Smart Meter Synthetic Data Generator development in python using FBProphet

Datasæt

Beskrivelse

Data-science is a key component of modern science since it fuels AI, data analytics, etc. As the electrical grid has been modernised into a smart grid, it has also become increasingly dependent on data science to monitor and control grid activity. Realistic data is essential to evaluating the algorithm's workability but it is difficult to obtain real smart meter data due to strict privacy and security policies of many countries. Using the prophet library synthetic data sets are generated in prediction-based Synthetic Data Generator GUI. For that source CSV (real-time) file is used to generate synthetic data in CSV format depending upon the number meter and number days to be calculated. Using FB prophet, time series data can be forecast based on an additive model that integrates seasonality, yearly, weekly, and daily trends, as well as holiday effects into non-linear trends. The algorithm is most effective when there are several seasons of historical data and strong seasonal effects in the data series. FB prophet provide automated forecast for the time series data along with seasonality and trend removals.
Dato for tilgængelighed2022
ForlagCode Ocean

Citationsformater