基于指数分布优化器的混合光伏-温差系统最大功率点跟踪

Hybrid PV-TEG system maximum power point tracking based on an exponential distribution optimizer

为解决混合光伏-温差(photovoltaic thermoelectric generator,PV-TEG)系统的最大功率点跟踪(maximum power point tracking,MPPT)问题以提高能源转换效率和利用率,提出了一种基于指数分布优化器(exponential distribution optimizer,EDO)的混合PV-TEG系统MPPT技术。EDO通过模拟指数分布的随机变化来搜索潜在的解空间,由于随机性,算法可有效避免在局部遮蔽条件(partial shading condition,PSC)下陷入局部最优,并在搜索空间中广泛探索以找到最优解。算例研究包括启动测试、太阳辐照度阶跃变化、随机变化、香港地区四季实际算例4个部分,并与其他5种算法进行对比分析,以较为全面地验证所提EDO技术在混合系统MPPT应用中的可行性和有效性。仿真结果表明,采用EDO的混合PV-TEG系统在不同运行条件下均能稳定、高效地实现最优越的MPPT性能,尤其是在春季低辐照度的条件下,EDO产生的能量分别超过蜻蜓算法(dragonfly algorithm,DA)、增量电导法(incremental conductance method,INC)、扰动观测法(perturbation observation method,P&O)能量输出的68.85%、66.13%和59.69%。

To solve the maximum power point tracking(MPPT)problem of a hybrid photovoltaic-thermoelectric generator(PV-TEG)system to improve energy conversion efficiency and utilization,a hybrid PV-TEG system MPPT technique based on an exponential distribution optimizer(EDO)algorithm is proposed.The EDO searches the potential solution space by modeling random variations of the exponential distribution.Because of the randomness,the algorithm can effectively avoid falling into a local optimum under partial shading condition(PSC),and explores extensively in the search space to find the optimal solution.The case study includes four parts:start-up test,step change of solar irradiance,stochastic variation,and actual cases of four seasons in Hong Kong region,and is compared to and analyzed with five other algorithms to more comprehensively verify the feasibility and effectiveness of the proposed EDO technique in the application of hybrid system MPPT.The simulation results show that the hybrid PV-TEG system with EDO can achieve superior MPPT performance stably and efficiently in different operating conditions.In particular,the energy generated by EDO in low irradiance conditions in spring exceeds 68.85%,66.13%,and 59.69%of the energy outputs of the dragonfly algorithm(DA),incremental conductance method(INC),and perturbation observation method(P&O),respectively.