Time series prediction of reservoir bank landslide failure probability considering the spatial variability of soil properties

Historically,landslides have been the primary type of geological disaster worldwide.Generally,the stability of reservoir banks is primarily affected by rainfall and reservoir water level fluctuations.Moreover,the stability of reservoir banks changes with the long-term dynamics of external disastercausing factors.Thus,assessing the time-varying reliability of reservoir landslides remains a challenge.In this paper,a machine learning(ML)based approach is proposed to analyze the long-term reliability of reservoir bank landslides in spatially variable soils through time series prediction.This study systematically investigated the prediction performances of three ML algorithms,i.e.multilayer perceptron(MLP),convolutional neural network(CNN),and long short-term memory(LSTM).Additionally,the effects of the data quantity and data ratio on the predictive power of deep learning models are considered.The results show that all three ML models can accurately depict the changes in the time-varying failure probability of reservoir landslides.The CNN model outperforms both the MLP and LSTM models in predicting the failure probability.Furthermore,selecting the right data ratio can improve the prediction accuracy of the failure probability obtained by ML models.

Prediction and Output Estimation of Pattern Moving in Non-Newtonian Mechanical Systems Based on Probability Density Evolution

A prediction framework based on the evolution of pattern motion probability density is proposed for the output prediction and estimation problem of non-Newtonian mechanical systems,assuming that the system satisfies the generalized Lipschitz condition.As a complex nonlinear system primarily governed by statistical laws rather than Newtonian mechanics,the output of non-Newtonian mechanics systems is difficult to describe through deterministic variables such as state variables,which poses difficulties in predicting and estimating the system’s output.In this article,the temporal variation of the system is described by constructing pattern category variables,which are non-deterministic variables.Since pattern category variables have statistical attributes but not operational attributes,operational attributes are assigned to them by posterior probability density,and a method for analyzing their motion laws using probability density evolution is proposed.Furthermore,a data-driven form of pattern motion probabilistic density evolution prediction method is designed by combining pseudo partial derivative(PPD),achieving prediction of the probability density satisfying the system’s output uncertainty.Based on this,the final prediction estimation of the system’s output value is realized by minimum variance unbiased estimation.Finally,a corresponding PPD estimation algorithm is designed using an extended state observer(ESO)to estimate the parameters to be estimated in the proposed prediction method.The effectiveness of the parameter estimation algorithm and prediction method is demonstrated through theoretical analysis,and the accuracy of the algorithm is verified by two numerical simulation examples.

A Cascading Fault Path Prediction Method for Integrated Energy Distribution Networks Based on the Improved OPA Model under Typhoon Disasters

In recent times,the impact of typhoon disasters on integrated energy active distribution networks(IEADNs)has received increasing attention,particularly,in terms of effective cascading fault path prediction and enhanced fault recovery performance.In this study,we propose a modified ORNL-PSerc-Alaska(OPA)model based on optimal power flow(OPF)calculation to forecast IEADN cascading fault paths.We first established the topology and operational model of the IEADNs,and the typical fault scenario was chosen according to the component fault probability and information entropy.The modified OPA model consisted of two layers:An upper-layer model to determine the cascading fault location and a lower-layer model to calculate the OPF by using Yalmip and CPLEX and provide the data to update the upper-layer model.The approach was validated via the modified IEEE 33-node distribution system and two real IEADNs.Simulation results showed that the fault trend forecasted by the novel OPA model corresponded well with the development and movement of the typhoon above the IEADN.The proposed model also increased the load recovery rate by>24%compared to the traditional OPA model.

Performance assessment of genetic programming(GP)and minimax probability machine regression(MPMR)for prediction of seismic ultrasonic attenuation

The determination of seismic attenuation （s） （dB/cm） is a challenging task in earthquake science. This article employs genetic programming （GP） and minimax probability machine regression （MPMR） for prediction of s. GP is developed based on genetic algo- rithm. MPMR maximizes the minimum probability of future predictions being within some bound of the true regression function. Porosity （n） （%）, permeability （k） （millidarcy）, grain size （d） （μm）, and clay content （c） （%） have been considered as inputs of GP and MPMR. The output of GP and MPMR is s. The developed GP gives an equation for prediction of s. The results of GP and MPMR have been compared with the artificial neural net- work. This article gives robust models based on GP and MPMR for prediction of s.

