Sangshu Technology has signed a cooperation agreement with the State Grid to deeply customize and transform the optimization of large power systems based on the mathematical optimization solver COPT.

To improve asset utilization efficiency, effectively reduce users' electricity costs, lower electricity prices, and accelerate the orderly competition among power production enterprises, China is vigorously promoting the construction of the electricity market. Large power companies represented by the State Grid Corporation bear the fundamental mission of ensuring safe, economical, clean, and sustainable power supply, which is related to the lifeline of the national economy and national energy security.

Shanshu Technology has signed a research and development cooperation agreement with the Global Energy Internet Development Cooperation Organization of the State Grid Corporation. The research and development focuses on a specialized solver for the joint optimization of hydro-thermal power generation units considering network security constraints, with the core goal of delivering an efficient parallel solver suitable for secure constrained unit combinations, and accelerating algorithms for the model based on the solver.

Security Constrained Unit Commitment is an important component of the technical support system for electricity market operations. For large power systems, security constrained unit commitment considers constraints such as power balance, network security, unit capacity, operational reserve limits, and ramp-up and ramp-down rates. This problem is mathematically a large-scale mixed-integer programming (MIP) problem, which is complex and computationally intensive. To meet the requirements of electricity market trading time, it is crucial to improve the computational efficiency of security constrained unit commitment using advanced operations research optimization techniques, parallel technology, and AI technology.

Shanshu Technology leverages its core advantages in operations research optimization modeling and underlying solver acceleration, based on its self-developed mathematical optimization solver COPT, which is also the first industrial-grade solver in China, to deeply customize and transform the optimization model solving algorithms for power systems. This assists the State Grid in reconstructing the hydro-thermal power generation unit optimization model and introduces artificial intelligence branch-and-bound methods, combined with graph parallel computing technology to enhance the overall solving efficiency and stability of power system optimization, thereby achieving the goal of an efficient, robust, autonomous, and controllable power system optimization solver.