摘要：【中文】本发明提供的一种基于对抗训练机制的多语言实体关系抽取方法及系统，将目标实体对相关的各语言中的目标句子分别编码到各语言对应的独立语义空间和所有语言对应的一致性语义空间，获得目标句子中蕴含的各语言独立的信息和跨语言的一致的信息；再分别采用各语言独立的注意力机制和各语言间一致的注意力机制衡量每个目标句子相对各关系类型的注意力权重，最终结合所有目标句子相对各关系类型的注意力权重获得各关系类型对应的全局概率，从各关系类型对应的全局概率中选取出最大概率，最终即可根据最大概率对应的关系类型预测目标实体对之间的关系。该方法及系统能够深层地利用多语言间的互补性，有效提高了多语言场景下的关系抽取结果的准确性。 【EN】The invention provides a multilingual entity relation extraction method and system based on an antagonistic training mechanism, which respectively encode target sentences in related languages of a target entity into independent semantic spaces corresponding to the languages and consistent semantic spaces corresponding to all the languages, and acquire independent information of each language and consistent information of cross-language contained in the target sentences; and then respectively measuring the attention weight of each target sentence relative to each relationship type by adopting an independent attention mechanism of each language and a consistent attention mechanism among the languages, finally obtaining the global probability corresponding to each relationship type by combining the attention weights of all the target sentences relative to each relationship type, selecting the maximum probability from the global probabilities corresponding to each relationship type, and finally predicting the relationship between the target entity pairs according to the relationship type corresponding to the maximum probability. The method and the system can deeply utilize the complementarity among multiple languages, and effectively improve the accuracy of the relation extraction result in a multi-language scene.

摘要：【中文】本发明公开了鉴定多对生物分子间相互作用调控因子的方法。本发明运用多价互作大分子建立相变体系，并将每对互作生物分子中的一个生物分子与形成相变的多价大分子之一进行共价连接，使其聚集于相变液滴中；同时将每个生物分子的互作配体分别与不同报告基团相连，创建一系列重组生物分子，这些重组生物分子可与相变液滴中的相应生物分子形成生物分子对而发生相互作用进而聚集于相变液滴中，可通过检测不同报告基团的信号以确定相应生物分子对间是否具有相互作用。在此基础上，通过向体系中加入生物分子互作抑制剂，并通过检测相变液滴中不同报告基团信号强度的变化实现多靶点生物分子互作抑制剂的同步高通量筛选。 【EN】The invention discloses a method for identifying a plurality of pairs of biomolecular interaction regulatory factors. The invention uses multivalent interacting macromolecules to establish a phase change system, and one biomolecule in each pair of interacting biomolecules is covalently connected with one of the multivalent macromolecules forming phase change, so that the biomolecules are gathered in phase change liquid drops; meanwhile, the interaction ligands of each biomolecule are respectively connected with different report groups to create a series of recombinant biomolecules, the recombinant biomolecules can form biomolecule pairs with corresponding biomolecules in the phase-change liquid drop to interact and further gather in the phase-change liquid drop, and whether the corresponding biomolecule pairs have interaction or not can be determined by detecting signals of different report groups. On the basis, the biomolecule interaction inhibitor is added into the system, and the synchronous high-throughput screening of the multi-target biomolecule interaction inhibitor is realized by detecting the change of signal intensity of different reporter groups in the phase-change liquid drop.

