摘要：【中文】本发明公开了一种能够避免堵塞的农业用稻草粉碎装置，包括外壳、储料箱、电机、第一搅拌齿轮、通风道和透风板，所述外壳的右侧表面固定连接有储料箱，且储料箱的上表面连接有盖板，并且外壳的内部分别开设有内腔和输送道，所述外壳的内部固定设置有电机，且电机的输出端安装有圆柱块，所述圆柱块的外表面固定连接有滚筒，且滚筒的外表面连接有传输带。该能够避免堵塞的农业用稻草粉碎装置，该粉碎装置可通过第一锥形齿轮与第二锥形齿轮的啮合，使得支撑柱带动螺旋状的搅拌机进行旋转，从而通过叶轮螺旋旋转的原理将物料均匀的向前推移，从而物料则会被均匀的排入到出料口，且通过搅拌桨的转动，能够有效的避免堵塞。 【EN】The invention discloses an agricultural straw smashing device capable of avoiding blockage, which comprises a shell, a storage box, a motor, a first stirring gear, a ventilation channel and a ventilation plate, wherein the right side surface of the shell is fixedly connected with the storage box, the upper surface of the storage box is connected with a cover plate, an inner cavity and a conveying channel are respectively arranged inside the shell, the motor is fixedly arranged inside the shell, a cylindrical block is arranged at the output end of the motor, a roller is fixedly connected to the outer surface of the cylindrical block, and a conveying belt is connected to the outer surface of the roller. This can avoid straw reducing mechanism for agricultural of jam, the meshing of the first conical gear of this reducing mechanism accessible and second conical gear for the support column drives spiral helicine mixer and rotates, thereby passes the even forward of material through the principle of impeller spiral rotation, thereby the material then can be by even discharge gate of going into, and the rotation through the stirring rake, can effectually avoid blockking up.

摘要：【中文】本发明公开了一种可对外壳进行回收的农业生产用剥豆装置，包括支撑架、主壳体、进料口、豆粒收集盒和外壳收集盒，所述支撑架的上方设置有主壳体，且主壳体的上方连接有进料口，并且支撑架的下方设置有固定板，所述固定板的下方设置有豆粒收集盒，且豆粒收集盒的左侧设置有外壳收集盒，所述主壳体的内部贯穿有传动轴，且传动轴的外侧固定有剥豆辊，并且剥豆辊位于主壳体的内部，所述挡板的前侧设置有滑槽，且滑槽开设在主壳体的内壁上。该可对外壳进行回收的农业生产用剥豆装置设置有防护板，在加料时物料可推动防护板下落至主壳体内，同时防护板可在扭力弹簧的作用下进行复位，对进料口进行遮挡，避免豆粒飞溅到外界，增加装置的功能性。 【EN】The invention discloses a bean peeling device for agricultural production, which can recover hulls and comprises a support frame, a main shell, a feed inlet, a bean particle collecting box and a shell collecting box, wherein the main shell is arranged above the support frame, the feed inlet is connected above the main shell, a fixing plate is arranged below the support frame, the bean particle collecting box is arranged below the fixing plate, the shell collecting box is arranged on the left side of the bean particle collecting box, a transmission shaft penetrates through the interior of the main shell, a bean peeling roller is fixed on the outer side of the transmission shaft and is positioned in the main shell, a sliding groove is arranged on the front side of the baffle, and the sliding groove is formed in the inner wall of the main shell. This agricultural production that can retrieve to hull is with shelling beans device is provided with the guard plate, and the material can promote the guard plate and fall to main casing internal when reinforced, and the guard plate can reset under torque spring's effect simultaneously, shelters from the feed inlet, avoids the beans grain to splash to the external world, increases the functional of device.

