An active-learning, black-box optimizer with virtual strain sensors found a Francis-turbine startup sequence that cut the measured maximum strain cycle amplitude by 42% versus the standard sequence.
Swirl number based transposition of flow-induced me- 24 chanical stresses from reduced scale to full-size Francis turbine runners,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Active Learning-Based Optimization of Hydroelectric Turbine Startup to Minimize Fatigue Damage
An active-learning, black-box optimizer with virtual strain sensors found a Francis-turbine startup sequence that cut the measured maximum strain cycle amplitude by 42% versus the standard sequence.