Mechanical Engineering Undergraduate | IIITDM Kurnool

Developing engineering capabilities in the physical-to-digital data pipeline for rotating machinery health monitoring through signal processing, embedded sensing, and data-driven diagnostics.

Repository: Tier_3_PDm_sys

Purpose

Python-based framework for deterministic vibration analysis and machinery fault diagnostics.

Key Capabilities

• Time-domain feature extraction (RMS, Kurtosis, Crest Factor)
• FFT spectral analysis and Hilbert Transform envelope analysis
• Bearing defect frequency calculations (BPFO, BPFI, BSF, FTF)

Technology

Python • NumPy • SciPy • Pandas • Matplotlib

Current Scope

Validated using controlled simulated vibration signals to verify diagnostic algorithms prior to hardware implementation.

Repository: DRP project

Purpose

Experimental edge-to-cloud monitoring platform for acquiring, storing, and visualizing multi-sensor machine data.

Key Capabilities

• ESP32-based multi-sensor data acquisition
• REST API backend for sensor telemetry
• Cloud data logging and interactive web dashboards

Technology

ESP32 • Embedded C++ • Node.js • Express.js • MongoDB Atlas • React

Current Scope

Prototype implementation based on an HTTP request-response architecture. Industrial communication protocols (MQTT, Modbus, OPC UA) are outside the current implementation.

Repository: Bearing-Fault-Diagnosis-System

Purpose

Research exploration of deep learning techniques for vibration-based bearing fault diagnosis.

Key Capabilities

• Sliding-window signal segmentation
• STFT spectrogram generation
• CNN-based fault classification using PyTorch
• Cross-domain evaluation to study model generalization

Technology

PyTorch • TorchVision • NumPy • SciPy

Current Scope

Models trained on the CWRU bearing dataset and evaluated on the IMS bearing dataset to investigate the impact of operational domain shift on classification performance.

My repositories follow a common engineering workflow:

Physical System → Sensors & Data Acquisition → Signal Processing → Feature Extraction → Data Transmission → Machine Learning & Diagnostics → Visualization

Each repository represents one stage of this physical-to-digital engineering pipeline.

• Mechanical Engineering Undergraduate (2nd Year)
• Projects are experimental engineering and research prototypes
• Focused on understanding engineering principles before production-scale implementation
• Repository technologies map directly to implemented source code
• No enterprise-scale, industrial deployment, or production claims

• Machinery Health Monitoring
• Condition Monitoring
• Signal Processing
• Embedded Systems
• Industrial IoT
• Machine Learning for Engineering
• Cyber-Physical Systems