摘要：【中文】本发明公开了一种水体质量监测及水体感官愉悦度评价方法，分析影响水体感官愉悦度的特征参数，确定表征水体感官质量的水体性状特征的水质指标种类。建立水体感官愉悦度分指数评价体系，并获得区间水体感官愉悦度分指数极值对应的水质指标的浓度限值。然后获取所述水质指标的监测值C_i，通过区间插值计算得到相应的水体感官愉悦度分指数WCI_i。确定所述水质指标的权重值a_i，且通过得到水体感官愉悦度指数WCI用以表征待测水体的水体感官愉悦度，实现水体感官质量的评价。本发明能够定量评价城市水体的感官质量，具有准确、高效、客观的优点。 【EN】The invention discloses a water quality monitoring and water sensory pleasure degree evaluation method, which analyzes characteristic parameters influencing the sensory pleasure degree of a water body and determines the water quality index types of water character characteristics representing the sensory quality of the water body. And establishing a water body sensory pleasure degree index evaluation system, and obtaining the concentration limit value of the water quality index corresponding to the extreme value of the water body sensory pleasure degree index in the interval. Then obtaining the monitoring value C of the water quality index_iAnd obtaining the corresponding water body sensory pleasure degree index WCI through interval interpolation calculation_i. Determining the weight value a of the water quality index_iAnd is andby passingAnd obtaining a water body sensory pleasure index WCI for representing the water body sensory pleasure of the water body to be detected and realizing the evaluation of the water body sensory quality. The method can quantitatively evaluate the sensory quality of the urban water body, and has the advantages of accuracy, high efficiency and objectivity.

摘要：【中文】本发明公开了一种非结构化道路条件下的行车环境态势评估方法及相应的评估装置。所述方法包括下述的步骤：步骤S1、获取行车环境要素信息，所述行车环境要素信息包括行车环境威胁要素信息和任务吸引要素信息；步骤S2、采用势能场和速度场方法，建立量化的行车环境要素态势模型；以及步骤S3、评估行车环境态势，其中，所述行车环境要素态势模型以行车环境威胁态势场U_p来反映行车环境威胁要素对自车的威胁，U_p＝U_pf+U_pv(1)。 【EN】The invention discloses a driving environment situation assessment method under an unstructured road condition and a corresponding assessment device. The method comprises the following steps: step S1, acquiring driving environment element information, wherein the driving environment element information comprises driving environment threat element information and task attraction element information; s2, establishing a quantitative driving environment element situation model by adopting a potential energy field and speed field method; and step S3, evaluating the driving environment situation, wherein the driving environment element situation model is a driving environment threat situation field U_pTo reflect the threat of the driving environment threat elements to the self vehicle, U_p＝U_pf+U_pv(1)。

摘要：【中文】本发明公开了柔性莫来石纤维气凝胶材料及其制备方法。制备柔性莫来石纤维气凝胶材料的方法包括：将聚合物材料与溶剂混合，得到聚合物溶液；将聚合物溶液与莫来石前驱体和催化剂混合，得到纺丝前驱体溶液；对纺丝前驱体溶液进行溶液喷射纺丝处理，得到含有聚合物材料、莫来石前驱体和催化剂的复合纤维气凝胶材料；对复合纤维气凝胶材料进行热处理，得到柔性莫来石纤维气凝胶材料。该柔性莫来石纤维气凝胶材料具有良好的柔韧性、可压缩性、耐火性、耐高低温性和隔热保温性能，并且不存在粉化的问题；该柔性莫来石纤维气凝胶材料制备过程简单可控，成本低，效率高，可重复性好，在防火服、航空航天、高温空气过滤等领域具有良好的工业化应用前景。 【EN】The invention discloses a flexible mullite fiber aerogel material and a preparation method thereof. The method for preparing the flexible mullite fiber aerogel material comprises the following steps: mixing a polymer material with a solvent to obtain a polymer solution; mixing the polymer solution with a mullite precursor and a catalyst to obtain a spinning precursor solution; carrying out solution jet spinning treatment on the spinning precursor solution to obtain a composite fiber aerogel material containing a polymer material, a mullite precursor and a catalyst; and carrying out heat treatment on the composite fiber aerogel material to obtain the flexible mullite fiber aerogel material. The flexible mullite fiber aerogel material has good flexibility, compressibility, fire resistance, high and low temperature resistance and heat insulation performance, and has no pulverization problem; the preparation process of the flexible mullite fiber aerogel material is simple and controllable, the cost is low, the efficiency is high, the repeatability is good, and the flexible mullite fiber aerogel material has a good industrial application prospect in the fields of fire-proof clothing, aerospace, high-temperature air filtration and the like.

