Scholarlink

<p>It is difficult to regulate groundwater and monitor water use in urban settings. It is feasible to create a model that precisely forecasts the water quality in an urban setting by applying certain strategies. Use information from monitoring wells to forecast the level of groundwater, including temperature, chemical concentrations, and pH values. This research work, Water Management Prediction using Deep Convolutional Spiking Neural Network Optimized with Red Fox Optimization Algorithm based on IoT (WMP-DCSNN-RFO) is proposed. Initially, real-time water quality data are collected from various sensor nodes using IoT (Internet of Things) devices deployed at different locations and it is preprocessed using Z-Score Normalization (Z-SN) for data normalization and cleaning of the input data.Then, Deep Convolutional Spiking Neural Network (DCSNN) is used to predict the water quality as excess, critical, moderated, and semi-critical by capturing complex spatial–temporal patterns in water quality data for effective Water Management. The proposed DCSNN does not show any adaption of optimization methods and RFOA is utilized to optimize DCSNN's weight parameters, preventing premature convergence and ensuring global optimality in dynamic and changing environmental conditions. The proposed method is implemented and evaluated using several performance metrics.The experimental results demonstrate that proposed technique performs well by achieving an accuracy of 99.27%, precision of 98.75%, sensitivity of 98.08%, and F1-score of 99.12% than existing models The proposed methodology significantly improves prediction accuracy while reducing computational time, making it a robust solution for real-time, large-scale water quality monitoring.</p>

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

Water Management Prediction using Deep Convolutional Spiking Neural Network Optimized with Red Fox Optimization Algorithm Based on IoT

  • 期刊论文

  • 发表日期：

    2025年6月6日

查看全文
  • V. Sushmitha Vadone,
  • Sibi Shaji,
  • Meenakshi Sundaram

摘要

It is difficult to regulate groundwater and monitor water use in urban settings. It is feasible to create a model that precisely forecasts the water quality in an urban setting by applying certain strategies. Use information from monitoring wells to forecast the level of groundwater, including temperature, chemical concentrations, and pH values. This research work, Water Management Prediction using Deep Convolutional Spiking Neural Network Optimized with Red Fox Optimization Algorithm based on IoT (WMP-DCSNN-RFO) is proposed. Initially, real-time water quality data are collected from various sensor nodes using IoT (Internet of Things) devices deployed at different locations and it is preprocessed using Z-Score Normalization (Z-SN) for data normalization and cleaning of the input data.Then, Deep Convolutional Spiking Neural Network (DCSNN) is used to predict the water quality as excess, critical, moderated, and semi-critical by capturing complex spatial–temporal patterns in water quality data for effective Water Management. The proposed DCSNN does not show any adaption of optimization methods and RFOA is utilized to optimize DCSNN's weight parameters, preventing premature convergence and ensuring global optimality in dynamic and changing environmental conditions. The proposed method is implemented and evaluated using several performance metrics.The experimental results demonstrate that proposed technique performs well by achieving an accuracy of 99.27%, precision of 98.75%, sensitivity of 98.08%, and F1-score of 99.12% than existing models The proposed methodology significantly improves prediction accuracy while reducing computational time, making it a robust solution for real-time, large-scale water quality monitoring.