Amelioration of Carbon Removal Prediction for an Activated Sludge Process using an Artificial Neural Network (ANN)

A dynamic simulation model of the Ankara central wastewater treatment plant (ACWTP) was evaluated for the prediction of effluent COD concentrations. Firstly, a mechanistic model of the municipal wastewater treatment process was developed based on Activated Sludge Model No. 1 (ASM1) by using a GPS‐X computer program. Then, the mechanistic model was combined with a feed‐forward back‐propagation neural network in parallel configuration. The appropriate architecture of the neural network models was determined through several iterative steps of training and testing of the models. Both models were run with the data obtained from the plant operation and laboratory analysis to predict the dynamic behavior of the process. Using these two models, effluent COD concentrations were predicted and the results were compared for the purpose of evaluation of treatment performance. It was observed that the ASM1 ANN model approach gave better results and better described the operational conditions of the plant than ASM1.     

人工神经网络法在烃源岩测井评价中的应用

运用有机地球化学方法分析岩芯、岩屑样品的有机炭含量存在着昂贵、时且不准确等问题。利用测井方法的优点是经济、准确。在测井评价中使用人工神经网络法具有极大的优越性和适用性。本文结合Kohonen和BP网络方法，完成了塔里木台盆区19口井的寒武、奥陶系烃源岩层段的识别与评价，并通过测井资料处理成果和岩芯有机地化资料、地质录井情况的相互检验，证实，其本上能够满足评价烃源岩的需要，从而为利用测井资料进行烃源岩评价做出了新的尝试。     

Estimating Curie Point Depth and Heat Flow Map for Northern Red Sea Rift of Egypt and Its Surroundings, from Aeromagnetic Data

In this study, we aim to map the Curie point depth surface for the northern Red Sea rift region and its surroundings based on the spectral analysis of aeromagnetic data. Spectral analysis technique was used to estimate the boundaries (top and bottom) of the magnetized crust. The Curie point depth (CPD) estimates of the Red Sea rift from 112 overlapping blocks vary from 5 to 20 km. The depths obtained for the bottom of the magnetized crust are assumed to correspond to Curie point depths where the magnetic layer loses its magnetization. Intermediate to deep Curie point depth anomalies (10–16 km) were observed in southern and central Sinai and the Gulf of Suez (intermediate heat flow) due to the uplifted basement rocks. The shallowest CPD of 5 km (associated with very high heat flow, ~235 mW m^?2) is located at/around the axial trough of the Red Sea rift region especially at Brothers Island and Conrad Deep due to its association with both the concentration of rifting to the axial depression and the magmatic activity, whereas, beneath the Gulf of Aqaba, three Curie point depth anomalies belonging to three major basins vary from 10 km in the north to about 14 km in the south (with a mean heat flow of about 85 mW m^?2). Moreover, low CPD anomalies (high heat flow) were also observed beneath some localities in the northern part of the Gulf of Suez at Hammam Fraun, at Esna city along River Nile, at west Ras Gharib in the eastern desert and at Safaga along the western shore line of the Red Sea rift. These resulted from deviatoric tensional stresses developing in the lithosphere which contribute to its further extension and may be due to the opening of the Gulf of Suez and/or the Red Sea rift. Furthermore, low CPD (with high heat flow anomaly) was observed in the eastern border of the study area, beneath northern Arabia, due to the quasi-vertical low-velocity anomaly which extends into the lower mantle and may be related to volcanism in northern Arabia. Dense microearthquakes seem to occur in areas where the lateral gradients of the CPD are steep (e.g. entrance of the Gulf of Suez and Brothers Island in the Red Sea). These areas may correspond to the boundaries between high and low thermal regions of the crust. Thus, the variations in the microseismic activity may be closely related to thermal structures of the crust. Indeed, shallow cutoff depths of seismicity can also be found in some geothermal areas (e.g. western area of Safaga city along the Red Sea coastal region and at Esna city along the River Nile). These facts indicate that the changes in the thickness of the seismogenic layer strongly depend on temperature. Generally, the shallow Curie point depth indicates that some regions in our study area are promising regions for further geothermal exploration particularly in some localities along the River Nile, Red Sea and Gulf of Suez coastal regions.     

