Abstract: With the implementation of the emission peak and carbon neutrality national strategy,distributed PV power stations,have developed rapidly as an effective supplementary application application from of centralized PV power stations. Due to the fact that distributed PV power stations are mostly built on the roofs of buildings such as factories,parks,and stations,they face difficulties in cleaning PV modules,such as a high risk of falling from heights by cleaning personnel,multiple cleaning times. This paper takes a rooft distributed PV power station as an example,first analyzes the problems existing in traditional manual cleaning method,and then proposes a solution to use intelligent cleaning robots for cleaning,and verifies the actual application effect of intelligent cleaning robots. The research results show that the total power generation capacity of PV arrays cleaned by intelligent cleaning robots can be increased by 21.2%. Under this increase,for a 5.324 MW roof distributed PV power station,using intelligent cleaning robots can increase the monthly power generation capacity by 637400 kWh compared to manual cleaning methods,the annual increase in power generation capacity revenue is 507700 yuan,which means that the initial investment cost of intelligent cleaning robots can be recovered in 1～2 years. In the hope that the research results can provide reference for improving the operation and maintenance efficiency and reducing the operation and maintenance costs of distributed PV power stations.