Autonomous Vehicle Platoons In Urban Road Networks:A Joint Distributed Reinforcement Learning and Model Predictive Control Approach

In this paper, platoons of autonomous vehicles operating in urban road networks are considered. From a methodological point of view, the problem of interest consists of formally characterizing vehicle state trajectory tubes by means of routing decisions complying with traffic congestion criteria. To this end, a novel distributed control architecture is conceived by taking advantage of two methodologies: deep reinforcement learning and model predictive control. On one hand, the routing decisions are obtained by using a distributed reinforcement learning algorithm that exploits available traffic data at each road junction. On the other hand, a bank of model predictive controllers is in charge of computing the more adequate control action for each involved vehicle. Such tasks are here combined into a single framework:the deep reinforcement learning output(action) is translated into a set-point to be tracked by the model predictive controller;conversely, the current vehicle position, resulting from the application of the control move, is exploited by the deep reinforcement learning unit for improving its reliability. The main novelty of the proposed solution lies in its hybrid nature: on one hand it fully exploits deep reinforcement learning capabilities for decisionmaking purposes;on the other hand, time-varying hard constraints are always satisfied during the dynamical platoon evolution imposed by the computed routing decisions. To efficiently evaluate the performance of the proposed control architecture, a co-design procedure, involving the SUMO and MATLAB platforms, is implemented so that complex operating environments can be used, and the information coming from road maps(links,junctions, obstacles, semaphores, etc.) and vehicle state trajectories can be shared and exchanged. Finally by considering as operating scenario a real entire city block and a platoon of eleven vehicles described by double-integrator models, several simulations have been performed with the aim to put in light the main f eatures of the proposed approach. Moreover, it is important to underline that in different operating scenarios the proposed reinforcement learning scheme is capable of significantly reducing traffic congestion phenomena when compared with well-reputed competitors.

Predicting the probability distribution of Martian rocks mechanical property based on microscale rock mechanical experiments and accurate grain-based modeling

The exploration of Mars would heavily rely on Martian rocks mechanics and engineering technology.As the mechanical property of Martian rocks is uncertain,it is of utmost importance to predict the probability distribution of Martian rocks mechanical property for the success of Mars exploration.In this paper,a fast and accurate probability distribution method for predicting the macroscale elastic modulus of Martian rocks was proposed by integrating the microscale rock mechanical experiments(micro-RME),accurate grain-based modeling(AGBM)and upscaling methods based on reliability principles.Firstly,the microstructure of NWA12564 Martian sample and elastic modulus of each mineral were obtained by micro-RME with TESCAN integrated mineral analyzer(TIMA)and nanoindentation.The best probability distribution function of the minerals was determined by Kolmogorov-Smirnov(K-S)test.Secondly,based on best distribution function of each mineral,the Monte Carlo simulations(MCS)and upscaling methods were implemented to obtain the probability distribution of upscaled elastic modulus.Thirdly,the correlation between the upscaled elastic modulus and macroscale elastic modulus obtained by AGBM was established.The accurate probability distribution of the macroscale elastic modulus was obtained by this correlation relationship.The proposed method can predict the probability distribution of Martian rocks mechanical property with any size and shape samples.

Federated Learning Model for Auto Insurance Rate Setting Based on Tweedie Distribution

In the assessment of car insurance claims,the claim rate for car insurance presents a highly skewed probability distribution,which is typically modeled using Tweedie distribution.The traditional approach to obtaining the Tweedie regression model involves training on a centralized dataset,when the data is provided by multiple parties,training a privacy-preserving Tweedie regression model without exchanging raw data becomes a challenge.To address this issue,this study introduces a novel vertical federated learning-based Tweedie regression algorithm for multi-party auto insurance rate setting in data silos.The algorithm can keep sensitive data locally and uses privacy-preserving techniques to achieve intersection operations between the two parties holding the data.After determining which entities are shared,the participants train the model locally using the shared entity data to obtain the local generalized linear model intermediate parameters.The homomorphic encryption algorithms are introduced to interact with and update the model intermediate parameters to collaboratively complete the joint training of the car insurance rate-setting model.Performance tests on two publicly available datasets show that the proposed federated Tweedie regression algorithm can effectively generate Tweedie regression models that leverage the value of data fromboth partieswithout exchanging data.The assessment results of the scheme approach those of the Tweedie regressionmodel learned fromcentralized data,and outperformthe Tweedie regressionmodel learned independently by a single party.

