Machine learning prediction models for ground motion parameters and seismic damage assessment of buildings at a regional scale

This study examines the feasibility of using a machine learning approach for rapid damage assessment of rein-forced concrete(RC)buildings after the earthquake.Since the real-world damaged datasets are lacking,have limited access,or are imbalanced,a simulation dataset is prepared by conducting a nonlinear time history analy-sis.Different machine learning(ML)models are trained considering the structural parameters and ground motion characteristics to predict the RC building damage into five categories:null,slight,moderate,heavy,and collapse.The random forest classifier(RFC)has achieved a higher prediction accuracy on testing and real-world damaged datasets.The structural parameters can be extracted using different means such as Google Earth,Open Street Map,unmanned aerial vehicles,etc.However,recording the ground motion at a closer distance requires the installation of a dense array of sensors which requires a higher cost.For places with no earthquake recording station/device,it is difficult to have ground motion characteristics.For that different ML-based regressor models are developed utilizing past-earthquake information to predict ground motion parameters such as peak ground acceleration and peak ground velocity.The random forest regressor(RFR)achieved better results than other regression models on testing and validation datasets.Furthermore,compared with the results of similar research works,a better result is obtained using RFC and RFR on validation datasets.In the end,these models are uti-lized to predict the damage categories of RC buildings at Saitama University and Okubo Danchi,Saitama,Japan after an earthquake.This damage information is crucial for government agencies or decision-makers to respond systematically in post-disaster situations.

Prediction and optimization of flue pressure in sintering process based on SHAP

Sinter is the core raw material for blast furnaces.Flue pressure,which is an important state parameter,affects sinter quality.In this paper,flue pressure prediction and optimization were studied based on the shapley additive explanation(SHAP)to predict the flue pressure and take targeted adjustment measures.First,the sintering process data were collected and processed.A flue pressure prediction model was then constructed after comparing different feature selection methods and model algorithms using SHAP+extremely random-ized trees(ET).The prediction accuracy of the model within the error range of±0.25 kPa was 92.63%.SHAP analysis was employed to improve the interpretability of the prediction model.The effects of various sintering operation parameters on flue pressure,the relation-ship between the numerical range of key operation parameters and flue pressure,the effect of operation parameter combinations on flue pressure,and the prediction process of the flue pressure prediction model on a single sample were analyzed.A flue pressure optimization module was also constructed and analyzed when the prediction satisfied the judgment conditions.The operating parameter combination was then pushed.The flue pressure was increased by 5.87%during the verification process,achieving a good optimization effect.

Study on Multi-Scale Blending Initial Condition Perturbations for a Regional Ensemble Prediction System

An initial conditions （ICs） perturbation method was developed with the aim to improve an operational regional ensemble prediction system （REPS）. Three issues were identified and investigated： （1） the impacts of perturbation scale on the ensemble spread and forecast skill of the REPS; （2） the scale characteristic of the IC perturbations of the REPS; and （3） whether the REPS＇s skill could be improved by adding large-scale information to the IC perturbations. Numerical experiments were conducted to reveal the impact of perturbation scale on the ensemble spread and forecast skill. The scales of IC perturbations from the REPS and an operational global ensemble prediction system （GEPS） were analyzed. A ＂multi-scale blending＂ （MSB） IC perturbation scheme was developed, and the main findings can be summarized as follows： The growth rates of the ensemble spread of the REPS are sensitive to the scale of the IC perturbations; the ensemble forecast skills can benefit from large-scale perturbations; the global ensemble IC perturbations exhibit more power at larger scales, while the regional ensemble IC perturbations contain more power at smaller scales; the MSB method can generate IC perturbations by combining the small-scale component from the REPS and the large-scale component from the GEPS; the energy norm growth of the MSB-generated perturbations can be appropriate at all forecast lead times; and the MSB-based REPS shows higher skill than the original system, as determined by ensemble forecast verification.

