An expert system approach to short-term load forecasting for Reliance Energy Limited, Mumbai

Abstract

Abstract-Economically efficient generation scheduling requires accurate forecasting of load. In this paper, we propose a Short Term Load Forecasting program for Reliance Energy Limited (REL) in Mumbai region. The method is based on a similar day approach. The development of forecast engine involves 4-steps. The first step involves discussion with domain experts (Utility Engineers) to extract and learn the rules regarding system behaviour. In the next step, these rules are refined by statistical analysis. A linear prediction model for each day of week is developed. The third step involves an adaptive implementation of the rules. The parameters of the linear model are learned from previous data by solving an optimization problem. Quadratic Programming is used with redundancy factor 2. The final step involves fine-tuning of forecast by re-shaping the forecast as the reference day using Fast Fourier Transform, filtering and smoothening by 3-point moving average technique. Normalization is done using dc component of reference day.Since the parameters are learnt from past few weeks data, the seasonal variations due to change in season like winter, summer are better modeled. Detailed study of the results of the forecast program, the overall Mean Absolute Percentage Error (MAPE) of the forecasted data is 2.89 over an interval from Aug'04 to May'05 which is reasonable.

A Multidisciplinary Approach for the Development of Smart Distribution Networks

“…However, these models tend to work only for energy systems that are well behaved. On the other hand, the AI models are based on expert systems [88], artificial neural networks (ANNs) [89][90][91], support vector machines, fuzzy logic [92], and genetic algorithm [87]. Currently, the most widely used methods are those based on ANNs, applied in this area since the 1980s.…”

Combination approaches for short term load forecasting

“…Parameters of the linear model are learned from the previous data by solving an optimization problem. In other words, the forecasting model comprise of linear combination of load data of reference days and a temperature correction [5]. Weights are optimized with the objective being minimization of sum of the squared error while applying model to similar days in the past 6-10 weeks.…”

“…LetL ∈ R Ns be the actual load andF ∈ R Ns a combination forecast where N s is the number (an integer multiple of N = 96) of samples, then L =F +¯ (5) where¯ is the error vector. To minimize mean square error, an optimization problem can be formulated as…”

“…Complexity of the forecasting can be simplified with the use of Artificial Intelligence (AI) techniques. Expert system based methods derive forecasting rules with the help of utility experts and application of optimization methods to fit the model to the forecast as accurately as possible [5]. Artificial Neural Network (ANN) based methods imply machine learning from historical data using mathematical algorithms.…”

Short term load forecast of Kano zone using artificial intelligent techniques

“…Intelligent techniques provide better alternative. These include various paradigm that uses artificial intelligence methods such as expert system [3], support vector machine (SVM) [4][5], Fuzzy Logic [6] and neural network. Expert system is a knowledge base method that builds load forecasting by imitating experience and human expertise.…”