Artificial protozoa optimizer (APO) is a novel meta-heuristic algorithm with high optimization efficiency and robustness. It abstracts mathematical models from the foraging, dormancy, and reproductive behaviors of protozoa to guide the evolution of the population. Also, it faces the dilemma of premature convergence. To solve this problem, the paper presents an efficient artificial protozoa optimizer with quasi-affine transformation evolutionary (QTAPO). The QTAPO uses the evolution matrix in the quasi-affine transformation evolution algorithm (QUATRE) to update the positions of the particles. This improvement greatly enhances the global exploration capability of the algorithm and prevents it from falling into the local optimum. The paper compares QTAPO with six intelligent algorithms on the public test set CEC2022. The experimental results demonstrate its excellent performance. Finally, QTAPO is applied to intrusion detection in wireless sensor networks. It improves the accuracy rate and recall rate compared to other algorithms.