This App allows to perform regression analysis from two variables. You will be able to observe trends, dependencies between variables and detect outliers.
Getting startedTo create a new scatter plot, just set up the following configuration options as desired. Once configured, click on "Draw Scatter plot" button to get the scatter plot.
- Frequency selector: select desired frequency you want to use in the scatter plot (hourly, daily, weekly or monthly). Parameters will be filtered based on this selection.
- Date range picker: select the period of time for the scatter plot analysis. You can use the presets available to move quickly (today, yesterday, last 7 days, last month...)
- X and Y axis devices and parameters: specify which device and parameter you want to use in the X and Y scatter plot axis. Remember that parameter is filtered by the above selected frequency.
- "Cut at" values: if any value introduced, Scatter plot App will divide the X axis in two zones, getting two different regression formulas. For example, if you are analyzing a whole year for Temp vs consumption, you can split the regression for Winter and Summer.
- Weekday filter: You can analyze separately your regression trends filtering by weekday. Pick the days that are included in the “Red” or “Yellow” regressions, or unselect them if you don’t want to include.
- Scatter plot results: once generated the scatter plot, you will see the analysis results at the bottom of the page, separated by scatter filters (Red or Yellow). R² values are shown in brackets
- Total Energy: amount of energy analyzed. Basically, is the sum of each value on Y axis
- Total Waste: quantification of the differences between regression lines and Y values, can be assimilated to potential wasted energy.
- Potential savings: percentage of waste energy versus total energy
ExampleIn the following example, we are going to compare the main consumption versus the Outdoor temperature. We select 1 year of data and daily frequency. In this example, we will filter by week days and split the regression.
Conclusions: If our facility is closed during Sundays, we can uncheck and hide those days from the analysis. However, we can select Sundays in the Yellow regression, to analyse those particular days. Changing these set up, we can get more accurate regressions.
In this example, Scatter plot App is calculating a Potential Savings of 6%. With the mouse cursor we can detect which are the particular days where the consumption exceeded the expected