CN106295069A - A kind of data digging method in helical gear designs - Google Patents

Abstract

The invention discloses a kind of data digging method in helical gear designs, the method comprises the steps: step 1) carry out helical gear design knowledge classification；2) helical gear design data mining framework is built；3) helical gear case library is set up；4) helical gear rule base is set up；5) gear retrieval mode is set up.Present invention reduces design and the R&D cycle of product, adapted to the rhythm that new detection technology is fast-developing, the research and development for new product provide reference frame.

Description

A Data Mining Method Used in Helical Gear Design

technical field

The invention belongs to the technical field of helical gear design, in particular to a data mining method used in helical gear design.

Background technique

As an important transmission mode in mechanical transmission, helical gear transmission is widely used in transportation, machine manufacturing, aerospace and other fields, so the design quality of helical gear plays a vital role in the performance of the entire gear transmission system The role of the traditional helical gear design method is a relatively cumbersome and repetitive work.

Data mining (English: Data mining), also translated as data mining, data mining. It is a step in database knowledge discovery (English: Knowledge-Discovery in Databases, referred to as: KDD). Data mining generally refers to the process of searching for information hidden in a large amount of data through algorithms. Data mining is often associated with computer science and accomplishes the above goals through methods such as statistics, online analytical processing, intelligence retrieval, machine learning, expert systems (relying on past rules of thumb), and pattern recognition.

Using data mining technology, the helical gear example library and rule library are established, which shortens the product design and development cycle, adapts to the rapid development of modern product design, and provides a reference for the development of new products.

Contents of the invention

The technical problem to be solved by the present invention is: in order to overcome the above-mentioned defects of the prior art, a data mining method for helical gear design is proposed, which shortens the product design and development cycle, and adapts to the rapid development of modern product design The rhythm provides a reference for the development of new products.

The technical scheme adopted in the present invention is: a kind of data mining method that is used in the design of helical gear, as shown in Figure 1, it is characterized in that realization steps are as follows:

Step 1. According to the mechanical industry standards and technical specifications, helical gear design rules and experience guidelines, and helical gear design expertise, classify helical gear design knowledge, including parameters such as helical gear basic design, mechanical characteristics, and manufacturing process;

Step 2. Build a data mining framework for helical gear design. As shown in Figure 2, the bottom layer of the system is a general system layer, which is mainly responsible for parametric modeling and finite element analysis of helical gears, and provides data query, storage and editing functions; the middle layer of the system For the CAD/CAM data mining layer, using data mining algorithms, based on the helical gear instance reasoning algorithm, according to the similarity analysis between each instance, the designer extracts the instance of interest, and analyzes the relevant design criteria and methods to provide a basis for the design of new products Provide design basis and store new design methods and rules into the rule base; the upper layer of the system is the design tool layer, which mainly guides designers to manage design knowledge, add, edit and delete operations, guide users to quickly complete design tasks, and improve design efficiency ;

Step 3: Establish a helical gear instance database, as shown in Figure 3, the expression technology form of the instance is the concrete embodiment of the product information model, and according to the hierarchical product information model, the composition of the helical gear instance is described as follows:

Case＝{ID, Code, Index, Matter, Model Topology, Assembly}

Among them, ID is the unique identification code of the instance, Code represents the code, Index is the instance index, the product instance content (Matter) is composed of the name, category, main rule parameters and storage path of each standard part and self-made part, and the order of each part It expresses the generation sequence of each part during product design, that is, the design process Model Topology expresses the geometric topology information of the instance, and Assembly is the assembly related information of the product instance;

The helical gear example library is divided into a standard part example library, a tooth profile modification example library, a tooth direction modification example library and a deformed gear example library;

Step 4. Establish a helical gear rule library. The helical gear rules mainly include the design rules of helical gears, which are divided into rule serial number, condition part and conclusion part;

Step 5, helical gear retrieval, as shown in Figure 4, the instance index is designed based on the CBR theory, and according to the design characteristics, product characteristics and processing characteristics of the instance information, the index vocabulary of the helical gear is introduced according to the feature information, see Table 1;

Table 1 Example Index Vocabulary

The advantage of the present invention compared with prior art is:

(1), the present invention shortens the product design and development cycle, adapts to the rhythm of the rapid development of modern product design, and provides a reference for the development of new products;

(2) The present invention combines data mining technology to make the design more efficient and accurate, make the information more comprehensive, and avoid unnecessary waste.

