Neural Networks for Energy Optimization of Production Processes in Small and Medium Sized Enterprises
Research output: Contributions to collected editions/works › Chapter
Standard
Advances and New Trends in Environmental Informatics. ed. / Volker Wohlgemuth; Stefan Naumann; Grit Behrens; Hans-Knud Arndt; Maximilian Höb. Springer, 2023. p. 129-145.
Research output: Contributions to collected editions/works › Chapter
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Neural Networks for Energy Optimization of Production Processes in Small and Medium Sized Enterprises
AU - Willenbacher, Martina
AU - Wohlgemuth, Volker
AU - Risch, Lisa
PY - 2023
Y1 - 2023
N2 - Due to the highly dynamic development processes of manufacturing companies (economic, demographic, sociological, ecological-biological processes), there are high requirements to find scientific answers to environmentally specific questions, considering the profitability and ensuring ongoing operation, and to integrate the developed models to improve the efficiency of the use of materials for energy reduction into the process flow. Decision-making is thus hampered on the one hand by the achievement of solutions in shortened innovation and production cycles and on the other hand by the complexity of the systems and processes of the environmental sector. Furthermore, there are often organizational obstacles and personnel difficulties in the introduction of intelligent algorithms in SMEs. This article describes the conception and development of an artificial neural network for the optimization of production processes regarding the reduction of energy under the aspect of quality assurance for manufacturing SMEs. It describes the development and implementation of the model for the analysis and adaptation of parameter settings to machines in the production process, which determines the ideal configuration to reduce energy consumption and improve quality. In the test of the model on four machines of a plastic-producing SME, it was proven that a total annual energy saving of 50,000 kWh can be achieved.
AB - Due to the highly dynamic development processes of manufacturing companies (economic, demographic, sociological, ecological-biological processes), there are high requirements to find scientific answers to environmentally specific questions, considering the profitability and ensuring ongoing operation, and to integrate the developed models to improve the efficiency of the use of materials for energy reduction into the process flow. Decision-making is thus hampered on the one hand by the achievement of solutions in shortened innovation and production cycles and on the other hand by the complexity of the systems and processes of the environmental sector. Furthermore, there are often organizational obstacles and personnel difficulties in the introduction of intelligent algorithms in SMEs. This article describes the conception and development of an artificial neural network for the optimization of production processes regarding the reduction of energy under the aspect of quality assurance for manufacturing SMEs. It describes the development and implementation of the model for the analysis and adaptation of parameter settings to machines in the production process, which determines the ideal configuration to reduce energy consumption and improve quality. In the test of the model on four machines of a plastic-producing SME, it was proven that a total annual energy saving of 50,000 kWh can be achieved.
KW - Sustainability sciences, Communication
UR - https://www.mendeley.com/catalogue/68fe2d97-a03e-33ea-ac0b-cfb81080fcf3/
U2 - 10.1007/978-3-031-18311-9_8
DO - 10.1007/978-3-031-18311-9_8
M3 - Chapter
SN - 978-3-031-18313-3
SN - 978-3-031-18310-2
SP - 129
EP - 145
BT - Advances and New Trends in Environmental Informatics
A2 - Wohlgemuth, Volker
A2 - Naumann, Stefan
A2 - Behrens, Grit
A2 - Arndt, Hans-Knud
A2 - Höb, Maximilian
PB - Springer
ER -