Deep Artificial Neural Networks as a Tool for the Analysis of Seismic Data

The number of seismological studies based on artificial neural networks has been increasing. However, neural networks with one hidden layer have almost reached the limit of their capabilities. In the last few years, there has been a new boom in neuroinformatics associated with the development of third-generation networks, deep neural networks. These networks operate with data at a higher level. Unlabeled data can be used to pretrain the network, i.e., there is no need for an expert to determine in advance the phenomenon to which these data correspond. Final training requires a small amount of labeled data. Deep networks have a higher level of abstraction and produce fewer errors. The same network can be used to solve several tasks at the same time, or it is easy to retrain it from one task to another. The paper discusses the possibility of applying deep networks in seismology. We have described what deep networks are, their advantages, how they are trained, how to adapt them to the features of seismic data, and what prospects are opening up in connection with their use.     

Least-squares optimum-order regional geothermal gradient determination from deep borehole temperature data with application to the Western Desert of Egypt

Summary This paper discusses an approach to determining the least-squares optimum order of a polynomial fit to temperature-gradient data. The optimum is reached when two successive least-squares regional maps are sufficiently similar. This similarity is measured by the correlation factor between the polynomials.The approach was applied to the available deep bore-hole temperature data in the central part of the Northern Western Desert of Egypt. The fitting polynomial was found to be of the second degree. Assuming the thermal conductivity to be uniformly 2·2 Wm^–1K^–1 for the stratigraphic section in the Western Desert, the heat-flow density ranges between 35 and 70 mW/m².Presented at the International Meeting on Terrestrial Heat Flow and the Structure of Lithosphere, Bechyn Castle, Czech Republic, September 2 – 7, 1991.     

Evidence for a ∼ 200–100 ka meteorite impact in the Western Desert of Egypt

In this study, we present evidence, in the form of unusual silicate glasses, for a meteorite impact event ∼ 200–100 ka in the Dakhleh Oasis, Western Desert, Egypt. These glasses, known locally as Dakhleh Glass, were derived from the shock melting of a series of unconsolidated sediments underlain by interbedded carbonates, sandstones and phosphate-rich lithologies. Hypervelocity impact in to a volatile-rich target resulted in the production of impact glasses with CaO and Al₂O₃ contents of up to ∼ 25 and 18 wt.%, respectively. Other notable properties include the presence of globules of immiscible calcite and pyrrhotite melt phases, shattered quartz grains, and fragments of silicified plant matter. Dating of geoarchaeological artefacts associated with the Dakhleh Glass support preliminary ⁴⁰Ar/³⁹Ar data, indicating a ∼ 200–100 ka age for the impact event. Geoarchaeological evidence indicates that archaic Homo sapiens and early modern humans continually inhabited this region of the Western Desert during this period. While it is unclear at present whether the Dakhleh Glass was formed during a cratering event or a large aerial burst, the effect on the environment and inhabitants of Dakhleh would have been catastrophic.     

基于专家知识的神经元网络方法及其在潜在震源区定量划分中的应用

将专家知识和神经元网络方法结合起来，利用专家知识和规则构造规则样本的方法来提高神经元网络学习样本的质量，通过神经元网络对规则样本的学习，形成基于专家知识的神经元网络模型。在潜在震源区划分专家系统中有关规则的基础上，将基于专家知识的神经元网络方法引入到潜在震源区定量划分中，并以首都圈地区为例，对其潜在震源区作了定量划分，结果表明，规则样本能较好的反映专家的知识和规则，利用构造规则样本的方法，可以提高     

Challenges in constraining pluvial events and hominin activity: Examples of ESR dating molluscs from the Western Desert,Egypt

