A rapid ultra-micro internal leakage detection method based on digital twins for hydraulic cylinders of hydro turbine speed control system

• A method to build a DT model of a ultra-micro internal leakage test system for hydraulic cylinders has been developed. • By combining RSM, MOAHA and TOPSIS, optimal estimation of unknown parameters in hydraulic systems is achieved. • A novel 2DSW algorithm is introduced to reconstruct simulation data for generating highly reliable DT data. • DT data ensures a strong linear relationship between pressure decay rate and ultra-micro internal leakage rate. • The stable pressure decay threshold improves the detection accuracy of ultra-micro internal leakage of hydraulic cylinders. Rapid and precise detection of ultra-micro internal leakage in hydraulic cylinders within hydro turbine speed control systems (HTSCS) is critical for ensuring the stable operation of power generation unit. The pressure decay test (PDT) method provides an efficient inspection approach for acceptance test and periodic offline maintenance of hydraulic cylinders. However, the poor stability of the pressure decay threshold limits the practicality and effectiveness of the PDT method. To address these issues, this paper proposes a research framework for rapid ultra-micro internal leakage detection of hydraulic cylinders of HTSCS based on digital twin (DT). First, a surrogate model between unknown parameters in the hydraulic test system and simulation-experiment errors was constructed. Combining the multi-objective artificial hummingbird algorithm (MOAHA) with the technique for order preference by similarity to ideal solution (TOPSIS) enabled the estimation of optimal parameter combinations. Second, a novel two-dimensional signal warping (2DSW) optimization algorithm is introduced to reconstruct simulation data and generate highly reliable DT data. Finally, a multi-stage data processing scheme based on simulation, experimental, and DT data is designed, systematically validating the proposed method’s effectiveness. Experimental results demonstrate that DT data significantly enhance the linear relationship between the pressure decay rate and the ultra-micro internal leakage rate. Under high-pressure conditions, the standard deviation and coefficient of variation of the pressure decay threshold indicator in the PDT method decreased by 67.46% and 77.29%, respectively, with significantly improved stability. This research framework establishes a systematic solution integrating a mechanism, data, and decision-making for the stable operation of HTSCS.