<p>Slime mould algorithm (SMA) is a novel meta-heuristic algorithm, which seeks the optimal solution by simulating the foraging behavior and morphological changes of slime moulds. However, when solving the global optimization problem, SMA has the problem of insufficient population diversity and single search direction, which leads to low convergence accuracy of the algorithm. This paper proposes an enhanced slime mould algorithm based on quasi-conjugate search mechanism (QCSMA). Specifically, a fitness-distance balanced regeneration mechanism is proposed to relax the selection pressure of the population and produce the potential virtual point for drawing individuals to be redistributed in other areas, enhancing the algorithm’s global search capability. Next, a quasi-conjugate search mechanism is proposed, which involves two key calculations. One is to utilize the fitness improvement as the reward feedback to estimate the gradient direction of the considered individual without relying on mathematical derivation information. Then, the previously mentioned direction is dynamically aggregated with the historical search direction to better guide the individual to converge to the region where the optimal solution is located. To validate the performance of the proposed algorithm, experiments are carried out on two function test sets, IEEE CEC2017 and IEEE CEC2021. In addition, QCSMA is compared with 23 other different algorithms. The experimental results show that the new algorithm exhibits competitive results compared to other algorithms, indicating that QCSMA has better optimization ability, stability, and convergence speed. To further verify the optimization performance of QCSMA in real applications, experiments are carried out in two engineering optimization problems, and the effectiveness of QCSMA is proved by the results.</p>