摘要：本发明涉及一种煤矿钻孔防塌孔管网和防塌孔方法，管网是在螺旋弹簧的螺旋臂上固定有螺旋的网片，或者在螺旋弹簧的螺旋臂上固定有柔性透气层。另外，本发明利用管网留在孔内，由于管网有一定强度，可以支撑煤体，保持钻孔形状。利用可折叠钻头，使管网内径与钻头匹配，同时管网能更容易地随钻杆插入钻孔。

摘要：【中文】本发明公开了一种开关组合模块及家用电器，其中开关组合模块包括开关压块和微动开关，至少一个微动开关设置在开关压块四周；开关压块在侧面形成有凸缘，微动开关设有触头，凸缘和触头一一对应；开关压块的转动，使部分或者全部凸缘同时与对应的触头接触。本发明所述的开关组合模块，通过一个简单的旋转结构，实现了对多个微动开关的同步控制，而且占用空间较小，成本廉价。 【EN】The invention discloses a switch combined module and a household appliance, wherein the switch combined module comprises a switch pressing block and a microswitch, and at least one microswitch is arranged around the switch pressing block; the switch pressing block is provided with a flange on the side surface, the microswitch is provided with contacts, and the flange corresponds to the contacts one to one; and the switch pressing block rotates to enable part or all of the flanges to be in contact with the corresponding contacts at the same time. The switch combined module realizes synchronous control of a plurality of micro switches through a simple rotating structure, and has the advantages of small occupied space and low cost.

摘要：【中文】本说明书实施例提供一种目标文本检测方法、模型训练方法、装置及设备。在获取待检测文本后，基于待检测文本中各词语的n元拼音以及n元笔画得到各词语的第一词向量，然后将第一词向量输入到预先训练的目标文本检测模型，检测文本是否为目标文本。由于第一词向量是基于所述词语的n元拼音以及n元笔画得到，因而具有“音”和“形”的特性，且更细粒度表现了词语之间的关联，因而在根据第一词向量来检测文本是否为目标文本时，可以更加准确地检测出一些形近和音近的目标文本。 【EN】The embodiment of the specification provides a target text detection method, a model training method, a device and equipment. After the text to be detected is obtained, a first word vector of each word is obtained based on n-element pinyin and n-element strokes of each word in the text to be detected, and then the first word vector is input into a pre-trained target text detection model to detect whether the text is the target text. Because the first word vector is obtained based on the n-element pinyin and the n-element strokes of the words, the first word vector has the characteristics of 'sound' and 'shape', and the association between the words is expressed in a finer granularity, so that when whether the text is the target text or not is detected according to the first word vector, the target texts with the shape close to the shape.

摘要：【中文】本发明公开了一种节能型烟道脱硫装置，包括依次连接的料仓、给料机和下料靴；还包括布袋除尘器；所述布袋除尘器通过第一烟道与锅炉的空预器连接；所述锅炉的高温省煤器和低温省煤器之间的第二烟道上设置第一脱硫喷射点，所述烟道上靠近布袋除尘器的位置设置第二脱硫喷射点，所述第一脱硫喷射点处设置用于粉体输送分配的第一分配器，所述第二脱硫喷射点处设置用于粉体输送分配的第二分配器；所述第一分配器与第二分配器均通过送粉管道与下料靴的出粉口连接，本脱硫装置改造成本低、吸附效果好、吸附效率高、脱硫剂使用量少，脱硫成本低。 【EN】The invention discloses an energy-saving flue desulfurization device, which comprises a storage bin, a feeder and a discharging shoe which are connected in sequence; also comprises a bag-type dust collector; the bag-type dust collector is connected with an air preheater of the boiler through a first flue; a first desulfurization spraying point is arranged on a second flue between a high-temperature economizer and a low-temperature economizer of the boiler, a second desulfurization spraying point is arranged on the position, close to a bag-type dust collector, on the flue, a first distributor for powder conveying and distribution is arranged at the first desulfurization spraying point, and a second distributor for powder conveying and distribution is arranged at the second desulfurization spraying point; first distributor and second distributor all are connected through the powder outlet of sending powder pipeline with the unloading boots, and this desulphurization unit reforms transform with low costs, adsorption effect is good, adsorption efficiency is high, the desulfurizer use amount is few, and the desulfurization is with low costs.

摘要：【中文】本说明书实施例提供一种文本检测方法、相似度计算方法、模型训练方法、装置及设备。获取第二文本以及待检测的第一文本，生成所述第一文本以及所述第二文本中各词语的向量集合，所述向量集合包括所述词语的词向量以及n‑元笔画向量，将所述第一文本中各词语、所述第二文本中各词语以及所述向量集合输入到预先训练的文本相似度计算模型，以计算所述第一文本和第二文本的相似度，基于所述相似度以及所述第二文本的类别确定所述第一文本是否为目标类别的文本。基于文本各词语的n元笔画向量计算文本相似度，可以从更粒度地提取词语之间的关联，也解决了预测中出现新词语的问题，可以有效检测测出目标类别的文本。 【EN】The embodiment of the specification provides a text detection method, a similarity calculation method, a model training method, a device and equipment. The method comprises the steps of obtaining a second text and a first text to be detected, generating a first text and a vector set of words in the second text, wherein the vector set comprises word vectors of the words and n-element stroke vectors, inputting the words in the first text, the words in the second text and the vector set into a pre-trained text similarity calculation model to calculate the similarity of the first text and the second text, and determining whether the first text is a text of a target category or not based on the similarity and the category of the second text. The text similarity is calculated based on the n-element stroke vectors of all words of the text, the association between the words can be extracted more granularly, the problem of new words in prediction is solved, and the text of the target category can be effectively detected and detected.

