There has been increasing interest in reducing carbon footprints globally in recent years. Hence increasing share of green energy and energy efficiency are promoted by governments. Therefore, optimizing energy consumption is becoming more critical for people, companies, industries, and the environment. Predicting energy consumption more precisely means that future energy management planning can be more effective. To date, most research papers have focused on predicting residential building energy consumption; however, a large portion of the energy is consumed by industrial machines. Prediction of energy consumption of large industrial machines in real time is challenging due to concept drift, in which prediction performance deteriorates over time. In this research, a novel data-driven method multi-regime approach (MRA) was developed to better predict the energy consumption for industrial machines. Whereas most papers have focused on finding an excellent prediction model that contradicts the no-free-lunch theorem, this study concentrated on adding potential concept drift points into the prediction process. A real-world dataset was collected from a semi-autonomous grinding (SAG) mill used as a data source, and a deep neural network was utilized as a prediction model for the MRA method. The results proved that the MRA method enables the detection of multi-regimes over time and provides a highly accurate prediction performance, thanks to the dynamic model approach.
Industrial machinery is a significant energy consumer, and its CO 2 emissions have increased dramatically in recent years. Therefore, energy efficiency is becoming crucial for businesses, governments, as well as the planet. Estimating the power consumption of industrial machines with greater accuracy assists management and optimizes machine operation parameters. Real-time industrial machine datasets present several challenges, such as changes in the data over time, unknown running conditions, missing data, etc. Most research publications focus on the accuracy of traditional static models of forecasting; however, prediction performance deteriorates over time because data evolves. We implemented deep learning as a prediction model for three distinct real-world industrial datasets. The proposed method, dynamic modeling with memory (DMWM), improved overall prediction performance compared with conventional approaches by identifying concept drifts and optimizing the number of required models in response to industrial datasets' recurring machine energy consumption patterns.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.