摘要
针对传统负荷模型参数辨识方法计算过程繁琐,计算时间长的难点,提出了一种基于多项式逼近的负荷模型参数辨识方法。通过多项式逼近的配点法构造能显式描述负荷计算功率与负荷模型参数间关系的多项式逼近函数,将复杂的负荷模型参数辨识问题转化为简单的显式参数最优化问题,并通过非线性规划算法求得使多项式逼近功率与实测功率最接近的负荷模型参数。与传统方法相比,该方法不仅可以描述负荷功率与负荷参数间的非线性特征,分析负荷参数变化对负荷功率的影响,还可以准确辨识负荷模型参数。实际电网算例表明,该方法能有效提高负荷模型参数辨识的速度,具有较好的实用性。
To solve the difficulties of cumbersome and time-consuming calculation process in traditional measurement based method, a parameter identification method based on polynomial approximation is proposed. Firstly, a polynomial approximation function is constructed to describe explicitly the relationship between load power and load parameters by collocation method. Then the complex load parameter identification problem is transformed into a simple explicit parameter optimization problem. Finally, the load parameters minimizing the differences between the calculated power and measured power are solved through nonlinear programming algorithm. Compared with traditional load identification method, the proposed method is advantageous of not only describing the nonlinear relationship between load power and load parameters and analyzing the influence of parameters on load power, but also identifying the load parameters effectively and accurately. Field test study shows that the speed of load modeling and feasible in practical application is improved by this method.
作者
沈一鸣
陈宇杰
郭成
孟贤
吴浩
鞠平
SHEN Yi-ming;CHEN Yu-jie;GUO Cheng;MENG Xian;WU Hao;JU Ping(College of Electrical Engineering,Zhejiang University,Hangzhou 310027,China;Electric Power Research Institute,Yunnan Power Grid,Kunming 650217,China)
出处
《能源工程》
2022年第2期71-77,共7页
Energy Engineering
基金
国家自然科学基金资助项目(51837004)。
关键词
负荷建模
多项式逼近
参数辨识
实测数据
load modeling
polynomial approximation
parameter identification
measured data
