Body-based intelligent reasoning method for box plane processing scheme

Abstract

The invention belongs to the technical field of box part process decision and computer aided process design (CAPP), and relates to an intelligent reasoning method for a box plane machining scheme based on a body. The method specifically comprises the following steps that 1, a box plane machining scheme is constructed to generate a body; 2, an assertion formula set APS of the relation between the part and the machined face is constructed; 3, extracting feature information of the box type part and the machining surface of the box type part, and constructing a part design feature assertion formula set APD and a machining surface design feature assertion formula set AFD; step 4, establishing an SWRL inference rule generated by the box plane processing scheme; and 5, constructing an instantiated ontology knowledge base generated by the box plane processing scheme, and reasoning the instantiated ontology knowledge base by using a reasoning engine according to the SWRL reasoning rule base to generate the box plane processing scheme.

Description

Technical Field

[0001] The present invention belongs to the technical field of process decision-making for box parts and computer-aided process planning (CAPP), and relates to an intelligent reasoning method for the planar machining scheme of boxes based on ontology. Background Technique

[0002] As the basic parts of a machine, box parts have a relatively complex structure, mainly due to the thin and unevenly distributed box walls, as well as various geometric-shaped planes and numerous holes. These characteristic elements need to meet high-precision manufacturing requirements. In the field of traditional process design, the process planning of box planes often relies on designers to consult design manuals or accumulate personal experience. This method is inefficient, and due to individual differences in designers' understanding, it is easy to lead to uncertainty in design results, thereby affecting the consistency of process design. This current situation has become a key factor restricting the intelligent process of process decision-making for box parts.

[0003] Aiming at the problems of low efficiency and uncertainty in the manual decision-making of the box planar machining scheme and the difficulty in sharing and transmitting relevant process knowledge between heterogeneous CAPP systems, to provide the intelligence of the decision-making for the box planar machining scheme, the ontology technology is introduced into the decision-making of the box part planar machining scheme. First, according to the concepts and relationships in the decision-making process of the machining scheme, a meta-ontology model for the decision-making of the box planar machining scheme is constructed using the Ontology Web Language (OWL). Then, according to the part feature information and process knowledge, an inference rule for the box planar machining process decision-making is constructed using the Semantic Web Rule Language (SWRL). Finally, according to the constructed SWRL inference rule, an inference engine is used to automatically generate the box planar machining scheme. The realization of the automatic generation of the box planar machining scheme is conducive to the sharing and reuse of process experience knowledge related to the box planar machining scheme, and improves the intelligence of the decision-making for the part machining scheme. Summary of the Invention

[0004] The technical problem to be solved by the present invention is the low level of intelligence in the decision-making of the existing box part machining scheme. Aiming at this problem, the present invention proposes an intelligent reasoning method for the box planar machining scheme based on ontology. To solve the above problem, the present invention is realized through the following technical solutions, including the following steps: Step 1: According to the domain knowledge in the decision-making process of the box planar machining scheme, construct an ontology knowledge base system for the box planar machining scheme. The attribute relationships between classes in the knowledge base provide a consistent description of the domain knowledge and experience in the decision-making process of the box planar machining scheme. Step 2: Based on the ontology knowledge base constructed in Step 1, according to the domain knowledge in the decision-making process of the box planar machining scheme, construct a SWRL rule base for the intelligent reasoning of the box planar machining scheme. Step 3: For the decision-making example of the box body plane machining plan, establish a part drawing model and extract the part structure to obtain several surface information. Represent the constraint relationship between the box body part of the instance and its machining surfaces through OWL and establish an assertion formula set A PS ; Step 4: According to the part drawing model established in Step 3, extract the design feature information related to the box body plane machining plan and establish an assertion formula set A PD and the assertion formula set A FD represent the subordinate relationship between the part and these machining feature information; Step 5: According to the assertion formula set A established in Step 3 PD , the assertion formula set A FD and the part drawing model, instantiate the relevant concept and attribute information respectively, construct the classes, subclasses, individuals and the relationships between individuals in the box body plane machining plan generation instance, and obtain the instantiated ontology knowledge base for the box body plane machining plan generation; establish the SWRL rule base for the intelligent reasoning of the box body plane machining plan, and use the inference engine built into the ontology editing software Protégé 5.5 to reason about the instantiated ontology knowledge base obtained in Step 5 to generate the instance box body plane machining plan.

