UPS Power Measurement and Prediction Towards Video Surveillance Systems

The UPS is advantageous and necessary for critical loads. Circuits and sensors are created and put into place for UPS protection to supply the load with power continuously. The battery voltage sensor, which will activate a light if the DC voltage of a battery connected to the UPS equals the charging voltage of the UPS and so prevent the battery from overcharging, was the missing sensor or alarm in the UPS. This sensor, along with numerous other sensors, is added to the UPS to improve stability and dependability. It is possible to anticipate UPS metrics like battery status voltage, current, and mode of the UPS that is currently operating. It is crucial to know exactly where the downed node is in the network of the security system. The voltage sensor that reads the immediate voltage readings of the backup battery and the hardware system in turn delivers this data over the internet for monitoring purposes is the missing piece for continuous monitoring of the UPS parameters. A further problem in confirming the proper operation of the surveillance system is the prediction of UPS metrics such battery status voltage, current, and mode of the UPS that is now operating. The creation and implementation of a hardware UPS monitoring system is carried out. The hardware implementation of the design is followed with creating and implementing a web-based system using ThingSpeak to track the UPS’s parameters. Further three algorithms namely Linear regression, SVM and random forest algorithms are designed and put into use to categorize UPS parameters.