Qingping Li

Title: Qingping Li and Data-Driven Equipment Benchmark Prediction.

Introduction

Qingping Li is an inventor based in Shanghai, China. His work includes a patent for applying machine learning to predict benchmark values for unit equipment, supporting more automated and data-informed equipment operation.

Latest Patents

Li is associated with 1 patent: “Method for Predicting Benchmark Value of Unit Equipment Based on XGBoost Algorithm and System Thereof.” The invention uses historical equipment-operation data to build a dataset, applies random forest out-of-bag estimation to assess feature importance, removes less important features, and standardizes the remaining data. It then uses Bayesian hyperparameter optimization to develop an XGBoost prediction model. Real-time operating data can be supplied to the model to predict benchmark values for equipment parameters. The patent describes the approach as offering strong generalization, prediction accuracy, and operating speed, while improving automation.

Career Highlights

Li works at Huaneng Shanghai Combined Cycle Power Co., Ltd. His patent reflects an application of machine learning and operational data analysis to equipment benchmarking in the power sector.

Collaborations

Li’s listed coworkers include Yongkang Wang and Gang Xu. The available information identifies them as coworkers but does not specify their roles in the patent.

Conclusion

Qingping Li’s patent applies XGBoost and feature-selection techniques to predict equipment benchmark values, connecting data-driven innovation with improved automation in power-plant operations.