Multi-Year Simulations and Experimental Seasonal Predictions for Rainy Seasons in China by Using a Nested Regional Climate Model (RegCM_NCC) Part Ⅱ:The Experimental Seasonal Prediction

A nested regional climate model has been experimentally used in the seasonal prediction at the China National Climate Center （NCC） since 2001. The NCC/IAP （Institute of Atmospheric Physics） T63 coupled GCM （CGCM） provides the boundary and initial conditions for driving the regional climate model （RegCM_NCC）. The latter has a 60-km horizontal resolution and improved physical parameterization schemes including the mass flux cumulus parameterization scheme, the turbulent kinetic energy closure scheme （TKE） and an improved land process model （LPM）. The large-scale terrain features such as the Tibetan Plateau are included in the larger domain to produce the topographic forcing on the rain-producing systems. A sensitivity study of the East Asian climate with regard to the above physical processes has been presented in the first part of the present paper. This is the second part, as a continuation of Part Ⅰ. In order to verify the performance of the nested regional climate model, a ten-year simulation driven by NCEP reanalysis datasets has been made to explore the performance of the East Asian climate simulation and to identify the model＇s systematic errors. At the same time, comparative simulation experiments for 5 years between the RegCM2 and RegCM_NCC have been done to further understand their differences in simulation performance. Also, a ten-year hindcast （1991-2000） for summer （June-August）, the rainy season in China, has been undertaken. The preliminary results have shown that the RegCM_NCC is capable of predicting the major seasonal rain belts. The best predicted regions with high anomaly correlation coefficient （ACC） are located in the eastern part of West China, in Northeast China and in North China, where the CGCM has maximum prediction skill as well. This fact may reflect the importance of the largescale forcing. One significant improvement of the prediction derived from RegCM_NCC is the increase of ACC in the Yangtze River valley where the CGCM has a very low, even a negative, ACC. The reason behind this improvement is likely to be related to the more realistic representation of the large-scale terrain features of the Tibetan Plateau. Presumably, many rain-producing systems may be generated over or near the Tibetan Plateau and may then move eastward along the Yangtze River basin steered by upper-level westerly airflow, thus leading to enhancement of rainfalls in the mid and lower basins of the Yangtze River. The real-time experimental predictions for summer in 2001, 2002, 2003 and 2004 by using this nested RegCM-NCC were made. The results are basically reasonable compared with the observations.

Possible Sources of Forecast Errors Generated by the Global/Regional Assimilation and Prediction System for Landfalling Tropical Cyclones.PartⅠ:Initial Uncertainties

This paper investigates the possible sources of errors associated with tropical cyclone （TC） tracks forecasted using the Global/Regional Assimilation and Prediction System （GRAPES）. The GRAPES forecasts were made for 16 landfaIling TCs in the western North Pacific basin during the 2008 and 2009 seasons, with a forecast length of 72 hours, and using the default initial conditions （＂initials＂, hereafter）, which are from the NCEP-FNL dataset, as well as ECMWF initials. The forecasts are compared with ECMWF forecasts. The results show that in most TCs, the GRAPES forecasts are improved when using the ECMWF initials compared with the default initials. Compared with the ECMWF initials, the default initials produce lower intensity TCs and a lower intensity subtropical high, but a higher intensity South Asia high and monsoon trough, as well as a higher temperature but lower specific humidity at the TC center. Replacement of the geopotential height and wind fields with the ECMWF initials in and around the TC center at the initial time was found to be the most efficient way to improve the forecasts. In addition, TCs that showed the greatest improvement in forecast accuracy usually had the largest initial uncertainties in TC intensity and were usually in the intensifying phase. The results demonstrate the importance of the initial intensity for TC track forecasts made using GRAPES, and indicate the model is better in describing the intensifying phase than the decaying phase of TCs. Finally, the limit of the improvement indicates that the model error associated with GRAPES forecasts may be the main cause of poor forecasts of landfalling TCs. Thus, further examinations of the model errors are required.

Development of regional prediction information system of coal and gas outburst

Coal and gas outburst information system is based on-Geographic Information System(GIS), with which the relation among mine geological structure, coal features, stress field and coal and gas outburst were researched, and also the relation between gas distributed condition and dangerous degrees. Various prediction method, index and technique were applied to realize the data visualization; the accuracy of region prediction was increased. The system has successfully applied in Huainan minging area and Pingdingshan minging area.

