Search

In Industry 4.0, the production data obtained from the Internet of Things has reached the magnitude of big data with the emergence of advanced information and communication technologies. The massive and low-value density of big data challenges traditional clustering and correlation analysis. To solve this problem, a big data-driven correlation analysis based on clustering is proposed to improve energy and resource utilisation efficiency in this paper. In detail, the production units with abnormal and energy-intensive consumption can be classified by using clustering analysis. Additionally, feature extraction is carried out based on clustering analysis and the same cluster data is migrated to the training data set to improve correlation analysis accuracy. Then, correlation analysis can balance the relationship between energy supply and demand, which can reduce carbon emission and enhance sustainable competitiveness. The sensitivity analysis results show that the feature extraction method can improve the correlation analysis accuracy compared to the original analysis model. In conclusion, the big data-driven correlation analysis based on clustering can uncover the potential relationship between energy consumption and product yield, thus improving the efficiency of energy and resources. • An architecture of big data clustering and correlation analysis is proposed. • A PCA-guided ISODATA is proposed for data pre-processing and processing. • The feature extraction method has been proposed based on clustering. • A correlation analysis model is established for data mining. • The models can improve the scheduling efficiency of energy and resources. [ABSTRACT FROM AUTHOR]

The circular economy plays an important role in energy-intensive industries, aiming to contribute to ethical sustainable societal development. Energy demand response is a key actor for cleaner production and circular economy strategy. In the Industry 4.0 context, the advanced technologies (e.g. cloud computing, Internet of things, cyber-physical system, digital twin and big data analytics) provide numerous opportunities for the implementation of a cleaner production strategy and the development of intelligent manufacturing. This paper presented a framework of data-driven sustainable intelligent/smart manufacturing based on demand response for energy-intensive industries. The technological architecture was designed to implement the proposed framework, and multi-level demand response models were developed based on machine, shop-floor and factory to save energy cost. Finally, an application of ball mills in a slurry shop-floor of a partner company was presented to demonstrate the proposed framework and models. Results showed that the energy efficiency of ball mills can be greatly improved. The energy cost of the slurry shop-floor saved approximately 19.33% by considering electricity demand response using particle swarm optimisation. This study provides a practical approach to make effective and energy-efficient decisions for energy-intensive manufacturing enterprises. • A framework of data-driven sustainable intelligent manufacturing is proposed. • The framework is to achieve a circular economy for energy-intensive industries. • Multi-level demand response models based on PSO are developed to save energy costs. • The framework and models can benefit the environment, economy and society. [ABSTRACT FROM AUTHOR]

Improving energy and resource efficiency in manufacturing is an important goal for enterprises to sustain their competitive advantages. Predictive production planning is a new solution to achieve such goal, following the direct improvement of energy efficiency and indirect energy savings through better scheduling. With the emergence of new information and communication technologies under the background of Industry 4.0, the amount of various energy and resource data obtained through Internet of Things is reaching the magnitude of big data. It poses a challenge to traditional data processing and mining methods for predictive production. To solve this challenge, in this paper, the big data driven predictive production planning is proposed to improve energy and resource efficiency for energy-intensive manufacturing industries. Additionally, the cube-based energy consumption models and long short-term memory based energy prediction model are established for data preprocessing and mining correspondingly. An industrial case study is presented to process energy big data and predict energy consumption parameters and production status. The performance evaluation results indicate that the proposed long short-term memory models outperform back propagation neural network, autoregressive moving average and support vector regression. Based on data preprocessing and forecasting results, the energy and resource efficiency could be improved during the whole manufacturing process for energy-intensive manufacturing industries. • Big data driven predictive production planning is proposed for cleaner production. • The cube-based energy consumption models are established to preprocess data. • The long short-term memory based energy prediction models are used for data mining. • Models can accurately predict energy and production status with periodicity. • The long short-term memory models outperform traditional forecasting methods. [ABSTRACT FROM AUTHOR]