[0005] Step 1 includes: Step 1.1: According to the relevant machining process experience of the box body plane machining plan decision-making, construct the constraint relationship between the classes of the box body plane machining plan based on ontology and the inherent attributes of the classes, where the attributes of the classes include two categories: object attributes and data attributes; Step 1.2: Use the OWL2 language to construct the terminology axiom set TBOX of the box body plane machining plan based on ontology, and represent TBOX in the form of the sub-nodes of the parent class, the inherent attributes of the class and the semantic knowledge, so as to form the box body plane machining plan ontology knowledge base system.

[0006] The specific content of Step 2 is: Step 2.1: Construct the ontology model for the box body plane machining plan generation.

[0007] Step 2.2: Construct the inference rules for the box body plane machining plan decision-making.

[0008] Compared with the prior art, the present invention has the following characteristics: 1. The ontology knowledge base and the inference rule base for the box body plane machining plan decision-making inference can be shared and reused, ensuring the portability and scalability of the process knowledge.

[0009] 2. By using the domain knowledge of the box body plane machining plan and the corresponding constraint relationships, the hierarchical relationship of the intelligent reasoning of the box body plane machining plan based on ontology is made clear and can be better interpreted by the computer.

[0010] 3. A knowledge base description framework and an intelligent inference rule base for the intelligent reasoning of the box body plane machining plan based on ontology are constructed, providing a feasible method for the intelligent reasoning of the box body plane machining plan based on ontology. Description of the Drawings

[0011] Figure 1 It is the intelligent reasoning flow chart of the box body plane machining plan.

[0012] Figure 2 It is the hierarchical relationship of the decision-making classes of the box body plane machining plan.

[0013] Figure 3 It is the hierarchical relationship of the decision-making attributes of the box body plane machining plan.

[0014] Figure 4 It is the instance diagram of the box body part.

[0015] Figure 5 It is the inference result diagram of the machining plan. Detailed Implementation Manner

[0016] The following takes the box body part as an example for a brief description, as shown in the appendix Figure 4 shown. The intelligent reasoning process of the box body plane machining plan based on ontology is as shown in the appendix Figure 1 shown, and specifically includes the following steps: Step 1: Consult the relevant process knowledge in the field of decision-making design of the box body plane machining plan. According to the document materials, expert experience knowledge and reusable ontology in this field, obtain the domain knowledge of the decision-making design of the box body plane machining plan. According to the obtained domain knowledge of the decision-making design of the box body plane machining plan, construct the classes and their hierarchical relationships for generating the box body plane machining plan; first, define the important classes in the domain, as well as the class attributes and the relationships between classes. Classes are the core of ontology and are used to describe the concepts in the domain. Based on these explicitly defined classes and relationships, implicit knowledge can be obtained through a certain reasoning mechanism.

[0017] Some ontology classes and the hierarchical relationships between classes are as shown in the appendix Figure 2 , where owl:thing is an abstract class in the ontology editing tool Protégé and is the parent class of all classes; its subclasses Part, FDM, PDF, and PP represent the box body part, the design features of the box body part, the design features of the machined surface, and the machining plan respectively.

[0018] The subclasses of Part include machined surface (MP); the subclasses of FDM include part material (PM), heat treatment (HT), production type (TP); the subclasses of PDF include precision grade (PF), geometric constraint (JH), etc.; the subclasses of PP include flat surface machining plan (FMP), machining method (FMM).

[0019] In the field of decision-making for flat surface machining plans of boxes, attributes are an important part for building the ontology knowledge base. Concepts are difficult to fully express all the information content of the concept, so it is necessary to define the specific attributes of the concept; when constructing the ontology of the ontology language and tools, object-type attributes and data-type attributes are mainly defined; OWL is used to represent the ontology model and establish the attribute relationships between concepts.

[0020] The attributes of classes are used to describe the relationships between the same and different classes in the field of decision-making for flat surface machining plans of boxes; these attributes include object-type attributes (Object-Property) and data-type attributes (Datatype-Property). Object-type attributes are used to express the constraint relationships between classes, while data-type attributes are used to represent the inherent numerical characteristic attributes of classes, generally used to represent the numerical size and characteristics of specific classes, usually including integer (int) and floating-point (float), string (string) as shown Figure 3 below.