Monthly Extended Predicting Experiments with Nonlinear Regional Prediction.PartⅠ:Prediction of Zonal Mean Flow

Systematic errors have recently been founded to be distinct in the zonal mean component forecasts, which account for a large portion of the total monthly-mean forecast errors. To overcome the difficulty of numerical model, the monthly pentad-mean nonlinear dynamic regional prediction models of the zonal mean geopotential height at 200, 300, 500, and 700 hPa based on a large number of historical data （NCEP/NCAR reanalysis data） were constituted by employing the local approximation of the phase space reconstruction theory and nonlinear spatio-temporal series prediction method. The 12-month forecast experiments of 1996 indicated that the results of the nonlinear model are better than those of the persistent, climatic prediction, and T42L9 model either over the high- and mid-latitude areas of the Northern and Southern Hemispheres or the tropical area. The root-mean-square of the monthly-mean height of T42L9 model was considerably decreased with a change of 30.4%, 26.6%, 82.6%, and 39.4%, respectively, over the high- and mid-latitudes of the Northern Hemisphere, over the high- and mid-latitudes of the Southern Hemisphere, over the tropics and over the globe, and also the corresponding anomaly correlation coefficients over the four areas were respectively increased by 0.306-0.312, 0.304-0.429, 0.739-0.746, and 0.360-0.400 （averagely a relative change of 11.0% over the globe） by nonlinear correction after integration, implying that the forecasts given by nonlinear model include more useful information than those of T42L9 model.

Monthly Extended Predicting Experiments with Nonlinear Regional Prediction.PartⅡ:Improvement of Wave Component Prediction

Based on Chen et al. （2006）, the scheme of the combination of the pentad-mean zonal height departure nonlinear prediction with the T42L9 model prediction was designed, in which the pentad zonal heights at all the 12-initial-value-input isobar levels from 50 hPa to 1000 hPa except 200, 300, 500, and 700 hPa were derived from nonlinear forecasts of the four levels by means of a good correlation between neighboring levels. Then the above pentad zonal heights at 12 isobar-levels were transformed to the spectrum coefficients of the temperature at each integration step of T42L9 model. At last, the nudging was made. On account of a variety of error accumulation, the pentad zonal components of the monthly height at isobar levels output by T42L9 model were replaced by the corresponding nonlinear results once more when integration was over. Multiple case experiments showed that such combination of two kinds of prediction made an improvement in the wave component as a result of wave-flow nonlinear interaction while reducing the systematical forecast errors. Namely the monthly-mean height anomaly correlation coefficients over the high- and mid-latitudes of the Northern Hemisphere, over the Southern Hemisphere and over the globe increased respectively from 0.249 to 0.347, from 0.286 to 0.387, and from 0.343 to 0.414 （relative changes of 31.5%, 41.0%, and 18.3%）. The monthly-mean root-mean-square error （RMSE） of T42L9 model over the three areas was considerably decreased, the relative change over the globe reached 44.2%. The monthly-mean anomaly correlation coeffi- cients of wave 4-9 over the areas were up to 0.392, 0.200, and 0.295, with the relative change of 53.8%, 94.1%, and 61.2%, and correspondingly their RMSEs were decreased respectively with the rate of 8.5%, 6.3%, and 8.1%. At the same time the monthly-mean pattern of parts of cases were presented better.

IMPROVEMENT OF REGIONAL PREDICTION OF SEA FOG ON GUANGDONG COASTLAND USING THE FACTOR OF TEMPERATURE DIFFERENCE IN THE NEAR-SURFACE LAYER

The relationship between the factor of temperature difference of the near-surface layer(T_(1000 hPa)-T_(2m))and sea fog is analyzed using the NCEP reanalysis with a horizontal resolution of l°xl°(2000 to 2011) and the station observations(2010 to 2011).The element is treated as the prediction variable factor in the GRAPES model and used to improve the regional prediction of sea fog on Guangdong coastland.(1) The relationship between this factor and the occurrence of sea fog is explicit:When the sea fog happens,the value of this factor is always large in some specific periods,and the negative value of this factor decreases significantly or turns positive,suggesting the enhancement of warm and moist advection of air flow near the surface,which favors the development of sea fog.(2) The transportation of warm and moist advection over Guangdong coastland is featured by some stages and the jumping among these states.It also gets stronger over time.Meanwhile,the northward propagation of warm and moist advection is quite consistent with the northward advancing of sea fog from south to north along the coastland of China.(3) The GRAPES model can well simulate and realize the factor of near-surface temperature difference.Besides,the accuracy of regional prediction of marine fog,the relevant threat score and Heidke skill score are all improved when the factor is involved.

