ICACT20230034 Question.1
Questioner: saocd@viettel.com.vn 2023-02-21 오후 7:56:00 |
ICACT20230034 Answer.1
Answer by Auhor mahdi.firoozjaei@uwindsor.ca 2023-02-21 오후 7:56:00
Chrome  Click!! |
| Thank you for your presentation. Based on analysis and evaluation for your model, you only use 61 days of historical data to estimate. Why do you use such short data and when the data is larger, maybe a year, the accuracy of your model can change? |
Yes, it is. A short range the measurment days were used at that stage. This paper is a progressive work and represents the initial results of this work.
We are working on bigger dataset of measurement. Its results will be published in its final verfsion paper.
Definitely, more evaluation items are needed for accourate clustering. We extend the classification by adding more time classes and use Daily load factor more than the devices' usage data points. |
ICACT20230034 Question.2
Questioner: arifulislamro@gmail.com 2023-02-21 오후 8:07:38 |
ICACT20230034 Answer.2
Answer by Auhor mahdi.firoozjaei@uwindsor.ca 2023-02-21 오후 8:07:38
Chrome Click!! |
| What is the importance of Time-Series Load Data Analysis for User Power Profiling? |
Time series classification has many applications. Time-series load data analysis can be used for feature-based classification. In this case, these series provide statistical features representation which can be used for profiling a smart grid user's power usage. |
ICACT20230034 Question.3
Questioner: yiweimaa@gmail.com 2023-02-22 오후 2:12:01 |
ICACT20230034 Answer.3
Answer by Auhor mahdi.firoozjaei@uwindsor.ca 2023-02-22 오후 2:12:01
Chrome Click!! |
| Why did you choose to use DTW algorithm to profile a consumer’s power consumption behaviour and analyzes the load data provided by smart meters? |
We selected DTW, due to its properties, i.e., signal warping invariability and implementation simplicity. By clustering the consumer’s load data with the DTW algorithm, load forecasting is performed by correlating the historical load data and time series of the new power usage data. |
