Achieving environmental sustainability via an integrated shampoo optimized BiLSTM-Transformer model for enhanced time-series forecasting

TL;DR

This paper introduces a new deep learning model combining BiLSTM and Transformer for better time-series forecasting in power systems.

Contribution

The novel BiLSTM-Transformer model uses Shampoo optimization for improved convergence and forecasting accuracy.

Findings

BiLSTM-Transformer outperforms existing models in time-series forecasting benchmarks.

The model effectively captures both short-term and long-range dependencies in meteorological data.

It supports sustainable energy planning and smart grid operations with reliable predictions.

Abstract

Accurate forecasting plays a vital role in enhancing the efficiency of power systems, ensuring better resource management, and supporting strategic decision-making. This work presents BiLSTM-Transformer, a hybrid deep learning model that integrates Bidirectional Long Short-Term Memory (BiLSTM) networks with Transformer architecture to improve predictive performance in complex time-series tasks. The model employs a second-order optimization approach using Shampoo, which strengthens convergence stability and promotes better generalization during training. By effectively modeling both short-term variations and long-range dependencies in meteorological data, BiLSTM-Transformer achieves superior forecast accuracy across multiple evaluation benchmarks. The results highlight its potential as a reliable tool for supporting sustainable energy planning and smart grid operations.

Figures