Reliability evaluation is an important foundation for power distribution system planning and operation. With the rapid development of new power systems

the scale access of multiple types of distributed generation has become the future development trend of power distribution systems. Intermittent renewable energy sources such as wind turbines and photovoltaics

as well as variations in the risk of component failures due to environmental factors

increase the system uncertainty. Based on the probability expectation and probability density evolution methods

this paper proposes an analytical method for reliability evaluation of distribution systems with multiple distributed generations under long-term and short-term time-varying failure rates to evaluate the reliability more accurately and efficiently. The short-term evaluation is categorized into two scenarios

general weather and extreme weather

so as to evaluate the instantaneous and average reliability of the distribution system. The effectiveness and efficiency of the proposed method is verified in the evaluation of the modified IEEE 123-node test system.