摘要
在智能电网实时定价研究中,为了提高用户偏好刻画的精度,在二次效用函数的基础上提出了一种与用户实际情况更贴切的分段效用函数。以所有用户的效用之和减去电力供应商成本后的结果作为目标函数,发电容量为约束构造分段实时定价的系统模型。针对模型中分段效用函数的整体不光滑性引入一种光滑化方法,可以将分段光滑的函数全局近似光滑化以优化模型求解的有效性。光滑后的效用函数代入到系统模型中,可直接使用拉格朗日乘子法对模型进行求解。仿真结果验证了分段实时定价策略的有效性和合理性。
Modeling users＇ preferences plays an important role in real - time pricing for smart grid. In order to improve the accuracy of the preferences model, in this paper, we proposed a piecewise utility function on the basis of quadratic utility function, which is more appropriate to the users＇ reality. Adopted the sum of all utility functions mi- nus the cost imposed to the energy provider as the objective function, while the consumption levels of all users were coupled via the limited available generation capacity to structure the piecewise real - time pricing model. At the same time, we introduced a smoothing method for the piecewise smooth function. This method can construct a new smooth function to approximate the original function to improve the validity of solution. Substituted the new smooth utility function into the system model, it can directly use the Lagrange multiplier method to solve the model. Finally, simulation results verify the validity and rationality of the piecewise real - time pricing strategy.
出处
《计算机仿真》
CSCD
北大核心
2016年第4期171-175,205,共6页
Computer Simulation
基金
国家自然科学基金项目(11171221)
美国IBM公司共享大学项目(SUR)
关键词
智能电网
实时定价
效用函数
光滑化方法
拉格朗日乘子
Smart gird
Real - time pricing
Utility function
Smoothing method
Lagrange multiplier
