基于类脑模型与深度神经网络的目标检测与跟踪技术研究

目标检测与跟踪技术广泛应用于交通、医疗、安保和航天等领域.目前，目标检测与跟踪技术面临目标微弱、背景复杂、目标被遮挡等挑战.同时，随着脑科学研究的不断深入，人们对人脑视觉系统的理解逐渐透彻，利用类脑计算解决复杂背景下高精度目标检测与跟踪问题成为相关领域的重要研究方向.本文结合神经工程导向的类脑模型和计算机工程导向的深度神经网络(Deep Neural Networks, DNNs)，提出多种基于类脑模型与深度神经网络的目标检测与跟踪算法，包括：基于演算侧抑制的目标检测算法，基于结构 对比度(Structure Contrast, SC)视觉注意模型的弱小目标检测算法和基于记忆机制与分层卷积特征的目标跟踪算法.实验结果表明，将类脑模型和深度神经网络应用于目标检测和跟踪领域，有利于实现复杂条件下的高精度目标检测和鲁棒性目标跟踪.     

Effect of curing condition on bonding quality of scarf-repaired composite laminates

Both single-face vacuum bag curing (SVC) and double-face vacuum bag curing (DVC) can be used in scarf repair of composite structures. But different curing conditions caused by the sealing state may affect the bonding quality of scarf-repaired structures. In this paper, the effect of curing condition on bonding quality of scarf-repaired laminates was experimentally investigated in terms of surface profiles, moisture absorption curves and section profiles. In order to further explore the moisture absorption mechanism, finite element model of the repaired laminates using DVC was established with moisture diffusion of both the adhesive and composite laminates considered. This model was verified by experimental results. Based on the model of DVC case, the model of SVC case was built by changing moisture absorption parameters of the adhesive. Results show that SVC reduces the bonding quality, mainly reflecting in more adhesive inner voids and patch-to-parent dislocation. And SVC increases moisture absorption rate and moisture equilibrium content of the adhesive, and its effect on the former is far greater than that on the latter.     

A boundary vector cell model of place field repetition

Hippocampal place cells are thought to form the neural substrate of a global cognitive map. However, in multicompartment mazes, these cells exhibit locally repeating representations, undermining the global cognitive map view of place cells. This phenomenon appears to be related to the repetitive layout of these mazes, but still no hypothesis adequately explains it. Here, we use a boundary vector cell (BVC) model of place cell firing to model the activity of place cells in numerous multicompartment environments. The activity of modeled place cells bears a striking resemblance to experimental data, replicating virtually every major experimental result. Our results support the BVC model and indicate that locally repeating place cell firing could result purely from local geometry.     

直升机惯性传感器结构的模态优化

提出了一种基于迭代抽样和径向基插值的自适应代理模型方法。这种自适应方法以减少仿真计算数量和提高代理模型自适应能力为目的,使用多岛遗传算法选择新增样本点并使新增样本点位于设计空间的稀疏区域,使得所有的样本点均匀分布于设计空间。标准误差用来判断代理模型的精度大小以决定是否对代理模型进行更新。这种自适应代理模型结合多岛遗传算法对直升机的惯性传感器结构模态进行优化。用拉丁超立方抽样方法选择10个样本点构建初始的代理模型,自适应代理模型的计算结果表明2%的误差条件下需要额外增加7个样本点。优化结果表明不同的权重系数对最优模态特性的影响很大,惯性传感器结构的一至六阶模态值更加远离直升机的激励频率。     

基于Kriging模型梯度解析解的改进一次二阶矩方法

改进一次二阶矩（AFOSM）法是一种基于功能函数梯度的结构可靠性分析方法，鉴于其对隐式函数的梯度较难求解，提出了一种基于Kriging模型梯度解析解的AFOSM方法，利用Kriging代理模型的解析表达式推导得到功能函数对输入变量的梯度解析解，为AFOSM中设计点的确定提供高精度的梯度信息。通过Kriging与AFOSM的结合，很好地解决了基于有限元模型的隐式情况下梯度计算量相当大、可靠性分析难的问题。数值与工程算例验证了所提Kriging梯度解析解的较高精确性，同时验证了所提基于Kriging解析解的AFOSM结构可靠性分析方法的正确性与较高精度。     

Application of a PCA-DBN-based surrogate model to robust aerodynamic design optimization

