Analysis of Classifier Performance on Aggregated Energy Status Data: Unterschied zwischen den Versionen

Aus SDQ-Institutsseminar
Keine Bearbeitungszusammenfassung
Keine Bearbeitungszusammenfassung
 
Zeile 5: Zeile 5:
|betreuer=Dominik Werle
|betreuer=Dominik Werle
|termin=Institutsseminar/2018-06-22
|termin=Institutsseminar/2018-06-22
|kurzfassung=Non-intrusive load monitoring(NILM) algorithms aim at disaggregating the energy dataset created by households to appliance level. However, there is no conventional way of quantifying and representing the tradeoff between the quality of analyses, such as the accuracy of the disaggregated consumption curves, and the load on the available resources. Thus, it is hard to plan the underlying infrastructure and resources for the analysis system and to find the optimal configuration of the system. In this thesis, we want to plan such a system which can, on the one hand, assess the quality of different analyzes. On the other hand, allocates the runtime behavior for multiple analysis based on varying configuration parameters and changed characteristics of the input dataset. Besides, be aware of the real scenario we added artificial noise for simulating the noisy data, which could be generated by low-quality sensors, and observing how it impacts the analyses' qualities.
|kurzfassung=Non-intrusive load monitoring (NILM) algorithms aim at disaggregating consumption curves of households to the level of single appliances. However, there is no conventional way of quantifying and representing the tradeoff between the quality of analyses, such as the accuracy of the disaggregated consumption curves, and the load on the available computing resources. Thus, it is hard to plan the underlying infrastructure and resources for the analysis system and to find the optimal configuration of the system. This thesis introduces a system that assesses the quality of different analyses and their runtime behavior. This assessment is done based on varying configuration parameters and changed characteristics of the input dataset. Varied characteristics are the granularity of the data and the noisiness of the data. We demonstrate that the collected runtime behavior data can be used to choose reasonable characteristics of the input data set.
}}
}}

Aktuelle Version vom 28. Mai 2018, 16:35 Uhr

Vortragende(r) Dou Beibei
Vortragstyp Masterarbeit
Betreuer(in) Dominik Werle
Termin Fr 22. Juni 2018
Vortragsmodus
Kurzfassung Non-intrusive load monitoring (NILM) algorithms aim at disaggregating consumption curves of households to the level of single appliances. However, there is no conventional way of quantifying and representing the tradeoff between the quality of analyses, such as the accuracy of the disaggregated consumption curves, and the load on the available computing resources. Thus, it is hard to plan the underlying infrastructure and resources for the analysis system and to find the optimal configuration of the system. This thesis introduces a system that assesses the quality of different analyses and their runtime behavior. This assessment is done based on varying configuration parameters and changed characteristics of the input dataset. Varied characteristics are the granularity of the data and the noisiness of the data. We demonstrate that the collected runtime behavior data can be used to choose reasonable characteristics of the input data set.