{"currentpage":1,"firstResult":0,"maxresult":10,"pagecode":5,"pageindex":{"endPagecode":5,"startPagecode":1},"records":[{"abstractinfo":"针对板带轧机液压AGC系统在线故障诊断问题,建立了一种基于非线性自回归滑动平均模型(NARMA)的<span style=\"color:red\">递归</span><span style=\"color:red\">神经网络</span>,通过AIC定阶法确定模型阶次.运用生产实际数据,通过动态学习算法完成对<span style=\"color:red\">网络</span>的训练,使<span style=\"color:red\">网络</span>映射系统的动力学特性.该<span style=\"color:red\">网络</span>模型避免了故障的自学习,能够很好地实现故障检测.试验研究证明了该<span style=\"color:red\">神经网络</span>方法进行轧机液压AGC系统在线故障诊断的可行性和有效性.","authors":[{"authorName":"董敏","id":"ab6de779-d07f-486a-bbe0-8a9fdf0e2c87","originalAuthorName":"董敏"},{"authorName":"刘才","id":"7062c546-24a3-465a-be08-cd02ef4422fb","originalAuthorName":"刘才"},{"authorName":"李国友","id":"ec3e1b60-bd84-4b60-a972-a0a74e095218","originalAuthorName":"李国友"},{"authorName":"张伟","id":"c9572bed-640f-4d78-bfb4-406438f4b774","originalAuthorName":"张伟"}],"doi":"","fpage":"45","id":"3d393ddc-c68a-4a14-bab0-60551a478651","issue":"5","journal":{"abbrevTitle":"GT","coverImgSrc":"journal/img/cover/GT.jpg","id":"27","issnPpub":"0449-749X","publisherId":"GT","title":"钢铁"},"keywords":[{"id":"5c9f8659-e5c1-4f46-a1d0-1a6ab90b22b3","keyword":"液压AGC","originalKeyword":"液压AGC"},{"id":"8f945f97-eb2f-4854-8c9d-50c6506f552b","keyword":"NARMA","originalKeyword":"NARMA"},{"id":"a185c2db-6af8-44a1-aad3-ee2cafe418ca","keyword":"<span style=\"color:red\">递归</span><span style=\"color:red\">神经网络</span>","originalKeyword":"递归神经网络"},{"id":"8a66f100-3e54-493f-8226-ca00daa636f0","keyword":"故障诊断","originalKeyword":"故障诊断"}],"language":"zh","publisherId":"gt200505012","title":"轧机液压AGC系统基于<span style=\"color:red\">神经网络</span>的传感器故障诊断技术","volume":"40","year":"2005"},{"abstractinfo":"<span style=\"color:red\">神经网络</span>应用于系统建模时要考虑两个关键问题:一是采用的<span style=\"color:red\">神经网络</span>类型;二是当<span style=\"color:red\">神经网络</span>类型确定以后,确定<span style=\"color:red\">网络</span>的输入向量,这两个问题是紧密关联的.相对于BP<span style=\"color:red\">神经网络</span>收敛太慢,具有泛化能力的缺点,DRNN<span style=\"color:red\">网络</span>(对角<span style=\"color:red\">递归</span><span style=\"color:red\">神经网络</span>)能实现动态非线性映射,具有记忆功能,可以追踪模型的变化,具有更好的预测效果.首先用关联规则方法挖掘出一些与质量指标有关的工艺条件,再结合现场工人的实际经验,找到模型的输入输出,再运用DRNN对82B钢进行建模和优化,取得了很好的预测效果.","authors":[{"authorName":"李方方","id":"cf8a1530-aac4-4103-86b6-f2f35b63c312","originalAuthorName":"李方方"},{"authorName":"赵英凯","id":"225e73e1-4937-4fe6-b930-0b4e290662b9","originalAuthorName":"赵英凯"},{"authorName":"俞辉","id":"a799d7e3-21f0-4026-aa1d-61997f18192c","originalAuthorName":"俞辉"}],"doi":"","fpage":"59","id":"d3572a45-6f07-4779-bb83-7d10ae532095","issue":"8","journal":{"abbrevTitle":"GTYJXB","coverImgSrc":"journal/img/cover/GTYJXB.jpg","id":"30","issnPpub":"1001-0963","publisherId":"GTYJXB","title":"钢铁研究学报"},"keywords":[{"id":"6d650ec5-361c-45b3-bc81-e46ca7cfeffe","keyword":"DRNN<span