Recursive construction of output-context fuzzy systems for the condition monitoring of electrical hotspots based on infrared thermography

Information granularity model for evolving context-based fuzzy system

“…Nevertheless, the initial preferences are often arbitrarily decided, and the designed model might not be able to achieve a desirable performance because the granularity depends on the distribution of the application problem. Therefore, granular computing with sequential decision making was proposed in [26,27] where finer granulation level with more detailed information is considered. A relationship between progressive computing and granular computing was proposed in SSEM [9], ECSFS [10], and top-down progressive computing [28].…”

“…Information granule has to be translated into significant operational frameworks of granular computing in a way that realizes the interpretability-accuracy tradeoff. Interpretability-accuracy tradeoff of fuzzy models is essential, although these contradicting objectives remain an open problem [27][28][29]. Therefore, the conflict situation needs to be handled.…”

Peculiarities Of Application Of Smart Grid Technology In Systems For Monitoring And Diagnostics Of Heat-andpower Engineering Objects

“…в работах [6][7][8][9][10] приведены средства контроля и методы повышения эффективности процесса сжигания топлива в котлоагрегатах. значительное внимание уделяется методам инфракрасной диагностики теплоэнергетического оборудования [11,12]. Актуальны разработки систем диагностирования технического состояния трансформаторного оборудования высокого напряжения [13].…”

An Examination of Thermal Features’ Relevance in the Task of Battery-Fault Detection

“…Similarly, the combination of the spot temperature with the temperature difference of two similar components and the maximum admissible temperature of the component was used in [9,10], the difference of the maximum and the minimum temperatures in the object, and the surrounding area, respectively, was designed. In [11], the authors used statistical features computed making use the pixel intensities of the image regions. The considered features were the maximum, minimum, mean and the standard deviation of intensity level.…”