A Cloud-Integrated Appliance Recognition Approach over Internet of Things

Abstract

The production, distribution, and consumption of energy have been receiving a lot of attention in the past few years. We have witnessed the emergence of an increasing number of energy-saving technologies and standards aimed at saving energy consumption of home appliances. To reap higher energy savings, home energy management systems were attempted to monitor and coordinate the individual energy saving activities of home appliances cost-effectively. To achieve this, the working status of each power load has to be recognized and then synchronized at run-time. In this context, low-cost and stand-alone electronic home appliance recognition technologies have been widely explored to identify different types of appliances being used and analyze the power consumption of appliances' operations. It is common that many appliances are used at the same time. However, these recognition technologies cannot accurately identify electronic home appliances operating in parallel at run-time. Diverse, multiple electronic home appliance recognition technologies encounter a range of new challenges. To address these challenges, we have proposed an algorithm that recognizes multiple diverse electronic home appliances concurrently via wave-form recognitions at run-time. To address deployment and response delay issues, we designed a prototype system that includes a smart socket, a non-invasive data acquisition module, and an Internet of Things (IoT) cloud-enabled back-end which provides scalable communication and computation capacity. In contrast to existing systems, the proposed system uses an embedded system that has a low energy overhead and allows high scalability. To evaluate our system, we conducted our experimental tests in an environment consisting of daily home appliances. The experimental results show that the total recognition rate of appliances operating in parallel can reach 86.14% compared to the recognition rate of single appliance which can reach 96.14%.