Automatic Procedure for Determining Control Parameters for Ship-to-Shore Cranes

Abstract: The objective of this thesis has been to create a tuning functionality that can determine the control parameters for a crane that loads and unloads containers from ships. To accomplish this a model of the crane has been developed in MapleSIM and exported into MATLAB’s Simulink. In MATLAB cost functions have been developed and evaluated to later be used in optimisation algorithms to find the optimal parameters. There are two sets of control parameters that the script needs to determine, the first is composed of 6 subsets of 4 co-dependent parameters that need to be optimised together. The other set consists of 66 parameters that are independent of each other. A few different optimisation methods have been considered, but for the first set a particle swarm optimisation was used and for the second set, due to the cost function, a form of binary search was possible to use. As concluding results, we found that the underlying model of the crane was not accurate enough to determine the first set of control parameters. The second set of control parameters are similar enough to the real values and can possibly be used on a real crane.