A motif-based classification algorithm for identifying solar panel installations

With increasing energy requirements and limitation of non-renewable resources for traditional electricity generation and transmission, many households and premises across the world have installed solar systems. Power companies require information about solar panel installations to regulate the whole power system. In this paper, we propose a motif-based classification algorithm for identifying whether a customer has installed the solar panels. Firstly, we symbolize our time-series data with alphabets and classify those data. Then we evaluate our method by checking error rates of different settings. Later, we test our algorithm with different training and testing datasets. The motif-based classification algorithm analyzes electricity consumption data of households. Results show that our motif-based classification algorithm for identifying solar panel installations have a very good accuracy.