Search

Geospatial artificial intelligence (GeoAI) has emerged as a subfield of GIScience that uses artificial intelligence approaches and machine learning techniques for geographic knowledge discovery. The non-regularity of data structures has recently led to different variants of graph neural networks in the field of computer science, with graph convolutional neural networks being one of the most prominent that operate on non-euclidean structured data where the numbers of nodes connections vary and the nodes are unordered. These networks use graph convolution – commonly known as filters or kernels – in place of general matrix multiplication in at least one of their layers. This paper suggests spatial regression graph convolutional neural networks (SRGCNNs) as a deep learning paradigm that is capable of handling a wide range of geographical tasks where multivariate spatial data needs modeling and prediction. The feasibility of SRGCNNs lies in the feature propagation mechanisms, the spatial locality nature, and a semi-supervised training strategy. In the experiments, this paper demonstrates the operation of SRGCNNs with social media check-in data in Beijing and house price data in San Diego. The results indicate that a well-trained SRGCNN model is capable of learning from samples and performing reasonable predictions for unobserved locations. The paper also presents the effectiveness of incorporating the idea of geographically weighted regression for handling heterogeneity between locations in the model approach. Compared to conventional spatial regression approaches, SRGCNN-based models tend to generate much more accurate and stable results, especially when the sampling ratio is low. This study offers to bridge the methodological gap between graph deep learning and spatial regression analytics. The proposed idea serves as an example to illustrate how spatial analytics can be combined with state-of-the-art deep learning models, and to enlighten future research at the front of GeoAI. [ABSTRACT FROM AUTHOR]

Microvibration induced by natural disturbance and human activities has an adverse effect on the operation of the large-scale and ultraprecise facilities in the world. Under such circumstances, a passive vibration control method is generally deployed for such vibration-sensitive facilities, taking the High Energy Photo Source (HEPS) in Beijing as an example, a 3 m-thick mass concrete layer forming a ring foundation was cast at the facility, where a 1 m-thick reinforced concrete slab (RC slab) lies. Since microvibration control plays a crucial role in the operation of such large-scale scientific and ultraprecise facilities and few studies have been reported for large-scale concrete layer as antimicrovibration devices, this paper presents four field tests in Beijing, China, to evaluate the vibration control capacity of a mass concrete layer. Based on a large number of field tests, the effect of applying the concrete layer is discussed, and a reference is provided for the construction of similar facilities. The vibration signals, generated by shock excitation and ambient excitation, are measured through a highly sensitive and high-accuracy vibration acquisition system. It is concluded that the existence of the 1 m-thick RC slab has little influence on the microvibration signal frequency distribution in the vertical direction and that the signals from the concrete layer and subsoil differ by approximately 10 Hz in the vertical direction while differing by approximately 5 Hz in the horizontal direction. The microvibration control ability of the concrete layer is favorable in a higher frequency band over 20 ~ 30 Hz and more than 50% attenuation can be gained through the concrete layer; however, the microvibration control ability is not significant below 20 ~ 30 Hz. The vibration levels across different heights of the concrete layer section are the same. To prevent adverse vibration disturbance below 20 ~ 30 Hz, it is suggested that the traffic and road surface conditions should be taken into consideration when choosing the construction location. In addition, a long-term monitoring shows that 75% vibration energy at the site is firmly related to the construction activities which are approximately 1.4 km from the site. [ABSTRACT FROM AUTHOR]

To study urban air quality, this paper proposes a novel categorical time series model, which is based on a linear combination of bounded Poisson distribution and discrete distribution to describe the dynamic and systemic features of air quality, respectively. Daily air quality level data of three major cities in China, including Beijing, Shanghai and Guangzhou, are analyzed. It is concluded that the air quality in Beijing is the worst among the three cities but is gradually improving, and its dynamics is also the most pronounced. Theoretically, the design of our model increases the flexibility of the probabilistic structure while ensuring a dynamic feedback mechanism without high computational stress. We estimate the parameters through an adaptive Bayesian Markov chain Monte Carlo sampling scheme and show the satisfactory finite sample performance of the model through simulation studies. [ABSTRACT FROM AUTHOR]

