news

from september 7 to 10, the theme of the conference was “transformation and development for a green future”.2024 global energy transformation conference held。

the reporter of daily economic news noticed that the digital transformation in the energy field, especially the power field, became one of the hot topics discussed by the guests, and the use of big data technology to achieve green, low-carbon and high-quality development received widespread attention. at the event booth, the reporter observed that big data empowering the power field has also become the focus of attention of many companies.

how does digitalization empower the power distribution link? what challenges does the new power system based on new energy bring to power regulation? how does digital technology improve the accuracy and efficiency of power dispatch? in response to these topics, reporters from the national business daily (hereinafter referred to as nbd) interviewedguo mengjie, vice president and secretary of the board of directors of beijing tsinghua keyueconducted an exclusive interview.

the contradictions of the power system's difficulty in absorbing power during lunch, ensuring supply during the evening peak, and regulating operation are becoming increasingly prominent.

NBD：the national development and reform commission and the national energy administration issued the "guiding opinions on strengthening the construction of power grid peak loading energy storage and intelligent dispatching capabilities"it points out that the construction of intelligent dispatching capabilities should be promoted. how are big data and artificial intelligence technologies applied in the power intelligent dispatching system? how do they improve the accuracy and efficiency of dispatching?

guo mengjie: big data and artificial intelligence technologies are two very popular technologies at present, and have been applied to many scenarios and businesses.

the main purpose of big data technology is to manage, process and analyze large-scale, high-dimensional, complex and rapidly changing data, and the data generated by the power system just meets these characteristics.

specifically, it is reflected in data collection and storage. during the operation of the power system, a large amount of data is generated, including real-time operation data such as voltage, current, power, frequency, and non-electrical quantity data such as equipment status and protection signals. these data are characterized by fine time granularity, real-time, and massive. big data technology can efficiently collect and store these massive data, providing a basis for subsequent analysis and processing.

in the data analysis phase, the collected data can be deeply analyzed and mined to discover the patterns and correlations in the data, providing decision support for the operation and maintenance of the power system. at the same time, the equipment status can be monitored and warned in real time, potential faults can be discovered in time, and measures can be taken to prevent the expansion of faults and accidents.

in addition, after collecting various types of data from the power system in real time, big data technology can monitor the real-time data of the power system operation, ensure the accuracy and timeliness of the data, lay a good foundation for subsequent work based on artificial intelligence, optimized decision-making and other technologies, and improve the accuracy and efficiency of related businesses.

artificial intelligence technology has a wider range of uses, such as time series prediction, classification decision-making, multi-modal large models, etc. in the power system, time series prediction is the most widely used, and large models and decision optimization technologies are also emerging in many scenarios.

specifically, time series forecasting refers to new energy power forecasting and load forecasting. both are the top priorities for the basic work of power system dispatching and operation. all decision optimization and stable operation of the power grid require accurate forecasts of the two. artificial intelligence technology can establish a time series forecasting model based on historical data and other relevant factors, such as weather, season, consumer electricity habits, etc., to achieve high-precision forecasts of future power and load changes.

in terms of big models and decision optimization technologies, which are of general concern to everyone, since the business rules and goals of different scenarios in the power system are not completely consistent and the models are different, traditional physical modeling alone will cost a lot of manpower and material resources, while the use of generative big models can effectively predict and make decisions. business personnel can complete the input through language descriptions, and after receiving the ai ​​big model, it automatically establishes a mathematical model for prediction and decision-making, which can effectively reduce the difficulty of model construction and realize the comprehensive promotion of "generalized" solutions.

with the continuous improvement of artificial intelligence technology and its popular application in power systems, it can not only assist people in load and power forecasting, formulate more accurate data boundaries, achieve supply and demand balance in scheduling plans, and effectively ensure the stable operation of the power grid, but also realize cross-system and cross-regional data sharing and coordinated scheduling, optimize the configuration and utilization of power resources, and improve the operating efficiency of the entire power system. in the future, big models will simplify the business processes of the power system and bring great convenience to the entire industry.

nbd: the construction of new power systems has put forward new requirements for power dispatching. what challenges does this bring to the current dispatching operation? what technologies are needed for power dispatching under the new situation?

guo mengjie: my country's new power system is being built at an accelerated pace. the construction of a new power grid with ac/dc hybrid and main and micro grids is being promoted in an all-round way. new energy sources are developing at a high speed and achieving high-level consumption. however, at the same time, the power structure, grid form, and load characteristics are undergoing extensive and profound changes, and the grid dispatching and operation are facing new situations and new challenges.

on the one hand, the strong random characteristics of the power supply are significant. with the rapid growth of new energy installed capacity, the randomness, intermittency and volatility of new energy output in time and space dimensions have become more prominent. combined with the characteristics of "quiet and windless", "no light at night" and susceptibility to rain, snow and ice, the operation and regulation of the power grid are facing outstanding challenges. at the same time, new energy power generation devices that lack inertia and autonomous voltage reference show obvious low interference resistance and weak support, which also brings challenges to the safety and operation control of the power grid itself.