Description of drawings

Fig. 1 is the realization flow chart of the method of the present invention;

Fig. 2 is a block diagram of a data mining platform in the present invention;

Fig. 3 is the classification of helical gears in the present invention;

Fig. 4 is a flow chart of CBR retrieval in the present invention.

detailed description

In order to clearly illustrate the technical features of the solution, the solution will be described below through a specific implementation mode and in conjunction with the accompanying drawings.

A kind of data mining method that the present invention is used in helical gear design, concrete steps are as follows:

Step 1. According to the mechanical industry standards and technical specifications, helical gear design rules and experience guidelines, and helical gear design expertise, classify helical gear design knowledge, including parameters such as helical gear basic design, mechanical characteristics, and manufacturing process;

Step 2. Build a data mining framework for helical gear design. As shown in Figure 2, the bottom layer of the system is a general system layer, which is mainly responsible for parametric modeling and finite element analysis of helical gears, and provides data query, storage and editing functions; the middle layer of the system For the CAD/CAM data mining layer, using data mining algorithms, based on the helical gear instance reasoning algorithm, according to the similarity analysis between each instance, the designer extracts the instance of interest, and analyzes the relevant design criteria and methods to provide a basis for the design of new products Provide design basis and store new design methods and rules into the rule base; the upper layer of the system is the design tool layer, which mainly guides designers to manage design knowledge, add, edit and delete operations, guide users to quickly complete design tasks, and improve design efficiency ;

Step 3: Establish a helical gear instance database, as shown in Figure 3, the expression technology form of the instance is the concrete embodiment of the product information model, and according to the hierarchical product information model, the composition of the helical gear instance is described as follows:

Case＝{ID, Code, Index, Matter, Model Topology, Assembly}

Among them, ID is the unique identification code of the instance, Code represents the code, Index is the instance index, the product instance content (Matter) is composed of the name, category, main rule parameters and storage path of each standard part and self-made part, and the order of each part It expresses the generation sequence of each part during product design, that is, the design process Model Topology expresses the geometric topology information of the instance, and Assembly is the assembly related information of the product instance;

An example in the helical gear tooth profile modification example library is expressed as follows:

Case1={Helical Gear_001, HGPM001, HGPM, m=0.5, z=34, α=20°, h_a=1, c=0.25, β=30°, tip=0, root=0, NULL, Precision=8, Material＝45,Path＝D:./HelicalGear/Data Mining,Profile Modification,NULL}

The helical gear example library is divided into a standard part example library, a tooth profile modification example library, a tooth direction modification example library and a deformed gear example library, see Table 1;

Table 1 Partial list of helical gear example library

Step 4. Establish a helical gear rule library. The helical gear rules mainly include the design rules of helical gears, which are divided into rule serial number, condition part and conclusion part;

There are many rules in helical gear design, some of which are shown below:

(1) If the designed gear is a modified gear, then output all modified gear information;

(2) If a helical gear with a helix angle of 30 degrees is designed, then output the information of all helical gears with a helix angle equal to 30 degrees;

(3) If the helical gear under heavy load conditions is designed, then output the helical gear information under all heavy load conditions;

(4) If the tooth surface hardness HB=350, then the main failure mode of the gear is pitting corrosion.

Step 5, helical gear retrieval, as shown in Figure 4, the instance index is designed based on the CBR theory, according to the design characteristics, product characteristics and processing characteristics of the instance information, according to the characteristic information, the index vocabulary of the helical gear is introduced, see Table 2; recently Neighbor instance retrieval is the simplest and most commonly used method in CBR.

