煤化工用水及废水处理分析

概述了煤化工高能耗、高污染、高耗水的产业特点,从工艺、意识、来源分析了造成产业特点的原因。为了减少煤化工废水中有毒害物质的含量,介绍了常用的废水处理方法,展望了煤化工用水及废水处理发展趋势。     

新智能化算法及其在中医诊断数字化中的应用

对中医临床诊断数据的特性进行了研究,提出了病证及其特征数据(症状或体征数据)的一些特性指标:病证的相似度、复杂度、隐蔽度,特征数据的贡献度、常见度、显隐性.另外,研究了基于多维关联规则提取诊断经验的方法.在此基础上,研究了一种非充分条件下复杂数据智能化处理拓展算法,该算法的实现模型嵌入了模糊竞争神经网络.该算法在复杂的中医诊断数字化中得到了应用,结果表明,该算法可以较好地处理复杂数据.     

移动位置服务中一种空间智能发布系统模型

与现有的位置服务模式相比,基于Push(推)模式的位置服务有着更加广阔的市场前景,是目前国内外研究的焦点.该模式的位置服务依赖于现存的空间发布系统,而目前空间发布系统由于条件的限制不能有效地完成空间事件的预订和发布.阐述了目前空间发布系统中两种常见的空间事件以及空间事件的预订,随后提出一种空间智能发布系统模型,有效地解决了无线通信的带宽限制问题并提高了空间发布系统的性能.     

浅析降低农村电网线损的技术措施

阐述了农村电网线损按性质的分类,从电力网络的功率分布、经济运行、电网改造等方面论述了降低线损的技术措施,并介绍了在线损管理中采取的一些较有成效的技术措施。     

基于神经网络的白酒勾兑目标规划算法优化

针对白酒勾兑过程中,现有的白酒勾兑目标规划算法难以确定权系数(优先因子)的缺点,本文提出了利用人工神经网络对目标规划算法进行改进和优化,选择三层前向BP神经网络结构,并通过选取理化指标向量与\     

钢/铝异种金属点焊研究进展综述

基于电阻点焊过程中工艺信号特征,在不同间距、不同间隙和不同间距与间隙3种条件下,引入相关性分析方法分析工艺信号与熔核直径之间的相关性,并建立基于麻雀搜索算法-BP神经网络(sparrow search algorithm- back propagation neural network, SSA-BP)的电阻点焊质量预测模型,将功率、焊接电流、焊接电压和动态电阻作为预测模型输入特征. 结果表明,经麻雀搜索算法优化后的BP神经网络在测试集上的决定系数R²、均方误差(mean-square error, MSE)、均方根误差(root mean square error, RMSE)和平均绝对误差(mean absolute error, MAE)分别为0.95,1.55,1.24和0.90,均优于BP模型. 获得了功率、焊接电流、焊接电压和动态电阻与熔核直径的映射关系,可为焊接的工艺参数设计提供依据.

     

基于混合数据增强的MSWI过程燃烧状态识别

国内城市固废焚烧(Municipal solid waste incineration, MSWI)过程通常依靠运行专家观察炉内火焰识别燃烧状态后再结合自身经验修正控制策略以维持稳定燃烧, 存在智能化水平低、识别结果具有主观性与随意性等问题. 由于MSWI过程的火焰图像具有强污染、多噪声等特性, 并且存在异常工况数据较为稀缺等问题, 导致传统目标识别方法难以适用. 对此, 提出一种基于混合数据增强的MSWI过程燃烧状态识别方法. 首先, 结合领域专家经验与焚烧炉排结构对燃烧状态进行标定; 接着, 设计由粗调和精调两级组成的深度卷积生成对抗网络(Deep convolutional generative adversarial network, DCGAN)以获取多工况火焰图像; 然后, 采用弗雷歇距离(Fréchet inception distance, FID)对生成式样本进行自适应选择; 最后, 通过非生成式数据增强对样本进行再次扩充, 获得混合增强数据构建卷积神经网络以识别燃烧状态. 基于某MSWI电厂实际运行数据实验, 表明该方法有效地提高了识别网络的泛化性与鲁棒性, 具有良好的识别精度.     

