Photovoltaic optimizer boost converters: Temperature influence and electro-thermal design

Abstract

ObjectivePhotovoltaic (PV) systems can operate in presence of not uniform working conditions caused by continuously changing temperature and irradiance values and mismatching and shadowing phenomena. The more the PV system works in these conditions, the more its energy performances are negatively affected. Distributed Maximum Power Point Tracking (DMPPT) converters are now increasingly used to overcome this problem and to improve PV applications efficiency. A DMPPT system consists in a DC–DC converters equipped with a suitable controller dedicated to the Maximum Power Point Tracking (MPPT) of a single PV module. It is arranged either inside the junction-box or in a separate box close to the PV generator. Many power optimizers are now commercially available. In spite of different adopted DC–DC converter topologies, the shared interests of DMPPT systems designers are the high efficiency and reliability values. It is worth noting that to obtain so high performances converters, electronic components have to be carefully selected between the whole commercial availability and appropriately matched together. In this scenario, an electro-thermal design methodology is proposed and a reliability study by means of the Military Handbook 217F is carried out. MethodThe developed DMPPT converters design method is constituted by many steps. In fact, beginning from installation site, PV generators and load data, this process selects power optimizers commercially available devices and it verifies their electro-thermal behavior to the aim to identify a set of suitable components for DMPPT applications. Repeating this process many times, many different feasible solutions can be found. An elaboration step follows to the “optima” power optimizer recognition among the whole obtained converters. In this case, a multi-objective optimization, consisting in the maximization of the solutions European efficiency and in the minimization of their cost, is executed and all not dominated solutions with respect to at least one of the two objectives are selected. The strength of the described method is represented by accurate PV generators and optimizer devices models. In detail, in the developed models particular attention is reserved to the thermal factor and to the quantification of the temperature action on devices parameters and performances. In fact, in such multiple and continuous changing working conditions, the temperature influence on components behavior can considerably vary their properties causing the whole converter performances worsening. The other important aspect, the converter reliability, is estimated by the reliability prediction model Military Handbook 217F. ResultsThe proposed tool is applied to Diode Rectification (DR) boosts and Synchronous (SR) boosts design. To completely characterize the obtained solutions their efficiency, cost and reliability performances are evaluated. In detail, Pareto fronts in terms of European efficiency and cost are identified for the SR and DR cases. Among the whole not dominated solutions, a SR converter characterized by a European efficiency of 97.1% and a DR boost characterized by a European efficiency of 95.5% are chosen. Their cost is comparable and equal about to $11. Then their reliability performances are evaluated by means of the Military Handbook 217F Notice 2. The carried out analysis shows that, for the same device cost, the SR solution represents the best one if efficiency is the most critical aspect. DR boost is, instead, the optimum solution if reliability represents the tighter requirement. ConclusionThe proposed DMPPT converters methodology permits to design families of feasible power optimizers. This process is applied to two boost versions, so two sets of power optimizers are obtained and a trade-off solution is chosen for each set. To correctly select the more suitable optimizer, a characterization in terms of efficiency, cost and reliability is carried out. In detail, the SR optimizer is characterized by lower losses and higher efficiency than the DR one. On the other hand, the DR boost results more reliable than the SR converter. So the optimum solution has to be chosen on the base of the most critical requirement. Practical implicationThe developed method can represent a useful tool to design DMPPT optimizers able to assure high level performances in terms of economical and technical aspects. It can be applied to many commercially available PV generators and, without loss of generality, it can be used with different DC–DC converter topologies.