Research On The Intelligent Power Demand Response Of Industrial Users

With China’s increasingly serious energy problem, demand response has become an increasingly important role in the power system for power industry sustainable and the optimize utilization of social resources and energy. The demand response as a demand side resource has the remarkable effect for improving load and energy efficiency, reducing power system peak valley difference. With the development of smart grid technology, the promotion and application of demand response has been promoted greatly. Industrial users become a good platform of demand response because for its large individual power consumption, load adjustment faster, round-the-clock electricity, etc. The multi-agent simulation modeling method comes from the development of artificial intelligence, through the simulation of microscopic behavior reflects the macro system, and provides a feasible approach to study the mechanism of demand response and the implementation effect.This thesis introduces the definition, background and basic theory of demand response, analyzes the demand response measures based on price and incentive. At the same time, introduces the basic theory of smart grid, focuses on the study of which can provide technical support for demand response. According to the characteristics of industrial users, this thesis summarizes several typical industry load characteristics and the demand response potential. Some problems such as types of industrial users, the way of response, response of load classification are studied. In the aspect of intelligent theory, this thesis introduces the basis of intelligent theory, agent and multi-agent related concepts and theories. The multi-agent theory is applied to the price and incentive forms of demand response, and makes two demand intelligent response systems.In the study of demand response based on the price, I summarize the traditional price demand response mode. For some of these shortcomings, this thesis proposes an improved model based on time-of-use pricing, and builds a TOU price adjustment system by using the theories of multi-agent system. The supply side can adjust the electricity pricing in accordance with certain rules, according to user response to different price. At the same time the users make response to adjust production according to different electricity price. The supply and demand sides make two-way interaction based on this. In order to prevent excessive price adjustment, the supervision system of the intervention function is designed. Finally, this system achieves the desired load reduction under multilateral interaction effect.In the study of demand response based on the incentive, this thesis use a demand side bidding model which different from the traditional way of signing a contract, for studying the project of interruptible load. According to the characteristics of industrial users, this thesis designs a power adjustment module in the user side. The module can adjust production according to different demand, in order to provide different capacity of interruptible load. On this basis, using the artificial neural network intelligent prediction technology predicts the bid winning price in order to maximize the interests of users. Finally, the system will use interruptible load applied to congestion management in the electricity market. This thesis constructs an interruptible load bidding model of congestion management system, and providing a reference for the application of the incentive intelligent demand response. Also this system provides a reference on study and implementation of bidding interruptible load in the liberalization of the electricity market for future.