This project belongs to the field of power systems and automation, involving advanced measurement, demand response, user interaction, high-speed real-time communication, intelligent hardware, data mining and other technologies.
Smart grid is an important part of the national energy strategy and one of the high-tech industries identified by the Shanghai City Government. Smart electricity consumption is an important part of the smart grid. Through intelligent power consumption systems, implementing flexible dispatch and control of loads, distributed power supplies and energy storage devices can effectively improve the efficiency of power grid energy and facilities, promote economic operation and orderly power use, and have important implications for power grid construction and operation, energy conservation and environmental protection, and power quality management has far-reaching implications and great significance.
Supported by the National Natural Science Foundation of China and Shanghai City's key scientific and technological research projects, the research team carried out research work on two-way interactive intelligent power consumption and flexible load control technology. The main innovations of the project include:

1. For the first time, a refined short-term load prediction method for block-level loads is proposed. The chaotic time series method is innovatively used to effectively construct a load dynamics model that includes mechanism-driven and data-driven through phase space reconstruction. It overcomes the limitations of relying only on a certain factor to predict power loads, reduces the impact of original data anomalies and data noise on prediction accuracy, and the prediction error is less than 5%.
2. A load identification technology based on non-intrusive load monitoring is innovatively proposed. A sliding window bilateral cumulative sum method is used to detect load start-up and stop transient events, the problem is transformed into a 0-1 quadratic programming model, and the model is solved by implicit enumeration method. Significantly reduces the number of metering sensors and reduces communication costs. The innovatively proposed load classification method based on information entropy segmented aggregation approximation and spectral clustering can effectively carry out load classification and dimension reduction processing to evaluate the dispatchable capacity of loads participating in demand response, and is helpful to the formulation of demand response control strategies.
3. Innovatively proposed a low-carbon and multi-objective flexible load control technology and energy management strategy. Optimization goals including low-carbon benefit parameters are introduced, and a low-carbon comprehensive benefit model is established. By constructing distributed cluster control of loads and distributed power sources, flexible loads participate in power grid regulation and two-way flow of energy between distributed clusters and main grid are realized. Control strategy, effectively achieving multi-objective optimization of low-carbon benefits and economic benefits.
4. Developed key equipment for intelligent electricity consumption such as intelligent load controllers and two-way interactive terminals. The load controller has built-in precision integrated circuits, processors, metering chips and ZigBee wireless communication, which can collect data and control on and off of various electrical equipment. Two-way interactive terminals realize the reception, storage, real-time display and analysis of information and data of smart meters and smart controllers, complete two-way interaction between users and power companies, implement flexible scheduling and control of loads, distributed power supplies and energy storage devices, and support demand response mechanism.
   This project has published 15 papers in SCI/EI journals and conferences, and applied for 6 patents (2 of which have been authorized). On the basis of theoretical research and talent training, we explored the linkage between technology transformation and engineering application, and initially formed a complete "industry-university-research" cooperation chain. In the past three years, the new products jointly developed by the project team and the cooperating units have added an added value of 28.1178 million yuan, an added profit of 9.8413 million yuan, and an added tax revenue of 1.673 million yuan.
   After expert evaluation organized by Shanghai City Science and Technology Commission and novelty search by science and technology information department, the research of this project is novel. The effects of technical achievements transformation, product marketing and practical application show that the load prediction, identification and control technology in this project is pioneering, and the project and product-related technologies are at the leading level in China. The project research has improved my country's intelligent electricity technology level and key technical equipment development capabilities, and achieved significant economic benefits and good social benefits.