混沌优化神经网络方法及其在地震多属性研究中的应用

针对传统的基于梯度下降法BP神经网络中存在非线性多极值目标函数易陷于局部最优解的问题,提出了一种权值学习混沌优化的神经网络方法。非线性动力系统具有初值敏感性、遍历性的特性,采用基于混沌和梯度反传训练相结合来训练网络,可以使网络的连接权在不断迭代过程中自适应演化。实际过井地震剖面地震多属性研究实践表明,所提出的混沌优化学习方法可以克服传统方法的不足,提高预测能力。     

Bicharacteristic quality control in manufacturing

In this paper, the importance of bicharacteristic quality control has been discussed, when it is desired to exercise process control on manufacturing products that present two measurable characteristics which can suffer variations and on the other hand are considered significant for the functioning or the acceptance of the product. The verification that the process is in control or not is made by examining periodically the items of the samples drawn and plotting the points representing the mean of the characteristic in analysis on the corresponding charts. Using the computational program developed, a practical application is carried out for a mechanical manufacturing component that presents two critical characteristics. Also, a programmable mathematical model has been developed, aiming to computationally determine the sample size, and the interval between successive samples, which minimize the expected total cost. Idealized numerical application illustrates the use of the cost model that optimizes these parameters.     

裂缝发育带地震识别预测技术研究进展

裂缝性储层在全球有广泛的分布，裂缝油气藏勘探在油气勘探中占有重要位置。但是裂缝的成因复杂，其发育程度受多种因素控制。地震资料中虽包含有裂缝信息，但是因裂缝因素引起的地震响应要远远小于其他地质因素引起的响应，因此利用地震资料识别裂缝存在困难。虽然如此，由于地震资料在横向连续性上的优势，人们还是认为地震方法在裂缝描述和预测中具有不可替代的作用。本文简要分析了利用地震资料进行裂缝研究的难点和对策，并总结了全球利用地震资料进行裂缝研究的主要技术和方法，为我国进一步开展这项研究工作提供借鉴。     

房屋分层分户图信息管理软件的开发及相关技术

本文对开发房屋分层分户图管理系统的主要技术与方法如平台的选择、各种类型的建筑面积的计算及分摊、图形与属性数据的关联及在产权产籍其它管理软件中对此系统的嵌入等进行了介绍 ,并提出了相应的解决方法。实践表明 ,这些方法对房屋分层分户图管理系统走向实用不仅是可行的 ,而且是方便高效的。     

Chemical quality and sensory attributes of marinated Pacific saury (Cololabis saira) during vacuum-packaged storage at 4 °C

This study was carried out to evaluate the chemical changes and sensory attributes of Pacific saury (Cololabis saira), brined (12% NaCl brine solution) or marinated (12% NaCl + 2% acetic acid; or 12% NaCl + 3% acetic acid solutions) followed by vacuum-packaging and storage at 4 °C for 90 days. The chemical analysis revealed a significant reduction in the pH value, total volatile bases nitrogen (TVBN), and trimethylamine (TMA) contents in marinated versus brined fillets. Lipid oxidation, as indicated by the 2-thiobarbituric acid (TBA) values, was significantly delayed in marinated fillets in comparison with the brined fillets. The growth rate of psychrotrophic bacteria was significantly (P < 0.05) reduced in marinated versus brined fillets. No significant differences were detected for the sensory attributes between the two marinating conditions although the overall acceptability was significantly higher in marinated versus brined fish. Both conditions of the marinating process resulted in an extension of the shelf life of the product to more than 90 days versus only 60 days for the control brined fillets. The study concluded that marination of Pacific saury can delay the undesirable chemical changes, retard lipid oxidation, improve the sensory attributes and extend the shelf life of the product during refrigerated storage.     

Design through selection: a method that works

In this paper a practical approach to design, based on the concept of selection, is presented. The approach involves: first, the generation of alternative concepts using ideation tecniques; second, the selection of the ‘most-likely-to-succeed’ concepts for further development into feasible alternatives; third, the formulation and solution of selection-decision-support problems to rank the feasible alternatives in order of preference using multiple attributes.The method presented in this paper is a combination of the methods proposed by Pugh and by Mistree and Muster. The former method is appropriate for use in concept selection, which is characterized by many alternatives and essentially insight-based ‘soft’ information. The latter method is appropriate when there are few alternatives and a mix of science-based ‘hard’ and insight-based ‘soft’ information. The method presented by Mistree and Muster is therefore used to formulate and solve the selection-decision-support problem. The design example used in this paper is a modified version of that used by Pugh.     

Development of the Hybrid Weight Assessment System for Multiple Quality Attributes

When designing a product or process, a design engineer is often faced with a number of quality attributes and hence prioritizing the quality attributes is an inevitable task. Development of an adequate weight assessment system for multiple attributes in the early stage of design is crucial, since the choice of paths to quality enhancement depends heavily upon the type of weight assessment system used. Although several weight assessment methods are available in the literature, there is room for improvement. Most weight assessment methods reported in the research community have been developed from the perspective of either customers or a design engineer. This article proposes a hybrid weight assessment system by combining the perspectives of both parties, the customer and a design engineer, in such a way that uncertainty of human perceptions can be modeled using the concepts of fuzzy sets and Monte Carlo simulation. A model is presented and a numerical example is discussed.     

试析科学传播中内在的伦理属性

科学传播本质上作为人与人之间的交往活动,具有伦理属性。一是传播主体与受众具有道德判断,二是科学传播模型的改变受社会道德的影响,三是科学知识本身的道德属性。但与科学研究不同的是,科学传播除了具有的科学性、真实性和规范性之外,还应具有传播方式公平性、传播动机正义性及传播态度批判性。     

基于SVR和地震属性的构造煤厚度定量预测

为了提高谱分解和甜面属性组合预测构造煤厚度的精度和可靠性,利用回归型支持向量机(SVR)的非线性处理能力,将SVR和地震属性相结合,研究采区构造煤厚度定量预测方法。在SVR预测模型建立时,以正演模型数据为基础,通过训练和测试获得SVR预测模型的主要参数;结合井旁道数据,建立了采区构造煤厚度SVR预测模型。通过输入实际谱分解属性和甜面属性,定量预测了构造煤厚度。相对于传统地震属性预测来说,本次所预测的构造煤厚度精度较高、误差较小。当核函数类型为径向基核函数、输入为谱分解属性和甜面属性时,预测模型的预测效果最好。由于模型建立时未考虑地震资料信噪比的影响,预测模型不能克服其造成的不确定性。     

Google Earth公路三维信息管理系统的构建

公路三维信息管理系统可以提高公路建设的管理水平,基于Google Earth二次开发构建的公路三维信息管理系统,可实现构造物空间属性的查询,属性信息的实时更新,为公路网规划、公路项目建设及公路运营管理提供方便。