An Enhanced Extreme Learning Machine for Dissolved Oxygen Prediction in Wireless Sensor Networks

Liang Kuang,Hui Zhu,Pei Shi,Chi Hua,Beijing Chen

doi:10.1109/access.2020.3033455

Abstract

Water quality monitoring using Wireless Sensor Networks (WSNs) is essential in aquaculture water quality management. In the field of water quality monitoring, dissolved oxygen (DO) is a key parameter, and its prediction can provide decision support for aquaculture production, thereby reducing farming risk. However, it is difficult to build a precise prediction model, and existing methods of DO prediction neglect the importance of analyzing DO content. To address this problem, this study proposes a hybrid DO prediction model, named KIG-ELM, which is composed of K-means, improved genetic algorithm (IGA), and extreme learning machine (ELM). This model is based on edge computing architecture, in which data acquisition, processing and dissolved oxygen prediction are distributed in sensing nodes, routing nodes and server respectively. Sensing technique and clustering operation are applied in the process of data acquisition and processing. Meanwhile, an optimized extreme learning machine is implemented for DO prediction. We evaluate the efficiency and accuracy of proposed prediction approach in a practical aquaculture on massive water quality data. Experimental results show that the hybrid model achieves significant prediction results and can meet the needs of practical production and management.

Highlights

In recent years, the development of Wireless Sensor Networks (WSNs) has promoted the progress of smart fishery
In this paper we propose a new multi-parameter Dissolved oxygen (DO) prediction method, named KIG-extreme learning machine (ELM), which is based on edge computing to improve prediction accuracy and efficiency
This paper constructs a new multi-parameter prediction method KIG-ELM to forecast the DO content in aquaculture based on edge computing architecture

Summary

Introduction

The development of WSNs has promoted the progress of smart fishery. To reduce the data transfer volume and bandwidth caused by these smart devices, edge computing provides guidance and strategy [1]–[3]. As a special application of WSNs, the study of water quality monitoring has become an important tool for sustainable development of smart fishery [4]–[6]. Based on edge computing architecture [7]–[9], water quality monitoring using WSNs can monitor and analyze culture. Dissolved oxygen (DO) concentration is a critical water quality parameter in aquaculture, which needs to be monitored in real time in intensive aquaculture [10]. Accurate and effective prediction of dissolved oxygen plays an important directional role in production, which can produce

Objectives

Methods

Results

Conclusion