The prediction and evaluation of environmental vibrations generated by traffic loads have been playing more and more important role during the planning and designing of high-quality traffic system. From the traffic source to sensitive building, the subsoil is the one of key propagation media, so the determination of soil dynamic parameters is very crucial in studying the transmitting and attenuation of environmental vibration. Due to the spatial variability of natural soil layers, the soil dynamic parameters are actually not certain but probabilistic with stochastic characteristics. However, most current inversion methods of soil parameters cannot consider the uncertainty. In this paper, an efficient inversion method is proposed to obtain the random distribution of soil dynamic parameters based on the combination of field experiment, forward soil modelling, and Bayesian theory. Take the shear modulus and damping ratio of subsoil as the typical parameters, the experimental dispersion curve and spatial attenuation curve can be obtained by field experiment, while the corresponding theoretical ones can be calculated by the thin-layered soil model with perfectly matched layer. The combining of experimental and theoretical curves generates the likelihood function. A prior probability distribution model is firstly established, and subsequently the posterior probability distribution model is deduced by genetic algorithm, Monte Carlo-Markov chain algorithm, and Bayesian theory. In order to verify the proposed inversion method, an on-site experiment based on Multichannel Analysis of Surface Waves(MASW) was conducted in an open field in Beijing. The results show the mean values of inversed soil parameters were close to the determined parameters provided by geological exploration. Compared to the traditional determined inversion, the proposed stochastic inversion method can consider the variety of soil parameters within small depth by providing probability distribution, mean and root-mean-square error, which has greater applicability in the further probabilistic prediction of environmental vibration. • Bayesian theory and Monte Carlo Markov chain algorithm can be well applied to soil parameter inversion. • The proposed inversion method can efficiently reflect the random variation of soil parameters. • The Metropolis Hastings algorithm is introduced to improve the computational efficiency. • The method is based on non-intrusive data and has more significant engineering value than traditional geological survey. [ABSTRACT FROM AUTHOR]

The district heating system (DHS) is one of the essential carriers for coordinating renewable energy with fossil energy and achieving flexible energy consumption. Considering the uncertainty of both the renewable energy units' power and the consumers' aggregate loads impact the transport process of the DHS, the thermal characteristics of the DHS transportation and the node temperature response under the source-load uncertainty need to be quantified and analyzed. This paper proposed a method to calculate the nodes' temperature dynamic response of the DHS under multiple source-load uncertainty scenarios. The prediction and combination methods of supply and demand probability intervals of source and load nodes under varying credibility levels were investigated. Additionally, the effects of source-load uncertainty boundary conditions on the DHS transport process were analyzed. The temperature response of the nodes was calculated based on the established dynamic DHS model under multiple source-load uncertainty scenarios. A DHS in Beijing was selected for case validation and analysis, and it has 90 nodes and 109 pipes. The simulation results showed that the relative average error of node flow rate in the dynamic DHS model was 3.02%, and the average difference in node temperature was 1 °C. The probability distribution semi-analytical method and the Bayesian credible interval calculation method can accurately describe the uncertainty on both source and load sides. The models and algorithms proposed in this paper could effectively quantify the nodes' dynamic temperature response under the source-load uncertainty. • Proposed uncertainty quantification models for the source and load nodes of DHS. • Developed a novel credible interval calculation method for nodes' thermal power. • Quantified nodes' temperature dynamic response considering source-load uncertainty. [ABSTRACT FROM AUTHOR]

The security travel of freight vehicles is of high societal concern and is the key issue for urban managers to effectively supervise and assess the possible social security risks. With continuous improvements in motion-based technology, the trajectories of freight vehicles are readily available, whose unusual changes may indicate hidden urban risks. Moreover, the increasing high spatial and temporal resolution of trajectories provides the opportunity for the real-time recognition of the abnormal or risky vehicle motion. However, the existing researches mainly focus on the spatial anomaly detection, and there are few researches on the real-time temporal anomaly detection. In this paper, a grid-based algorithm, which combines the spatial and temporal anomaly detection, is proposed for tracing the risk of urban freight vehicles trajectory by considering local temporal window. The travel time probability distribution of vehicle historical trajectory is analyzed to meet the time complexity requirements of real-time anomaly calculation. The developed methodology is applied to a case study in Beijing to demonstrate its accuracy and effectiveness. [ABSTRACT FROM AUTHOR]

Turbulent flow over complex urban morphology poses challenges to utilization of distributed wind energy in cities. In this paper, turbulence intensity (TI), which is often used to guide wind turbine design and evaluate wind power production, is analyzed by high-frequency observations on a 325-m tower located in the downtown of Beijing, China. The results show that the characteristics of TI, including the diurnal variation, the statistical distribution, and the relations between TI and wind speeds near the ground are significantly different from these at higher heights. It is also found that the normal turbulence model (NTM) with the parameters recommended by the IEC standard cannot fit the data well. The new NTM parameters are given by the least-square fittings. Besides, a three-parameter model is found to fit the data well and performs slightly better than the NTM with the least-square fitting parameters. This research indicates that the NTM parameters in the IEC standard should be re-evaluated in urban areas, and different turbulence model parameters are recommended for higher and lower hub heights. [ABSTRACT FROM AUTHOR]

