Abstract: The concept of edge computing model and its advantages in data processing and analysis are studied in view of the problems that exist in real-time,privacy protection and energy consumption in data analysis and processing of cloud computing model. A new edge computing system architecture is proposed,and unloading,operation and scheduling strategies are designed which can complete the tasks of the system.According to the improvement direction and application prospect of the proposed new edge computing system,a new power data acquisition system based on cloud edge-end collaboration is constructed through analysis and research. The system architecture is divided into three layers from top to bottom: the master station of the cloud acquisition system,the intelligent terminal at the side and the meter equipment at the end side. The meter equipment is responsible for collecting power data and reporting it to the intelligent terminal at the side. The intelligent terminal at the side is responsible for preprocessing the data,while the master station of the cloud acquisition system is responsible for collecting the preprocessed data and processing it in a unified manner. Based on the new cloud-edge-end collaborative data acquisition system proposed in this paper,the master station of the cloud acquisition system hosts intelligent terminal devices at the side end, and supports sending applications to the intelligent terminal at the side end for edge calculation. The intelligent terminal device at the side end manages the meter device,and supports uploading the data collected by the meter device to the edge end for cache and nearby processing,analyzing and decision-making. This paper verifies that the proposed new cloud-edge-end collaborative power data acquisition system can not only solve the problem of real-time processing and response of edge applications,but also reduce the bandwidth pressure of data communication between cloud and edge end.