CN104462205A - Self-adaptive manufacturing system organizing method based on dynamic knowledge network - Google Patents

Abstract

Translated fromChinese

本发明公开了一种基于动态知识网的制造系统自适应组织方法，首先定义动态知识网，建立动态知识网模型，然后基于制造系统中的动态因素设计动态组织的触发规则，并根据知识点的状态构造相应的静态知识子网。综合信息匹配度、功能匹配度及功能完善度定义了静态知识子网的匹配度，为知识网的动态组织提供决策依据。在知识网的动态组织过程中，首先检测知识点的状态，对于状态异常的知识点建立静态知识子网并以此为目标子网，计算目标子网与知识库中资源静态子网的匹配度，根据触发规则选择匹配度最高且大于阈值的资源子网利用多重集的运算更新知识网的结构。本发明使制造企业具备快速响应的能力，提高了其市场竞争力。

The invention discloses a method for self-adaptive organization of a manufacturing system based on a dynamic knowledge network. Firstly, a dynamic knowledge network is defined, a dynamic knowledge network model is established, and then dynamic organization trigger rules are designed based on dynamic factors in the manufacturing system. The state constructs the corresponding static knowledge subnetwork. Comprehensive information matching degree, function matching degree and function perfection degree define the matching degree of static knowledge subnet, and provide decision-making basis for the dynamic organization of knowledge network. In the process of dynamic organization of the knowledge network, the status of the knowledge points is detected first, and a static knowledge subnet is established for the knowledge points with abnormal status as the target subnet, and the matching degree between the target subnet and the resource static subnet in the knowledge base is calculated , select the resource subnetwork with the highest matching degree and greater than the threshold value according to the trigger rule, and update the structure of the knowledge network by using the multiset operation. The invention enables the manufacturing enterprise to have the ability of rapid response and improves its market competitiveness.

Description

Translated fromChinese

一种基于动态知识网的制造系统自适应组织方法A Method of Adaptive Organization of Manufacturing System Based on Dynamic Knowledge Network

技术领域technical field

本发明涉及一种基于动态知识网的制造系统自适应组织方法，属于智能制造系统及知识表示技术领域。The invention relates to an adaptive organization method of a manufacturing system based on a dynamic knowledge network, and belongs to the technical field of intelligent manufacturing systems and knowledge representation.

背景技术Background technique

知识化制造系统是2000年提出的一种高智能制造系统，将先进制造模式以知识网表示并转化为先进制造知识，通过知识网与Agent网之间的映射关系来实现。若将各种制造模式转化为先进制造知识纳入制造系统，则这种系统可随时根据需要选择最合适的先进制造模式或若干先进制造模式的恰当组合，以适应各种各样和不断变化的环境和制造企业。智能制造为系统自主、智能、动态地适应环境变化和企业需求提供了良好的平台。近年来，针对智能制造系统的控制理论和方法取得一些成果，建立了多重集理论及自重构算法，提出了基于用户功能需求的知识网的自动生成方法，建立了表达式优化模型。但已有的研究中用来表示制造模式的知识网是静态的，难以描述制造系统的动态过程。近几年，研究者结合智能制造单元的高智能特性构建了动态调度的自适应控制策略；考虑到生产环境的不确定性，构建了具有自适应和自学习特性的基于多Agent可互操作的动态调度系统。这些工作促进了智能制造系统的自适应控制研究，但并未触及智能制造系统本身结构在系统自适应过程中的动态组织和变化。Knowledge-based manufacturing system is a high-intelligence manufacturing system proposed in 2000. The advanced manufacturing model is represented by knowledge network and transformed into advanced manufacturing knowledge, which is realized through the mapping relationship between knowledge network and agent network. If various manufacturing modes are transformed into advanced manufacturing knowledge and incorporated into the manufacturing system, the system can choose the most suitable advanced manufacturing mode or the appropriate combination of several advanced manufacturing modes according to the needs at any time to adapt to various and changing environments and manufacturing companies. Intelligent manufacturing provides a good platform for the system to autonomously, intelligently and dynamically adapt to environmental changes and enterprise needs. In recent years, some achievements have been made in the control theory and method of intelligent manufacturing system. The multiset theory and self-reconfiguration algorithm have been established, the automatic generation method of knowledge network based on user functional requirements has been proposed, and the expression optimization model has been established. However, the knowledge network used to represent the manufacturing model in the existing research is static, and it is difficult to describe the dynamic process of the manufacturing system. In recent years, researchers have combined the high intelligence characteristics of intelligent manufacturing units to construct adaptive control strategies for dynamic scheduling; considering the uncertainty of the production environment, a multi-Agent interoperable control strategy with adaptive and self-learning characteristics has been constructed. Dynamic scheduling system. These works have promoted the research on adaptive control of intelligent manufacturing systems, but have not touched the dynamic organization and changes of the structure of intelligent manufacturing systems during the process of system adaptation.

发明内容Contents of the invention

本发明所要解决的技术问题是针对背景技术中的不足，提出一种基于动态知识网的制造系统自适应组织方法，使得智能制造系统结构根据生产环境参数的变化动态调整，提高生产企业的竞争力。The technical problem to be solved in the present invention is to address the shortcomings in the background technology, and propose an adaptive organization method for the manufacturing system based on a dynamic knowledge network, so that the structure of the intelligent manufacturing system can be dynamically adjusted according to the changes in the production environment parameters, and the competitiveness of the production enterprise can be improved. .

本发明为解决上述技术问题采用以下技术方案：The present invention adopts the following technical solutions for solving the problems of the technologies described above:

一种基于动态知识网的制造系统自适应组织方法，包括以下几个步骤：A method for adaptive organization of manufacturing system based on dynamic knowledge network, including the following steps:

步骤1)，定义动态知识网，建立动态知识网模型；Step 1), define the dynamic knowledge network, and establish the dynamic knowledge network model;

步骤2)，设计动态事件触发规则，建立动态规则集；Step 2), designing dynamic event trigger rules and establishing dynamic rule sets;

步骤3)，构建静态知识子网及知识库中资源集的虚拟静态知识子网；Step 3), constructing the static knowledge subnet and the virtual static knowledge subnet of the resource set in the knowledge base;

步骤4)，定义静态知识子网的匹配度，为动态组织提供决策依据；Step 4), define the matching degree of the static knowledge subnet, and provide decision-making basis for the dynamic organization;

步骤5)，检测动态知识网中知识点状态，根据静态知识子网定义构建异常知识点的静态子网并以此为目标知识子网；根据静态知识子网的匹配度计算资源子网与目标知识子网的匹配度；Step 5), detect the status of knowledge points in the dynamic knowledge network, construct the static subnet of abnormal knowledge points according to the definition of the static knowledge subnet, and use this as the target knowledge subnet; calculate the resource subnet and target according to the matching degree of the static knowledge subnet Matching degree of knowledge subnetwork;

步骤6)，根据步骤5)的计算结果将匹配度最高且大于阈值的目标知识子网根据多重集的运算规则更新动态知识网中的异常知识点，实现动态组织过程。Step 6), according to the calculation result of step 5), update the abnormal knowledge points in the dynamic knowledge network in the target knowledge subnet with the highest matching degree and greater than the threshold according to the operation rules of the multiset, and realize the dynamic organization process.

作为本发明一种基于动态知识网的制造系统自适应组织方法进一步的优化方案，步骤1)中动态知识网DKM定义为DKM＝{P,M,R,F_P,F_M,F_R,S_P,T,Q,F,D}；As a further optimization scheme of the self-adaptive organization method of the manufacturing system based on the dynamic knowledge network of the present invention, the dynamic knowledge network DKM in step 1) is defined as DKM={P,M,R,F_P ,F_M ,F_R ,S_P ,T,Q,F,D};

其中，P＝{p₁,p₂,…,p_m}是一个由m个知识点组成的有限知识点集，R＝{r₁,r₂,…,r_n}是一个由n个复合联系组成的有限复合联系集，m、n均为大于1的自然数，F_P为功能集，F_R为复合联系流，M为信息联系集，F_M为信息联系流，S_P为状态集，F为动态流，T为事件集，Q为资源集，D为决策点集；Among them, P={p₁ ,p₂ ,…,p_m } is a finite knowledge point set composed of m knowledge points, R={r₁ ,r₂ ,…,r_n } is a compound knowledge point set composed of n A finite set of compound relations composed of relations, Both m and n are natural numbers greater than 1,_FP is a function set,_FR is a compound connection flow, M is an information connection set, F_M is an information connection flow, S_P is a state set, F is a dynamic flow, and T is an event set, Q is the resource set, D is the decision point set;

P、F_P、R、F_R、M、F_M是动态知识网的静态组成部分，S_P、F、T、Q、D是动态知识网的动态组成部分。P, F_P , R,_FR , M, F_M are the static components of the dynamic knowledge network, and S_P , F, T, Q, D are the dynamic components of the dynamic knowledge network.

