Data Challenge

Prognosis of Rotor Parts Fly-off

Background

Large rotating machines, such as compressors, steam turbines, gas turbines, etc., are critical equipment in many process industries such as energy, chemical, and power generation. Due to the high rotating speed and tremendous momentum of the rotor, the centrifugal force may lead to the loose or flying apart of the rotor parts, which brings a great threat to the operation safety of the equipment. The fault usually causes damage to multi-stage stationary and non-stationary blades or impellers. This type of fault could lead to high maintenance cost and significant economic loss. However, the capture of the early symptom of rotor faults is relatively difficult and has become a worldwide challenge in the field of prognostics. It will bring significant safety value and economic benefits if the big data-based methods can detect and predict incipient faults.

Competition task

Step 1: Identify which unit/units has/have the symptom of the centrifugal-force-led-components-flying-apart-fault (the target fault) according to the model trained by the training data.

Step 2: Sort the time sequence of the data according to the intensity of fault symptom.