Quick Start 2: Begin Analyzing Your Engine Data

APPROXIMATELY 12 MINUTES TO COMPLETE

Important – Make sure you have completed Quick Start 1 before starting Quick Start 2.

For this Quick Start, you’re going to use some of the insight gained from Quick Start 1 and create some simple calculations for the bearing temperature data. These calculations will enable you to track and identify important data aggregations in bearing temperature. We suspect there may be issues with Cylinder 3 because we’re using a new bearing manufacturer. Since Cylinder is 8 is the most stable cylinder, we will use that as a baseline. Also, we want to compare how Cylinder 3 is doing compared to the average of all the cylinders.

Analyzing Devices using computations

In this Quick Start, you create queries that direct AXON Predict to read the sampled engine data and calculate the average, maximum, and minimum bearing temperatures.

For all cylinders collectively (except Cylinder 3), AXON Predict will calculate the:

  • Average bearing temperature during every 30-second time window.
  • Minimum bearing temperature during every 30-second time window.
  • Maximum bearing temperature during every 30-second time window.

Then, it performs the same calculations for Cylinder 3, separately.

You’ll use these computed values to compare the cylinder group against Cylinder 3, to detect differences in that device. Then, you’ll graph these values using three multi sensor views to get a visual representation of those calculation outputs.

VIDEO FOR THIS QUICK START

The steps

  1. Browse to your AXON Predict instance. First, you will create a new dashboard.
  2. Click the Dashboard Options button  and select New Dashboard from the drop down list.
  3. Click Open Blank Dashboard. A new dashboard will open.You are ready to start creating your compute questions.
  4. Click the Ask a Question about your data button . The Questions panel displays:
  5. Click Add. This opens the question wizard, which guides you through the process for creating analytics questions for your devices and data.
  6. Under Patterns Gallery, click Compute and drag the compute over time template to the "Drop analytics here" workspace.

    The equation builder appears, prompting you to select a device group.

    Note: The Pattern Query panel shows the (read-only) JSON configuration for the selected compute template. When you are finished creating the compute question, the finalized configuration is submitted to AXON Predict to run the analytics question. For example:

    As you recall, in Quick Start 1 you created two device groups: one with all devices and one with only Cylinder 3. In Quick Start 2, you will create one more device group: for all devices except Cylinder 3.

  7. In the device group box, click Create New Group. This opens the New Device Group page.
  8. In the displayed New Device Group page, use these steps to create a device group containing 7 of the 8 cylinder devices:
    1. Click System Organization.
    2. Click Group by device. (The list shows all 8 cylinder devices.)
    3. One by one, drag each cylinder device - EXCEPT for Cylinder 3 - to the workspace.
    4. Type a name for the device group: 7Cylinders_group.
    5. Click Create.

    The process to create the device group looks similar to this:

  9. Select to persist the device group data. (Note that other device groups were already set to persist data. If you find that they have stopped persisting data, make sure to set them to start persisting data again.)

  10. Click TAKE ME BACK. You see the equation builder with all defined device groups.
  11. Select the new device group, 7Cylinders_group, and click Apply.

    First, you're going to create the average compute for devices defined by the 7Cylinders_group. This question needs to calculate the data for the group as a whole rather than for individual devices in the group.

  12. Change the window to calculate every 30 seconds and press Enter. Leave the sliding time period at 1 second.

  13. In the Group by field, select All.
  14. Create the equation as follows:
    1. Click in the equation text box, then click the avg key.
    2. From the list of displayed sensors, select BearingTemp.
    3. Click the ) key. The equation is complete and appears similar to the following:
    4. Click CREATE.
  15. Type a descriptive pattern name, such as avgTemp_7Cylinders, and click Create.

    You see the message, "Question successfully created."

    (Note: For these Quick Starts, all questions you create need to persist data. You leave the default selection for "Save history for this pattern.")

    Now you will create two more questions for the 7Cylinders_group using the same template, compute over time.

  16. Click RESET, add the compute over time template to the workspace, and follow steps 11-15, but instead use the calculator keys to create an equation that calculates the maximum BearingTemp for the 7Cylinders_group every 30 seconds.

    Type a descriptive pattern name, such as maxTemp_7Cylinders, and click Create.

    Your equation will look similar to the following:

  17. Click RESET, add the compute over time template to the workspace, and follow steps 11-15 but instead use the calculator keys to create an equation that calculates the minimum BearingTemp for the 7Cylinders_group every 30 seconds.

