No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
LSTM-AdaBoost Electricity Sales Forecasting Model Based on One-Dimensional Time Series Input
Forecasting wind speed is a very important part in weather forecasting. Because of the nonlinear behaviors of nature and climate changes, wind speed prediction becomes a challenging task, particularly country like Bangladesh where lots of areas are costal and season changes in frequent. This study is done to make an attempt to predict the wind speed using two very potential and wide frames of statistical data mining and machine learning approaches; Support Vector Regression (SVR) and Artificial Neural Network (ANN) with back propagation technique. 7years (2008–2014) historical dataset of wind speed of Chittagong costal area were collected from Bangladesh meteorological division (BMD) for undertaking the experiment. Leaky ReLu function was applied as the rectifier to the input data to control the thresholds and activations of neurons in MLP. The aim of this study was to propose a model that can predict short term wind speed with maximum accuracy. Finally, after considerable amount of experimentations the outcome from this study is; our proposed SVR and ANN models are able to predict wind speed with more than 99% accuracy in short term prediction. Moreover, ANN can outperform SVR in some situations with highest 99.80% accuracy. But SVR models are best suited for overall wind speed forecasting in different horizons with highest 99.60% accuracy. These results outperform the performances of previous recent works that are mentioned in literature review and reference sections of this paper.
As the power system is facing a transition towards a more intelligent, flexible and interactive system with higher penetration of renewable energy generation, load forecasting, especially short-term load forecasting for individual electric customers plays an increasingly essential role in the future grid planning and operation. Other than aggregated residential load in a large scale, forecasting an electric load of a single energy user is fairly challenging due to the high volatility and uncertainty involved. In this paper, we propose a long short-term memory (LSTM) recurrent neural network (RNN) based framework, which is the latest and one of the most popular techniques of deep learning, to tackle this tricky issue. The proposed framework is tested on a publicly available set of real residential smart meter data, of which the performance is comprehensively compared to various benchmarks including the state-of-the-arts in the field of load forecasting. As a result, the proposed LSTM approach outperforms the other listed rival algorithms in the task of short-term load forecasting for individual residential households.