摘要：【中文】靶向BTK降解化合物在治疗自身免疫系统疾病中的应用。本发明提出了化合物在制备药物中的用途，所述药物用于治疗或预防自身免疫疾病，该化合物为式I所示化合物或其立体异构体、几何异构体、互变异构体、氮氧化物、水合物、溶剂化物、代谢产物、药学上可接受的盐或前药，该化合物对自身免疫疾病有良好的治疗和预防效果。X‑Y‑Z式I。 【EN】Use of a targeted BTK degrading compound in the treatment of an autoimmune disease. The invention provides application of a compound in preparing a medicament for treating or preventing autoimmune diseases, wherein the compound is a compound shown in a formula I or a stereoisomer, a geometric isomer, a tautomer, a nitrogen oxide, a hydrate, a solvate, a metabolite, a pharmaceutically acceptable salt or a prodrug thereof, and the compound has good treatment and prevention effects on the autoimmune diseases. X-Y-Z is formula I.

摘要：【中文】一种线性自抗扰控制的动态无扰切换方法，属于工业自动化技术领域。该方法包括一阶线性自抗扰控制器和二阶线性自抗扰控制器的动态无扰切换方法。该动态无扰切换方法分别包括的跟踪算法，以及自抗扰控制与比例‑积分‑微分控制、手动控制三者之间的切换逻辑。本发明所提出的动态无扰切换方法，不仅可以在被控量等于设定值的稳态情况下，也可以在被控量不等于设定值的动态调节过程中实现无扰切换，保证控制量在切换过程中无跳变。仿真测试结果验证了该动态无扰切换方法的有效性，弥补了既有无扰切换方法的不足。 【EN】A dynamic undisturbed switching method for linear active disturbance rejection control belongs to the technical field of industrial automation. The method comprises a dynamic undisturbed switching method of a first-order linear active disturbance rejection controller and a second-order linear active disturbance rejection controller. The dynamic undisturbed switching method comprises a tracking algorithm and switching logics among active disturbance rejection control, proportional-integral-derivative control and manual control. The dynamic undisturbed switching method provided by the invention can realize undisturbed switching not only under the steady-state condition that the controlled quantity is equal to the set value, but also in the dynamic regulation process that the controlled quantity is not equal to the set value, thereby ensuring that the controlled quantity has no jump in the switching process. The simulation test result verifies the effectiveness of the dynamic undisturbed switching method, and makes up the defects of the existing undisturbed switching method.

摘要：【中文】本申请涉及一种碰撞严重性的预测方法，通过构建第一学习模型，并将整车数据和碰撞事故场景特征数据输入第一学习模型得出碰撞加速曲线，能够充分利用路段的道路交通情况和车辆自身数据，对车辆碰撞加速度进行实际计算。通过构建第二学习模型，并将乘员特征数据、约束系统特征数据和碰撞加速曲线输入至第二学习模型，得出不同人体部位的碰撞动力学曲线，可以综合考虑乘员特征和约束系统特征等实际碰撞场景特征，大大提升乘员碰撞严重性预测的可靠性。最终，通过不同人体部位的碰撞动力学曲线进行标量转化，实现碰撞严重性的量化。本申请涉及的碰撞严重性的预测方法，可以在考虑综合因素的前提下实现快速、准确的乘员碰撞严重性预测。 【EN】The application relates to a method for predicting the severity of a collision, which can be used for obtaining a collision acceleration curve by constructing a first learning model and inputting the data of the whole vehicle and the scene characteristic data of the collision accident into the first learning model, and can be used for fully utilizing the road traffic condition of a road section and the data of the vehicle per se to actually calculate the collision acceleration of the vehicle. By constructing the second learning model and inputting the passenger characteristic data, the restraint system characteristic data and the collision acceleration curve into the second learning model, collision dynamics curves of different human body parts are obtained, the passenger characteristics, the restraint system characteristics and other actual collision scene characteristics can be comprehensively considered, and the reliability of passenger collision severity prediction is greatly improved. Finally, scalar conversion is carried out through collision dynamic curves of different human body parts, and the quantification of the severity of the collision is realized. The method for predicting the severity of the collision can realize rapid and accurate prediction of the severity of the collision of the passenger on the premise of considering comprehensive factors.

