Generate engine-failure predictions and gain insight into your data with machine learning.
This guide will upload captured data from an Edge MicroServer (EMS) "Engine Simulator" to ThingWorx Analytics Builder.
Following the steps in this guide, you will create an analytical model, and then refine it based on further information from the Analytics platform.
We will teach you how to determine whether or not a model is accurate and how you can optimize both your data inputs and the model itself.
NOTE: This guide's content aligns with ThingWorx 9.3. The estimated time to complete ALL parts of this guide is 60 minutes
YOU'LL LEARN HOW TO
Load an IoT dataset
Generate machine learning predictions
Evaluate the analytics output to gain insight
Step 1: Scenario
In this guide, we’re continuing the same MotorCo scenario, where an engine can fail catastrophically in a low-grease condition.
In previous guides, you’ve gathered and exported engine vibration-data from an Edge MicroServer (EMS).
The goal of this guide is to now import that previously-exported Comma-Separated Values (.csv) data intoThingWorx Analytics, and thencreate an analytical model for predictive maintenance.
Analytical model creation can be extremely helpful for theautomotive segmentin particular. For instance, each car that comes off the factory line could have an EMS constantly sending data from which an analytical model couldautomatically detect engine trouble.
This could enable your company to offer anengine monitoring subscription serviceto your customers.
This guide will show you how to build an analytic model of your engine to facilitate this monitoring service.
Step 2: Upload Simulated Data
This guide assumes that you are using either thehosted trial(with has both Foundation and Analytics pre-installed) or a combination of the Foundation and Analyticsdownloadable installers.
To confirm that Foundation is communicating with Analytics, perform the following steps:
On the ThingWorx Foundation left-side navigation column, clickAnalytics > Analytics Builder > Settings.
At the top-right in theAnalytics Server Versionfield, ensure that you seean appropriate version number.
NOTE: If you use your own dataset, it's possible that you're results in the following steps will differ from those created by the provided-dataset.
If you were unable to generate a 30,000+ entry dataset in the last guide, then you may download testCSVfile.csv attached here,instead.
You will also need to download and extract vibration_metadata.zip which describes each column of the dataset.
On the left, clickAnalytics Builder > Data.
In theDataset Namefield, entersimulated_dataset.
In theFile Containing Dataset Datasection, search for and selecttestCSVfile.csv.
In theFile Containing Dataset Field Configurationsection, search for and selectvibration_metadata.json.
Note that the time it takes to import the dataset is determined by its size.
Step 3: Simulated Signals and Profiles
TheSignalssection of ThingWorx Analytics looks for themost statistically correlatedsingle fieldin the dataset which relates to your selected goal.
This doesn't necessarily indicate that it is the cause of your goal, whether maximizing or minimizing. It just means that the dataset indicates that this single field happens to correlate with the goal that you desire.
Unfortunately, our results aren't very good. Or, more accurately, they'retoo good.
Our simulated dataset has some noise in it from adding random values to our five frequency bands on each our two sensors. However, ThingWorx Analytics has instantly seen through that noise and discarded it. Instead, it's only detected that s2_fb5 isn't relevant.