LSTM-based Flow Prediction
Hongzhi Wang, Yang Song, Shihan Tang
TL;DR
This work tackles industrial flow prediction from high-dimensional time series by introducing an LSTM framework enhanced with periodicity awareness and multivariate tuning. The approach decomposes into a data-conversion module that reduces dimensionality and constructs supervised sequences, an LSTM modeling module for sequence forecasting, and a tuning module that leverages cycle detection to iteratively refine predictions. The method demonstrates substantial accuracy gains on a real-world industrial boiler dataset, achieving a RMSE improvement of 54.05% over a vanilla LSTM and outpacing RF, BP, and CNN baselines. By explicitly exploiting periodic patterns and cross-variable dependencies, the approach offers a practical, scalable solution for predictive maintenance and production optimization in industry.
Abstract
In this paper, a method of prediction on continuous time series variables from the production or flow -- an LSTM algorithm based on multivariate tuning -- is proposed. The algorithm improves the traditional LSTM algorithm and converts the time series data into supervised learning sequences regarding industrial data's features. The main innovation of this paper consists in introducing the concepts of periodic measurement and time window in the industrial prediction problem, especially considering industrial data with time series characteristics. Experiments using real-world datasets show that the prediction accuracy is improved, 54.05% higher than that of traditional LSTM algorithm.