[0021] Step 2: Based on the ontology knowledge base constructed in Step 1, according to the domain knowledge of the decision-making process for flat surface machining plans of boxes, construct a SWRL rule base for intelligent reasoning of flat surface machining plans of boxes; the representative rules are as follows: Rule1-1: Part(?x) ^ RoughDatum(?y) ^ hasRD(?x,?y) ->hasRD(?x,Sh) Rule1-2: Part(?x) ^ PType(?y) ^ hasPT(?x,Lb) ->hasFD(?x,1f2h) Rules 1-1 and 1-2 describe the selection of positioning references during the flat surface machining of boxes. If the production type of the box part is large batch, then the rough reference is the spindle hole, and the fine reference is one plane and two holes.

[0022] Rule 2-1: If a part (?x) is made of gray cast iron (Gcr), is heat-treated by natural aging (Na), the machined flat surface (?y) has a surface roughness value_Ra(?y,?b) where swrlb:lessThanOrEqual(?b, 12.5) and swrlb:greaterThan(?b, 3.2), the accuracy class of the part is 11, has a geometric constraint of parallelism (PA), and the tolerance grade range is 6 - 9 (Value_TG(?y,?b) where swrlb:lessThanOrEqual(?b, 9) and swrlb:greaterThan(?b, 6)), then the machining plan for the surface is CMP2, and the machining methods are rough milling (Rm) and semi-finish milling (SFm) in sequence. Rule 2-1: Describe the selection of the machining plan for the box body plane. If the box body material is gray cast iron, the heat treatment is natural aging treatment, the surface roughness range of the machined surface is 3.2 - 12.5, the accuracy grade is 11, the geometric constraint is parallelism, and the grade range is 6 - 9, then the machining plan for the machined surface is CMP2, and the machining methods are rough milling and semi-finish milling in sequence.

[0023] Step 3: For the decision-making example of the box body plane machining plan, establish a part drawing model and extract the part structure to obtain the information of the machined surface. Represent the constraint relationship between the shaft parts of the instance and their respective machined surfaces through OWL and establish an assertion formula set A. PS 。

[0024] Assertion formula set A PS : A PS = { MachinedFlat (SA), MachinedFlat (SB), MachinedFlat (SC), MachinedFlat (SD)}; Step 4: According to the part drawing model established in Step 3, extract the part feature information related to the decision-making of the box body plane machining plan, establish the assertion formula set A CS and the assertion formula set A CF represent the subordinate relationship between the part and these machining feature information.

[0025] Assertion formula set A PD : A PD = { Parts(p1), hasPT (Lb), isPMof(Gcr), isHTof(Na)}; Assertion formula set A FD : A FD = {Value_Of_Ra(SA, 1.6), Value_Of_Ra(SB, 6.3), Value_Of_Ra(SC, 0.8), Value_Of_Ra(SD, 6.3), Value_IT(SA, f), Value_IT(SB, f), Value_IT(SC, f), Value_IT(SD, f), hasFPT(SA,FN), hasFPT(SB,PA), hasFPT (SC,SQ), Value_TG (SA, 7), Value_ TG (SB, 8), Value_ TG (SC, 11),}; Step 5: According to the set of assertion formulas A established in Step 3 PD , the set of assertion formulas A FD and the part drawing model, instantiate the relevant concepts and attribute information respectively, construct the classes, subclasses, individuals and the relationships between individuals in the generation instance of the box body plane machining plan, and obtain the instantiated ontology knowledge base for the generation of the box body plane machining plan; establish the SWRL rule base for the intelligent reasoning of the box body plane machining plan, and use the Drools inference engine to reason about the instantiated ontology knowledge base obtained in Step 5 to generate the plane machining plan for the instance box body parts.

[0026] To ensure the integrity of the decision-making ontology knowledge base for the box body plane machining plan constructed, all individuals belonging to a class need to be added to the already constructed class. Add individuals SA, SB, SC, SD under the class Part, add individuals St, Gcr under the class PM; add individuals Ma, Na under the class HT; add individuals Pa, Fl under the class JH; add individuals Finish-milling, FinishTurning, Grinding, RoughMill, RoughTurning, Semi-finishTurning under the class FMM.

