A multi-level monitoring mechanism for inland ships sewage based on software-defined cloud-edge-end collaborative architecture

Abstract

In response to the issues of large data processing volume, weak real-time capability, narrow monitoring scope, and insufficient utilization of monitoring data for business applications within traditional ship sewage monitoring mechanisms, this paper proposes a novel ship sewage monitoring mechanism. Combining the advantages of cloud-edge collaborative computing and software defined network architecture, a multi-level ship sewage monitoring mechanism for real-time monitoring of inland ship wastewater is presented. The proposed system offers flexibility for large-scale deployment while effectively managing the collaborative workflow between system data, operational processes, and business applications. A task offloading model and multi-agent deep deterministic policy gradient based task scheduling scheme is introduced to optimize resource allocation within the monitoring system. Through a case study conducted on the Wuhan section of the Yangtze River, the effectiveness and superiority of this approach are validated. The results demonstrate the practical utility of the inland ship wastewater monitoring system and its associated processing methods. Compared with the traditional centralized architecture, the end-to-end delay of the proposed architecture is reduced by 42%, the packet sending rate is increased by 15%, and the energy consumption is reduced by 23%. The research contributes to the advancement of environmental monitoring technology, offering insights into optimizing resource allocation and enhancing real-time monitoring capabilities for inland waterway water quality management.