Abstract The decentralised control of multicellular converters is an alternative to the usual control techniques. Multicellular converters use interleaved carriers to reduce filtering elements and potentially improve transient response. Decentralised interleaving techniques are scalable, avoid a complex interleaving controller for a high number of cells, and simplify reconfiguration cases like phase shading. Circular chain of communication (ring, or daisy chain) approaches have been proposed in the literature and implemented in concrete applications. However, the analytical study of the stability and dynamic response of this system involving several identical phase-delay local controllers connected in a dedicated communication chain with their close neighbours has not been conducted yet. This paper presents a behavioural discrete model for a digitally implemented decentralised interleaving device. The eigenvalue study gives the stability criterion and convergence speed to choose the appropriate parameters of the controllers. A modal decomposition technique dissociates the various types of differential interactions to observe their respective time response. Simulation results demonstrate that the system is unconditionally stable when all differential modes are properly damped. An analytical expression for the final disposition of the carriers in steady state depending on the start-up condition is established. Lastly, a more precise operator to overcome the singular discontinuity of the model is presented and discussed. Experimental validation by FPGA implementation have been done for reconfiguration and start-up cases.

中文翻译：

---

多蜂窝转换器数字迭代分散交织的特征分解

摘要 多单元变换器的分散控制是对通常控制技术的一种替代。多单元转换器使用交错载波来减少滤波元件并潜在地改善瞬态响应。分散式交织技术是可扩展的，避免了大量单元的复杂交织控制器，并简化了相位着色等重新配置情况。循环通信链（环形或菊花链）方法已在文献中提出并在具体应用中实施。然而，尚未对该系统的稳定性和动态响应进行分析研究，该系统涉及在专用通信链中与其近邻连接的几个相同的相位延迟本地控制器。本文提出了一种数字实现的分散式交织设备的行为离散模型。特征值研究给出了选择合适的控制器参数的稳定性准则和收敛速度。模态分解技术分离各种类型的微分相互作用以观察它们各自的时间响应。仿真结果表明，当所有微分模式都被适当阻尼时，系统是无条件稳定的。建立了依赖于启动条件的稳态载流子最终配置的解析表达式。最后，提出并讨论了一个更精确的算子来克服模型的奇异不连续性。已针对重新配置和启动情况通过 FPGA 实现进行了实验验证。