Efficient Training of Large-scale Industrial Fault Diagnostic Models through Federated Opportunistic Block Dropout
This work addresses communication inefficiencies in federated learning for industrial fault diagnostics, enabling more practical deployment across companies, though it is incremental as it builds on existing FL and dropout methods.
The paper tackles the challenge of efficiently training large-scale industrial fault diagnostic models across multiple institutions with sensitive distributed data by developing Federated Opportunistic Block Dropout (FEDOBD), which reduces communication overhead by over 70% while maintaining model performance at over 85% test F1 score.
Artificial intelligence (AI)-empowered industrial fault diagnostics is important in ensuring the safe operation of industrial applications. Since complex industrial systems often involve multiple industrial plants (possibly belonging to different companies or subsidiaries) with sensitive data collected and stored in a distributed manner, collaborative fault diagnostic model training often needs to leverage federated learning (FL). As the scale of the industrial fault diagnostic models are often large and communication channels in such systems are often not exclusively used for FL model training, existing deployed FL model training frameworks cannot train such models efficiently across multiple institutions. In this paper, we report our experience developing and deploying the Federated Opportunistic Block Dropout (FEDOBD) approach for industrial fault diagnostic model training. By decomposing large-scale models into semantic blocks and enabling FL participants to opportunistically upload selected important blocks in a quantized manner, it significantly reduces the communication overhead while maintaining model performance. Since its deployment in ENN Group in February 2022, FEDOBD has served two coal chemical plants across two cities in China to build industrial fault prediction models. It helped the company reduce the training communication overhead by over 70% compared to its previous AI Engine, while maintaining model performance at over 85% test F1 score. To our knowledge, it is the first successfully deployed dropout-based FL approach.