Abstract: A design method of oil pump based on knowledge and visualization technology is proposed. In the program design stage, either instance-based design method or model-based design method may be used. The design method of the instance will use a large number of program design case bases to support the design of the main parameters. The design of the model-based solution is to utilize the long- Design model for design. Due to the complexity of the fuel injection system, there are a large number of empirical coefficients in the design model. We use the database knowledge discovery technology, mining design knowledge from the previous design tasks, product structure and commissioning evaluation database to find out the design rule of choice. The process of fuel injection is animated, the result of injection performance is evaluated by fuzzy evaluation technique, and the redesign is assisted by fuzzy redesign expert system. A constraint-based approach to product family design is proposed to support detailed design and evaluated by assembly simulation. This article has studied these key technologies, developed the corresponding software, and got the application in the factory.
Key words: computer aided design; simulation; instance-based design; knowledge base data discovery; fuzzy reasoning CLC number: TP391.72; TK242.8 Document code: A.
A design method of oil pump based on knowledge and visualization technology
At present, the scientific research units in the oil pump industry focus on the basic research on the injection mechanism and the test mechanism. The production units focus on the product imitation and modification, and seldom study the design and development process of the oil pump products. Most domestic manufacturers of new products from design to stereotyping experienced a long time, resulting in long product development cycle, poor quality.
In view of the shortcomings in the product development process of the oil pump production industry, we put forward the oil pump design method based on knowledge and visualization technology (Figure 1). By improving and perfecting the design process of oil pump products, this method can make the product development process model cover user needs, preliminary design, injection performance analysis, detailed design and assembly analysis.
As can be seen from Figure 1, in the design stage, the design of the main key parameters can be based on instance-based design method or model-based design method. The former will make full use of the program design case base, retrieve the program according to the appropriate retrieval method and change the instances according to the demand changes; the latter design the different empirical coefficients based on the mathematical model. After the program design, the injection performance simulation can be carried out, and the modification of some main parameters is decided by the performance evaluation and redesign system. Detailed design, mainly through the constraint-based product family design method, the structural design of all components and all parts of the parameter design, you can perform assembly simulation analysis. Various requirements to meet, you can get more satisfactory digital products optimized pump program.
Application of simulation technology and artificial intelligence technology, will make product development to the height of the "virtual product development", and in the design phase can consider the design of many activities in the post-event, thereby shortening the product development cycle and improve the test success rate. 2 based on the example of the pump design
Since the domestic production of oil pump nipples in the 1950s, we have gone through a road from imitation to retrofit technology development, which takes full advantage of design experience and examples. Therefore, Case-Based Design (CBD) More suitable method. The design process is based on design tasks, selected from the prototype prototype instance of the prototype and generated to the current work area [1,2] . Instance Prototypes Retrieve similar instances from instance space based on their instance retrieval model. If a similar instance is found, it is extracted and mapped to a similar target scenario; if it is verified that the target scenario fully satisfies the current design task requirements, the instance can serve as the final design match the design task and store in the instance repository; if not, Design task requirements, then analyze it and modify it until all design requirements are met. Thus, the design of instance-based is divided into instance expression, similar instance retrieval, instance modification, instance verification and so on.
The design example consists of the data of the design instance, the solution knowledge of the instance and the index of the instance. Oil pump design uses object-oriented knowledge representation method, that is, a variety of single knowledge expression methods (rules, frameworks and processes), in accordance with the principles of object-oriented design to form a mixed knowledge representation. With the pump as the center, the static attributes and dynamic behavior characteristics, as well as the design process knowledge, are "encapsulated" in the structure of the presentation object. This method allows the complex oil pump object to be decomposed into several simple objects and becomes a tree structure (as shown in FIG. 2). At the same time, a complex example can be gradually broken down into simpler examples (Figure 3).
![Illustration 2]()
Figure 2 injection system tree Figure 3 Example of the decomposition of the structure
There are two cases of decomposition: ① If the detailed design is supported, it can be decomposed to the part level, but the structure of the sample library may be huge and complex. ② If the design is supported, it can be decomposed into the structure that can carry out the design evaluation (injection performance evaluation) Features and main features of a key parameter level.
