光伏发电是实现双碳目标的坚实保障，随着碳达峰的不断推进，光伏发电产业步入快车道。本文以双级式并网光伏发电系统为研究对象，研究最大功率点跟踪（Maximum Power Point Tracking，MPPT）算法以实现前级最大功率输出，研究并网逆变器控制技术以实现后级母线电压的稳定和并网电流的高质量入网。

        针对局部遮阴情况下引起光伏阵列多峰值的P-U特性，而传统MPPT方法极易陷入局部最优这一问题，研究基于改进蝴蝶优化算法（Improved Butterfly Optimization Algorithm，IBOA）的MPPT方法和基于混沌变异黏菌算法结合算术算法（Chaos Variable Slime Mould Algorithm Combined With Arithmetic Optimization Algorithm，CVSMA-AOA）的MPPT方法。针对BOA收敛速度慢、追踪过程震荡幅度大和稳态精度差等问题，利用BOA调节参数少这一特性，在未引入额外调节参数的情况下改进BOA的整体性能。采用自适应切换概率平衡BOA搜索的整个过程；采用平衡权重因子增强BOA前期全局寻优效果；采用精英引导变异降低BOA陷入局部最优的概率。仿真实验表明基于IBOA的MPPT方法可以实现光伏最大功率发电，相比BOA，追踪稳态时间大幅缩减。

        针对SMA收敛速度慢、稳态精度欠缺问题，利用SMA多重搜索机制，混合多种优化策略进一步提高SMA对目标函数的寻优性能，引入高斯变异提高算子跃变范围和收敛速度；引入算术算法中低分散性的加减粒子提高局部开发能力；引入混沌初始化得到位置更为均匀的种群。据各案例环境下仿真实验分析，基于CVSMA-AOA的MPPT方法具有较强的收敛趋势，均稳定在0.1s即可锁定MPP位置，且具有较高的静态MPPT效率。

        针对单相LCL型并网逆变器，控制电路中电压外环采用改进PI控制，电流内环应用多谐振准PR控制，同时电流环中引入电容电流反馈实现有源阻尼。锁相环采用改进SOGI-PLL，降低对电网电压中直流分量的敏感度，提高并网系统的稳定性。搭建并网光伏发电仿真模型，据仿真分析，电流环采用多谐振准PR控制，相比传统PI控制，并网电流谐波含量降低0.19%。通过搭建逆变实验平台，验证了电流控制方法的正确性。

        Photovoltaic power generation is a solid guarantee for achieving the dual-carbon goal. With the continuous advancement of carbon peak, the photovoltaic power generation industry has entered the fast lane. This paper takes the two-stage grid-connected photovoltaic power generation system as the research object, studies the Maximum Power Point Tracking ( MPPT ) algorithm to achieve the maximum power output of the front stage, and studies the grid-connected inverter control technology to achieve the stability of the rear bus voltage and the high-quality grid-connected current.

        Aiming at the multi-peak P-U characteristics of photovoltaic array caused by partial shading, the traditional MPPT method is easy to fall into the local optimum. A MPPT method based on Improved Butterfly Optimization Algorithm ( IBOA ) and a photovoltaic MPPT method based on Chaos Variable Slime Mould Algorithm Combined With Arithmetic Optimization Algorithm ( CVSMA-AOA ) are proposed. Aiming at the problems of slow convergence speed, large oscillation amplitude and poor steady-state accuracy of BOA, the overall performance of BOA is improved without introducing additional adjustment parameters by using the characteristic of less adjustment parameters of BOA. The adaptive switching probability is used to balance the whole process of BOA search. The balanced weight factor is used to enhance the global optimization effect of BOA in the early stage. The elite guided mutation is used to reduce the probability of BOA falling into local optimum. The simulation results show that the MPPT method based on IBOA can realize the maximum power generation of photovoltaic, and the tracking steady-state time is greatly reduced compared with BOA.

