基于R(λ)学习的孤岛微电网智能发电控制

通过研究各种分布式电源的发电特性,搭建了含风电、光伏发电、飞轮储能、小水电、微型燃气轮机与负荷的微电网负荷频率控制(Load Frequency Control, LFC)模型,其中小水电和微型燃气轮机为调频机组。将大型互联电网中的集中式自动发电控制(Automatic Generation Control, AGC)原理引入微电网,并结合基于平均报酬模型的多步R(?)学习算法,提出了一种孤岛运行模式下基于强化学习的AGC控制器,以实现对微网的智能发电控制与频率调整。仿真试验分析表明,与PI控制、Q学习和Q(?)学习相比,所提出的R(?)控制器具有快速收敛特性和良好的动态性能以及较强的模型适应性。     

基于模糊比例?微分控制的永磁直驱风电机组频率调节

在分析系统不同等效惯量和阻尼对电网频率影响的基础上,提出了基于模糊比例-微分(proportional differential,PD)控制的含飞轮储能单元的永磁直驱风电机组频率调节控制策略,以动态调节电网等效惯量和阻尼。建立了系统小信号稳定性模型,以确定模糊PD控制器的输出论域,并通过相应仿真验证了理论分析的正确性。通过对1台含飞轮储能单元的2MW永磁直驱风电系统进行仿真研究表明,所提控制策略能辅助电网频率调节,减小电网受扰动后的频率波动,有助于增强电网的频率稳定性。     

互联电网AGC的分数阶PID控制

控制性能标准(CPS)下的互联电网自动发电控制(AGC)的控制策略已经取得一些进展.分数阶PID的较强的鲁棒性和非线性适应能力的特征已引起研究人员关注.文中研究了将分数阶PID应用于互联电网AGC控制的问题,并且总结了参数λ、μ的整定规律.仿真结果表明,互联电网AGC的分数阶PID控制系统具有鲁棒性强和非线性适应能力好的优点,并且在CPS指标考核上也比传统PID控制器强.     

基于PSO-GSA算法的含DFIG互联系统AGC优化控制研究

传统双馈感应风力发电机(DFIG)的解耦控制使其无法响应电网的频率变化。随着风电渗透率的不断提高,电网调频压力不断增大,有必要对含DFIG互联系统AGC优化控制进行研究。首先建立了将风电作为"负的负荷"的两区域AGC模型,通过引入改进的虚拟惯性控制使DFIG具有更好的频率响应的能力。同时以快速消除系统区域控制偏差和风机转速偏差为目的,采用PSO-GSA算法对控制区PID控制器和DFIG转速控制单元PI控制器参数进行优化。仿真结果表明,单个区域受负荷扰动时,风电参与调频时能提供更多的有功功率支撑以减小同步机调频出力,能有效缓解同步机调频压力。PSO-GSA算法较PSO和GSA迭代速度快且适应度值更好,基于PSO-GSA参数优化后的控制器对系统区域频率偏差、联络线功率变化和区域控制偏差信号的超调量和调节时间都有明显改善,增强了系统的稳定性。     

基于社会学习自适应细菌觅食算法的互联电网AGC最优PI/PID控制器设计

AGC控制器的参数对电网频率控制的动态性能具有重要影响。不合适的控制器参数将可能使得电网在遭遇较大的负荷扰动时失去频率稳定。针对互联电网AGC控制器参数优化整定问题,提出了一种基于社会学习自适应细菌觅食算法的最优PI/PID控制器设计方法。该方法将社会学习机制及自适应步长策略引入到标准细菌觅食算法中,通过改进细菌寻优过程中的趋化、群聚及繁殖等操作,提高算法的收敛速度及寻优精度。建立两区域互联电网AGC系统仿真模型,采用所提算法优化整定其PI/PID控制器参数。仿真结果验证了所提方法的有效性。     