An efficient method employing a Principal Component Analysis (PCA)-Deep Belief Network (DBN)-based surrogate model is developed for robust aerodynamic design optimization in this study. In order to reduce the number of design variables for aerodynamic optimizations, the PCA technique is implemented to the geometric parameters obtained by parameterization method. For the purpose of predicting aerodynamic parameters, the DBN model is established with the reduced design variables as input and the aerodynamic parameters as output, and it is trained using the k-step contrastive divergence algorithm. The established PCA-DBN-based surrogate model is validated through predicting lift-to-drag ratios of a set of airfoils, and the results indicate that the PCA-DBN-based surrogate model is reliable and obtains more accurate predictions than three other surrogate models. Then the efficient optimization method is established by embedding the PCA-DBN-based surrogate model into an improved Particle Swarm Optimization (PSO) framework, and applied to the robust aerodynamic design optimizations of Natural Laminar Flow (NLF) airfoil and transonic wing. The optimization results indicate that the PCA-DBN-based surrogate model works very well as a prediction model in the robust optimization processes of both NLF airfoil and transonic wing. By employing the PCA-DBN-based surrogate model, the developed efficient method improves the optimization efficiency obviously.     

基于变刚度变阻尼的柔性喷管动态模型

针对柔性接头动态迟滞曲线受控制系统控制位置精度和动态响应速度影响较大的问题,基于电液伺服机构和柔性接头变刚度变阻尼模型,构建了柔性喷管的电液伺服机构-变刚度变阻尼模型,将其和电液伺服机构-定刚度定阻尼模型进行了对比。分析了电液伺服机构主要参数、柔性接头工作参数等对电液伺服机构-柔性接头系统动态特性的影响。分析结果表明:电液伺服机构-变刚度变阻尼模型所构造的迟滞曲线可更准确地与实验结果相吻合,并符合迟滞曲线随频率变化的规律,反馈系数、放大器静态放大系数、电液伺服机构增益、滑阀流量增益等参数对系统动态特性的影响更为明显。该模型为固体火箭发动机电液伺服机构-柔性接头系统动态特性的调整提供理论依据。     

变体辅助的无人机栖落机动模糊控制设计

无人机（Unmanned aerial vehicle， UAV）的栖落机动是一种大幅度的俯仰运动，易引起升降舵操纵力矩饱和。本文以变体方式增强无人机的俯仰操纵能力，并研究其对应的控制设计方法。首先对栖落机动建立了纵向动力学模型，并通过采用轨迹线性化和张量积变换方法转换得到T-S模糊模型。基于Lyapunov稳定理论和平方和方法，设计了满足控制输入约束的栖落机动多项式模糊控制器。对非变体与变体下的栖落机动控制过程进行了仿真，结果验证了控制律的有效性，并且表明变体辅助的无人机具有更强的操纵性能，能提高栖落机动中升降舵的抗饱和能力。     

移15°双十二相隐极同步整流发电机数学模型的研究

分别在abc坐标系和dq0坐标系下建立了移15°双十二相隐极同步整流发电机的基本数学方程,推导了电机电压、磁链、输出功率、电磁转矩和转子运动方程。在MATLAB/Simulink中建立了移15°双十二相隐极同步整流发电机的仿真模型。对两路输出进行了不平衡负载仿真和试验。仿真与试验结果相比基本一致,验证了移15°双十二相隐极同步整流发电机数学模型的正确性。这为双十二相同步整流发电机运行性能和励磁控制系统的研究奠定了基础。     

Fault diagnosis based on measurement reconstruction of HPT exit pressure for turbofan engine

Aero-engine gas path health monitoring plays a critical role in Engine Health Management(EHM). To achieve unbiased estimation, traditional filtering methods have strict requirements on measurement parameters which sometimes cannot be measured in engineering. The most typical one is the High-Pressure Turbine(HPT) exit pressure, which is vital to distinguishing failure modes between different turbines. For the case of an abrupt failure occurring in a single turbine component, a model-based sensor measurement reconstruction method is proposed in this paper. First,to estimate the missing measurements, the forward algorithm and the backward algorithm are developed based on corresponding component models according to the failure hypotheses. Then,a new fault diagnosis logic is designed and the traditional nonlinear filter is improved by adding the measurement estimation module and the health parameter correction module, which uses the reconstructed measurement to complete the health parameters estimation. Simulation results show that the proposed method can well restore the desired measurement and the estimated measurement can be used in the turbofan engine gas path diagnosis. Compared with the diagnosis under the condition of missing sensors, this method can distinguish between different failure modes, quantify the variations of health parameters, and achieve good performance at multiple operating points in the flight envelope.