作为本发明一种基于动态知识网的制造系统自适应组织方法进一步的优化方案，步骤3)的具体步骤为：As a further optimization scheme of a dynamic knowledge network-based manufacturing system adaptive organization method of the present invention, the specific steps of step 3) are:

首先将基于知识点p_i的静态知识子网SSKM_pi定义为6元组：First, the static knowledge subnetwork SSKM_pi based on the knowledge point_pi is defined as a 6-tuple:

${\mathrm{<span><span>SSKM</span>SSKM</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>=</span>=</span><span><span>\{</span>\{</span>{\mathrm{<span><span>P</span>P</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>,</span>,</span>{\mathrm{<span><span>M</span>m</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>,</span>,</span>{\mathrm{<span><span>R</span>R</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>,</span>,</span>{\mathrm{<span><span>F</span>f</span>}}_{{\mathrm{<span><span>P</span>P</span>}}_{\mathrm{<span><span>pi</span>p</span>}}}<span><span>,</span>,</span>{\mathrm{<span><span>F</span>f</span>}}_{{\mathrm{<span><span>M</span>m</span>}}_{\mathrm{<span><span>pi</span>p</span>}}}<span><span>,</span>,</span>{\mathrm{<span><span>F</span>f</span>}}_{{\mathrm{<span><span>R</span>R</span>}}_{\mathrm{<span><span>pi</span>p</span>}}}<span><span>\}</span>\}</span>$

其中，i＝1,…,m，P_pi表示与知识点p_i具有联系的有限知识点集，M_pi是定义在P_pi上的复合联系集，R_pi是定义在P_pi上的信息联系集，表示P_pi上的所有功能的有限集，是定义在集合M_pi上的所有信息流的有限集，是定义在集合R_pi上的所有继承流和信息流的有限集；Among them, i=1,...,m, P_pi represents the limited knowledge point set connected with knowledge point p_i , M_pi is the compound relation set defined on P_pi , R_pi is the information relation defined on P_pi set, denote the finite set of all functions on P_pi , is a finite set of all information flows defined on the set M_pi , is a finite set of all inheritance flows and information flows defined on the set R_pi ;

然后对备用资源点的存储形式进行处理，将资源视为一种特殊的知识点，构造资源集的虚拟静态知识子网。Then, the storage form of the spare resource points is processed, the resource is regarded as a special knowledge point, and the virtual static knowledge subnet of the resource set is constructed.

作为本发明一种基于动态知识网的制造系统自适应组织方法进一步的优化方案，步骤4)中静态知识子网SSW与SSV的匹配度为$\mathrm{mat}(\mathrm{SSW},\mathrm{SSV})=\frac{f_M(\mathrm{SSW},\mathrm{SSV})}{\mathrm{\&epsiv;}(\mathrm{SSW},\mathrm{SSV})\&CenterDot;\mathrm{\&gamma;}(\mathrm{SSW},\mathrm{SSV})},$其中，f_M(SSW,SSV)为功能匹配度，ε(SSW,SSV)为功能完善度，γ(SSW,SSV)为信息匹配度。As a further optimization scheme of the self-adaptive organization method of the manufacturing system based on the dynamic knowledge network of the present invention, the matching degree of the static knowledge subnetwork SSW and SSV in step 4) is$\mathrm{mat}(\mathrm{SSW},\mathrm{SSV})=\frac{f_m(\mathrm{SSW},\mathrm{SSV})}{\mathrm{\&epsiv;}(\mathrm{SSW},\mathrm{SSV})\&Center\; Dot;\mathrm{\&gamma;}(\mathrm{SSW},\mathrm{SSV})},$ Among them, f_M (SSW, SSV) is the function matching degree, ε(SSW, SSV) is the function perfection degree, and γ(SSW, SSV) is the information matching degree.

作为本发明一种基于动态知识网的制造系统自适应组织方法进一步的优化方案，步骤6)中实现动态组织过程的具体步骤为：As a further optimization scheme of a dynamic knowledge network-based manufacturing system adaptive organization method of the present invention, the specific steps for realizing the dynamic organization process in step 6) are:

步骤6.1)，查询知识点p_i的状态，若转步骤6.2)，否则转步骤6.1)查询下一个知识点的状态；Step 6.1), query the state of knowledge point p_i , if Go to step 6.2), otherwise go to step 6.1) to query the status of the next knowledge point;

步骤6.2)，根据定义构造基于知识点p_i的静态知识子网SSW，并将其视为目标知识子网，触发相应事件；Step 6.2), according to the definition, construct the static knowledge subnetwork SSW based on the knowledge point p_i , regard it as the target knowledge subnetwork, and trigger the corresponding event;

步骤6.3)，构造资源集中备用资源的虚拟知识子网，并以虚拟子网的形式存储于知识库中；Step 6.3), constructing a virtual knowledge subnetwork of backup resources in resource concentration, and storing it in the knowledge base in the form of a virtual subnetwork;

步骤6.4)，计算步骤6.3)中的虚拟知识子网与步骤6.2)中目标知识子网的匹配度，获得匹配度最高mat(·)的虚拟知识子网SSV；Step 6.4), calculate the matching degree between the virtual knowledge subnetwork in step 6.3) and the target knowledge subnetwork in step 6.2), and obtain the virtual knowledge subnetwork SSV with the highest matching degree mat(·);

步骤6.5)，根据触发规则判断mat(·)，若mat(·)≥θ，转步骤6.6)否则转步骤6.7)，其中，θ为阈值，由专家经验或先验知识获取；Step 6.5), judge mat(·) according to the trigger rule, if mat(·)≥θ, go to step 6.6), otherwise go to step 6.7), where θ is the threshold, obtained by expert experience or prior knowledge;

步骤6.6)，利用多重集的运算理论从DKM中删除SSW，添加SSV，更新动态知识网结构，转步骤6.1)；Step 6.6), use the operation theory of multiple sets to delete SSW from DKM, add SSV, update the dynamic knowledge network structure, and turn to step 6.1);

步骤6.7)，触发DKM中相应的其他分支事件，事件处理结束转步骤6.1)。Step 6.7), trigger other corresponding branch events in DKM, and turn to step 6.1) after the event processing is completed.

本发明采用以上技术方案与现有技术相比，具有以下技术效果：Compared with the prior art, the present invention adopts the above technical scheme and has the following technical effects:

1.本发明提出的动态知识网模型突破了以往研究中用以表示制造模式的知识网的静态性，能够描述制造系统的动态过程；1. The dynamic knowledge network model proposed by the present invention breaks through the static nature of the knowledge network used to represent the manufacturing model in previous studies, and can describe the dynamic process of the manufacturing system;

2.本发明提供了静态知识子网的匹配度函数，给出一种基于信息匹配度、功能匹配度和完善度匹配度的匹配度量方法；2. The present invention provides the matching degree function of the static knowledge subnet, and provides a matching measurement method based on information matching degree, function matching degree and complete degree matching degree;

3.知识网的动态组织方法能够根据生产企业所面临的生产要素的变化动态调整知识网的结构，指导企业调整生产模式，使制造企业具备快速响应的能力，提高市场企业的竞争力。3. The dynamic organization method of the knowledge network can dynamically adjust the structure of the knowledge network according to the changes of the production factors faced by the production enterprise, guide the enterprise to adjust the production mode, enable the manufacturing enterprise to have the ability of rapid response, and improve the competitiveness of the market enterprise.

附图说明Description of drawings

图1是本发明的动态知识网模型；Fig. 1 is the dynamic knowledge network model of the present invention;

图2(a)、图2(b)、图2(c)是本发明的静态知识子网并运算中子网结构；Fig. 2 (a), Fig. 2 (b), Fig. 2 (c) are static knowledge subnet of the present invention and subnet structure in operation;

图3(a)、图3(b)、图3(c)是本发明的静态知识子网差运算中子网结构；Fig. 3 (a), Fig. 3 (b), Fig. 3 (c) are subnet structures in the static knowledge subnetwork difference operation of the present invention;

图4是本发明的动态组织流程图；Fig. 4 is a dynamic organization flowchart of the present invention;

图5是本发明的动态组织使能工具界面。Fig. 5 is a dynamic organization enabling tool interface of the present invention.

具体实施方式Detailed ways

下面结合附图对本发明的技术方案做进一步的详细说明：Below in conjunction with accompanying drawing, technical scheme of the present invention is described in further detail:

本发明是一种知识网的动态组织方法，包括以下内容：The present invention is a kind of dynamic organization method of knowledge network, comprises the following contents:

1.定义动态知识网DKM为11元组：DKM＝{P,M,R,F_P,F_M,F_R,S_P,T,Q,F,D}，其中：P＝{p₁,p₂,…,p_m}是一个由m个知识点组成的有限知识点集。R＝{r₁,r₂,…,r_n}是一个由n个复合(从属与信息)联系组成的有限复合联系集，M＝{m₁,m₂,…,m_k}是一个由除了父子知识点之外的知识点间k个信息联系组成的有限信息联系集，R∩M＝Φ，Φ是空集。F_P＝{F_p1,F_p2,……,F_pm}是一个定义在集合P上的所有功能的有限集，和i＝1,2,…,m，是知识点p_i的所有功能的有限集，是F_P的一个子集。F_R＝{F_r1,F_r2,……,F_rn}是一个定义在集合R上的所有继承流和信息流的有限集，和j＝1,2,…,n，F_rj是复合联系r_j的所有继承流和信息流的有限集，是F_R的一个子集。F_M＝{F_m1,F_m2,……,F_ml}是一个定义在集合M上的所有信息流的有限集，和l＝1,2,…,k,F_ml是信息联系m_l的所有信息流的有限集，是F_M的一个子集。S_P＝{S_p1,S_p2,……,S_pm}是定义在集合P上的所有状态的有限集，和i＝1,2,…,m,S_pi是知识点p_i的所有状态的有限集，是S_P的一个子集。T＝{T₁,T₂,……,T_e}是由e个事件组成的有限事件集。Q＝{Q₁,Q₂,……,Q_g}是由g个资源组成的有限资源集。F＝{F₁,F₂,……,F_σ}是由σ个动态流组成的动态流关系集，D＝{D₁,D₂,……,D_e}是由e个决策点组成的有限决策点集，用于标识决策位置及存储动态事件触发的决策规则R_u。1. Define the dynamic knowledge network DKM as an 11-tuple: DKM={P,M,R,F_P ,F_M ,F_R ,S_P ,T,Q,F,D}, where: P={p₁ , p₂ ,…,p_m } is a limited knowledge point set composed of m knowledge points. R={r₁ ,r₂ ,…,r_n } is a finite compound relation set composed of n compound (subordination and information) relations, M={m₁ ,m₂ ,…,m_k } is a limited information connection set composed of k information connections between knowledge points except parent-child knowledge points, R∩M=Φ, Φ is an empty set. F_P ={F_p1 ,F_p2 ,...,F_pm } is a finite set of all functions defined on the set P, and i=1,2,...,m, It is a finite set of all functions of knowledge point p_i , and is a subset of F_P. F_R ={F_r1 ,F_r2 ,...,F_rn } is a finite set of all inheritance flows and information flows defined on the set R, and j=1,2,...,n, F_rj is a finite set of all inheritance flows and information flows of the composite relation r_j , and is a subset of_FR . F_M ={F_m1 ,F_m2 ,...,F_ml } is a finite set of all information flows defined on the set M, And l=1,2,...,k, F_ml is a finite set of all information flows of information connection m_l , and is a subset of F_M. S_P = {S_p1 ,S_p2 ,...,S_pm } is a finite set of all states defined on the set P, and i=1,2,...,m, S_pi is a finite set of all states of knowledge point_pi , and is a subset of S_P. T={T₁ , T₂ ,...,T_e } is a finite event set consisting of e events. Q={Q₁ , Q₂ ,...,Q_g } is a finite resource set composed of g resources. F＝{F₁ ,F₂ ,...,F_σ } is a dynamic flow relationship set composed of σ dynamic flows, D={D₁ ,D₂ ,...,D_e } is a limited set of decision points consisting of e decision points, used to identify decision positions and store decision rules R_u triggered by dynamic events.

2.建立动态知识网模型。动态知识网由静态和动态两部分组成，其模型如图1所示。静态组成部分由知识点、功能集、联系集以及联系流组成。每一先进制造系统都是某一先进制造模式的具体实现，都有若干功能模块和子功能模块组成，将所有功能模块映射为知识点集，将所有模块能实现的功能映射为功能集，将功能模块之间或功能模块与子功能模块之间的信息联系关系映射为联系集与联系流集，即可获得动态知识网的静态组成部分。动态部分由状态集、动态流、事件集、资源集和决策点集组成。为了描述知识点的运行状态定义了状态集，其元素可以描述知识点的工作状态、需求状态以及供应状态等，根据不同制造系统知识粒度选择的不同进行相应的定义。事件集体现知识网动态组织过程中可能发生的事件，如机器故障、生产任务更新等。若事件集中有事件发生将引发知识网结构的变化，使知识网动态适应生产环境的变化。资源集是由存储在知识库中的备用资源组成，如备用加工机器，紧急供应物料等。当有事件发生时备用资源与事件主体进行匹配，若匹配度满足匹配阈值的要求则触发相应事件，更新知识网的结构。决策点集确定了动态知识网中需要进行决策的位置，且若规则集中决策规则满足则触发相应的事件。2. Establish a dynamic knowledge network model. The dynamic knowledge network consists of static and dynamic parts, and its model is shown in Figure 1. The static component consists of knowledge point, function set, relationship set and relationship flow. Each advanced manufacturing system is a specific realization of an advanced manufacturing model, and consists of several functional modules and sub-functional modules. All functional modules are mapped to knowledge point sets, and the functions that can be realized by all modules are mapped to functional sets. The static components of the dynamic knowledge network can be obtained by mapping the information relationship between modules or between functional modules and sub-functional modules into a relationship set and a relationship flow set. The dynamic part consists of state sets, dynamic flows, event sets, resource sets, and decision point sets. In order to describe the operating state of knowledge points, a state set is defined, and its elements can describe the working state, demand state, and supply state of knowledge points, etc., and corresponding definitions are made according to the selection of knowledge granularity in different manufacturing systems. The event collection reflects the events that may occur during the dynamic organization of the knowledge network, such as machine failures, production task updates, and so on. If an event occurs in the event set, it will trigger a change in the structure of the knowledge network, making the knowledge network dynamically adapt to changes in the production environment. The resource set is composed of spare resources stored in the knowledge base, such as spare processing machines, emergency supply materials, etc. When an event occurs, the backup resources are matched with the event subject, and if the matching degree meets the matching threshold requirements, the corresponding event is triggered to update the structure of the knowledge network. The decision-making point set determines the location where decision-making needs to be made in the dynamic knowledge network, and if the decision-making rules in the rule set are satisfied, the corresponding event will be triggered.

3.设计动态事件触发规则，建立动态规则集。DKM具有动态组织能力。当系统中出现意外情况时，DKM根据决策点的动态规则触发相应的事件，同时从系统的资源集中查找匹配资源替换异常知识点，使系统能够自适应生产要素的动态变化，实现知识网的动态组织过程。3. Design dynamic event trigger rules and establish dynamic rule sets. DKM has the ability of dynamic organization. When an unexpected situation occurs in the system, DKM triggers the corresponding event according to the dynamic rules of the decision point, and at the same time finds matching resources from the system resource set to replace the abnormal knowledge point, so that the system can adapt to the dynamic changes of production factors and realize the dynamic knowledge network organizational process.

4.构建动态知识网静态知识子网及知识库中资源集的知识子网。4. Construct the static knowledge subnet of the dynamic knowledge network and the knowledge subnet of the resource set in the knowledge base.

