/SmartHome-HouseholdAppliance-usageRecommender-andMonitoringSystem-HARMS

The system partially implementing H/S approach, feeded with two different dataset, the recommendation system implemented using pattern mining algorithm(SVM), sql queries, and Apriori algorithm

Primary LanguagePHP

SmartHome-HouseholdAppliance-usageRecommender-andMonitoringSystem-HARMS

KSA electricity consumption continues to increase as the population grows, leading to a higher amount of electricity consumed per person. There are several factors contributing to this problem such as, appliances are left unattended and misuse of lights. The increase in consumption caused by these factors has a significant impact on the increasing monthly electric bill. With this, we developed HARMS, a monitoring and recommendation system for household appliances that can help reduce energy consumption. HARMS offers a smart solution that uses machine learning (ML) and a recommender system to track inhabitant's normal and unusual behavior in terms of appliances usage to manage energy consumption. Despite the fact that only 16% of the population in Saudi Arabia uses an energy consumption system to save energy, the KSA General Authority for Statistics reported in 2019 that 46% are willing to buy a power saver or energy conservation system. This motivates us to develop the system to cater for those who are willing to conserve energy to reduce monthly electric bill and to aid them in monitoring and managing their electric consumption.

HARMS will be divided into four modules: (1) Mobile application that will serve as an intermediary software in between system operations to provide real time appliances recommendation to use, monitoring appliances usage and energy consumption. (2) Central Control Module (CCM), it will process data from censors to provide recommendation to the user. Monitors and computes energy consumption which can be accessed using the system dashboard. The recommendation is achieved by using the Apriori algorithm and SVM. The Apriori is used to extract the user's patterns and based upon, the generated association rule will be fed to SVM to generate a recommendation. (3) Sensor module, which are interconnected modules that plug with appliances to get data from each appliance and send it to the CCM. (4)Lastly, the simulation module, which will be divided into two parts, first part will represent the home environment, and second part which control panel that will represent the home appliances and sensors.

We demonstrate our ongoing work on HARMS in which studies the daily inhabitant's energy consumption behavior. The system composes of hardware and software that can be handled and utilized to consume energy more efficiently. It will be developed to fit and maximize the inhabitant's comfort level through monitoring the inhabitant’s energy usage. The inhabitant will be fully aware of all of the appliances that turn on via receiving frequent notifications on the Android application. In addition, the system will notify the user of any appliance he forgot to turn it off. Thus, it will utilize electricity efficiently and not be consumed at unnecessary expenses. User awareness of the devices used, and the average power consumed weekly would drive to cut down the high electricity cost and optimize the energy efficiency. It is expected from the system to reduce the energy consumption by real-time monitoring and controlling different devices automatically. That will lead to reducing the monthly electricity bill for the customer.

https://drive.google.com/drive/folders/1vkLabu6oQPpNqG0oMITVfVaQ8zAiJJwm?usp=sharing