The paper presents kinematic characteristics of the double 4-link coupler system, used in actual powertrain of low-floor trams (NGT6-Kr). The spatial kinematic model of the couplings was formulated assuming ideal joints and rigid members. The constraints equations of the mechanism were solved iteratively and differentiated to obtain the Jacobian matrix. The mobility and singularity analysis of the coupler mechanism was performed on the basis of the Jacobian matrix.

Kinematic characteristics of the single and double coupler system were analyzed for gross angular and linear axle displacements (misalignments), taking the advantage of the fully nonlinear model. The coupling system was evaluated based on criteria describing homokinetics, balancing and clearance demands, and angular displacements in the joints. These criteria were determined for different design parameters like: coupler proportions, platform shift and angle, middle shaft length.

Variation in powertrain parameters caused by dimensioning, manufacturing and assembly inaccuracies may prevent model-based virtual sensors from representing physical powertrains accurately. Data-driven virtual sensors employing machine learning models offer a solution for including variations in the powertrain parameters. These variations can be efficiently included in the training of the virtual sensor through simulation. The trained model can then be theoretically applied to real systems via transfer learning, allowing a data-driven virtual sensor to be trained without the notoriously labour-intensive step of gathering data from a real powertrain. This research presents a training procedure for a data-driven virtual sensor. The virtual sensor was made for a powertrain consisting of multiple shafts, couplings and gears. The training procedure generalizes the virtual sensor for a single powertrain with variations corresponding to the aforementioned inaccuracies. The training procedure includes parameter randomization and random excitation. That is, the data-driven virtual sensor was trained using data from multiple different powertrain instances, representing roughly the same powertrain. The virtual sensor trained using multiple instances of a simulated powertrain was accurate at estimating rotating speeds and torque of the loaded shaft of multiple simulated test powertrains. The estimates were computed from the rotating speeds and torque at the motor shaft of the powertrain. This research gives excellent grounds for further studies towards simulation-to-reality transfer learning, in which a virtual sensor is trained with simulated data and then applied to a real system.