Abstract: The paper proposes a hybrid long short-term memory (LSTM) model based on improved grid search (IGS) algorithm (LSTM-IGS) model for precise day-ahead wholesale market price forecasting at a 5 ...

The main questions this project seeks to answer include: How to effectively integrate LSTM-based flow rate prediction into the token bucket algorithm? Can the adaptive token bucket algorithm result in ...

In this paper, a deep-learning model based on two-stage attention mechanism over LSTM is proposed to forecast a day-ahead PV power. In addition, the Bayesian optimization algorithm is applied to ...

a novel optimized LSTM model is proposed in this article. The model’s purpose is to predict the stability of smart grids. The findings of the experiments are then contrasted with more contemporary ...

This project implements a Long Short-Term Memory (LSTM) model using MindSpore to predict traffic volume based on historical data and various weather conditions. The dataset contains traffic volume ...

The first type is a set of experiments targeting and evaluating the proposed feature selection algorithm. Whereas the second type is a set of experiments that assessed the optimized LSTM model, which ...