style=\"color:red\">网络</span>","originalKeyword":"DRNN网络"},{"id":"03065db2-e2c8-4b8f-918e-214b378ce4bf","keyword":"关联规则","originalKeyword":"关联规则"},{"id":"f9ec0a67-7b22-4875-8ffc-772dcef489fb","keyword":"预估","originalKeyword":"预估"},{"id":"e6983adc-618d-4f10-984a-30d3a6379b65","keyword":"82B钢","originalKeyword":"82B钢"}],"language":"zh","publisherId":"gtyjxb200708014","title":"对角<span style=\"color:red\">递归</span><span style=\"color:red\">神经网络</span>在82B钢力学性能预估中的应用","volume":"19","year":"2007"},{"abstractinfo":"<span style=\"color:red\">神经网络</span>应用于系统建模时要考虑两个关键问题：一是采用的<span style=\"color:red\">神经网络</span>类型；二是当<span style=\"color:red\">神经网络</span>类型确定以后，确定<span style=\"color:red\">网络</span>的输入向量，这两个问题是紧密关联的。相对于BP<span style=\"color:red\">神经网络</span>收敛太慢，具有泛化能力的缺点，DRNN<span style=\"color:red\">网络</span>(对角<span style=\"color:red\">递归</span><span style=\"color:red\">神经网络</span>)能实现动态非线性映射，具有记忆功能，可以追踪模型的变化，具有更好的预测效果。首先用关联规则方法挖掘出一些与质量指标有关的工艺条件，再结合现场工人的实际经验，找到模型的输入输出，再运用DRNN对82B钢进行建模和优化，取得了很好的预测效果。","authors":[{"authorName":"李方方","id":"775c7a8e-aa62-4897-8fd3-4f7e9dc6e7e8","originalAuthorName":"李方方"},{"authorName":"赵英凯","id":"52fa1a7b-3b77-48e2-af4a-8e21a7af957c","originalAuthorName":"赵英凯"},{"authorName":"俞辉","id":"b5b48cf7-ad3c-4387-8e8e-cc1c1924aae2","originalAuthorName":"俞辉"}],"categoryName":"|","doi":"","fpage":"59","id":"f1cadabe-c802-4bcc-8d63-ec9410c441e1","issue":"8","journal":{"abbrevTitle":"GTYJXB","coverImgSrc":"journal/img/cover/GTYJXB.jpg","id":"30","issnPpub":"1001-0963","publisherId":"GTYJXB","title":"钢铁研究学报"},"keywords":[{"id":"aba90a6d-6c7a-477f-9adf-f8019b73d289","keyword":"DRNN<span style=\"color:red\">网络</span>;关联规则;预估;82B钢","originalKeyword":"DRNN网络;关联规则;预估;82B钢"}],"language":"zh","publisherId":"1001-0963_2007_8_4","title":"对角<span style=\"color:red\">递归</span><span style=\"color:red\">神经网络</span>在82B钢力学性能预估中的应用","volume":"19","year":"2007"},{"abstractinfo":"在实验的基础上建立基于<span style=\"color:red\">神经网络</span>的连铸保护渣性能预测模型,经对比分析表明,用<span style=\"color:red\">神经网络</span>对连铸保护渣性能进行训练和学习后,<span style=\"color:red\">神经网络</span>方法能有效和比较准确地对保护渣性能进行预测.","authors":[{"authorName":"向嵩","id":"58f02dff-c592-4da7-b4c1-09f8f0908b0b","originalAuthorName":"向嵩"},{"authorName":"王雨","id":"43eea525-9488-4f6b-bfd0-2a78c7fb9b4b","originalAuthorName":"王雨"}],"doi":"10.3969/j.issn.1005-4006.2005.05.016","fpage":"43","id":"4989b298-192d-4f2d-92c7-550a0121b193","issue":"5","journal":{"abbrevTitle":"LZ","coverImgSrc":"journal/img/cover/LZ.jpg","id":"52","issnPpub":"1005-4006","publisherId":"LZ","title":"连铸"},"keywords":[{"id":"22766b87-5901-4a63-a0cc-1f74d54f3055","keyword":"","originalKeyword":""}],"language":"zh","publisherId":"lz200505016","title":"利用<span