WO2022007380A1 - 面向"互联网+"环境机械产品个性化设计模式匹配方法 - Google Patents

面向"互联网+"环境机械产品个性化设计模式匹配方法 Download PDF

Info

Publication number
WO2022007380A1
WO2022007380A1 PCT/CN2021/070982 CN2021070982W WO2022007380A1 WO 2022007380 A1 WO2022007380 A1 WO 2022007380A1 CN 2021070982 W CN2021070982 W CN 2021070982W WO 2022007380 A1 WO2022007380 A1 WO 2022007380A1
Authority
WO
WIPO (PCT)
Prior art keywords
design
order
module
design pattern
pattern matching
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/070982
Other languages
English (en)
French (fr)
Inventor
张树有
裘乐淼
王自立
顾叶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University
Original Assignee
Zhejiang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University filed Critical Zhejiang University
Priority to JP2021544343A priority Critical patent/JP7171094B2/ja
Priority to US17/475,288 priority patent/US20220004677A1/en
Publication of WO2022007380A1 publication Critical patent/WO2022007380A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0621—Electronic shopping [e-shopping] by configuring or customising goods or services
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the invention belongs to the field of customized design of mechanical products, and in particular relates to a matching method for the personalized design mode of mechanical products in an "Internet +" environment.
  • the traditional product customization design mode and mass production line no longer satisfy the design and production of customized products.
  • the mass customization design mode configures product modules according to user needs by presetting product modules and module families, and combines products that meet user needs.
  • the design mode of mass customization will also be difficult to meet the needs of users.
  • the existing product modules and module families cannot fully cover the individual needs of users, and the future demand trends cannot be accurately predicted, so the module library cannot be updated in advance.
  • the design mode of mass customization cannot be adapted.
  • the purpose of the present invention is to provide a pattern matching method for the personalized design of mechanical products facing the "Internet +" environment, aiming at the deficiencies of the prior art.
  • a kind of oriented "Internet +" environment mechanical product personalized design pattern matching method comprises the following steps:
  • req i denotes the i th demand characteristics
  • r i represents the normalized value of the i th needs demand characteristics
  • n is an number of characteristic needs
  • M represents the number of historical orders in the design pattern matching case library
  • order j represents the order feature vector of the jth order
  • p kj represents the design pattern adopted by the kth module in the jth order
  • pattern j represents the jth order
  • the matching result of each module design pattern in the order 1 ⁇ k ⁇ m
  • 0 ⁇ P(order * ) ⁇ 1 is a constant
  • P( ⁇ k ) represents the probability that the kth module in the design pattern matching case library X adopts the ⁇ k design pattern:
  • X( ⁇ k ) represents the order set of the kth module in X using the ⁇ k design pattern
  • is the number of elements in X( ⁇ k );
  • ⁇ k ) Conditional probabilities of choosing different design patterns for the kth module according to the new order eigenvector order *:
  • r i * represents the i-th order * normalized value of the demand characteristics of the demand, 1 ⁇ i ⁇ n;
  • X ( ⁇ k) i is the X k th [epsilon] k module is designed i-th order mode a normalized value needs demand characteristics of a set of r i, are the mean and variance of the set X( ⁇ k ) i, respectively;
  • the method of the present invention studies the individualized design pattern matching method of each design module in the customized product design according to Bayes' theorem from the perspective of probability. By calculating the user satisfaction probability values of different design modules using different design modes, the corresponding design modes of each design module are matched, which reduces the trial and error situation when designers select design modes based on design experience, and improves design efficiency.
  • the present invention improves the intelligence and operability in the matching process of individualized design patterns.
  • Figure 1 is the flow chart of personalized design of "Internet +" environmental mechanical products
  • Figure 2 is a flow chart of the personalized design pattern matching of "Internet +" environmental mechanical products
  • Figure 3 is a process diagram of the implementation process of personalized design of the elevator car system in the "Internet +" environment.
  • the Internet + design platform converts user needs into product orders, and transmits them to designers through a web server.
  • the designer Based on the electronic order, the designer matches the design mode of each module of the product according to the user satisfaction probability matching method of the personalized design mode.
  • the "Internet +" personalized design mode of environmental mechanical products includes: configuration-by-order design mode, deformed-by-order design mode, and generation-by-order design mode, and the steps are as follows:
  • the complex products are decomposed according to the design modules, and the modules are designed one by one.
  • the design pattern of each module is obtained by matching the order content with the product module library.
  • the Internet + design platform is a design platform provided for the personalized design of "Internet +" environmental mechanical products. Users and designers jointly participate in product customization design. Generally, users put forward or modify needs, and designers transform according to needs. Orders are custom designed, users participate in the whole process of design, and real-time feedback with designers to complete products that meet user needs. Specifically, a method for matching patterns of individual design patterns of mechanical products oriented to the "Internet+" environment of the present invention includes the following steps:
  • the customized design process of mechanical products is completed on the Internet + design platform.
  • design modes namely, the configuration-to-order design mode, the deformation-to-order design mode, and the generation-to-order design mode.
  • the designer According to the order content and the existing product module matching the design mode, the designer will design the product module one by one according to the matching result and complete the design.