摘要：【中文】本发明公开了一种投料速度可控的农业用种子分级装置，包括装置主体外壳、内腔、加热芯、电机和按压杆，所述装置主体外壳的上表面开设有入料口，且装置主体外壳的内部分别开设有内腔、第一分级腔、储料箱、第二分级腔和第三分级腔，所述装置主体外壳的外表面开设有入风口。该投料速度可控的农业用种子分级装置，该分级装置通过具有多个筛孔的筛布的震动，能够使得种子被多个筛孔进行多个大小的分级，且在螺旋桨对种子运输的过程中能够持续进行筛选，提高分级效率，此时可通过控制电机的转速，来达到控制筛布震动的频率，筛布快速震动时则会快速对种子进行筛选投料，筛布慢速震动时则会慢速对种子进行筛选投料。 【EN】The invention discloses an agricultural seed grading device with controllable feeding speed, which comprises a device main body shell, an inner cavity, a heating core, a motor and a pressing rod, wherein a feeding port is formed in the upper surface of the device main body shell, the inner cavity, a first grading cavity, a material storage box, a second grading cavity and a third grading cavity are formed in the device main body shell respectively, and an air inlet is formed in the outer surface of the device main body shell. This throw controllable agricultural seed grading plant of material speed, this grading plant is through the vibrations of the sieve cloth that has a plurality of sieve meshes, can be so that the seed carries out the classification of a plurality of sizes by a plurality of sieve meshes, and can continuously filter at the screw to the in-process of seed transportation, improve classification efficiency, the rotational speed of accessible control motor this moment, reach the frequency of control sieve cloth vibrations, then can filter the seed fast when sieve cloth shakes fast and throw the material, then can filter the seed at a slow speed when sieve cloth shakes slowly and throw the material.

摘要：【中文】本发明公开了一种多智能体系统的量化控制方法，所述多智能体系统包括一个领导者和N个跟随者；所述控制方法包括以下步骤：建立多智能体系统的系统模型，包括跟随者的系统模型以及领导者的系统模型；建立多智能体系统的误差系统，包括第i个跟随者与领导者输出之间的误差，以及第i个跟随者系统的q阶状态与虚拟控制器之间的误差；设计多智能体系统的控制器，包括为第i个跟随者设计第一个虚拟控制器至第q个虚拟控制器，然后为第i个跟随者设计控制输入。本发明解决了现有技术中不能保证在控制器与智能体之间带宽受限情况下的系统稳定性的问题。 【EN】The invention discloses a quantitative control method of a multi-agent system, wherein the multi-agent system comprises a leader and N followers; the control method comprises the following steps: establishing a system model of the multi-agent system, which comprises a system model of a follower and a system model of a leader; establishing an error system of the multi-agent system, wherein the error system comprises an error between the output of the ith follower and the leader and an error between the q-order state of the ith follower system and the virtual controller; a controller for a multi-agent system is designed, including designing first through qth virtual controllers for an ith follower, and then designing control inputs for the ith follower. The invention solves the problem that the system stability under the condition of limited bandwidth between the controller and the intelligent agent cannot be ensured in the prior art.

摘要：【中文】本发明公开了一种相似药品推荐方法及系统、服务器及介质，该方法包括：获取药企在设定时间段的销售数据，获取每一种药品的用户列表集合；采用相似算法评价药品与参考药品之间的相似度，获取相似度排名前第一设定阈值的药品；采用条件概率算法评价药品与参考药品之间的相似度，获取相似度排名前第二设定阈值的药品；分别给予相似度以设定的权重，选取相似度排名前第三设定阈值的药品推荐。本发明采用现有的销售数据作为分析基础，通过相似算法和条件概率算法评价销售数据中药品的相似度，且通过权重融合以调整药品的相似度，进而判断与参考药品更为相似的药品，其可提高相似药品的评价准确性，并满足相似药品的数量需求，提高药企的销售利润。 【EN】The invention discloses a similar medicine recommending method and system, a server and a medium, wherein the method comprises the following steps: acquiring sales data of a medicine enterprise in a set time period, and acquiring a user list set of each medicine; evaluating the similarity between the medicine and a reference medicine by adopting a similarity algorithm, and acquiring the medicine with a first set threshold before the similarity ranking; evaluating the similarity between the medicine and a reference medicine by adopting a conditional probability algorithm, and acquiring the medicine with a second set threshold before the similarity ranking; and respectively giving the similarity with set weight, and selecting the medicine recommendation with the third set threshold before the similarity ranking. The method adopts the existing sales data as an analysis basis, evaluates the similarity of the medicines in the sales data through a similarity algorithm and a conditional probability algorithm, adjusts the similarity of the medicines through weight fusion, further judges the medicines more similar to the reference medicines, can improve the evaluation accuracy of the similar medicines, meets the quantity requirements of the similar medicines, and improves the sales profits of medicine enterprises.

