Lattice structures, characterized by lightweight properties, high specific strength and specific stiffness, and excellent vibration-damping and energy-absorbing efficiency, have been widely applied in the design of critical loadbearing components for high-end equipment in aerospace and deep-sea fields. However, for traditional lattice structures, the unit cells can only be described by a single parameter at the microscopic scale, and their distribution is confined to the assumption of uniformity at the macroscopic scale. This leads to underutilization of the design space, which restricts the enhancement of mechanical properties and fails to meet the stringent requirements for the extreme lightweight design of critical load-bearing components. In this paper, a multi-scale topology optimization design method for multi-parameter lattice structures is proposed. At the microscopic scale, the topological configuration design of multi-parameter lattice unit cells is achieved using an approximate model-assisted particle swarm optimization (PSO) algorithm. At the macroscopic scale, the topological distribution of lattice unit cells is optimized via a parametric level set-based topology optimization method. The proposed method realizes the coupling of microscopic lattice material design and macroscopic structural optimization, maximizing material potential and enhancing the mechanical properties of lattice structures. Numerical examples show that compared with traditional single-parameter lattice structures, the mechanical properties of the structures optimized using the proposed method are enhanced by 53.42%. Compared with lattice structures optimized via single-scale optimization (either microscopic or macroscopic), the performance is improved by 48.07% and 12.69%, respectively. This indicates that lattice structures designed with multi-scale optimization exhibit significantly superior load-bearing capacity. The proposed method significantly expands the design space of lattice structures and effectively enhances their mechanical properties, thus holding significant application potential for the lightweight design of structural components in key fields like aerospace.