高比例新能源电网短期日负荷数据扰动控制算法

A Disturbance Control Algorithm for Short-Term Daily Load Data of High-Penetration New Energy Power Grids

中文关键词:高比例新能源电网  短期日负荷数据  分布式模型预测控制器  天牛须算法  数据扰动控制

keywords:high-penetration new energy power grid  short-term daily load data  DMPC  beetle antennae search algorithm  data disturbance control

中文摘要:为降低计算复杂度，设计一种高比例新能源电网短期日负荷数据扰动控制算法。新能源发电系统被视为独立区域，实施分布式控制。依据模型预测控制思想，设计了分布式模型预测控制器（distributed model predictive control，DMPC），并对子区域系统的线性时不变状态空间方程进行离散化处理。通过设定预测和控制时域，利用状态、控制和扰动矩阵等参数，预测未来时刻的状态变量。以最小化输出电力负荷波动为目标，构建目标函数；采用天牛须算法改进基本粒子群算法，并求解目标函数，获取最佳扰动控制参数。实验测试结果表明：对于实验电网中的水利、风力、光伏、火电4个区域，设计方法的超调量分别低于3.0 kW、2.8 kW、1.6 kW、1.8 kW，整体调节时间低于182 s。

Abstract:To reduce computational complexity， this paper proposes a disturbance control algorithm for short-term daily load data in high-penetration new energy power grids. New energy generation systems are regarded as independent regions， with distributed control implemented. Based on the idea of model predictive control， a distributed model predictive controller （DMPC） is designed， and the linear time-invariant state-space equations of the sub-region systems are discretized. By defining the prediction and control horizons and utilizing parameters including state， control， and disturbance matrices， the future state variables are predicted. To minimize the fluctuation of the output power load， an objective function is established. The basic particle swarm optimization algorithm is improved using the beetle antennae search algorithm， which is then employed to solve the objective function and obtain the optimal disturbance control parameters. Test results show that for the four regions （hydropower， wind power， photovoltaic power， and thermal power） in the experimental power grid， the overshoots of the proposed method are less than 3.0 kW， 2.8 kW， 1.6 kW， and 1.8 kW， respectively， and the overall settling time is less than 182 s.