  • design parameters, configuration structure, configuration rules and configuration modules match suitable modules from the configuration module library to form a customized design scheme for mechanical products configured according to the order;
  • the design scheme is transmitted to the user, and it is evaluated whether the scheme meets its own needs. If so, output the customized design scheme of mechanical products to complete the design. If not, return to step 2 to reconfigure, or replace the other two design modes.
  • the order deformation design mode can be used
  • the designer first judges whether the new module meets the performance. If it does not meet the requirements, it needs to optimize the structure or go to step 5-6. If it meets the requirements, then feedback to the user and repeat steps 4-6 to fully meet the user's needs.
  • the design scheme needs to be modified for real-time feedback evaluation by users, and there is no product module that meets the design requirements in the module library, the generation-to-order design mode can be used. design the module;
  • the designer selects the module that meets the product performance from the massive generative design modules, transmits it to the user for evaluation, and collaboratively selects the module that fully meets the user's needs; if the user changes the needs during this process, go to step 4;
  • Generative modules model, document, structure, and rules, and integrate into the product design resource library.
  • FIG. 2 it is a flow chart of the user satisfaction probability matching method of the personalized design mode in the present invention.
  • the described matching design mode according to the order content and the existing product module adopts the user satisfaction probability matching method of the personalized design mode, and the steps of the method are as follows:
  • the design patterns adopted by the k modules d k , and pattern j represents the set of design pattern matching results p kj of each module in the jth order record.
  • r i * represents the normalized user personalized demand value of the i-th demand feature req i * in order * . Due to the continuity of r i * , it satisfies the distribution law of the probability density function,
  • ⁇ k ) is jointly determined by n r i values in order * , based on the assumption of attribute conditional independence.
  • 0 ⁇ P(order * ) ⁇ 1 represents the evidence factor, which is a constant; therefore, comparing the size of P( ⁇ k
  • ⁇ k 1, 2, 3 ⁇ as the matching result
  • the designer completes the customized design scheme of mechanical products, and directly feedbacks it to the user on the Internet + design platform.
  • the user can experience the product performance and determine whether it meets the needs. If it is not satisfied, it will be fed back to the designer in real time to redesign the part that does not meet the requirements; if it is satisfied, it is determined that the product design plan will be transferred to the manufacturing stage.
  • the present invention carries out the customized design of mechanical products based on user orders, drives the design based on the needs of a single user, and the user participates in the entire design process through the Internet + design platform, and communicates with designers in real time, so that the final design scheme meets the personalized needs of users.
  • the three personalized design modes of mechanical products proposed by the present invention configuration design mode according to order, deformation design mode according to order, and generative design mode according to order, which meet the customized design requirements of almost all mechanical products.
  • the three design patterns can effectively alleviate the contradiction between demand individuation and production scale.
  • the design pattern matching stage eliminates the need for designers to judge the modular design pattern that meets the order requirements based on experience, improving design efficiency and user satisfaction with design results.
  • elevators are widely used in life. There are great differences in elevator systems in different buildings, so users' needs for elevators are also personalized.
  • the elevator is divided into a driving system, a suspension system, a landing door system, a car system and other systems, each of which can be independently designed and assembled.
  • order * )( ⁇ 24 1, 2, 3) to determine which design pattern to choose.
  • the specific calculation steps are as follows:
  • FIG. 3 it is a process diagram of the implementation process of the personalized design of the elevator car system in the corresponding "Internet +" environment of the present invention.
  • the personalized demand put forward by the user in the "Internet +" environment is transformed into a new electronic order, and the order feature vector order* of the user's personalized demand generated by the electronic order is based on the personalized design pattern matching method proposed by the present invention.
  • the user satisfaction probabilities of the three design modes are calculated one by one, and the design mode matching is carried out.
  • the results show that 91% of the modules pass the configuration-to-order design mode, 7% of the modules pass the modification-to-order design mode, and 2% of the modules pass the design mode.
  • the build-to-order design pattern is shown in Figure 3(a).
  • the design modules such as car frame, car, protective cover and car wall can all adopt the configuration-to-order design mode.
  • the existing configuration rule library configure the module library and the configuration structure of the elevator car system, and match the components that meet the order requirements from the module library from top to bottom, and form the specific design scheme of each module configured according to the order, as shown in Figure 3 (b).
  • Figures 3(b)-(d) all take a certain component as an example, and the other components are designed one by one according to the matching results of the design mode. Finally, the designers integrate each module, and users and designers design on the Internet + Real-time mutual feedback is carried out on the platform, and designers revise or redesign the design results until they fully meet user needs.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种面向"互联网+"环境机械产品个性化设计模式匹配方法,该方法将用户订单量化为特征向量,将机械产品分解成模块,并构建历史案例库,根据机械产品各模块采用不同设计模式时用户满意的概率得到根据用户订单匹配的设计模式方案。该方法从概率学角度出发,依据贝叶斯定理研究产品定制设计中各个设计模块的个性化设计模式匹配方法;通过计算不同设计模块采用不同设计模式的用户满意概率值,匹配出各个设计模块对应的设计模式,减少设计人员靠设计经验选择设计模式时的试错情况,提高设计效率;提高了个性化设计模式匹配过程中的智能性与操作性。