Receiving <0.1 mm/y of precipitation, Egypt’s hyperarid Western Desert, today lacks naturally occurring surface water. Artesian spring deposits, tufa deposited by springs and carbonate-rich silty lacustrine sediment attest that oases in the Western Desert had surface water during the Pleistocene. Paleolithic artefacts, fossil ungulate teeth, and snails occurring within the Pleistocene deposits and dotting the surface record times when higher rainfall and/or groundwater tables during pluvial events allowed surface water to exist in wetlands, small ponds and lakes, enabling hominin habitation. Archaeological finds ranging from Early to Later Stone Age (ESA–LSA) occur in gravel lags, within sedimentary deposits, and on the older geomorphic surfaces. Near Kharga, large tufa deposits ranging from a few hectares to more than 10 km² in area, such as Matana and Medauwara, dot the edge of the Libyan Plateau. Molluscs were dated using standard ESR protocols. To test for reworked fossils, multiple samples from a single sample were dated independently. In some units at Medauwara, multiple gastropod populations from different times have been preserved, while others appear to only preserve a single population. To see the effects of the cosmic dose rate on ESR ages, ages were calculated using zero cosmic dose rate, the full modern cosmic dose rate, and time-averaged cosmic and sedimentary dose rates. For gastropods from Matana, no significant difference in ESR ages resulted from different cosmic dose rate assumptions. Therefore, at Matana 2, the shells dated at 27.7 ± 1.9 assuming time-averaged external dose rates, while at Matana 3, they averaged 65.1 ± 4.1 ka, suggesting that water was present for hominin use at times during OIS 2 and 4.     

Modeling and Prediction of Monthly Total Ozone Concentrations by Use of an Artificial Neural Network Based on Principal Component Analysis

In the work discussed in this paper we considered total ozone time series over Kolkata (22°34′10.92″N, 88°22′10.92″E), an urban area in eastern India. Using cloud cover, average temperature, and rainfall as the predictors, we developed an artificial neural network, in the form of a multilayer perceptron with sigmoid non-linearity, for prediction of monthly total ozone concentrations from values of the predictors in previous months. We also estimated total ozone from values of the predictors in the same month. Before development of the neural network model we removed multicollinearity by means of principal component analysis. On the basis of the variables extracted by principal component analysis, we developed three artificial neural network models. By rigorous statistical assessment it was found that cloud cover and rainfall can act as good predictors for monthly total ozone when they are considered as the set of input variables for the neural network model constructed in the form of a multilayer perceptron. In general, the artificial neural network has good potential for predicting and estimating monthly total ozone on the basis of the meteorological predictors. It was further observed that during pre-monsoon and winter seasons, the proposed models perform better than during and after the monsoon.     

南海北部第四系厚度分布图的编制与简要说明

收集南海北部大量多道地震和钻井资料,编制了该区域首幅第四系厚度分布图。第四系等厚线总体呈NEE向,特别是珠江口外大陆架最明显,莺歌海、琼东南呈NW-EW-NE向弧形,珠江口大陆架南部存在次一级NW向等厚线。第四系等厚线分布形态明显受到断裂构造控制。     

Seismic Source Classification by Means of a Sonogram- Correlation Approach: Application to Data of the RSM Seismic Network (Central Italy)

—?In this paper, we apply a new sonogram correlation approach for seismic source classification. In order to test the technique, we use the waveforms recorded by 6 among 16 of the stations of the Rete Sismometrica Marchigiana (RSM) network, relevant to the vertical component of 365 earthquakes that occurred in the analyzed area from April 1996 until September 1999. All the sonograms of each station are cross-correlated in a single station procedure and the calculated degrees of resemblance are used to classify the events by means of a bridging technique. The results obtained with the single-station approach are integrated into a network classification. We identified three families in the investigated area. Once the efficiency of the method is tested, we analyzed the families obtained for the selected subset of events. Considering the temporal and the spatial distribution of the events forming each family, we classified them as quarry blasts. This classification, even if relevant to a subset of earthquakes occurring in Central Italy, modifies the present knowledge of the seismicity distribution in the area and allows estimation of the errors affecting the hypocenter locations provided by the RSM network. An extension of this new technique to the whole Central Italy earthquake catalog promises an interesting review of the seismicity in the area.     

The Quaternary development of tributary channels to the Nile River at Kom Ombo area,Eastern Desert of Egypt,and their implication for groundwater resources

