The fierce competition in the machine tools market provokes a demand for the fast, accurate, and reliable production of machined components. This can be achieved using virtual prototyping technology which provides a computer model of the machine tool. To build an accurate virtual model of the machine tool, dynamic properties of all parts of the structure, including joints, should be identified. Mechanical joints have considerable effects on the dynamics of machine tools and ignoring them results in deviations between the virtual model and the physical structure analog.

This study proposes new analytical and experimental methods based on modal analysis and sensor development for joint dynamics identification in bolted lap joints. An analytical joint identification (AJI) technique, which employs the modal parameters of the structure, is developed using Euler-Bernoulli theory for determining the joint properties in transverse direction. The obtained results are compared with those of the inverse receptance coupling (IRC) approach. Numerical and experimental investigations verify the accuracy of the proposed methods.

A new experimental approach is proposed to determine the damping of bolted lap joints in translational and torsional directions. Because of the complex nature of the joint, this element is isolated through the addition of a mechanical resonator to the bolted structure and the frequency response function (FRF) of the combined system is used for joint damping identification. This approach overcomes difficulties associated with microslip analysis of bolted lap joints. The method is verified after comparing the results with those of the hysteresis loop approach.

In addition, a novel nanocomposite-based sensor is developed for strain and force measurements that can be used for identifying joint dynamics. Exhibiting both piezoelectric and piezoresistive properties, the developed sensors are capable of measurements over a wide frequency range. To improve the accuracy and frequency bandwidth of the sensor, the piezoresistive and piezoelectric signals are fused. A 3D random walk model and a 2D finite element model are used to elucidate the piezoresistive and piezoelectric behaviour of the sensor, respectively. The experimental results show that the developed sensor is capable of measuring both static and high frequency loads through a fused piezoelectric/piezoresistive output, which is a unique feature of the sensor.

The proposed nanocomposite sensor is implemented in the joint interface and used for contact force measurement and joint dynamics identification through two different experimental approaches. The first approach uses the contact force and displacement data to build the hysteresis loop and subsequently extract the joint parameters. The second method employs an efficient calculation algorithm to identify the stick-slip transition in the joint interface and then obtain the energy loss and joint properties. The obtained results from both approaches are compared to each other to investigate the accuracy of these methods in determining the joint parameters.

The results of the joint dynamics identification techniques can be employed in building a database for bolted lap joints. This database can then be used in the design process in order to increase the correlation between the virtual models and actual structures.