Quality Control of Thermally Modified Western Hemlock Wood Using Near-Infrared Spectroscopy and Explainable Machine Learning.

The quality control of thermally modified wood and identifying heat treatment intensity using nondestructive testing methods are critical tasks. This study used near-infrared (NIR) spectroscopy and machine learning modeling to classify thermally modified wood. NIR spectra were collected from the surfaces of untreated and thermally treated (at 170 °C, 212 °C, and 230 °C) western hemlock samples. An explainable machine learning approach was practiced using a TreeNet gradient boosting machine. No dimensionality reduction was performed to better explain the feature ranking results obtained from the model and provide insight into the critical wavelengths contributing to the performance of classification models. NIR spectra in the ranges of 1100-2500 nm, 1400-2500 nm, and 1700-2500 nm were fed into the TreeNet model, which resulted in classification accuracy values (test data) of 94.35%, 89.29%, and 84.52%, respectively. Feature ranking analysis revealed that when using the range of 1100-2500 nm, the changes in wood color resulted in the highest variation in NIR reflectance amongst treatments. As a result, associated features were given higher importance by TreeNet. Limiting the wavelength range increased the significance of features related to water or wood chemistry; however, these predictive models were not as accurate as the one benefiting from the impact of wood color change on the NIR spectra. The developed framework could be applied to different applications in which NIR spectra are used for wood characterization and quality control to provide improved insights into selected NIR wavelengths when developing a machine learning model.

The Role of Drying Schedule and Conditioning in Moisture Uniformity in Wood: A Machine Learning Approach.

Monitoring the moisture content (MC) of wood and avoiding large MC variation is a crucial task as a large moisture spread after drying significantly devalues the product, especially in species with high green MC spread. Therefore, this research aims to optimize kiln-drying and provides a predictive approach to estimate and classify target timber moisture, using a gradient-boosting machine learning model. Inputs include three wood attributes (initial moisture, initial weight, and basic density) and three drying parameters (schedule, conditioning, and post-storage). Results show that initial weight has the highest correlation with the final moisture and possesses the highest relative importance in both predictive and classifier models. This model demonstrated a drop in training accuracy after removing schedule, conditioning, and post-storage from inputs, emphasizing that the drying parameters are significant in the robustness of the model. However, the regression-based model failed to satisfactorily predict the moisture after kiln-drying. In contrast, the classifying model is capable of classifying dried wood into acceptable, over-, and under-dried groups, which could apply to timber pre- and post-sorting. Overall, the gradient-boosting model successfully classified the moisture in kiln-dried western hemlock timber.

Prediction of Mechanical Properties of Artificially Weathered Wood by Color Change and Machine Learning.

Color parameters were used in this study to develop a machine learning model for predicting the mechanical properties of artificially weathered fir, alder, oak, and poplar wood. A CIELAB color measuring system was employed to study the color changes in wood samples. The color parameters were fed into a decision tree model for predicting the MOE and MOR values of the wood samples. The results indicated a reduction in the mechanical properties of the samples, where fir and alder were the most and least degraded wood under weathering conditions, respectively. The mechanical degradation was correlated with the color change, where the most resistant wood to color change exhibited less reduction in the mechanical properties. The predictive machine learning model estimated the MOE and MOR values with a maximum R2 of 0.87 and 0.88, respectively. Thus, variations in the color parameters of wood can be considered informative features linked to the mechanical properties of small-sized and clear wood. Further research could study the effectiveness of the model when analyzing large-sized timber.

Deformation mechanics of self-expanding venous stents: Modelling and experiments.

Deformation properties of venous stents based on braided design, chevron design, Z design, and diamond design are compared using in vitro experiments coupled with analytical and finite element modelling. Their suitability for deployment in different clinical contexts is assessed based on their deformation characteristics. Self-expanding stainless steel stents possess superior collapse resistance compared to Nitinol stents. Consequently, they may be more reliable to treat diseases like May-Thurner syndrome in which resistance against a concentrated (pinching) force applied on the stent is needed to prevent collapse. Braided design applies a larger radial pressure particularly for vessels of diameter smaller than 75% of its nominal diameter, making it suitable for a long lesion with high recoil. Z design has the least foreshortening, which aids in accurate deployment. Nitinol stents are more compliant than their stainless steel counterparts, which indicates their suitability in veins. The semi-analytical method presented can aid in rapid assessment of topology governed deformation characteristics of stents and their design optimization.

A Mesoscopic Analytical Model to Predict the Onset of Wrinkling in Plain Woven Preforms under Bias Extension Shear Deformation.

A mesoscopic analytical model of wrinkling of Plain-Woven Composite Preforms (PWCPs) under the bias extension test is presented, based on a new instability analysis. The analysis is aimed to facilitate a better understanding of the nature of wrinkle formation in woven fabrics caused by large in-plane shear, while it accounts for the effect of fabric and process parameters on the onset of wrinkling. To this end, the mechanism of wrinkle formation in PWCPs in mesoscale is simplified and an equivalent structure composed of bars and different types of springs is proposed, mimicking the behavior of a representative PWCP element at the post-locking state. The parameters of this equivalent structure are derived based on geometric and mechanical characteristics of the PWCP. The principle of minimum total potential energy is employed to formluate the model, and experimental validation is carried out to reveal the effectiveness of the derived wrinkling prediction equation.

Modeling and simulation of linear and nonlinear MEMS scale electromagnetic energy harvesters for random vibration environments.

The simulation results for electromagnetic energy harvesters (EMEHs) under broad band stationary Gaussian random excitations indicate the importance of both a high transformation factor and a high mechanical quality factor to achieve favourable mean power, mean square load voltage, and output spectral density. The optimum load is different for random vibrations and for sinusoidal vibration. Reducing the total damping ratio under band-limited random excitation yields a higher mean square load voltage. Reduced bandwidth resulting from decreased mechanical damping can be compensated by increasing the electrical damping (transformation factor) leading to a higher mean square load voltage and power. Nonlinear EMEHs with a Duffing spring and with linear plus cubic damping are modeled using the method of statistical linearization. These nonlinear EMEHs exhibit approximately linear behaviour under low levels of broadband stationary Gaussian random vibration; however, at higher levels of such excitation the central (resonant) frequency of the spectral density of the output voltage shifts due to the increased nonlinear stiffness and the bandwidth broadens slightly. Nonlinear EMEHs exhibit lower maximum output voltage and central frequency of the spectral density with nonlinear damping compared to linear damping. Stronger nonlinear damping yields broader bandwidths at stable resonant frequency.

A biomechanical model to assess the contribution of pelvic musculature weakness to the development of stress urinary incontinence.

A biomechanical model of the female pelvic support system was developed to explore the contribution of pelvic floor muscle defect to the development of stress urinary incontinence (SUI). From a pool of 135 patients, clinical data of 26 patients with pelvic muscular defect were used in modelling. The model was employed to estimate the parameters that describe the stiffness properties of the vaginal wall and ligament tissues for individual patients. The parameters were then implemented into the model to evaluate for each patient the impact of pelvic muscular defect on the vaginal apex support and the bladder neck support, a factor that relates to the onset of SUI. For the modelling analysis, the compromise of pelvic muscular support was demonstrated to contribute to vaginal apex prolapse and bladder neck prolapse, a condition commonly seen in SUI patients, while simulated conditions of restored muscular support were shown to help re-establish both vaginal apex and bladder neck supports. The findings illustrate the significance of pelvic muscle strength to vaginal support and urinary continence; therefore, the clinical recommendation of pelvic muscle strengthening, such as Kegel exercises, has been shown to be an effective treatment for patients with SUI symptoms.