Focus on power grid control artificial intelligence to help build "three types and two networks"

On September 7, a seminar on artificial intelligence application technology in the field of power grid control was held in Beijing. The meeting was hosted by the State Electric Power Dispatching and Control Center and organized by the China Electric Power Research Institute and the State Grid Energy Internet Technology Research Institute. It is committed to promoting the integration of artificial intelligence and power grid control business, using artificial intelligence to drive the innovation of traditional control technology models, and promoting control operation and management to a new level.

State Grid Corporation of China implements the new energy security strategy of "four revolutions and one cooperation". The new form of the power grid, the new changes in regulation and operation, and the market-oriented reform of electricity provide space for the development and application of new technologies. Chen Guoping, chief engineer of the company, said at the seminar that we should combine domestic reality, grasp the laws of the power grid, seize the opportunities of new energy consumption and market-oriented reform, promote the application of artificial intelligence technology in the field of power grid regulation, and help my country's energy and power technology development and progress. Li Mingjie, director of the National Electric Power Dispatching and Control Center, believes that national and provincial dispatching should propose artificial intelligence application topics based on demand, publicly release them to the society, and jointly tackle key problems through industry, academia and research. Actual regulation work should carry out artificial intelligence application technology innovation and management innovation according to local conditions to support the implementation of the company's "three types and two networks, world-class" strategy.

Exploring the need for artificial intelligence in power grid regulation

Over the past two decades, my country's energy consumption has continued to increase, electricity has grown rapidly, the energy revolution has accelerated, the UHV national interconnected power grid has taken shape, and the integrated power grid operation has become increasingly prominent. Power grid dispatching and operation is an extremely complex systematic project. Ge Rui, director of the General Office of the National Electric Power Dispatching and Control Center, believes that artificial intelligence, as a cutting-edge technology, can simulate human thinking and quickly handle complex and multi-dimensional problems, and has a relatively broad application prospect in power grid operation and control.

Nowadays, power electronic equipment is increasingly used in power grids, and the power supply, power grid and load structure have undergone profound changes. In view of the current status of power grid development, Yan Jianfeng, chief engineer of the State Grid Simulation Center of China Electric Power Research Institute, believes that the digital era in the field of energy and power has arrived. Multiple factors in various links of energy production, transmission and consumption are widely connected and integrated to form a new open and ecological system. Artificial intelligence will empower the digitalization of energy and power.

At present, the uncertainty of power grid operation has increased significantly, and the difficulty of real-time control and operation has continued to increase. However, real-time control and operation of power grids is based on experience-based dispatching. Dai Zemei, deputy director of the Nanrui Research Institute and director of the Two-Grid Integration Center, believes that the application of artificial intelligence technology aims to improve the accuracy of prediction and analysis, reduce repetitive dispatching work, and improve the efficiency of power grid operation. However, the requirements for safety and reliability of power grid control and operation are extremely high, and it is necessary to focus on the safe and reliable application of artificial intelligence.

Faced with the new generation of complex power grids, more and more power electronic equipment is sensitive to voltage, the system inertia is getting lower and lower, and the frequency and voltage regulation capabilities are seriously insufficient. The traditional dispatch decision-making method based on pure model drive is difficult to meet the requirements. "The development prospects of artificial intelligence technology are broad. The cross-section safety AO (intelligent machine dispatcher) system, which has been developed for 17 years, has been applied in Guangdong Central Dispatching. It is just the beginning of dispatching from automation to intelligence, and it is still very preliminary." Sun Hongbin, a professor at Tsinghua University, said at the discussion meeting that this AO technical architecture can be further used to solve problems such as voltage safety and frequency safety. It has universality and broad application prospects, but it also faces many major technical challenges.

Promote the integration of artificial intelligence and power grid control services

In January this year, State Grid Corporation of China proposed the strategic goal of "three types and two networks, world-class", and will make full use of modern information technologies and advanced communication technologies such as mobile Internet and artificial intelligence to create a ubiquitous power Internet of Things with comprehensive status perception, efficient information processing, and convenient and flexible application.

In March, the company released the Outline for the Construction of Ubiquitous Power Internet of Things, which included artificial intelligence in the scope of technical research and core products. In May, the National Control Center organized a meeting to promote the dispatching of ubiquitous power Internet of Things and proposed the application of artificial intelligence based on the control cloud.

The State Grid Tianjin Electric Power Dispatching and Control Center is based on intelligent human-computer interaction, cloud-edge collaborative computing, big data intelligent analysis and other aspects to enrich the application scenarios of dispatching and control business and adapt to the new needs of power grid dispatching. Guo Lingxu, deputy director of the Equipment Monitoring Department of the State Grid Tianjin Electric Power Dispatching and Control Center, said that in the future, it is necessary to closely combine the construction of the control cloud with the application of artificial intelligence, make full use of the cloud computing power and big data storage capabilities of the control cloud, create intelligent application scenarios, and expand the application space of artificial intelligence.

State Grid Qinghai Electric Power has introduced blockchain technology into shared energy storage auxiliary service transactions, using the four core technologies of blockchain distributed storage, encryption algorithm, consensus mechanism and smart contract to solve the problems of credibility and clearing and settlement of power market transactions. Dong Ling, deputy director of the State Grid Qinghai Electric Power Dispatching and Control Center, said that the next step will be to build a Qinghai unified power market transaction data sharing business platform based on cloud platform architecture and blockchain technology to promote the application of blockchain in the transaction market.

At the seminar, representatives from Huawei, Alibaba, iFLYTEK and other companies said that they have completed the implementation of artificial intelligence technology in their respective companies, which will release new productivity potential.

Huawei's self-developed artificial intelligence platform and self-developed artificial intelligence chips are autonomous and controllable. Zhang Song, general manager of Huawei Cloud EI Service Product Solutions, said that artificial intelligence is becoming a driving force for the intelligentization of technology companies' industries, and can help companies quickly integrate artificial intelligence technology into the industry.

"The technology of Alibaba Cloud DAMO Academy's City Brain Laboratory can achieve cognitive understanding and inductive reasoning. The City Brain artificial intelligence open innovation platform can realize video image search, abnormal event recognition, scene recognition, city modeling, and city deduction." Zhou Chang, technical director of Alibaba Cloud DAMO Academy's City Brain Laboratory, took the City Brain project practice carried out by Alibaba Cloud in various places as an example to share the experience of implementing artificial intelligence technology scenarios.

iFlytek has built a unified service platform for artificial intelligence capabilities based on the intelligent control application practice of advanced voice technology. Liu Guoguang, director of the company's energy business line, said that this platform can provide core perception and cognitive support for scenarios such as power grid dispatch management, operation, information retrieval, and identity authentication, and deeply optimize and continuously train language models, acoustic models, and semantic libraries based on the characteristics of the power industry and dispatching fields.

This seminar deeply integrated artificial intelligence technology with control business needs, enhanced the consensus on demand cooperation, and was conducive to further promoting the application of artificial intelligence technology in the field of power grid control.