Description

面向“互联网+”环境机械产品个性化设计模式匹配方法 技术领域
本发明属于机械产品定制设计领域,尤其涉及一种面向“互联网+”环境机械产品个性化设计模式的匹配方法。
背景技术
专家预测,未来一半以上的产品将是定制个性化产品。随着工业技术与人民生活水平的不断提升,物质需要愈加广泛,传统的单一产品不再满足人们个性化的需求。传统的产品定制设计模式与批量生产流水线不再满足定制产品的设计与生产。大规模定制的设计模式通过预设产品模块与模块族,根据用户需求进行产品模块配置,组合出满足用户需求的产品。但是随着需求的个性化不断增多,产品订单的来源从群体慢慢转为单个用户,大规模定制的设计模式也将会难以满足用户的需求。已有的产品模块与模块族不能完全涵盖用户的个性化需求,且对未来的需求趋势还未能做到精确的预测导致模块库不能提前更新。同时,对于复杂定制机械产品或装备,大规模定制的设计模式还无法适应。
发明内容
本发明的目的在于针对现有技术的不足,提供一种面向“互联网+”环境机械产品个性化设计模式匹配方法。
本发明的目的是通过以下技术方案来实现的:一种面向“互联网+”环境机械产品个性化设计模式匹配方法,包括以下步骤:
(1)构建用户个性化需求的订单特征向量order:
order={(req 1,r 1),(req 2,r 2),...,(req i,r i),...,(req n,r n)}
其中,req i表示第i个需求特征,r i表示第i个需求特征的归一化需求值,n为需求特征数;
(2)将机械产品分解成m个模块d 1~d m,构建产品分解模块集D={d 1,d 2,...,d m};
(3)根据满足用户需求的设计模式方案的历史订单记录,构建设计模式匹配案例库X:
Figure PCTCN2021070982-appb-000001
pattern j={p 1j,p 2j,...,p kj,...,p mj}
其中,M表示设计模式匹配案例库中历史订单的数量,order j表示第j个订单的订单特征向量;p kj表示第j个订单中第k个模块采用的设计模式,pattern j表示第j个订单中各个模块设计模 式匹配结果,1≤k≤m;
(4)针对新的订单特征向量order *,机械产品第k个模块采用不同设计模式时用户满意概率P(ε k|order *)(ε k=1,2,3)为:
Figure PCTCN2021070982-appb-000002
其中,ε k=1表示第k个模块采用按订单配置设计模式,ε k=2表示第k个模块采用按订单变形设计模式,ε k=3表示第k个模块采用按订单生成设计模式;0<P(order *)≤1为常数;P(ε k)表示设计模式匹配案例库X中第k个模块采用ε k设计模式的概率:
Figure PCTCN2021070982-appb-000003
其中,X(ε k)表示X中第k个模块采用ε k设计模式的订单集合,|X(ε k)|为X(ε k)中的元素个数;P(order *|ε k)为第k个模块根据新的订单特征向量order *选择不同设计模式的条件概率:
Figure PCTCN2021070982-appb-000004
其中,r i *表示order *中第i个需求特征的归一化需求值,1≤i≤n;X(ε k) i是X中第k个模块采用ε k设计模式的订单中第i个需求特征的归一化需求值r i的集合,
Figure PCTCN2021070982-appb-000005
分别是集合X(ε k) i的均值与方差;
(5)将P(ε k|order *)最大概率值对应的设计模式作为第k个模块的设计模式匹配结果
Figure PCTCN2021070982-appb-000006
由于P(order *)为常数,比较{P(ε k|order *)|ε k=1,2,3}的大小即比较{P(ε k)P(order *|ε k)|ε k=1,2,3}的大小;
(6)得到机械产品的所有模块的设计模式匹配结果,形成最终的设计模式匹配结果
Figure PCTCN2021070982-appb-000007
进一步地,所述设计模式ε k=1,2,3;其中,ε k=1表示第k个模块采用按订单配置设计模式,ε k=2表示第k个模块采用按订单变形设计模式,ε k=3表示第k个模块采用按订单生成设计模式。
本发明的有益效果是:
1.本发明方法从概率学角度出发,依据贝叶斯定理研究产品定制设计中各个设计模块的 个性化设计模式匹配方法。通过计算不同设计模块采用不同设计模式的用户满意概率值,匹配出各个设计模块对应的设计模式,减少设计人员靠设计经验选择设计模式时的试错情况,提高设计效率。
2.本发明提高了个性化设计模式匹配过程中的智能性与操作性。
附图说明
图1是“互联网+”环境机械产品个性化设计流程图;