For prediction of urban remote sensing surface temperature, cloud, cloud shadow and snow contamination lead to the failure of surface temperature inversion and vegetation‐related index calculation. A time series prediction framework of urban surface temperature under cloud interference is proposed in this paper. This is helpful to solve the problem of the impact of data loss on surface temperature prediction. Spatial and temporal variation trends of surface temperature and vegetation index are analyzed using Landsat 7/8 remote sensing data of 2010 to 2019 from Beijing. The geographically weighed regression (GWR) method is used to realize the simulation of surface temperature based on the current date. The deep learning prediction network based on convolution and long short‐term memory (LSTM) networks was constructed to predict the spatial distribution of surface temperature on the next observation date. The time series analysis shows that the NDBI is less than −0.2, which indicates that there may be cloud contamination. The land surface temperature (LST) modeling results show that the precision of estimation using GWR method on impervious surface and water bodies is superior compared to the vegetation area. For LST prediction using deep learning methods, the result of the prediction on surface temperature space distribution was relatively good. The purpose of this study is to make up for the missing data affected by cloud, snow, and other interference factors, and to be applied to the prediction of the spatial and temporal distributions of LST. This article is categorized under:Technologies > Machine Learning [ABSTRACT FROM AUTHOR]

• Spatiotemporal usage patterns of station-free shared bikes are explored. • A topic-based two-stage algorithm is proposed to discover city functional regions. • Region functions are labeled by static POI data and dynamic mobility patterns. Station-free shared bike (SFSB) is a new travel mode that shared bikes are allowed to park in any proper places. It implies that the users are more likely to park the SFSB as close as their destinations. This advantage makes the SFSB data to be an ideal source to understand the land-use distribution. Inspired by the idea in text mining, this paper proposes a topic-based two-stage SFSB data mining algorithm to understand the SFSB user's travel behavior and to discover city functional regions. A region, a function and human mobility patterns are treated as a document, a topic and words, respectively. Then, a region is represented by a distribution of functions, and a function is featured by a distribution of mobility patterns. The point-of-interest data is combined to annotate the clustered regions to discover the real functions. At last, the proposed method is tested using 14-day SFSB data in Beijing and the results are estimated by the Satellite Map data. The proposed methods and the results can be applied to infer the individual's travel purpose through functional regions and to improve land-use planning, etc. [ABSTRACT FROM AUTHOR]

The catastrophic Sanyanyu and Luojiayu debris flows, which were induced by heavy rain, struck Zhouqu County in Gannan Prefecture, Gansu Province, at approximately midnight, 7 August 2010 (Beijing time, UTC + 8), causing 1765 fatalities and huge economic loss. The ZHQ seismic station is located approximately 170 m west of the outlet of the Sanyanyu gully, and its power system was destroyed by the Sanyanyu debris flow when its leading edge reached the vicinity of the seismic station. In this paper, seismic signals recorded approximately 10 min before its termination are collected and analyzed to study the Sanyanyu debris flow. A double-exponential model is first proposed to quantitatively characterize seismic energy distributions in the frequency domain, which reveals that the peak frequency of seismic signals is around 5 Hz. Influenced by the Doppler effect, the peak frequency of the N–S component is the highest, and the U–D component is the lowest. Time–frequency analysis is applied to the seismic signals. From the spectrogram, it is easily observed that the formation time of the Sanyanyu debris flow is around 23:33:10. The entire debris flow is divided into three phases with distinct frequency characteristics, using 23:36:20 and 23:37:35 as crucial times. The frequency energy distributions in the first two phases are relatively stable, and are constrained in 0–8.8 Hz and 0–17.8 Hz, respectively. For the third phase, the upper boundary of frequency energy increases in a nearly linear manner, reaching approximately 35 Hz at the end. We calculate synthetic seismograms of the Sanyanyu debris flow. Generally, synthetic seismograms have morphological features and characteristics in key stages similar to those of the actual seismic records, and their maximum values are of the same magnitude. Our results suggest that the seismic source of a debris flow can be represented by a single-force model, and reveal that real-time monitoring and rapid identification of potential debris flows using broadband seismic network records is possible, which can provide approximately 15 min of pre-event warning for local residents and hopefully save many lives. [ABSTRACT FROM AUTHOR]

