Analysis of Sewage Volume by Decomposition of Time Series Water Temperature Data

  • Sato Katsumi
    日本大学生産工学部土木工学科 日本下水道協会特別会員
  • Nakane Susumu
    中日本建設コンサルタント株式会社水工技術本本部 日本下水道協会特別会員
  • Takahashi Iwahito
    日本大学生産工学部土木工学科 日本下水道協会特別会員
  • Hosaka Seiji
    日本大学生産工学部環境安全工学科 日本下水道協会特別会員
  • Morita Hiroaki
    日本大学生産工学部土木工学科 日本下水道協会特別会員

Bibliographic Information

Other Title
  • 時系列水温データの成分分解による下水量解析
  • ジケイレツ スイオン データ ノ セイブン ブンカイ ニ ヨル ゲスイリョウ カイセキ

Search this article

Abstract

<p> Based on the characteristics of the measured water temperature by the water temperature method, which is one of the methods for infiltration and inflow, the water temperature measured using the moving average method was decomposed. And the decomposed water temperature was divided into fine weather water temperature and rainy weather water temperature. The difference in water temperature distribution between fine weather and rainy weather is expressed as a non-excess probability (power of detection), and this probability can be used as an index to evaluate the effect of infiltration of storm-water between sewage water. By using a neural network to estimate the water temperature drop due to rainfall in the measured water temperature, and then analyze the ratio of the total water temperature drop due to rainfall to the total water temperature measured during the measurement period. It was confirmed that the non-exceedance probability represents the decrease in water temperature due to rainfall. Then we conducted a flow rate survey at some water temperature measurement points. And also estimated the infiltration of storm-water ratio (= total infiltration of storm-water rate/total flow rate) using a neural network, and confirmed that there was a correlation with the non-excess probability. Therefore, by using a neural network, it was confirmed that the nonexcess probability expresses the infiltration of storm-water ratio.</p>

Journal

Details 詳細情報について

Report a problem

Back to top