Distributed Economic MPC for Synergetic Regulation of the Voltage of an Island DC Micro-Grid

In this paper,distributed model predictive control(DMPC) for island DC micro-grids(MG) with wind/photovoltaic(PV)/battery power is proposed,which coordinates all distributed generations(DG) to stabilize the bus voltage together with the insurance of having computational efficiency under a real-time requirement.Based on the feedback of the bus voltage,the deviation of the current is dispatched to each DG according to cost over the prediction horizon.Moreover,to avoid the excessive fluctuation of the battery power,both the discharge-charge switching times and costs are considered in the model predictive control(MPC) optimization problems.A Lyapunov constraint with a time-varying steady-state is designed in each local MPC to guarantee the stabilization of the entire system.The voltage stabilization of the MG is achieved by this strategy with the cooperation of DGs.The numeric results of applying the proposed method to a MG of the Shanghai Power Supply Company shows the effectiveness of the distributed economic MPC.

Modeling load distribution for rural photovoltaic grid areas using image recognition

Expanding photovoltaic(PV)resources in rural-grid areas is an essential means to augment the share of solar energy in the energy landscape,aligning with the“carbon peaking and carbon neutrality”objectives.However,rural power grids often lack digitalization;thus,the load distribution within these areas is not fully known.This hinders the calculation of the available PV capacity and deduction of node voltages.This study proposes a load-distribution modeling approach based on remote-sensing image recognition in pursuit of a scientific framework for developing distributed PV resources in rural grid areas.First,houses in remote-sensing images are accurately recognized using deep-learning techniques based on the YOLOv5 model.The distribution of the houses is then used to estimate the load distribution in the grid area.Next,equally spaced and clustered distribution models are used to adaptively determine the location of the nodes and load power in the distribution lines.Finally,by calculating the connectivity matrix of the nodes,a minimum spanning tree is extracted,the topology of the network is constructed,and the node parameters of the load-distribution model are calculated.The proposed scheme is implemented in a software package and its efficacy is demonstrated by analyzing typical remote-sensing images of rural grid areas.The results underscore the ability of the proposed approach to effectively discern the distribution-line structure and compute the node parameters,thereby offering vital support for determining PV access capability.

Finer topographic data improves distribution modeling of Picea crassifolia in the northern Qilian Mountains

The Qilian Mountains, a national key ecological function zone in Western China, play a pivotal role in ecosystem services. However, the distribution of its dominant tree species, Picea crassifolia (Qinghai spruce), has decreased dramatically in the past decades due to climate change and human activity, which may have influenced its ecological functions. To restore its ecological functions, reasonable reforestation is the key measure. Many previous efforts have predicted the potential distribution of Picea crassifolia, which provides guidance on regional reforestation policy. However, all of them were performed at low spatial resolution, thus ignoring the natural characteristics of the patchy distribution of Picea crassifolia. Here, we modeled the distribution of Picea crassifolia with species distribution models at high spatial resolutions. For many models, the area under the receiver operating characteristic curve (AUC) is larger than 0.9, suggesting their excellent precision. The AUC of models at 30 m is higher than that of models at 90 m, and the current potential distribution of Picea crassifolia is more closely aligned with its actual distribution at 30 m, demonstrating that finer data resolution improves model performance. Besides, for models at 90 m resolution, annual precipitation (Bio12) played the paramount influence on the distribution of Picea crassifolia, while the aspect became the most important one at 30 m, indicating the crucial role of finer topographic data in modeling species with patchy distribution. The current distribution of Picea crassifolia was concentrated in the northern and central parts of the study area, and this pattern will be maintained under future scenarios, although some habitat loss in the central parts and gain in the eastern regions is expected owing to increasing temperatures and precipitation. Our findings can guide protective and restoration strategies for the Qilian Mountains, which would benefit regional ecological balance.

Virtually coupled train set control subject to space-time separation:A distributed economic MPC approach with emergency braking configuration

The emerging virtual coupling technology aims to operate multiple train units in a Virtually Coupled Train Set(VCTS)at a minimal but safe distance.To guarantee collision avoidance,the safety distance should be calculated using the state-of-the-art space-time separation principle that separates the Emergency Braking(EB)trajectories of two successive units during the whole EB process.In this case,the minimal safety distance is usually numerically calculated without an analytic formulation.Thus,the constrained VCTS control problem is hard to address with space-time separation,which is still a gap in the existing literature.To solve this problem,we propose a Distributed Economic Model Predictive Control(DEMPC)approach with computation efficiency and theoretical guarantee.Specifically,to alleviate the computation burden,we transform implicit safety constraints into explicitly linear ones,such that the optimal control problem in DEMPC is a quadratic programming problem that can be solved efficiently.For theoretical analysis,sufficient conditions are derived to guarantee the recursive feasibility and stability of DEMPC,employing compatibility constraints,tube techniques and terminal ingredient tuning.Moreover,we extend our approach with globally optimal and distributed online EB configuration methods to shorten the minimal distance among VCTS.Finally,experimental results demonstrate the performance and advantages of the proposed approaches.