图2是“互联网+”环境机械产品个性化设计模式匹配流程图;
图3是“互联网+”环境电梯轿厢系统个性化设计实现过程图。
具体实施方式
以下结合附图和实施例对本发明作进一步的详细描述。
如图1所示,是本发明相应的“互联网+”环境机械产品个性化设计流程图,步骤如下:
(1.1)用户通过多源终端构建的“互联网+”环境,提出自己对机械产品的个性化需求。互联网+设计平台将用户需求转化为产品订单,通过网络服务器传递给设计人员。
(1.2)设计人员以电子订单为依据,根据个性化设计模式用户满意概率匹配方法,匹配出产品各个模块的设计模式。所述的“互联网+”环境机械产品个性化设计模式包括:按订单配置设计模式、按订单变形设计模式、按订单生成式设计模式,其步骤如下:
(1)复杂产品按设计模块进行分解,按模块逐一设计。每个模块的设计模式通过订单内容与产品模块库匹配获得。
(2)选择按订单配置设计模式,根据产品设计参数、配置规则、配置模板从模块库中匹配出符合设计要求的模块。
(3)选择按订单变形设计模式,定位待变形的模块,从模块库中检索相似配置模块,选择最优移植母版与移植备选模块,将满足设计要求的可用结构进行结构特征分割,提取母版性能参数差异,将可用结构移植(替换)到母版上,重建约束,在满足性能要求下,重构与邻接模块的标准接口,形成按订单变形的模块设计方案。
(4)选择按订单生成式设计模式,用于模块库中完全不匹配的模块生成式设计,设置约束条件、边界条件、载荷条件,生成多种满足条件的生成式设计结果,对生成的模块进行性能模拟,并选择满足订单需求的模块作为模块设计方案。
(5)对每个模块逐次设计,采用(2)-(4)的设计模式进行设计,完成全部设计后,进行复杂产品的模块方案融合,生成机械产品定制设计方案。
(1.3)设计人员完成设计方案后,用户通过“互联网+”平台实时互反馈,设计人员修改设计方案直至满足用户需求;此外,用户可以直接参与设计全过程,在模块逐次设计过程中即可 提出更改意见供设计人员修改。最终完成的设计方案应完全满足用户的个性化需求。
所述的互联网+设计平台是为“互联网+”环境机械产品个性化设计所提供的设计平台,由用户与设计人员共同参与产品定制设计,一般是用户提出或修改需求,设计人员根据需求转化的订单进行定制设计,用户参与设计全过程,与设计人员实时互反馈以完成满足用户需求的产品。具体地,本发明一种面向“互联网+”环境机械产品个性化设计模式匹配方法包括以下步骤:
1.“互联网+”环境下用户根据自身的个性化需求,提交需求并根据预设模板生成订单;
2.机械产品的定制设计过程在互联网+设计平台上完成,共有三种设计模式,分别是按订单配置设计模式、按订单变形设计模式和按订单生成式设计模式。根据订单内容与已有产品模块匹配设计模式,设计人员根据匹配结果对产品模块逐一进行并完成设计。
所述的按订单配置设计模式的步骤如下,
A1.个性化订单参数输入,转化为机械产品设计参数;
A2.根据设计参数、配置结构、配置规则和配置模块,从配置模块库中匹配出适合模块,形成按订单配置的机械产品定制设计方案;
A3.设计方案传输至用户端,评价方案是否满足自身需求,若满足则输出机械产品定制设计方案完成设计,若不满足则返回步骤2重新配置,或者更换另两种设计模式。
所述的按订单变形设计模式的步骤如下,
B1.个性化订单参数输入,转化为机械产品设计参数;
B2.在按订单配置设计无法满足订单要求,用户实时反馈评价需要修改设计方案,和模块库中只有局部满足设计需求的产品模块等情况下,可采用按订单变形设计模式;
B3.定位待变形的产品模块,并检索模块库中相似模块;
B4.相似模块评估,选择最优移植模块母版与移植备选模块;
B5.分析最优移植模块母版与设计参数的性能差异,提取移植备选模块中可用结构并进行分割;
B6.将移植备选模块中分割的可用结构移植或替换至最优移植模块母版中,重建结构约束;
B7.设计人员先判断新模块是否满足性能,不满足需进行结构优化或转入步骤5-6,若满足则与用户互反馈,重复步骤4-6,以致完全符合用户需求。
B8.将按订单变形设计的新模块与邻接模块之间的接口标准化;
B9.重复步骤3-8,完成其余模块变形设计,输出按订单变形的机械产品定制设计方案。
所述的按订单生成式设计模式的步骤如下,
C1.个性化订单参数输入,转化为机械产品设计参数;
C2.在按订单配置设计与订单变形设计均无法满足订单要求,用户实时反馈评价需要修改设计方案,和模块库中不存在满足设计需求的产品模块等情况下,可采用按订单生成式设计模式设计该模块;
C3.定位需要更改的模块或者零件,根据订单要求更新产品订单库,便于之后生成式模块入库存储;
C4.利用产品设计资源库辅助按订单生成式设计模式,设置约束要求,边界条件,载荷条件等,生成海量模块生成式设计结果;
C5.设计人员从海量生成式设计模块中选择满足产品性能的模块,传输给用户进行评价,协同选择出完全满足用户需求的模块;若在此过程中用户更改需求,转到步骤4;