This paper studies the location–allocation problem of distribution centers (DCs) for fresh products, in which it requires to determine the number and locations of DCs and design the allocation scheme for fresh products. The main challenge in optimizing this problem is to handle the uncertain demands with only partially known probability distribution information. To address this challenge, this paper first characterizes the ambiguous probability distributions of fresh products demands by an ambiguity set. Then, from different perspectives, this paper proposes two distributionally robust (DR) location–allocation models to find the optimal decisions that perform best in view of the worst-case distribution within the ambiguity set. The first model is with ambiguous expectation, which is transformed into a mixed-integer linear programming (MILP) model via employing the linear decision rule technique. The second model is with ambiguous chance constraints. This model is a semi-infinite chance constraint programming problem, which is transformed into a mixed-integer second-order cone programming (MISOCP) model by deriving safe approximations of the ambiguous chance constraints. To verify the feasibility and effectiveness of the proposed models, a case study about Hema fresh chain stores in Beijing is addressed. The experimental results demonstrate that the optimal decision provided by the proposed optimization methods can immunize against ambiguous probability distributions of demands with a small cost price. • Two DR location–allocation models of DCs for fresh products are raised. • The distribution information of uncertain demands is only partially known. • The DR model with ambiguous expectation is transformed into an MILP model. • The DR model with ambiguous chance constraints is transformed into an MISOCP model. • A case study shows the feasibility and effectiveness of the proposed DR models. [ABSTRACT FROM AUTHOR]

• An imputation model imputes specific missing values in multivariate time series. • Our method outperforms the comparison method when specific values are missing. • Our experiments provide selection principles of hyperparameters. • Our experiment determines the best training method with introduction of noise. Data missing is a ubiquitous phenomenon in multivariate time series analysis because of failure measurements, improper installation or other unfavorable factors. Most imputation methods, such as statistical methods and machine learning methods, are mainly for dealing with random or continuous missing, but do not take into consideration the non-random specific missing situation. This paper proposes an imputation model based on the variational auto-encoders (VAE) and shift correction for specific missing values, which is also extended to the β -VAE model. The shift correction is used to correct the original probability distribution deviation caused by specific values concentrated missing. Then taking the meteorological and air quality data sets in Beijing as the example, this paper mainly explores two common specific missing situations, including large values missing and small values missing. The experimental results show that the proposed model can effectively impute specific missing values, which has high imputation accuracy and robustness. [ABSTRACT FROM AUTHOR]

This paper presents a probability distribution model of vehicles' travel time index (TTI) in urban road segments that depicts traffic flow dispersion. Compared with traditional models, it can depict traffic flow dispersion with a uniform model under conditions of different road speeds or road lengths. Based on it, a dynamic model that depicts how the TTI distribution parameters vary with speed of road segment is proposed. Furthermore, parameters of the dynamic model are used to construct a characteristic vector to reveal a new inherent feature of road segments. The tests of goodness of fit for the two proposed models were carried out with data collected from 66,000 floating cars. The test area covered 40,400 urban road segments in the urban area of Beijing. Test results showed that more than 98% of the tested bins accepted the proposed TTI probability distribution model, and more than 90% of the tested road segments accepted the proposed dynamic model. The proposed models would be applied in traffic simulation, congestion forecast, signal control, and other relevant fields in the future. [ABSTRACT FROM AUTHOR]

This paper develops a mathematical model to calculate the average waiting time for passengers transferring from rail transit to buses based on the statistical analysis of primary data collected in Beijing. An important part of the average waiting time modelling is to analyse the distributions of passenger arrival rates. It is shown that the lognormal and gamma distributions have the best fit for direct transfer and non-direct transfer passengers, respectively. Subsequently, an average waiting time model for transferring passengers is developed based on passenger arrival rate distributions. Furthermore, case studies are conducted for two scenarios with real and estimated data, resulting in relative errors of −3.69% and −3.77%, respectively. Finally, the paper analyses the impacts of bus headway, the headway of rail cars, and the proportion of direct transfer passengers on average waiting time. [ABSTRACT FROM AUTHOR]

