PROGRAM AND POWER CONSUMPTION PREDICTION DEVICE

PROBLEM TO BE SOLVED: To propose a method for highly accurately predicting power consumption when a measurement frequency is changed according to a weather condition concerning a wireless sensor network.SOLUTION: A rainfall prediction value at each prediction time in a prediction period is calculated by probability calculation based on rainfall data being an observation value of past rainfall. A rainfall prediction value is compared with a predetermined threshold, so as to predict a measurement frequency in a prediction period. Measurement is performed at a predicted measurement frequency and an operation of each sensor is predicted to transfer measurement data to an aggregation device. Then, a prediction value of power consumption in a prediction period is calculated.SELECTED DRAWING: Figure 3 【課題】ワイヤレスセンサネットワークにおいて、気象状況に応じて計測頻度を変更する場合の消費電力の予測を高精度に実現し得る手法を提案すること。【解決手段】過去の雨量の観測値である雨量データに基づく確率演算によって、予測期間における各予測時刻の雨量の予測値を算出し、この雨量の予測値を所定の閾値と比較することで、予測期間における計測頻度を予測する。そして、予測した計測頻度で計測を行って計測データを集約装置へ転送するように各センサの動作を予測して予測期間における消費電力の予測値を算出する。【選択図】図３