基于Ziegler-Nichols优化算法的火电机组负荷频率PID控制研究

随着风电和光伏接入电网的规模不断扩大,电网负荷峰谷差逐渐增加,电网的频率控制出现超调量过大、调节速度过慢等一系列问题。本文针对互联电力系统的负荷分配以及频率控制问题,提出考虑发电机组分配系数的互联电力系统负荷频率控制方法。首先,建立单个区域包含多个火电机组的两区域电网自动发电控制（AGC）系统仿真模型,根据各发电机组的煤耗特性建立经济性目标函数,并且获得最优的出力分配系数;然后,基于联络线功率和频率偏差控制（TBC-TBC）的控制模式,采用经Ziegler-Nichols相关算法优化后的PID控制,并对优化后PID控制器的鲁棒性进行分析研究;最后,基于两区域系统仿真模型,在一定的负荷扰动工况下进行仿真验证。仿真结果表明:该分配方法能够有效节省煤耗,本文算法优化后的PID控制器在频率恢复速度方面具有一定的优势,并且鲁棒性良好。     

含风电的交直流互联电网AGC两级分层模型预测控制

随着大规模风电接入交直流互联电网,传统的自动发电控制(AGC)方法难以有效地抑制风功率波动带来的频率稳定问题。为此,提出基于两级分层模型预测的AGC策略。该两级分层控制方法在下层对多个区域电网采用分散式模型预测控制;在上层对下层分散的控制器采用动态协调控制方式。以含多电源的两区域交直流互联电网AGC模型为例,仿真结果表明:与集中式模型预测控制和分散式模型预测控制方法相比,文中所提控制策略不仅对频率和联络线功率等具有良好的控制效果,还兼具高可靠性。     

互联电网AGC控制研究进展

综述了互联电网AGC控制研究进展。归纳并分析了6个研究方向:基于经典控制器的结构与参数优化的AGC控制;基于模糊控制器及改进的AGC控制;基于CPS信息的AGC控制;多类型发电机组协调下的AGC控制;利用SMES的AGC控制;利用TCPS的AGC控制。     

基于多变量广义预测理论的互联电力系统负荷频率协调控制体系

基于多变量广义预测理论建立了互联电力系统储能与AGC的负荷频率协调控制体系。区域协调控制器作为储能与AGC的上层控制器,在系统受控自回归积分滑动平均模型的基础上利用广域测量系统的实时信息预测频率及联络线功率的动态轨迹,并在充分考虑储能与AGC自身约束,如储能容量、火电机组爬坡率、备用容量及最小稳定出力的情况下对储能与AGC进行了最优的动态协调。仿真结果表明,该控制体系不仅能够有效改善系统的稳定性,而且对负荷扰动具有较强的鲁棒性。     

在策略SARSA算法在互联电网CPS最优控制中的应用

CPS标准下的互联电网的自动发电控制（AGC）系统是一个典型的不确定随机系统,提出一种新型应用于AGC系统的随机最优控制策略——基于SARSA算法的互联电网CPS最优控制策略。SARSA算法更多地考虑控制策略动作的花费,倾向于避开进入随机系统中搜索一些危险的状态,相对于离策略（off-policy）的Q学习算法,在策略（on-policy）SARSA算法的学习试错过程中对受控系统影响更小。仿真结果表明,基于SARSA算法的CPS控制器较Q学习算法有更好的鲁棒性和适应性,能够获得更佳的CPS指标。     

Ant Colony Optimized Fuzzy Control Solution for Frequency Oscillation Suppression

This paper presents a novel approach in addressing a critical power system issue, i.e., automatic generation control (AGC) in a smart grid scenario. It proposes the design and implementation of an optimized fuzzy logic controller (FLC) for AGC of interconnected power network. There are three different sources of power generation considered in the two-area interconnected model of power system network. First area is equipped with a single reheat thermal unit and a superconducting magnetic energy storage (SMES) unit, while another area has a hydro-unit with SMES. A multi-stage optimization strategy for the optimal solution of FLC for tie-line and frequency oscillation suppression is proposed in this paper using an ant colony optimization technique. The optimization of FLC is carried out in four different stages. The first stage is the optimization of range of input and output variables; the second stage is the optimization of membership function; the third and fourth stages are the optimization for rule base and rule weight optimization, respectively. The performance of the proposed controller is also compared with another control approaches to stabilize P_tie-line and Δf oscillations; these are the Ziegler–Nichols-tuned proportional–integral–derivative (PID) controller and genetic algorithm optimized PID controller. A comprehensive analysis of the traditional techniques and proposed techniques is presented on the basis of major dynamic performance parameters, i.e., settling time and peak overshoot.     

