Power Monitoring System Using Machine Learning
Power Monitoring System using Machine Learning, which has feature of finding the extra load usage on specific loads usingRaspberry Pi
Power Monitoring | Over usage alert system
Shipping: 4 to 8 working days from the date of purchase
Complete Hardware Kit
Demo Video-Embedded Below
PPT (20 Slides)
!!! Online Support !!!
100 in stock
In this power monitoring system, the power of the bulb is monitored using a machine learning technique. This system consists of raspberry pi3 b+, ADC, 6volt transformer, bulb holder and current sensor. In which raspberry pi plays as mini-computer to this system, the ADC (MCP 3008) IC is used to convert the analog value to digital value, the 6-volt transformer is used to reduce the 230 volts to 6 volts and last the bulb holder is used to hold the bulb.
In the existing system, the classification is done based on manually calculating the analog value and do the calculation based upon it.
In this proposed system, machine learning is used with Sci-kit learn, in which linear regression is used to classify the power of the bulb.
In this application the raspberry pi is interfaced with ADC (MCP 3008) IC, from the current sensor is connected to ADC IC, the transformer is connected with the bulb holder and then to the current sensor, all the ground is shorted to one pin.
In this project, the data set is created for the testing bulbs, the first data set consists of four classes such as NOLOAD, MEDIUM, NORMAL and RISK. The second data set consist of three class such as NOLOAD, NORMAL, RISK. From there the class and the parameter are split separately and send them in Logistic Regression then the prediction is done based on the trained model the class is printed as output in a display.
- Raspberry Pi
- 6-volt transformer
- Power supply
- ADC (MCP 3008) IC
- Current Sensor
- Testing bulbs
- Raspbian OS with libraries installed
- SD Card Formatter
- Etcher/Win32 Disk imager
In the project, you can able to see the output obtained which classifies the power of the bulb with more than 90% accuracy with 4 classes.