The accurate short-term traffic flow prediction can provide timely and condition information which help one to make travel decision mitigate the jam. Deep learning (DL) provides a new paradigm for analysis of big data generated by urban daily traffic. In this paper, we propose novel end-to-end deep architecture consists two modules. We combine convolution LSTM form Conv-LSTM module extract spatial-temporal information. Furthermore, Bi-directional is also adopted analyze historical point get...

Due to factors such as the availability of assembly equipment, technology, and site management level, prefabricated building construction safety accidents often occur. To ensure buildings effectively reduce accident rate, concept resilience is introduced into buildings. Based on absorption capacity, adaptation recovery optimization a comprehensive evaluation index system for established. By combining prior knowledge with structural learning parameter learning, dynamic Bayesian network (DBN)...

Prefabricated buildings that are more environmentally friendly have been vigorously promoted by the Chinese government because of reduced waste and carbon emissions during construction process. Most processes prefabricated completed in component factory, but safety risks hoisting process cannot be ignored. In this paper, initial framework a Bayesian Network (BN) is obtained from combination improved Human Factors Analysis Classification System Model (HFACS) BN. The similarity aggregation...

Construction accident investigation reports contain critical information, but extracting useful insights from the voluminous Chinese text is challenging. Traditional methods rely on expert judgment, which leads to time-consuming and potentially inaccurate results. To overcome this problem, we propose a novel approach that combines mining techniques latent Dirichlet allocation (LDA) models analyze standardized in construction industry. The proposed method integrates an information entropy...

Glyptostrobus pensilis is a critically endangered living fossil plant species of the Mesozoic era, with high scientific research and economic value. The aim this study was to assess impact climate change on potential habitat area G. in East Asia. MaxEnt (maximum entropy) model optimized by ENMeval data package used simulate distribution habitats since last interglacial period (LIG, 120–140 ka). results showed that has prediction accuracy under receiver operating characteristic curve (AUC)...

Abstract Background Predicting the potential habitat of Phytolacca americana, a high-risk invasive species, can help provide scientific basis for its quarantine and control strategies. Using optimized MaxEnt model, we applied latest climate data, CMIP6, to predict distribution risk zones their change patterns P . americana under current future (SSP126, SSP245, SSP585) conditions, followed by invasion analysis. Results The predictions model based on R language optimization were highly...

Current research on personal protective equipment (PPE) detection has mainly focused hard hats, overlooking the of reflective clothing. Therefore, this study aims to address gap comprehensively. We achieve by creating a novel dataset using semi-automatic labeling techniques and enhancing YOLOv5 model. The consists four categories, assessing presence hats Additionally, we introduce an attention mechanism improved loss function tackle challenges related detecting clothing overlapping frames....

As an efficient measure to protect the environment and reduce energy consumption, promoting prefabricated buildings can lead sustainable development of world economy long-term economic benefits for society. Nevertheless, China, it is immature because high cost difficulties in promotion. The government’s strategies management are crucial building policies. This paper establishes a quadripartite evolutionary game model composed construction units, property developers, homebuyers, government...

In order to achieve a comprehensive evaluation of the environmental impact ecological restoration projects (ERP) under current destruction and coastal areas, this paper takes into account positive negative indicators on environment; analyzes benefits ERP; establishes index system for marine ERP from ecological, economic, social perspectives through DPSIR model. On basis, cloud model Monte Carlo simulation are used obtain assessment grade construction period, short-term operation, long-term...

As countries around the world pay more and attention to sustainable development of construction industry, prefabricated building model has become best type achieve energy conservation emission reduction. However, entails higher technical requirements, workers involved in must be trained reduce risks. For China, where demographic dividend is gradually disappearing, how quickly promote industrializing process an urgent issue. This research focuses on training management building. First, facial...

Scholars and engineers have increasingly focused on the safety of steel-structure buildings. An accurate analysis can substantially reduce collapse probability these This paper proposes a novel risk model to assess The vector entropy method weight clustering were used improve controlled interval memory (CIM) model. proposed has advantages straightforward modeling approach, strong adaptability, logical relationships. new strategy improves reliability stability CIM results when maximum...

Addressing the challenges of intelligent decision support within traditional management approaches, this study presents a novel conceptual framework for enhancing sustainability in construction behavioral safety management. By integrating Bayesian network (BN) modeling and ontology, our enables robust decision-making fosters knowledge sharing. Key to approach is encoding storage BN-modeled properties facilitating formal representation knowledge. Leveraging SWRL rules reasoning judgment,...

A great number of renewable energy sources and nonlinear loads are integrated into the distribution systems through power electronic converters, which leads to serious harmonic distortions. Correctly determining location is primary task manage pollution users develop a strategy for suppression on system level. Existing localization methods leverage time-domain waveforms voltage currents across each feeder. Meanwhile, an accurate knowledge network topology line impedances also required....

To resolve the issue of difficultly in effectively balancing output performance improvement, cost reduction, and efficiency improvement a medium-voltage modular multilevel converter (MMC), novel MMC (NMMC) topology based on high- low-frequency hybrid modulation is proposed this study. Each arm NMMC contains high-frequency sub-module composed heterogeneous cross-connect module (HCCM) N − 1 sub-modules half-bridge converters. The bridge HCCM study adopts SiC MOSFET devices, while commutation...

With the rapid development of distribution networks and increasing demand for electricity, pressure power supply medium- low-voltage (M&LVDNs) is increasingly significant, especially considering large scale customers at (LV) level. In this paper, an outage sequence optimization method (LVDNs) that considers importance users proposed. The aims to develop optimal strategy LV in case medium-voltage (MV) failure events. First, a multi-dimensional indicator system constructed, are ranked...

Youth Park community is the first international safe in Mainland China, and its success has led to widespread adoption of communities across country. The Safe Community was analyzed empirically terms community's history, ideas, procedures, propulsion system safety promotion programs. Its experiences are expected serve as an example for other communities.