基于AHP / GA的储能装置参与电网调频控制器参数优化

针对电力系统调频过程中火电机组响应速度慢、不适合参与短周期调频的问题,提出一种基于层次分析法(AHP)和遗传算法(GA)相结合的优化算法用于电池储能控制器参数优化,使控制器能更好地控制电池储能装置并参与调频。通过AHP确定最大偏差幅值、稳态偏差、调节时间之间的权重大小,构造出一个GA适应度函数,再由GA进行寻优计算得到最佳的控制器参数。借助MATLAB/Simulink对储能装置参与电网调频的两区域系统进行仿真。仿真结果表明,优化后的控制器可以有效地控制储能装置并辅助AGC进行调频,能够及时响应扰动,相较于传统以时间绝对偏差乘积积分(ITAE)准则作为适应度函数的参数优化效果更好。     

Load frequency control of an autonomous hybrid power system by quasi-oppositional harmony search algorithm

This paper deals with a novel quasi-oppositional harmony search algorithm (QOHSA) based design of load frequency controller for an autonomous hybrid power system model (HPSM) consisting of multiple power generating units and energy storage units. QOHSA is a novel improved version of music inspired harmony search algorithm for obtaining the best solution vectors and faster convergence rate. In this paper, the efficacy of the proposed QOHSA is adjudged for optimized load frequency control (LFC) of an autonomous HPSM. The studied HPSM consists of renewable/non-renewable energy based generating units such as wind turbine generator, solar photovoltaic, solar thermal power generator, diesel engine generator, fuel cell with aqua-electrolyzer while energy storage units consists of battery energy storage system, flywheel energy storage system and ultra-capacitor. Gains of the conventional controllers such as integral (I) controller, proportional–integral (PI) controller and proportional–integral–derivative (PID) controller (installed as frequency controller one at a time in the proposed HPSM) is optimized using QOHSA to mitigate any frequency deviation owing to sudden generation/load change. In order to corroborate the efficacy of QOHSA, performance of QOHSA to design optimal LFC is compared with that of other well-established technique such as teaching learning based optimization algorithm (TLBOA). The comparative performances of the HPSM under the action of QOHSA/TLBOA based optimized conventional controllers (I or PI or PID) are investigated and compared in the present work. It is found that the QOHSA tuned frequency controllers improves the overall dynamic response in terms of settling time, overshoot and undershoot in the profile of frequency deviation and power deviation of the studied HPSM.     

Automatic generation control of multi area thermal system using Bat algorithm optimized PD–PID cascade controller

This article presents automatic generation control (AGC) of an interconnected multi area thermal system. The control areas are provided with single reheat turbine and generation rate constraints of 3%/min. A maiden attempt has been made to apply a Proportional derivative–Proportional integral derivative (PD–PID) cascade controller in AGC. Controller gains are optimized simultaneously using more recent and powerful evolutionary computational technique Bat algorithm (BA). Performance of classical controllers such as Proportional Integral (PI) and Proportional Integral Derivative (PID) controller are investigated and compared with PD–PID cascade controller. Investigations reveal that PI, and PID provide more or less same response where as PD–PID cascade controller provides much better response than the later. Dynamic analysis has also been carried out for the controllers in presence of random load pattern, which reveals the superior performance of the PD–PID cascade controller. Sensitivity analysis reveals that the BA optimized PD–PID Cascade controller parameters obtained at nominal condition of loading, size and position of disturbance and system parameter (Inertia constant, H) are robust and need not be reset with wide changes in system loading, size, position of disturbance and system parameters. The system dynamic performances are studied with 1% step load perturbation in Area1.     

