A liquid-impulse neural network model based on heterogeneous fusion of multimodal information for interpretable rotating machinery fault diagnosis

Key Points

• Interpretable models enhance accuracy in diagnosing faults, fostering trust in machinery operations.
• The liquid-impulse neural network integrates various types of data, improving diagnostic outcomes significantly.
• Heterogeneous fusion of information from different sources is crucial for effective fault diagnosis in rotating machinery.
• This highlights the need for advanced models capable of interpreting complex data to predict malfunctions.