Doktorsavhandling

Chengqian Zhang, Marin teknik

Voyage Planning Framework for Autonomous Inland Waterway Vessels: Ship Performance Modelling and Operational Analysis

Översikt

Inland waterway transport offers substantial potential to mitigate greenhouse gas emissions and congestion from road freight transport. The development of advanced inland vessels equipped with clean energy systems and a high degree of automation represents a promising direction for future transport networks. Nevertheless, inland waterways are often restricted by shallow water and limited manoeuvring space. Hence, the deployment of full-scale autonomous vessels requires careful consideration of environmental and operational challenges to ensure safety and energy efficiency. To enhance the automation of inland shipping, this thesis provides an in-depth analysis of how shallow and confined water affects ship resistance, manoeuvring, control implementation, and energy efficiency. The aim of this thesis is to develop a holistic simulation framework for ship performance prediction and operational analysis.

The development of such a comprehensive framework requires modelling of vessel characteristics with hydrodynamic and river hydraulic effects. Firstly, this thesis develops a novel ship energy performance model explicitly tailored for resistance and energy consumption prediction of inland waterway vessels. It aims to generate fast and accurate predictions based on a collection of purely empirical methods. Secondly, this thesis develops a Manoeuvring Modelling Group-based manoeuvring model for motion prediction under shallow water and bank effects. Building upon this, a systematic control design was conducted with consideration of ship hydrodynamic characteristics in confined waterways. A comparison of a conventional controller and an advanced model predictive control was performed to evaluate their performance and robustness in tackling path-following tasks under complex environments with various disturbances. With these models and control methods, an integrated voyage planning framework (VPF) is proposed for analysing vessel operations. It captures a vessel’s dynamic performance with energy consumption analysis under coupled interactions between ship hydrodynamics, river hydraulics, and motion control. Based on the analysis of ship energy performance, this thesis proposes a particle swarm optimisation (PSO) module for fuel optimisation in inland waterways.

Validation studies with full-scale trial measurements revealed that the energy performance prediction model achieved promising accuracy, ensuring a mean absolute error below 10% based on finite input parameters. It was also observed that disturbances in inland waterways, such as currents and bank effects, significantly affect a vessel’s course stability, which should be carefully considered in the control system development of autonomous vessels operating in narrow channels. A series of case studies showcased the VPF’s capabilities for enabling a wide range of applications, such as route assessment, evaluation of control design, and operational energy efficiency analysis. It was shown that the PSO-based optimisation achieved an average of 5.7% fuel savings in shallow water operations with water depth-aware speed initialisation methods.