Icon

Chapter 11

6 workflows in this folder and its subfolders

Icon
01_​Preprocessing_​Time_​Alignment_​and_​Visualization

September 2021 · Preprocessing,Time Alignment and Visualization This workflow preprocesses and visualizes sensor data for anomaly detection: Read FFT preprocessed data files with date, time, FFT frequency, and FFT amplitude Standardize the data by binning the frequencies and averaging the data by sensor, frequency b…

Icon
02_​Time_​Series_​AR_​Training

September 2021 · Time Series AR Training This workflow trains an auto-regressive model for anomaly detection: Filter the data to training data covering only normal functioning Loop over each frequency column at a time Train an auto-regressive model using 10 past values as predictors Calculate in-sample prediction er…

Icon
03a_​Time_​Series_​AR_​Deployment

February 2021 · Anomaly Detection. Time Series AR Deployment This workflow deploys a previously trained auto-regressive model for anomaly detection: Select the date for deployment. Two months of its past values must be available. Loop over each frequency column Apply the previously trained auto-regressive model to…

Icon
03b_​Time_​Series_​AR_​Visualization

September 2021 · Time Series AR Visualization This workflow visualizes the performance of previously trained auto-regressive models for anomaly detection: Filter the data to the maintenance window Loop over each frequency column Apply the previously trained model to the data Calculate 1st level alarms based on the p…

Icon
04_​Creating_​a_​ControlChart_​of_​a_​Time_​Series

February 2021 · Anomaly Detection. Control Chart This workflow performs anomaly detection using a control chart: Calculate the "normal conditions" as the average +/- 12 times the corresponding standard deviation Raise a 1st level alarm if a sensor exceeds the band for normal conditions on a single frequency band Ra…

Icon
Send_​Email_​to_​start_​checkup

August 2018 · Send email to start checkup This workflow simply sends an email. It is called via a Call Workflow node from a parent workflow. TAGS: anomaly detection,time series analysis,IoT,Internet of Things