Application of a 2 Parameter Weibull Distribution in Modeling of State Holding Time in HIV/AIDS Transition Dynamics

This study investigates the application of the two-parameter Weibull distribution in modeling state holding times within HIV/AIDS progression dynamics. By comparing the performance of the Weibull-based Accelerated Failure Time (AFT) model, Cox Proportional Hazards model, and Survival model, we assess the effectiveness of these models in capturing survival rates across varying gender, age groups, and treatment categories. Simulated data was used to fit the models, with model identification criteria (AIC, BIC, and R2) applied for evaluation. Results indicate that the AFT model is particularly sensitive to interaction terms, showing significant effects for older age groups (50 - 60 years) and treatment interaction, while the Cox model provides a more stable fit across all age groups. The Survival model displayed variability, with its performance diminishing when interaction terms were introduced, particularly in older age groups. Overall, while the AFT model captures the complexities of interactions in the data, the Cox model’s stability suggests it may be better suited for general analyses without strong interaction effects. The findings highlight the importance of model selection in survival analysis, especially in complex disease progression scenarios like HIV/AIDS.

The effects of cold region meteorology and specific environment on the number of hospital admissions for chronic kidney disease:An investigate with a distributed lag nonlinear model

Objective:To explore the effects of daily mean temperature(°C),average daily air pressure(hPa),humidity(%),wind speed(m/s),particulate matter(PM)2.5(μg/m3)and PM10(μg/m3)on the admission rate of chronic kidney disease(CKD)patients admitted to the Second Affiliated Hospital of Harbin Medical University in Harbin and to identify the indexes and lag days that impose the most critical influence.Methods:The R language Distributed Lag Nonlinear Model(DLNM),Excel,and SPSS were used to analyze the disease and meteorological data of Harbin from 01 January 2010 to 31 December 2019 according to the inclusion and exclusion criteria.Results:Meteorological factors and air pollution influence the number of hospitalizations of CKD to vary degrees in cold regions,and differ in persistence or delay.Non-optimal temperature increases the risk of admission of CKD,high temperature increases the risk of obstructive kidney disease,and low temperature increases the risk of other major types of chronic kidney disease.The greater the temperature difference is,the higher its contribution is to the risk.The non-optimal wind speed and non-optimal atmospheric pressure are associated with increased hospital admissions.PM2.5 concentrations above 40μg/m3 have a negative impact on the results.Conclusion:Cold region meteorology and specific environment do have an impact on the number of hospital admissions for chronic kidney disease,and we can apply DLMN to describe the analysis.

COMBINATION OF DISTRIBUTED KALMAN FILTER AND BP NEURAL NETWORK FOR ESG BIAS MODEL IDENTIFICATION

By combining the distributed Kalman filter （DKF） with the back propagation neural network （BPNN）,a novel method is proposed to identify the bias of electrostatic suspended gyroscope （ESG）. Firstly,the data sets of multi-measurements of the same ESG in different noise environments are ＂mapped＂ into a sensor network,and DKF with embedded consensus filters is then used to preprocess the data sets. After transforming the preprocessed results into the trained input and the desired output of neural network,BPNN with the learning rate and the momentum term is further utilized to identify the ESG bias. As demonstrated in the experiment,the proposed approach is effective for the model identification of the ESG bias.

Analysis and Optimum Design of Differential Inductors Using Distributed Capacitance Model

A distributed capacitance model for monolithic inductors is developed to predict the equivalently parasitical capacitances of the inductor.The ratio of the self-resonant frequency (f SR) of the differential-driven symmetric inductor to the f SR of the single-ended driven inductor is firstly predicted and explained.Compared with a single-ended configuration,experimental data demonstrate that the differential inductor offers a 127% greater maximum quality factor and a broader range of operating frequencies.Two differential inductors with low parasitical capacitance are developed and validated.

