Modeling of high voltage induction motor cooling system using linear regression mathematical models.

Area of Science:

• Electrical Engineering
• Thermal Management
• Predictive Maintenance

Background:

• High voltage induction motors (HVIMs) face challenges in power efficiency due to control strategy failures.
• Effective cooling systems are critical for HVIMs, preventing overheating and plant shutdowns.
• Existing HVIM cooling system designs are often limited and developed in isolation.

Purpose of the Study:

• To develop a dynamic model of the HVIM cooling system for performance evaluation, failure diagnosis, and prediction.
• To address limitations in current HVIM external cooling system designs.
• To propose an integrated modeling approach for improved thermal management.

Main Methods:

• Developed an electrical model using established mathematical principles.
• Created a thermal model combining mathematical and linear regression techniques to predict motor temperature.
• Designed a modified cooler model to supply cold air temperature data for monitoring.
• Integrated electrical, thermal, and cooler models into a unified HVIM cooling system model.

Main Results:

• The integrated model accurately represents the HVIM cooling system's actual signals.
• Validation using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) showed high accuracy.
• The model achieved less than 1% error tolerance, meeting industrial expert expectations.

Conclusions:

• The proposed combined modeling approach significantly improves thermal performance representation for HVIMs.
• The validated model is suitable for quality maintenance, fault identification, and prediction.
• This integrated model enhances the reliability and efficiency of high voltage induction motors.