It accurately predicts the cooling demand of the building at different times of the day, and combines the periods of high and low electricity prices (peak-valley price difference) with the operating efficiency rules of the air-conditioning host itself to automatically calculate the optimal solution: when to directly use air-conditioning for cooling, when to use off-peak electricity at night for cold storage and release cooling during peak electricity prices in the daytime, while ensuring that the air-conditioning always operates at the highest efficiency. The superposition of multiple optimizations improves the energy efficiency ratio of the cold source system operation, and ultimately saves 35% of the operating costs.