二氧化碳爆破储层改造近炮孔处岩石性质变化及温度分布规律研究

传统的地热井井身结构不适用于利用二氧化碳爆破技术进行干热岩储层建造,因此本文基于二氧化碳爆破致裂技术建造干热岩储层的新型井身结构工艺进行了进一步探究。结合该工艺水平井段（炮孔）钻进过程实际工况,对钻井后水平井段（炮孔）附近干热岩储层的物理力学性质变化规律及储层温度分布特征进行了岩石力学实验和数值模拟研究。研究结果表明钻井液降温产生的热应力作用是储层岩石性质发生变化、出现热损伤区的主要原因;冷却过程中炮孔附近储层呈现出“快速降温区-平缓降温区-未降温区”的区域性温度分布特征;储层初始温度和冷却时长对快速降温区温度分布以及热损伤区范围影响较大。本研究可以为二氧化碳爆破致裂干热岩储层建造提供理论指导。     

Sequential Gaussian simulation for geosystems modeling: A machine learning approach

Sequential Gaussian Simulation(SGSIM)as a stochastic method has been developed to avoid the smoothing effect produced in deterministic methods by generating various stochastic realizations.One of the main issues of this technique is,however,an intensive computation related to the inverse operation in solving the Kriging system,which significantly limits its application when several realizations need to be produced for uncertainty quantification.In this paper,a physics-informed machine learning(PIML)model is proposed to improve the computational efficiency of the SGSIM.To this end,only a small amount of data produced by SGSIM are used as the training dataset based on which the model can discover the spatial correlations between available data and unsampled points.To achieve this,the governing equations of the SGSIM algorithm are incorporated into our proposed network.The quality of realizations produced by the PIML model is compared for both 2D and 3D cases,visually and quantitatively.Furthermore,computational performance is evaluated on different grid sizes.Our results demonstrate that the proposed PIML model can reduce the computational time of SGSIM by several orders of magnitude while similar results can be produced in a matter of seconds.     

河北某铁矿掘进爆破参数的优化与实践

针对河北某铁矿投产初期巷道掘进爆破效果差、效率低的问题,重点从炮孔间距和布置上对爆破参数进行了优化。生产实践证明,优化后的掘进爆破参数达到了提高巷道掘进效率、降低爆破成本的目的。优化后的爆破参数已在该矿实际生产中推广应用,为今后掘进爆破参数进一步优化积累了经验。     

Prediction of blast fragmentation using multivariate analysis procedures

An extensive multivariate analysis procedure for prediction of blast fragmentation distribution is presented. Several blasts performed in various mines and rock formations in the world are brought together and evaluated. Blast design parameters, the modulus of elasticity, in situ block size are considered to perform multivariate analysis. The hierarchical cluster analysis is used to separate the blasts data into different groups of similarity. Group memberships were checked by the discriminant analysis. The multivariate regression analysis was applied to develop prediction equations for the estimation of the mean particle size of muckpiles. Two different prediction equations were developed based on the rock stiffness. Validation of the proposed equations on various mines is presented and the capability of the prediction equations was compared with one of the most applied fragmentation distribution models appearing in the blasting literature. Prediction capability of the proposed models was found to be strong. Diversity of the blasts data used is one of the most important aspects of the developed models. The models are not complex and suitable for practical use at mines. Copyright © 2010 John Wiley & Sons, Ltd.     

Artificial Intelligence and headland-bay beaches

Headland-bay beaches are a typical feature of many of the world's coastlines. Their curved planform has aroused much interest since the early days of Coastal Engineering. Modelling this characteristic planform is a task of great interest, not least in relation to projects of coastal structures whose effects on the shoreline must be studied from the planning stages. In this work, Artificial Intelligence is applied to this task—in particular, artificial neural networks (ANNs). Unlike conventional planform models, they are not based on a given mathematical expression of the shoreline curve. Instead, they learn from experience (from a number of training cases) how the planform of a headland-bay beach is shaped, with due regard to the obliquity of incident waves. Three artificial neural networks, with different input/output structures, are implemented and subsequently trained with a number of bays. Once trained, they are tested for validation on other headland-bay beaches. Finally, the most performing neural network is compared with a state-of-the-art planform model.     