[0027] According to the SWRL rule base for the intelligent reasoning of the box body part plane machining plan established in Step 2, use the Drools inference engine to reason about the instantiated ontology knowledge base obtained in Step 5. The reasoning process is implemented in the software protégé, and the reasoning result of the instance box body part plane machining plan obtained is added to the instantiated ontology knowledge base. The reasoning result of the machining plan is as Figure 5 shown.

[0028] The present invention uses OWL to define the ontology and reasoning rules for the intelligent reasoning of the box body plane machining plan. The various characteristic information of the box body parts and their machined surfaces are described by precision, material, tolerance grade, etc., and the design intention can be retrieved from the design characteristic information. Use the rule semantic reasoning mechanism to reason about the decision-making ontology knowledge base for the box body plane machining plan, identify the design intention, and thus improve the sharing and reuse of the decision-making model for the box body part plane machining plan.

[0029] The above content is a further detailed description of the present invention in combination with specific engineering examples. It cannot be determined that the specific implementation manner of the present invention is limited thereto. For those of ordinary skill in the technical field to which the present invention belongs, without departing from the concept of the present invention, several simple deductions or substitutions can still be made, which should all be regarded as belonging to the protection scope of the present invention.

Claims

1. An intelligent reasoning method for box plane processing scheme based on ontology, characterized in that: It consists of the following steps: Step 1: Based on the domain knowledge of the decision-making process of the box plane processing plan, a box plane processing plan ontology knowledge base system is constructed. The attribute relationship between classes in the knowledge base provides a consistent description of the domain knowledge experience of the decision-making process of the box plane processing plan; Step 2: Based on the ontology knowledge base constructed in step 1, and according to the domain knowledge of the decision-making process of the box plane processing plan, a SWRL rule base for intelligent reasoning of the box plane processing plan is constructed; Step 3: For the decision-making example of the box plane processing plan, establish a part drawing model and extract the part structure to obtain the information of the processed surface. Use OWL to represent the constraint relationship between the box part of the example and its processing surfaces and establish the assertion formula set A. PS ; Step 4: Based on the part drawing model established in step 3, extract the design feature information related to the box plane processing plan and establish the assertion formula set A PD and assertion formula set A FD Indicates the subordinate relationship between parts and these processing feature information; Step 5: According to the assertion formula set A established in step 3 PD , assertion formula set A FD and part drawing model, instantiate the relevant concepts and attribute information respectively, construct the classes and subclasses, individuals and the relationships between individuals in the box plane processing plan generation instance, and obtain the ontology knowledge base for the instantiated box plane processing plan generation; establish the SWRL rule base for intelligent reasoning of the box plane processing plan, use the Drools reasoning engine to reason with the instantiated ontology knowledge base obtained in step 5, and generate an instance box plane processing plan.

2. According to the ontology-based intelligent reasoning method for box plane processing schemes according to claim 1, it is characterized in that: In step 1, the top-level classes in the knowledge base for generating the box plane machining plan based on the ontology include: box parts, machined surfaces, part features, and machining plans.

3. According to claim 1, the intelligent reasoning method for box plane processing scheme based on ontology is characterized in that Step 1 specifically includes: Step 1.1: Based on the relevant mechanical processing experience of the box plane processing plan decision, the constraint relationship between the classes of the box plane processing plan based on the ontology and the inherent attributes of the class are constructed, where the class attributes include object attributes and data attributes; Step 1.2: Use OWL language to construct the box plane processing scheme terminology axiom set TBOX based on ontology, and express TBOX in the form of child nodes of the parent class, the inherent attributes of the class and based on semantic knowledge, thus forming the box plane processing scheme ontology knowledge base system.

4. According to the ontology-based intelligent reasoning method for box plane processing schemes of claim 1, it is characterized in that: Step 3 further includes the following process: using the reasoning ability of the instantiated ontology knowledge base itself to infer implicit relationships and verifying the consistency of the instantiated ontology knowledge base.

5. The intelligent reasoning method for box plane processing scheme based on ontology according to claim 1 is characterized in that: When the ontology-based intelligent reasoning method of the box plane processing solution is applied to the engineering example, the engineering case and the reasoning result can be added to the instantiated ontology knowledge base, which can realize the self-update of the knowledge base.

Images