Aiming at the problem of low efficiency caused by the insufficient layout planning of heliostats in the new type of tower-type solar thermal power generation system

this study proposes a two-stage genetic optimization algorithm combining Monte Carlo ray tracing and collision detection based on the radial staggered distribution theory. On the basis of the traditional genetic algorithm

the algorithm updates the receiver position by feeding back the heliostat layout and further optimizes the heliostat parameters through genetic iteration. The algorithm takes into account energy loss factors such as the light cone effect and shading obstruction. Based on the two-dimensional normal distribution model

the energy of the rays inside the light cone is discretized to construct a Monte Carlo optical simulation model

and the optical efficiency of the system is evaluated using a collision detection algorithm. A simulation was carried out for a circular heliostat field with a radius of 350 meters in Jiuquan

Gansu

and a heliostat field layout with 2977 heliostats of different installation heights and sizes was designed. The shading obstruction efficiency can reach 96.623%

the optical efficiency is 71.358%

the average annual heat power output per unit area of the heliostat is 0.690271 kW/m²

and the average annual total heat power output of the heliostat field is 42.977625 MW. Compared with the improved grey wolf algorithm

the two-stage genetic optimization algorithm improves the four indicators of shading obstruction efficiency

optical efficiency

average annual heat power output per unit area of the heliostat

and average annual total heat power output of the heliostat field by 13.94%

27.59%

6.92%

and 29.04%

respectively.