Anomaly Detection based Condition Monitoring of UAV IC-Engine based on Deep Learning Approach

In this project we will design a Condition Monitoring System which will provide a data through which it will be easy to predict and observe health of an UAV IC Engine to prevent unexpected outages, improve overall reliability and reduce repairing costs. This Condition Monitoring System is bas

Project Title

Anomaly Detection based Condition Monitoring of UAV IC-Engine based on Deep Learning Approach

Project Area of Specialization

Artificial Intelligence

Project Summary

In this project we will design a Condition Monitoring System which will provide a data through which it will be easy to predict and observe health of an UAV IC Engine to prevent unexpected outages, improve overall reliability and reduce repairing costs.

This Condition Monitoring System is based on an ‘Anomaly Detection Based on Deep Learning Approach’. Anomaly detection in high dimensional data is becoming a fundamental research problem that has various applications in the real world. Many existing anomaly detection techniques fail to retain sufficient accuracy due to so-called “big data” characterized by high-volume, and high-velocity data generated by variety of sources. That’s why we are approaching anomaly detection through deep learning approach. Deep Learning network allow the abstraction of complex problems and enable more accuracy of fault diagnosis and prognosis (e.g., remaining useful life prediction).

Deep Learning has shown superior ability in feature learning, fault classi?cation and fault prediction with multilayer nonlinear transformations. We are focusing on Auto-Encoder (AE), which is a neural network model that uses a function to map input data into their short/compressed version subsequently decoded into a closest version of the original input.

The Condition Monitoring System will continuously monitor the behavior of IC Engine and it will predict the life span of the engine accordingly. If engine will behave abnormally, Anomaly Detection will consider it as an error and will provide a data about the remaining life of the engine, this will prevent any unplanned downtime and a costly repairment of the IC Engine

Project Objectives

1. To Develop and Test deep Auto-encoders on Simulated vibration by the mid of Q2
2.  To Develop a Test Rig on UAV Engine with vibration sensors and Data Acquisition system and fault insertion mechanism by the mid of Q3
3. To Integrate Auto-Encoder Model with vibration data from Test Rig by the start by the mid of Q4

Project Implementation Method

A systems engineering approach will be utilized towards the design and development of the proposed project. The project will be broken down in to modules and submodules spanned across the four quarters.

Q1: Development of deep autoencoder models using offline vibration dataset.

Q2: Development of test setup consisting of UAV IC engine, MEMS vibration sensors and data acquisition system. In this phase, vibration data will be logged and dataset will be prepared and pre-processed (cleaned).

Q3: In this phase, deep autoencoders will be trained in the dataset gathered in the Q3. The autoencoders will be tested in this phase.

Q4: Development of online anomaly detection system

Benefits of the Project

Most engine or systems of avionics don’t approach predictive maintenance because they are replaced when they are malfunctioned, sometimes these errors occur in between their flight which can also result in devastated ways. The project is designed to prevent any unplanned downtime of machinery equipment or devices like engines or any avionics system, it would degrade or interrupt a company’s core business, potentially resulting in signi?cant penalties and unmeasurable reputation loss.

Condition monitoring of IC Engine will monitor the health of engine through Anomaly Detection and will provide a detailed output of its running condition and if any abnormality is detected, Anomaly Detection will provide how long can an engine operate without any chances of causalities or failure of Engine

Technical Details of Final Deliverable

The final deliverable will consist of following.

1. Deep autoencoder algorithm for anomaly detection
2. Online data acquisiton and anomaly detection system with IC engine test rig

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Transportation

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Good Health and Well-Being for People, Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com