Probability Prediction Model for Landslide Occurrences in Xi'an, Shaanxi Province, China

Landslides are increasing since the 1980s in Xi'an, Shaanxi Province, China. This is due to the increase of the frequency and intensity of precipitation caused by complex geological structures, the presence of steep landforms, seasonal heavy rainfall, and the intensifcation of human activities. In this study, we propose a landslide prediction model based on the analysis of intraday rainfall(IR) and antecedent effective rainfall(AER). Primarily, the number of days and degressive index of the antecedent effective rainfall which affected landslide occurrences in the areas around Qin Mountains, Li Mountains and Loess Tableland was established. Secondly, the antecedent effective rainfall and intraday rainfall were calculated from weather data which were used to construct critical thresholds for the 10%, 50% and 90% probabilities for future landslide occurrences in Qin Mountain, Li Mountain and Loess Tableland. Finally, the regions corresponding to different warning levels were identified based on the relationship between precipitation and the threshold, that is; "A" region is safe, "B" region is on watch alert, "C" region is on warning alert and "D" region is on severe warning alert. Using this model, a warning program is proposed which can predict rainfall-induced landslides by means of real-time rain gauge data and real-time geo-hazard alert and disaster response programs. Sixteen rain gauges were installed in the Xi'an region by keeping in accordance with the regional geology and landslide risks. Based on the data from gauges, this model accurately achieves the objectives of conducting real-time monitoring as well as providing early warnings of landslides in the Xi'an region.

Prediction of chaotic time series based on modified minimax probability machine regression

Long-term prediction of chaotic time series is very difficult,for the Chaos restricts predictability.in this paper a new method is studied to model and predict chaotic time series based on minimax probability machine regression （MPMR）. Since the positive global Lyapunov exponents lead the errors to increase exponentially in modelling the chaotic time series, a weighted term is introduced to compensate a cost function. Using mean square error （MSE） and absolute error （AE） as a criterion, simulation results show that the proposed method is more effective and accurate for multistep prediction. It can identify the system characteristics quite well and provide a new way to make long-term predictions of the chaotic time series.

Application of probability integral method in ground deformation prediction

In order to study the law of mining subsidence and ground movement, to provide the basis of coal mining under building, railway and water, we used the probability integration method to make comprehensive evaluation of ground stability. Take Yingcheng Coal Mine of Jiutai as an example. Mining-induced movement and horizontal movement are analyzed on the basis of the measurement data. The resuhs of prediction can pro- vide reference and basis for prevention of coal mining subsidence and future restoration and treatment.

Fatigue Life Prediction of Horizontal Press Frame Based on Statistical Probability and Its Redesign

Horizontal press as an important part of hydro-forming machine is used to output the horizontal force to keep the high internal pressure during tube hydro-forming. However,the horizontal press frame is usually mounted on the press bed and not pre-stressed. Meanwhile it will be subjected to the reaction force caused by liquid pressure. Stresses are concentrated severely on the assemble region due to deformation,and total fatigue life will decrease. In order to predict the total fatigue life of the frame,the simulations are used firstly to determine to stress concentration region,and then strain gauge measurements are carried out under different loads. Next,the methods of statistical probability are conducted to calculate the fatigue life based on long-term load history. Finally a structure with the considerable longer fatigue life is redesigned.

ICIC_Prediction:基于因果关系全局动态特性的预测方法

因果关系的预测是因果关系研究的重要内容和主要应用。现有的很多预测方法以寻找最优预测方程或最小特征变量集合为目的,以简化计算。提出一种新的可用于处理政策干预的因果关系预测方法ICIC_Prediction,不局限于利用马尔科夫毯等特征变量集合,而是从因果关系网络结构出发,利用因果关系系统及其采样数据的动态全局特性,预测目标变量在当前采样中的取值。通过在NIPS 2008"因果与预测"的评测会议上发布的四个不同类型的数据集上的对比实验,分析并展示了ICIC_Prediction方法的优势和特点。

Improving Multi-model Ensemble Probabilistic Prediction of Yangtze River Valley Summer Rainfall

Seasonal prediction of summer rainfall over the Yangtze River valley（YRV） is valuable for agricultural and industrial production and freshwater resource management in China, but remains a major challenge. Earlier multi-model ensemble（MME） prediction schemes for summer rainfall over China focus on single-value prediction, which cannot provide the necessary uncertainty information, while commonly-used ensemble schemes for probability density function（PDF） prediction are not adapted to YRV summer rainfall prediction. In the present study, an MME PDF prediction scheme is proposed based on the ENSEMBLES hindcasts. It is similar to the earlier Bayesian ensemble prediction scheme, but with optimization of ensemble members and a revision of the variance modeling of the likelihood function. The optimized ensemble members are regressed YRV summer rainfall with factors selected from model outputs of synchronous 500-h Pa geopotential height as predictors. The revised variance modeling of the likelihood function is a simple linear regression with ensemble spread as the predictor. The cross-validation skill of 1960–2002 YRV summer rainfall prediction shows that the new scheme produces a skillful PDF prediction, and is much better-calibrated, sharper, and more accurate than the earlier Bayesian ensemble and raw ensemble.

