Key Skills

Software

Appen

Data Annotation Tech

Google Cloud Vertex AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Image

Top Task Types

Action Recognition

Audio Recording

Classification

Computer Programming Coding

Data Collection

Freelancer Overview

my recent project developed and evaluated a cost-effective computer system for fault classification of induction motor (IM) rolling bearings using current signal data analysis. This system leverages machine learning models, comparing a Multi-Layer Perceptron (MLP) neural network with a Random Forest classifier. Key features were extracted from spectral and envelope analysis of motor current data to train the models. To address challenges such as noisy data, poor data quality, and overfitting, I implemented robust preprocessing techniques like filtering and feature selection to enhance data quality. These works were supported by Google Colab’s cloud-based resources and Python machine learning frameworks (TensorFlow, Scikit-learn), enabling efficient model training and evaluation. The result was a scalable, low-cost solution for motor fault diagnosis, demonstrating its potential for real-world industrial applications. Building on this work, I have further research ideas about optimizing induction motor parameters through pre-trained fault classification models, aiming to acquire datasets for training other motors without the need for experimental work.

Labeling Experience

Education

Work History