脉冲编码调制型气动位置伺服系统的研究

提出广义脉冲编码调制（ＧＰＣＭ）控制和自校正加速度观测．针对气动位置伺服系统，设计出自校正加速度观测器，并成功地应用到气动位置伺服控制中．研究表明：广义脉冲编码调制控制和自校正加速度观测改善了气动位置伺服控制系统的动、静态特性，提高了定位精度．     

二次充填开采矿压显现规律及充填效果分析

7623工作面是邢台矿现有的唯一分层充填开采工作面,开采煤层属于厚煤层,且该面地质条件复杂。通过矿压观测,找到了充填工作面充填体下二次开采矿压显现规律,通过对矿压显现规律与充填效果的分析,有效地减少了工作面片帮和冒顶事故的发生,同时强化两巷巷帮及三角地带的稳定性。     

Influence of the SiC matrix introduction time on the microstructure and mechanical properties of Cf/Hf0.5Zr0.5C-SiC ultra-high temperature composites

C_f/Hf_0.5Zr_0.5C-SiC composites were prepared by introducing Hf_0.5Zr_0.5C matrix (11 cycles) and SiC matrix (9 cycles) into the carbon cloth preform through precursor impregnation and pyrolysis (PIP) process. The influence of the introduction time of SiC matrix on the microstructure and mechanical properties of C_f/Hf_0.5Zr_0.5C-SiC composites was studied, and the results show that with the increase of the PIP cycles of the SiC matrix introduced before Hf_0.5Zr_0.5C matrix, the composite open porosity decreased, and the flexural strength and modulus presented an obvious upward trend. CS45 sample, which has 4 cycles of PIP SiC introduced in advance, has the highest flexural strength, flexural modulus and interfacial shear strength of 402.73 ± 35.73 MPa, 56.92 ± 3.97 GPa and 100.88 ± 7.79 MPa, respectively. Hf_0.5Zr_0.5C matrix has a loose and porous structure, so when more SiC matrix was introduced in advance, its covering effect on the surface of fibers led to less intra-bundle pores and thusly denser composite structure, and due to the compactness of SiC matrix, better overall bonding of fiber, interface and matrix was achieved, as well as better load transfer effect, which led to obvious interfacial debonding and cracking based on the in-situ SEM observation during flexural tests. While in the sample without pre-introduced SiC, the cracking occurred mainly between the interface and porous matrix and the overall performance of the material was poor.     

大渡河过饱和溶解气体原型观测研究

高坝泄洪会导致水中的总溶解气体(total dissolved gas,TDG)过饱和,其不利影响会持续较长范围,可能导致下游鱼类患气泡病甚至死亡.随着大量高坝的建成运行,这一生态环境问题更受关注.基于对大渡河水电站A的3年过饱和TDG原型观测数据,分析过饱和TDG生成释放的影响因素及其作用规律,并开展泄水方式优化的分...     

基于改进指针网络的卫星对地观测任务规划方法

针对卫星观测任务规划问题约束复杂、求解空间大和输入任务序列长度不固定的特点,使用深度强化学习（DRL）方法对卫星观测任务规划问题进行求解. 综合考虑时间窗口约束、任务间转移机动时间和卫星电量、存储约束,对卫星观测任务规划问题进行建模. 基于指针网络（PN）的运行机制建立序列决策算法模型,使用Mask向量来考虑卫星观测任务规划问题中的各类约束,并通过Actor Critic强化学习算法对模型进行训练,以获得最大的收益率. 借鉴多头注意力（MHA）机制的思想对PN进行改进,提出多头注意力指针网络（MHA-PN）算法. 根据实验结果可以看出,MHA-PN算法显著提高了模型的训练速度和泛化性能,训练好的MHA-PN算法模型可以直接对输入序列进行端到端的推理,避免传统启发式算法迭代求解的过程,具有较高的求解效率.     

密集观测场景下的敏捷成像卫星任务规划方法

针对密集观测场景下敏捷成像卫星任务规划问题求解空间大、输入任务序列较长的特点,综合考虑时间窗口约束、任务转移时卫星姿态调整时间、存储约束和电量约束,对敏捷成像卫星任务规划问题进行建模. 提出融合IndRNN和Pointer Networks的算法模型（Ind-PN）对敏捷成像卫星任务规划问题进行求解,使用多层的IndRNN结构作为算法模型的解码器. 基于Pointer Networks机制对输入任务序列进行选择,使用Mask向量考虑敏捷成像卫星任务规划问题中的各类约束. 基于Actor Critic强化学习算法对算法模型进行训练,以获得最大的观测收益率. 实验结果表明,对于密集观测场景下的任务规划,Ind-PN算法的收敛速度更快,可以获得更高的观测收益率.     

储层天然裂缝识别技术研究

针对裂缝性油气藏勘探开发的热点、重点和难点问题进行了论述。研究认为裂缝形成机理及发育特征识别技术成为广泛研究的热点和低孔、低渗储层评价的关键。同时,储层裂缝发育机制的复杂性决定了裂缝发育特征识别技术的难度,特别是天然裂缝的识别技术的研究仍是当前研究的难点,其对天然裂缝分布特征的认识准确与否,直接关系到对裂缝性储层的评价,影响地质储量的计算和经济评价,甚至关系到一个油气田能否顺利投入开发。     

In situ optical microscopy observation of the growth and rearrangement behavior of surface holes in the breath figure process

Using in situ optical microscope observations of the surface of polyphenylene oxide (PPO)/chloroform solution in the breath figure (BF) process, the detailed formation phenomenon of surface of the BF pattern was obtained. This phenomenon directly supports the “bursting hypothesis”. Based on the “bursting hypothesis” and additional experimental data, the mechanism for the late BF process of the PPO/chloroform system is discussed. Moreover, the formation process of defects on the film surface was also directly observed, and the defect formation mechanism of the system is discussed. This work can help to further understand the formation of patterns and defects on the surface of BF films.     

塔山矿综放工作面煤壁片帮规律分析研究

为了掌握特厚复杂煤层综放开采工作面煤壁片帮规律,在分析综放面回采工艺和机械设备的基础上,通过观测工作面周期来压期间和非周期来压期间煤层推进度与片帮深度及片帮率的关系,得出了采面片帮深度和片帮率的分布规律,为改进回采工艺和采面安全生产提供了可靠的科学数据。     

基于强跟踪滤波器的永磁同步电动机状态观测

针对永磁同步电动机(permanent magnet synchronous motor,PMSM)研究其在外界作用发生突变情况下的状态观测问题。首先基于理想状态下PMSM的数学模型,采用强跟踪滤波器(the strong tracking filter,STF)算法估计电动机的转速、转角、磁链和转矩,并通过仿真比较STF算法与扩展卡尔曼滤波器(extended Kalman filter,EKF)算法的估计性能,结果表明STF与EKF相比对突加给定起动以及突加扰动等因素引起的状态快速变化能够实现更快速、更准确的跟踪。最后在考虑电动机的铁耗情况下分别通过这两种方法进行状态观测,所得结果与上述结论相符,进一步表明将STF应用于实际电动机进行状态观测的可行性和优越性。