Loading...

Locally linear neuro-fuzzy (LLNF) electricity price forecasting in deregulated power markets

Abdollahzade, M ; Sharif University of Technology | 2010

1237 Viewed
  1. Type of Document: Article
  2. Publisher: 2010
  3. Abstract:
  4. The disguise of traditional monopolistic electricity markets into deregulated competitive ones has made 'price forecasting' a crucial strategy for both producers and consumers: for the producers, to maximize their profit and hedge against price volatilities and for the consumers to manage their utility. Electricity price forecasting has thus emerged as a progressive field of study and numerous researches have been conducted to improve and optimize the price forecast methods. This paper proposes a precise and computationally efficient method to perform price forecasting in deregulated power markets. A locally linear neuro-fuzzy model is developed for price forecasting. The model is trained by a locally linear model tree (LOLIMOT) learning algorithm. An appropriate input selection based on correlation analysis is considered to develop the model. In order to investigate the performance of the proposed model, three major power markets almost serving as global benchmarks for price forecasting studies are tested and the results are verified using real observed data for various forecasting scenarios. Furthermore, the performance of the suggested model is comprehensively compared to the most recent studies available in the literature using various assessment criteria. Comparisons demonstrate the superiority of the proposed model, in terms of its performance and accuracy, for application in real world forecasting problems
  5. Keywords:
  6. Electricity markets ; Locally linear model tree algorithm ; Locally linear neuro-fuzzy model ; Assessment criteria ; Computationally efficient ; Correlation analysis ; Deregulated power markets ; Electricity market ; Electricity price forecasting ; Forecasting problems ; Input selection ; Locally linear model trees ; Neuro-Fuzzy ; Neuro-Fuzzy model ; Observed data ; Power markets ; Price forecasting ; Price forecasts ; Price volatility ; Benchmarking ; Electric load forecasting ; Forecasting ; Learning algorithms ; Deregulation ; Algorithms ; Commerce ; Costs ; Electricity ; Forecasts ; Forestry ; Marketing ; Profitability
  7. Source: International Journal of Innovative Computing, Information and Control ; Volume 6, Issue 9 , 2010 , Pages 4203-4218 ; 13494198 (ISSN)
  8. URL: http://www.ijicic.org/09-0342-1.pdf