电力系统稳定器参数优化的改进算法

在多机系统多运行方式下,用非线性规划方法进行电力系统稳定器(PSS)参数优化时,特征根的稳定程度会随着参数的调整而不断变化,以致收敛过程中存在振荡现象.为减缓振荡,对非线性规划目标函数进行了改进,增加了优化过程中所计及的特征根数量,并根据各特征根的稳定程度选择相应的权重系数,大幅度减少了优化计算的迭代次数.在一个8机系统的算例上进行试算的结果表明,改进后的算法可有效改善迭代过程的收敛特性.     

基于小生境优化的变权重粒子群算法在PMU优化配置中的应用

针对相量测量单元优化配置问题展开研究,提出了一种基于小生境优化的变权重粒子群算法来解决该问题.所提算法以离散粒子群算法为基础,通过基于共事函数的小生境技术优化初始粒子群,改进权重系数的取值函数来提高算法的收敛效果,最后进行冗余度比较得出最优方案.该算法与常用算法相比,提高了计算的收敛速度和全局性,实现了算法多峰性,可通过分析问题模型有效得出相量测量单元(PMU)优化配置方案.并通过IEEE 14母线系统和新英格兰39母线系统仿真实例验证了所提方法的有效性.     

励磁附加调差系数对电力系统小干扰稳定的影响分析及其参数优化

多机系统中,发电机附加调差系数的变化不仅会影响电力系统稳态特性,还会影响系统的动态特性。在分析附加调差环节对电力系统稳态特性影响的基础上,以发电机附加调差系数为控制量,并以系统机电振荡模式的阻尼比为约束条件,用系统机电振荡模式的特征值构建目标函数,建立多机系统附加调差系数的优化模型,利用非线性规划方法求解优化问题。从而协调多机系统中的附加调差系数,提供最大阻尼。算例分析表明,利用所述方法,可保证多机系统中附加调差系数同时满足稳态和小干扰稳定的要求。     

遗传算法在有功安全经济调度中的应用

利用遗传算法(简称GA)求解电力系统有功经济调度问题。优化模型中考虑了线路N安全性约 束，对遗传算法中适合度函数、变异概率取值及收敛判据等方面进行了探讨，进一步拓展了 电力系统安全经济调度计算方法的应用前景。最后对IEEE　30节点6机系统进行了数字仿真 ，证明了该方法的简捷和有效。     

电力系统环保经济负荷分配的模糊自修正粒子群算法

针对标准粒子群算法易陷入局部最优、收敛过早的缺陷,提出了一种模糊自修正粒子群算法。通过利用模糊推理机制建立了粒子适应度值隶属度函数,在每次寻优过程中,使得各粒子根据自身当前适应度隶属度函数值来修正惯性权重的取值,而不是把惯性权重作为全局变量,对同一代粒子使用相同的惯性权重;这充分考虑了各粒子自身的性能,可以进一步改善早熟的缺陷,增强全局搜索能力,从而可以获取更好的目标值。将该算法用于求解电力系统经济负荷分配问题,兼顾考虑了燃料成本和环境成本;在求解此问题时,为了更精确地处理功率平衡约束,根据寻优过程中等式约束偏差量的大小不断调整罚系数取值,并以此建立相应的罚函数。算例结果表明,模糊自修正粒子群算法对比标准粒子群算法有较强的全局搜索能力,有更可靠的优化计算结果,进而体现了该方法的有效性和优越性。     

基于改进遗传算法的多目标无功优化

阐述了用于无功优化的改进遗传算法,在已有改进简单遗传算法的基础上,提出在含有多个目标的目标函数中采用线性变化和指数变化规律的越界罚系数,并对适应度函数进行模拟退火修正以保持种群的多样性和加快收敛;采用遗传因子自适应变化和改进的变异操作,可使遗传算法的全局优化和局部寻优能力大为提高.IEEE14节点系统的仿真计算结果表明,该方法在计算速度和收敛能力上优于简单遗传算法,且罚系数采用指数规律变化比采用定值或线性变化规律时收敛能力有明显改善.     

计及负荷动态特性的电压稳定概率特征根分析

概率特征根分析可以将系统的多种运行方式融入到一次数值计算中,但是现有的概率特征根分析只考虑了负荷的电压指数依赖特性模型.负荷的动态特性是造成电压不稳定的主要原因,为此基于一般性的多机系统表达技术(GMR),将概率特征根分析进行扩展以包含负荷的动态特性模型,进而根据特征根的概率分布、动态负荷相关比以及特征根对节点电压的电压失稳模式系数确定系统的电压薄弱点,从而完成了系统多运行方式下的小干扰电压稳定性分析.最后,对五节点系统进行了计算,考察了在所确定的电压薄弱点进行电压控制的效果.     