style=\"color:red\">神经网络</span>预测连铸保护渣性能","volume":"","year":"2005"},{"abstractinfo":"简略介绍了人工<span style=\"color:red\">神经网络</span>的基本思想及特点,综述了人工<span style=\"color:red\">神经网络</span>在材料性能预测、工艺参数优化、相变规律的研究、微观组织模拟等领域的应用情况.对人工<span style=\"color:red\">神经网络</span>研究中存在的问题进行了分析,展望了其应用前景.","authors":[{"authorName":"赖静","id":"789d1363-8cb1-457c-81ba-3257d59af78e","originalAuthorName":"赖静"},{"authorName":"王清","id":"2e44c95e-e141-49af-8b16-58af159b2aa5","originalAuthorName":"王清"},{"authorName":"孙东立","id":"99381c44-1bc5-42e7-89b3-757de8f825b7","originalAuthorName":"孙东立"}],"doi":"10.3969/j.issn.1001-4381.2006.z1.121","fpage":"458","id":"7f0a71a9-ab61-4bdb-9299-44d9808aa7e5","issue":"z1","journal":{"abbrevTitle":"CLGC","coverImgSrc":"journal/img/cover/CLGC.jpg","id":"9","issnPpub":"1001-4381","publisherId":"CLGC","title":"材料工程"},"keywords":[{"id":"c117048d-1982-4342-9138-0137871b3a85","keyword":"人工<span style=\"color:red\">神经网络</span>","originalKeyword":"人工神经网络"},{"id":"5626655d-c84d-410a-afae-b0532c5886dc","keyword":"性能","originalKeyword":"性能"},{"id":"a4c88884-ac9a-4a49-a2d7-1cb3e0889ec9","keyword":"工艺","originalKeyword":"工艺"},{"id":"e1723322-039d-427c-a5fb-70f5e105af2e","keyword":"相变","originalKeyword":"相变"},{"id":"46e7c5d0-2f30-4633-8895-20dc3d5801ca","keyword":"组织模拟","originalKeyword":"组织模拟"}],"language":"zh","publisherId":"clgc2006z1121","title":"人工<span style=\"color:red\">神经网络</span>在材料研究中的应用","volume":"","year":"2006"},{"abstractinfo":"<span style=\"color:red\">神经网络</span>是一种旨在模仿人脑结构及其功能的软计算方法.它具有优良的容错性、自组织性和突出的非线性映射能力.最近几年,<span style=\"color:red\">神经网络</span>开始渗透到膜技术领域,为膜过程的研究和应用提供了一种全新的表达工具.目前,<span style=\"color:red\">神经网络</span>涉及到的膜过程包括MF、UF、NF以及RO等.在简要介绍了<span style=\"color:red\">神经网络</span>的结构与理论的基础上,分别评述了<span style=\"color:red\">神经网络</span>在这些膜过程中应用的研究现状,并展望了今后的研究方向.","authors":[{"authorName":"熊日华","id":"3472b7b5-bea8-496e-9570-4de8887c4792","originalAuthorName":"熊日华"},{"authorName":"王世昌","id":"8ef0ea5c-3faf-494a-8fb2-62d3a50a231e","originalAuthorName":"王世昌"}],"doi":"10.3969/j.issn.1007-8924.2003.06.011","fpage":"44","id":"aa1e482d-25ea-4682-8cd2-bcc687696405","issue":"6","journal":{"abbrevTitle":"MKXYJS","coverImgSrc":"journal/img/cover/MKXYJS.jpg","id":"54","issnPpub":"1007-8924","publisherId":"MKXYJS","title":"膜科学与技术   "},"keywords":[{"id":"41a584ce-49dc-40e2-aa00-9236a43b324d","keyword":"<span