A dataset from a hydraulically actuated forestry crane for data-driven stress estimation (PATU655Stress)

Hydraulic forestry cranes operate under continuously varying loads, where coupling between hydraulic actuation, mechanical motion, and structural response affects performance, safety, and fatigue life. Accurate estimation of internal states and structural stresses remains challenging due to nonlinear hydraulics, structural flexibility, and limited sensing, and publicly available experimental datasets from real machines are scarce. This paper presents PatuCrane655, an experimental dataset collected from a real PATU 655 forestry crane. The dataset includes time-synchronized measurements of joystick inputs, hydraulic pressures, actuator displacements, boom angles, and strain-gauge signals at structurally critical locations, recorded under multiple payloads, hydraulic settings, and crane configurations. To demonstrate its applicability, a baseline sequence-based LSTM model was applied to predict structural strains from measured inputs, showing that the dataset supports accurate data-driven stress estimation. The dataset is intended to support data-driven modelling, state estimation, and structural stress prediction for hydraulic cranes.