The comprehensive optimization of thermodynamic and economic performances is significant for the engineering application of ocean thermal energy conversion (OTEC). Motivated by this, this paper develops a thermo-economic OTEC model and conducts a sensitivity analysis of the OTEC system concerning its thermodynamic and economic performances. Specifically, the impact of warm-seawater temperature and cold-seawater pumping depth on the net thermal efficiency and the total investment cost are investigated. The results indicate that, an increase in warm-seawater temperature and cold-seawater pumping depth can improve the net thermal efficiency and a higher installed capacity is beneficial to the system economics. Building on these, a design optimization method with considering the on-design and off-design conditions is proposed in this paper, and the dynamic variation of warm-seawater temperature are considered in this method. In multi-objective optimization procedure, with the objective functions being the average net thermal efficiency and unit power cost within the operational cycle, the non-dominated sorting genetic algorithm II (NSGA-II) is employed to maximize the net thermal efficiency and minimize the unit power investment cost, resulting in the Pareto front. The net thermal efficiencies of OTEC systems using ammonia and R245fa as working fluids are 4.13% and 3.8%, respectively. This represents an improvement of 19.4% and 57.0%, respectively, compared to traditional optimization methods that do not account for off-design conditions.