Equity concerns have been an important obstacle to adopting congestion pricing, in both developed and developing countries. However, the existing evidence on the equity effects of congestion pricing has come only from developed countries. In this paper, we shed light on the distributional consequences of a congestion pricing scheme currently under consideration in Beijing. We find that under this scheme, which covers the areas within the city's third ring road, a very small proportion of motorized trips would be subject to the full congestion charge. The directly affected individuals typically have higher household incomes and are wealthier than individuals who are not directly affected by the congestion pricing scheme. This finding reflects the fact that individuals who drive to work in Beijing are relatively wealthy. More important, we find that the Suits index for the congestion charge is 0.027, indicating that the congestion charge is slightly progressive. [ABSTRACT FROM AUTHOR]

Based on the stochastic concept for freeway capacity, the procedure of capacity estimation is developed. Due to the fact that it is impossible to observe the value of the capacity and to obtain the probability distribution of the capacity, the product-limit method is used in this paper to estimate the capacity. In order to implement estimation of capacity using this technology, the lifetime table based on statistical methods for lifetime data analysis is introduced and the corresponding procedure is developed. Simulated data based on freeway sections in Beijing, China, were analyzed and the results indicate that the methodology and procedure are applicable and validated. [ABSTRACT FROM AUTHOR]

Abstract: This paper is concerned with vertical distributions of mean wind speed and atmospheric turbulence characteristics over a typical urban area. The wind data were measured from more than 30 anemometers installed at 15 different height levels on the 325m high Beijing Meteorological Tower during numerous windstorms. Profiles of mean wind speed are presented based on the field measurements and are compared with empirical models’ predictions. Aerodynamic parameters of atmospheric boundary layer are determined from the measured wind speed profiles. Furthermore, wind velocity data in longitudinal, lateral and vertical directions, which were recorded by ultrasonic anemometers at three height levels during windstorms, are analyzed and discussed. Atmospheric turbulence information such as turbulence intensity, gust factor, turbulence integral length scale and power spectral densities of the three-dimensional fluctuating wind velocity are presented and used to evaluate the adequacy of the existing theoretical and empirical models. A comparative study between the measured profiles of turbulence parameters and those estimated by and is performed. The objective of this study aims to provide useful information on boundary layer wind characteristics for the wind-resistant design of tall buildings and high-rise structures in urban areas. [Copyright &y& Elsevier]

In this paper, we applied a fractal approach to analyze the traffic data collected from the Beijing Yuquanying. The power spectrum, the empirical probability distribution function, the statistical moment scaling function and the autocorrelation function are used as indicators to investigate the presence of the fractal. The results from the fractal identification methods indicate that these data exhibit fractal behavior. A fractal framework seemed well suited for description of the data observed here, but its suitability for general traffic systems was not clear. [Copyright &y& Elsevier]

The purpose of this paper is to investigate the applicability of Bayesian Multi‐net Classifier (BMC) to classify remote sensing data. BMC is based on Bayesian Network (BN), which is a graphical model encoding probabilistic relationships among variables of interest. Different from the BNC that has a mere network, a BMC has as many local Bayesian Networks as the predefined classes, which means that the probabilistic relationships among the features can be different for different classes. Classification is done by computing the probability of the class, given the particular instance of the features, and then predicting the class with the highest posterior probability. This method was validated using a Landsat ETM+ image of Beijing acquired on 1 May 2003. Based on the confusion matrix, overall accuracy, Kappa statistic, total normalized probability of misclassification (TNPM), and McNemar's test, classification results of BMC were compared with those of MLC and BNC in the case study. The comparison results show that BMC performs slightly better than MLC and similar to BNC. The local Bayesian Networks of BMC can also lead to a better understanding of the dependencies between bands for different classes. [ABSTRACT FROM AUTHOR]

The decomposition of a point process is useful for the analysis of spatial patterns and in the discovery of potential mechanisms of geographic phenomena. However, when a local repulsive cluster is present in a complex heterogeneous point process, the traditional solution, which is based on clustering, may be invalid for decomposition because a repulsive pattern is not subject to a specific probability distribution function and the effects of aggregative and repulsive components may be counterbalanced. To solve this problem, this paper proposes a method of decomposing repulsive clusters in complex point processes with multiple heterogeneous components. A repulsive cluster is defined as a set of repulsive density-connected points that are separated by a certain distance at a small scale and aggregated at a large scale simultaneously. The H-function is used to identify repulsive clusters by determining the repulsive distance and extracting repulsive points for further clustering. Through simulation experiments based on three datasets, the proposed method has been shown to effectively perform repulsive cluster decomposition in heterogeneous point processes. A case study of the point of interest (POI) dataset in Beijing also indicates that the method can identify meaningful repulsive clusters from types of POIs that represent different service characteristics of shops in different local regions. [ABSTRACT FROM AUTHOR]