Probabilistic neural network algorithm for using radon emanations as an earthquake precursor

The investigation throughout the world in past two decades provides evidence which indicate that significance variation of radon and other soil gases occur in association with major geophysical events such as earthquake. The traditional statistical algorithm includes regression to remove the effect of the meteorological parameters from the raw radon and anomalies are calculated either taking the periodicity in seasonal variations or periodicity computed using Fast Fourier Transform. In case of neural networks the regression step is avoided. A neural network model can be found which can learn the behavior of radon with respect to meteorological parameter in order that changing emission patterns may be adapted to by the model on its own. The output of this neural model is the estimated radon values. This estimated radon value is used to decide whether anomalous behavior of radon has occurred and a valid precursor may be identified. The neural network model developed using Radial Basis function network gave a prediction rate of 87.7%. The same was accompanied by huge false alarms. The present paper deals with improved neural network algorithm using Probabilistic Neural Networks that requires neither an explicit step of regression nor use of any specific period. This neural network model reduces the false alarms to zero and gave same prediction rate as RBF networks.     

Natural disaster management mechanisms for probabilistic earthquake loss

High rates of urbanization, environmental degradation, and industrial development have affected all nations worldwide, but in disaster-prone areas, the impact is even greater serving to increase the extent of damage from natural catastrophes. As a result of the global nature of environmental change, modern economies have had to adapt, and sustainability is an extremely important issue. Clearly, natural disasters will affect the competitiveness of an enterprise. This study focuses on natural disaster management in an area in which the direct risks are posed by the physical effects of natural disasters such as floods, droughts, tsunamis, and rising sea levels. On a local level, the potential impact of a disaster on a company and how much damage (loss) it causes to facilities and future business are of concern. Each company must make plans to mitigate predictable risk. Risk assessments must be completed in a timely manner. Disaster management is also very important to national policy. Natural disaster management mechanisms can include strategies for disaster prevention, early warning (prediction) systems, disaster mitigation, preparedness and response, and human resource development. Both governmental administration (public) and private organizations should participate in these programs. Participation of the local community is especially important for successful disaster mitigation, preparation for, and the implementations of such measures. Our focus in this study is a preliminary proposal for developing an efficient probabilistic approach to facilitate design optimization that involves probabilistic constraints.     

Rock mass excavatability estimation using artificial neural network

One important decision in design of surface mine is the selection of mine equipment and plant. Demand for mechanical excavation is growing in mining industry because of its high productivity and excavation in large scale with lower costs. Several models have been developed over the years to evaluate the ease of excavation and machine performance against rock mass properties. Due to complexity of excavation process and large number of effective parameters, approaches made for this purpose are essentially empirical. There are many uncertainties in results of these models. An attempt is made in this paper to revise the exisiting models. Neural network models for estimation of rock mass excavatability and production rate of VASM-2D excavating machine at Limestone quarry in Retznei, Austria, is presented. Input parameters of this model are Uniaxial compressive strength, tensile strength and discontinuities spacing of rocks. Output is the specific excavation rate per power consumption (bcm/Kwh) as the productivity indicator. Average of deviation between actual data and results estimated by neural network model was only 15% which is in an acceptable range.     

Seismic loss estimation tool as rapid survey for prioritizing buildings for disaster preparedness: case study to hospital buildings

Natural Hazards - Hospital buildings must be fully operational after the earthquake to protect the lives of patients as well as to provide emergency care and medical treatment to the victims....     

基于RS/GIS的城市财产保险洪灾损失评估研究

如何快速,有效地进行投保户洪涝灾害损失评估是保险行业急需解决的一个重要课题.从洪涝灾害的成灾机理出发,针对保险公司对具体受灾体理赔需求,提出了计算每个投保户洪灾损失率方法,建立基于遥感(RS)和地理信息系统(GIS)的城市财产保险洪涝灾害损失评估模型.建模时较全面地考虑了与投保物性质有关的承灾体易损度和与投保物所处环境有关的地基承载力等因素,并使用RS/GIS将其定量化提取.最后使用广东省深圳市洪灾数据进行模型检验.验证结果表明,模型对于各个投保户均能得到较好的精度.     