Multi-year Simulations and Experimental Seasonal Predictions for Rainy Seasons in China by Using a Nested Regional Climate Model (RegCM_NCC). Part Ⅰ: Sensitivity Study

A modified version of the NCAR/RegCM2 has been developed at the National Climate Center （NCC）, China Meteorological Administration, through a series of sensitivity experiments and multi-year simulations and hindcasts, with a special emphasis on the adequate choice of physical parameterization schemes suitable for the East Asian monsoon climate. This regional climate model is nested with the NCC/IAP （Institute of Atmospheric Physics） T63 coupled GCM to make an experimental seasonal prediction for China and East Asia. The four-year （2001 to 2004） prediction results are encouraging. This paper is the first part of a two-part paper, and it mainly describes the sensitivity study of the physical process paraxneterization represented in the model. The systematic errors produced by the different physical parameterization schemes such as the land surface processes, convective precipitation, cloud-radiation transfer process, boundary layer process and large-scale terrain features have been identified based on multi-year and extreme flooding event simulations. A number of comparative experiments has shown that the mass flux scheme （MFS） and Betts-Miller scheme （BM） for convective precipitation, the LPMI （land surface process model I） and LPMII （land surface process model Ⅱ） for the land surface process, the CCM3 radiation transfer scheme for cloud-radiation transfer processes, the TKE （turbulent kinetic energy） scheme for the boundary layer processes and the topography treatment schemes for the Tibetan Plateau are suitable for simulations and prediction of the East Asia monsoon climate in rainy seasons. Based on the above sensitivity study, a modified version of the RegCM2 （RegCM_NCC） has been set up for climate simulations and seasonal predictions.

Optimal Initial Error Growth in the Prediction of the Kuroshio Large Meander Based on a High-resolution Regional Ocean Model

Based on the high-resolution Regional Ocean Modeling System(ROMS) and the conditional nonlinear optimal perturbation(CNOP) method, this study explored the effects of optimal initial errors on the prediction of the Kuroshio large meander(LM) path, and the growth mechanism of optimal initial errors was revealed. For each LM event, two types of initial error(denoted as CNOP1 and CNOP2) were obtained. Their large amplitudes were found located mainly in the upper 2500 m in the upstream region of the LM, i.e., southeast of Kyushu. Furthermore, we analyzed the patterns and nonlinear evolution of the two types of CNOP. We found CNOP1 tends to strengthen the LM path through southwestward extension. Conversely,CNOP2 has almost the opposite pattern to CNOP1, and it tends to weaken the LM path through northeastward contraction.The growth mechanism of optimal initial errors was clarified through eddy-energetics analysis. The results indicated that energy from the background field is transferred to the error field because of barotropic and baroclinic instabilities. Thus, it is inferred that both barotropic and baroclinic processes play important roles in the growth of CNOP-type optimal initial errors.

Possible Sources of Forecast Errors Generated by the Global/Regional Assimilation and Prediction System for Landfalling Tropical Cyclones. Part Ⅱ: Model Uncertainty

This paper investigates the possible sources of errors associated with tropical cyclone(TC) tracks forecasted using the Global/Regional Assimilation and Prediction System(GRAPES). In Part I, it is shown that the model error of GRAPES may be the main cause of poor forecasts of landfalling TCs. Thus, a further examination of the model error is the focus of Part II.Considering model error as a type of forcing, the model error can be represented by the combination of good forecasts and bad forecasts. Results show that there are systematic model errors. The model error of the geopotential height component has periodic features, with a period of 24 h and a global pattern of wavenumber 2 from west to east located between 60?S and 60?N. This periodic model error presents similar features as the atmospheric semidiurnal tide, which reflect signals from tropical diabatic heating, indicating that the parameter errors related to the tropical diabatic heating may be the source of the periodic model error. The above model errors are subtracted from the forecast equation and a series of new forecasts are made. The average forecasting capability using the rectified model is improved compared to simply improving the initial conditions of the original GRAPES model. This confirms the strong impact of the periodic model error on landfalling TC track forecasts. Besides, if the model error used to rectify the model is obtained from an examination of additional TCs, the forecasting capabilities of the corresponding rectified model will be improved.