(a).基于知识点p_i的静态知识子网SSKM_pi(Static Sub-Knowledge Mesh of p_i)定义为6元组：其中，P_pi表示与知识点p_i具有联系的有限知识点集，M_pi是定义在P_pi上的复合联系集，R_pi是定义在P_pi上的信息联系集，表示P_pi上的所有功能的有限集，是定义在集合M_pi上的所有信息流的有限集，是定义在集合R_pi上的所有继承流和信息流的有限集；$P_{\mathrm{pi}}\&Subset;P,$$M_{\mathrm{pi}}\&Subset;M,R_{\mathrm{pi}}\&Subset;R,F_{P_{\mathrm{pi}}}\&Subset;F_P,F_{M_{\mathrm{pi}}}\&Subset;F_M,F_{R_{\mathrm{pi}}}\&Subset;F_R.$(a). The static knowledge subnetwork SSKM_pi (Static Sub-Knowledge Mesh of p_i ) based on the knowledge point p_i is defined as a 6-tuple: Among them, P_pi represents a limited set of knowledge points related to knowledge points_pi , M_pi is a compound relation set defined on P_pi , R_pi is an information relation set defined on P_pi , denote the finite set of all functions on P_pi , is a finite set of all information flows defined on the set M_pi , is a finite set of all inheritance flows and information flows defined on the set R_pi ;$P_p\&Subset;P,$$m_p\&Subset;m,R_p\&Subset;R,f_{P_p}\&Subset;f_P,f_{m_p}\&Subset;f_m,f_{R_p}\&Subset;f_R.$

(b).考虑备用资源无法孤立的存储于知识库中，必须明确此资源所有的联系信息与继承关系，因此需要对备用资源点的存储形式进行处理。将资源视为一种特殊的知识点，构造资源的虚拟静态知识子网。即将资源视为实知识点，将静态知识网定义中的其他知识点视为虚拟知识点，且将所有联系和流关系均视为虚拟关系，从而实现不便存储的孤立资源到特殊的虚拟静态知识子网的映射。(b). Considering that standby resources cannot be stored in the knowledge base in isolation, all contact information and inheritance relations of this resource must be clarified, so the storage form of standby resource points needs to be processed. Treat resources as a special kind of knowledge point and construct a virtual static knowledge subnetwork of resources. That is to say, resources are regarded as real knowledge points, other knowledge points in the definition of static knowledge network are regarded as virtual knowledge points, and all connections and flow relations are regarded as virtual relations, so as to realize the transformation of isolated resources that are inconvenient to store into special virtual static knowledge Mapping of subnets.

(c).静态知识子网SSN上的多重集SSN_M定义为SSN_M＝{α₁x₁,α₂x₂,…,α_tx_t}，其中，{x₁,x₂,…,x_t}为6个有限集P_pi、R_pi、M_pi、和组成的具有t个元素的大集合，α₁,α₂,…,α_t为有界实数，称为元素系数。对于q＝1,2,…,t，正的α_q表示元素x_q的个数，α_q＝0表示无元素x_q(0x_q也可视为空集)，负的α_q表示欠了|α_q|个元素x_q。在实际应用中，元素系数通常取有限整数。(c). The multiset SSN_M on the static knowledge subnetwork SSN is defined as SSN_M = {α₁ x₁ ,α₂ x₂ ,…,α_t x_t }, where {x₁ ,x₂ ,…, x_t } are six finite sets P_pi , R_pi , M_pi , and A large set composed of t elements, α₁ , α₂ ,…, α_t is a bounded real number, called element coefficient. For q=1,2,...,t, a positive α_q means the number of elements x_q , α_q =0 means no element x_q (0x_q can also be regarded as an empty set), and a negative α_q means owed |α_q | elements x_q . In practical applications, element coefficients usually take finite integers.

5.定义静态知识子网的匹配度，为动态组织提供决策依据。5. Define the matching degree of the static knowledge subnet, and provide decision-making basis for the dynamic organization.

静态知识子网SSW＝{x_w1,x_w2,…,x_wt}与SSV＝{x_v1,x_v2,…,x_vs}的匹配度为The matching degree of static knowledge subnet SSW＝{x_w1 ,x_w2 ,…,x_wt } and SSV＝{x_v1 ,x_v2 ,…,x_vs } is

$\mathrm{<span><span>mat</span>mat</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>,</span>,</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span><span><span>=</span>=</span>\frac{{\mathrm{<span><span>f</span>f</span>}}_{\mathrm{<span><span>M</span>m</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>,</span>,</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>}{\mathrm{<span><span>\&epsiv;</span>\&epsiv;</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>,</span>,</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span><span><span>\&CenterDot;</span>\&Center\; Dot;</span>\mathrm{<span><span>\&gamma;</span>\&gamma;</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>,</span>,</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>}$

其中，f_M(SSW,SSV)为功能匹配度，ε(SSW,SSV)为功能完善度，γ(SSW,SSV)为信息匹配度。各功能函数描述如下：Among them, f_M (SSW, SSV) is the function matching degree, ε(SSW, SSV) is the function perfection degree, and γ(SSW, SSV) is the information matching degree. Each functional function is described as follows:

(a).信息匹配度γ(SSW,SSV)定义为：(a). Information matching degree γ (SSW, SSV) is defined as:

$\mathrm{<span><span>\&gamma;</span>\&gamma;</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>,</span>,</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span><span><span>=</span>=</span>\left\{\begin{array}{cc}\frac{\mathrm{<span><span>I</span>I</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>}{\mathrm{<span><span>I</span>I</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>)</span>)</span>}<span><span>,</span>,</span>& \mathrm{<span><span>I</span>I</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>)</span>)</span><span><span><</span><</span>\mathrm{<span><span>I</span>I</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>\\ \mathrm{<span><span>1</span>1</span>}<span><span>,</span>,</span>& \mathrm{<span><span>I</span>I</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>)</span>)</span><span><span>\&GreaterEqual;</span>\&Greater\; Equal;</span>\mathrm{<span><span>I</span>I</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>\end{array}\right.$

其中，I(SSW)表示静态知识子网SSW＝{x₁,x₂,…,x_t}的信息量，定义为Among them, I(SSW) represents the information volume of the static knowledge subnetwork SSW={x₁ ,x₂ ,…,x_t }, defined as

$\mathrm{<span><span>I</span>I</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSN</span>SSN</span>}<span><span>)</span>)</span><span><span>=</span>=</span>\underset{\mathrm{<span><span>i</span>i</span>}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}}{\overset{\mathrm{<span><span>t</span>t</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}}{\mathrm{<span><span>\&beta;</span>\&beta;</span>}}_{\mathrm{<span><span>i</span>i</span>}}\mathrm{<span><span>C</span>C</span>}<span><span>(</span>(</span>{\mathrm{<span><span>\&alpha;</span>\&alpha;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>)</span>)</span>{\mathrm{<span><span>\&tau;</span>\&tau;</span>}}_{\mathrm{<span><span>i</span>i</span>}}{\mathrm{<span><span>\&delta;</span>\&delta;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>\&CenterDot;</span>\&CenterDot;</span>{\mathrm{<span><span>log</span>log</span>}}_{\mathrm{<span><span>b</span>b</span>}}<span><span>(</span>(</span>\frac{\underset{\mathrm{<span><span>\&zeta;</span>\&zeta;</span>}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}}{\overset{\mathrm{<span><span>t</span>t</span>}}{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}}{\mathrm{<span><span>\&beta;</span>\&beta;</span>}}_{\mathrm{<span><span>\&zeta;</span>\&zeta;</span>}}\mathrm{<span><span>C</span>C</span>}<span><span>(</span>(</span>{\mathrm{<span><span>\&alpha;</span>\&alpha;</span>}}_{\mathrm{<span><span>\&zeta;</span>\&zeta;</span>}}<span><span>)</span>)</span>{\mathrm{<span><span>\&tau;</span>\&tau;</span>}}_{\mathrm{<span><span>\&zeta;</span>\&zeta;</span>}}{\mathrm{<span><span>\&delta;</span>\&delta;</span>}}_{\mathrm{<span><span>\&zeta;</span>\&zeta;</span>}}}{{\mathrm{<span><span>\&beta;</span>\&beta;</span>}}_{\mathrm{<span><span>i</span>i</span>}}\mathrm{<span><span>C</span>C</span>}<span><span>(</span>(</span>{\mathrm{<span><span>\&alpha;</span>\&alpha;</span>}}_{\mathrm{<span><span>i</span>i</span>}}<span><span>)</span>)</span>{\mathrm{<span><span>\&tau;</span>\&tau;</span>}}_{\mathrm{<span><span>i</span>i</span>}}{\mathrm{<span><span>\&delta;</span>\&delta;</span>}}_{\mathrm{<span><span>i</span>i</span>}}}<span><span>+</span>+</span>\mathrm{<span><span>1</span>1</span>}<span><span>)</span>)</span>$

