A life maximization problem fulfilling heterogeneous coverage requirements of the targets with different priorities in directional sensor networks which consist of nodes with different parameters in terms of the sensing radius, sensing angle of view and energy is addressed. The lifespan of the networks can be extended by organizing the sensor nodes into different cover sets and switching between these cover sets, which can cover all targets and satisfy their coverage quality requirements. Firstly, the life maximization problem is proved to be NP-Hard. Then enhanced harmony search algorithm (shortly for EHS) is used to solve the mentioned problem. In the proposed EHS, chaotic differential mutation operator is used instead of pitch adjustment operator in the harmony search (HS) algorithm to improve the diversity of the population. The local optimization ability of the EHS can be improved by combining harmony search algorithm with simulated annealing algorithm. To valid the performance of the proposed algorithm a greedy algorithm for life maximization problem is also proposed. Simulation results show that the proposed EHS achieves better performance than the primitive harmony algorithm and greedy algorithm, which proves the effectiveness of the proposed algorithm.