Estimation of radon as an earthquake precursor: A neural network approach

An artificial neural networks (ANN) approach combined with Fourier Transform based selection of time period in the time series Radon Emission Data has been presented and shown to improve event prediction rates and reduce false alarms in Earthquake Event Identification over the traditional multiple linear regression techniques. The paper presents a neural networks system using radial basis function (RBF) network as an alternative to traditional statistical regression technique in isolating Radon Emission Anomaly caused by seismic activities. The RBF model has been developed to accept and predict earthquakes events based on a known data set of Radon Emanation, Metrological parameters and actual earthquake events. Subsequently, the model was tested and evaluated on a future data set and a prediction rate of 87.8%, if a reduced false alarm was achieved, the results obtained are better than the traditional techniques.     

Seismic random noise attenuation using artificial neural network and wavelet packet analysis

In this paper, we present a method for attenuating background random noise and enhancing resolution of seismic data, which takes advantage of semi-automatic training of feed forward back propagation (FFBP) artificial neural network (ANN) in a multiscale domain obtained from wavelet packet analysis (WPA). The images of approximations and details of the input seismic sections are calculated and utilized to train neural network to model coherent events by an automatic algorithm. After the modeling of coherent events, the remainder data are assumed to be related to background random noise. The proposed method is applied on both synthetic and real seismic data. The results are compared with that of the adaptive Wiener filter (AWF) in synthetic shot gather and real common midpoint gather and also with that of band-pass filtering on real common offset gather. The comparison indicates substantially higher efficiency of the proposed method in attenuating random noise and enhancing seismic signals.     

Scenario-based earthquake loss estimation for the city of Cairo,Egypt

This paper describes a multi-tiered loss assessment methodology to estimate seismic monetary implications resulting from structural damage to the building population in Greater Cairo. After outlining a ground-shaking model, data on geological structures and surface soil conditions are collated using a considerable number of boreholes to produce a classification of different soil deposits. An inventory database for the existing building stock is also prepared. The seismic vulnerability of representative reinforced concrete building models, designed according to prevalent codes and construction practices, is evaluated. Capacity spectrum methods are utilised for assessing the structural performance through a multi-level damage scale. A simplified methodology for deriving fragility curves for non-ductile reinforced concrete building classes that typically constitute the building population of the city is adopted. In addition, suitable fragility functions for unreinforced masonry constructions are selected and used for completing the loss model for the study area. The results are finally used to build an event-based loss model caused by possible earthquakes in the region.     

Seismic damage and loss estimation for central Indo-Gangetic Plains,India

Natural Hazards - The Indo-Gangetic Plains which lies between the Himalayan mountain ranges and peninsular India is considered to be the region of great concern due to its thick sediments and...     

Application of neural network and ANFIS model for earthquake occurrence in Iran

This study examined the spatial-temporal variations in seismicity parameters for the September 10^th, 2008 Qeshm earthquake in south Iran. To this aim, artificial neural networks and Adaptive Neural Fuzzy Inference System (ANFIS) were applied. The supervised Radial Basis Function (RBF) network and ANFIS model were implemented because they have shown the efficiency in classification and prediction problems. The eight seismicity parameters were calculated to analyze spatial and temporal seismicity pattern. The data preprocessing that included normalization and Principal Component Analysis (PCA) techniques was led before the data was fed into the RBF network and ANFIS model. Although the accuracy of RBF network and ANFIS model could be evaluated rather similar, the RBF exhibited a higher performance than the ANFIS for prediction of the epicenter area and time of occurrence of the 2008 Qeshm main shock. A proper training on the basis of RBF network and ANFIS model might adopt the physical understanding between seismic data and generate more effective results than conventional prediction approaches. The results of the present study indicated that the RBF neural networks and the ANFIS models could be suitable tools for accurate prediction of epicenteral area as well as time of occurrence of forthcoming strong earthquakes in active seismogenic areas.     

Integrated model for earthquake risk assessment using neural network and analytic hierarchy process:Aceh province,Indonesia

Catastrophic natural hazards,such as earthquake,pose serious threats to properties and human lives in urban areas.Therefore,earthquake risk assessment(ERA)is indispensable in disaster management.ERA is an integration of the extent of probability and vulnerability of assets.This study develops an integrated model by using the artificial neural network–analytic hierarchy process(ANN–AHP)model for constructing the ERA map.The aim of the study is to quantify urban population risk that may be caused by impending earthquakes.The model is applied to the city of Banda Aceh in Indonesia,a seismically active zone of Aceh province frequently affected by devastating earthquakes.ANN is used for probability mapping,whereas AHP is used to assess urban vulnerability after the hazard map is created with the aid of earthquake intensity variation thematic layering.The risk map is subsequently created by combining the probability,hazard,and vulnerability maps.Then,the risk levels of various zones are obtained.The validation process reveals that the proposed model can map the earthquake probability based on historical events with an accuracy of 84%.Furthermore,results show that the central and southeastern regions of the city have moderate to very high risk classifications,whereas the other parts of the city fall under low to very low earthquake risk classifications.The findings of this research are useful for government agencies and decision makers,particularly in estimating risk dimensions in urban areas and for the future studies to project the preparedness strategies for Banda Aceh.     