快速求解多机系统机电振荡模式的新算法

提出了一种快速求解多机系统机电振荡模式的新方法,既能避免维数灾问题又能防止出现丢根现象。首先,通过建立发电机的经典模型,计算出多机系统的各振荡模式(特征根虚部)和振荡模态(特征向量);再用等值阻尼系统表示发电机的励磁控制作用对机电振荡模式的影响,推导出只用机组惯性常数M和阻尼系数D及相应的特征向量表示的机电模式的实部估算公式,由此公式估算出相应的实部,从而得到各振荡模式的特征根的初值;根据Heffron-Phillips全系统的状态矩阵和已经估算出的特征根,用反幂法快速求出系统状态矩阵的根,即为此多机系统的全部机电模式特征根。最后,用ENGLAND10机系统算例证明了该方法的有效性。     

基于改进SMA法求解多机系统机电模式新方法

利用改进SMA法求解多机系统机电振荡模式特征根可以避免维数灾问题,但是改进SMA法对特征根初值有一定的要求,否则会出现丢根现象.为了解决这一问题,本文将快速估算特征根法和改进SMA法结合起来,以快速估算法计算得到的机电模式特征根作为改进SMA法的特征根初值.最后利用新英格兰10机系统算例证明了本方法的有效性.     

基于改进鸟群算法的微电网并网优化调度研究

以分析微电网并网优化调度中不同模型对优化的影响和解决传统智能算法在多目标函数寻优时易早熟收敛、陷入局部最优、收敛速度和精度差等问题为目的,采用二元对比定权法将以运行成本、环境污染处理费用为目标的多目标优化模型转化为单目标优化模型,并应用了一种新型生物启发式群智能算法——鸟群算法,该算法以鸟类觅食、警惕和飞行行为为依据,其性能优于粒子群和微分进化算法,因其认知和群体行为调节参数易使种群收敛精度和迭代次数偏大,通过线性微分递减策略改进惯性权重、线性调整认知系数和社会系数来改进鸟群算法,以此对不同模型进行优化。本文对不同目标函数的优化结果和两种算法的仿真结果做了对比,验证了多目标优化模型和改进算法的有效性。     

PSS and FDS simultaneous tuning

A proposition is made to design and coordinate multiple stabilizers in order to enhance the electromechanical transient behavior of power systems. A technique for tuning FACTS-based stabilizers as an optimization problem is presented. The stabilizers’ parameters are obtained through the minimization of an objective function based on second order eigen-sensitivities, and set up by some dominant eigenvalues and their left and right-eigenvectors calculated for different operating conditions. The effectiveness and robustness of the procedure are demonstrated through digital simulations on a 16-machine power system embedding an STATCOM.     

Substructure Preservation Sylvester-based Model Order Reduction with Application to Power Systems

A new substructure preservation Sylvester-based model order reduction technique with application to power systems is presented in this article. The new approach is intended for multiple-input–multiple-output linear time invariant systems, given in the form of state-space realization with the objective of obtaining a proper reduced-order model (complexity reduction), preserving the dominant eigenvalues of the full-order model as a subset in the reduced model, and maintaining a minimum steady-state error. The proposed reduction method is performed based on transforming the system state matrix into a special form, taking into account the dominant eigenvalues, while the rest of the model transformation is derived utilizing the Sylvester equation formula. Once the system is transformed, the reduced-order model is obtained by truncating the less dominant eigenvalues using the singular perturbation technique. To evaluate the potential of the new approach, results of the proposed technique are compared to some of the well-known methods for model order reduction and relatively recently published work. Results comparison shows the superiority of the new method especially in terms of time convergence.     

Pole placement approach for robust optimum design of PSS and TCSC-based stabilizers using reinforcement learning automata

Power system stability enhancement via robust optimum design of power system stabilizers (PSSs) and thyristor controlled series capacitor (TCSC)-based stabilizers is thoroughly investigated in this paper. The design problem of PSS and TCSC-based stabilizers is formulated as an optimization problem where a reinforcement learning automata-based optimization algorithm is applied to search for the optimal setting of the proposed PSS and CSC parameters. A pole placement based objective function is considered to shift the dominant system eigenvalues to the left in the s-plane. For evaluation of the effectiveness and robustness of the proposed stabilizers, their performances have been examined on a weakly connected power system subjected to different disturbances, loading conditions, and system parameter variations. The nonlinear simulation results and eigenvalues analysis demonstrate the high performance of the proposed stabilizers and their ability to provide efficient damping of low frequency oscillations. In addition, it is observed that the proposed CSC has greatly improved the voltage profile of system under severe disturbances.     