        Aiming at the problems of slow convergence speed and lack of steady-state accuracy of SMA, the multi-search mechanism of SMA is used, and a variety of optimization strategies are mixed to further improve the optimization performance of SMA on the objective function. Gaussian mutation is introduced to improve the jump range and convergence speed of the operator. The addition and subtraction particles with low dispersion in the arithmetic algorithm are introduced to improve the local development ability. Chaos initialization is introduced to obtain a more uniform population. According to the analysis of simulation experiments in each case environment, the MPPT method based on CVSMA-AOA has a strong convergence trend, which is stable at 0.1 s to lock the MPP position, and has high static MPPT efficiency.

        For the single-phase LCL-type grid-connected inverter, the improved PI control is used in the voltage outer loop of the control circuit, the multi-resonant quasi-PR control is used in the current inner loop, and the capacitive current feedback is introduced in the current loop to achieve active damping. The phase-locked loop uses an improved SOGI-PLL to reduce the sensitivity to the DC component in the grid voltage and improve the stability of the grid-connected system. The simulation model of grid-connected photovoltaic power generation is built. According to the simulation analysis, the current loop adopts multi-resonant quasi-PR control. Compared with the traditional PI control, the harmonic content of grid-connected current is reduced by 0.19 %. The correctness of the current control method is verified by building an inverter experimental platform.

［1］黄震, 谢晓敏, 张庭婷. “双碳”背景下我国中长期能源需求预测与转型路径研究[J]. 中国工程科学, 2022, 24(06): 8-18.

［2］Sai Krishna G, Moger T. Reconfiguration Strategies for Reducing Partial Shading Effects in Photovoltaic Arrays: State of the Art[J]. Solar Energy, 2019, 182: 429-452.

［3］黄晟, 王静宇, 郭沛, 等. 碳中和目标下能源结构优化的近期策略与远期展望[J]. 化工进展, 2022, 41(11): 5695-5708.

［4］周宏春. 中国低碳经济发展现状及展望[J]. 科技导报, 2022, 40(21): 6-12.

［5］邹才能, 陈艳鹏, 熊波, 等. 碳中和目标下中国新能源使命[J]. 中国科学院院刊2023, 38(01): 48-58.

［6］Martin A D, Vazquez J R, Cano J M. MPPT in PV Systems under Partial Shading Conditions using Artificial Vision[J]. Electric Power Systems Research, 2018, 162: 89-98.

［7］张东宁. 基于改进电导增量法的光伏最大功率点跟踪策略研究[J]. 太阳能学报, 2022, 43(08): 82-90.

［8］Kumar V, Singh M. Derated Mode of Power Generation in PV System Using Modified Perturb and Observe MPPT Algorithm[J]. Journal of Modern Power Systems and Clean Energy, 2021, 9(05): 1183-1192.

［9］罗茜, 陈卓, 郝正航, 等. 基于新型趋近律的光伏MPPT控制策略[J]. 电力系统保护与控制, 2023, 51(05): 139-153.

［10］葛传九, 武鹏, 董祥祥, 等.基于布谷鸟算法的光伏MPPT改进[J]. 太阳能学报, 2022, 43(10): 59-64.

［11］Nickabadi A, Ebadzadeh M M,Safabakhsh R. A Novel Particle Swarm Optimization Algorithm with Adaptive Inertia Weight[J]. Applied soft computing, 2011, 11(4): 3658-3670.

［12］Mirjalili S. Dragonfly Algorithm:A Eew Meta-heuristic Optimization Technique for Solving Single-objective,Discrete, and Multi-objective Problems[J]. Neural Computing and Applications, 2016, 27(4): 1053-1073.

［13］Mirjalili S, Lewis A. The Whale Optimization Algorithm[J]. Advances in engineering software, 2016, 95: 51-67.

［14］李世光, 夏杰, 李雪杨, 等. 改进APSO算法在光伏MPPT控制中的应用[J]. 电测与仪表, 2019, 56(22): 19-24.

［15］孙恺, 刘光宇. 基于改进PSO算法的光伏发电系统MPPT控制策略[J]. 传感器与微系统, 2021, 40(01): 49-52.

