With the development of urban electrification and renewable energy technologies, distributed power sources play an important role in distribution systems, but their random volatility and uncertainty bring challenges to system stability. To solve this problem, this paper proposes a new Probabilistic Load Flow(PLF) calculation method based on the concept of Cyber-Physical System(CPS) integration. This method comprehensively perceives the dynamic operation of the system through the CPS fusion model, takes into account the random volatility of distributed power generation, constructs the output probability model of wind and photovoltaic power generation, and uses the Copula theory to establish a joint model of the output correlation of wind and photovoltaic power plants. In addition, a Gaussian mixture model is proposed to reveal the probabilistic characteristics of the load. On this basis, a new probabilistic power flow calculation method is proposed by adopting the probability density evolution equation and using the finite difference method for efficient solution. This method provides an accurate and efficient calculation approach for power distribution systems considering distributed generation, which helps to ensure the safe and stable operation of the system.