Conundrum is an ISV partner for NVIDIA in the Industrial AI Predictive Maintenance domain. This container is a demo container using Conundrum's Deep Learning based AI platform that uses CUDA 10 with PyTorch and TensorFlow DL frameworks. The demo involves real-time data simulation of randomly picked aircraft from NASA's FD001 dataset, Inference model for failure probability and a dashboard with widgets displaying engine malfunction warnings, sensor influence, model quality and GPU usage.
The training model is using time-series data and RNN/Recurrent Highway Networks.
• In the Tags section, locate the container image release that you want to run.
• In the Pull command, click the icon to copy the docker pull command.
• Open a command prompt and paste the pull command. The pulling of the container image begins. Ensure the pull completes successfully before proceeding to the next step.
• Get IP addr of the system - look for eth0 - 10.#.#.# (command: ip addr)
• Run the image using this command: docker run -p 80:8000 -e USER=nvidia -e PASSWORD=nvidia2019 nvcr.io/nvidia/conundrum-aircraftengine-cuda10-pytorch11:latest
• Start Chrome web browser and connect to container using IP address from earlier step - for login: user=nvidia, pw=nvidia2019 (URL: http://enter IP addr here)
Reference: A. Saxena and K. Goebel (2008). "Turbofan Engine Degradation Simulation Data Set", NASA Ames Prognostics Data Repository (https://ti.arc.nasa.gov/tech/dash/groups/pcoe/prognostic-data-repository/ case #6), NASA Ames Research Center, Moffett Field, CA
Data Set: FD001 Train trajectories: 100 Test trajectories: 100 Conditions: ONE (Sea Level) Fault Modes: ONE (HPC Degradation)
Data Set: FD002 Train trajectories: 260 Test trajectories: 259 Conditions: SIX Fault Modes: ONE (HPC Degradation)
Data Set: FD003 Train trjectories: 100 Test trajectories: 100 Conditions: ONE (Sea Level) Fault Modes: TWO (HPC Degradation, Fan Degradation)
Data Set: FD004 Train trjectories: 248 Test trajectories: 249 Conditions: SIX Fault Modes: TWO (HPC Degradation, Fan Degradation)
Data sets consists of multiple multivariate time series. Each data set is further divided into training and test subsets. Each time series is from a different engine – i.e., the data can be considered to be from a fleet of engines of the same type. Each engine starts with different degrees of initial wear and manufacturing variation which is unknown to the user. This wear and variation is considered normal, i.e., it is not considered a fault condition. There are three operational settings that have a substantial effect on engine performance. These settings are also included in the data. The data is contaminated with sensor noise.
The engine is operating normally at the start of each time series, and develops a fault at some point during the series. In the training set, the fault grows in magnitude until system failure. In the test set, the time series ends some time prior to system failure. The objective of the competition is to predict the number of remaining operational cycles before failure in the test set, i.e., the number of operational cycles after the last cycle that the engine will continue to operate. Also provided a vector of true Remaining Useful Life (RUL) values for the test data.
The data are provided as a zip-compressed text file with 26 columns of numbers, separated by spaces. Each row is a snapshot of data taken during a single operational cycle, each column is a different variable. The columns correspond to: