With the large-scale penetration of renewable energy, the safe and stable operation of power grid and economic dispatch are facing great challenges. How to realize accurate perception of internal load characteristics of power users is an important technical difficulty to support power demand side management. For this reason, this article is based on extensive IOT technology in electricity, and a non-invasive power load monitoring method (NILM) based on cloud edge collaboration is proposed. Through the two-level architecture of “edge identification” and “cloud correction”, the method effectively overcomes the contradiction between the weak computing capability of edge terminal and the heavy communication pressure of cloud master station. The experimental results show that the method can effectively improve the accuracy and reliability of traditional NILM method by considering the influence of external factors.

This paper introduces a smart grid environment authorization service architecture based on XACML and SAML standards. Software prototypes are deployed with the platform as a service (PaaS) framework Appscale. Appscale is an open source AppEngine framework released by Google that can execute applications in the cloud to solve the problems caused by the distributed computing model. This paper mainly evaluates its scalability. Through the different configuration of all distributed databases, the test of the system running load is completed. The test results show that the number of nodes deployed by Appscale varies from very unstable behavior to waiting scalability.