Progress and Challenge of the Short-Term Climate Prediction

The experience of developing a short-term climate prediction system at the Institute of Atmospheric Science of the Chinese Academy of Sciences is summarized,and some problems to be solved in future are discussed in this paper.It is suggested that a good system for short-term climate prediction should at least consist of (1) well-tested model(s),(2) sufficient data and good methods for the initialization and assimilation,(3) a good system for quantitative corrections,(4) a good ensemble prediction method,and (5) appropriate prediction products,such as mathematical expectation,standard deviation,probability,among others.

Pattern recognition prediction of coal and gas outburst hazard in the sixth mine of Hebi

Based on the systematical analysis influence factors of coal and gas outburst, the main factors and their magnitude was determined by the corresponding methods.With the research region divided into finite predicting units,the internal relation between the factors and the hazard of coal and gas outburst,that was combination model of influence factors,was ascertained through multi-factor pattern recognition method.On the basis of contrastive analysis the pattern of coal and gas outburst between prediction region and mined region,the hazard of every predication unit was determined.The mining area was then divided into coal and gas outburst dangerous area,threaten area and safe area re- spectively according to the hazard of every predication unit.Accordingly the hazard of mining area is assessed.

Typhoon/Hurricane/Tropical Cyclone Disasters: Prediction, Prevention and Mitigation

Since 1972 Rita typhoon attacked on Dalian Port and induced severe catastrophe, we were studied on statistical prediction model of typhoon induced wave height and wind speed. With an increasing tendency of the natural hazards frequency and intensity, risk assessment of some design codes for coastal defence infrastructures should be of paramount importance influencing the economic development and a lot of lifes in China. Comparison between existing extreme statistical model like Gumbel, Weibull, P-III distribution or Probable Maximum Typhoon/Hurricane (PMT/PMH), Design Basis Flood (DBF) with our 1975-1980 proposed (CEVD) model showed that all the planned, designed and constructed coastal infrastructures accepted the traditional safety regulations are menaced by possibility of future ty-phoon/hurricane disasters and cannot satisfy the safety requirements with the increasing tendency of the extreme natural hazards. Our first publication in US (J. of Waterway Port Coastal & Ocean Eng. ASCE, 1980, ww4) proposed an new model “Compound Extreme Value Distribution” used for China sea, after then the model was used in “Long term Distribution of Hurricane Characteristics” for Gulf of Mexico & Atlantic coasts, U.S. (OTC.1982). 2005 hurricane Katrina, Rita and 2012 hurricane Sandy induced disasters proved 1982 CEVD and CEVD has been developed into Multivariate Compound Extreme Value Distribution (MCEVD). 2006 MCEVD predicted extreme hazards in New Orleans, Gulf of Mexico and Philadelphia areas. 2013 typhoon Fitow induced disaster in China also proved MCEVD 2006 predicted results.

Preliminary Study on Probabilistic Prediction of Seismic Hazard in a Period of 10 Years

Many uncertainty factors need be dealt with in the prediction of seismic hazard for a 10-year period.Restricted by these uncertainties,the result of prediction is also uncertain to a certain extent,so the probabilistic analysis method of seismic hazard should be adopted.In consideration of the inhomogeneity of the time,location,and magnitude of future earthquakes and the probabilistic combination of the background of long-term seismic hazard(geology,geophysical field,etc.)and the precursors of earthquake occurrence,a model of probabilistic prediction of seismic hazard in a period of 10 years s proposed.Considering the inhomogeneity of data and earthquake precursors for different regions in China,a simplified model is also proposed in order to satisfy the needs of different regions around the country.A trial in North China is used to discuss the application of the model.The method proposed in this paper can be used in the probabilistic prediction of seismic hazard in a period of 10 years.According to

Prediction of Smear Positive TB Cases at Different Types of Designated Microscopy Centres, Karnataka, India