C6.重复步骤3-5,完成其余模块生成式设计,形成按订单生成式的机械产品定制设计方案;
C7.生成式模块进行模型、文档、结构、规则建模,并融入产品设计资源库。
如图2所示,是本发明中个性化设计模式用户满意概率匹配方法的流程图。所述的根据订单内容与已有产品模块匹配设计模式,采用个性化设计模式的用户满意概率匹配方法,该方法步骤如下:
2.1.构建用户个性化需求的订单特征向量order={(req 1,r 1),(req 2,r 2),...,(req i,r i),...,(req n,r n)};其中,req i表示订单特征向量中的第i个需求特征,r i表示req i对应的归一化用户个性化需求值,n为需求特征数。对同类机械产品,不同订单特征向量中的需求特征相同,但由于个性化的需求,不同订单相同需求特征下的归一化需求值r皆不同。
2.2.构建产品分解模块集D,每个机械产品在定制设计前,都需要分解产品模块,通过模块设计的思想,按模块逐一设计,同类机械产品其模块划分结果也相同D={d 1,d 2,...,d m};其中,d k表示产品划分的第k个模块,共划分为m个模块,1≤k≤m。
2.3.根据满足用户需求的订单记录构成的设计案例库,构建设计模式匹配案例库
Figure PCTCN2021070982-appb-000008
pattern j={p 1j,p 2j,...,p kj,...,p mj};其中,M表示设计模式匹配案例库中记录的订单数量;p kj表示第j个订单记录中第k个模块d k采用的设计模式,pattern j表示第j个订单记录中各个模块的设计模式匹配结果p kj的集合。
2.4.输入新的订单特征向量order *,比较D中的第k个模块的三种设计模式对应的用户满意概率P(ε k|order *)(ε k=1,2,3),选择用户满意概率最高的设计模式,包括以下子步骤:
2.4.1找到X中第k个模块的设计模式匹配结果的集合{p kj|j=1~M},根据该集合中的不同设计模式将X划分成X(ε k),ε k=1,2,3;其中,ε k表示第k个模块选择的设计模式,ε k=1表示第k个模块采用按订单配置设计模式,ε k=2表示第k个模块采用按订单变形设计模式,ε k=3表示第k个模块采用按订单生成设计模式;X(ε k)表示第k个模块采用ε k设计模式的案例集合。
2.4.2找到X中第i个需求特征的归一化用户个性化需求值的集合{r ij|j=1~M},根据步骤2.4.1得到的X(ε k)将该集合划分为X(ε k) i,则X(ε k) i表示第k个模块采用ε k设计模式时对应的订单记录中第i个需求特征的归一化用户个性化需求值的集合。
2.4.3计算在X中第k个模块选择不同设计模式ε k=1,2,3的概率P(ε k):
Figure PCTCN2021070982-appb-000009
其中,|X(ε k)|表示X(ε k)中元素的个数。
2.4.4计算集合X(ε k) i的均值
Figure PCTCN2021070982-appb-000010
与方差
Figure PCTCN2021070982-appb-000011
2.4.5计算新的订单特征向量order *中需求特征req i *对应的r i *值不同时第k个模块选择不同设计模式的条件概率P(r i *|ε k)为:
Figure PCTCN2021070982-appb-000012
式中,r i *表示order *中第i个需求特征req i *的归一化用户个性化需求值,由于r i *的连续性,满足概率密度函数的分布规律,
Figure PCTCN2021070982-appb-000013
2.4.3第k个模块根据新的订单特征向量order *选择不同设计模式的条件概率P(order *|ε k)为:
Figure PCTCN2021070982-appb-000014
式中,P(order *|ε k)由order *中的n个r i值联合决定,基于属性条件独立性假设。
2.4.4基于贝叶斯定理得到第k个模块采用不同设计模式时用户满意概率P(ε k|order *)为:
Figure PCTCN2021070982-appb-000015
式中,0<P(order *)≤1表示证据因子,为常数;因此,比较P(ε k|order *)的大小就是比较 {P(ε k)·P(order *|ε k)}的大小,选择{P(ε k)·P(order *|ε k)|ε k=1,2,3}中的最大值对应的设计模式,作为匹配结果
Figure PCTCN2021070982-appb-000016
2.5依次对每个订单特征向量进行计算与匹配,形成最终的设计模式匹配结果
