Abstract: In recent years, the construction of big data cloud platform has caused the urgent concern of the photovoltaic power generation industry. This paper first sorted out the idea of ​​building a private cloud platform for operation and management of big data in photovoltaic power generation group; then proposed the construction plan of private cloud platform for operation and management of big data, including the overall architecture, big data architecture, and centralized monitoring architecture; and finally proposed the composition of the platform application system. Program.
0 Preface
In the context of the current environment and energy crisis, China has introduced a series of policies to promote photovoltaic power generation. As of the end of September 2015, the installed capacity of photovoltaic power generation in China reached 37.95 million kW, of which, 31.7 million kW of photovoltaic power stations and 6.25 million kW of distributed photovoltaics [1]. According to the latest GlobalData report, China is expected to be the world's largest installed base of photovoltaics in 2015. China has already emerged a number of leading companies in the photovoltaic power generation industry, such as the 5 major power generation groups and GCL, Chint, Beijing and Qingfa.
With the rapid growth of the scale of photovoltaic power plants put into operation by photovoltaic power generation groups, facing the huge data environment and the complex data types of photovoltaic power plants in the power station equipment, conventional technologies can no longer meet the collection, storage, processing, and statistical analysis of massive data. . Photovoltaic Power Generation Group urgently needs to simultaneously plan and launch the Big Data Cloud Platform [2] project. It needs to sort out the ideas for constructing a big data private cloud platform to ensure that the construction ideas, technical routes, and business plans adopted by informatization are in line with the next 5 years or even more. Long-term development trend.
1. Construction ideas
The construction of the private cloud platform for big-data operation and management of photovoltaic power generation requires the integration of existing technologies accumulated in cloud computing, big data, and the Internet of Things [3]. Based on business needs, a set of coverage will be built using enterprise application integration EAI technology. Group company's cloud platform for operation and management of photovoltaic power plants. The cloud management platform is based on professional, leading, mature, and stable load balancing technology, virtualized network technology, and cloud service architecture. The software and hardware resources are flexibly expanded to meet the requirements for the rapid development of the photovoltaic power business.
The overall design of the cloud platform is based on the principle of using open source technologies, hierarchization, modularization, componentization, and configurability. It effectively integrates production control and other business application technology systems, and fully meets the new requirements for IT development in the next five years in technology selection. There is good support for new hardware and software technologies.
The cloud platform is encapsulated by basic software services to form a unified underlying support platform. Through modularization and componentized application configuration and a small amount of secondary development, all the functional requirements are fulfilled. The flexible deployment can be implemented according to actual conditions and can be implemented step by step. Through the dynamic configuration of parameters, it can adapt to changes in the organizational structure, enable flexible adjustment of the organizational structure, support group-level multi-organization and multi-level management and modelling, and support the process, data, and business functions among multiple organizations. Interoperability.
The cloud platform is guided by resource integration, relies on big data collection, storage, processing, and analysis as the basis of technology, and integrates the comprehensive advantages of real-time database, thematic analysis library (OLAP), and business intelligence (BI), and establishes the full-business support decision for the group company. Supported application systems, with the platform SOA/ESB data bus as the core, providing data interface services and application integration services for other related applications.
2. Operational Management Big Data Private Cloud Platform Construction Plan
2.1, overall architecture
The photovoltaic power plant operation management cloud platform is an IT basic resource comprehensive service platform composed of elastic computing, storage, network, virtualization, load balancing, and hot backup redundancy. The cloud platform provides on-demand computing, storage, and networking capabilities that can be used to quickly build a variety of applications, reduce the company's overall IT costs and management difficulties, so that the company can more specifically be their own business development and innovation.
Figure 1 Overall Operation of the Big Data Cloud Platform
The cloud platform adopts a modular design. It will use the cloud operating system and blade servers to build cloud computing resource pools to achieve unified centralized management of hardware and software, and to decentralize services, so that resources can be used more efficiently, system processing delays can be shorter, and hardware expansion can be achieved. More simple and quick.
In terms of architecture design and construction, the cloud management platform will follow the industry-promulgated standards and regulations, adopt advanced and mature technologies, and fully consider openness and scalability in terms of system architecture, basic platform, and communication protocols, as well as security and availability. Management and virtualization technology applications [4].
