Research On Massive Data Processing Of Intermittent Energy And Energy Management Technology

With the increasing depletion of coal, oil and other non-renewable resources and the increasingly serious global environmental and climatic change, priority and vigorous to the development of renewable energy in the energy revolution is underway. Wind, solar and other renewable energy have gradually become an important part of the world’s energy, however, the intermittent and volatility feature is an important cause to restrict the development of renewable energy. It is helpful for the grid schedule department to make the daily operation mode and adjust scheduling plan according to the precise control of intermittent energy output through the rational use of energy management technology and power prediction, and this would weaken the adverse effects of intermittent energy on the grid to ensure reliable and safe operation of the power system. However, with the in-depth study of energy management technology, the achievement of control strategies leads to increasing demand on massive data processing. It is both difficult to achieve this goal and needs to pay a high cost of hardware and software by using the traditional single/multi-threaded architecture platform. Currently cloud platform architecture for handling massive data processing has been successful in the field of Internet, and such architecture has the advantages of massive data processing capability, high resource utilization and low application cost. We can properly solve these problems if we can apply it to the energy management systems of intermittent energy.Regional energy management technology of intermittent energy, represented by wind power and photovoltaic power has been studied in this paper based on the massive data processing technology. In wind power, for example, combining the achievement of our team in the aspects on wind power monitoring and wind power prediction, this paper introduces the active control strategy based on wind speed information dynamic classification for WTGS in wind farm and the output power loss calculation model caused by the failures or regulation. The massive data processing and computation requirements to implement the strategy and computational model are analyzed. This paper introduces the solution combination of memory database processing technology, traditional relational database processing technology and Hadoop cloud platform according to the design features of intermittent energy management platform. The main work and achievements are as follows:(1) This paper analyzed the problems of intermittent energy in the process of energy management, proposed the concept of regional control, and studied the characteristics of massive data generated by intermittent energy in the regional energy management process. A set of complex data storage and processing solutions for intermittent energy management system was proposed, combined with traditional data storage and processing technology, the distributed data storage and massive data processing technology.(2)The issues of intermittent energy in the process of active power regulation were studied to solve these problems. With wind power, for example, the active control strategy based on wind speed information dynamic classification for WTGS in wind farm was proposed and the idea for applying the thoughts on photovoltaic power was put forwad. The results show that the active control strategy based on wind speed information dynamic classification for WTGS in wind farm could achieve more precise control objectives under the dispatching, at the same time, the frequency of WTGS starting and stopping was reduced and the service life of WTGS was effectively extended.(3) The output power loss situation of wind power caused by generator failures or regulation was studied in this paper and the calculation model and method of the output power loss of wind farms and wind turbins were proposed. The MapReduce-based implementation scheme about the calculation model and method of the output power loss of wind farms and wind turbins was proposed. An expert system database with large amounts of historical data was establised to present the relationships among wind speed, wind direction and power output. The power loss model was built based on the expert database. The wind power output losses caused by wind turbine failures or abandoned wind can be exactly calculated by this model, and the calculation of the effectiveness of the MapReduce-based implementation scheme about the calculation model and method of the output power loss of wind farms and wind turbins was analyzed. This paper provides research ideas for cloud computing and massive data processing technology used in the calculation of massive data proce of intermittent energy. The results show that this solution has excellent processing performance in the case of massive data processing close to200GB.(4) The private cloud platform design of large-scale intermittent energy management system to support regional regulation was proposed based on the idea of private cloud platform in the massive data technology and the research results our team achieved in the aspect of intermittent monitor, operation management and power forecasting. The secondary energy management system development based on this solution was done. The environment consisting of 10 wind turbines in a wind farm in Inner Mongolia was test, the results show that this was a safe and reliable system,which was easy to operate, and the operational efficiency was greatly improved compared with the original system and achieve the desired target of the platform.