Energy optimisation in manufacturing systems using virtual engineering-driven discrete event simulation

Abstract

Increasing the cost of electricity and the global obligations for efficient use of energy have added additional pressure to industrial companies in an already challenging market. Manufacturing companies are adopting methods to have a greater agility to respond quicker to the market dynamics and varying demands by changing production configurations. In recent years, deploying virtual engineering design approaches and extensive simulation methods have facilitated an early insight into how a system may perform, in advance of the physical build. However, the impact of system reconfiguration on the cost of energy consumption is typically unknown. Much research has been carried out on the development of new energy-efficient drives (e.g. motors and actuators) and also the possibility of turning off the drives when idle to save energy. However, in this article, the authors propose a new method of energy saving for engineering production lines by fine tuning the low level device motions to optimise energy consumption. An integration method between virtual engineering design and simulation modelling is proposed, and as a result a simulation method for energy optimisation is developed. In this article the method of interpreting virtual design data for use in simulation modelling of a production assembly line in an automotive industry is discussed. Furthermore, a developed algorithm for optimising energy usage based on adjusting dynamic properties of the system components (e.g. accelerations, torque and mass) is discussed and the result of implementing the concept in an experimental application in the powertrain industry is reported.

Industry 4.0 and Climate Change—Exploring the Science-Policy Gap

“…• Issues hindering efficient activity scheduling techniques in machine-to-machine communications for more energy efficiency [37]; • workflow for integrating energy data into material flow simulation [38]; • simulation method for energy optimization [39]; and • algorithm adjusting dynamic properties of robots to reduce energy consumption [40].…”

“…Although showing some of the common shortcomings of a research prototype, the module they present is a promising step toward the combined simulation of energy and material flows in discrete production planning. Ghani et al (2012) propose a new simulation method to optimize energy on engineering production lines by fine tuning low-level device motions. In the same paper, the authors also present a method to integrate virtual engineering design and simulation modeling.…”

Discrete Event Simulation and Virtual Reality Use in Industry: New Opportunities and Future Trends

“…Ren and Guan [37] make a contribution to real-time DES use in their case study the construction of a concrete dam. Ghani et al [38], [39] provide a methodology for Virtual Engineering (VE) and DES combined use, utilizing XML and a normalization schema for model data exchange. Although VE mainly concentrates on the use and manipulation of 3-D CAD models, the integration approach detailed by Ghani et al [38], [39] could be applied to DES and VR.…”

“…Ghani et al [38], [39] provide a methodology for Virtual Engineering (VE) and DES combined use, utilizing XML and a normalization schema for model data exchange. Although VE mainly concentrates on the use and manipulation of 3-D CAD models, the integration approach detailed by Ghani et al [38], [39] could be applied to DES and VR. The possibility to change simulation parameters while a simulation is running via an AR environment is the subject that is explored by Dorozhkin et al [40].…”

Calibration of Stochastic Computer Simulators Using Likelihood Emulation

“…There is a broad range of examples of stochastic simulation models to which the proposed calibration method can contribute. Ghani, Monfared, and Harrison (2012) coupled virtual engineering and simulation models to efficiently minimize energy costs for a manufacturing process; here calibration can ensure that various worker and shift pattern constraints are met, allowing so-called "what-if" scenarios to be explored for further energy savings. Gillespie (2007) simulated chemical interactions of molecules over time; calibrating inputs ensures that numbers of molecules for different species match target data at given times and, given the simulation model's structure, allows molecules' positions and velocities at intermediate times to be understood.…”