Cost and Reliability Design Consideration for CHB-based Converter Based on Decomposition Optimization

In this paper, an optimal solution is introduced for CHB-based inverters to improve reliability while the total cost is reduced. The trade-off between the conflicting objectives (minimizing cost and maximizing reliability) can be solved using decomposition optimization methods such as goal attainment and goal programming. In the proposed optimization framework, the objective functions are degraded into one single-objective function, repeated for 200 different weight factors. Thus, an accurate approximation of the Pareto set is determined. The proposed optimization scheme can control both reliability and total cost within the allowable range. In addition, the number of redundant cells is determined to make the converter more reliable. Eventually, the optimal number of cascade cells is determined as (13,5) for cold redundancy and (7,4) for hot one.