Advancements in Heating Load Prediction: A Neural Network Approach in Biotechnological Facility Management

Authors

  • Xiaoran Tao School of environment and energy engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China

DOI:

https://doi.org/10.5912/jcb1726

Abstract

Efficient energy utilization and reduction of carbon emissions in heating systems can be significantly enhanced through accurate predictions of heating loads. This study introduces a sophisticated method for forecasting heating loads by leveraging historical weather and heating data. Initial steps include the use of principal component analysis to interpolate missing values within the data set, ensuring completeness and enhancing the reliability of the input data. Subsequently, the refined data undergoes processing through a radial base function neural network, which has been further optimized using computational mutual information techniques and nearest neighbor algorithms to improve its predictive accuracy. Experimental evaluations of this method demonstrate its robust capability to effectively handle missing data and provide highly accurate heating load forecasts. This approach not only contributes to more sustainable energy management in heating systems but also offers potential applications in biotechnological facility management where precise control of environmental conditions is crucial. The methodology’s effectiveness in improving prediction accuracy presents a valuable tool for optimizing energy consumption and reducing operational costs in various biotechnological applications.

Published

2025-01-24