Study On The Real-time Electricity Price Based On The Smart Grid

Energy is the important material basis which the human depends on for survival and development, but the energy will be exhausted gradually and the environment has been worsened increasingly with the progress and development of human society. These issues trouble the further development of human civilization, and they remind human to improve the efficiency of energy development and utilization and to strength environmental protection at the same time. Electric energy is facile to transmit as a clean secondary energy, which occupies a large proportion in energy structure. Human use fossil fuel to generate power and transmit it to various regions to meet the users'consumption needs by grid. In order to satisfy the demand for electric energy when develop social economic and adapt to the diversiform evolution of power generation and consumption, building a safe, reliable, economical, efficient, environmentally friendly and secure usage smart grid is dealt as the national strategy.Smart grids are different in the purpose and focus between different countries because of their actual situation. In smart grid, power companies reduce the peak load and encourage the users to consume power when the load is low by implementing real-time electricity price to guide users to participate in power system operation and management. The main purpose is energy conservation and emission reduction by shifting the peak load and improving the utilization efficiency of network equipment and power generation efficiency while reducing the electricity bills. In order to boost real-time electricity price, this paper analyzed the power load and electricity price. A method based on the Particle Swarm Optimization with Extended Memory (PSOEM) and Support Vector Regression (SVR) was proposed for short-term power load forecast. This paper proposed the principle to establish real-time electricity price and test the feasibility by simulation.Firstly, this paper investigates the present home and abroad development situation of the smart grid, and our country's power structure with coal-based brought the environment problem and other problems. So the research on real-time electricity price based on smart grid is very important. Secondly, the composition and characteristics of power load were analyzed deeply, and the basic principles of power load forecasting and electricity prices in power market were expatiated in this paper. To forecast short-term power load accurately,quickly and efficiently, a method based on the Particle Swarm Optimization with Extended Memory (PSOEM) and Support Vector Regression (SVR) was proposed for short-term power load forecast, taking the historical load data as model inputs. PSOEM had more extensive capability of global optimization than PSO owing to higher accuracy and convergence rate. In order to reduce blindness and inefficiency, PSOEM was used to optimize the parameters of the SVR with compounding kernels, and obtained an optimum PSOEM-SVR model to forecast the load. Use the historical load data some city of Sichuan province to test the performance in accuracy and stability through simulation. This paper analyzed the users'response to electricity price under the conditions of the electricity market. At last, this paper proposed extend real-time electricity price based on the grid load rate, and test the validity by simulation.