The aim of this thesis is to develop, validate and deploy numerical modelling techniques to further understand and optimise the resin infusion and curing strategies of wind turbine blade manufacturing. To further develop the accuracy and understanding of the flow phenomina that occur during wind turbine blade manufacturing, three key goals are achieved:• A model is developed and validated that is accurate in the simulation of resin infusion into blade relevant geometry and under appropriate conditions without the presence of resin distribution blocks.• The effect of blocks on infusing flow are isolated and quantified under relevant geometry and conditions on two scales: scale of a single block, and a wind turbine blade subcomponent. The interactions between these two scales is studied, demonstrating the effect of each flow mechanism on the infused structure.• These models are implemented into blade relevant conditions using the same materials, resins and processes to optimise the infusion and cure strategies of blade resin infusion processes.This work integrates and validates the time-temperature-cure-viscosity relationship into a fully flexible CFD solver (named VRTMFoam) built upon the openFoam framework, allowing for the specific traits of individual resins to be assessed under conditions akin to that seen in the manufacturing of wind turbine blades by SGRE. VRTMFoam was validated in stages; attaining a 0.04% error when simulating isothermal, non-curing flows, and attaining less than 5% error when simulating non-isothemal curing flows; a significant step closer to full blade complexity models. The strategic inclusions of processes such as heat transfer between the resin and reinforcement phase also allow it to act as a test-bed for further development into the optimisation of the infusion and curing process.The time-temperature-cure relationship is also developed as a stand-alone system, allowing for a light weight, lumped parameter version of the modelling technique to be used to optimise the post-infusion temperature profiles. Using this technique a curing strategy with a 40% reduction in processing time was developed for the manufacturing of a key subcomponent of the Siemens Gamesa B108 blade.The effects of both resin distribution grooves and the bulk balsa core material itself is investigated. From this, it is shown that the balsa core absorbs resin into its structure, inturn slowing the infusing flow along its surface. Each standard size, 0.6m by 1.2m block absorbs around 1.8kg of resin, equating to over 200kg of the SGRE B108 weight being resin held within the balsa, around 30% more than the expected weight of the balsa. Whilst this does increase the weight of the structure significantly, it does however mean that the high interface area between the balsa and the composite contributes significantly to the interface strength. The passing of resin through the balsa also couples the infusing flow on top and below the core material, with the balsa ensuring that the flow front on one side cannot move significantly faster than another (proven for the structures investigated, theorised for larger layup structures). This builds in some ’safety factor’ into an infusion, meaning that if any issues occur on one side of the balsa, the flow of resin from the other side would ensure that the flow would still reach both sides of the core.Whilst distribution grooves do move resin at the scale of one block, the main distribution mechanism found within this body of work is actually the cracks and gaps between the balsa blocks, with the channels acting as a secondary distribution method at the scale of a block. From this, two scales of the flow are defined, macroscale; governs the bulk of the flow at a larger scale, moving along the cracks and gaps between blocks, and then the second mesoscale, local scale where the grooves distribute resin at the scale of one block with the flow often not just being in the shortest path from inlet to outlet.
Oliver Morgan-Clague (Wed,) studied this question.