Classifiers vs. input variables—The drivers in image classification for land cover mapping

The study investigates the performance of image classifiers for landscape-scale land cover mapping and the relevance of ancillary data for the classification success in order to assess and to quantify the importance of these components in image classification. Specifically tested are the performance of maximum likelihood classification (MLC), artificial neural networks (ANN) and discriminant analysis (DA) based on Landsat7 ETM+ spectral data in combination with topographic measures and NDVI. ANN produced high accuracies of more than 75% also with limited input information, while MLC and DA produced comparable results only by incorporating ancillary data into the classification process. The superiority of ANN classification was less pronounced on the level of the single land cover classes. The use of ancillary data generally increased classification accuracy and showed a similar potential for increasing classification accuracy than the selection of the classifier. Therefore, a stronger focus on the development of appropriate and optimised sets of input variables is suggested. Also the definition and selection of land cover classes has shown to be crucial and not to be simply adaptable from existing land cover class schemes. A stronger research focus towards discriminating land cover classes by their typical spectral, topographic or seasonal properties is therefore suggested to advance image classification.     

尖紫蛤的人工育苗

1997年 7月开始进行尖紫蛤人工育苗的研究 ,1999年 11月培育出壳长 4 mm左右的幼苗近30万只 ,首次获得尖紫蛤人工育苗的成功     

Wave hindcasting by coupling numerical model and artificial neural networks

By coupling numerical wave model (NWM) and artificial neural networks (ANNs), a new procedure for wave prediction is proposed. In many situations, numerical wave modeling is not justified due to economical consideration. Although incorporation of an ANN model is inexpensive, such a model needs a long time period of wave data for training, which is generally inconvenient to achieve. A proper combination of these two methods could carry the potentials of both. Based on the proposed approach, wave data are generated by a NWM by means of a short period of assumed winds at a concerned point. Then, an ANN is designed and trained using the above-mentioned generated wind-wave data. This ANN model is capable of mapping wind-velocity time series to wave height and period time series with low cost and acceptable accuracy. The method was applied for wave hindcasting to two different sites; Lake Superior and the Pacific Ocean. Simulation results show the superiority of the proposed approach.     

Radiation of lamp and optimized experiment using artificial light in the Arctic Ocean

A winter optical experiment by an artificial lamp was conducted in the Amundsen Bay of Arctic Ocean from November of 2007 to January of 2008.The radiation field emitted from an artificial lamp was measured and is introduced in this paper ,and the optimized experiment project is discussed.It is demonstrated that the minimum size allowed of the lamp is determined by both the field of view(FOV) of optical instrument and the measuring distance from the lamp.Some problems that might influence on the experimen...     

Effects of prediction accuracy of the proportion of vegetation cover on land surface emissivity and temperature using the NDVI threshold method

Predicting land surface energy budgets requires precise information of land surface emissivity (LSE) and land surface temperature (LST). LST is one of the essential climate variables as well as an important parameter in the physics of land surface processes at local and global scales, while LSE is an indicator of the material composition. Despite the fact that there are numerous publications on methods and algorithms for computing LST and LSE using remotely sensed data, accurate prediction of these variables is still a challenging task. Among the existing approaches for calculating LSE and LST, particular attention has been paid to the normalised difference vegetation index threshold method (NDVI^THM), especially for agriculture and forest ecosystems. To apply NDVI^THM, knowledge of the proportion of vegetation cover (P_V) is essential. The objective of this study is to investigate the effect of the prediction accuracy of the P_V on the estimation of LSE and LST when using NDVI^THM. In August 2015, a field campaign was carried out in mixed temperate forest of the Bavarian Forest National Park, in southeastern Germany, coinciding with a Landsat-8 overpass. The P_V was measured in the field for 37 plots. Four different vegetation indices, as well as artificial neural network approaches, were used to estimate P_V and to compute LSE and LST. The results showed that the prediction accuracy of P_V improved using an artificial neural network (R²_CV = 0.64, RMSE_CV = 0.05) over classic vegetation indices (R²_CV = 0.42, RMSE_CV = 0.06). The results of this study also revealed that variation in the accuracy of the estimated P_V affected calculation results of the LSE. In addition, our findings revealed that, though LST depends on LSE, other parameters should also be taken into account when predicting LST, as more accurate LSE results did not increase the prediction accuracy of LST.