In a deregulated electricity market, generators have to optimally bid to maximize their profit under incomplete information of other competing generators. This paper addresses an optimal bidding strategy of a thermal generator in a uniform price spot market considering a precise model of nonlinear operating cost function and minimum up/down constraints of unit commitment. The bidding behaviors of other competing generators are described using normal probability distribution function. Bidding strategy of a generator for each trading period in a day-ahead market is solved by fuzzy adaptive particle swarm optimization (FAPSO), where inertia weight is dynamically adjusted using fuzzy evaluation. FAPSO can dynamically follow the frequently changing market demand and supply in each trading interval. The effectiveness of the proposed approach is tested with examples and the results are compared with the solutions obtained using genetic algorithm (GA) approach and other versions of PSO.