Table 2 Example Index Vocabulary

Definition of similarity: use dist(A, B), sim(A, B) to represent the distance and similarity between A and B respectively, in the nearest neighbor instance retrieval, sim(A, B)=1-dist(A, B). The calculation of similarity is essentially a measure of the distance between attributes, then sim(A, B) should satisfy the following conditions and properties:

sim (A, B) ∈ [0, 1]; sim (A, B) = 1, if and only if A = B, namely reflexive; sim (A, B) = sim (B, A), namely Symmetry; sim (A, B) ≥ sim (A, C) + sim (C, B)-1, that is, the triangle inequality relationship.

Similarity calculation:

(1) Determine the similarity of the value sim(a, b)=1/|a-b|

(2) Similarity of fuzzy concepts

Map fuzzy concepts into numerical representations, such as process performance {excellent, good, medium, poor} and {1, 0.75, 0.5, 0.25} to establish a mapping relationship, so that fuzzy concepts can be converted into numerical values for similarity calculation.

Because each instance generally has multiple feature attributes, the similarity calculation is the overall similarity calculation, and the similarity calculation formula is as follows:

$\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; \&lsqb;</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; \&lsqb;</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; \&rsqb;</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>}{\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}$

Among them, a _i and b _i respectively indicate that the two instances of a and b have n feature attributes, and E _i indicates the important factors of each feature attribute, see Table 3.

Let the newly designed helical gear X (2,20,18,20,15,6, modified, left, inner), according to the description in Table 3, the helical gear modulus and meshing mode are mandatory attributes, so HG0010 And HG0086 does not participate in the instance test matching similarity calculation, calculated according to the above formula, the results are as follows:

Sim(X, HG0030) = 0.465

Sim(X, HG0050) = 0.754

Sim(X, HG0053) = 0.510

Then the instance HG0050 has the largest similarity, and it is the nearest neighbor matching object of the newly designed helical gear X.

Table 3 Helical gear characteristic parameter attributes

The present invention is not limited to the above-mentioned specific implementation methods, and changes, modifications, additions or substitutions made by those skilled in the art within the essential scope of the present invention should also belong to the protection scope of the present invention.

Claims (1)

1. the data digging method in helical gear designs, it is characterised in that realize step as follows:

Step one, according to mechanical industry standards and guidelines, helical gear design rule and Experience norms, helical gear design specialist Knowledge, classifies to helical gear design knowledge；

Step 2, building helical gear design data mining framework, system bottom is general-purpose system layer, main is responsible for helical gear ginseng Numberization modeling and finite element analysis, it is provided that the inquiry of data, storage and editting function；System middle level is CAD/CAM data mining Layer, utilizes data mining algorithm, and based on helical gear case-based reasoning algorithm, according to the similarity analysis between each example, designer carries Taking example interested, and analyze relevant design criterion, method, the design for new product provides design considerations, by new design Method and rule are stored in rule base；High system level is design tool layer, mainly guide design person's management design knowledge, add, Editor and deletion action, guide user to be rapidly completed design objective, improve design efficiency；

Step 3, setting up helical gear case library, the expression technology form of example is the concrete embodiment of product information model, according to point Layer passs the product information model on rank, and helical gear example composition is described as follows:

Case={ID, Code, Index, Matter, Model Topology, Assembly}

Wherein, ID is the exclusive identification code of example, and Code represents coding, and Index is example index, in Matter is product example Holding, product example content is by each standardized element and the self-control title of part, generic, main parameter of regularity and store path group Become, the genesis sequence of each part during product design of the sequential expression of each part, i.e. design process Model Topology table Reaching the geometric topology information of example, Assembly is the assembling conjunction information of product example；

Described helical gear case library be divided into standard component case library, flank profil practice Buddhism or Taoism case library, axial modification case library and deformation tooth Wheel case library；

Step 4, setting up helical gear rule base, helical gear rule mainly includes helical gear design rule, is divided into regular sequence Number, condition part and conclusion part；

Step 5, helical gear are retrieved, and example index is designed based on CBR theory, for design feature, the product of example information Feature and machining feature, according to characteristic information, release helical gear index word.