| 220kV network locates in the middle of the voltage series of Chinese power system.With the fast development of the installed generation capability and load,the development of 220 kV system displays a diverse pattern introduced by functional difference between sending and receiving sides as well as the uneven regional economy.In light of the diverse development pattern of 220 kV system,traditional network System planning exposes its own limitation by the corresponding application onto multi-scenario planning cases,and the feasibility is not ensured,while the planning principle cannot be summarized.220 kV system planning with multi-scenario constraints is a complex problem integrated with model building,algorithm selection and study case generation.There are still few researches related to this topic.Therefore,this paper focuses on the problem mentioned above.For model building,this paper uses super-quantile theory to transfer the LOLP constraints and integrates short current constraint into a bilayer model.For algorithm selection,a mixed integer convex optimization solving method is applied to solve the master problem,while a heuristic method is used to solve the slave problem.The N-1 principle is verified by a post-checking algorithm with Monte Carlo Method.For study case generation and analysis,modified IEEE standard systems are used to prove the performance of the algorithm in large system and to study the computational cost.An actual study case of System planning in year 2030 generated from eastern China power system is used to prove the feasibility of the algorithm.With a multi-scenario 4 bus system in consideration of sending-receiving and regional differences,the typical network System planning principle is summarized.The method proposed in this paper to deal with LOLP constraints could be further applied to model other system planning uncertainties.The multi-scenario diverse application principle could also be helpful for researchers to explore 220 kV system planning problems. |
PDF Full Text Request