Reliable assessment of the fatigue resistance of asphalt pavement with a long-term service is critically crucial for the rational formulation of original pavement utilization strategies in reconstruction and expansion projects. Currently, the pavement performance evaluation indicators are mainly used to guide pavement preventive maintenance, and its applicability in reconstruction and expansion projects of the freeway is limited. This paper aims to propose an evaluation method of fatigue resistance of asphalt concrete utilizing semi-circular bending (SCB) tests and stereo digital image correlation (stereo-DIC) techniques. A total of 27 asphalt concrete cores were drilled from the three freeways (K84, K124, and K165) with a service life of more than 20 years, and the SCB specimens were produced to conduct the SCB fracture and fatigue tests. During the SCB test, the stereo-DIC technique was employed to monitor the evolution process of the strain distribution and crack length for the specimens. K-dimension tree neighbor-searching algorithm (K-d tree algorithm) was used to effectively measure the change of crack length corresponding to each fatigue load cycle. Meanwhile, the strain threshold of asphalt concrete crack initiation was determined by the bilinear softening cohesive zone model (CZM) to ensure the accuracy of the crack length calculated by the K-d tree algorithm. Furthermore, the relationship between crack growth rate and stress intensity, which was used to fit the Paris law parameters, was determined. The CZM and DIC results indicated that the strain threshold of asphalt concrete crack should be set as 2000 με when using the K-d tree algorithm to determine the crack length. With the stress ratio increase, the Paris law parameter A increased wavily, and the parameter n decreased steadily, while the threshold of the stress intensity factor increased steadily. The Paris law master curves could characterize the fatigue performance of various road sections at a wide load range. The residual fatigue life of K84, K124, K165 the road sections were 2.13E + 08, 3.57E + 08, and 1.02E + 07, respectively.

@en

Geothermal energy with abundance and large quantity can partially cover building heating/cooling loads and promote the carbon-neutrality transitions. Shallow geothermal ventilation (SGV) system, with a little initial investment cost, is one of promising technologies to partly replace the conventional air-conditioning system for air pre-cooling/pre-heating. This paper reviews applications of SGV system for improving thermal performance over latest two decades, which mainly includes the reclassification of SGV system, coupling with other advanced energy-saving technologies, application potentials for building cooling/heating under various weather conditions. Heat transfer mechanism and mathematical modelling techniques have been reviewed, together with in-depth analysis on current research trends, existing limitations, and recommendations of SGV system. Phase change materials, with considerable latent energy density, can stabilize the thermal performance with high reliability. The review identifies that optimization designs and advanced approaches need to be investigated to address the existing urgent issues of SGV system (e.g., large land occupation, difficulty in centralized collection of condensate water timely for horizontal buried pipe, bacteria growth, polluted supply air, and high construction cost for vertical buried pipe). A plenty of studies show that the SGV system could greatly expand the application scope and improve system energy efficiency by combining with other energy-saving technologies. This paper will provide some guidelines for the scientific researchers and engineers to keep track on recent advancements and research trends of SGV system for the building thermal performance enhancement and pave path for future research works.

@en

In personnel recognition based on radar, significant research exists on statistical features extracted from the micro-Doppler signatures, whereas research considering other domains and information such as phase is less developed. This paper presents the use of deep learning methods to integrate both phase and magnitude features from range profiles and spectrogram. The temporal features of both domains are separately extracted using a stack of Long Short Term Memory (LSTM) layers. Then, the extracted features are aggregated in the corresponding domains and pass through a series of dense layers with SoftMax classifier. Finally, the information from the two domains is fused with a soft fusion approach to improve the performance further. Preliminary results show that the proposed network with soft fusion can achieve 85.5% accuracy in personnel recognition with six subjects@en