基于改进直接灰色模型的泵车摆缸泄漏预测

提出一种基于参数和时间响应序列的改进直接灰色模型IDGM(1,1),对泵车摆缸泄漏的趋势进行预测分析,并与传统灰色模型GM(1,1)和直接灰色模型DGM(1,1)的预测结果进行比较。结果表明,该方法可以准确地预测摆缸泄漏故障的劣化趋势,其预测精度明显优于传统灰色模型GM(1,1)和直接灰色模型DGM(1,1)的预测精度,可为摆缸泄漏故障的主动维护提供重要理论依据。     

考虑内摩擦角影响的开口管桩极限承载力理论分析

针对桩内土塞的存在导致开口管桩极限承载力力计算的困难,在别列赞策夫理论的基础上,将开口管桩的桩壁和桩内土塞分别按刚、柔基础对待,考虑持力层土内摩擦角对桩壁和土塞的影响,推导出了持力层为砂土的开口管桩单桩极限承载力公式。选取12组算例对该承载力公式进行计算,结果表明：当持力层土的内摩擦角越大、桩长越长、桩内径越小时,土塞端阻应力和桩壁端阻应力的差值就越大;当持力层土内摩擦角大于30°时,用该表达式计算出的开口管桩极限承载力值比用 JGJ94—2008《建筑桩基技术规范》方法计算值大很多,说明用该规范计算的承载力还有很大的安全储备。     

西北村镇建筑冬季室内期望温度研究

对乌鲁木齐、西宁、西安、天水、银川5个典型西北地区村镇住宅建筑各主要功能房间的使用情况、采暖情况、温度、居民热需求、衣着和活动强度等进行测试与问卷调查,以得到主要功能房间间歇采暖时间段、采暖与否两种情况下居民的期望温度、室内活动强度与服装热阻对居民期望温度的影响。结果表明：采暖和非采暖情况下,卧室期望温度分别约为15～19℃和12～14℃;采暖和非采暖情况下,客厅期望温度分别约为16～19℃和11～13℃。冬季间歇采暖时间段可取为：主卧为19：00—次日8：30,次卧为20：30—次日8：00,客厅为8：30—19：00,但考虑到时差原因,采暖时段可作相应调整。     

一种基于 Pareto 云隶属度的 MOPSO 算法

为定量解决非支配解排序问题,并兼顾多目标粒子群优化算法（multi-objective particle swarm optimization,MOPSO）的收敛性和多样性,提出了一种基于 Pareto 云隶属度的 MOPSO 算法。利用 Logistic 混沌映射优化种群的初始空间分布并融合布谷鸟搜索（cuckoo search,CS）指导粒子跳出局部陷阱,以增强算法的全局寻优能力。首次提出云向量评价 Pareto 最优解集方法,采用云隶属度评价准则对粒子适应度值进行量化评价。依据云隶属度选取个体最优和群体最优,平衡全局开发与开采,进而实现外部档案维护。测试函数集 ZDT 的实验结果表明,改进算法在收敛性和多样性方面较 MOPSO 和 NSGA-Ⅱ有一定优势。     

线性时滞多智能体系统的最优一致性研究

将一致性问题与代价函数相结合,研究了带有时滞的线性多智能体系统最优一致性问题。首先,采用模型变换将一致性问题转换成稳定性问题。进而,利用控制理论和线性矩阵不等式方法推导出控制器存在的一般性条件和代价函数性能指标上界。然后,将问题转换成凸优化问题,得到最优性能指标的一致性控制器。最后,通过仿真实例,验证了算法和判据的正确性。     

Reductive leaching of low-grade manganese ore with pre-processed cornstalk

Cornstalk is usually directly used as a reductant in reductive leaching manganese. However, low utilization of cornstalk makes low manganese dissolution ratio. In the research, pretreatment for cornstalk was proposed to improve manganese dissolution ratio. Cornstalk was preprocessed by a heated sulfuric acid solution (1.2 M of sulfuric acid concentration) for 10 min at 80°C. Thereafter, both the pretreated solu-tion and the residue were used as a reductant for manganese leaching. This method not only exhibited superior activity for hydrolyzing corn-stalk but also enhanced manganese dissolution. These effects were attributed to an increase in the amount of reductive sugars resulting from lignin hydrolysis. Through acid pretreatment for cornstalk, the manganese dissolution ratio was improved from 50.14%to 83.46%. The pre-sent work demonstrates for the first time the effective acid pretreatment of cornstalk to provide a cost-effective reductant for manganese leaching.     

New Seismic Attribute Technology for Predicting Dissolved Pore-Fracture of Deeply Buried Platform Margin Reef-Beach System in Northeast Sichuan Basin,China

The large reef complexes of the Upper Permian Changxing Formation, with a significant breakthrough for petroleum exploration, are an important target for petroleum exploration in the Yu-anba area of th...     

How does the retailer make pricing decision when consumers have justice reference price

A series of studies demonstrate that consumers have fairness judgments about the retailer's price which will impact their shopping decisions.Thus,it is necessary for the retailer to take consumers' fairness concerns into account when setting his pricing policies.We assume that when the retailer's price is lower than a consumer's justice reference price,the consumer is likely to sense a positive price unfairness that will lead to increased consumer utility,and when the retailer's price is higher than a consumer's justice reference price,the consumer is likely to sense a negative price unfairness that will have a negative effect on consumer utility.According to the information conditions of the consumers' justice reference price,the retailer should consider three situations:certain information,random information and partial information.In all situations,we show that the retailer has a unique optimal pricing strategy.Finally,through numerical examples,we find that our distribution-free policies perform almost the same as the results under the distributions that maximize the entropy.Our results reveal that it is important for the retailer to consider consumers' fairness concerns,otherwise he may suffer losses when p_c is very small.     

基于改进WNN的随动负载模拟器研究

为了抑制随动系统负载模拟器的多余力矩对力矩加载性能的影响,提出了一种基于改进的自学习函数扩展小波神经网络智能控制器。该控制器在误差大时采用Bang-Bang控制,误差小时采用基于函数扩展的小波神经网络和模糊补偿控制;同时采用基于改进的差分演变算法来估计控制器的参数;考虑到计算复杂度和控制系统性能,设计了隐含神经元数目的自学习算法。系统仿真实例结果表明,动静态响应性能均达到双十指标,该控制策略具有可行性和合理性,可以提高该系统力矩加载的跟踪性能和控制精度。