其中，β_i为元素x_i的权系数，β_i＞0，表示x_i的重要程度；α_i为元素x_i的多重数，$C\left({\mathrm{\&alpha;}}_i\right)=\left\{\begin{array}{cc}{\mathrm{\&alpha;}}_i& {\mathrm{\&alpha;}}_i>0\\ 0& {\mathrm{\&alpha;}}_i\&le;0\end{array}\right.;$τ_i为元素x_i的功能数；δ_i为元素x_i所具有复合联系和信息联系的联系数；b＞1为对数底数，以下从略，并约定0·log_b∞＝0。若静态知识子网的元素个数有限，则信息量I(·)收敛。Among them, β_i is the weight coefficient of element x_i , and β_i > 0 indicates the importance of x_i ; α_i is the multiplicity of element x_i ,$C\left({\mathrm{\&alpha;}}_i\right)=\left\{\begin{array}{cc}{\mathrm{\&alpha;}}_i& {\mathrm{\&alpha;}}_i>0\\ 0& {\mathrm{\&alpha;}}_i\&le;0\end{array}\right.;$ τ_i is the functional number of element x_i ; δ_i is the number of composite and information links of element x_i ; b>1 is the logarithmic base, which is omitted below, and it is agreed that 0·log_b ∞=0. If the number of elements in the static knowledge subnet is limited, the amount of information I(·) converges.

(b).功能匹配度f_M(SSW,SSV)定义为：(b). Function matching degree f_M (SSW, SSV) is defined as:

${\mathrm{<span><span>f</span>f</span>}}_{\mathrm{<span><span>M</span>m</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>,</span>,</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span><span><span>=</span>=</span>\frac{{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}_{\mathrm{<span><span>i</span>i</span>}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}}^{\mathrm{<span><span>t</span>t</span>}}{\mathrm{<span><span>\&rho;</span>\&rho;</span>}}_{\mathrm{<span><span>wi</span>wi</span>}<span><span>,</span>,</span>\mathrm{<span><span>V</span>V</span>}}{\mathrm{<span><span>\&tau;</span>\&tau;</span>}}_{\mathrm{<span><span>wi</span>wi</span>}<span><span>,</span>,</span>\mathrm{<span><span>V</span>V</span>}}<span><span>+</span>+</span>{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}_{\mathrm{<span><span>j</span>j</span>}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}}^{\mathrm{<span><span>s</span>the\; s</span>}}{\mathrm{<span><span>\&rho;</span>\&rho;</span>}}_{\mathrm{<span><span>vj</span>vj</span>}<span><span>,</span>,</span>\mathrm{<span><span>W</span>W</span>}}{\mathrm{<span><span>\&tau;</span>\&tau;</span>}}_{\mathrm{<span><span>vj</span>vj</span>}<span><span>,</span>,</span>\mathrm{<span><span>W</span>W</span>}}}{{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}_{\mathrm{<span><span>i</span>i</span>}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}}^{\mathrm{<span><span>t</span>t</span>}}{\mathrm{<span><span>\&tau;</span>\&tau;</span>}}_{\mathrm{<span><span>wi</span>wi</span>}}<span><span>+</span>+</span>{\mathrm{<span><span>\&Sigma;</span>\&Sigma;</span>}}_{\mathrm{<span><span>j</span>j</span>}<span><span>=</span>=</span>\mathrm{<span><span>1</span>1</span>}}^{\mathrm{<span><span>s</span>the\; s</span>}}{\mathrm{<span><span>\&tau;</span>\&tau;</span>}}_{\mathrm{<span><span>vj</span>vj</span>}}}$

其中，τ_wi为元素x_wi(i＝1,2,…,t)的功能数；τ_vj为元素x_vj(j＝1,2,…,s)的功能数；ρ_wi,V为SSV中与元素x_wi功能近似相同的近似系数，ρ_wi,V∈[0,1]，ρ_wi,V＝1表示功能完全相同，ρ_wi,V＝0表示功能完全不同；τ_wi,V为SSV中与元素x_wi近似相同的功能数；ρ_vj,W为SSW中与元素x_vj功能近似相同的近似系数；τ_vj,W为SSW中与元素x_vj近似相同的功能数。Among them, τ_wi is the functional number of element x_wi (i=1,2,…,t); τ_vj is the functional number of element x_vj (j=1,2,…,s); ρ_wi,V is the SSV The approximation coefficients in which the function is approximately the same as the element x_wi , ρ_wi,V ∈ [0,1], ρ_wi,V = 1 means the function is completely the same, ρ_{wi, V} = 0 means the function is completely different; τ_{wi, V} is The number of approximately the same function as the element x_wi in SSV; ρ_vj,W is the approximate coefficient of approximately the same function as the element x_vj in SSW; τ_vj,W is the number of approximately the same function as the element x_vj in SSW.

(c).功能完善匹配度ε(SSW,SSV)定义为：(c). The function perfect matching degree ε(SSW, SSV) is defined as:

$\mathrm{<span><span>\&epsiv;</span>\&epsiv;</span>}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>,</span>,</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span><span><span>=</span>=</span>\left\{\begin{array}{cc}\frac{\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&lambda;</span>\&lambda;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>)</span>)</span>}{\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&lambda;</span>\&lambda;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>}<span><span>,</span>,</span>& \stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&lambda;</span>\&lambda;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>)</span>)</span><span><span>></span>></span>\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&lambda;</span>\&lambda;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>\\ \mathrm{<span><span>1</span>1</span>}<span><span>,</span>,</span>& \stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&lambda;</span>\&lambda;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSW</span>SSW</span>}<span><span>)</span>)</span><span><span>\&le;</span>\&le;</span>\stackrel{<span><span>\&OverBar;</span>\&OverBar;</span>}{\mathrm{<span><span>\&lambda;</span>\&lambda;</span>}}<span><span>(</span>(</span>\mathrm{<span><span>SSV</span>SSV</span>}<span><span>)</span>)</span>\end{array}\right.$

其中，为综合完善度。静态知识子网SSW＝{x_w1,x_w2,…,x_wt}的综合完善度定义为λ_wi∈[0,1]为元素x_wi(i＝1,2,…t)的功能完善度，可以根据先验知识或专家根据元素的完善程度划分等级，采用模糊分析方法获得。in, for comprehensive perfection. The comprehensive perfection degree of static knowledge subnetwork SSW＝{x_w1 ,x_w2 ,…,x_wt } is defined as λ_wi ∈ [0,1] is the functional perfection of element x_wi (i=1,2,…t), which can be obtained by fuzzy analysis method based on prior knowledge or experts according to the degree of perfection of elements.

6.静态知识子网的多重集运算。6. Multiset operation of static knowledge subnet.

