Abstract
The energy efficient operation of a manufacturing system is important for sustainable development of industry. Apart from the device and process level, energy saving methods at the system level has attracted increasing attention with the rapid growth of the industrial Internet of things technology, which makes it possible to sense and collect real-time data from the production line and provide more opportunities for online control for energy saving purposes. In this paper, a dynamic adaptive fuzzy reasoning Petri net is proposed to decide the machine energy saving state considering the production information of a discrete stochastic manufacturing system. Fuzzy knowledge for energy saving operations of a machine is represented in weighted fuzzy production rules with certain values. The rules describe uncertain, imprecise, and ambiguous knowledge of machine state decisions. This makes an energy saving sleep decision in advance when a machine has the inclination of starvation or blockage, which is based on the real-time production rates and level of connected buffers. A dynamic adaptive fuzzy reasoning Petri net is formally defined to implement the reasoning process of the machine state decision. A manufacturing system case is used to demonstrate the application of our method and the results indicate its effectiveness for energy saving operation purposes.
Highlights
It has been shown that 37% of the energy consumption and 17% of carbon dioxide (CO2 ) emissions come from the industrial sector [1]
By representing the operation knowledge in weighed fuzzy production rules (WFPRs), this paper representing the operation knowledge in weighed fuzzy production rules (WFPRs), this paper presents presents a dynamic adaptive fuzzy reasoning Petri net (DAFRPN) to represent and infer machine a dynamic adaptive fuzzy reasoning Petri net (DAFRPN) to represent and infer machine states for states for energy saving decisions considering real-time buffer levels and production rates
A control method based on dynamic adaptive fuzzy reasoning Petri nets is proposed to make a decision on machine states for reducing idle times
Summary
It has been shown that 37% of the energy consumption and 17% of carbon dioxide (CO2 ) emissions come from the industrial sector [1]. The energy consumption of industries has an annual growth rate of 1.3% from 2013 to 2025 [2]. The manufacturing industry is responsible for 38% of energy-related CO2 emissions in China [3]. 18–26% of the total energy consumption in manufacturing industries, i.e., 25-37 EJ, can be saved if proper actions are taken [6]. The energy-saving potential in manufacturing industries worldwide is estimated to be 20% until 2050 [7]. Energy management decisions are more and more important for manufacturing industries all over the world [8]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
