A study on optimizing class A power quality measurements for low-end devices using Mixed Radix FFT combined with resampling

In recent years, rapid advancements in AI technology, along with increasing data usage and power consumption, have heightened concerns about power quality in data centers, where outages can disrupt networks and cause economic losses. This calls for reliable power quality monitoring. IEC 61000-4-30 Class A devices ensure precise measurement and monitoring of power quality parameters, supporting advanced features, including high-order harmonic analysis and flicker. However, these functions demand heavy computational resources, requiring high-spec hardware that increases costs and limits their wider adoption for monitoring. This study addresses these challenges by optimizing the aggregation process, a critical step in Class A compliance, for low-end devices. By analyzing IEC 61000-4-30 standard, we propose a method using Mixed Radix FFT combined with Resampling to reduce computational loads while meeting requirements. The proposed approach enables Class A functionality without a high-spec CPU or an FPGA, lowering the cost of power quality meters and promoting their wider adoption in power quality monitoring applications.