短模型上的明渠非恒定流试验

水工整体模型试验中,明渠非恒定流至今大多被简化为恒定流,而效果有时并不理想。本文指出,只要设置固定式曲线尾门,并有必要的动态量测设备,便可模拟原型非恒定流;即使模型偏短,也能得出比恒定流试验更接近原型的成果。在同一模型上,既做恒定流试验,又进行必要的非恒定流试验,二者相辅相成,必将有助于试验水平的提高。作为例证,本文介绍了泄洪调度导致航道水面波动的一项试验成果,并以岩基冲刷、通航水流条件、围堰过水、河床冲淤等研究课题为例,讨论了采用“短模型非恒定流试验方式”的必要性与可行性。     

Quantum immune clonal coevolutionary algorithm for dynamic multiobjective optimization

The existing algorithms to solve dynamic multiobjective optimization (DMO) problems generally have difficulties in non-uniformity, local optimality and non-convergence. Based on artificial immune system, quantum evolutionary computing and the strategy of co-evolution, a quantum immune clonal coevolutionary algorithm (QICCA) is proposed to solve DMO problems. The algorithm adopts entire cloning and evolves the theory of quantum to design a quantum updating operation, which improves the searching ability of the algorithm. Moreover, coevolutionary strategy is incorporated in global operation and coevolutionary competitive operation and coevolutionary cooperative operation are designed to improve the uniformity, the diversity and the convergence performance of the solutions. The results on test problems and performance metrics compared with ICADMO and DBM suggest that QICCA has obvious effectiveness and advantages which shows great capability of evolving convergent, diverse and uniformly distributed Pareto fronts.     

打车软件混合随机均衡模型时间可靠性研究*

为了探究打车软件出现后对乘客打车行为的影响,揭示是否使用打车软件对乘客候车时间长短的变化规律,基于路网混合随机均衡模型理论,在考虑打车软件对出租车司机搜索行为影响和随机变化的OD对需求条件下,构建基于弹性需求的打车软件随机用户均衡模型;然后结合乘客等车、司机空驶时间和可靠性计算模型,引入连续平均（MSA）算法和随机路径流量分配（Dial）算法进行求解;最后通过算例验证了所建模型和算法的有效性。研究结果表明不同需求弹性下、是否使用打车软件对会乘客候车时间及候车时间可靠性产生明显的差异。     

利用X射线脉冲星的星间差分联合定位方法简

Relative navigation traditionally using inter-satellite links demands higher inter-satellite time synchronization, and has no ability of absolute positioning and timing. In this paper, we propose a relative navigation model of X-ray pulsar-based co-localization. The main satellite receives the radiation pulse signal from X-ray pulsars, and sends the arrival time and individual pulse's feature to adjacent satellites through inter-satellite links. Then by detecting the inter-satellite time difference of X-ray pulsar signals we can establish the location equation to find the inter-satellite relative position. We deduce the error correction equation of the pulse TOA prediction model and analyze the identifiability of individual pulses of the pulsar. The method proposed in this paper is less influenced by the TOA prediction model, ephemeris error, etc. It is helpful for achieving on-line self-adaption correction of the TOA prediction model and combined orbit determination of satellite constellation.     

TIG弧焊机器人系统在宇航大型铝合金贮箱箱底拼焊中的应用

对TIG弧焊机器人应用于宇航大型铝合金贮箱底的拼焊技术进行了初步探讨，介绍了机器人系统的构成，研究了机器人的示教，再现过程，选定了焊拉工艺参数，认为弧焊机器人焊接宇航大型铝合金贮箱箱底有很好的应用价值与前景。     

堆焊结构的氢浓度分布及其对剥离断裂的影响

采用有限元分析方法计算堆焊结构中氢浓度的分布，并将电解充氢试验条件下试块中氢扩散行为的计算结果和高压釜试验条件的计算结果进行对比分析。计算结果表明，采用电解充氢方法，可以在熔合线附近获得与高压釜充氢数值相近的氢浓度峰值，而到达这一峰值的时间较短。通过解析模型比较了不同试样条件和电解充氢条件对试验结果的影响。结果显示，增大充氢电流密度，延长充氢时间，减薄母材厚度，均使氢浓度峰值升高，从而使堆焊层剥离     

