Abstract: Considering that the large thermal effect produced by chips in the process of testing will damage the reliability of the chips, in this paper, a new method using improved quantum-inspired evolutionary algorithm of multi populations is proposed to optimize the test power consumption. The proposed method is meant to reduce the test cost and prevent the chip from being damaged due to excessive power consumption. In order to avoid the single population can't guarantee the diversity of populations, this paper combines multiple populations operation with algorithm and adopts the strategy of crossover in excellent individuals to improve the ability of global optimization. Taking ITC'02 standard circuit as the test object, the experiment results demonstrate that the proposed method can converge to the optimal solution quickly, which can reduce the total test time and improve the test efficiency, in addition, the test power consumption is optimized under the constraint of the test time.