摘要：【中文】本说明书实施例提供一种分类模型的训练、对象分类方法及装置，在训练方法中，获取带分类标签的样本。在嵌入层中，确定样本的特征向量以及分类标签的标签向量。在卷积层中，基于若干不同宽度的卷积窗口，对样本的特征向量进行多次卷积处理，得到多个卷积结果。在池化层中，计算各卷积结果与分类标签的标签向量之间的相似度，并基于计算得到的相似度，确定对应于各卷积结果的注意力权重值。基于对应于各卷积结果的注意力权重值，对各卷积结果进行加权平均池化操作，得到池化结果。将池化结果作为样本的样本表示向量，并至少基于样本表示向量以及分类标签的标签向量，确定预测损失。基于预测损失，调整分类模型的参数。 【EN】The embodiment of the specification provides a method and a device for training a classification model and classifying an object. In the embedding layer, a feature vector of the sample and a tag vector of the class tag are determined. In the convolutional layer, based on a plurality of convolutional windows with different widths, the feature vectors of the samples are subjected to convolutional processing for a plurality of times to obtain a plurality of convolutional results. In the pooling layer, a similarity between each convolution result and the label vector of the classification label is calculated, and an attention weight value corresponding to each convolution result is determined based on the calculated similarity. And performing weighted average pooling operation on each convolution result based on the attention weight value corresponding to each convolution result to obtain a pooled result. The pooling result is taken as a sample representation vector of the samples, and a prediction loss is determined based on at least the sample representation vector and a label vector of the classification labels. Based on the predicted loss, parameters of the classification model are adjusted.

摘要：本发明公开了一种旋转射流抛光装置，包括射流喷嘴部件，射流喷嘴部件包括旋转接头、转动主轴、抛光喷嘴和贯穿式电机；转动主轴为空心轴，转动主轴的两端分别与旋转接头的输出端和抛光喷嘴相连；贯穿式电机包括中空转子和定子，中空转子转动时，通过转动主轴带动抛光喷嘴旋转；抛光加工时，首先，将射流喷嘴部件安装与运动平台或是工业机器人机械臂上，供液系统依次通过供液管路、旋转接头的输入端、输出端和转动主轴的空心内腔，最终将抛光液到达抛光喷嘴，并以一定压力和速度喷射到工件表面。本发明可以减小射流抛光产生的高频误差，保证射流冲击工件表面形成的去除函数是回转对称且中心去除量最大的类高斯型去除函数，同时提高射流抛光的效率。

摘要：本发明公开了一种复杂型面器件超精密射流抛光装置，包括供液系统和安装在六轴运动平台上的抛光喷嘴部件，所述供液系统将预混合的抛光液以指定的压力和速度通过所述射流喷嘴喷射到固定在工作台上的待加工工件的表面上，所述抛光液中的颗粒与待加工工件的表面相互作用，进行材料的纳米去除，最终实现对待加工工件的表面材料进行去除和修整加工。本发明中射流喷嘴具有偏心单孔结构、偏心多孔结构和非偏心多孔结构多种不同的结构形式，可以根据需要进行选择。采用本发明射流抛光装置进行抛光加工，可以保证在复杂型面器件超精密射流抛光过程中供液系统能够提供压力、浓度、温度、流量、PH值等稳定的抛光液，保证工件表面的抛光质量。

摘要：本发明公开了一种基于预混磨料的脉冲水射流抛光方法，利用脉冲水射流抛光装置实现，脉冲水射流抛光装置包括磨料搅拌单元和脉冲喷射单元，磨料搅拌单元包括搅拌器控制器，脉冲喷射单元包括脉冲装置、脉冲控制电机、脉冲控制器和用于喷射射流水柱喷嘴，通过控制搅拌电机、供液电机和的脉冲控制电机的状态及运动工作台的运动，以改变工件相对射流水柱的位置，通过压力控制器调节射流水柱与工件表面的接触压力，使得射流水柱与工件表面的接触压力和速度发生瞬态变化，从而改变射流水柱与工件表面的接触状态。本发明可以使得射流水柱与工件表面实现断续脉冲接触，以实现可控制断续磨料水射流，以确保抛光液接触过程实时可控。

摘要：一种基于改进yolov5的智慧工地火灾烟雾预警方法，包括以下步骤：步骤1：利用摄像头获取前端感知视频流数据；步骤2：对得到的视频流数据进行图像预处理以改善图像质量并得到待检测图像集合序列；步骤3：由步骤2获得的图像集合序列得到训练集，验证集，测试集；步骤4：构建基于yolov5的火灾检测模型，然后由步骤3中获取的训练集对所构建的基于yolov5的火灾检测模型进行训练，以获取训练后的基于yolov5的火灾检测模型；步骤5：运用步骤4中训练后的基于yolov5的火灾检测模型对待检测图像集合序列进行检测，获得检测结果；通过以上步骤实现工地火灾烟雾的预警。