［16］商立群, 李帆. 基于自适应布谷鸟搜索和扰动观察法的光伏最大功率点跟踪[J]. 电力系统保护与控制, 2022, 50(08): 99-107.

［17］陈斌, 王俊江, 赵明胤, 等. 基于改进鲸鱼优化算法的光伏发电系统MPPT控制研究[J]. 电力系统及其自动化学报, 2023, 35(02): 19-26.

［18］张晓强, 刘宜罡, 邹应全, 等. 基于自适应神经网络控制的光伏MPPT算法改进[J]. 太阳能学报, 2019, 40(11): 3095-3102.

［19］Arora S, Singh S. Butterfly Optimization Algorithm: A Novel Approach for Global Optimization[J]. Soft Computing, 2019, 23(3): 715-734.

［20］Sankalap Aroraa, Satvir Singh. An Improved Butterfly Optimization Algorithm with Chaos[J]. Journal of Intelligent & Fuzzy Systems, 2017, 32(1): 1079-1088.

［21］Arora S, Singh S. An Improved Butterfly Optimization Algorithm for Global Optimization[J]. Advanced Science, Engineering and Medicine, 2016, 8(9): 711-717.

［22］Shi Zijian, Wang Xueqing, Palomares I, et al. A Novel Consensus Model for Multi-attribute Large-scale Group Decision Making based on Comprehensive Behavior Classification and Adaptive Weight Updating[J]. Knowledge-Based Systems, 2018, 158: 96-208.

［23］郑旭彬, 李梦达, 梁智超, 等. 局部遮荫下改进蝴蝶算法的MPPT控制仿真研究[J]. 计算机仿真, 2022, 39(01): 96-101.

［24］刘凯, 代永强. 融合变异策略的自适应蝴蝶优化算法[J]. 计算机应用研究, 2022, 39(01): 134-140+145.

［25］Li S, Chen H, Wang M, et al. Slime Mould Algorithm: A New Method for Stochastic Optimization[J]. Future Generation Computer Systems. 2020, 111(1): 300-323.

［26］贾鹤鸣, 刘宇翔, 刘庆鑫, 等. 融合随机反向学习的黏菌与算术混合优化算法[J].计算机科学与探索, 2022, 16(05): 1182-1192.

［27］刘宇凇, 刘升. 无迹西格玛点引导的拟反向黏菌算法及其工程应用[J]. 计算机应用研究, 2022, 39(09): 2709-2716.

［28］肖亚宁, 孙雪, 李三平, 等. 基于混沌精英黏菌算法的无刷直流电机转速控制[J].科学技术与工程, 2021, 21(28): 12130-12138.

［29］董密, 胡佳盛, 杨建, 等. 基于改进黏菌优化算法的光伏多峰MPPT控制策略[J/OL]. 控制理论与应用: 1-9[2022-11-28].

［30］毛无穷. 单相光伏逆变并网及孤岛检测策略研究[D]. 北京:华北电力大学(北京), 2016.

［31］杨文燕. 单相并网式微型光伏逆变器的设计[D]. 北京:北京交通大学, 2018.

［32］A.Hasanzadeh, O.C.Onar, H.Mokhtari, et al. A Proportional-resonant Controller-based Wireless Control Strategy with A Reduced Number of Sensors for Parallel-operated UPSs[J]. IEEE Transactions on Power Electronics, 2010,25(1):468-478.

［33］梁文勇, 吴大伟, 汪滔, 等. 两相静止坐标系下并网逆变器改进重复控制方案[J]. 电力电子技术, 2020, 54(06): 112-116.

［34］周振雄, 石硕. 基于DRNN自整定准PR控制的光伏并网系统研究[J]. 太阳能学报, 2017, 38(11): 2932-2940.

［35］梁理程, 丁欣, 谢李兵, 等. 基于状态观测器的单相并网逆变器复合控制[J/OL]. 电源学报: 1-12[2023-01-03].