The dry wadis ‘ephemeral channel’ constituting the main tributaries to the Nile River in Kom Ombo are structurally and tectonically controlled and exhibit complex drainage pattern. This complicated drainage pattern is inherited from the morpho‐tectonic evolution of the ancestral Nile River (‘Protonile’), which drained the Eastern Desert during the Middle Pleistocene. A digital elevation model derived from Shuttle Radar Topography Mission data is used to delineate the contemporary drainage networks and their catchments. Satellite images acquired during a flash flood event were used to validate the delineated watershed divides and flow pathways, particularly where the courses of dry wadis are interlocked. Currently, the westward flow of Wadi Abu‐Suberah is derived from a small area in the Eastern Desert, as the palaeo‐upper reaches of this wadi were captured due to tectonic movements along NW/SE and N/S faults by wadis in the Kharit and Elewa areas. The influence of these tectonic movements on groundwater distribution is also shown where the deep Nubian aquifer discharges its water into the Quaternary aquifer through fault planes. The northward flowing ‘Protonile’ main course has tectonically shifted from the Gallaba plain in the Western Desert, eastward to the current Nile River course. This shift has produced several cut‐off segments of the palaeo‐tributary drainage that was originally flowing westward towards the old ‘Protonile’ main course in the Gallaba plain. However, these segments have reversed their surface run‐off flow directions eastward towards the current Nile course; they could include potential groundwater resources, as their alluvium may be still recharged by the Nile River. Copyright © 2010 John Wiley & Sons, Ltd.     

Active Source Seismic Identification and Automatic Picking of the P-wave First Arrival Using a Convolutional Neural Network

In seismic data processing, picking of the P-wave first arrivals takes up plenty of time and labor, and its accuracy plays a key role in imaging seismic structures. Based on the convolution neural network (CNN), we propose a new method to pick up the P-wave first arrivals automatically. Emitted from MINI28 vibroseis in the Jingdezhen seismic experiment, the vertical component of seismic waveforms recorded by EPS 32-bit portable seismometers are used for manually picking up the first arrivals (a total of 7242). Based on these arrivals, we establish the training and testing sets, including 25,290 event samples and 710,616 noise samples (length of each sample:2s). After 3,000 steps of training, we obtain a convergent CNN model, which can automatically classify seismic events and noise samples with high accuracy (> 99%). With the trained CNN model, we scan continuous seismic records and take the maximum output (probability of a seismic event) as the P-wave first arrival time. Compared with STA/LTA (short time average/long time average), our method shows higher precision and stronger anti-noise ability, especially with the low SNR seismic data. This CNN method is of great significance for promoting the intellectualization of seismic data processing, improving the resolution of seismic imaging, and promoting the joint inversion of active and passive sources.     

稀疏台网的中小地震震源机制解反演研究——以2017年1月4日西藏仲巴4.7级地震为例

本文应用ISOLA近震全波形方法,以2017年1月4日西藏仲巴4.7级地震为例,反演稀疏台网记录的中小地震震源机制解。该地震反演所得最佳双偶机制参数为：节面Ⅰ的走向109°/倾角85°/滑动角-177°,节面Ⅱ的走向19°/倾角88°/滑动角-4°,最佳矩心位置为30.590°N、83.784°E,最佳质心深度为6km,矩震级M_W4.6。震源机制反演结果表明此次地震是一次走滑型为主的事件,其与震源区域附近历史地震震源机制解具有相同性质。本文还应用CSPS初动扫描法,利用P波初动资料和近震波形联合约束反演此次地震的震源机制,并与ISOLA近震全波形反演结果进行比较,结果表明,联合少量台站的的三分量波形数据,能够定量地判断最佳震源机制解,降低了P波初动反演结果的非唯一性,同时也约束了由于少量台站参与全波形反演引起的解的不稳定性。本文研究为中小地震震源参数测定提供了一种简单有效的方法,具有较高的稳定性和可靠性。     

BP神经网络算法考虑激活函数后对强震路基塌陷变形预测研究

当前强震后铁路路基变形预测中,相关算法未能考虑激活函数的非线性属性,造成非线性变形特征数据提取不完整,且其特征数据存在偏差,陷入局部最优解。文章提出BP神经网络算法考虑激活函数后对强震路基塌陷变形预测方法,采用双极S因子补偿ReLU非线性激活函数,优化BP神经网络算法,解决非线性路基变形特征数提取问题。利用数据标准化归一方法,对已修正提取的全部变形特征数据进行偏差数据归一,得到路基变形特征数据集合,结合强震后路基变形连接权值计算路基变形预测值,完成强震路基塌陷变形预测。结合实测结果,在matlab下进行预测实验,结果表明所提混合方法可以有效地对水平地震作用下铁路路基塌陷变形程度进行预测,且预测值在误差允许范围内,为铁路的安全运行以及实时维护提供重要依据。