DCU

Topic:

Optimising Energy Efficiency in Manufacturing – Data Driven Modelling.

Synopsis:

With the greater availability of data from manufacturing tools and systems there is an opportunity to optimise the use of the energy resources available so that manufacturing has a lower impact on the environment.  The secondary effect is that it should also reduce the cost of manufacturing by allowing the energy consumption match the demands of production and environmental protection more closely.  To enable this requires the development of modelling approaches which can be driven by the data.  Once developed the models can be used to determine the best feasible solution to allow efficient use of energy to meet production demands.

Takeaways:

• Modelling is required to understand the system behaviour in response to demand for product
• Energy modelling must be combined with production demands
• Process energy consumption must be mapped to outputs
• Outputs from modelling process must be designed to enable implementation

Submission:

When designing a new manufacturing line or facility, the focus is often on ensuring that production of quality product is achieved with the minimal capital investment to provide rapid return on investment.  With this focus, often the operational impact on the environment is often overlooked entirely, or driven to meet specific regulatory restrictions.  Indeed, in the meeting of regulations, concentration on the measurable indicators of compliance with regulations often means that other impacts are ignored completely.

Optimising the consumption of energy in manufacturing is not a simple task, as the requirements of production rate and quality often lead to conservative setting of control parameters to ensure the product is available when the customer demands.  The significant differences between processes and manufacturing systems mean that the development of a “silver bullet” solution to reducing the energy consumption and environmental impact across a wide range of industries is difficult.  However, with the development of control systems and networks more data than ever before is becoming available.  In this context there is the possibility to develop model driven optimization approaches where measurements from the system can be used to improve the operation.

One such example was applied in the context of a job-shop environment where parallel stations were used to process product.  Depending on the mix of product, the capacity of the stations was not always fully occupied however, as the time to bring a station to production was significant, all of the stations were left in a ready state with considerable energy wasted while no product was being processed.  Modelling of the flow of material towards this station by reviewing the statistical progress of parts through the whole factory allowed the development of a strategy which enabled some stations to be switched to a low energy consumption mode when they were not required by considering the status of work in progress upstream.  Using simulation modelling the effectiveness of this approach was examined and the point where the upstream flow should be monitored was established so that the operators could easily determine when the stations could be switched in and out of production.

A more recent development looks at the water-energy nexus, where the energy associated with ensuring waste water meets environmental standards before discharge has a significant impact in its own right.  Often such plant is designed and installed to ensure it meets the legal requirements with little consideration of the efficiency of the process.  Indeed, operators are often unaware of the specific energy consumption associated with such processes as they are often part of background facilities rather than direct processing of product.  Here monitoring approaches and energy consumption models have been developed with allow the details to be understood.  The next stage in this work is to link this energy modelling to specific water quality outputs across different technologies.  Once completed this will enable modelling of the most appropriate solution to ensure that the standards are met with minimal energy consumption.

About Paul Young.

Dr Paul Young has been actively researching in the field of manufacturing since the early 1990’s.  He has worked extensively with a number of industries to model and improve their manufacturing systems through the application of data driven models.  In addition he has worked in the area of vehicle design and vibration analysis.  Across all of these disciplines the approach is that of analysing the available data from the system and then using a model based approach develop both an understanding of the characteristics and apply some form of optimisation to enable improvement.

He is a lecturer in the School of Mechanical and Manufacturing Engineering, where he teaches modules in mechanics, design, system analysis and optimisation.

About the Advanced Processing Technology Research Centre in DCU

The Advanced Processing Technology Research Centre (APT) focuses on state of the art research activities in the areas of Production Technology, Sustainable Technology, Micro and Nano Technology, and Advanced Engineering Materials. The APT is a leading international research centre which as a primary goal strives to provide significant translational benefit to the wider community. Research projects undertaken within APT are conducted to a world class level and support local and internationally based enterprises. The APT research group has established a strong infrastructure of equipment and people in the area of processing technologies at DCU. APTs education and outreach events include seminars and courses which enable the transfer of processing technologies knowledge to the broader community.