［36］周媛媛. 光伏发电系统中最大功率点跟踪控制算法研究[D]. 西安:西安理工大学, 2021.

［37］刘春娟, 郑丽君, 孙赟赟, 等. 基于改进型细菌觅食算法的MPPT[J]. 太阳能学报, 2021, 42(09): 83-89.

［38］王青云. 光伏微网失配分析与智能MPPT控制[D]. 杭州: 杭州电子科技大学, 2022.

［39］赵斌, 袁清, 王力, 等. 基于改进蚁狮算法的光伏多峰值MPPT控制[J]. 太阳能学报, 2021, 42(09): 132-139.

［40］李彦苍, 卜英乔, 朱海涛, 等. 融合最优邻域扰动和反向学习策略的蝴蝶优化算法[J]. 中国科技论文, 2021, 16(11): 1181-1188.

［41］郑洪清, 冯文健, 周永权. 融合正弦余弦算法的蝴蝶优化算法[J]. 广西科学, 2021, 28(02): 152-159.

［42］高鹏, 李佳田, 杨汝春, 等. 遥感图像分割的低维纹理特征算子与双变异蝴蝶优化算法[J]. 武汉大学学报(信息科学版), 2023, 48(01): 165-174.

［43］刘景森, 马义想, 李煜. 改进蝴蝶算法求解多维复杂函数优化问题[J]. 电子学报, 2021, 49(06): 1068-1076.

［44］Abualigah L, Diabat A, Mirjalili S, et al. The Arithmetic Optimization Algorithm[J]. Computer Methodsin Applied Mechanics and Engineering, 2021, 376: 113609.

［45］刘庆利, 商佳乐, 曹娜, 等. 基于改进MVO算法的多无人机任务分配研究[J/OL]. 控制工程: 1-8[2022-11-28].

［46］贾鹤鸣, 姜子超, 彭晓旭, 等. 基于改进鬣狗优化算法的多阈值彩色图像分割[J]. 计算机应用与软件, 2020, 37(05): 261-267.

［47］魏昕, 冯锋. 基于高斯-柯西变异的帝国竞争算法优化[J]. 计算机科学, 2021, 48(S2): 142-146.

［48］黄华娟, 程前, 韦修喜, 等. 融合Jaya高斯变异的自适应乌鸦搜索算法[J/OL]. 山东大学学报(工学版): 1-12[2022-11-28].

［49］崔文峰, 胡森军, 李武华, 等. 基于有源钳位的无变压器型单相光伏并网逆变器[J]. 电工技术学报, 2015, 30(16): 26－32.

［50］雒向东, 崔剑波. 光伏并网逆变器LCL滤波器参数优化设计[J]. 沈阳工业大学学报, 2016, 38(06): 601-605.

［51］张兴. 太阳能光伏并网发电及其逆变控制[M]北京. 机械工业出版社, 2018: 174-175.

［52］梁理程, 丁欣,谢李兵, 等. 基于状态观测器的单相并网逆变器复合控制[J]. 电源学报.

［53］王志浩. 光伏并网逆变器关键技术研究[D]. 恩施: 湖北民族大学, 2022.

［54］肖蕙蕙, 向文凯, 郭强, 等. 三相Vienna整流器自抗扰准PR控制[J]. 重庆理工大学学报(自然科学), 2022, 36(06): 224-230.

［55］鲁改凤, 杜帅, 欧钰雷, 等. 自适应谐波消除的准PR并网逆变器优化控制[J]. 电测与仪表, 2019, 56(19): 59-63.

［56］范宝奇, 罗晓曙, 廖志贤, 等. 神经网络准PR光伏并网逆变器控制技术[J]. 电力系统及其自动化学报, 2016, 28(03): 30-34.

［57］冉立泽. 单相并网逆变器控制策略研究[D]. 哈尔滨:哈尔滨工业大学, 2020.

［58］陶慧, 赵世彬, 贾春华. 准PR调节下单相光伏并网逆变器的非线性研究[J]. 太阳能学报, 2022, 43(10): 21-28.