On the Edge Embedded Fault-Detection in Electro-mechanical Systems
Industrial machines are often subject of wear and tear which can result in catassrophic situation, and limit the life and efficiency of the machine. The disturbances and damages to the machine may be undetectable for some time, but cause long term damage. Our project aims to be able to
2025-06-28 16:28:41 - Adil Khan
On the Edge Embedded Fault-Detection in Electro-mechanical Systems
Project Area of Specialization Artificial IntelligenceProject SummaryIndustrial machines are often subject of wear and tear which can result in catassrophic situation, and limit the life and efficiency of the machine. The disturbances and damages to the machine may be undetectable for some time, but cause long term damage.
Our project aims to be able to detect and classify the anomaly in machines, both domestic and industrial. The faults like defective bolt, unexpected current/voltage spike, torn canveyer belts etc.
Once the fault is classified, we can set it up for predictive maintenance. Thus, we can fix the machine before it leads to permanent damage.
Project Objectives- Detect any electrical and mechanical amomaly in industrial as well as domestic machines.
- Remote monitering of health of machines.
- Predicitive maintenance
The system consists of two nodes. A sensore node, and a processor node.
The sensor node consists of multiple sensors that collect data and wirelessly treansmits the data to the processor node.
The data from the sensor node is ued to train the a Machine Learning / Artificail Intellence algorithm, which is then deployed on the processor node. The process node will then process on the incoming data from the sensor node and will classify the state of the machine if it is faulty or not. and in case of faulty, it will be abl;e to classify the type of fault.
Benefits of the Project- Predicitve maintenance of industrial and dimestic machines
- Cost reduction for monitering
- Remote health monitering
The processor node in NVIDIA Nano Jetson, which will run a Convolution Nueral Network (CNN) or LSTM, depecnding upon the collected data.
The sensor niode is Arduino Uno proessor which is which controls the sensors and collect the data. Tne data is transmitted via ESP8266 wifi module to processor node.
The AI algorithm i.e CNN and LSTM will than process the data, and draw the conlcusion if the machine is faulty or not, and in case of any fault, it will classify the tyoe if fault.
Final Deliverable of the Project HW/SW integrated systemCore Industry ManufacturingOther Industries IT Core Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT)Sustainable Development Goals Affordable and Clean Energy, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 62500 | |||
| Sensors | Equipment | 15 | 500 | 7500 |
| Arduino Uno | Equipment | 1 | 1500 | 1500 |
| NVIDIA Nano jetson | Equipment | 1 | 32000 | 32000 |
| Test rigging system | Equipment | 1 | 7000 | 7000 |
| Breadboards/wire | Equipment | 10 | 300 | 3000 |
| overheads | Miscellaneous | 1 | 8000 | 8000 |
| motor | Equipment | 1 | 3500 | 3500 |