Assessment of dynamic stability of large-scale power system by Rayleigh's quotient

A method of assessing dynamic stability of large-scale power system by Rayleigh's quotient is proposed. One-machine infinite-bus systems show damping torque characteristics similar to diagonal components of operational transfer matrices for original multimachine systems, which means that design of PSS with one-machine systems controls those components. An expression for damping constants of oscillation modes is derived based on an energy function and its time derivative for a simplified system representation. This expression demonstrates that oscillations do not necessarily become unstable even if there are some generators with negative damping; and the effect of damping torque is determined by eigenvectors. The expression is generalized with Rayleigh's quotient, and a method of estimating eigenvalues of large-scale power systems is proposed. With this method, approximate eigenvalues are refined to accurate eigenvalues. Only a specified number of eigenvalue analyses are required irrespective of the number of generators, hence much calculation is saved. Finally, this method is applied to a 107-machine system to verify its effectiveness.     

Optimal Power System Stabilizers design via Cuckoo Search algorithm

Cuckoo Search (CS) algorithm is introduced in this paper for optimal Power System Stabilizers (PSSs) design in a multimachine power system. The PSSs parameter tuning problem is formulated as an optimization problem which is solved by CS Algorithm. An eigenvalues based objective function involving the damping ratio, and the damping factor of the lightly damped electromechanical modes is considered for the PSSs design problem. The performance of the proposed CS based PSSs (CSPSS) has been compared with Genetic Algorithm (GA) based PSSs (GAPSS) and the Conventional PSSs (CPSS) under different operating conditions and disturbances. The results of the developed CSPSS are verified through time domain analysis, eigenvalues and performance indices. Also, the effectiveness of the proposed algorithm in providing good damping characteristics is confirmed.     

异步发电机自激建压临界值的计算

以自激异步发电机ab轴等值电路为基础，结合系统稳定性理论说明了异步发电机自激建压的条件。对系统的特征值进行分类，定义特征值实部函数，并对不同自激电容和转速下的函数值进行计算。以特征值实部函数绝对值作为目标函数，采用一维搜索优化方法对样机的建压临界电容和临界转速进行计算。新方法与传统方法计算结果一致，验证了新方法的正确性。     

Coordinated design of probabilistic PSS and SVC damping controllers

This paper presents an application of probabilistic theory to the coordinated design of power system stabilizers (PSSs) and FACTS controllers, taking static VAr system (SVC) as an example. The aim is to enhance the damping of multi electro-mechanical modes in a multimachine system over a large and pre-specified set of operating conditions. In this work, conventional eigenvalue analysis is extended to the probabilistic environment in which the statistical nature of eigenvalues corresponding to different operating conditions is described by their expectations and variances. Probabilistic sensitivity indices (PSIs) are used for robust damping controller site selection and for optimization objective functions. A probabilistic eigenvalue-based objective function is employed for coordinated design of PSS and SVC controller parameters. The effectiveness of the proposed controllers is demonstrated on an 8-machine system.     

基于粒子群优化算法的VSC-HVDC系统的控制参数优化策略

建立了双端基于电压源型换流器的高压直流输电(VSC-HVDC)系统的小信号模型;利用特征根分析法,基于线性化状态空间模型,将与系统特征根直接相关的振荡模式和衰减模式的罚函数作为目标函数;提出了基于粒子群优化算法的VSC-HVDC系统的控制参数优化策略,对整个系统的控制参数同时进行整体优化。仿真结果验证了小信号模型的正确性;优化后的系统在小扰动、大扰动、潮流反转及故障情况下均具有较高的控制精度,整个系统的稳态与暂态特性均得到较大改善。     

A simple and fast algorithm for determining asymptotic stability of linear autonomous systems

The method introduced here determines explicitly whether a given system has only eigenvalues with negative real parts, without determining these eigenvalues, but rather by determining the eigenvalues of a related symmetric matrix which are always real and hence easy to calculate. The new method is similarly applied to linear autonomous discrete systems where all the eigenvalues of the system matrix are required to be positioned within the unit circle in the z-plane.     

A practical closed form solution for systems with variable eigenvalues via bilinear system theory

This paper treats systems in state variable formulation with non-constant, parameter controlled system matrices. The synthesis of a system with controlled eigenvalues (ECS) is given. The synthesized system is a commutative bilinear system. Its solution has a closed form and is based on the solution of just one time invariant system although as many arbitrary time functions are involved as the system has independent states. The ECS is homologous to any system with a system matrix being an arbitrary, possibly time-dependent function of a single constant system matrix. All results are deduced for multiple eigenvalues of the system matrix including single eigenvalues as a special case. They are fully analogous to the solution of time invariant systems by means of the Laplace transformation.