(1)静态知识子网并运算。静态知识子网SSW_a结构如图2(a)所示，静态知识子网SSW_b结构如图2(b)所示,其多重集分别为：(1) Static knowledge subnetwork and operation. The structure of the static knowledge subnet SSW_a is shown in Figure 2(a), and the structure of the static knowledge subnet SSW_b is shown in Figure 2(b), and its multiple sets are:

SSW_a＝SSW_Ma＝{a,b,c,d,e,f,g,h,i；a-b,a-c,a-d,b-e,b-g,c-f,c-g,d-h,d-i；SSW_a = SSW_Ma = {a, b, c, d, e, f, g, h, i; ab, ac, ad, be, bg, cf, cg, dh, di;

b-c,c-d,c-h,e-f,g-h；}b-c, c-d, c-h, e-f, g-h; }

SSW_b＝SSW_Mb＝{a,c,d,g,h,i,j,k,l,m；a-c,a-d,a-j,c-g,d-h,d-i,j-k,j-l,j-m；SSW_b = SSW_Mb = {a,c,d,g,h,i,j,k,l,m; ac,ad,aj,cg,dh,di,jk,jl,jm;

c-h,c-j,g-m,l-m；}c-h, c-j, g-m, l-m; }

其中，a,b,c分别表示知识点，a-b表示知识点a和b之间的联系，以此类推。这两个静态知识子网的组合可通过两个相应多重集的并集来实现：Among them, a, b, c represent knowledge points respectively, a-b represents the connection between knowledge points a and b, and so on. The combination of these two static knowledge subnetworks can be realized by the union of two corresponding multisets:

$\begin{array}{c}{\mathrm{SSW}}_{\mathrm{Ma}+b}={\mathrm{SSW}}_{\mathrm{Ma}}+{\mathrm{SSW}}_{\mathrm{Mb}}=\{a,b,c,d,e,f,g,h,i;a-b,a-c,a-d,b-e,b-g,c-f,c-g,\\ d-h,d-i;b-c,c-d,c-h,e-f,g-h;\}\\ +\{a,c,d,g,h,i,j,k,l,m;a-c,a-d,a-j,c-g,d-h,d-i,j-k,j-l,j-m;\\ c-h,c-j,g-m,l-m;\}\\ =\{2a,b,2c,2d,e,f,2g,2h,2i,j,k,l,m;a-b,2a-c,2a-d,a-j,b-e,b-g,\\ c-f,2c-g,2d-h,2d-i,j-k,j-l,j-m;b-c,c-d,2c-h,c-j,e-f,g-h,g-m,l-m;\}\end{array}$根据多重集与静态知识子网的映射关系，则该合并的静态知识子网的形式表示为：$\begin{array}{c}{\mathrm{SSW}}_{\mathrm{Ma}+b}={\mathrm{SSW}}_{\mathrm{Ma}}+{\mathrm{SSW}}_{\mathrm{MB}}=\{a,b,c,d,e,f,g,h,i;a-b,a-c,a-d,b-e,b-g,c-f,c-g,\\ d-h,d-i;b-c,c-d,c-h,e-f,g-h;\}\\ +\{a,c,d,g,h,i,j,k,l,m;a-c,a-d,a-j,c-g,d-h,d-i,j-k,j-l,j-m;\\ c-h,c-j,g-m,l-m;\}\\ =\{2a,b,2c,2d,e,f,2g,2h,2i,j,k,l,m;a-b,2a-c,2a-d,a-j,b-e,b-g,\\ c-f,2c-g,2d-h,2d-i,j-k,j-l,j-m;b-c,c-d,2c-h,c-j,e-f,g-h,g-m,l-m;\}\end{array}$ According to the mapping relationship between the multiset and the static knowledge subnetwork, the form of the merged static knowledge subnetwork is expressed as:

SSW_a+b＝{a,b,c,d,e,f,g,h,i,j,k,l,m；a-b,a-c,a-d,a-j,b-e,b-g,c-f,c-g,d-h,d-i,SSW_a+b ={a,b,c,d,e,f,g,h,i,j,k,l,m; ab,ac,ad,aj,be,bg,cf,cg,dh, di,

j-k,j-l,j-m；b-c,c-d,c-h,c-j,e-f,g-h,g-m,l-m；}j-k, j-l, j-m; b-c, c-d, c-h, c-j, e-f, g-h, g-m, l-m; }

并由此得到静态知识子网合并后生成新的如图2(c)所示的静态知识子网。And thus get the static knowledge subnetwork merged to generate a new static knowledge subnetwork as shown in Figure 2(c).

(2)静态知识子网差。静态知识子网SSW_a+b及静态知识子网SSW_j结构分别如图3(a)和图3(b)所示，其对应的多重集表示分别为：(2) The static knowledge subnet is poor. The structures of the static knowledge subnetwork SSW_a+b and the static knowledge subnetwork SSW_j are shown in Figure 3(a) and Figure 3(b) respectively, and their corresponding multiset representations are:

SSW_Ma+b＝{2a,b,2c,2d,e,f,2g,2h,2i,j,k,l,m；a-b,2a-c,2a-d,a-j,b-e,b-g,SSW_Ma+b = {2a,b,2c,2d,e,f,2g,2h,2i,j,k,l,m; ab,2a-c,2a-d,aj,be,bg,

c-f,2c-g,2d-h,2d-i,j-k,j-l,j-m；b-c,c-d,2c-h,c-j,e-f,g-h,g-m,l-m；}c-f, 2c-g, 2d-h, 2d-i, j-k, j-l, j-m; b-c, c-d, 2c-h, c-j, e-f, g-h, g-m, l-m; }

SSW_Mj＝{j,k,l,m；j-k,j-l,j-m；l-m；}SSW_Mj = {j,k,l,m;jk,jl,jm;lm;}

从SSW_a+b中删除SSW_j通过多重集的差集运算实现：Deleting SSW_j from SSW_a+b is achieved by subtraction of multisets:

W_Ma+b-j＝W_Ma+b-W_Mj＝{2a,b,2c,2d,e,f,2g,2h,2i,j,k,l,m；a-b,2a-c,2a-d,a-j,b-e,W_Ma+bj =W_Ma+b -W_Mj ={2a,b,2c,2d,e,f,2g,2h,2i,j,k,l,m; ab,2a-c,2a-d, aj,be,

b-g,c-f,2c-g,2d-h,2d-i,j-k,j-l,j-m；b-c,c-d,2c-h,c-j,e-f,g-h,g-m,l-m；}b-g, c-f, 2c-g, 2d-h, 2d-i, j-k, j-l, j-m; b-c, c-d, 2c-h, c-j, e-f, g-h, g-m, l-m; }

-{j,k,l,m；j-k,j-l,j-m；l-m；}-{j,k,l,m;j-k,j-l,j-m;l-m;}

＝{2a,b,2c,2d,e,f,2g,2h,2i；a-b,2a-c,2a-d,a-j,b-e,b-g,c-f,2c-g,2d-h,2d-i；={2a,b,2c,2d,e,f,2g,2h,2i; a-b,2a-c,2a-d,a-j,b-e,b-g,c-f,2c-g,2d-h,2d-i;

b-c,c-d,2c-h,c-j,e-f,g-h,g-m；}b-c, c-d, 2c-h, c-j, e-f, g-h, g-m; }

根据多重集与静态知识子网的映射关系可推出W_a+b-j，但{a-j；c-j,g-m}中所有联系都是单向的，即静态知识子网W_a+b-j是不完备的。但{a-j；c-j,g-m}恰好是W_a+b-j-c与W_j之间的割集C_(a+b-j-c)(j)。根据多重集的差集运算，从W_a+b-j去掉割集C_(a+b-j-c)(j)，就可以得到完备的静态知识子网W_a+b-j-c。According to the mapping relationship between multiset and static knowledge subnetwork, W_a+bj can be deduced, but all connections in {aj; cj, gm} are unidirectional, that is, static knowledge subnetwork W_a+bj is incomplete. But {aj; cj,gm} is exactly the cut set C_(a+bjc)(j) between W_a+bjc and W_j . According to the multiset subtraction operation, the complete static knowledge subnetwork W_a+bjc can be obtained by removing the cut set C_(a+bjc)(j) from W_a+ bj.

W_Ma+b-j-c＝W_Ma+b-j-C_(a+b-j-c)(j)W_Ma+bjc ＝W_Ma+bj -C_(a+bjc)(j)

＝{2a,b,2c,2d,e,f,2g,2h,2i；a-b,2a-c,2a-d,a-j,b-e,b-g,c-f,2c-g,2d-h,2d-i；={2a,b,2c,2d,e,f,2g,2h,2i; a-b,2a-c,2a-d,a-j,b-e,b-g,c-f,2c-g,2d-h,2d-i;

b-c,c-d,2c-h,c-j,e-f,g-h,g-m；}-{a-j；c-j,g-m；}b-c,c-d,2c-h,c-j,e-f,g-h,g-m;}-{a-j;c-j,g-m;}

＝{2a,b,2c,2d,e,f,2g,2h,2i；a-b,2a-c,2a-d,b-e,b-g,c-f,2c-g,2d-h,2d-i；={2a,b,2c,2d,e,f,2g,2h,2i; a-b,2a-c,2a-d,b-e,b-g,c-f,2c-g,2d-h,2d-i;

b-c,c-d,2c-h,e-f,g-h；}b-c, c-d, 2c-h, e-f, g-h; }

根据静态知识子网与多重集的映射，可得到如图3(c)所示的W_a+b-j-c：According to the mapping between the static knowledge subnetwork and the multiset, W_a+bjc as shown in Figure 3(c) can be obtained:

W_a+b-j-c＝{a,b,c,d,e,f,g,h,i；a-b,a-c,a-d,b-e,b-g,c-f,c-g,d-h,d-i；W_a+bjc = {a, b, c, d, e, f, g, h, i; ab, ac, ad, be, bg, cf, cg, dh, di;

b-c,c-d,c-h,e-f,g-h；}b-c, c-d, c-h, e-f, g-h; }

7.概括知识网动态组织方法，具体流程如图4所示，其具体步骤为：7. Summarize the dynamic organization method of the knowledge network. The specific process is shown in Figure 4. The specific steps are:

步骤A查询知识点p_i(i＝1,…,m)的状态，若转步骤B,否则转步骤A。Step A queries the state of knowledge points p_i (i=1,...,m), if Go to step B, otherwise go to step A.

