Purpose

Power converters are an integral part of the energy conversion process in solar photovoltaic (PV) systems which is used to match the solar PV generation with the load requirements. The increased penetration of renewable invokes intermittency in the generated power affecting the reliability and continuous energy supply of such converters. DC-DC converters deployed in solar PV systems impose stringent restrictions on supplied power, continuous operation and fault prediction scenarios by continuously observing state variables to ensure continuous operation of the converter.

Design/methodology/approach

A converter deployed for a mission-critical application has to ensure continuous regulated output for which the converter has to ensure fault-free operation. The fault diagnostic algorithm relies on the measurement of a state variable to assess the type of fault. In the same line, a predictive controller depends on the measurement of a state variable to predict the control variable of a converter system to regulate the converter output around a fixed or a variable reference. Consequently, both the fault diagnosis and the predictive control algorithms depend on the measurement of a state variable. Once measured, the available data can be used for both algorithms interchangeably.

Findings

The objective of this work is to integrate the fault diagnostic and the predictive control algorithms while sharing the measurement requirements of both these control algorithms. The integrated algorithms thus proposed could be applied to any converter with a single inductor in its energy buffer stage.

Originality/value

laboratory prototype is created to verify the feasibility of the integrated predictive control and fault diagnosis algorithm. As the proposed method combine the fault detection algorithm along with predictive control, a load step variation and manual fault creation methods are used to verify the feasibility of the converter as with the simulation analysis. The value for the capacitors and inductors were chosen based on the charge-second and volt-second balance equations obtained from the steady-state analysis of boost converter.

中文翻译：

---

光伏系统中基于单电感器的DC-DC转换器的集成预测控制和故障诊断算法

目的

功率转换器是太阳能光伏（PV）系统中能量转换过程不可或缺的一部分，用于使太阳能光伏发电与负载需求相匹配。可再生能源渗透率的提高会引起发电功率的间歇性，从而影响此类转换器的可靠性和持续的能源供应。太阳能光伏系统中部署的DC-DC转换器通过不断观察状态变量以确保转换器连续运行，对供电，连续运行和故障预测方案施加了严格的限制。

设计/方法/方法

部署用于关键任务应用的转换器必须确保连续稳定的输出，而转换器必须确保其无故障运行。故障诊断算法依靠状态变量的测量来评估故障的类型。在同一行中，预测控制器依赖于状态变量的测量值来预测转换器系统的控制变量，以围绕固定或可变参考值调节转换器输出。因此，故障诊断和预测控制算法都取决于状态变量的测量。一旦测量，可用数据就可以互换地用于两种算法。

发现

这项工作的目的是集成故障诊断和预测控制算法，同时共享这两种控制算法的测量要求。由此提出的集成算法可以应用于在其能量缓冲级中具有单个电感器的任何转换器。

创意/价值

创建实验室原型以验证集成预测控制和故障诊断算法的可行性。由于所提出的方法将故障检测算法与预测控制相结合，因此采用负载阶跃变化和人工故障创建方法，与仿真分析一样，可以验证变频器的可行性。电容器和电感器的值是根据从升压转换器的稳态分析中获得的充电秒和伏秒平衡方程式选择的。