多模态医学图像融合超声检查系统的设计与实现

超声检查系统所呈图像中缺乏三维结构信息,且商业的医学图像分析软件价格昂贵。为此,开发一款综合病灶部位的三维模型和二维超声图像各自特征以及对应空间关系的检查系统。基于开源的医学图像处理软件、商用的电磁定位系统和超声成像系统,在局域网环境下,遵循OpenIGTLink协议进行数据共享的超声检查系统。该系统主要由超声图像信息共享模块、超声图像显示模块、超声探头实时跟踪模块、2D/3D配准融合模块组成。实验结果表明,该系统可初步应用于临床教学和临床辅助诊断等领域,具有价格低廉、扩展性强、维护方便等优势。     

短时记忆与CenterTrack的车辆多目标跟踪

目的 车辆多目标跟踪是智能交通领域关键技术,其性能对车辆轨迹分析和异常行为鉴别有显著影响。然而,车辆多目标跟踪常受外部光照、道路环境因素影响,车辆远近尺度变化以及相互遮挡等干扰,导致远处车辆漏检或车辆身份切换（ID switch,IDs）问题。本文提出短时记忆与CenterTrack的车辆多目标跟踪,提升车辆多目标跟踪准确度（multiple object tracking accuracy,MOTA）,改善算法的适应性。方法 利用小样本扩增增加远处小目标车辆训练样本数;通过增加的样本重新训练CenterTrack确定车辆位置及车辆在相邻帧之间的中心位移量;当待关联轨迹与检测目标匹配失败时通过轨迹运动信息预测将来的位置;利用短时记忆将待关联轨迹按丢失时间长短分级与待匹配检测关联以减少跟踪车辆IDs。结果 在交通监控车辆多目标跟踪数据集UA-DETRAC （University at Albany detection and tracking）构建的5个测试序列数据中,本文方法在维持CenterTrack优势的同时,对其表现不佳的场景获得近30%的提升,与YOLOv4-DeepSort（you only look once—simple online and realtime tracking with deep association metric）相比,4种场景均获得近10%的提升,效果显著。Sherbrooke数据集的测试结果,本文方法同样获得了性能提升。结论 本文扩增了远处小目标车辆训练样本,缓解了远处小目标与近处大目标存在的样本不均衡,提高了算法对远处小目标车辆的检测能力,同时短时记忆维持关联失败的轨迹运动信息并分级匹配检测目标,降低了算法对跟踪车辆的IDs,综合提高了MOTA。     

Feature selection and parameter optimization for support vector machines: A new approach based on genetic algorithm with feature chromosomes

Support vector machines (SVM) are an emerging data classification technique with many diverse applications. The feature subset selection, along with the parameter setting in the SVM training procedure significantly influences the classification accuracy. In this paper, the asymptotic behaviors of support vector machines are fused with genetic algorithm (GA) and the feature chromosomes are generated, which thereby directs the search of genetic algorithm to the straight line of optimal generalization error in the superparameter space. On this basis, a new approach based on genetic algorithm with feature chromosomes, termed GA with feature chromosomes, is proposed to simultaneously optimize the feature subset and the parameters for SVM.To evaluate the proposed approach, the experiment adopts several real world datasets from the UCI database and from the Benchmark database. Compared with the GA without feature chromosomes, the grid search, and other approaches, the proposed approach not only has higher classification accuracy and smaller feature subsets, but also has fewer processing time.     

FPN推理模型在上下文感知系统中的应用研究

研究模糊Petri网模型在上下文感知系统中的快速推理算法.针对模糊Petri网的通用推理算法不能解决“或逻辑”问题,提出了模糊Petri网推理算法从变迁开始搜索,区别于通常从库所开始搜索的策略,有效解决了“或逻辑”问题;并在不增加虚库所和虚变迁的前提下对变迁先排序后再进行推理,降低了算法的空间和时间复杂度.最后,对模糊Petri网模型和推理算法进行仿真,实验结果验证了该算法的有效性.