摘要：【中文】本发明公开了一种药品的知识图谱构建方法及系统、服务器及介质，该方法包括：获取药品的属性数据；根据药品的属性数据获取药品之间的相互作用，并构建知识图谱。本发明提供的药品的知识图谱构建方法及系统、服务器及介质，其通过获取药品的属性数据，并根据属性数据得到药品之间的相互作用，当患者购买药品时可通过药品之间的相互作用所构建的知识图谱知晓该种药品的搭配用药、禁忌用药关系，其降低了患者购药不慎的危险性，并降低了药店的成本。 【EN】The invention discloses a method and a system for establishing a knowledge graph of a medicine, a server and a medium, wherein the method comprises the following steps: acquiring attribute data of the medicine; and acquiring the interaction among the medicines according to the attribute data of the medicines, and constructing a knowledge graph. According to the method and the system for establishing the knowledge graph of the medicine, the server and the medium, the attribute data of the medicine is obtained, the interaction among the medicines is obtained according to the attribute data, and when a patient purchases the medicine, the relationship between matched medicines and contraindicated medicines of the medicine can be known through the knowledge graph established by the interaction among the medicines, so that the risk of the patient who carelessly purchases the medicine is reduced, and the cost of a pharmacy is reduced.

摘要：【中文】本发明公开了一种基于大数据的药品通用名清洗方法及系统、服务器及介质，该方法包括构建标准通用名数据库：利用网络爬虫技术爬取药品的通用名数据，获取已清洗过的药品的通用名数据，将通用名数据合并排重、清洗以获得标准通用名数据，导入标准通用名数据库；对每一个新增的药企数据进行通用名清洗：将药企新增的通用名与标准通用名数据进行匹配，若满足匹配要求则入相应的数据库，并更新标准通用名数据库。本发明通过构建标准通用名数据库，并将药企新增的通用名与标准通用名数据库中的标准通用名数据进行匹配，以实现自动化的对每一个新增的药企数据中的通用名进行清洗，其在保证准确率的基础上，减少了人力投入，降低了药品通用名清洗成本。 【EN】The invention discloses a big data-based medicine universal name cleaning method and system, a server and a medium, wherein the method comprises the following steps of constructing a standard universal name database: the universal name data of the medicines are crawled by utilizing a web crawler technology, the cleaned universal name data of the medicines are obtained, the universal name data are combined, duplicated and cleaned to obtain standard universal name data, and the standard universal name data are imported into a standard universal name database; and (3) carrying out universal name cleaning on each newly added medicine enterprise data: and matching the newly added universal name of the medicine enterprise with the standard universal name data, if the matching requirement is met, entering the corresponding database, and updating the standard universal name database. According to the invention, the standard universal name database is established, and the newly-added universal names of the medicine enterprises are matched with the standard universal name data in the standard universal name database, so that the automatic cleaning of the universal names in each newly-added medicine enterprise data is realized, and on the basis of ensuring the accuracy, the manpower input is reduced, and the cleaning cost of the medicine universal names is reduced.