Regional Prediction of Impending Debris Flow Based on Doppler Weather Radar

Debris flow prediction is one of the important means to reduce the loss caused by debris flow. This paper built a regional prediction model of impending debris flow based on regional environmental background （including topography, geology, land use, and etc.）, rainfall and debris flow data. A system of regional prediction of impending debris flow was set up on ArcGIS 9.0 platform according to the model. The system used forecast precipitation data of Doppler weather radar and observational precipitation data as its input data. It could provide a prediction about the possibility of debris flow one to three hours before it happened, and was put into use in Liangshan Meteorological Observatory in Sichuan province in the monsoon of 2006.

The ‘Two oceans and one sea' extended range numerical prediction system with an ultra-high resolution atmosphere-ocean-land regional coupled model

The‘Two Oceans and One Sea’area(West Pacific,Indian Ocean,and South China Sea;15°S–60°N,39°–178°E)is a core strategic area for the‘21st Century Maritime Silk Road’project,as well as national defense.With the increasing demand for disaster prevention and mitigation,the importance of 10–30-day extended range prediction,between the conventional short-term(around seven days)and the climate scale(longer than one month),is apparent.However,marine extended range prediction is still a‘blank point’in China,making the early warning of marine disasters almost impossible.Here,the authors introduce a recently launched Chinese national project on a numerical forecasting system for extended range prediction in the‘Two Oceans and One Sea’area based on a regional ultra-high resolution multi-layer coupled model,including the scientific aims,technical scheme,innovation,and expected achievements.The completion of this prediction system is of considerable significance for the economic development and national security of China.

Evaluation of a Regional Ensemble Data Assimilation System for Typhoon Prediction

An ensemble Kalman filter(EnKF) combined with the Advanced Research Weather Research and Forecasting model(WRF) is cycled and evaluated for western North Pacific(WNP) typhoons of year 2016. Conventional in situ data, radiance observations, and tropical cyclone(TC) minimum sea level pressure(SLP) are assimilated every 6 h using an 80-member ensemble. For all TC categories, the 6-h ensemble priors from the WRF/EnKF system have an appropriate amount of variance for TC tracks but have insufficient variance for TC intensity. The 6-h ensemble priors from the WRF/EnKF system tend to overestimate the intensity for weak storms but underestimate the intensity for strong storms. The 5-d deterministic forecasts launched from the ensemble mean analyses of WRF/EnKF are compared to the NCEP and ECMWF operational control forecasts. Results show that the WRF/EnKF forecasts generally have larger track errors than the NCEP and ECMWF forecasts for all TC categories because the regional simulation cannot represent the large-scale environment better than the global simulation. The WRF/EnKF forecasts produce smaller intensity errors and biases than the NCEP and ECMWF forecasts for typhoons, but the opposite is true for tropical storms and severe tropical storms. The 5-d ensemble forecasts from the WRF/EnKF system for seven typhoon cases show appropriate variance for TC track and intensity with short forecast lead times but have insufficient spread with long forecast lead times. The WRF/EnKF system provides better ensemble forecasts and higher predictability for TC intensity than the NCEP and ECMWF ensemble forecasts.

Study of perturbing method in regional BGM ensemble prediction system

Based on an Ensemble Prediction System with the BGM method on the regional numerical prediction model AREM, Static State Perturbation （SSP, including Initial Random Perturbation and Perturbation Restriction） and Dynamic State Perturbation （DSP） are designed. The impacts of both perturbations on precipitation prediction are studied by analyzing a slrong precipitation process oc- curring during July 20-21, 2008. The results show that both SSP and DSP play a positive role in prediction of mesoscale precipita- tion, such as lowering the （missing） rate of precipitation prediction. SSP is mainly helpful for the 24-hour prediction, while DSP can improve both 24-hour and 48-hour prediction. DSP is better than the two SSPs in the hit rate of regional precipitation prediction. However, the former also has a little higher false alarm rate than the latter. DSP enlarges in some extent the dispersion of EPS, which is good for EPS.

