An Extensible Approach for Non-intrusive Load Disaggregation and Occupancy Detection using Smart Meter Data with Deep Machine Learning Techniques.
Energy saving becomes one of the significant topics because of the spread of climate and energy challenges globally. In recent times, the demand for electricity has increased in households because of the use of household appliances which becomes a major concern as the energy generation does not increase that much. In this situation it is necessary to track the consumers’ daily power usages in house to save and control this resource. This project emphasis on smart metering data which provides information about the consumption of electricity for household appliances. We will use deep machine learning techniques to analyze the usage pattern of the appliances. Non-intrusive load monitoring (NILM) is a set of methodologies and techniques that analyze the total aggregated smart meter measurements and break them down into the individual consumptions by devices present in a household which is called appliance modeling. With the help of NILM the detail information about the detail consumption of individual device will sent to the household owner thus they can change their energy consumption behavior. With the smart meter data it is also possible to detected occupancy as the home’s pattern of electricity usages normally changes when occupants are present due to their interact with electrical loads. The overall load during peak hour becomes too much higher than the generation of electricity and often load-shedding occurs. With the process discussed here, it is possible to change the consumers’ behaviors of appliance usages which will definitely reduce the peak time load. The main goal of this project is to apply deep learning techniques on smart meter data to make the best use of energy generation by rescheduling of using home appliances and reducing misuse of electricity.
Funding Agency: Khulna University Research Cell.
Total Amount: 240000/=
|25 June, 2019