W-LSTMix: A Hybrid Modular Forecasting Framework for Trend and Pattern Learning in Short-Term Load Forecasting

SHIVAM DWIVEDI, Anuj Kumar, Harish Kumar Saravanan, Pandarasamy Arjunan

Keywords: Energy Informatics, Smart Grid, Energy Forecasting, Short-term Load Forecasting (STLF), Demand-side Load Management, Time Series Foundation Models (TSFM), Machine Learning

TL;DR: W-LSTMix uses wavelets with enhanced N-BEATS to learn trends and patterns, boosting time series forecasting accuracy.

Abstract:

Motivated by the need for a lightweight, gener-
alizable forecasting model in intelligent build-
ing energy management and reliable microgrid
control. We propose W-LSTMix—a modular
and lightweight hybrid architecture designed to
generalize across diverse building types, with
only ∼ 0.13 M parameters integrates wavelet-
based signal decomposition, neural basis expan-
sion, and patch-based temporal mixing for effi-
cient building-level load forecasting. The model
separately forecasts decomposed time series com-
ponents: long-term trends are modeled using an
LSTM-enhanced N-BEATS stack, while resid-
ual and seasonal patterns are captured through
an MLP-Mixer-enhanced N-BEATS structure.
We compare our model against state-of-the-art
TSFMs such as Lag-Llama, Moirai, Chronos, and
Tiny Time Mixers in zero-shot and fine-tuned
settings, under domain-specific training and test-
ing. In both comparisons, our model consistently
outperforms the baselines, demonstrating robust
generalization capabilities and suitability for real-
world intelligent energy management.

Submission Number: 45