Integrid - Bridging the Gap

32 In addition to this data, an algorithm was applied to measure the permissible load that each transformer can be subject to without significantly accelerating the equipment aging rate. This was done to compute the ratio between the nominal transformer load and the measured load. After computing the ratio, two transformers were compared that had failures with others which were still working. The results of this analysis are shown in Figures 4.10 and 4.11, respectively. Each figure depicts the distribution of the load ratio from 2015-2019, as well as the transformer calendar age (see Figures 4.10 and 4.11 top right corner) at the end of our analysis. From the results it was possible to conclude that transformers that operated more frequently close to the nominal load or even surpassing it, failed at a lower calendar age than transformers still operating. Nevertheless, it is essential to underline that these results were obtained from a small sample. Further studies need to be carried out to validate these conclusions. With this in mind, the team is developing a traffic light system that will analyse the load ratio for a sample of more than 5,000 dry transformers. The goal is to highlight the transformers under more load and then analyse these same transformers in more detail. KPI calculation The following metrics were then calculated as a result of the HLUC03 demonstration. A brief description is also provided. Figure 4.10 – Load ratio distribution for transformers with failures Figure 4.11 – Load ratio distribution for transformers without failures Table 4.3 – Summary of KPI results KPI Result Description Nominal load ratio 15% (average) This KPI computes the ratio between the transformer nominal load and the transformer load at a specific point in time.