스마트그리드를 위한 Hybrid DataBase 기반의 MDMS 설계 및 구현

Abstract

세계적으로 에너지 자원의 고갈 문제가 발생 하면서 대안으로 지능형전력망(스마트그리드, Smart Grid)이 미래의 그린에너지 산업으로 부각되었고 전력에너지 분야의 중요한 이슈로 등장하게 되었다. 지능형전력망은 기존의 전력망에 정보기술(Information Technology)을 접목하여 전력공급자와 전력 소비자가 서로 실시간으로 정보를 교환해 에너지 효율을 극대화하는 차세대 전력망이며 이를 구현하기 위하여 Energy Storage Devices(ESD), Advanced Superconducting Transmission Cables, Smart Substation and Smart Transformers, Advanced Metering Infrastructure (AMI), Home Area Networks(HAN)의 5개 주요 기술 분야가 논의 되고 있다. 최근에는 지능형전력망을 위한 인프라 연구로 AMI(Advanced Metering Infrastructure)가 제시되었고 AMI는 홈네트워크, 스마트 미터, 지능형전력 서비스 네트워크, MDMS(미터 데이터 관리 시스템, Meter Data Management System)등의 기술 분야로 구성된다. 본 논문에서는 여러 가지 물리적 인프라를 활용해 미터 데이터를 수집하고 이를 처리 및 가공, 분석 하여 가치정보로 변환하는 MDMS 서버를 설계 및 구현하였다. 또한 전력사 고객의 전력사용량 수집 주기가 15분 단위로 짧아져 수집되는 데이터 양이 방대해 지고 있는 실정에 DB성능 개선을 위해 Hybrid DataBase를 적용하였고 상대적으로 접근빈도가 낮은 1개월 기준 이후 데이터를 접근 빈도가 높은 Hot Data영역에서 Cold Data영역인 디스크로 이동 하여 관리 하였다. 그 결과 Hybrid DataBase기반 MDMS는 기존 MDMS 서버와 비교하여 대용량 데이터를 수용하면서 Hybrid MMDB-Disk Based 적용 시 약 1.2배, Hybrid MMDB-Memory Based 적용 시 약 4배, Hybrid MMDB-Memory + Disk Based 적용시 약 2배에 가깝게 MDMS의 DB 처리 능력이 향상되었다.| Due to the depletion of energy resource criteria over the world, as a solution, the Smart Grid has been magnified as the future green energy industry, and it became the important issue in the electrical energy field. The Smart Grid interconnects the conventional electrical grid with Information Technology to lead real-time bi-directional information exchange between power supplier and consumer to optimize the energy efficiency, and to implement this, there are five major technological fields to be discussed: Energy Storage Devices (ESD), Advanced Superconducting Transmission Cables, Smart Substation and Smart Transformers, Advanced Metering Infrastructure (AMI), and Home Area Networks (HAN). Recently, as the research on infrastructure for the Smart Grid, AMI(Advanced Metering Infrastructure) was proposed, and this AMI is composed of technological fields such as home network, smart meter, intelligent electrical service network, and MDMS (Meter Data Management System). In this paper, the MDMS server to collect, process, manipulate, and analyze the meter data by using various physical infrastructures and to convert them into the meaningful information was designed and implemented. Furthermore, as the electric use data collection cycle has been shortened by the electric power company to make the size of data to be enormous, the Hybrid Database to improve the DB performance was applied. The data more than one month old with comparatively low access were moved from the hot data region to cold data region and managed independently. As a result, the MDMS based on Hybrid Database maintained the large capacity compare to conventional MDMS server. And it showed increasement of performance about 1.2 times the conventional DB process under Hybrid MMDB-Disk-based application, about 4 times the conventional under Hybrid MMDB-Memory-based application, and about 2 times the conventional under Hybrid MMDB-Memory + Disk-based application.; Due to the depletion of energy resource criteria over the world, as a solution, the Smart Grid has been magnified as the future green energy industry, and it became the important issue in the electrical energy field. The Smart Grid interconnects the conventional electrical grid with Information Technology to lead real-time bi-directional information exchange between power supplier and consumer to optimize the energy efficiency, and to implement this, there are five major technological fields to be discussed: Energy Storage Devices (ESD), Advanced Superconducting Transmission Cables, Smart Substation and Smart Transformers, Advanced Metering Infrastructure (AMI), and Home Area Networks (HAN). Recently, as the research on infrastructure for the Smart Grid, AMI(Advanced Metering Infrastructure) was proposed, and this AMI is composed of technological fields such as home network, smart meter, intelligent electrical service network, and MDMS (Meter Data Management System). In this paper, the MDMS server to collect, process, manipulate, and analyze the meter data by using various physical infrastructures and to convert them into the meaningful information was designed and implemented. Furthermore, as the electric use data collection cycle has been shortened by the electric power company to make the size of data to be enormous, the Hybrid Database to improve the DB performance was applied. The data more than one month old with comparatively low access were moved from the hot data region to cold data region and managed independently. As a result, the MDMS based on Hybrid Database maintained the large capacity compare to conventional MDMS server. And it showed increasement of performance about 1.2 times the conventional DB process under Hybrid MMDB-Disk-based application, about 4 times the conventional under Hybrid MMDB-Memory-based application, and about 2 times the conventional under Hybrid MMDB-Memory + Disk-based application.