The optimal integration of work and heat has attracted increasing attention due to its paramount significance in achieving considerable energy and cost savings. However, the strong interactions between temperature, pressure, heat and work, in addition to unclassified stream identities, represent great challenges for optimization of work-heat exchange networks (WHEN). Quantities of binary variables and non-convexities to cope with highly nonlinear relations and identification of stream identities lead to a complex mathematical model which cannot ensure the global optimal solution and even be unable to obtain feasible solutions. To surpass these difficulties, this paper proposes a sequential approach for WHEN synthesis based on a superstructure-based model combined with meta-heuristic strategies. The work exchange networks (WEN) configuration is firstly derived by solving a MINLP model based on the superstructure method. Afterwards, several meta-heuristic strategies in terms of thermodynamic analysis are proposed for the identification of streams identities (cold or hot), thus facilitating the subsequent heat integration through heat exchange networks (HEN) synthesis between the determined cold and hot streams. A case study is conducted to access the efficacy of our proposed method, where the results show that the sequential approach is highly efficient for WHEN synthesis with more mechanical and thermal energy recovery as well as considerable savings in total annual cost.