GACNet: A Generative Adversarial Capsule Network for Regional Epitaxial Traffic Flow Prediction

With continuous urbanization,cities are undergoing a sharp expansion within the regional space.Due to the high cost,the prediction of regional traffic flow is more difficult to extend to entire urban areas.To address this challenging problem,we present a new deep learning architecture for regional epitaxial traffic flow prediction called GACNet,which predicts traffic flow of surrounding areas based on inflow and outflow information in central area.The method is data-driven,and the spatial relationship of traffic flow is characterized by dynamically transforming traffic information into images through a two-dimensional matrix.We introduce adversarial training to improve performance of prediction and enhance the robustness.The generator mainly consists of two parts:abstract traffic feature extraction in the central region and traffic prediction in the extended region.In particular,the feature extraction part captures nonlinear spatial dependence using gated convolution,and replaces the maximum pooling operation with dynamic routing,finally aggregates multidimensional information in capsule form.The effectiveness of the method is evaluated using traffic flow datasets for two real traffic networks:Beijing and New York.Experiments on highly challenging datasets show that our method performs well for this task.

Effectiveness analysis of the prediction of regional debris flow susceptibility in post-earthquake and drought site

Debris-flow disasters occurred frequently after the Mw 8.0 Wenchuan earthquake on 12 May 2008 in Sichuan Province, China. Based on historical accounts of debris-flow disaster events, it found that debris flow occurrence is closely related to the impact of earthquakes and droughts, because earthquakedrought activities can increase the loose solid materials, which can transform into debris flows under the effect of rainstorms. Based on the analysis of historical earthquake activity(frequency, magnitude and location), drought indexes and the trend of climate change(amount of rainfall), a prediction method was established, and the regional debris flow susceptibility was predicted. Furthermore, in a debris flow-susceptible site, effective warning and monitoring are essential not only from an economicpoint of view but are also considered as a frontline approach to alleviate hazards. The advantages of the prediction and early monitoring include(1) the acquired results being sent to the central government for policy making;(2) lives and property in mountainous areas can be protected, such as the 570 residents in the Aizi valley, who evacuated successfully before debris flows in 2012;(3) guiding the government to identify the areas of disasters and the preparation for disaster prevention and mitigation, such as predicting disasters in high-risk areas in the period 2012-2017, helping the government to recognize the development trend of disasters;(4) the quantitative prediction of regional debris-flow susceptibility, such as after the Wenchuan earthquake, can promote scientific and sustainable development and socioeconomic planning in earthquake-struck areas.

Analysis and Prediction of Regional Electricity Consumption Based on BP Neural Network

Electricity consumption forecasting is one of the most important tasks for power system workers,and plays an important role in regional power systems.Due to the difference in the trend of power load and the past in the new normal,the influencing factors are more diversified,which makes it more difficult to predict the current electricity consumption.In this paper,the grey system theory and BP neural network are combined to predict the annual electricity consumption in Jiangsu.According to the historical data of annual electricity consumption and the six factors affecting electricity consumption,the gray correlation analysis method is used to screen the important factors,and three factors with large correlation degree are selected as the input parameters of BP neural network.The power forecasting model uses nearly 18 years of data to train and validate the model.The results show that the gray correlation analysis and BP neural network method have higher accuracy in power consumption prediction,and the calculation is more convenient than traditional methods.

SOUTH CHINA REGIONAL SHORT RANGE CLIMATE PREDICTION MODEL AND ITS PERFORMANCE

In this paper, a newly established "South China Regional Short Range Climate Prediction Model System" is introduced and its performance is analyzed in real case simulation. It shows that the system has a good performance and suitable for short range climate modeling. The model simulates well the monthly mean, pentad mean and daily field, pentad mean and daily field and can depict more details than coarse resolution analyses. Weather systems and information can pass into and out of the model domain through lateral boundaries without notable damping. Almost all of the weather and climate changes can be reflected in the simulation, in which both the changing tendencies, amplitudes, speeds, and phases are consistent with the real cases. The simulated precipitation is much close to the observed one, both in the extent, position and in the intensity of rainfall. In addition, some smaller precipitation centers could also be reflected in the simulation.