Integrated Optimization and Design of a Variable-Speed Pumped Storage Electric Machine and Converter System

In response to the national strategic goals of “carbon peaking and carbon neutrality” and to enhance the safety and stability of the new power system, China is vigorously promoting the construction of pumped storage power stations. Variable speed pumped storage technology, with its advantages of superior power characteristics, good adaptability to water head, and flexible working condition transitions, has become the most advanced pumped storage technology in the world today. It is also a critical “bottleneck” issue that China must focus on overcoming.

In the development and operation of such new units, there are still problems such as poor coordination between main and auxiliary machinery, unclear operating mechanisms of the units, and a lack of specialized performance evaluation tools. These issues not only hinder the research and development of key equipment for domestically-produced units but also limit the characteristic analysis and status assessment after the units are put into operation. Against this background, this project plans to conduct research on the design optimization and mechanistic twin technology of variable speed pumped storage units. This includes proposing a water-machine-electric coupling multi-physics field simulation method for variable speed pumped storage units, overcoming the technical challenges of integrated design for AC excitation generator-motor and inverter systems, establishing a mechanistic twin modeling theory and platform for variable speed pumped storage units, and constructing multi-physics field digital twin sub-models for key equipment of the units. These outcomes will provide strong support for the design, manufacturing, and operation and maintenance diagnostics of large variable speed pumped storage units, enabling real-time and accurate perception of the internal operating states of the units. It will also lead to the establishment of a device-level twin simulation platform domestically, aiding the localization of research and operation of variable speed pumped storage units.

The sub-project topics include:

Researching accelerated transient process simulation methods for motor-inverter systems based on field-circuit coupling analysis principles. This involves establishing reduced-order equivalent circuit models of the motor on the stator side under different frequency AC excitation sources. The aim is to maximize the calculation speed of the electromagnetic circuit characteristics of the motor-inverter system during transient processes while ensuring the accuracy of motor parameter calculations, thereby reducing the simulation time of the motor-inverter system’s transient processes.

Studying rapid sensitivity analysis methods for motor structural parameters using machine learning algorithms. This involves establishing a rapid optimization method for motor structural parameters based on surrogate models, incorporating ideas from artificial intelligence, physical modeling, and convex and non-convex optimization. A general motor-inverter system structural parameter optimization software based on Ansys software will be developed, capable of considering both steady-state and transient process simulations to accurately and quickly optimize motor structural parameters. Based on this optimization software, performance simulation verification and prototype reduction test methods will be proposed for the structural parameter optimization design of AC excitation generator-motor and inverter systems used in large-capacity variable speed pumped storage units. Software functionality and performance verification will be conducted for the structural optimization design of the AC excitation generator-motor and inverter system used in large-capacity variable speed pumped storage units.