Electrical Energy Storage (EES) can decouple energy production from its consumption and is urgently needed by both conventional energy system for load leveling and renewable energy system for intermittency smoothing. Currently, Pumped Hydro Energy Storage (PHES) and Compressed Air Energy Storage (CAES) are the main technologies employed, but they both suffer from high capital cost, geographical constraints and environmental issues. Therefore, many innovative concepts of EES technologies have been proposed recently, including the Adiabatic Compressed Air Energy Storage (A-CAES), Pumped Thermal Energy Storage (PTES), Isothermal Compressed Air Energy Storage (I-CAES) and Liquid Air Energy Storage (LAES). All of these EES store electricity in the forms of thermal or mechanical energy, and are most suitable for large-scale energy storage. As a result, they have received intensive treatment from both the industry and academia, but a comparative study and optimization of these large-scale thermo-mechanical energy storage systems from a thermo-economic perspective has so far been lacking, which forms the major part of this thesis. In this thesis, a complete set of system models have been developed for each EES technology, incorporating both thermodynamic and economic factors with due consideration for the constraints and variation ranges of each parameter. Then parametric studies are carried out for each system to analyze the impact of each parameter on the system performance (e.g., efficiency and unit cost). Loss and cost distributions are given for the representative cases, and the detailed explanation and potential improvement are also provided. After that, thermoeconomic optimizations are carried out for each individual system, and the trade-off between efficiency and unit cost as well as the main factors controlling this trade-off are revealed. Finally, these optimized systems are compared with each other, and new EES systems that combined the merits of existing ones are proposed as well. The results show that A-CAES and I-CAES tend to have higher system efficiency and lower unit cost than PTES and LAES, but the PTES and LAES enjoys higher energy density and more siting freedom. More information about the component efficiency and cost is required for an accurate comparison between A-CAES and I-CAES, and between PTES and LAES.