A Hybrid Convolutional and Recurrent Neural Network for Multi-Sensor Pile Damage Detection with Time Series.

Machine learning (ML) algorithms are increasingly applied to structure health monitoring (SHM) problems. However, their application to pile damage detection (PDD) is hindered by the complexity of the problem. A novel multi-sensor pile damage detection (MSPDD) method is proposed in this paper to extend the application of ML algorithms in the automatic identification of PDD. The time-series signals collected by multiple sensors during the pile integrity test are first processed by the traveling wave decomposition (TWD) theory and are then input into a hybrid one-dimensional (1D) convolutional and recurrent neural network. The hybrid neural network can achieve the automatic multi-task identification of pile damage detection based on the time series of MSPDD results. Finally, the analytical solution-based sample set is utilized to evaluate the performance of the proposed hybrid model. The outputs of the multi-task learning framework can provide a detailed description of the actual pile quality and provide strong support for the classification of pile quality as well.

Pile Damage Detection Using Machine Learning with the Multipoint Traveling Wave Decomposition Method.

The in-hole multipoint traveling wave decomposition (MPTWD) method is developed for detecting and characterizing the damage of cast in situ reinforced concrete (RC) piles. Compared with the results of MPTWD, the results of the in-hole MPTWD reconstruction technique are found ideal for evaluating the lower-part pile integrity and are further utilized to establish a data-driven machine-learning framework to detect and quantify the degree of damage. Considering the relatively small number of field test samples of the in-hole MPTWD method at this stage, an analytical solution is employed to generate sufficient samples to verify the feasibility and optimize the performance of the machine learning modeling framework. Two types of features extracted by the distributed sampling and statistical and signal processing techniques are applied to three machine-learning classifiers, i.e., logistic regression (LR), extreme gradient boosting (XGBoost) and multilayer perceptron (MLP). The performance of the data-driven machine-learning framework is then evaluated through a specific case study. The results demonstrate that all three classifiers perform better when employing the statistical and signal processing techniques, and the total of 24 extracted features are sufficient for the machine-learning algorithms.

Torsional Low-Strain Test for Nondestructive Integrity Examination of Existing High-Pile Foundation.

Low-strain tests are widely utilized as a nondestructive approach to assess the integrity of newly piled foundations. So far, the examination of existing pile foundations is becoming an indispensable protocol for pile recycling or post-disaster safety assessment. However, the present low-strain test is not capable of testing existing pile foundations. In this paper, the torsional low-strain test (TLST) is proposed to overcome this drawback. Both the upward and downward waves are considered in the TLST wave propagation model established in this paper so that a firm theoretical basis is grounded for the test signal interpretations. A concise semi-analytical solution is derived and its rationality is verified by comparisons with the existing solutions for newly piled foundations and the finite element results. The main conclusions of this study can be drawn as follows: (1). by placing the sensors where the incident wave is applied, the number of reflected signals can be minimized; (2). the defects can be more evidently identified if the incident wave/sensors are input/installed close to the superstructure/pile head.

Mono- and co-substrate utilization kinetics using mono- and co-culture of Clostridium beijerinckii and Clostridium saccharoperbutylacetonicum.

The effect of co-culturing C. beijerinckii and C. saccharoperbutylacetonicum for H2 production using mono- and co-substrates of glucose, starch, and cellulose was assessed. Monod kinetic parameters (K, maximum specific substrate utilization rate; and Ks, half-saturation constant) of the C. beijerinckii, C. saccharoperbutylacetonicum, and the co-culture were determined. Co-cultures utilizing glucose competed for the substrate, but showed enhancement for utilizing starch. The maximum values for K on glucose and starch were 0.48g substrate/gVSS.h achieved by C. saccharoperbutylacetonicum mono-culture and 0.39g substrate/gVSS.h achieved by the co-culture, respectively. The average Ks for all mono- and co-culture experiments was 0.93±0.03g/L. Acetate, butyrate, and propionate were the main fermentation products for all experiments. Maximum H2 production yields on glucose (2.69mol/molglucose) and starch (1.07mol/molhexose) were achieved by C. beijerinckii and C. saccharoperbutylacetonicum mono-cultures, respectively; however, neither culture was able to degrade cellulose as a mono-substrate.

Comparative assessment of single-stage and two-stage anaerobic digestion for the treatment of thin stillage.

A comparative evaluation of single-stage and two-stage anaerobic digestion processes for biomethane and biohydrogen production using thin stillage was performed to assess the impact of separating the acidogenic and methanogenic stages on anaerobic digestion. Thin stillage, the main by-product from ethanol production, was characterized by high total chemical oxygen demand (TCOD) of 122 g/L and total volatile fatty acids (TVFAs) of 12 g/L. A maximum methane yield of 0.33 L CH(4)/gCOD(added) (STP) was achieved in the two-stage process while a single-stage process achieved a maximum yield of only 0.26 L CH(4)/gCOD(added) (STP). The separation of acidification stage increased the TVFAs to TCOD ratio from 10% in the raw thin stillage to 54% due to the conversion of carbohydrates into hydrogen and VFAs. Comparison of the two processes based on energy outcome revealed that an increase of 18.5% in the total energy yield was achieved using two-stage anaerobic digestion.