The integration of the "energy revolution" and the "digital revolution" takes a new step
-observations of the artificial intelligence and power big data forum
In Xiaohe Street, Gongshu District, Hangzhou City, Zhejiang Province, 326 elderly people living alone will be the first in the country if an accident occurs The online power cloud computing service can calculate the abnormal power consumption, input the abnormality into the "City Eye·Cloud Public Administration" platform, and the community grid staff will immediately come to the door to check and send care to the elderly.
At the power engineering site in Guangzhou, Guangdong Province, if the staff does not wear a safety helmet, they will be immediately captured by a camera with edge computing; if there are problems such as crossing the safety boundary and incorrect use of safety tools and equipment, the smart safety monitoring system The background artificial intelligence recognition will also alert, so that there is no opportunity for security risks.
Because the fan generally has the problem of inaccurate wind, the maintenance is expensive because the fan is relatively high, and the efficiency of positioning the problem is very low. The application of SPIC's fan "yaw to wind optimization" can detect the inaccuracy of the fan to the wind in time, and automatically adjust it in the case of a small angle.
Recently, at the 2020 Artificial Intelligence and Electric Power Big Data Forum jointly hosted by China Electric Power Development Promotion Association and State Grid Big Data Center, the reporter heard many cases of artificial intelligence and big data in electric power applications. From the initial small test to the full-scale advancement today, the power companies’ artificial intelligence and big data have achieved gratifying gains in the fast-track progress.
Big data applications are placed high hopes
1155 centralized monitoring substations, with more than 260,000 monitoring signals per capita; if the 10 kV line is out of power, it will affect 1,000 to 2,000 residents, and if the 35 kV line is out of power, it will affect 5,000 to 10,000 residents— —As a mega-city grid, Shanghai Power Grid has the characteristics of large data volume, high load density, and wide impact of power outages. "Under the traditional centralized monitoring mode, there are difficulties such as limited manual monitoring efficiency, insufficient operation trend analysis methods, and low accuracy of fault research. It is urgent to carry out centralized monitoring technology and application research based on big data to improve centralized monitoring efficiency and grid security. "State Grid Shanghai Electric Power Company Li Xiongli said at the forum.
Shanghai Power Grid's ongoing "Power Grid Intelligent Monitoring and Decision-Making Diagnosis Based on Equipment Panoramic Data" introduces big data intelligent analysis methods, breaks the bottleneck problem of substation data transmission, enhances the power grid panoramic data aggregation ability, and improves the power grid monitoring data cleaning and identification ability Exploring key factors affecting failures, early warning of equipment defects and failure trends; improving data intelligent correction capabilities, enhancing power grid panoramic data aggregation capabilities, optimizing fault intelligent diagnosis methods, ensuring the safety of equipment and power grid operation, and further improving the level of grid regulation and operation.
Li Xiongli gave an example that the project is based on fault fingerprint real-time fault intelligent diagnosis, through clustering algorithm to obtain the optimal fingerprint library, real-time diagnosis algorithm finds and matches the fault "second level" diagnosis output, and the average fault handling time is shortened to 50%. %. The intelligent correction of remote signals based on discrete neural network effectively solves the problem of interference of false remote signals on fault judgment, and improves the accuracy of fault diagnosis to 99.15%. The big data system selected 310 historical data from 2014 for training to form a database of early warning experts. During 2019, Typhoon Lichma predicted 198 failures (243 actual failures), and the prediction accuracy rate reached 81.48%.
"Now data has penetrated into all aspects of our society, becoming a new driving force for business decision-making, new content of goods and services, and a new means of governance for all members of society. At present, there are still many differences in the level of digital economy development in various industries, while the power industry It is probably in the outbreak period, and the next step will be to lead the way." Shen Liang, deputy director of the State Grid Big Data Center, said at the forum.
"The core of the future energy system is solar energy and wind power. Renewable energy is indirect. In addition, more application scenarios of energy such as photovoltaics may be distributed in the future. So how can it be integrated into the current energy system? Effective use? Digital technology must be adopted." Fan Jihong, Director of Science and Technology Innovation of the State Power Investment Group, Secretary of the Party Committee of the Central Research Institute, and Chairman of the Board of Directors, also has high hopes for power big data.
State Power Investment Corporation established the Group Big Data Center in 2017. At the end of 2018, the China Energy Convergence Smart Technology Company was established to promote the construction of the energy industry Internet, and it also led the China Smart Energy Industry Alliance. Fan Jihong said"The company’s big data application is bringing direct effects, which is very obvious in our group. Through the current application of big data, it will directly benefit the entire production and operation and give us some pricing and transaction judgments in the electricity market. , And even the management of the entire company brings great benefits.”
“A development engine of the digital economy, we think it is industrial integration.” Shen Liang further stated that in March this year, the central government put forward the concept of new infrastructure, “new The specific embodiment of the Industrial Internet in the energy industry in "Infrastructure" is the Energy Internet. Energy Internet is a deep integration of digital technology, big data applications and energy and power technology. Electric power big data is also an act that conforms to the development of the digital economy and serves the country's development strategy. It is also an important part of building a growth point for electric power companies and establishing an energy Internet.
