Deep learning based ensemble approach for probabilistic wind power forecasting

A new model to deal with the short-term generation scheduling problem for hydrothermal systems is proposed. Using genetic algorithms (GAs), the model handles simultaneously the subproblems of short-term hydrothermal coordination, unit commitment, and economic load dispatch. Considering a scheduling horizon period of a week, hourly generation schedules are obtained for each of both hydro and thermal units. Future cost curves of hydro generation, obtained from long and mid-term models, have been used to optimize the amount of hydro energy to be used during the week. In the genetic algorithm (GA) implementation, a new technique to represent candidate solutions is introduced, and a set of expert operators has been incorporated to improve the behavior of the algorithm. Results for a real system are presented and discussed.

/pdf/short-term-hydrothermal-generation-scheduling-model-using-a-4ta48c1dyd.pdf

Short-term hydrothermal generation scheduling model using a genetic algorithm

Short-term hydro-thermal scheduling using particle swarm optimization method

The volatility and uncertainty of wind power often affect the quality of electric energy, the security of the power grid, the stability of the power system, and the fluctuation of the power market. In this case, the research on wind power forecasting is of great significance for ensuring the better development of wind power grids and the higher quality of electric energy. Therefore, a lot of new forecasting methods have been put forward. In this paper, a new forecasting model based on a convolution neural network and LightGBM is constructed. The procedure is shown as follows. First, we construct new feature sets by analyzing the characteristics of the raw data on the time series from the wind field and adjacent wind field. Second, the convolutional neural network (CNN) is proposed to extract information from input data, and the network parameters are adjusted by comparing the actual results. Third, in consideration of the limitations of the single-convolution model in predicting wind power, we innovatively integrated the LightGBM classification algorithm at the model to improve the forecasting accuracy and robustness. Finally, compared with the existing support vector machines, LightGBM, and CNN, the fusion model has better performance in accuracy and efficiency.

/pdf/a-model-combining-convolutional-neural-network-and-lightgbm-2x1hhbnj13.pdf

A Model Combining Convolutional Neural Network and LightGBM Algorithm for Ultra-Short-Term Wind Power Forecasting

Fast evolutionary programming techniques are applied for the solution of a short-term hydrothermal scheduling problem. Evolutionary programming (EP)-based algorithms with Gaussian and other mutation techniques have been developed and tested on a multi-reservoir cascaded hydroelectric system having prohibited operating zones and a thermal unit with valve point loading. Numerical results show that all of the EP algorithms are capable of finding very nearly global solutions within a reasonable time but an EP algorithm with better of Gaussian and Cauchy mutations appears to be the best amongst all EPs in terms of convergence speed, solution time, and minimum cost.

Fast Evolutionary Progranuning Techniques for Short-Term Hydrothermal Scheduling

In this paper a diploid genotype based genetic algorithm (GA) is applied to solve the short-term scheduling of hydrothermal systems. The proposed genetic algorithm uses a pair of binary strings with the same length to represent a solution to the problem. The crossover operator is carried out by means of the separating and recombining technique, which is of the same effect of that of uniform crossover. The dominance mechanism in the algorithm is realized by a simple Boolean algebra calculation. Simulation results show that the proposed algorithm has a strong ability to maintain gene diversity in a limited population due to the diploid chromosomal structure accompanying the dominance mechanism. This ability improves the overall performance and avoids premature convergence. The model can concurrently tackle the requirements of power balance, water balance and water traveling time between cascaded power stations, which are more difficult for other approaches to manage. Several examples are used to verify the validity of the algorithm.

A diploid genetic approach to short-term scheduling of hydro-thermal system

High-quality wind power prediction intervals (PIs) are of utmost importance for system planning and operation. To improve the reliability and sharpness of PIs, this paper proposes a new approach in which the original wind power series is first decomposed and grouped into components of reduced order of complexity using ensemble empirical mode decomposition and sample entropy techniques. The methods for the prediction of these components with extreme learning machine technique and the formation of the overall optimal PIs are then described. The effectiveness of proposed approach is demonstrated by applying it to real wind farms from Australia and National Renewable Energy Laboratory. Compared to the existing methods without wind power series decomposition, the proposed approach is found to be more effective for wind power interval forecasts with higher reliability and sharpness.

Yong-Gang Wu

Papers

A diploid genetic approach to short-term scheduling of hydro-thermal system

An Advanced Approach for Construction of Optimal Wind Power Prediction Intervals