步骤B根据定义构造基于知识点p_i的静态知识子网SSW，并将其视为目标知识子网，触发相应事件。Step B constructs the static knowledge subnetwork SSW based on the knowledge point p_i according to the definition, regards it as the target knowledge subnetwork, and triggers corresponding events.

步骤C构造资源集中备用资源的虚拟知识子网，并以虚拟子网的形式存储于知识库中。Step C constructs a virtual knowledge subnetwork of backup resources in the resource set, and stores it in the knowledge base in the form of a virtual subnetwork.

步骤D根据定义计算步骤C中的虚拟知识子网与步骤B中目标知识子网的匹配度，获得匹配度最高mat(·)的虚拟知识子网SSV。Step D calculates the matching degree between the virtual knowledge subnetwork in step C and the target knowledge subnetwork in step B according to the definition, and obtains the virtual knowledge subnetwork SSV with the highest matching degree mat(·).

步骤E根据触发规则，判断mat(·)，若mat(·)≥θ(θ为阈值，由专家经验或先验知识获取)，转步骤F否则转步骤G。Step E judges mat(·) according to the triggering rules, if mat(·)≥θ (θ is the threshold, obtained from expert experience or prior knowledge), go to step F, otherwise go to step G.

步骤F利用多重集的运算理论从DKM中删除SSW，增加SSV，更新动态知识网结构。转步骤A。Step F uses the operation theory of multiset to delete SSW from DKM, add SSV, and update the dynamic knowledge network structure. Go to step A.

步骤G触发DKM中相应的其他分支事件，事件处理结束转步骤A。Step G triggers other corresponding branch events in the DKM, and the event processing is completed and then go to step A.

8.利用C#编程语言和SQL数据库，开发知识网动态组织使能工具。在智能制造系统中，知识网是对实际制造系统的抽象反映。实际制造系统中存在着各种生产模块与管理模块，而且在各部件、模块之间有着不同的联系，因此在知识网中既要描述这些实际的功能模块，又要能反映它们之间的联系。实际上，这些功能模块和联系就是对应于知识网中知识点及知识点之间的信息联系或者复合联系。知识网数据库应该能够存储制造系统包含的所有知识，主要基表结构如以下表1、表2、表3所示。8. Use C# programming language and SQL database to develop dynamic organization enabling tools for knowledge network. In the intelligent manufacturing system, the knowledge network is an abstract reflection of the actual manufacturing system. There are various production modules and management modules in the actual manufacturing system, and there are different connections between the components and modules. Therefore, it is necessary to describe these actual functional modules and reflect the connections between them in the knowledge network. . In fact, these functional modules and connections correspond to knowledge points and information connections or compound connections between knowledge points in the knowledge network. The knowledge network database should be able to store all the knowledge contained in the manufacturing system, and the main base table structures are shown in Table 1, Table 2, and Table 3 below.

表1知识网公共属性基表结构Table 1 Structure of public attribute base table of knowledge network

字段fieldNumber/charNumber/char关键词Key words非空non empty备注RemarkkwkwChar(20)Char(20)√√√√知识点号Knowledge point numberrootrootChar(20)Char(20)√√√√根知识点号root knowledge point numberversionversionChar(10)Char(10)√√√√版本VersionkwmkwmChar(100)Char(100)知识点名Knowledge roll callxnkwwxyaChar(3)Char(3)虚拟知识点virtual knowledge pointfhkwfhkwChar(3)Char(3)复合知识点Compound knowledge pointsdckwdckwNumber(5，2)Number(5, 2)多重知识点数multiple knowledge pointsagentagentChar(100)Char(100)Agent名Agent namefzcxwxyaChar(100)Char(100)仿真程序名emulator namexnagentxnagentChar(3)Char(3)虚拟AgentVirtual Agent

表2知识点间继承流和信息流基表结构Table 2 Inheritance flow and information flow base table structure between knowledge points

列名column nameNumber/charNumber/char关键词Key words非空non empty列名column namerootrootChar(20)Char(20)√√√√根知识点号root knowledge point numberversionversionChar(10)Char(10)√√√√版本VersioninkwinkwChar(20)Char(20)√√√√输入知识点号Enter the knowledge point numberoutkwoutkwChar(20)Char(20)√√√√输出知识点号Output knowledge point numberfzkwwxyaChar(3)Char(3)√√√√父子知识点间Between father and son knowledge pointsinfkwinfkwChar(200)Char(200)√√√√信息informationdcinfdcinfNumber(5，2)Number(5, 2)多重信息数multiple information

表3知识点功能描述基表结构Table 3 Knowledge point function description base table structure

字段fieldNumber/charNumber/char关键词Key words非空non empty备注RemarkrootrootChar(20)Char(20)√√√√根知识点号root knowledge point numberversionversionChar(10)Char(10)√√√√版本VersionkwkwChar(20)Char(20)√√√√知识点号Knowledge point numbergnmsgnmsChar(240)Char(240)√√√√功能描述Functional descriptiondcgndcgnNumber(5，2)Number(5, 2)多重功能数multiple function number

使能工具开发采用模块化设计理念，各功能模块功能独立，且具有良好的接口环境，便于模块之间的灵活调用。静态知识子网的多重集运算封装在SQL的存储过程中，以提高查询速度和模块的可重用性。以磨床车间管理系统为例，使能工具的功能界面如图5所示。使能工具输出相应的控制信号驱动磨床车间的自动搬运小车和机械手等执行机构实现制造系统结构的自适应调整。The development of enabling tools adopts a modular design concept, each functional module has independent functions, and has a good interface environment, which is convenient for flexible calls between modules. The multiset operation of the static knowledge subnet is encapsulated in the stored procedure of SQL to improve the query speed and the reusability of the module. Taking the grinding machine workshop management system as an example, the functional interface of the enabling tool is shown in Figure 5. Enable tools to output corresponding control signals to drive actuators such as automatic transfer trolleys and manipulators in the grinder workshop to achieve adaptive adjustment of the manufacturing system structure.

Claims (5)