Figure PCTCN2021070982-appb-000017
指导设计人员设计。
3.设计人员完成机械产品定制设计方案,在互联网+设计平台上直接反馈给用户,用户可以体验产品性能,确定是否满足需求。若不满足,则实时反馈给设计人员将不满足需求的部分重新设计;若满足,则确定产品设计方案转向制造阶段。
本发明按用户订单为基础进行机械产品的定制设计,以单个用户需求驱动设计,且用户通过互联网+设计平台参与设计全过程,与设计人员实时互反馈,使最后设计方案满足用户个性化需求。本发明提出的三种机械产品个性化设计模式:按订单配置设计模式,按订单变形设计模式,按订单生成式设计模式,满足了几乎全部的机械产品的定制设计需求。三种设计模式有效地缓解需求个性化与生产规模化之间的矛盾。设计模式匹配阶段,省去了设计人员根据经验判断满足订单需求的模块设计模式,提高设计效率与设计结果的用户满意度。
电梯作为一种高度个性化的定制机械产品,广泛运用在生活中,不同建筑的电梯系统存在较大差异,因此用户对电梯的需求也是个性化的。电梯分为驱动系统、悬挂系统,层门系统,轿厢系统等多个系统,每个系统都可独立设计与组装,本发明选择个性化程度最高的轿厢系统作为具体实施例进行说明。
结合一个简化的例子说明该方法的实施过程。表1为简化后的局部电梯设计模式匹配案例库,新的个性化订单特征向量order *={(载重,0.54),(场景,0.25),(速度,0.61),(楼层,0.45),(装潢,0.14)},以编号24模块的设计模式匹配为例,需比较P(ε 24|order *)(ε 24=1,2,3)三个值的大小,确定选择何种设计模式。具体计算步骤如下:
表1:电梯设计模式匹配案例库(简化局部)
Figure PCTCN2021070982-appb-000018
P(ε 24=1)=0.6,P(ε 24=2)=0.4,P(ε 24=3)=0
Figure PCTCN2021070982-appb-000019
Figure PCTCN2021070982-appb-000020
Figure PCTCN2021070982-appb-000021
Figure PCTCN2021070982-appb-000022
Figure PCTCN2021070982-appb-000023
Figure PCTCN2021070982-appb-000024
Figure PCTCN2021070982-appb-000025
Figure PCTCN2021070982-appb-000026
Figure PCTCN2021070982-appb-000027
Figure PCTCN2021070982-appb-000028
设P(order *)=1,则P(ε 24=1|order *)=0.574,P(ε 24=2|order *)=3.06×10 -84,P(ε 24=3|order *)=0,因此编号24模块应使用按订单配置的设计模式。其余模块设计模式匹配的计算方式与此类似。
如图3所示,是本发明相应的“互联网+”环境电梯轿厢系统个性化设计实现过程图。
(1)用户在“互联网+”环境下提出的个性化需求,转化为新的电子订单,电子订单生成的用户个性化需求的订单特征向量order*,根据本发明提出的个性化设计模式匹配方法,按照以划分的设计模块逐一计算三种设计模式用户满意概率,进行设计模式匹配,其结果显示有91%模块通过按订单配置设计模式,7%模块通过按订单变形设计模式,2%模块通过按订单生成式设计模式,如图3(a)所示。
(2)匹配结果中轿架,轿厢,防护罩,轿壁等设计模块都可采用按订单配置设计模式。通过已有的配置规则库,配置模块库与电梯轿厢系统的配置结构,从模块库中自上而下匹配满足订单需求的零部件,形成按订单配置的各模块具体设计方案,如图3(b)所示。
(3)按订单变形设计模式以绳轮为例说明设计流程。先从模块库中检索出相似配置模块,相似配置方案评估后,选择最优移植母版与移植备选模块,从移植备选模块中提取可用结构, 从最有移植母版中提取性能参数差异,将可用结构通过结构特征分割后移植到母版上,重建约束,并进行结构优化以满足订单需求,新的变形模块与邻接模块之间的接口标准化,形成按订单变形设计的绳轮具体设计方案,如图2(c)所示。
(4)按订单生成式设计模式以拨架体为例进行说明设计流程。根据关于拨架体的订单信息设置约束条件、边界条件、载荷条件后,按订单生成式设计模式可以得到多种满足条件的生成式设计结果,对其进行性能模拟,选择满足订单需求的拨架体设计方案,如图3(d)所示。
(5)图3(b)-(d)都是以某一零部件为例,其余零部件按设计模式匹配结果逐一设计,最后由设计人员集成各模块,由用户与设计人员在互联网+设计平台上进行实时互反馈,设计人员对设计结果进行修改或重新设计,直至完全满足用户需求。