The construction of big data has been accelerated
"Electric power big data has four characteristics: huge scale, various types, high authenticity, and good industrial foundation. With the construction of the energy Internet, the coverage, perception depth and data quality of electric power big data will continue to improve , The application foundation is becoming more and more perfect. If we integrate these data with external information such as meteorology, economy, finance, etc., we can bring out the value of 1+1>2." Shen Liang, deputy director of the State Grid Big Data Center, mentioned power Said the prospect of big data.
He introduced that after the establishment of the State Grid Big Data Center, it paid special attention to the mining of the value of electric power big data, and built a "3D + 4 layer" electric power big data value system, covering the "ideal layer-brand layer-system layer-product layer" "3 dimensions, 4 levels, and 6 "Shu e" brands (Shu Jing e, Shu Rong e, Shu Ke e, Shu Neng e, Shu Ce e, and Shu E e). As the big data application service platform of Writer Grid Corporation, Digital Intelligence State Network has been launched in early April this year, providing services to authoritative users, covering functions such as data products, data resources, and capability opening. Up to now, 1 headquarters channel and 8 provincial channels have been built, 203 typical data products such as power economic index, non-home inspection, and 79 model algorithms have been put on the shelves; 71 sets of business system data and external data resources have been put on the shelves 9 types of data, including meteorological and economic data, integrated data analysis tools, enterprise report center, etc. The
reporter learned that big data has been applied in many aspects in the State Grid. This year, the State Grid Big Data Center took the lead in developing production-resumption related indexes, using index analysis methods to clean, model, and calculate billions of data, and calculate the degree of power recovery, the production rate, the over-production rate, and the failure rate on a monthly basis. Prosperity index and other analysis reports are compiled and submitted to relevant state ministries and commissions to continuously evaluate the recovery of small and micro enterprises. In response to the situation in this year’s national census that the enumerators did not know which households were occupied, the State Grid Corporation, in accordance with the commission of the relevant departments, combined with the residents’ files and the household electricity meter data in the last 30 days, developed a recognition algorithm for the residential situation of residents in different provinces. Created a "home code" to determine whether someone lives regularly in the home. In order to support the implementation of the company's digital transformation strategy and accelerate the construction of a world-class enterprise with global competitiveness, China Southern Power Grid has upgraded the company's 2019 version of its digital transformation action plan twice. The "China Southern Power Grid Digital Transformation and Digital Grid Construction Action Plan (2020)" develops "Digital Southern Power Grid" as a "digital grid" construction, and clearly proposes "cloud and digital integration" as the core, with cloudification and microservices. Based on the technical route, action items such as building a company's service sharing center, building a public cloud, and establishing a digital platform operation system have been added. According to Chen Jun, Chief Information Officer of Guangdong Power Grid Corporation, China Southern Power Grid has provided cloud-based storage, computing, network and other basic resources and resources for China Southern Power Grid’s data center, unified services and business applications since it launched the Nannet cloud platform in July 2019. The operating environment solves the problem of flexible resource allocation and rapid deployment. At this stage, the resources of all 9 CSG cloud sub-nodes and 3 stock clouds have been completed, as well as the unified management of 1162 servers across the country, providing the company with large-scale computing of 100,000 core vCPUs (virtual processors in computers) Capacity and 263 terabytes of memory space, data storage capacity reaches 1731 terabytes.
"Based on the artificial intelligence platform of China Southern Net Cloud, 97 artificial intelligence components are currently online on the entire network, and the components have been called 29.26 million times. They are widely used in typical application scenarios such as power transmission and transformation and enterprise management." He said.
According to reports, in 2020, China Southern Power Grid will upgrade its data center in a “integrated cloud and data” approach to create coverage data collection,With the ability to manage the entire life cycle of data in storage, processing, and application, the next step is to build a unified basic data base for all platforms and applications of China Southern Network. All data are collected on the data base to achieve data synchronization and consistency. In 2020, China Southern Power Grid will also initially complete the construction of a unified power grid data model for the power grid digital platform, creating a twin mirror image of the entire process of the power grid, and accumulating shared service components for grid business.
This year, State Power Investment Corporation established the group's industry innovation centers to professionally support digital transformation and promote the deep integration of water, fire, wind, light, nuclear and other business formats and digital technologies. According to Fan Jihong, most of these work are carried out around the application of big data. For the future big data work, he put forward his own ideas: "Our group has built many smart energy parks in China, but the consumer side connection does not reach every electricity facility, such as every air conditioner and air purification device in the house. No connection. In the future, the true digital energy system should penetrate into every most terminal energy use facility. In this way, we can realize a smart society and wisdom based on the smart energy system. "Z3z
" At present, the power industry is actively implementing the new energy security strategy of'Four Revolutions, One Cooperation', and promoting the integrated development of the energy revolution and the digital revolution. Since the establishment of the 4th Council of the Council, it will serve the power industry. Planning to develop and promote the integration and application of a new generation of digital technology in the power industry as the core business direction, and actively advocate the service concept of "cross-border, integration, value-added, and sharing"." You Min, Executive Vice President and Secretary-General of China Electric Power Development Promotion Association, at the forum The above pointed out that in the future, it will cooperate wholeheartedly with the State Grid Big Data Center to jointly create an ecological construction and service platform for the "Artificial Intelligence and Big Data Committee". As the third platform, the association will have a lot to do.
Author: Zhao Ran
Editor: Wang Ping
proof: in glowing
