The Real-Time Pricing Method For Smart Grids Based On Nonsmooth Equations

Demand side management in smart grid refers to management activities that encourage power users to actively participate in power operation and optimize power usage,thereby improving terminal power efficiency,achieving optimal resource allocation,and minimizing power supply costs.Demand response is the core of demand side management.It is a measure that encourages the users to participate in price setting.It plays an important role in the balance of supply and demand,energy conservation and emission reduction,and shifting the peak loading of smart grids.It can guide the development of the demand side management of smart grids.The real-time pricing is an effective method to achieve demand response management.Its implementation will have a profound impact on the operation and management of smart grids.At present,the researches on real-time pricing for smart grid are mainly based on the social welfare maximization model,which aims to maximize the utility of users while minimizing the cost of power providers.In this dissertation,optimization theory is used to research the real-time pricing problem based on social welfare maximization model.The models are built for fifferent conditions,and then algorithms are designed to solve the real-time pricing for smart grids.The main innovation work of this dissertation is as follows:Firstly,based on social welfare maximisation model,this dissertation considers the KKT system to establish the real-time pricing model for smart grids.A system of nonsmooth equations with the Lagrange multipliers is creatively established.The Lagrange multipliers,i.e.,the shadow prices are used to determine the basic price of electricity.A centralized algorithm based on the system of nonsmooth equations is designed to solve real-time pricing for smart grids.Compared to the dual method,the Lagrange multiplier is a decision variable and plays the same role as electricity consumption and generating capacity variables in the nonsmooth equations.The numerical results of Lagrange multipliers are more accurate in the nonsmooth equations.The proposed method reduces the deviation between the total electricity demand and production.Secondly,considering the differences between different users in electricity demand and consumption satisfaction,a multi-user social welfare maximization model is established.It is useful to solve the problem that the utility function is single in real-time pricing researchs.Simulation verifies that the model is reasonable and the proposed algorithm is feasible.Finally,considering the influence of uncertainty load,the electricity demand is divided into basic demand and uncertain demand.The social welfare maximization model with the uncertainty load is established.Then based on the social welfare maximization model,an online distributed algorithm is designed to solve the real-time pricing for smart grids.In the proposed method,the real-time pricing is considered on two sides: the users determine the electricity demand according to the price given by provider;the provider decides the price and production based on users demands.The method can effectively avoid the shortage of supply and demand,promote the user to actively participate in the demand side management of the smart grid,and improve the social welfare.In above researches,the smoothing Newton method is used to solve real-time price,electricity demand and production for smart grids.The proposed method can improve the social welfare,multiplier numerical-precision and computation speed of the algorithm.