Claims (2)

  1. 一种面向“互联网+”环境机械产品个性化设计模式匹配方法,其特征在于,包括以下步骤:(1)构建用户个性化需求的订单特征向量order:
    order={(req 1,r 1),(req 2,r 2),...,(req i,r i),...,(req n,r n)}
    其中,req i表示第i个需求特征,r i表示第i个需求特征的归一化需求值,n为需求特征数;
    (2)将机械产品分解成m个模块d 1~d m,构建产品分解模块集D={d 1,d 2,...,d m};
    (3)根据满足用户需求的设计模式方案的历史订单记录,构建设计模式匹配案例库X:
    Figure PCTCN2021070982-appb-100001
    pattern j={p 1j,p 2j,...,p kj,...,p mj}
    其中,M表示设计模式匹配案例库中历史订单的数量,order j表示第j个订单的订单特征向量;p kj表示第j个订单中第k个模块采用的设计模式,pattern j表示第j个订单中各个模块设计模式匹配结果,1≤k≤m;
    (4)针对新的订单特征向量order *,机械产品第k个模块采用不同设计模式ε k时用户满意概率P(ε k|order *)为:
    Figure PCTCN2021070982-appb-100002
    其中,0<P(order *)≤1为常数;P(ε k)表示设计模式匹配案例库X中第k个模块采用ε k设计模式的概率:
    Figure PCTCN2021070982-appb-100003
    其中,X(ε k)表示X中第k个模块采用ε k设计模式的订单集合,|X(ε k)|为X(ε k)中的元素个数;P(order *|ε k)为第k个模块根据新的订单特征向量order *选择不同设计模式的条件概率:
    Figure PCTCN2021070982-appb-100004
    其中,r i *表示order *中第i个需求特征的归一化需求值,1≤i≤n;X(ε k) i是X中第k个模块 采用ε k设计模式的订单中第i个需求特征的归一化需求值r i的集合,
    Figure PCTCN2021070982-appb-100005
    分别是集合X(ε k) i的均值与方差;
    (5)将P(ε k|order *)最大概率值对应的设计模式作为第k个模块的设计模式匹配结果
    Figure PCTCN2021070982-appb-100006
    由于P(order *)为常数,比较{P(ε k|order *)|ε k=1,2,3}的大小即比较{P(ε k)P(order *|ε k)|ε k=1,2,3}的大小;
    (6)得到机械产品的所有模块的设计模式匹配结果,形成最终的设计模式匹配结果
    Figure PCTCN2021070982-appb-100007
  2. 根据权利要求1所述面向“互联网+”环境机械产品个性化设计模式匹配方法,其特征在于,所述设计模式ε k=1,2,3;其中,ε k=1表示第k个模块采用按订单配置设计模式,ε k=2表示第k个模块采用按订单变形设计模式,ε k=3表示第k个模块采用按订单生成设计模式。
PCT/CN2021/070982 2020-07-06 2021-01-08 面向"互联网+"环境机械产品个性化设计模式匹配方法 Ceased WO2022007380A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2021544343A JP7171094B2 (ja) 2020-07-06 2021-01-08 「インターネット+」環境機械製品向けの個人化設計パターン整合方法
US17/475,288 US20220004677A1 (en) 2021-01-08 2021-09-14 Mechanical product personalized design pattern matching method oriented to internet + environment

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010640643.8 2020-07-06
CN202010640643.8A CN111815409B (zh) 2020-07-06 2020-07-06 面向“互联网+”环境机械产品个性化设计模式匹配方法

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/475,288 Continuation US20220004677A1 (en) 2021-01-08 2021-09-14 Mechanical product personalized design pattern matching method oriented to internet + environment

Publications (1)

Publication Number Publication Date
WO2022007380A1 true WO2022007380A1 (zh) 2022-01-13

Family

ID=72841594

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/070982 Ceased WO2022007380A1 (zh) 2020-07-06 2021-01-08 面向"互联网+"环境机械产品个性化设计模式匹配方法

Country Status (2)