2.2, Big Data Architecture
The central database uses real-time database, historical database (Oracle), business database (Oracle), open source Key-Value database (Redis), theme database (Oracle), and big data distributed storage (Hadoop) hybrid application [5]. It supports the real-time monitoring of the plant's operating status and meets all types of application-oriented and subject-oriented analysis requirements.
The database design organizes database management in an object-oriented manner that conforms to the natural mode of human thinking. It implements a monitoring model that is based on equipment, facilitates equipment maintenance and fault diagnosis, and improves the speed and efficiency of data retrieval and search.
Figure 2 Group Operations Management Big Data Architecture
2.3, centralized monitoring architecture
The data acquisition subsystem of the centralized control system is the support system for the data acquisition of photovoltaic power plant equipment. It uses C++ as the development language, and includes a variety of IEC60870-5 101, 102, 103, 104, Modbus, CDT, DISA, etc. Data communication protocol; Modeling conforms to IEC 61970 interface reference model, Common Information Model (CIM) and Component Interface Specification (CIS) requirements, conforms to international standards, and can be seamlessly integrated with SCADA systems as middleware; boost station, inverter Equipment, combiner box, box change, power prediction, AGC\AVC system, electricity metering, protection fault information and other system data access. The system supports the access of a variety of photovoltaic power plant equipment to meet the access requirements of different brands of equipment, with a variety of procedural resolution capabilities.
The centralized monitoring system adopts a three-level architecture, namely a centralized monitoring system at the station level, a centralized monitoring system at the regional level, and a centralized monitoring system at the headquarters. The station-level centralized monitoring system is used to collect real-time monitoring data from the PV power station integrated monitoring system and other related terminal equipments to achieve local data monitoring, historical data sampling and storage and key real-time data upload to the headquarters centralized monitoring system. The regional centralized monitoring system/headquarters centralized monitoring system obtains the key equipment monitoring data from the station centralized monitoring system for real-time operation of the power station, and is used to monitor the overall operation status of the power plants. The communication protocol between Level 3 systems adopts the IEC104 statute of the electric power standard. The real-time data collection frequency supports the second level according to the statutory requirements. It supports three modes of sending on changes, sending on loops, and summoning.
Figure 3 centralized monitoring architecture
3, platform application system composition
Photovoltaic Power Generation Operation Management Big Data Private Cloud Platform Application System includes headquarters/regional level operation management system, station level intelligent operation and maintenance management system, headquarters level/regional level remote centralized monitoring and management system, and station level/regional level/headquarters level online monitoring Intelligent diagnosis system.
Figure 4 platform application system composition
Figure 5 platform monitoring large-screen renderings
4 Conclusion
Using cloud computing platforms to achieve a management approach (including virtualization, on-demand distribution, dynamic management, flexible configuration), a technical system (Hadoop, distributed computing and storage), consistent R & D technology route (J2EE/C++) A service approach (including cloud platform basic services and module applications) can adapt to the rapid development of the company's photovoltaic power plant business and the rapid development of ICT technology.
Remarks:
Qi Shuqiang 1, Li Yonglong 1, Zhou Shuangquan 2, Wang Haifeng 2, Liu Baolin 2, Wei Hong 2
(1. Qinghai Green Power Group Co., Ltd.; 2. Beijing Jinhongtai Technology Co., Ltd.)
references
[1] Bloomberg, Xiaoyang. 2015 The most comprehensive data analysis of the photovoltaic industry & 2016 Outlook [EB/OL]. http://solar.ofweek.com/2016-01/ART-260009-8440-29057595.html, 2016- 1.
[2] Wang Dewen, Song Yaqi, Zhu Yongli. Smart grid information platform based on cloud computing [J]. Automation of Electric Power System, 201-34 (22): 7-12.
[3]Zeng Lingkang. Application of Internet of things and cloud computing in smart grid information communication scheduling [A].2012 Proceedings of the Power Communication Management and Smart Grid Communication Technology Forum [C], Beijing, 2012.
[4] Zhou Xiangfeng. Research on virtualized cloud computing security in smart grid[J]. Computer Security, 2013, (5): 71-73, 76.
[5] BAI Hongwei, MA Zhiwei, SONG Yaqi. Processing of smart grid status monitoring data based on cloud computing [J]. East China Electric Power, 201-39 (9): 1485-1487.
Ningbo Alite Lighting Technology Co.,Ltd , https://www.alite-tmwt.com