Working Condition Real-Time Monitoring Model of Lithium Ion Batteries Based on Distributed Parameter System and Single Particle Model

Lithium ion batteries are complicated distributed parameter systems that can be described preferably by partial differential equations and a field theory. To reduce the solution difficulty and the calculation amount, if a distributed parameter system is described by ordinary differential equations （ODE） during the analysis and the design of distributed parameter system, the reliability of the system description will be reduced, and the systemic errors will be introduced. Studies on working condition real-time monitoring can improve the security because the rechargeable LIBs are widely used in many electronic systems and electromechanical equipment. Single particle model （SPM） is the simplification of LIB under some approximations, and can estimate the working parameters of a LIB at the faster simulation speed. A LIB modelling algorithm based on PDEs and SPM is proposed to monitor the working condition of LIBs in real time. Although the lithium ion concentration is an unmeasurable distributed parameter in the anode of LIB, the working condition monitoring model can track the real time lithium ion concentration in the anode of LIB, and calculate the residual which is the difference between the ideal data and the measured data. A fault alarm can be triggered when the residual is beyond the preset threshold. A simulation example verifies that the effectiveness and the accuracy of the working condition real-time monitoring model of LIB based on PDEs and SPM.

A hardening load transfer function for rock bolts and its calibration using distributed fiber optic sensing

Confinement of rock bolts by the surrounding rock formation has long been recognized as a positive contributor to the pull-out behavior,yet only a few experimental works and analytical models have been reported,most of which are based on the global rock bolt response evaluated in pull-out tests.This paper presents a laboratory experimental setup aiming to capture the rock formation effect,while using distributed fiber optic sensing to quantify the effect of the confinement and the reinforcement pull-out behavior on a more local level.It is shown that the behavior along the sample itself varies,with certain points exhibiting stress drops with crack formation.Some edge effects related to the kinematic freedom of the grout to dilate are also observed.Regardless,it was found that the mid-level response is quite similar to the average response along the sample.The ability to characterize the variation of the response along the sample is one of the many advantages high-resolution fiber optic sensing allows in such investigations.The paper also offers a plasticity-based hardening load transfer function,representing a"slice"of the anchor.The paper describes in detail the development of the model and the calibration/determination of its parameters.The suggested model captures well the coupled behavior in which the pull-out process leads to an increase in the confining stress due to dilative behavior.

Potential global distribution of the guava root-knot nematode Meloidogyne enterolobii under different climate change scenarios using MaxEnt ecological niche modeling

In recent years,Meloidogyne enterolobii has emerged as a major parasitic nematode infesting many plants in tropical or subtropical areas.However,the regions of potential distribution and the main contributing environmental variables for this nematode are unclear.Under the current climate scenario,we predicted the potential geographic distributions of M.enterolobii worldwide and in China using a Maximum Entropy(MaxEnt)model with the occurrence data of this species.Furthermore,the potential distributions of M.enterolobii were projected under three future climate scenarios(BCC-CSM2-MR,CanESM5 and CNRM-CM6-1)for the periods 2050s and 2090s.Changes in the potential distribution were also predicted under different climate conditions.The results showed that highly suitable regions for M.enterolobii were concentrated in Africa,South America,Asia,and North America between latitudes 30°S to 30°N.Bio16(precipitation of the wettest quarter),bio10(mean temperature of the warmest quarter),and bio11(mean temperature of the coldest quarter)were the variables contributing most in predicting potential distributions of M.enterolobii.In addition,the potential suitable areas for M.enterolobii will shift toward higher latitudes under future climate scenarios.This study provides a theoretical basis for controlling and managing this nematode.

A DISTRIBUTED HYPERMAP MODEL FOR INTERNET GIS

The rapid development of Internet technology makes it possible to integrate GIS with the Internet,forming Internet GIS.Internet GIS is based on a distributed client/server architecture and TCP/IP & IIOP.When constructing and designing Internet GIS,we face the problem of how to express information units of Internet GIS.In order to solve this problem,this paper presents a distributed hypermap model for Internet GIS.This model provides a solution to organize and manage Internet GIS information units.It also illustrates relations between two information units and in an internal information unit both on clients and servers.On the basis of this model,the paper contributes to the expressions of hypermap relations and hypermap operations.The usage of this model is shown in the implementation of a prototype system.

