Scholarlink

In order to control the operation and performance of wastewater treatment plants, it is vital to develop a stable model. This control reduces errors and maintenance c and maintains environmental balance. This research is concerned with applying an artificial neural network model to predict the outputs of Karbala- Iraq’s main wastewater treatment plant. The data collected from station records weekly for 3 years consists of the main parameters entering and leaving the station’s performance by predicting the influencing values affecting effluent of the station, such as biological oxygen demand, chemical oxygen demand, and the amount suspended. The mode performance was evaluated using the root mean square error and the correlation coefficient. The results showed that the inputs explain about 85% of the biological oxygen demand, chemical oxygen demand, and 16% of the amount suspended and that the error rate between the predicted and real data was 1.6 on average error. The artificial neural network was determined successfully in the treatment plant's performance simulation.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Employing Artificial Neural Networks in Predicting the Performance of the Wastewater Treatment Plant in Kerbala

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Sara Galb Salman,
  • Muhammad Abduredha,
  • Basim Khalil Nile

摘要

In order to control the operation and performance of wastewater treatment plants, it is vital to develop a stable model. This control reduces errors and maintenance c and maintains environmental balance. This research is concerned with applying an artificial neural network model to predict the outputs of Karbala- Iraq’s main wastewater treatment plant. The data collected from station records weekly for 3 years consists of the main parameters entering and leaving the station’s performance by predicting the influencing values affecting effluent of the station, such as biological oxygen demand, chemical oxygen demand, and the amount suspended. The mode performance was evaluated using the root mean square error and the correlation coefficient. The results showed that the inputs explain about 85% of the biological oxygen demand, chemical oxygen demand, and 16% of the amount suspended and that the error rate between the predicted and real data was 1.6 on average error. The artificial neural network was determined successfully in the treatment plant's performance simulation.