Converter Capacitor Temperature Estimation Based on Continued Training LSTM under Variable Load Conditions.

Area of Science:

• Power Electronics
• Artificial Intelligence
• Condition Monitoring

Background:

• Capacitors are vital in power electronic converters but prone to damage from operational stress.
• Accurate capacitor temperature monitoring is critical for ensuring power converter reliability and preventing failures.

Purpose of the Study:

• To develop a novel method for precise estimation of capacitor core temperature.
• To enhance the adaptability of temperature estimation algorithms under variable load conditions in converters.

Main Methods:

• Utilizing a long short-term memory (LSTM) neural network algorithm for temperature prediction.
• Implementing a continued training mechanism to adapt the LSTM model to dynamic load variations.

Main Results:

• The proposed LSTM-based method achieved high accuracy in estimating capacitor core temperatures.
• The continued training mechanism demonstrated robustness in adapting to variable load conditions.

Conclusions:

• The novel LSTM approach offers a reliable solution for real-time capacitor temperature monitoring in power electronic converters.
• This method contributes to improved power converter diagnostics and operational safety.