A distributed runoff model for inland mountainous river basin of Northwest China

In order to predict the futuristic runoff under global warming, and to approach to the effects of vegetation on the ecological environment of the inland river mountainous watershed of Northwest China, the authors use the routine hydrometric data to create a distributed monthly model with some conceptual parameters, coupled with GIS and RS tools and data. The model takes sub-basin as the minimal confluent unit, divides the main soils of the basin into 3 layers, and identifies the vegetation types as forest and pasture. The data used in the model are precipitation, air temperature, runoff, soil weight water content, soil depth, soil bulk density, soil porosity, land cover, etc. The model holds that if the water amount is greater than the water content capacity, there will be surface runoff. The actual evaporation is proportional to the product of the potential evaporation and soil volume water content. The studied basin is Heihe mainstream mountainous basin, with a drainage area of 10,009 km 2 . The data used in this simulation are from Jan. 1980 to Dec. 1995, and the first 10 years' data are used to simulate, while the last 5 years' data are used to calibrate. For the simulation process, the Nash-Sutcliffe Equation, Balance Error and Explained Variance is 0.8681, 5.4008 and 0.8718 respectively, while for the calibration process, 0.8799, -0.5974 and 0.8800 respectively. The model results show that the futuristic runoff of Heihe river basin will increase a little. The snowmelt, glacier meltwater and the evaportranspiration will increase. The air temperature increment will make the permanent snow and glacier area diminish, and the snowline will rise. The vegetation, especially the forest in Heihe mountainous watershed, could lead to the evapotranspiration decrease of the watershed, adjust the runoff process, and increase the soil water content.

GLOBAL STABILITY OF SIRS EPIDEMIC MODELS WITH A CLASS OF NONLINEAR INCIDENCE RATES AND DISTRIBUTED DELAYS

In this article, we establish the global asymptotic stability of a disease-free equilibrium and an endemic equilibrium of an SIRS epidemic model with a class of nonlin- ear incidence rates and distributed delays. By using strict monotonicity of the incidence function and constructing a Lyapunov functional, we obtain sufficient conditions under which the endemic equilibrium is globally asymptotically stable. When the nonlinear inci- dence rate is a saturated incidence rate, our result provides a new global stability condition for a small rate of immunity loss.

Distributed modeling of direct solar radiation on rugged terrain of the Yellow River Basin

Due to the influences of local topographical factors and terrain inter-shielding, calculation of direct solar radiation （DSR） quantity of rugged terrain is very complex. Based on digital elevation model （DEM） data and meteorological observations, a distributed model for calculating DSR over rugged terrain is developed. This model gives an all-sided consideration on factors influencing th a resolution of 1 km × 1 km for thDSR. Using the developed model, normals of annual DSR quantity wie Yellow River Basin was generated, with DEM data as the general characterization of terrain. Characteristics of DSR quantity influenced by geographic and topographic factors over rugged terrain were analyzed thoroughly. Results suggest that： influenced by local topographic factors, i.e. azimuth, slope and so on, and annual DSR quantity over mountainous area has a clear spatial difference; annual DSR quantity of sunny slope （or southern slope） of mountains is obviously larger than that of shady slope （or northern slope）. The calculated DSR quantity of the Yellow River Basin is provided in the same way as other kinds of spatial information and can be employed as basic geographic data for relevant studies as well.

Framework of Distributed Coupled Atmosphere-Ocean-Wave Modeling System

In order to research the interactions between the atmosphere and ocean as well as their important role in the intensive weather systems of coastal areas, and to improve the forecasting ability of the hazardous weather processes of coastal areas, a coupled atmosphere-ocean-wave modeling system has been developed. The agent-based environment framework for linking models allows flexible and dynamic information exchange between models. For the purpose of flexibility, portability and scalability, the framework of the whole system takes a multi-layer architecture that includes a user interface layer, computational layer and service-enabling layer. The numerical experiment presented in this paper demonstrates the performance of the distributed coupled modeling system.

Distributed hydrological models for addressing effects of spatial variability of roughness on overland flow

In this study, we investigated the origin of the overland flow roughness problem and divided the current overland flow roughness research into three types, as follows: the first type of research takes into account the effects of roughness on the volume and velocity of surface runoff, flood peaks, and the scouring capability of flows, but has not addressed the spatial variability of roughness in detail; the second type of research considers that surface roughness varies spatially with different land usage types, land-cover conditions, and different tillage forms, but lacks a quantitative study of the spatial variability; and the third type of research simply deals with the spatial variability of roughness in each grid cell or land type. We present three shortcomings of the current overland flow roughness research, including(1) the neglect of roughness in distributed hydrological models when simulating the overland flow direction and distribution,(2) the lack of consideration of spatial variability of roughness in hydrological models, and(3) the failure to distinguish the roughness formulas in different overland flow regimes. To solve these problems,distributed hydrological model research should focus on four aspects in regard to overland flow: velocity field observations, flow regime mechanisms, a basic roughness theory, and scale problems.