A local energy company would like to achieve the following goals:
- Improve customer satisfaction
- Reduce costs
- Improve efficiency in its operations
- Decentralize decision making
In order to achieve the above objectives, the energy company would like to implement a smart meter program whereby energy consumption data from homes is continuously gathered and analyzed at the city, county, and state levels. Using the insights at the various levels the following abilities are anticipated to be enabled:
- At the city level, an improvement in the ability to rapidly respond to issues, resulting in improved customer satisfaction.
- At the county level, an improvement in the ability to efficiently manage the supply and demand of energy across its cities, resulting in improved operational efficiency.
- At the state level, the ability to use past historical insights and predict potential overloading and black-outs, resulting in reduced operation costs.
The energy company would also like to correlate energy consumption data with exogenous factors (weather data) that will enable it to make better decisions.
So in summary, our use-case scenario will process an input stream of smart meter data, along with an input stream of local weather data, and analyze these inputs to provide the required insights:
Data Format: Smart Meter Energy Utilization
The sample smart meter data is generated at each home every second. The sample data is provided in a comma-separated file with the following format:
|Timestamp||Smart Meter ID||City Code||Utilization|
Here's an example of the raw smart meter data:
Data Format: Weather Data
Similarly, the sample weather data is generated every second for each city, and includes temperature as well as humidity and precipitation. The sample data is provided in a comma-separated file with the following format:
Here's an example of the raw weather data:
Go to the next page for HSDP Tutorial #1: Simple Streaming Query.