振动沉管挤密碎石桩在闸基处理中的应用

振动沉管挤密碎石桩是近十几年来在挤密砂桩和振冲桩的基础上发展起来的地基处理新技术,具有设备简单、操作方便、成本低、施工快、无污染等特点，被工程界广泛使用，积累了一定的工程经验。其在共产主义闸地基处理中的应用即为一例。     

多特征自适应融合的粒子滤波跟踪算法

针对跟踪过程中目标形态不断变化或部分遮挡导致鲁棒性差的问题,提出一种基于多特征自适应融合的粒子滤波跟踪算法.该算法从视觉特征集中选取了描述能力强的2种特征,并将其按照与目标模型的多尺度相似度进行线性融合;为了减小跟踪漂移,通过计算当前目标模型与初始目标模型的多尺度相似度自适应地更新目标模型.大量仿真实验结果表明,文中算法可以鲁棒地跟踪到部分遮挡和形态变化的运动目标.     

利用RSS推送的主动服务解决藏书呆滞的研究

将通过系统提取的读者兴趣及需求信息与不断更新的RSS资源库进行比对,筛选出最能满足用户要求的RSS信息,并通过RSS主动推送技术,将资源及时发送给读者,是提高馆藏图书利用率,减少图书呆滞量的一种有效途径。     

带平衡约束矩形布局优化问题的深度强化学习算法

带平衡约束的矩形布局问题源于卫星舱设备布局设计,属于组合优化问题。深度强化学习利用奖赏机制,通过数据训练实现高性能决策优化。针对布局优化问题,提出一种基于深度强化学习的新算法DAR及其扩展算法IDAR。DAR用指针网络输出定位顺序,再利用定位机制给出布局结果,算法的时间复杂度是O（n³）;IDAR算法在DAR的基础上引入迭代机制,算法时间复杂度是O（n⁴）,但能给出更好的结果。测试表明DAR算法具有较好的学习能力,用小型布局问题进行求解训练所获得的模型,能有效应用在大型问题上。在两个大规模典型算例的对照实验中,提出算法分别超出和接近目前最优解,具有时间和质量上的优势。     

改进蚁群算法在路径规划中的应用

针对传统蚁群算法在机器人路径规划时存在收敛速度慢、易陷入局部最优等问题,提出了一种基于自适应归档更新的蚁群算法。根据路径性能指标建立多目标性能评估模型,对最优路径进行多指标优化;采用路径方案归档更新策略进行路径方案的更新和筛选,提高算法的收敛速度;当搜索路径进入不可行区域时,采用自适应路径补偿策略转移不可行路径节点,构造可行路径,减少死锁蚂蚁数量;若算法无法避开障碍或者进入停滞状态,则进行种群重新初始化,增加物种多样性,避免算法陷入局部最优。仿真实验表明,改进后的算法收敛速度更快、收敛精度更高、稳定性更好。     

Consistency retention method for CNC machine tool digital twin model

Computer Numerical Control Machine Tool (CNCMT) Digital Twin (DT) model is a carrier for complex, time-varying, coupled data of CNCMT, which can theoretically provide a time-varying high-fidelity model. However, there are still many difficulties in its implementation process. And the key issue is how to realize the updated DT model with performance attenuation and validate it. In order to solve this problem, a model consistency retention method for CNCMT DT model is studied and proposed in this paper. Firstly, the framework of consistency retention method for DT model is designed including both digital space and physical space. The principles of data management and performance attenuation update in digital space are elaborated. Then, the implementation method for consistency retention of CNCMT DT model is studied in terms of performance attenuation update workflow for wear and other damage separately. Finally, a case study for the establishment and application of high-fidelity test bench DT model that focusing on rolling guide-rail is carried out to show the implementation flow of the proposed method and verify its operability and effectiveness.     

Behavioral and electrophysiological activity of (Z,E)-7,9,11-dodecatrienyl formate,a mimic of the major sex pheromone component of carob moth,Ectomyelois ceratoniae

The behavioral and electrophysiological activity of a mimic [(Z,E)7,9,11-dodecatrienyl formate] of the major sex pheromone component [(Z,E) 9,11,13-tetradecatrienal] of carob moth was assessed. Wind-tunnel bioassays demonstrated that the formate was as effective as natural gland extracts, and significantly more effective than the trienal alone or than the trienal blended with two minor pheromone components, in evoking source contact. Dispensers containing the formate were as effective as trienal-containing blend lures in attracting males when placed at the same dosage in traps in date gardens. Single-cell recordings showed that at least two olfactory neurons, differentiated by spike amplitude, are located in the long trichoid hairs on male carob moth antennae. Dose-response relationships indicated that puffs from cartridges loaded with at least 0.1 g of the formate or of the trienal were necessary to elicit spiking by either the small or the large-spiking cell within a sensillum. Cross-adaptation studies demonstrated that both compounds stimulated the same large-spiking cell. The frequencies of spikes evoked from the large cell when stimulated by emissions from 0.1-g, 1-g, or 10-g cartridges of either the formate or the trienal were not significantly different, suggesting that the formate is an effective mimic of the trienal at the antennal receptor cell level.     

Continuous process control using neural networks

We present some adaptive control strategies based on neural networks that can be used for designing controllers for continuous process control problems. Specifically, a learning algorithm has been formulated based on reinforcement learning, a weakly supervised learning technique, to solve set-point control and control scheduling for continuous processes where the process cannot be modeled easily. It is shown how reinforcement learning can be used to learn the control strategy adaptively based on exploration of the control space without making assumptions about the process model. A new learning scheme, handicapped learning, was developed to learn a control schedule that specifies a schedule of set points. Applications studied include the control of a nonisothermal continuously stirred tank reactor at its unstable state and the learning of the daily time-temperature schedule for an environment controller. Experimental results demonstrate good learning performance, indicating that the learning algorithm can be used for solving transient startup and boundary value control problems.     

机器人三维路径规划问题的一种改进蚁群算法

本文研究移动机器人三维空间路径规划问题,针对三维空间的复杂地形特点,提出了一种基于改进蚁群算法的路径规划算法。文中首先描述了一种简单有效的环境建模方法,然后给出了算法在信息素呈现、路径点选取、信息素更新以及启发式函数设计等方面的改进方法。仿真结果证明了算法的可行性和可靠性。     

改进的蚁群算法在飞行路径模型中的应用

为了保障飞机安全地运行,FAA规定每架飞机在飞行一定时间后都要进行周期性检修。通常考虑到检修成本,任一种检修都要在指定的机场完成,因此要求飞机每飞行一定的小时数后都要返回到指定的维护基地进行定期检修。如何合理地进行飞机飞行路径的调度以使飞机周期性地返回指定检修机场进行检修是本文研究的关键。为了问题简化,本文主要针对单检修基地、单机型的检修路径调度问题进行研究。本文结合经典蚁群算法,对现有算法参数的设置及信息的更新等方面进行改进后应用到飞行路径模型中,并运用Matlab对该算法编程,运用具体数据进行计算实现,确保了飞机周期性返回检修机场,实现周期性检修。