on the other hand, it is difficult to coordinate the control of the main and distribution networks. the power system is gradually developing towards the "double high" direction of "high proportion of renewable energy" and "high proportion of power electronic equipment", and the system characteristics have changed profoundly. ac and dc, source, grid and load are coupled with each other, the scale of conventional power supply startup has decreased, the frequency and voltage support capacity of the power grid has weakened, the fault form of the power grid has become more complex, and stable control has become more difficult.

distributed photovoltaic, energy storage and active distribution networks are developing rapidly, and the mode of deep coordination with the main grid needs to be improved. the demand for dispatching services of local and county-level power grids has increased significantly, and the difficulty of management has increased. the fluctuation of new energy output leads to "tidal" fluctuations in power grid currents during the day. the contradictions of the power system's difficulty in absorbing power during the lunch break, ensuring supply during the evening peak, and operating and regulating are becoming increasingly prominent and are gradually becoming the norm.

in view of this situation, we believe that under the new situation, as a dispatching department to ensure the safe and stable operation of the power grid, more accurate source-load prediction technology is needed. at the same time, people are increasingly relying on source-grid-load-storage collaborative optimization technology based on new generation information technologies such as artificial intelligence to smooth out the fluctuations brought by new energy to the grid dispatch balance.

based on the new business forms and models of virtual power plants, as well as aggregated regulation mining and flexible interaction technologies, we can fully tap and promote the participation of flexible resources such as user load, power supply, and energy storage in the regulation of the power grid, guide them to actively interact with the dispatching department through the market price mechanism, and jointly maintain the safe and stable operation of the power grid.

electricity-carbon synergy requires the improvement of market mechanism design at the national level

nbd: in the future, when renewable energy will be connected to the power grid on a large scale, how can smart power generation technology help achieve accurate matching of power supply and demand? what is the difference between it and traditional power generation technology?

guo mengjie: the safety of the power grid depends on the coordinated operation of all links of "power generation, transmission, distribution, and power consumption", not just "power generation". with the large-scale access of renewable energy, especially the randomness and volatility of wind power generation and photovoltaic power generation, it is difficult to accurately calculate the peak load regulation, frequency regulation, and backup capacity of the power system, which is easy to cause problems such as wind abandonment, solar abandonment, and load shedding, bringing challenges to the safety and stability of the power grid.

traditional power plants find it difficult to meet the requirements of flexible operation of generators, intelligent trading, and ultra-low emissions for the “dual carbon” goals under the background of new power systems and power markets.

smart electricity sales technology is the application of intelligent technology to the process of power generation and electricity sales (trading). smart electricity sales uses digital signals as a carrier, integrates electronic information, intelligent control, cloud computing, big data technology, economics, operations research, etc. into power production and operation, and comprehensively improves the level of intelligent power generation operation and transaction decision-making.

in particular, the application of the new generation of artificial intelligence (ai) technology, big data technology, etc. in load forecasting and power forecasting can further improve the accuracy of new energy power generation and load forecasting, help improve the balance of supply and demand in the power grid, and promote stable operation of the power grid.

in addition, intelligent technologies for electricity sales and marketing can build more marketing models to truly meet diversified electricity demands, mobilize the coordinated development of "source, grid, load and storage", and thus promote the sustained and stable development of the power system.

nbd: electricity-carbon synergy is also a topic that has been discussed in the energy field. how do you think we can promote the effective connection between the two markets and open up the "last mile" of electricity-carbon synergy? can electricity-carbon synergy reduce electricity costs?

guo mengjie: my country's electricity market and carbon market have both developed significantly in recent years. in some pilot areas, an effective conversion mechanism has been initially established between green electricity and carbon emissions. by purchasing green electricity, enterprises can not only meet their energy needs, but also effectively offset some carbon emission indicators. however, few users can accurately explain the synergy between the carbon market and the green electricity market, precisely because synchronization and interoperability are still lacking.

we believe that, first of all, the power market and the carbon market need to be promoted in a coordinated manner in terms of market design and market mechanism construction. the national level should coordinate and formulate the overall goals and development ideas for the integration of the power and carbon markets, improve the design of market mechanisms, and ensure that the power market and the carbon market are consistent in terms of goals, tasks, and construction schedules. at the same time, policy adjustments should be strengthened to avoid policy conflicts and repeated incentives, and provide a solid policy foundation for market integration.

secondly, market information exchange is the key to achieving integration. further improving market transparency, unblocking market information exchange channels, strengthening public data information sharing between electricity and carbon markets, achieving interconnection and unified certification of environmental rights and interests data between the electricity market and the carbon market, and strengthening the mutual trust and sharing of data information, credit information and regulatory information can all help market players make better independent decisions when considering environmental rights and interests.

as for whether electricity-carbon synergy can reduce electricity costs, there may not be a direct and significant effect in the short term. after all, additional environmental rights and interests demands will bring additional costs to buyers, whether in green electricity trading or carbon trading.

however, in the long run, as renewable energy technology continues to mature and costs continue to fall, the share of green electricity in the electricity market will gradually increase. at the same time, the effective operation of the carbon market will increase the power generation costs of high-carbon emission units, thereby promoting the transformation of the electricity price structure to a low-carbon one. this transformation will also encourage users to increase the use of renewable energy. therefore, in the long run, electricity-carbon synergy will not only help promote the transformation of the energy structure, but will also reduce electricity costs for users.

daily economic news