The Platform Architecture Research For Power Grid Simulation Based On Management Of Equipment Parameters And Operating Data

Based on summarizing the current situation of data management architecture design in various simulation data management platforms of power system at home and abroad,this paper makes a detailed analysis of the problems existing in the design and application of data management architecture,clarifies the requirements of large interconnected power grid for simulation data management,proposes a solution that can meet the needs of China's power grid simulation computing data management from two aspects:key issues of data architecture and software architecture technology,and develops a data management platform for power grid simulation calculation.The research and development results of this paper have been applied to the State Grid Simulation and Computing Data Center,which has effectively solved the problems of long data preparation period and low work efficiency of ultra-large-scale power grid operation mode,unified the data sources of simulation calculation in different areas,realized source-end maintenance,centralized management and rolling update of power grid simulation data,improved the maintenance efficiency and quality of power grid simulation model data,and provided strong support for the safe and stable operation of UHV AC/DC hybrid power grid.The main research contents are as follows:(1)The layered design idea of power grid equipment model and mode model is proposed.The power grid equipment data model object and operation mode data model object are constructed by non-destructive decomposition method and based on the principle of eliminating low-level redundancy.The data structure of layered design of power grid equipment model and mode model can not only ensure the uniqueness of equipment model parameters,but also realize the flexibility of operation mode data.It is the most core and basic data structure design of this paper,and can be well applied to power grid simulation calculation data management.(2)The idea of life cycle management for power grid equipment is proposed.The rolling maintenance of model parameters is realized by the transition from planning power grid to current power grid.The method of extracting power grid model based on time dimension is studied to realize rolling maintenance of power grid equipment in large time span.For a long time,the requirement of rolling maintenance has not been well solved in the management of power network simulation computing data.This design provides a feasible and effective solution.(3)The idea of unifying the electrical structure division space and maintenance authority attribution space of all equipments is proposed.This paper studies the method of privilege management based on spatial dimension and collaborative work mode,and studies the method of modular mode splicing based on spatial dimension,which lays a key foundation for parallel equipment data maintenance,operation mode debugging and flexible realization of operation mode splicing across units.(4)In order to guarantee the quality of data maintenance source and trace the data evolution process,the parameter version and process audit mechanism are designed.The parameter version is an important supplement to the" Uniqueness of Device Model Parameters",that is,"Uniqueness of the Effective Device Model Parameters",while the parameters of the equipment model in the non-effective state(draft state or historical state)have diversity,which can meet the actual needs of the project.(5)Based on the concept of "Unified Platform,Source-side Maintenance,Parallel Collaboration,Avenue-to-Simplicity,and Simplicity-to-Computing",the overall architecture of the simulation computing data management platform is designed.The SSMAES integration framework is used to develop the platform.The development platform realizes a new work mode of data maintenance and data sharing in different places,improves work efficiency,and enhance the level of power grid simulation calculation and analysis.