应用一种改进BP神经网络算法预测密度曲线

利用测井数据与地震数据二者相结合进行综合分析,是地震勘探工作者的重要工作。可以通过分析井位处的地震数据与测井数据,提取地震的多个属性,建立一个与测井属性的统计关系。选取已改进的三层网络结构BP神经网络算法,在应用一个实际例子后表明,该算法的主要特点是收敛速度快、计算简单,同时还具有跳出局部最小的能力。应用此神经网络算法对某油田的二维地震数据进行了处理,提取了多种地震属性,并在井位置建立了地震属性与密度曲线的非线性关系,成功预测了剖面密度曲线,为了解储层状况提供了有益的资料。     

Investigation of general regression neural network architecture for grade estimation of an Indian iron ore deposit

The mineral resource estimation requires accurate prediction of the grade at location from limited borehole information. It plays the dominant role in the decision-making process for investment and development of various mining projects and hence become an important and crucial stage. This paper evaluvates the use of two distinct artificial neural network (ANN)-based models, general regression neural network (GRNN) and multilayer perceptron neural network (MLP NN), to improve the grade estimation from Koira iron ore region in Sundargarh district, Odisha. ANN-based models capture the inherent complex structure of mineral deposits and provide a reliable generalization of the iron grade. The ANN-based approach does not require any preliminary geological study and is free from any statistical assumption on the raw data before its application. The GRNN is a one-pass learning algorithm and does not require any iterative procedure for training less complex structure and requires only one learning parameter for optimization. In this investigation, the spatial coordinates and multiple lithological units were taken as input variables and the iron grade was taken as the output variable. The comparative analysis of these models has been carried out and the results obtained were validated with traditional geostatistical method ordinary kriging (OK). The GRNN model outperforms the other methods, i.e. MLP and OK, with respect to generalization and predictability of the grades at an un-sampled location.     

Seismic hazard evaluation at Algiers using Benouar's earthquake catalogue

This paper presents the evaluation of seismic hazard at the site of Algiers (capital of Algeria). In order to implement earthquake-resistant design codes, it is usually necessary to know the maximum dynamic load which a particular structure might experience during its economic life, or alternatively, the most probable return period of a specified design load. The evaluation of the seismic hazard at the site, based on peak ground motion acceleration and using Cornell's method and Benouar's earthquake Maghreb catalogue, in terms of return period, probability of exceedance of PGA, design ground motion and a response spectrum, is carried out for the City of Algiers and its surroundings. The response spectrum for Algiers presented in this paper is the first one realized in Algeria using revised Algerian data.     

Impact of local site conditions on portfolio earthquake loss estimation for different building types

This article presents a sensitivity analysis investigating the impact of using high-resolution site conditions databases in portfolio earthquake loss estimation. This article also estimates the effects of variability in the site condition databases on probabilistic earthquake loss ratios and their geographical pattern with respect to structural characteristics of different building types. To perform the earthquake loss estimation here, the OpenQuake software developed by Global Earthquake Model is implemented in Clemson University’s supercomputer. The probabilistic event-based risk analysis is employed considering several notional portfolios of different building types in the San Francisco area as the inventory exposure. This analysis produces the stochastic event sets worth for 10,000 years including almost 8000 synthetically simulated earthquakes. Then, the ground motion prediction equations are used to calculate the ground motion per event and incorporate the effect of five site conditions, on amplifying or de-amplifying the ground motions on notional building exposure locations. Notional buildings are used to account for various building characteristics in conformance with the building taxonomy represented in HAZUS software. The HAZUS damage functions are applied to model the vulnerability of various structural types of buildings. Finally, the 50-year average mean loss and probabilistic loss for multiple values for probability of exceedance (2, 10, 20, and 40%) in 50 years are calculated, and the impact of different site condition databases on portfolio loss ratios is investigated for different structural types and heights of buildings. The results show the aggregated and geographical variation of loss and loss ratio throughout the region for various site conditions. Comparing the aggregated loss and loss ratio, while considering different databases, represents normalized differences that are limited to 6% for all building taxonomy with various heights and for all PoEs. However, site-specific loss ratio errors are significantly greater and in some cases are more than 20%.     

