously evolving and where agents’ actions are non-deterministic, so called inherently non-stationary environments. When there are inconsistent results for agents acting on such an environment, learning and adapting is challenging. In this article, we propose P-MARL, an approach that integrates prediction and pattern change detection abilities into MARL and thus minimises the effect of non-stationarity in the environment. The environment is modelled as a time-series, with future estimates provided using prediction techniques. Learning is based on the predicted environment behaviour, with agents employing this knowledge to improve their performance in realtime. We illustrate P-MARL’s performance in a real-world smart grid scenario, where the environment is heavily influenced by non-stationary power demand patterns from residential consumers. We evaluate P-MARL in three different situations, where agents’ action decisions are independent, simultaneous, and sequential. Results show that all methods outperform traditional MARL, with sequential P-MARL achieving best results.

show abstract

Section: Power Demand Forecasting In Smart Gridsmentioning

confidence: 99%

Prediction-Based Multi-Agent Reinforcement Learning in Inherently Non-Stationary Environments

Marinescu

Dusparić

Clarke

2017

ACM Trans. Auton. Adapt. Syst.

View full text Add to dashboard Cite

show abstract

“…Other diverse approaches for electricity price and load forecasting include Self-Organizing Map (SOM) [34], hybrid Principal Component Analysis (PCA) [35], Data Association Mining (DAM) [36], the Bayesian Method [37], Fuzzy Inference [38], Multiple Regression [36], Kernel Machine [39], Neural Networks [32,40], Particle Swarm Optimization (PSO) [41], etc.…”

Section: Related Workmentioning

confidence: 99%

Ensemble Prediction Model with Expert Selection for Electricity Price Forecasting

Neupane¹,

Woon²,

Aung³

2016

Preprint

View full text Add to dashboard Cite

Day-ahead forecasting of electricity prices is important in deregulated electricity markets for all the stakeholders: energy wholesalers, traders, retailers, and consumers. Electricity price forecasting is an inherently difficult problem due to its special characteristic of dynamicity and non-stationarity. In this paper, we present a robust price forecasting mechanism that shows resilience towards aggregate demand response effect and provides highly accurate forecasted electricity prices to the stakeholders in a dynamic environment. We employ an ensemble prediction model in which a group of different algorithms participate in predicting the price for each hour of a day. We propose two different strategies, namely, Fixed Weight Method (FWM) and Varying Weight Method (VWM), for selecting each hour's expert algorithm from the set of participating algorithms. In addition, we utilize a carefully engineered set of features selected from a pool of features derived from information such as past electricity price data, weather data, and calendar data. The proposed ensemble model offers better results than both the Pattern Sequence-based Forecasting (PSF) method and our own previous work using Artificial Neural Networks (ANN) alone do on the datasets for New York, Australian, and Spanish electricity markets.

show abstract

“…Market demand has a large influence on electricity pricing and therefore it has been accounted for in the price prediction models [17], [11]. Motamedi et al [17] recognized the importance of studying the consumers' decisions and reactions when creating a forecasting framework.…”

Section: Related Workmentioning

confidence: 99%

“…Motamedi et al [17] recognized the importance of studying the consumers' decisions and reactions when creating a forecasting framework. Consequently, they proposed a joint price and demand prediction.…”

Section: Related Workmentioning

confidence: 99%

Energy cost forecasting for event venues

Zagar

Grolinger

Capretz

et al. 2015

2015 IEEE Electrical Power and Energy Conference (EPEC)

View full text Add to dashboard Cite

Abstract-Electricity price, consumption, and demand forecasting has been a topic of research interest for a long time. The proliferation of smart meters has created new opportunities in energy prediction. This paper investigates energy cost forecasting in the context of entertainment event-organizing venues, which poses significant difficulty due to fluctuations in energy demand and wholesale electricity prices. The objective is to predict the overall cost of energy consumed during an entertainment event. Predictions are carried out separately for each event category and feature selection is used to select the most effective combination of event attributes for each category. Three machine learning approaches are considered: k-nearest neighbor (KNN) regression, support vector regression (SVR) and neural networks (NN). These approaches are evaluated on a case study involving a large event venue in Southern Ontario. In terms of prediction accuracy, KNN regression achieved the lowest average error. Error rates varied greatly among different event categories.

show abstract

Electricity Price and Demand Forecasting in Smart Grids

Cited by 141 publications

References 33 publications

Prediction-Based Multi-Agent Reinforcement Learning in Inherently Non-Stationary Environments

Prediction-Based Multi-Agent Reinforcement Learning in Inherently Non-Stationary Environments

Ensemble Prediction Model with Expert Selection for Electricity Price Forecasting

Energy cost forecasting for event venues

Contact Info

Product

Resources

About