The extraction of the best example is to extract in the instance library the instance or instance fragment whose characteristics are most similar to the design task characteristics. Design tasks include design goals, constraints, and initial conditions. The search for similar instances is actually a decision in multi-attribute space, so the weighted sum based on index features is a typical search strategy. When an instance is represented as a collection, the distance between instances can be defined as the intersection of two instances; when an instance is represented as a state space vector, the distance between instances can be defined as the distance between vectors on the algebra. All three belong to the nearest neighbor search strategy. When the design examples have enough time, can also be based on neural network extraction strategy. The neural network-based extraction strategy can eliminate the interference of human factors through the training of multiple successful and failed instances, and has the ability of self-adaptation and self-learning.
3 Design knowledge from design process data mining
When the similarity of design examples is relatively low, the design method of CBD is not applicable, and the mapping relationship reflecting the design experience between performance, structure and parameters is included in the example organization, which is not conducive to the accumulation and dissemination of design experience. Product family-based design reflects design knowledge from constraints and can not guide the design process. Therefore, it is necessary to seek a scientific calculation model and simulation model to improve the product's self-development capability. Due to the particularity of diesel engine, its performance is good or not depends on the match between fuel injection, air intake and combustion, so there are many empirical formulas and empirical coefficients for design calculation model and injection performance model. Here, the use of database knowledge found
(Knowledge Discovery in Database, KDD) technology principles, from the previous product instance library, test conclusion library for product features, design intent and design knowledge.
After years of extensive application of computer aided design, production, management and database technology, product development department has recorded a large number of database related to product development, such as design task feature library, product list and pump commissioning evaluation documents. In order to convert the latent knowledge in the design process into explicit knowledge, the author uses the database knowledge discovery technology to excavate the design knowledge. KDD is a technology that obtains correct, novel, potentially useful and finally understandable patterns from the database [3] .
There are many empirical coefficients in the pump design and calculation model. The selection of these factors is directly related to the performance of the product and the scale of the design. Using the design task feature library, product specification, commissioning evaluation results and design calculation model, we recalculate all kinds of past design examples to get the design task feature library, the design process experience coefficient library corresponding to the task features, the main key parameter database and Design program evaluation conclusion library. KDD method through the design knowledge mining, oil pump design knowledge (Figure 4). The steps of data mining in the pump design are:
(1) Select and prepare the data to be excavated. According to the existing product design requirements, the main key parameters and test conclusions, the design experience coefficient of each product instance is inverted;
(2) data preprocessing. Establish a relational database to describe the product design requirements, the main key parameters and test conclusions, the other to build a database to describe the design experience coefficient of the corresponding product instance, to reduce the complexity of the data and to reorganize the data through purification, reduction, transformation, classification;
(3) Select the appropriate discovery methods and algorithms, such as classification, regression, etc;
(4) implement the discovery algorithm to get the association and credibility between the above databases;
(5) management and maintenance of knowledge, such as testing knowledge of the conflict.
Knowledge discovery in the process of oil pump design is the use of statistical techniques and database technology for the operation of the attribute-oriented, is the relationship between the attributes of awareness and discovery activities. Among them, there are mainly methods of cutting branches, branches, causalities and finding related ones. Shear branch is cut off to the task of finding no contribution or contribution rate is very low attribute domain; and branch is to carry out principal component analysis, the similar properties to be integrated; to find causal is the numerical representation of the property regression analysis; to find the relevant is to carry out the factor Analysis, looking for dependencies between disordered and discrete attributes. In the A-type pump design process data mining, found that the fuel injection flow coefficient and the host supporting plants, a factory in Shandong coefficient is lower than the average, while a factory in Dalian is higher than the average. This factor is closely related to the manufacturing precision of the orifice of the injection hole of the supporting plant. It is a comprehensive embodiment of the manufacturing method, manufacturing process and manufacturing equipment of the factory hole machining.
4 jet performance simulation
The fuel injection process of the pump has complex dynamic effects, which reflect the generation, propagation, reflection and superposition of pressure waves in the high-pressure oil circuit. Therefore, the injection performance is affected by many factors. If you rely on the observed data of jet performance calculations, it is difficult to find the exact law. To this end, the use of animation technology, the performance of a variety of factors reproduce the jet performance curve to describe and animated display reflects the injection process is conducive to test personnel to further understand the injection law is conducive to the designer of the design parameters Determine, thereby reducing the manufacture of physical prototypes.
