Modelling the startup of machine tools for energy efficient multi-sleep control policies. (July 2021)

• Record Type: Journal Article
• Authors: Frigerio, Nicla; Matta, Andrea
• Is Part Of: Journal of manufacturing systems. Volume 60(2021)
• Page Start: 337
• Page End: 349
• Publication Date: 2021-07
• Subjects: EE energy efficient -- AO Always On -- SP Single-sleep Switching Policy -- MSP Multi-sleep Switching Policy -- DES Discrete Event Simulation -- CNC Computer Numeric Control -- pdf probability density function -- IHR Increasing Hazard Rate -- DHR Decreasing Hazard Rate -- CI Confidence Interval -- GA Genetic Algorithm; Machine tools -- Energy efficient control -- Multiple sleeping states -- Time-dependent startup; Manufacturing processes -- Periodicals; Production engineering -- Data processing -- Periodicals; Robots, Industrial -- Periodicals; Production, Technique de la -- Informatique -- Périodiques; Robots industriels -- Périodiques; Electronic journals; 670.42
• DOI: 10.1016/j.jmsy.2021.05.013
• Languages: English
• ISSNs: 0278-6125

Abstract

Highlights: The control of machine state when machine is idle can reduce energy consumption. The model includes time-dependent startup duration to resume the service. The selection of which machine component to control is included in the model. Production constraints on machine are handled by the optimization problem. Multiple sleeping states improve the efficiency of a controlled machine. Abstract: Energy efficient control strategies have been of recent interest in literature as promising measures to reduce the energy consumed by production equipment during interruptions of part flow. A general time-based control policy is analyzed. Machine idle times are assumed stochastic and the expected value of the energy consumed per produced part is reduced while assuring a certain target of expected production rate. The startup time required to resume the service depends on the time period the machine stays in a low power consumption state. Since the startup is often related to one or more machine components, the optimal policy might be to partially switch the machine. Therefore, a policy acting at component level is also analyzed where the duration of the startup depends on the set of machine components that are switched off. The policy is compared to state-of-the-art policies and discussed for a set of simulated numerical cases representing several production environments.