Effective oscillation damping of an interconnected multi-source power system with automatic generation control and TCSC

Stabilizing area frequency and tie-line power oscillations in interconnected power systems are main concerns that have received significant attention in automatic generation control (AGC) studies. This paper deals with modeling and simulation of thyristor controlled series capacitor (TCSC) based damping controller in coordination with AGC to damp the oscillations and thereby, improve the dynamic stability. The contribution of TCSC in tie-line power exchange is formulated analytically for small perturbation and a systematic method based on the Taylor series expansion is proposed for modeling of TCSC based damping controller. The integral gains of AGC and TCSC parameters are optimized simultaneously using an improved particle swarm optimization (IPSO) algorithm through minimizing integral of time multiplied squared error (ITSE) performance index. The performance of the proposed TCSC–AGC coordinated controller is compared with case of AGC alone. A two-area interconnected multi-source power system, including TCSC located in series with the tie-line, is studied considering nonlinearity effects of generation rate constraint (GRC) and governor dead band (GDB). Simulation results show that proposed controller shows greater performance in damping of the oscillations and enhancing the frequency stability. Furthermore, sensitivity analyses are carried out against system loading condition, parametric uncertainties, and different perturbation patterns to show the robustness of TCSC–AGC.     

基于鲁棒方差优化的电力系统频率控制研究

随着新能源发电大规模并网,随机负荷扰动给电力系统稳定优化运行提出了新的挑战。针对自动发电控制过程中存在的随机扰动和参数摄动的问题,提出了一种基于鲁棒方差约束的状态反馈控制器的参数优化方法。根据鲁棒方差控制(Robust Variance Control,RVC)中不等式约束条件,分析了闭环系统在抑制随机扰动和提高阶跃扰动响应动态性能之间的博弈关系。构造了融合稳态状态方差和控制能量输出约束的优化问题,利用线性矩阵不等式(Linear Matrix Inequality,LMI)获得了鲁棒方差控制器参数。在此基础上,对配置的区域极点圆心坐标值和正标量两个参数进行遗传优化,得到性能指标最优的控制策略。以两区域电力系统模型为例,表明该方法能够有效抑制随机扰动并保持良好的控制性能和鲁棒性能。     

基于MVO算法与改进目标函数的电力系统负荷频率控制

针对风电并网时的随机波动功率、负荷频率控制(load frequency control, LFC)系统参数变化所引起的电力系统频率稳定问题,提出了一种基于智能优化算法与改进目标函数的互联电网LFC系统最优PID控制器设计方法。首先,分析了基于PID控制的含风电互联电力系统LFC闭环模型。其次,在时间乘误差绝对值积分(integral of time multiplied absolute error, ITAE)性能指标的目标函数中考虑了区域控制器的输出信号偏差,对优化目标函数进行改进。采用性能优良的多元宇宙优化(multi-verse optimizer, MVO)算法先计算后验证的思路,寻优获得最优PID控制器参数。最后,以两区域4机组互联电力LFC系统为例,仿真验证了基于MVO算法结合改进目标函数所获得的PID控制器,比基于MVO算法所获得的PID控制器,对阶跃负荷扰动、随机负荷扰动、风电功率偏差扰动以及系统的参数变化,具有相对较好的鲁棒性能。并且,对控制器参数也具有相对较好的非脆弱性指标。     

含抽水蓄能电站的互联电网负荷频率自抗扰优化控制研究

针对发电能源结构的多元化发展给互联电网负荷频率的稳定性控制带来较大的挑战,建立含抽水蓄能电站的两区域互联电网多元混合发电的负荷频率控制模型,提出一种基于粒子群优化算法的负荷频率线性自抗扰控制器参数整定优化策略,通过粒子群算法的迭代寻优计算获得最优的线性自抗扰控制器参数。考虑互联电网各区域发生不同的负荷扰动,在抽水蓄能电站的抽水和发电2种工况下,对所提出的控制方法进行系统仿真。仿真结果表明,通过粒子群算法优化的负荷频率线性自抗扰控制器,与传统PI控制器对比,前者具有更强的抗扰动能力和适应性,系统动态稳定性更好。