摘要：【中文】本发明公开了一种基于地理位置信息的机构命名排歧方法及装置，其中，该方法包括：对输入文本信息抽取生成机构信息集合，对输入文本信息挖掘生成具有地理位置指向性的信息，并结构化为地理位置指向性信息集合；根据机构信息集合、地理位置指向性信息集合和多地图API，将机构信息集合中的每个机构映射为不同的地理位置，将映射结果进行结构化生成结构化机构信息集合；根据相似度算法对结构化机构信息集合进行相似度计算，根据计算结果生成机构相似度矩阵；根据机构相似度矩阵判断机构信息集合中的不同机构是否属于同一机构，根据判断结果进行机构命名排歧，输出排歧结果。该方法能够辅助实现快速准确地实现学术成果中不同机构之间的命名排歧。 【EN】The invention discloses a mechanism naming disambiguation method and a mechanism naming disambiguation device based on geographical location information, wherein the method comprises the following steps: extracting input text information to generate a mechanism information set, mining the input text information to generate information with geographical position directivity, and structuring the information into a geographical position directivity information set; according to the mechanism information set, the geographical position directivity information set and the multi-map API, mapping each mechanism in the mechanism information set into different geographical positions, and structuring the mapping result to generate a structured mechanism information set; similarity calculation is carried out on the structured mechanism information set according to a similarity calculation method, and a mechanism similarity matrix is generated according to a calculation result; and judging whether different mechanisms in the mechanism information set belong to the same mechanism according to the mechanism similarity matrix, naming and disambiguating the mechanisms according to the judgment result, and outputting a disambiguation result. The method can assist in quickly and accurately realizing naming disambiguation among different mechanisms in academic achievements.

摘要：【中文】本申请涉及一种燃料电池双极板。所述燃料电池双极板包括基板。所述基板包括第一表面。所述第一表面开设流道。所述流道用于输送反应气体。所述流道中反应气体的流量一定。沿着所述反应气体在所述流道的输送方向，所述流道的截面面积逐渐减小，所述反应气体流速逐渐增大，所述流道的排水速率逐渐增大。所述燃料电池双极板使得所述流道各位置的排水速率与液态水量相匹配，有效避免局部膜干和局部水淹。所述燃料电池双极板保证气体扩散层内不发生积水的同时质子交换膜不会过干，燃料电池内部极化损失降低，提高了燃料电池的性能。 【EN】The present application relates to a fuel cell bipolar plate. The fuel cell bipolar plate includes a substrate. The substrate includes a first surface. The first surface is provided with a flow passage. The flow channel is used for conveying reaction gas. The flow rate of the reaction gas in the flow channel is constant. Along the conveying direction of the reaction gas in the flow channel, the cross-sectional area of the flow channel is gradually reduced, the flow speed of the reaction gas is gradually increased, and the drainage rate of the flow channel is gradually increased. The fuel cell bipolar plate enables the drainage rate of each position of the flow channel to be matched with the liquid water quantity, and effectively avoids local membrane dryness and local flooding. The fuel cell bipolar plate ensures that the proton exchange membrane is not excessively dry while water is not accumulated in the gas diffusion layer, the polarization loss in the fuel cell is reduced, and the performance of the fuel cell is improved.