style=\"color:red\">神经网络</span>","originalKeyword":"神经网络"},{"id":"d88fc47b-6712-449c-ba00-4622761f7c45","keyword":"微滤","originalKeyword":"微滤"},{"id":"624a077f-dd46-4e67-bd02-9c9bd226286d","keyword":"超滤","originalKeyword":"超滤"},{"id":"b5e05f95-87ba-4d9a-b4e2-1a7350ae72b6","keyword":"纳滤","originalKeyword":"纳滤"},{"id":"96b26f2d-cdb5-4258-9c06-e0ee97cce728","keyword":"反渗透","originalKeyword":"反渗透"}],"language":"zh","publisherId":"mkxyjs200306011","title":"<span style=\"color:red\">神经网络</span>在膜技术中的应用","volume":"23","year":"2003"},{"abstractinfo":"轧制过程中，针对4200轧机在轧件宽展变化自动预测和控制，分析了轧制过程中宽展变化的影响因素。在<span style=\"color:red\">神经网络</span>技术和现场实测数据的基础上，利用Matlab人工<span style=\"color:red\">神经网络</span>工具箱，应用GRNN广义回归<span style=\"color:red\">神经网络</span>建立宽展变化预测模型来提高轧制宽展变化预测的精度。结果表明，该方法建立的模型可以实现对宽展变化的预测，其预测精度有较大提高。","authors":[{"authorName":"孟令启","id":"0df6bc79-b808-45a1-813c-06824e975af3","originalAuthorName":"孟令启"},{"authorName":"孟梦","id":"7c6ec060-21a0-49d6-b8ae-f6e1d5a9a9f5","originalAuthorName":"孟梦"}],"categoryName":"|","doi":"","fpage":"23","id":"493171aa-6021-4a61-ba7d-c2630f2c623b","issue":"3","journal":{"abbrevTitle":"GTYJXB","coverImgSrc":"journal/img/cover/GTYJXB.jpg","id":"30","issnPpub":"1001-0963","publisherId":"GTYJXB","title":"钢铁研究学报"},"keywords":[{"id":"8d329922-578e-41f6-9d9b-8495f3e98789","keyword":"中厚板轧机;宽展;GRNN<span style=\"color:red\">神经网络</span>","originalKeyword":"中厚板轧机;宽展;GRNN神经网络"}],"language":"zh","publisherId":"1001-0963_2008_3_1","title":"基于GRNN<span style=\"color:red\">神经网络</span>的4200轧机宽展模型","volume":"20","year":"2008"},{"abstractinfo":"轧制过程中,针对4200轧机在轧件宽展变化自动预测和控制,分析了轧制过程中宽展变化的影响因素.在<span style=\"color:red\">神经网络</span>技术和现场实测数据的基础上,利用Matlab人工<span style=\"color:red\">神经网络</span>工具箱,应用GRNN广义回归<span style=\"color:red\">神经网络</span>建立宽展变化预测模型来提高轧制宽展变化预测的精度.结果表明,该方法建立的模型可以实现对宽展变化的预测,其预测精度有较大提高.","authors":[{"authorName":"孟令启","id":"1646277a-165a-43e6-a686-c5f2801532f6","originalAuthorName":"孟令启"},{"authorName":"孟梦","id":"6b81b794-74a8-446a-a6f3-4344fb8b50f5","originalAuthorName":"孟梦"}],"doi":"","fpage":"23","id":"6f67305b-2592-4089-848e-6e2f3346da50","issue":"3","journal":{"abbrevTitle":"GTYJXB","coverImgSrc":"journal/img/cover/GTYJXB.jpg","id":"30","issnPpub":"1001-0963","publisherId":"GTYJXB","title":"钢铁研究学报"},"keywords":[{"id":"a5ee9aec-dfbf-47e5-90e2-d0ca7e30b196","keyword":"中厚板轧机","originalKeyword":"中厚板轧机"},{"id":"4b9c5263-90bf-4a4a-8ab0-9e586cec6a05","keyword":"宽展","originalKeyword":"宽展"},{"id":"b8ee4990-0dd4-489f-ab28-33043c3fd9aa","keyword":"GRNN<span style=\"color:red\">神经网络</span>","originalKeyword":"GRNN神经网络"}],"language":"zh","publisherId":"gtyjxb200803006","title":"基于GRNN<span style=\"color:red\">神经网络</span>的4200轧机宽展模型","volume":"20","year":"2008"},{"abstractinfo":"针对板带材轧制过程是一个复杂的非线性过程、难以建立精确数学模型的特点，提出了厚度自动控制<span style=\"color:red\">神经网络</span>模糊智能方法，设计了输入为“编码” 的<span style=\"color:red\">神经网络</span>模糊控制器。