Additive manufacturing (AM) is a leading technology used in many fields, such as medicine and aerospace, to make prototypes and functional part fabrication. The energy requirements of the AM process are considerable and have serious consequences for environmental health and long-term viability. Research in both the private and public sectors has shifted its attention to the problem of predicting and optimising the amount of energy that AMs use.Material state, process operation, part and process design, working environment, and other factors all play a role in this problem. Existing research shows that the design-relevant aspects have a significant role in AM energy consumption (EC) modelling in reality, although this topic has not received enough attention. As a result, this research starts by analysing the design relevant features (DRFs) from the perspective of energy modelling.Before production, these features are normally decided by part designer (PD)and process operator (PO).An ANN driven cluster-aware enhanced spider monkey optimization algorithm (CAESMOA) is suggested to improve the energy utility relying on the novel modelling methodology. Deep learning is used to improve the global best of CAESMOA and solve a number of concerns, including speeding up search times. In order to verify the accuracy of the suggested modelling technique, DRFsare obtained from a functioning AM system in the production line. In our research, we use a normalisation strategy to filter out extraneous information. At the same time, optimization has been performed to direct PD and PO towards design and decision modifications that lessen the energy requirements of the specified AM system under investigation.The effectiveness of the suggested approach is examined, and the efficiency is also contrasted with that of other current methods. These statistics showed that our approach to energy optimization in AM delivered the most trustworthy outcomes.