#IOT technology #Author: Aline Jaimes
Smart building technology allow buildings to be more efficient, comfortable and easier to manage by connecting IOT devices, collaborate, analyze and use real time intelligence with IOT applications and solutions as part of a Smart Building platform. A challenge for the implementation of such technologies is the evaluation of the energy use efficiency under variable climatic conditions during a year.
#GOALS:
- Evaluate and predict energy efficiency performance from solar energy during a year and variable weather conditions. Tools to be used: Python Pandas, Scikit-Learn, Matplotlib
- Evaluate in what state of the USA solar energy is more or less efficient Tools to be used: Plotly or leaflet or D3 or Tableau (To be determined...)
#SPECIFICATIONS The project uses one year data set (HomeC) that contains the readings with a time span of 1 minute of house appliances in kW from a smart meter and weather conditions of that particular region. My approach will gather weather data from different states/climatic regions. The data will be used to train and parameterize the predictive models. #The variables to analyze are: time use [kW]Total energy consumption gen [kW]Total energy generated by means of solar or other power generation resources House overall [kW]overall house energy consumption Dishwasher [kW]energy consumed by specific appliance Furnace 1 [kW]energy consumed by specific appliance Furnace 2 [kW]energy consumed by specific appliance Home office [kW]energy consumed by specific appliance Fridge [kW]energy consumed by specific appliance Wine cellar [kW]energy consumed by specific appliance Garage door [kW]energy consumed by specific appliance Kitchen 12 [kW]energy consumption in kitchen 1 Kitchen 14 [kW]energy consumption in kitchen 2 Kitchen 38 [kW]energy consumption in kitchen 3 Barn [kW]energy consumed by specific appliance Well [kW]energy consumed by specific appliance Microwave [kW]energy consumed by specific appliance Living room [kW]energy consumption in Living room Solar [kW]Solar power generation temperature humidity visibility apparentTemperature pressure windSpeed cloudCover windBearing precipIntensity dewPoint precipProbability
#RESEARCH QUESTIONS
- How much the E-gen-Total energy generated (solar or other power generation resources) varies during the year under different weather conditions?
- Can random forest analysis determine what are the most important variables that regulate E-gen. For instance, Air temperature, humidity, net radiation, wind speed, pressure, etc. At what time frequency machine learning unsupervised models predict E-gen more accurately (monthly, daily, minutes) and why?
- What states are more likely to use E-gen more and less efficiently? Ultimately, calculate energy efficiency during a year and compare different states. Energy efficiency is defined as energy output, divided by the energy input, and expressed as a percentage.