Hierarchical Synthesis Methods of DC-DC Converters

In order to synthesize DC-DC converters for specific application requirements, most of the existing synthesis methods of DC-DC converters are summarized. As listed in Tab, these methods can be divided into primary and secondary synthesis according to the three levels of “component-circuit cell-converter”. The principles and steps of typical synthesis methods are highlighted, and the requirements, scope, key technologies, development trends, advantages, and disadvantages of these methods are analyzed and compared. Moreover, the feasibility of using computer-aided technology in the above synthesis methods is discussed. Tab Classification of Synthesis Methods of DC-DC Converters Classification Hierarchy Typical Methods Primary synthesis Component-circuit cell State equation method, flux balance method, graph search method Switching cell method, duality, cascading, layer and graft method, interleaving, R2P2 Secondary synthesis Circuit cell-converter method, replacing and inserting of impedance networks, serial-parallel-cascading connection of multioutput converters The primary synthesis method refers to the synthesis of circuit cells directly from separate components. A circuit cell is a full-featured but relatively basic DC-DC converter that can be used as a basic unit to synthesize other converters. Regarding primary synthesis methods, some conclusions are listed as follows. (1) The flux balance method synthesis DC-DC converters according to specific requirements (e.g., voltage gain), while the state equation method and graph search method do not. In addition, the latter two methods derive a large number of converters with various features that require further screening. (2) The state equation method and the flux balance method traverse numerical values, so the implementation programs are simple. However, the graph search method traverses the graphs, which is more difficult to be programmed. (3) Computer programs can be utilized in some steps of primary synthesis methods, but the utilization of the program still needs to be improved. In the future, full-featured software is expected to realize the whole process, from inputting the requirements to outputting the converter circuits. Secondary synthesis methods involve synthesizing new topologies from existing topologies by changing connections of circuit cells, combining circuit cells, and adding or removing components or cells. Regarding secondary synthesis methods, some conclusions are listed as follows. (1) Cascading, interleaving, R2P2 method and serial-parallel-cascading connection of multioutput converters only need to fill the converter cells into specific structures, which require little computer-aided technology. (2) Switching cell method, duality, layer and graft method have relatively clear steps, which are suitable for programming. (3) The replacing and inserting method of impedance networks is flexible, and some significant problems still need to be solved manually. Therefore, programming is not appropriate for this method currently. In conclusion, the primary synthesis method is usually suitable for synthesizing DC-DC converters with fewer components and simpler functions, providing prototypes with different functions for the secondary synthesis. On the other hand, the secondary synthesis method is suitable for synthesizing complex converters. In the future, most of the synthesis methods are expected to be programmed, which can select and execute appropriate synthesis methods intelligently to synthesize a DC-DC converter that meets the given requirements. © 2023 Chinese Machine Press. All rights reserved.