The fifth generation cellular network performance is measured by its spectral and energy efficiency (EE). Densifying the network might provide a solution. However, this will lead to tremendous network infrastructure and increased power consumption (PC). Cloud radio access network (C-RAN) has emerged as a solution to curtail the PC and deliver spectrum sharing in a cost-effective and energy-efficient way. Since the PC is an important key success for the upcoming generations, and in order to study and evaluate the power reductions achievable by C-RAN, a reliable power model (PM) is required. This study proposes a component, linear and parameterised PMs to explore the individual components relevant for PC analysis, particularly for C-RAN architecture. The models quantify the EE by capturing the power consumed by individual components, and the amount of power reduced in the network. The model shows a dramatic reduction in the cooling PC (87.4%), whereas the total PC is reduced to about (33.3%) compared with macro base stations deployment. Finally, accuracy comparison of the component and parameterised models has been presented.