Country Link
CN (1) CN111815409B (zh)
WO (1) WO2022007380A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115391745A (zh) * 2022-10-27 2022-11-25 国能大渡河大数据服务有限公司 一种基于概率匹配平均法的降水预报订正方法及系统
CN118887199A (zh) * 2024-08-15 2024-11-01 南通人民彩印有限公司 一种定制化印刷品的需求分析处理方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111815409B (zh) * 2020-07-06 2023-07-21 浙江大学 面向“互联网+”环境机械产品个性化设计模式匹配方法
CN112182406B (zh) * 2020-10-27 2024-09-24 上海设序科技有限公司 机械设计方案推荐方法及装置
CN115293610A (zh) * 2022-08-12 2022-11-04 桂林理工大学 一种基于个性化定制的配件订单生成系统
CN117440553B (zh) * 2023-12-20 2024-03-01 上海舒谦智能技术有限公司 一种多用途智能led场景灯控制系统及方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100145492A1 (en) * 2008-12-09 2010-06-10 The Boeing Company Automated Custom Design Generation
CN103679334A (zh) * 2013-10-14 2014-03-26 浙江工业大学 考虑生命周期影响因子的确定产品平台参数的方法
CN109166004A (zh) * 2018-08-03 2019-01-08 贵州大学 基于案例推理的产品个性化定制方法
CN109741120A (zh) * 2018-08-16 2019-05-10 贵州大学 用于用户需求的产品深度个性化定制方法
CN110348747A (zh) * 2019-07-15 2019-10-18 齐鲁工业大学 一种产品设计的集约化设计方法及系统
CN111815409A (zh) * 2020-07-06 2020-10-23 浙江大学 面向“互联网+”环境机械产品个性化设计模式匹配方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100145492A1 (en) * 2008-12-09 2010-06-10 The Boeing Company Automated Custom Design Generation
CN103679334A (zh) * 2013-10-14 2014-03-26 浙江工业大学 考虑生命周期影响因子的确定产品平台参数的方法
CN109166004A (zh) * 2018-08-03 2019-01-08 贵州大学 基于案例推理的产品个性化定制方法
CN109741120A (zh) * 2018-08-16 2019-05-10 贵州大学 用于用户需求的产品深度个性化定制方法
CN110348747A (zh) * 2019-07-15 2019-10-18 齐鲁工业大学 一种产品设计的集约化设计方法及系统
CN111815409A (zh) * 2020-07-06 2020-10-23 浙江大学 面向“互联网+”环境机械产品个性化设计模式匹配方法

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115391745A (zh) * 2022-10-27 2022-11-25 国能大渡河大数据服务有限公司 一种基于概率匹配平均法的降水预报订正方法及系统
CN115391745B (zh) * 2022-10-27 2023-01-20 国能大渡河大数据服务有限公司 一种基于概率匹配平均法的降水预报订正方法及系统
CN118887199A (zh) * 2024-08-15 2024-11-01 南通人民彩印有限公司 一种定制化印刷品的需求分析处理方法

Also Published As

Publication number Publication date
CN111815409A (zh) 2020-10-23
CN111815409B (zh) 2023-07-21

Similar Documents

Publication Publication Date Title
CN111815409B (zh) 面向“互联网+”环境机械产品个性化设计模式匹配方法
CN111240209B (zh) 构型动型控型优型联动响应的自适应组态方法及系统
Vong et al. Case-based reasoning and adaptation in hydraulic production machine design
CN109741120B (zh) 用于用户需求的产品深度个性化定制方法
Shen et al. Atelier: An automated analog circuit design framework via multiple large language model-based agents
CN120162847A (zh) 一种基于拓扑优化的个性化设计智能交互方法
CN101093517A (zh) 产品配置设计系统
CN117150851A (zh) 一种基于智能交互的结构设计分析优化一体化方法及系统
CN112347698A (zh) 一种基于飞机结构件的人机混合增强智能设计方法
CN120804844B (zh) 基于边缘-中心混合优化多智能体强化学习故障诊断方法
Lou et al. Development of an integrated knowledge-based system for mold-base design
US20220004677A1 (en) Mechanical product personalized design pattern matching method oriented to internet + environment
Xu et al. Research on an active knowledge push service based on collaborative intent capture
Zhang Text-to-form
CN120317215B (zh) 一种基于多模态的模拟电路性能驱动自动布线方法
CN120012681B (zh) 基于强化学习算法的模拟ic网表版图协同优化方法及系统
CN119886547B (zh) 一种基于航天装备模型的指标表征评估方法
Caiyuan et al. HPRODIE: using feature modelling and feature mapping to speed up progressive die design
Li et al. Mechanical design paradigm based on ACP method in parallel manufacturing
JP7171094B2 (ja) 「インターネット+」環境機械製品向けの個人化設計パターン整合方法
Fu et al. Rapid design system for agricultural machinery using knowledge-based engineering: Case study of a wheeled combine chassis
Fu et al. Research on knowledge-based assembly process planning method based on representation learning reasoning
CN120317841B (zh) 模具成本计算方法、装置、设备和存储介质
Shen et al. Changhao Yan1, Dian Zhou1, and Xuan Zeng1 1Affiliation not available
Chen Optimization Control of Construction Engineering Management Projects Based on Deep Learning Algorithm

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021544343

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21838759

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21838759

Country of ref document: EP

Kind code of ref document: A1