A novel multi-level hierarchy optimization algorithm for inner detector speed control

This paper proposes a novel nature-inspired algorithm called Multi-Level Hierarchy Optimization (MLHO) for solving optimization problems over continuous space. The MLHO algorithm is inspired by the hierarchy of nature, especially the hierarchy of biological populations. The entire algorithm structure is divided into four levels for iterative optimization, and the work of each level is global direction guidance, optimization-seeking task allocation, local optimal exploration, and broad domain exploration. Differential variation strategy and dynamic inertia factor are also designed to solve the problem of decreasing population diversity and slow convergence speed at the late stage of evolution. In order to validate and analyze the performance of MLHO, numerical experiments were conducted on benchmark problems in each dimension of CEC'20. In addition, comparisons with 4 state-of-the-art (SOTA) algorithms are executed. The results show that the performance of MLHO is significantly superior to, or at least comparable to the SOTA algorithms. At the same time, the feasibility and effectiveness of MLHO are also demonstrated for the speed control problem of the pipeline inner detector.