Stereo-vision-based mooring tension monitoring method for offshore floating photovoltaic platforms

• Developed a stereo-vision method for OFPV mooring-tension monitoring, reducing cost and improving stability. • Achieved sub-pixel marker detection via concentric rings, yielding 95% motion and 97% fairlead position accuracy. • Enabled multi-line mooring-tension monitoring via shared markers, reaching 95% accuracy in moderate and 88% in extreme seas. • Analyzed the impact of viewing angles on accuracy, verified method stability, and provided deployment recommendations. Conventional mooring tension monitoring methods rely on contact sensors, which are expensive and prone to corrosion-induced failure, compromising their stability and durability for long-term monitoring. To address these limitations, this study proposes a stereo-vision-based mooring tension monitoring method for offshore floating photovoltaic (OFPV) platforms. Stereo cameras are used to capture video and identify markers mounted on the platform, from which 3D coordinates are reconstructed. The platform’s motion is then calculated to determine the spatial positions of the fairleads. Based on these positions and the known coordinates of the anchors, mooring tensions can be calculated using a quasi-static approach. A scaled physical model for the actual OFPV platform was produced, and hydrodynamic experiments on the scaled model were conducted to evaluate the performance of the proposed method. The two-chain mooring experiment confirmed the feasibility of this method, and the monitoring accuracies for fairlead position and mooring tension were approximately 97% and 95%, respectively. Moreover, the effects of camera viewing angles and perspective distortion on measurement accuracy were also examined. Based on the four-chain mooring experiment, the stability of this method was validated by successfully enabling simultaneous tension monitoring for multiple mooring lines. Experiment results indicate that the measurement accuracy exceeds 95% and 88% under moderate and extreme conditions, respectively. Compared to conventional methods, this method significantly reduces the measurement cost and implementation difficulty, and shows broad application prospects in both model testing and prototype monitoring.