Translated fromChinese

1.一种基于动态知识网的制造系统自适应组织方法，其特征在于，包括以下几个步骤：1. A manufacturing system adaptive organization method based on dynamic knowledge network, is characterized in that, comprises the following steps:步骤1)，定义动态知识网，建立动态知识网模型；Step 1), define the dynamic knowledge network, and establish the dynamic knowledge network model;步骤2)，设计动态事件触发规则，建立动态规则集；Step 2), designing dynamic event trigger rules and establishing dynamic rule sets;步骤3)，构建静态知识子网及知识库中资源集的虚拟静态知识子网；Step 3), constructing the static knowledge subnet and the virtual static knowledge subnet of the resource set in the knowledge base;步骤4)，定义静态知识子网的匹配度，为动态组织提供决策依据；Step 4), define the matching degree of the static knowledge subnet, and provide decision-making basis for the dynamic organization;步骤5)，检测动态知识网中知识点状态，根据静态知识子网定义构建异常知识点的静态子网并以此为目标知识子网；根据静态知识子网的匹配度计算资源子网与目标知识子网的匹配度；Step 5), detect the status of knowledge points in the dynamic knowledge network, construct the static subnet of abnormal knowledge points according to the definition of static knowledge subnet and use this as the target knowledge subnet; calculate the resource subnet and target according to the matching degree of the static knowledge subnet Matching degree of knowledge subnetwork;步骤6)，根据步骤5)的计算结果将匹配度最高且大于阈值的目标知识子网根据多重集的运算规则更新动态知识网中的异常知识点，实现动态组织过程。Step 6), according to the calculation result of step 5), the target knowledge subnetwork with the highest matching degree and greater than the threshold is updated according to the operation rules of the multiset to update the abnormal knowledge points in the dynamic knowledge network to realize the dynamic organization process.2.根据权利要求1所述的一种基于动态知识网的制造系统自适应组织方法，其特征在于，步骤1)中动态知识网DKM定义为DKM＝{P,M,R,F_P,F_M,F_R,S_P,T,Q,F,D}；2. A method for adaptive organization of a manufacturing system based on a dynamic knowledge network according to claim 1, wherein the dynamic knowledge network DKM in step 1) is defined as DKM={P, M, R, F_P , F_M , F_R , S_P , T, Q, F, D};其中，P＝{p₁,p₂,…,p_m}是一个由m个知识点组成的有限知识点集，R＝{r₁,r₂,…,r_n}是一个由n个复合联系组成的有限复合联系集，m、n均为大于1的自然数，F_P为功能集，F_R为复合联系流，M为信息联系集，F_M为信息联系流，S_P为状态集，F为动态流，T为事件集，Q为资源集，D为决策点集；Among them, P={p₁ ,p₂ ,…,p_m } is a finite knowledge point set composed of m knowledge points, R={r₁ ,r₂ ,…,r_n } is a compound knowledge point set composed of n A finite set of compound relations composed of relations, Both m and n are natural numbers greater than 1,_FP is a function set,_FR is a compound connection flow, M is an information connection set, F_M is an information connection flow, S_P is a state set, F is a dynamic flow, and T is an event set, Q is the resource set, D is the decision point set;P、F_P、R、F_R、M、F_M是动态知识网的静态组成部分，S_P、F、T、Q、D是动态知识网的动态组成部分。P, F_P , R,_FR , M, F_M are the static components of the dynamic knowledge network, and S_P , F, T, Q, D are the dynamic components of the dynamic knowledge network.3.根据权利要求2所述的一种基于动态知识网的制造系统自适应组织方法，其特征在于，步骤3)的具体步骤为：3. a kind of manufacturing system adaptive organization method based on dynamic knowledge network according to claim 2, is characterized in that, the concrete steps of step 3) are:首先将基于知识点p_i的静态知识子网SSKM_pi定义为6元组：First, the static knowledge subnetwork SSKM_pi based on the knowledge point_pi is defined as a 6-tuple:${\mathrm{<span><span>SSKM</span>SSKM</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>=</span>=</span><span><span>\{</span>\{</span>{\mathrm{<span><span>P</span>P</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>,</span>,</span>{\mathrm{<span><span>M</span>m</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>,</span>,</span>{\mathrm{<span><span>R</span>R</span>}}_{\mathrm{<span><span>pi</span>p</span>}}<span><span>,</span>,</span>{\mathrm{<span><span>F</span>f</span>}}_{{\mathrm{<span><span>P</span>P</span>}}_{\mathrm{<span><span>pi</span>p</span>}}}<span><span>,</span>,</span>{\mathrm{<span><span>F</span>f</span>}}_{{\mathrm{<span><span>M</span>m</span>}}_{\mathrm{<span><span>pi</span>p</span>}}}<span><span>,</span>,</span>{\mathrm{<span><span>F</span>f</span>}}_{{\mathrm{<span><span>R</span>R</span>}}_{\mathrm{<span><span>pi</span>p</span>}}}<span><span>\}</span>\}</span>$其中，i＝1,…,m，P_pi表示与知识点p_i具有联系的有限知识点集，M_pi是定义在P_pi上的复合联系集，R_pi是定义在P_pi上的信息联系集，表示P_pi上的所有功能的有限集，是定义在集合M_pi上的所有信息流的有限集，是定义在集合R_pi上的所有继承流和信息流的有限集；Among them, i=1,...,m, P_pi represents the limited knowledge point set connected with knowledge point p_i , M_pi is the compound relation set defined on P_pi , R_pi is the information relation defined on P_pi set, denote the finite set of all functions on P_pi , is a finite set of all information flows defined on the set M_pi , is a finite set of all inheritance flows and information flows defined on the set R_pi ;然后对备用资源点的存储形式进行处理，将资源视为一种特殊的知识点，构造资源集的虚拟静态知识子网。Then, the storage form of the spare resource points is processed, the resource is regarded as a special knowledge point, and the virtual static knowledge subnet of the resource set is constructed.4.根据权利要求3所述的一种基于动态知识网的制造系统自适应组织方法，其特征在于，步骤4)中静态知识子网SSW与SSV的匹配度为$\mathrm{mat}(\mathrm{SSW},\mathrm{SSV})=\frac{f_M(\mathrm{SSW},\mathrm{SSV})}{\mathrm{\&epsiv;}(\mathrm{SSW},\mathrm{SSV})\&CenterDot;\mathrm{\&gamma;}(\mathrm{SSW},\mathrm{SSV})},$其中，f_M(SSW,SSV)为功能匹配度，ε(SSW,SSV)为功能完善度，γ(SSW,SSV)为信息匹配度。4. a kind of manufacturing system self-adaptive organization method based on dynamic knowledge network according to claim 3, is characterized in that, in step 4), the matching degree of static knowledge subnetwork SSW and SSV is$\mathrm{mat}(\mathrm{SSW},\mathrm{SSV})=\frac{f_m(\mathrm{SSW},\mathrm{SSV})}{\mathrm{\&epsiv;}(\mathrm{SSW},\mathrm{SSV})\&CenterDot;\mathrm{\&gamma;}(\mathrm{SSW},\mathrm{SSV})},$ Among them, f_M (SSW, SSV) is the function matching degree, ε(SSW, SSV) is the function perfection degree, and γ(SSW, SSV) is the information matching degree.5.根据权利要求4所述的一种基于动态知识网的制造系统自适应组织方法，其特征在于，步骤6)中实现动态组织过程的具体步骤为：5. A kind of manufacturing system self-adaptive organization method based on dynamic knowledge network according to claim 4, it is characterized in that, in step 6), the concrete steps that realize dynamic organization process are:步骤6.1)，查询知识点p_i的状态，若S_pi＝0转步骤6.2)，否则转步骤6.1)查询下一个知识点的状态；Step 6.1), query the state of the knowledge point p_i , if S_pi = 0 go to step 6.2), otherwise go to step 6.1) query the state of the next knowledge point;步骤6.2)，根据定义构造基于知识点p_i的静态知识子网SSW，并将其视为目标知识子网，触发相应事件；Step 6.2), according to the definition, construct the static knowledge subnetwork SSW based on the knowledge point p_i , regard it as the target knowledge subnetwork, and trigger the corresponding event;步骤6.3)，构造资源集中备用资源的虚拟知识子网，并以虚拟子网的形式存储于知识库中；Step 6.3), constructing a virtual knowledge subnetwork of backup resources in resource concentration, and storing it in the knowledge base in the form of a virtual subnetwork;步骤6.4)，计算步骤6.3)中的虚拟知识子网与步骤6.2)中目标知识子网的匹配度，获得匹配度最高mat(·)的虚拟知识子网SSV；Step 6.4), calculate the matching degree between the virtual knowledge subnetwork in step 6.3) and the target knowledge subnetwork in step 6.2), and obtain the virtual knowledge subnetwork SSV with the highest matching degree mat(·);步骤6.5)，根据触发规则判断mat(·)，若mat(·)≥θ，转步骤6.6)否则转步骤6.7)，其中，θ为阈值，由专家经验或先验知识获取；Step 6.5), judge mat(·) according to the trigger rule, if mat(·)≥θ, go to step 6.6), otherwise go to step 6.7), where θ is the threshold, obtained by expert experience or prior knowledge;步骤6.6)，利用多重集的运算理论从DKM中删除SSW，添加SSV，更新动态知识网结构，转步骤6.1)；Step 6.6), use the operation theory of multiple sets to delete SSW from DKM, add SSV, update the dynamic knowledge network structure, and turn to step 6.1);步骤6.7)，触发DKM中相应的其他分支事件，事件处理结束转步骤6.1)。Step 6.7), trigger other corresponding branch events in DKM, and turn to step 6.1) after the event processing is completed.