This paper presents an improved adaptive simulated annealing genetic algorithms of optimal sensor placement (OSP) based on Modal Assurance Criterion (MAC). Taking the maximum off-diagonal element of the MAC matrix as a target function, and a dualistic coding genetic algorithms is developed to solve the constraint problem. Improve the traditional simulation annealing algorithms, and use it as an operator of GA. Therefore, adaptive crossover and mutation probabilities are adopted to prevent premature convergence. Example results clearly indicate that this method can simultaneously optimize the placement and number of sensors, so it is feasible in optimal sensor placement of composite laminate aimed at damage diagnosis.