模糊神经网络在矿震预测中的应用

矿震同天然地震一样会给矿山生产及人身安全等带来重大灾难。也是目前尚不能准确预测的矿山灾害现象之一。根据现有的研究成果可知，矿震是一个多输入、多干扰、单输出的复杂系统。由于干扰项的存在，使利用建模、神经网络等手段对系统进行预测时会导致很大误差。模糊神经网络系统在建立对象输入、输出关系时与传统数学方法不同。即可以建立在无模型基础上，并利用其较强的学习训练特性，自动获取对象的输入、输出关系表达；可以将专家的评价语言作为系统的干扰项引入。这在一定程度上缓解了人为因素对预测结果的影响，且平滑了观测数据的随机性。文章利用改进的模糊神经网络及抚顺老虎台矿的矿震资料，对矿震最大震级的预测方法进行了探索。‘初步探讨了改进的模糊神经网络在矿震预测中的应用。得出在运用模糊神经网络进行预测时，为减小预测误差，应综合多种因素并提高专家评判语言的精确度的结论。指出在建立矿震系统预测模型时，利用干扰项将人为因素引入系统是必需的。通过实际应用证明其可行性。     

基于异质SVM神经网络的土壤盐渍化灾害预测模型

为研究银川平原普遍存在的土壤盐渍化问题,文章对银川平原的土壤盐渍化程度及潜在的发展趋势作出预测。利用Landsat 8 OLI数据与野外实测数据,选取地面高程、地下水位埋深、地下水溶解性总固体、植被指数、盐分指数及干旱指数为预测指标并提取指标值建立数据集,结合野外实测样点数据,建立基于异质支持向量机（Support Vector Machine,SVM)神经网络算法的盐渍化灾害预测模型。结果表明:（1）建立预测模型时,选择Radial Basis Funciton作为模型的核函数,c=100且g=3时预测精度最高可达85%;（2）研究区轻度盐渍化土壤面积约854 km2,中度盐渍化土壤面积约985 km2,重度盐渍化土壤面积约231 km2,主要分布在平罗县西大滩、银川芦花和吴忠苦水河地区;（3）银川平原北部的土壤盐渍化情况较严重且多分布于耕地周围的撂荒地以及地下水位埋藏较浅的地区,耕地资源中土壤盐渍化状况较严重,应注重耕地的合理灌溉与排水,增加土壤的可持续利用性。     

多属性神经网络反演预测煤层顶板岩性

煤层顶板岩性对于煤矿安全生产产生很大的影响。通过神经网络方法对自然伽马测井与地震属性（包括波阻抗属性）进行训练得到两者的非线性关系,并将其应用到整个地震数据中得到拟自然伽马体。相对于波阻抗属性,自然伽马参数可以很好地区分砂泥岩,从而更直接预测煤层顶板岩性,提高岩性预测的分辨率。     

Artificial neural network approach for estimation of surface specific humidity and air temperature using multifrequency scanning microwave radiometer

Microwave sensor MSMR (Multifrequency Scanning Microwave Radiometer) data onboard Oceansat-1 was used for retrieval of monthly averages of near surface specific humidity (Q _a) and air temperature (T _a) by means of Artificial Neural Network (ANN). The MSMR measures the microwave radiances in 8 channels at frequencies of 6.6, 10.7, 18 and 21 GHz for both vertical and horizontal polarizations. The artificial neural networks (ANN) technique is employed to find the transfer function relating the input MSMR observed brightness temperatures and output (Q _a andT _a) parameters. Input data consist of nearly 28 months (June 1999 – September 2001) of monthly averages of MSMR observed brightness temperature and surface marine observations ofQ _a andT _a from Comprehensive Ocean-Atmosphere Data Set (COADS). The performance of the algorithm is assessed with independent surface marine observations. The results indicate that the combination of MSMR observed brightness temperatures as input parameters provides reasonable estimates of monthly averaged surface parameters. The global root mean square (rms) differences are 1.0‡C and 1.1 g kg⁻¹ for air temperature and surface specific humidity respectively.