As a promising electricity storage system, Liquid Air Energy Storage (LAES) has the main advantage of being geographically unconstrained. LAES has a considerable potential in energy efficiency improvement by utilizing compression heat and integrating with other systems. In this work, the Stirling Engine (SE) is introduced to improve the energy efficiency of the LAES system. Three LAES-SE systems are modelled in Aspen HYSYS and optimized by the Particle Swarm Optimization (PSO) algorithm. The studied systems include (i) the LAES system with 3 compressors and 3 expanders (3C+3E) using an SE to recover the compression heat, (ii) the 3C+3E LAES system with LNG regasification and SE, and (iii) the 3C+3E LAES system with solar energy and SE. The optimization results show that the Round-Trip Efficiencies (RTEs) of the LAES-SE system and the LNG-LAES-SE systems are 68.2% and 73.7%, which are 3.2% and 8.7% points higher than the basic 3C+3E LAES-ORC system with an RTE of 65.0%. For the Solar-LAES-SE system, a revised RTE and the economic performance with solar energy input are optimized. The traditional RTE for the Solar-LAES-SE system, which only accounts for power produced and consumed in the discharging and charging sections, is 189% and 173% respectively, when optimized with respect to energy and economic performances. The revised RTE accounts for the integrated external sources, avoiding the confusing result that the RTE becomes larger than 100%. The energy and economic performances of the Solar-LAES-SE system are proved to be the best compared with the Solar-LAES-ORC and Solar energy directly heated-LAES systems.