Discharge Aware Optimization Method for Retaining State of Charge of Batteries using Knowledge Learning
Iyappan Murugesan, G. Kannan, M. Shanmuga Priya, Gunasekaran PrabhakarIntroduction:
Power battery’s storage and operational efficiency are measured using its State of Charge (SoC) and State of Health (SoH) metrics. The load handled by the batteries decides the persistence of SoC and SoH for which optimization is required.
Methods:
This article introduces a Discharge-aware Optimization method using Knowledge Learning (DOM-KL) to retain the sustainable SoC of load-induced batteries. Using the discharge factor for a load-induced battery operation and the duty cycle allocations, the SoC of the battery is computed in this method. Besides, the duty cycle merging operations are recommended with the aid of Internet of Things (IoT) computations. The knowledge learning retains the previous selection of load or duty cycle allocation through merging outcomes to reduce the abrupt discharge factor, even if the load is not available.
Results:
The proposed DOM-KL retains the SoC by 11.51% and SoH by 11.16% by reducing the degradation factor by 12.17% and mean square error by 12.38%, for the maximum duty cycle variant.
Discussion:
The proposed DOM-KL is significant for the varying duty cycle intervals and time factors, balancing the improvements on SoC by reducing its associated mean squared error and number of degrading cycles.
Conclusion:
The proposed method is reliable in terms of adaptability and efficiency regardless of the varying input battery load. Besides, the integration of diverse technologies helps to maintain the operational lifetime of the batteries.