Interactive Smart Grid Load Forecasting And Prcing Mechanism

With the constant development of economic, the contradiction betweenthe environment and economic development is more and more serious. It is ahot study that how to ensure the interests of development and reduce thepollution at the same time. So smart grid arises at the historic moment. It is agloble aim that constructing a grid which is reliable, safe, economic andoperating efficiently in order to reduce the costs of operation andconveniently for the distributed energy and new energy to access so that wecan reduce the dependence on fossil fuels. Electrical demand response use theinformation from demand side in real-time so power supply department canuse the electrovalence to change user traditional use way of electricity inorder to change load curve. Electrical demand response in smart grid mainlyuse two ways such as one based on incentive and another based on price.Using its main projects: the demand side response based on excitation bydirect load control (DLC), the price of demand side response based on thereal-time pricing (RTP), so as to achieve the purpose of demand sidemanagement.1.Based on the interaction of the smart grid information flow and gettingreal-time demand-side information, this paper used traditional demand-sidemanagement projects, fusing the technical features of the smart grid,established a simulation of an interactive smart grid system in Shanghaiincluding Photovoltaic、wind power generation, battery energy storage anddiesel generators. It Simulates the total PV array output power is150MWp,which is about1%of electricity generation in Shanghai and the total wind power is1000MWp, which is about9-10%of electricity generation inShanghai.By analyzing the device characteristics and studying theoreticalformula,establishing a mathematical model of a suitable device simulationsystem including each component.Using the HOMER software to get asimulation of distributed energy resources including interactive smart gridsystems, inputing energy resources and the equipment cost data to calculatethe annual output of the simulation system via the constructed mathematicalmodel.2.Processing the data get from simulation in HOMER software to obtainthe real load data sample of the distributed energy systems. In order toimprove the accuracy of load forecasting, in this paper, the auther uses datainterpolation expanded to96points. Combined with wavelet neural networkand the characteristics of distributed energy system, the auther design aforecasting model based on wavelet neural network for interactive smart gridload to verify the accuracy of the Shanghai area load data obtained fromHOMER software simulation. By analysing the data， the results showed thatthe distributed generation can play the role of peak load shaving directly.Distributed generation is one of the effective means in demand-sidemanagement based on motivation, and this paper makes some suggestions tothe distributed energy system.3.By analysing the results of load forecasting, the auther propose aconcept of quasi-real-time pricing based onload fluctuations. Comparing thepeak and valley price ladder performed in Shanghai now, combining theinteractive smart grid load data from the simulation including the distributedenergy systems in Shanghai region, and according to the characteristics of theShanghai residential electricity, based on price-based demand responseprogram, the auther propose a concept of quasi-real-time pricing andconstruct a quasi-real time pricing which is suitable for the situation inShanghai, and fix the matrix model of the elasticity of electricity aftercarrying out the quasi real-time pricing to study the electricity situation of theresidents in the implementation of the tariff situation. At last, analysing the effect that how the price will influence the the electric car charging anddischarging in the grid.