Artificial Intelligence: The Inevitable Choice for the Next-Generation Grid and Energy Internet

China Energy Storage Network News: On June 20, at the China Electric Power Research Institute's Artificial Intelligence Application Academic Seminar, academicians and experts from research institutes, universities, and enterprises in the fields of artificial intelligence and electricity gathered together to discuss the forefront of artificial intelligence technology, exchange ideas, sharpen wisdom, and offer suggestions for the development of artificial intelligence technology for State Grid Corporation of China.

At present, artificial intelligence is developing rapidly and has the conditions for implementation in various fields. The CPC Central Committee and the State Council attach great importance to the development of artificial intelligence technology and have elevated it to a national strategy. In July 2017, the State Council issued the "New Generation Artificial Intelligence Development Plan" and the Ministry of Science and Technology announced the first batch of artificial intelligence open innovation platforms to promote artificial intelligence innovation and large-scale application, and promote the deep integration of artificial intelligence with the real economy. As the core driving force of a new round of industrial transformation and a new engine for economic development, artificial intelligence technology will drive various industries to form new demands for intelligence, give birth to a large number of new intelligent technologies, new products, and new industries, and promote the leap of society from digitalization and networking to intelligence.

The development of smart grids also provides a broad platform for the application of artificial intelligence technology. Data-driven power artificial intelligence technology will play an increasingly important role and will become an important strategic direction for the development of power grids and an inevitable solution for the intelligent development of power grids.

Artificial intelligence technology is an important support for the construction of a new generation of power systems and an important engine for promoting innovation in power grid management methods.

Artificial intelligence technology has broad application prospects in the fields of power grid construction, operation, decision-making, and management. It will play a positive role in improving the control capabilities of large power grids, ensuring energy security, and better serving economic and social development.

At the seminar, Yang Jinbai, deputy general manager and member of the party group of State Grid Corporation of China, said that the company is making every effort to promote the deep integration of artificial intelligence and power grid production and operation, forming an artificial intelligence innovation system layout with unified planning by the headquarters, technical and equipment support provided by directly affiliated scientific research and industrial units, and implementation of applications by provincial (municipal) power companies, and clearly designated China Electric Power Research Institute as responsible for the application technology research of artificial intelligence in power grids. This is an important measure for the company to strengthen its R&D strength in the field of artificial intelligence applications and improve the layout of the artificial intelligence innovation system, which shows the company's confidence and determination to support the development of artificial intelligence technology.

As a scientific research unit directly under the State Grid Corporation of China, the China Electric Power Research Institute has been continuously conducting research on the application of artificial intelligence in the power sector for more than 30 years, and has achieved phased results in intelligent inspection of power transmission and transformation projects, auxiliary decision-making for power grid operation, and intelligent mining and pattern recognition of power distribution and consumption data.

Wang Jiye, deputy general manager (deputy president) of China Electric Power Research Institute, pointed out in his keynote report entitled "Key R&D Directions and Development Plans for Artificial Intelligence" that compared with the traditional power grid, the energy Internet is no longer an industrial system consisting of sources, networks, and loads, but an open and shared interconnected system that includes social factors and human will, with the characteristics of extensive interconnection, equal access, and the characteristics and effects of the Internet. Compared with the traditional Internet, the energy Internet is also fundamentally different. The traditional Internet is model-driven, while the energy Internet is technology-centered. Adopting data-driven artificial intelligence technology is an important means to support the new generation of power systems.

The keynote report elaborates on the layered architecture of the seven core technical directions of power artificial intelligence. Among them, intelligent sensing aims to build a full-service ubiquitous power Internet of Things, integrates advanced technologies and innovative research, and breaks through core technologies such as advanced perception, edge intelligence, network connection, and micro-source energy extraction; the artificial intelligence platform will provide a rich library of algorithms and models, and form a one-stop service for the entire process of artificial intelligence application research and development; big data will continue to study data-driven research methods, focusing on technical research and application needs and business model research such as big data unified platform, data analysis and mining, and unified data model; machine learning will build a machine learning algorithm library for power systems and develop machine learning application models for specific power business scenarios; computer vision will focus on the development of power image resource libraries, power image intelligent auxiliary annotation technology, power equipment detection and defect recognition algorithms, form equipment-level power vision modules, and realize real-time equipment recognition and defect detection; natural language processing will build a power theme word list, study power text feature extraction, ontology modeling, knowledge processing and reasoning technologies, build knowledge graphs in the fields of control, operation and inspection, and marketing, and ultimately form a new generation of power intelligent search and question-answering solutions; intelligent robots will focus on breakthroughs in core technologies such as intelligent algorithm encapsulation, autonomous recognition, autonomous behavior, autonomous learning, and human-machine collaboration to achieve autonomy and intelligence of power robots.

The seminar looked forward to the application scenarios of power artificial intelligence technology in four major areas, including new energy consumption, large power grid security and stability, emerging load perception and prediction, power asset management and intelligent operation and maintenance. It proposed to build the AlphaGo of the energy and power system, realize the unmanned driving of the energy and power system, and ultimately form the development goals of the digital and physical dual planes of the power grid.

The key technologies and development trends of artificial intelligence show that power artificial intelligence has broad prospects

The seminar invited Tan Jianrong, an academician of the Chinese Academy of Engineering, to introduce the current status and development trends of intelligent robots at home and abroad. He proposed that the essence of intelligent robots is to work together with people, and the key to improving the intelligence of robots lies in the development and application of artificial intelligence technology. Manufacturing is a basic industry that supports human survival and development, and countries around the world attach great importance to manufacturing. The core of intelligent manufacturing lies in improving the intelligence of machines, that is, the combination and application of artificial intelligence technology. Artificial intelligence should be based on knowledge base and knowledge engineering. "Without knowledge, artificial intelligence is empty, and becomes water without a source and a tree without roots." Tan Jianrong said that the reason why people have intelligence is because they have knowledge, and the core of artificial intelligence technology application is knowledge engineering.

Tan Jianrong believes that in the past, intelligent manufacturing was machine-centric; now, it is necessary to emphasize human-machine and brain-machine interaction. In the past, it emphasized individual intelligence, but now it is necessary to emphasize network-based group intelligence.

In addition, Tan Jianrong also sorted out the ten key technologies of intelligent equipment and intelligent robots. The first is the vision-based three-dimensional environmental perception and recognition technology, the second is the multi-source multi-channel sensing heterogeneous data fusion technology, the third is the global precise positioning technology based on environmental understanding, the fourth is the cluster interaction and control technology of multi-machine collaboration, the fifth is big data-driven fault diagnosis, the sixth is big data-driven multi-mode fusion learning, the seventh is robot motion path planning under multiple constraints, the eighth is multi-task adaptive collaborative group intelligence decision-making technology, the ninth is the product knowledge graph and knowledge network construction, and the tenth is the knowledge service technology of the electromechanical and hydraulic integrated cloud platform.

Li Juanzi, professor of the Department of Computer Science at Tsinghua University, gave an academic report titled "Knowledge Engineering and the Construction of Knowledge Graphs in Its Field", which explained that knowledge engineering is a very important branch of artificial intelligence. It is a transformation process from seeing data, understanding data, transforming knowledge and forming intelligence, which can help us make decisions and give us a better way of living or producing. She summarized the three key technologies of knowledge engineering: first, knowledge representation and calculation that combines knowledge guidance and data drive; second, platformization of knowledge graph technology and establishment of knowledge engine; third, diversification of knowledge graph applications and innovative knowledge applications.