Background: Under the Revised National Tuberculosis control Programme (RNTCP) in India, the designated microscopy centres (DMCs) form the basic unit of smear positive TB case detection in a district. There is a need by the programme managers to estimate the mean and range of smear positive tuberculosis (TB) cases that can be detected at DMCs located in different type of health facilities to channelize their resources. Methods: It is a cross-sectional study conducted in the state of Karnataka, India during January 2014 to December 2014 based on the compiled reports from past five years received from all the 30 districts of the state. The prediction was made based on the performance of these DMCs in the last five years using a modeling technique. Results: The proportions of the DMCs located at health facilities are Primary Health Institutions/Centres (PHIs)—73%, Tuberculosis Units (TUs)—15%, Medical colleges (MC)—7%, District TB centres (DTC)—3% and Private Practitioners (PP)—2%. The maximum number of cases that can be detected at DTC is 3621 (SD 54), TU is 9224 (SD 90), PHI is 20,412 (SD 135), PP is 859 (SD 26) and MC is 8322 (SD 84). Conclusion: The predicted values will essentially serve as a tool for the programme managers of Karnataka to plan, strategize and monitor the performance of DMCs in the state.

Bayesian prediction of potential depressions in the Erlian Basin based on integrated geophysical parameters

In this study,we analyzed the geological,gravity,magnetic,and electrical characteristics of depressions in the Erlian Basin.Based on the results of these analyses,we could identify four combined feature parameters showing strong correlations and sensibilities to the reservoir oil-bearing conditions:the average residual gravity anomaly,the average magnetic anomaly,the average depth of the conductive key layer,and the average elevation of the depressions.The feature parameters of the 65 depressions distributed in the whole basin were statistically analyzed:each of them showed a Gaussian distribution and had the basis of Bayesian theory.Our Bayesian predictions allowed the defi nition of a formula to calculate the posterior probability of oil occurrence in the depressions based on the combined characteristic parameters.The feasibility of this prediction method was verifi ed by considering the results obtained for the 22 drilled depressions.Subsequently,we were able to determine the oilbearing threshold of hydrocarbon potential for the depressions in the Erlian Basin,which can be used as a standard for quantitative optimizations.Finally,the proposed prediction method was used to calculate the probability of hydrocarbons in the other 43 depressions.Based on this probability and on the oil-bearing threshold,the fi ve depressions with the highest potential were selected as targets for future seismic explorations and drilling.We conclude that the proposed method,which makes full use of massive gravity,magnetic,electric,and geological data,is fast,eff ective,and allows quantitative optimizations;hence,it will be of great value for the comprehensive geophysical evaluation of oil and gas in basins with depression group characteristics.

The new method of prediction on mining subsidence and deformation

A new probability function of mining overlying strata and subsidence is put forward that has a general statistical significance based on the ideal stochastic medium displacement model. It establishes a new system of prediction on horizontal mining subsidence and deformation, which gives a new method for prediction on mining subsidence and deformation.

Synthetic Model for Prediction Seismic Risk in About 10 Years

In this paper, the relation that strong earthquakes and seismo-geological conditions bear with precursory phenomena has been analyzed. The information content concerning the long-term (i.e., a decade) seismic risk that various precursors can provide has been estimated by using the quantitative judgment method. On such a basis, a synthetic probability model for predicting the strong earthquake risk in about 10 years has been suggested. With the northern part of North China used as the test region, the feasibility of the model which is used for medium-term to long-term prediction has been proved.

Efficient Coding Unit and Prediction Unit Decision Algorithm for Multiview Video Coding

To aim at higher coding efficiency for multiview video coding, the multiview video with a modified high efficiency video coding（MV-HEVC）codec is proposed to encode the dependent views.However, the computational complexity of MV-HEVC encoder is also increased significantly since MV-HEVC inherits all computational complexity of HEVC. This paper presents an efficient algorithm for reducing the high computational complexity of MV-HEVC by fast deciding the coding unit during the encoding process. In our proposal, the depth information of the largest coding units（LCUs） from independent view and neighboring LCUs is analyzed first. Afterwards, the analyzed results are used to early determine the depth for dependent view and thus achieve computational complexity reduction. Furthermore, a prediction unit（PU） decision strategy is also proposed to maintain the video quality. Experimental results demonstrate that our algorithm can achieve 57% time saving on average,while maintaining good video quality and bit-rate performance compared with HTM8.0.

Model of Probabilistic Prediction for Life Loss

Based on discussion of the reasonableness of the seismic destruction described by using the death degree,the probabilistic curves of the different death degrees with different future times in several regions on the Chinese mainland have been obtained by applying the probabilistic model,which is in accord with present seismic destruction data and related data.The historical data of the life loss have been processed with the dynamic system model.Finally,the results from two kinds of data have been analyzed and discussed.