通过仿真证明了<span style=\"color:red\">神经网络</span>模糊控制的可行性，其控制精度优于常规方法。","authors":[{"authorName":"贾春玉","id":"ce3edd3c-8ffc-4a20-89d9-193f824f8df2","originalAuthorName":"贾春玉"}],"doi":"","fpage":"50","id":"8c2310d4-1c78-40a3-897f-78512b652db7","issue":"2","journal":{"abbrevTitle":"GTYJXB","coverImgSrc":"journal/img/cover/GTYJXB.jpg","id":"30","issnPpub":"1001-0963","publisherId":"GTYJXB","title":"钢铁研究学报"},"keywords":[{"id":"6dd019e5-6ca1-4a98-a467-b955d26ba90e","keyword":"<span style=\"color:red\">神经网络</span>模糊推理","originalKeyword":"神经网络模糊推理"},{"id":"a5db2e63-50a7-4296-9bf6-e9c92fde8c14","keyword":"编码","originalKeyword":"编码"},{"id":"3443e460-2443-4e97-a284-54d67e032193","keyword":"热轧带钢","originalKeyword":"热轧带钢"},{"id":"34491d80-fc87-40b3-936e-4f4ceff0092c","keyword":"厚度控制","originalKeyword":"厚度控制"}],"language":"zh","publisherId":"gtyjxb200102011","title":"基于<span style=\"color:red\">神经网络</span>模糊推理的智能厚度控制","volume":"13","year":"2001"},{"abstractinfo":"针对选择性激光烧结成型件变形大、精度较低的问题,将<span style=\"color:red\">神经网络</span>方法应用于选择性激光烧结(SLS)加工工艺的研究.根据SLS加工工艺的特点,研究的工艺参数包括:层厚、扫描间距、激光功率、扫描速度、环境温度、层与层之间的加工时间间隔和扫描方式.建立了SLS加工工艺参数与加工变形、收缩率之间的<span style=\"color:red\">神经网络</span>预测模型.实验结果与<span style=\"color:red\">神经网络</span>模型计算结果十分吻合,说明该<span style=\"color:red\">神经网络</span>模型能定量地反映出工艺参数与加工材料变形、收缩率之间的关系.","authors":[{"authorName":"王荣吉","id":"851b45da-4aeb-4daa-9dbe-ea4abef190c2","originalAuthorName":"王荣吉"},{"authorName":"王玲玲","id":"1221054f-80bd-47b9-89ee-b2de641d0d6e","originalAuthorName":"王玲玲"},{"authorName":"赵立华","id":"fffed2b0-5113-46dd-8486-0dd3249c4e7d","originalAuthorName":"赵立华"}],"doi":"","fpage":"452","id":"46c04892-998f-4fba-9988-5ee3c7bafebe","issue":"3","journal":{"abbrevTitle":"ZGYSJSXB","coverImgSrc":"journal/img/cover/ZGYSJSXB.jpg","id":"88","issnPpub":"1004-0609","publisherId":"ZGYSJSXB","title":"中国有色金属学报"},"keywords":[{"id":"2f532bc8-7210-4e3e-b0c7-31d5ef3b8e3e","keyword":"快速成型","originalKeyword":"快速成型"},{"id":"7ab453db-c6a1-4ff1-b0a3-06707be1c417","keyword":"选择性激光烧结","originalKeyword":"选择性激光烧结"},{"id":"a66e5985-4687-497d-b0a4-da29f1af7150","keyword":"工艺参数","originalKeyword":"工艺参数"},{"id":"1fa1f9e4-0797-48fb-8aa8-48ec697438ff","keyword":"<span style=\"color:red\">神经网络</span>","originalKeyword":"神经网络"}],"language":"zh","publisherId":"zgysjsxb200503022","title":"基于<span style=\"color:red\">神经网络</span>的快速成型工艺","volume":"15","year":"2005"}],"totalpage":325,"totalrecord":3248}