Science Popularization: how can virtual power plants help carbon neutralization

Recently, a company announced that it has made key achievements in the application of distributed energy commercial virtual power plants: on February 22, it launched its first distributed energy management core product smartrams, a digital energy technology product that enables the operation of virtual power plants, and officially opened the cloud in the product system "Wind and solar power forecasting" customer subscription service has become the industry's leading "wind and solar power forecasting" product that can achieve 100% cloud deployment.

Virtual power plant (VPP) is a power supply coordination management system, which realizes the aggregation, coordination and optimization of various der distributed geographically through advanced communication technology and software architecture, so as to participate in the power market and power grid operation as a special power plant. Its concept more emphasizes the function and effect of external presentation, updates the operation concept and produces social and economic benefits.
This new product smartrams It will improve the accuracy of new energy power forecasting, help distributed generation participate in energy trading, improve asset utilization, and establish a distributed energy data warehouse by providing accurate data services such as power forecasting, trading decision-making and demand side response, so as to provide strong technical support for improving the operation efficiency of virtual power plants and realize the balance between supply side and demand side.
The cloud "wind and solar power prediction" service launched this time is the front-end entrance of smartrams distributed energy management platform. This set of non intrusive new energy power station power prediction system can realize 100% cloud deployment through virtual weather station technology. It only needs one coordinate (station longitude and latitude information) and more than 6 months of historical measured data, and the cloud orders 14 The power prediction data of the whole station can be viewed in real time in the cloud in 10 days. The prediction time span can range from ultra short term in the next 4 hours, medium short term in the next 0-72 hours to the longest in the next 7 days (168 hours).
Compared with the traditional station deployment, the cloud deployment can make up for the lack of single terrain and unit model in the traditional local deployment, improve the prediction accuracy, significantly shorten the project cycle and reduce the operation and maintenance cost of the station, so as to achieve the goal of green and energy-saving low-carbon operation of energy enterprises. With the increase of data samples and the improvement of cloud computing capacity, in the future, cloud orders will be able to use the power forecasting service the next day after the order is paid, just like the current "next day arrival".
In the actual case, there are nearly 20 wind farms and more than 1500 distributed photovoltaic power plants in the jurisdiction of a power supply company. Due to the rapid growth of distributed energy in the area under its jurisdiction, customers urgently need to establish a distributed energy management system to comprehensively manage and analyze the carrying capacity of distributed energy within its jurisdiction. By using smartrams wind and solar power prediction service, each distributed energy station in the area is analyzed in the next 4 hours, 0-72 hours the next day At the same time, with the help of virtual weather station technology, subscribing to the station longitude and latitude numerical weather forecast, and using machine learning algorithm model intelligent adaptation, the expenditure of station environment meteorological instrument, station end server and other hardware equipment is greatly reduced; in addition, the difficulty of operation and maintenance of the whole project is reduced, and the accuracy of system prediction is improved, and the accuracy of ultra short term prediction is more than 85%.