    Type a descriptive pattern name, such as minTemp_7Cylinders, and click Create.

    Your equation will look similar to the following:

    You are finished creating compute questions for the 7Cylinders_group. Now you will create the same three compute questions again, but this time you will use data for the Cylinder 3 device.

    This question will calculate the data for the individual device, rather than as a group.

  18. Click RESET, and do the following:
    1. Add the compute over time template to the workspace.
    2. Select the Cylinder3_group (you created this device group in Quick Start 1), and click Apply.
    3. Click Group by Device.
    4. Create an equation that calculates the average BearingTemp for the Cylinder3_group every 30 seconds (using the calculator keys). The equation will look like this: avg(BearingTemp).
    5. Type a descriptive pattern name, such as avgTemp_Cylinder3, and click Create.
    Your equation will look similar to the following:
  19. Click RESET, and repeat step 18 but instead use the calculator keys to create an equation that calculates the maximum BearingTemp for the Cylinder3_group every 30 seconds.

    Type a descriptive pattern name, such as maxTemp_Cylinder3, and click Create.

    Your equation will look similar to the following:

  20. Click RESET, and repeat step 18 but instead use the calculator keys to create an equation that calculates the minimum BearingTemp for the Cylinder3_group every 30 seconds.

    Type a descriptive pattern name, such as minTemp_Cylinder3, and click Create.

    Your equation will look similar to the following:

    You are finished creating questions for this Quick Start.

    Now, you will view the analytics calculated by AXON Predict and compare the results to get insight into your devices. You need to open the realtime dashboard and then add visualization components to the page.

  21. Select Realtime.

    If you see Multi Sensor (defaulted from Quick Start 1 dashboard), then you will remove it from this Quick Start 2 dashboard.

  22. If applicable, locate the Muli Sensor component in the dashboard and click "X" (in the top, right corner) to delete it.
  23. Select Add a visualization component and, from the list of components, select and drag the Multi Sensor Explore and Performance Explore components to the workspace, one at a time.
  24. Click Ask a Question about your data. You see all the Compute questions you created. (If you click a question, you can see some of the details.)

    Now, you will begin analyzing the data and results of the compute questions to understand how Cylinder 3 (the suspect cylinder) compares with Cylinder 8 (your baseline cylinder).

  25. Click Discover Devices, then click Sensors (if needed), and drag the BearingTemp sensors from Cylinder 3 and Cylinder 8 to the Multi Sensor Explore component.
  26. Click Ask a Question about your data, and drag the Average compute for the 7 cylinders group to the left side of Performance Explore.
  27. Now, drag the Average compute for the Cylinder 3 group to the right side of Performance Explore. You see the data displayed in the two components:  

    (The Legend is shown in both of these images so you can see which devices and data are included in each component. You'll see your sensors are listed in the order in which you added them to the component.)

  28. Analyze the data in the components as follows:
    • In the Multi Sensor Explore, set the time period to compare the historical bearing temperature for the two cylinders.  (Note that you may not see data in the component until you select a time period.)
    • In the Performance Explore, set the time period for both sides of the component and see how the average bearing temp for cylinder 3 compares to the average bearing temp for the group of cylinders. You can view data from different time periods. For example, you may want to look at the last hour of data for both devices, and then zoom in on the last 5 minutes of data.
  29. You can add more components (like Multi Sensor, Multi Sensor Explore, or Performance Explore) to the dashboard and drop the other computed values in them to perform more analysis and comparisons.

    Your goal is to determine how Cylinder 3 data deviates from, or aligns with, the data for the other cylinders. To do this, you review the computed values from different time periods.

    Video: Reviewing data in multiple components

    As a last step in the Quick Start, you will save the dashboard so you can reuse it later.
  30. Click the Dashboard Options button  and select Save Dashboard from the drop down list.
  31. In the displayed Save Dashboard page, type a name for your dashboard and click Save.

Quick Start 2 is Complete!

Using the computed calculations and data visualizations, you can see how the bearing temperature for Cylinder 3 and the other cylinders varied during the sample data time period. 

In Quick Start 3, you will use these calculations to determine if and when a recently serviced device (Cylinder 3) is experiencing an issue with performance. You’re going to create analytics to look for specific temperature patterns in your devices. 

 

Important – Make sure you have completed Quick Start 2 before starting Quick Start 3.