文章摘要
薛花,胡英俊,董丙伟,等.基于分层控制的光—储—燃直流供电系统能量管理方法[J].电力系统自动化,2018,42(21):53-62. DOI: 10.7500/AEPS20170909006.
XUE Hua,HU Yingjun,DONG Bingwei, et al.Hierarchical Control Based Energy Management Method for Photovoltaic-Energy Storage- Fuel Cell DC Generation System[J].Automation of Electric Power Systems,2018,42(21):53-62. DOI: 10.7500/AEPS20170909006.
基于分层控制的光—储—燃直流供电系统能量管理方法
Hierarchical Control Based Energy Management Method for Photovoltaic-Energy Storage- Fuel Cell DC Generation System
DOI:10.7500/AEPS20170909006
关键词: 混合储能  模型预测  微分平滑  光伏
KeyWords: hybrid energy storage  model prediction  differential flatness  photovoltaic
上网日期:2018-09-14
基金项目:国家自然科学基金资助项目(51407114);上海市自然科学基金资助项目(15ZR1418200);上海市科技创新行动计划项目(16DZ0503300)
作者单位E-mail
薛花 上海电力学院电气工程学院, 上海市 200090 xuehua@shiep.edu.cn 
胡英俊 上海电力学院电气工程学院, 上海市 200090  
董丙伟 上海电力学院电气工程学院, 上海市 200090  
王育飞 上海电力学院电气工程学院, 上海市 200090  
摘要:
      为克服光伏发电的随机性和间歇性,实现高可靠性、高品质供电,提出一种基于光伏/蓄电池/超级电容/燃料电池的直流供电系统能量管理分层控制方法。针对混合直流供电系统的多参数、非线性特性,上层控制设计微分平滑方法直接补偿非线性分量,实现系统非线性动态可逆,确保在光伏出力大幅变化、负载突变或系统参数发生摄动情形下,使系统直流母线电压依然能够稳定运行,同时为下层控制提供总载荷电流参考轨迹;针对混合直流供电系统中多电源的不同特性,下层控制采用模型预测方法,以延长系统循环使用寿命为目标,设计加权因子,根据光伏出力和负载功率需求状况,动态调节储能单元充放电速率,使系统快速跟踪总载荷电流期望轨迹,实现各电源间功率的合理分配。基于MATLAB/Simulink的仿真结果验证了所提能量管理分层控制方法的可行性和有效性,该方法具有结构简单、动态响应快、鲁棒性高、稳定性好的特点。
Abstract:
      In order to overcome the random and intermittent of photovoltaic(PV)power generation, realize the high reliability and quality of power supply, a hierarchical control method based on photovoltaic/storage battery/super capacitor/fuel cell(FC)DC generation system is proposed. For the multi-parameter and non-linear characteristic of hybrid DC power supply system, the control of upper layer designs differential flatness method to compensate non-linear components directly, which realizes the non-linear dynamic reversibility of system, makes the DC bus voltage still stable under situations of great changes of PV output, loads and system parameters. Meanwhile, the upper layer provides reference track for total load current to lower layer. Aiming at different characteristics of multiple power sources in the hybrid DC power supply system, the control of lower layer utilizes model prediction method to design weighting factors to extend cycle life of the system. According to the conditions of photovoltaic output and load demand, the charging and discharging rate of storage unit is dynamically adjusted, which makes the system immediately track the current expectation of total loads to achieve the reasonable distribution of power in various sources. Based on the MATLAB/Simulink, the simulation results show the feasibility and effectiveness of proposed method, which has simple structure, rapid response, high robustness and good stability.
查看全文(Free!)   查看附录   查看/发表评论  下载PDF阅读器