ASARIMA: An Adaptive Harvested Power Prediction Model for Solar Energy Harvesting Sensor Networks

Harvesting energy from solar radiation has emerged as an effective approach to prolong the lifetime of outdoor harvesting sensor networks. The harvested must be carefully managed ensure that sufficient is available when scarce. For prediction problem power harvesting, this paper proposes adaptive seasonal auto-regressive integrated moving average model (ASARIMA) for prediction. A training set can adaptively adjusted by similarity historical data, and then we conduct difference data fitting based on obtain optimal parameters. Experimental results show ASARIMA performs better than other existing algorithms. If weather conditions are stable, error decreases more 70%. change sharply, 20% in comparison with those