Research on Power Marketing System Based on Intelligent Full Speech Recognition and Feature Analysis
- Resource Type
- Conference
- Authors
- Guan, Tairan; Huang, Jiangteng; Yang, Huidong; Yao, Menglang; Bao, Xufeng; Liu, Xinxin
- Source
- 2023 IEEE International Conference on Control, Electronics and Computer Technology (ICCECT) Control, Electronics and Computer Technology (ICCECT), 2023 IEEE International Conference on. :138-144 Apr, 2023
- Subject
- Communication, Networking and Broadcast Technologies
Fields, Waves and Electromagnetics
Power, Energy and Industry Applications
Robotics and Control Systems
Speech recognition
Computer architecture
Predictive models
Virtual machining
Power markets
Power systems
Servers
Full Speech Recognition
Power Marketing
Private Cloud Storage
Load Prediction
- Language
In this paper, a new electric power marketing information system is designed and developed by Silverstream platform with three layers including application server. Then, based on each server node in the speech recognition model, the system adjusts the virtual machine usage through the load prediction process. The image backup process is used to restore VM data when faults occur. The experimental results show that the load of different virtual machines is lower than 40% during the process of power marketing information storage. The VMS have the same load. This method can improve the blocking balance during information storage and improve storage efficiency.