区域频谱资源数据融合方法

为实现高效的数据融合,提出了一种多源频谱资源数据融合方法.该方法将多个监测节点的频谱可用资源评估信息进行融合,利用意见形成理论,对关心地域的频谱资源可用性进行评估,为频谱动态管理提供资源状态信息.通过对不同网络规模的仿真分析,结果表明：在意见形成理论的基础上,通过设定“骨干”节点,缩减了50%的融合时间;在网络节点不超过100个的情况下,通过30轮次的沟通即可完成频谱资源数据的融合.     

一种基于社交网络社区的组推荐框架

面向用户群组的推荐主要面临如何有意义地对群组进行定义并识别,以及向群组内用户进行有效推荐两大问题。该文针对已有研究在用户群组定义解释性不强等存在的问题,提出一种基于社交网络社区的组推荐框架。该框架利用社交网络结构信息发现重叠网络社区结构作为用户群组,具有较强的可解释性,并根据用户与群组间的隶属度制定了考虑用户对群组贡献与用户从群组获利的4种聚合与分配策略,以完成组推荐任务。通过在公开数据集上与已有方法的对比实验,验证了该框架在组推荐方面的有效性和准确性。     

Elevated Expression of CCN3 in Articular Cartilage Induces Osteoarthritis in Hip Joints Irrespective of Age and Weight Bearing

Osteoarthritis (OA) occurs not only in the knee but also in peripheral joints throughout the whole body. Previously, we have shown that the expression of cellular communication network factor 3 (CCN3), a matricellular protein, increases with age in knee articular cartilage, and the misexpression of CCN3 in cartilage induces senescence-associated secretory phenotype (SASP) factors, indicating that CCN3 promotes cartilage senescence. Here, we investigated the correlation between CCN3 expression and OA degenerative changes, principally in human femoral head cartilage. Human femoral heads obtained from patients who received total hip arthroplasty were categorized into OA and femoral neck fracture (normal) groups without significant age differences. Gene expression analysis of RNA obtained from femoral head cartilage revealed that CCN3 and MMP-13 expression in the non-weight-bearing part was significantly higher in the OA group than in the normal group, whereas the weight-bearing OA parts and normal cartilage showed no significant differences in the expression of these genes. The expression of COL10A1, however, was significantly higher in weight-bearing OA parts compared with normal weight-bearing parts, and was also higher in weight-bearing parts compared with non-weight-bearing parts in the OA group. In contrast, OA primary chondrocytes from weight-bearing parts showed higher expression of CCN3, p16, ADAMTS4, and IL-1β than chondrocytes from the corresponding normal group, and higher ADAMTS4 and IL-1β in the non-weight-bearing part compared with the corresponding normal group. Acan expression was significantly lower in the non-weight-bearing group in OA primary chondrocytes than in the corresponding normal chondrocytes. The expression level of CCN3 did not show significant differences between the weight-bearing part and non-weight-bearing part in both OA and normal primary chondrocytes. Immunohistochemical analysis showed accumulated CCN3 and aggrecan neoepitope staining in both the weight-bearing part and non-weight-bearing part in the OA group compared with the normal group. The CCN3 expression level in cartilage had a positive correlation with the Mankin score. X-ray analysis of cartilage-specific CCN3 overexpression mice (Tg) revealed deformation of the femoral and humeral head in the early stage, and immunohistochemical analysis showed accumulated aggrecan neoepitope staining as well as CCN3 staining and the roughening of the joint surface in Tg femoral and humeral heads. Primary chondrocytes from the Tg femoral head showed enhanced expression of Ccn3, Adamts5, p16, Il-6, and Tnfα, and decreased expression of Col2a1 and -an. These findings indicate a correlation between OA degenerative changes and the expression of CCN3, irrespective of age and mechanical loading. Furthermore, the Mankin score indicates that the expression level of Ccn3 correlates with the progression of OA.