A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Ultrasonic Power Load Forecasting Based on BP Neural Network
https://orcid.org/http://orcid.org/0000-0002-2431-6945
The use of power load forecasting in aluminum alloy ultrasonic-assisted casting systems can improve its working efficiency and stability and simultaneously improve the casting quality of the aluminum alloy as well. A power load forecasting model based on a back-propagation neural network was designed and embedded in an ultrasonic power supply, referred to as the new ultrasonic power supply; the ultrasonic power supply without power load forecasting was referred to as the traditional ultrasonic power supply. Both these power supplies were used in an experimental process of 7085 aluminum alloy ultrasonic-assisted casting. The power load range and harmonic frequency range were 953.01–1194.02 W and 16.03–19.1 kHz for the traditional ultrasonic power supply with an average grain size of 179.93 µm and 1073.1–1213.02 W and 17.94–20.04 kHz for the new ultrasonic power supply with an average grain size of 139.41 µm, respectively. The results of the ultrasonic-assisted alloy casting experiment showed that the design of the proposed power load forecasting model could improve the work efficiency of the assisted casting system as well as the quality of the aluminum alloy casting.
Ultrasonic Power Load Forecasting Based on BP Neural Network
https://orcid.org/http://orcid.org/0000-0002-2431-6945
The use of power load forecasting in aluminum alloy ultrasonic-assisted casting systems can improve its working efficiency and stability and simultaneously improve the casting quality of the aluminum alloy as well. A power load forecasting model based on a back-propagation neural network was designed and embedded in an ultrasonic power supply, referred to as the new ultrasonic power supply; the ultrasonic power supply without power load forecasting was referred to as the traditional ultrasonic power supply. Both these power supplies were used in an experimental process of 7085 aluminum alloy ultrasonic-assisted casting. The power load range and harmonic frequency range were 953.01–1194.02 W and 16.03–19.1 kHz for the traditional ultrasonic power supply with an average grain size of 179.93 µm and 1073.1–1213.02 W and 17.94–20.04 kHz for the new ultrasonic power supply with an average grain size of 139.41 µm, respectively. The results of the ultrasonic-assisted alloy casting experiment showed that the design of the proposed power load forecasting model could improve the work efficiency of the assisted casting system as well as the quality of the aluminum alloy casting.
Ultrasonic Power Load Forecasting Based on BP Neural Network
https://orcid.org/http://orcid.org/0000-0002-2431-6945
The use of power load forecasting in aluminum alloy ultrasonic-assisted casting systems can improve its working efficiency and stability and simultaneously improve the casting quality of the aluminum alloy as well. A power load forecasting model based on a back-propagation neural network was designed and embedded in an ultrasonic power supply, referred to as the new ultrasonic power supply; the ultrasonic power supply without power load forecasting was referred to as the traditional ultrasonic power supply. Both these power supplies were used in an experimental process of 7085 aluminum alloy ultrasonic-assisted casting. The power load range and harmonic frequency range were 953.01–1194.02 W and 16.03–19.1 kHz for the traditional ultrasonic power supply with an average grain size of 179.93 µm and 1073.1–1213.02 W and 17.94–20.04 kHz for the new ultrasonic power supply with an average grain size of 139.41 µm, respectively. The results of the ultrasonic-assisted alloy casting experiment showed that the design of the proposed power load forecasting model could improve the work efficiency of the assisted casting system as well as the quality of the aluminum alloy casting.
Ultrasonic Power Load Forecasting Based on BP Neural Network
J. Inst. Eng. India Ser. C
Chen, Xiwen (author) / Li, Xiaoqian (author) / Li, Ruiqing (author)
Journal of The Institution of Engineers (India): Series C ; 101 ; 383-390
2020-04-01
8 pages
Article (Journal)
Electronic Resource
English
Ultrasonic Power Load Forecasting Based on BP Neural Network
| Online Contents | 2019
Embedding based quantile regression neural network for probabilistic load forecasting
| DOAJ | 2018