Prospects for “intelligent business” derived from ubiquitous power Internet of Things

The construction of the ubiquitous power Internet of Things is the core task of the State Grid Corporation of China's new strategy of "three types and two networks, world-class". Since its proposal, it has attracted widespread attention from all walks of life and has become a new outlet in the capital market. This is due to the broad imagination space it brings in terms of industrial upgrading, capital investment, scale effect and ecological construction, involving equipment manufacturers, upstream and downstream enterprises, Internet companies and other parties. The ubiquitous power Internet of Things will provide a physical foundation, promote the improvement of quality and efficiency of traditional businesses, and derive new businesses, new formats and new business models. This new change, without exception, shows the characteristics of "intelligent business", and the core is driven by data and intelligence.

The basic logic is to continuously feedback and improve the business closed loop with data as the core element

The booming development of my country's consumer Internet and the rapid rise of the industrial Internet benefit from the long-term accumulation of various data resources, as well as the high computing power brought by cloud computing and the development of artificial intelligence algorithms. More directly, relying on the practical application and commercialization of artificial intelligence technology, machine learning algorithms represented by deep neural networks continue to flourish in scenarios such as AlphaGo and image recognition. They all take data as the core element, forming a closed-loop positive feedback of learning, prediction, application, and learning. The actual results are used as identification data feedback to "supervise" the improvement and optimization of the algorithm model, and continuously improve the accuracy from input to output. This thinking logic is similar to the logical system of Internet businesses such as Taobao.

The main goal of artificial intelligence at present is to make more accurate predictions. The essence is probabilistic calculations based on association analysis. The efforts of various new algorithms are aimed at improving accuracy and computing efficiency. If high-precision predictions are achieved through fast calculations in certain scenarios, Taobao merchants can recommend products that users really need, JD.com can pre-arrange the location of goods to achieve same-day delivery, and AVIC Cloud Network can improve the process in advance to reduce product defect rates. The core is to better meet the needs of users and related parties, and even create new businesses to accurately match users' potential needs. This is also the goal pursued by State Grid in building hub-type, platform-type, and sharing-type enterprises.

The ubiquitous power Internet of Things will bring together resources from all aspects and provide sufficient and effective information and data support through the widespread application of information technologies and intelligent technologies such as big data, the Internet of Things, edge computing, and blockchain. Its value needs to be better explored and utilized by building a data-based continuous feedback and improvement business closed loop.

The "intelligent business" derived from the ubiquitous power Internet of Things will be based on energy flow and data flow, relying on artificial intelligence technology to promote the continuous improvement of business flow, and mainly includes two categories: the first is smart energy service, which uses digital management to improve energy efficiency, mainly to intelligently optimize energy service businesses including power supply; the second is the digital economy in the power vertical field, which uses energy data to explore new growth points, mainly using electricity consumption data, equipment operation data, etc. as the main means of production, to generate new business, new business forms and new business models. For example, the "double 10 billion" inclusive finance launched by State Grid and China Construction Bank for small and micro enterprises uses electricity data to build a credit model to provide banks with credit investigation methods, and can subsequently use the actual loan repayment data of the loan enterprises to continuously improve the credit model.

Dedicated models that combine vertical domain expertise with artificial intelligence are core technologies

Although we are still in the weak AI stage, the application of AI technology in vertical fields such as face recognition has reached a very high level, and relies on the continuous feedback of new data to continuously improve accuracy. For example, graphic recognition is almost error-free, and it is difficult for humans to defeat AlphaGo. It can be foreseen that in the future, with the continuous improvement of AI algorithms and the development of AI-specific chips, specific intelligent analysis and prediction problems will be solved better and better, and even small smart terminals connected to the Internet of Things can achieve good results.

In the process of realizing intelligent scenarios, how to digitize, model, intelligentize, and softwareize the business is the key. Based on the current storage and processing capabilities, as much data as possible should be collected. Some data is useless at present, but may be useful in new scenarios in the future. Modeling and intelligence are to achieve the automation of analysis and prediction. By combining vertical field expertise with general artificial intelligence algorithms, special models from input to output are built to form the core technology of various Internet platforms.

The ubiquitous power Internet of Things will derive several platforms in various vertical fields such as power credit investigation. Whether it is energy services or digital economy, even if it is not necessary to establish an intelligent analysis model in the early stages of business development, the ultimate form will definitely use intelligent analysis as the core technology, which requires professional talents and artificial intelligence technical talents to cooperate in research and development. For example, power credit investigation can use the simplest electricity consumption as the only standard for determining credit; at a higher level, it can build an intelligent analysis model of the relationship between electricity consumption, load, electricity bill payment time, equipment operation and other data and credit rating, and more accurately predict the company's future loan repayment behavior.

Ubiquitous power Internet of Things is a new infrastructure to enhance strategic competitiveness

The power industry is a basic pillar industry for economic and social development. The power grid is the most widely covered physical network and a hub connecting the power production and consumption ends. The convenience of power transmission determines that the transformation and upgrading of various industries must rely on power to provide power, and a large amount of data from the whole society flows on the power grid platform. Therefore, power data is one of the most suitable data to depict the status of the macro-economy, meso-industry, micro-enterprises, and individual users. Moreover, power data is a physical quantity transmitted in real time by equipment, which is more accurate than other values.

In the future, the massive data collected by the ubiquitous power Internet of Things, supported by dedicated intelligent analysis technology, can provide a full range of "intelligent services" for the development of the entire economy and society, and become an important support for improving the strategic competitiveness of the country, industry and enterprises. From the perspective of serving the government, it can provide advanced, quantitative and highly accurate basis for situation analysis and policy formulation; from the perspective of serving the real economy, it can provide in-depth analysis services for power equipment manufacturers to upgrade their products; from the perspective of serving the energy industry, it can promote efficient and accurate matching of supply and demand, promote the consumption of new energy and improve energy efficiency; from the perspective of serving users, it can provide high-quality comprehensive energy services and data-derived services; from the perspective of power grid companies, it will promote the accelerated transformation and upgrading of traditional industries.

Looking ahead, with the development of the ubiquitous power Internet of Things, more and more people, objects and machines will be connected to this system, and the three types of resources, energy, data and computing power algorithms, will be deeply integrated, and together they will build a physical foundation for various new "smart businesses", just like cloud computing and 5G, providing universal and standardized basic services. The ubiquitous power Internet of Things will work together with the strong smart grid to play the role of energy Internet infrastructure and serve the high-quality development of the economy and society. (Author: Zuo Xinqiang is the director of the office of State Grid Energy Research Institute Co., Ltd.)