摘要：【中文】本发明公开了一种基于catboost算法与深度学习结合的排序推荐方法及系统、服务器及介质，该方法包括：定位候选集药品数据，获取离散特征和连续特征；构造新的组合特征，将组合特征作为模型的输入层；将离散特征向量转化得到castboost_vector,将组合特征和连续特征转化得到64维DNN特征向量，将两者特征向量合并一起；构建模型进行训练，全连接层的输出结果对药品进行排序推荐。本发明采用catboost算法与深度学习结合的方式，不仅保留了传统的记忆功能，还增加了泛化的能力，利用用户的标签数据和药品的特征数据定位候选集药品数据对特征召回，其降低了客户的等待时间，也利于提高点击率。 【EN】The invention discloses a sorting recommendation method and system, a server and a medium based on the combination of a catboost algorithm and deep learning, wherein the method comprises the following steps: positioning medicine data of the candidate set to obtain discrete characteristics and continuous characteristics; constructing new combination characteristics, and using the combination characteristics as an input layer of the model; converting the discrete feature vectors to obtain a castboost _ vector, converting the combined features and the continuous features to obtain 64-dimensional DNN feature vectors, and combining the feature vectors; and constructing a model for training, and sequencing and recommending the medicines according to the output result of the full connection layer. The method adopts the mode of combining the catboost algorithm and the deep learning, not only retains the traditional memory function, but also increases the generalization capability, positions the drug data of the candidate set to recall the features by utilizing the label data of the user and the feature data of the drug, reduces the waiting time of the client and is beneficial to improving the click rate.

摘要：【中文】本发明公开了一种基于大数据的睡眠会员唤醒方法及系统、服务器及介质，该方法包括：获取睡眠会员的用户属性标签数据及用户行为标签数据；采用多种不同算法对睡眠会员的用户属性标签数据和用户行为标签数据进行多种交叉验证，以得到不同算法对应的多个唤醒概率；对多个唤醒概率按权重融合并得到最终唤醒概率，若最终唤醒概率大于设定阈值，则向该睡眠会员发出营销信息。本发明采用多种算法以对睡眠会员的用户属性标签数据及用户行为标签数据进行交叉验证，以判断睡眠会员能够被唤醒的概率，并根据不同权重下的唤醒概率以修正最终唤醒概率，其针对唤醒概率高的睡眠会员进行营销，其极大的减少了营销成本，也提高了所有睡眠会员的唤醒体验效果。 【EN】The invention discloses a big data-based sleep member awakening method and system, a server and a medium, wherein the method comprises the following steps: acquiring user attribute label data and user behavior label data of the sleeping member; performing multiple cross validation on the user attribute tag data and the user behavior tag data of the sleeping member by adopting multiple different algorithms to obtain multiple awakening probabilities corresponding to the different algorithms; and fusing the plurality of awakening probabilities according to the weight to obtain a final awakening probability, and if the final awakening probability is greater than a set threshold value, sending marketing information to the sleeping member. The method adopts multiple algorithms to perform cross validation on the user attribute tag data and the user behavior tag data of the sleeping members so as to judge the probability that the sleeping members can be awakened, corrects the final awakening probability according to the awakening probabilities under different weights, and performs marketing aiming at the sleeping members with high awakening probability, so that the marketing cost is greatly reduced, and the awakening experience effect of all the sleeping members is also improved.

摘要：【中文】本发明公开了一种环绕式共轨直列变速箱，包括有变速箱和换挡调节箱，动力输入轴件，包括有动力输入轴及位于所述变速箱壳体内的副轴，且副轴其一端与所述换挡调节箱转动连接，动力输出轴件，倒挡轮组，组合旋转块，档位限位块，本发明结构简单、传动效率高、换挡简单，兼具手动斜齿变速箱和直齿序列变速箱以上两种变速箱的优点。 【EN】The invention discloses a surrounding common rail in-line gearbox, which comprises a gearbox and a gear shifting adjusting box, wherein a power input shaft part comprises a power input shaft and an auxiliary shaft positioned in a gearbox shell, one end of the auxiliary shaft is rotationally connected with the gear shifting adjusting box, a power output shaft part, a reverse gear wheel set, a combined rotating block and a gear limiting block.