Some simplifications are made in the mathematical model of the injection process, such as ignoring the elastic deformation of the part under the action of hydraulic pressure or impact; the fuel density ρ, the flow coefficient and the pressure wave propagation velocity α are considered as normal values; irrespective of the fluid inertia , The presence of leaks at the precision coupling and the influence of spring vibration [4,5] . The model appears as a set of differential equations.
The three aspects of the visualization process are as follows: ① In the performance simulation of oil pump, the data manipulation mainly takes different time steps to solve the differential equations and get the law of injection performance; ② The visualization mapping converts the calculation result into the displayed value Such as the displacement into the plunger, the output valve and the needle relative to the cam angle of displacement, into the law of injection, the pump pressure, nozzle pressure, the total amount of fuel, fuel injection, fuel supply law, Outlet valve flow cross-section relative to the amount of cam angle; ③ to curve and animation of these values [6] .
By solving the system of differential equations, many relationships reflecting jetting performance can be obtained. These performance results can also be animated simulation results (Figure 5). It can be animated to show that with the rotation of the cam (omitted from the figure), the plunger is caused to move up and down, resulting in a change in the hydraulic pressure, a change in the output valve lift, a change in the lift of the needle valve, a start point and an end of the injection Point, and five main fuel injection performance curve.
5 jet performance of the fuzzy evaluation and fuzzy redesign Whether it is CAxD (Computer Aided x-application Design) or DFx (Design for x-function), whether it is based on the constraint-based design method, or instance-based reasoning technology, the final design plan should be evaluated, multi-program sort And re-design. In the pump design stage, is to evaluate the injection performance.
The fuzzy evaluation of injection performance is as follows:
(1) The set of factors of the evaluation object U = {u 1, u 2, ..., u n}, where u i (i = 1,2, ..., n) , Tip pressure, injection rate, injection start and injection duration;
(2) The evaluation set? V = {v? 1, v? 2, ..., v? M}. The evaluation set is a set of levels, where v i (i = 1, 2, ..., m) are respectively good, good, fair, poor and poor.
(3) The set of weight factors corresponding to the indicator is A = {a? 1, a? 2, ..., a? N}. Here, a i (i = 1, 2, ..., n) are respectively the importance of u i in this factor set. The weight set is generally given by the experts. Taking r ij i as the membership degree of u i as v j, the fuzzy comprehensive evaluation model of design scheme can be obtained.
(4) After the weight set and the synthesis operator are determined, the design plan can be evaluated. If the evaluation results are not satisfied, we must redesign.
The key to re-design is how to use the failure information and related knowledge to feedback and complete the re-design of local and global tasks. The essence of re-design is to solve the three options of modifying parameters, modifying the value of parameters and making decisions on the back-level. Due to the complexity of the injection model, it is difficult to implement the re-design strategy of sensitivity and the re-design strategy of the correlation. However, the engineering community has accumulated a great deal of experience in the relationship between performance and structural parameters, which is conducive to the application of fuzzy reasoning in redesigning.
Oil pump redesign fuzzy reasoning is multidimensional, multiple fuzzy reasoning, where the generalized first search strategy for forward reasoning. According to the uncertainty evidence input by the system, some uncertainties in the knowledge base and fuzzy database are used to give some suggestions on how to modify the parameters according to a certain fuzzy inference strategy. Fuzzy redesign expert system uses relational database structure to describe the modifiable nature of the designable parameter xi, the performance evaluation factor uj and its rule credibility eij. Its reasoning rule base is represented by a relational database.
Considering the correlation between the rules, let the operator of synthesis be Dc, the operator of propagation be Dt, the activation threshold of each rule is αj, the acceptance threshold of each parameter modification conclusion is βi, and let e'j be the credible of each premise degree. Fuzzy redesign, the rules in the parameters xi reasoning is:
Thus we can get the change trend of each modified parameter and the credibility of the modified conclusion. When there are several reasoning conclusions, the credibility is greater than the adoption threshold, we must consider the control parameters to modify the choice of parameters, such as the conclusion of the credibility of the principle of manufacturing resources, modify the parameters of the principle of the principle of validity, continuous selection Limit principles and negative effects of such principles [7] .
The process of fuzzy redesign is as follows: ① The choice of modifying parameters can be determined by the principle of fuzzy reasoning and selection; ② The value of modified parameters is determined by modifying the credibility of the conclusions and referring to the principle of negative effects; ③ The parameter- Sexually determined. In the process of re-design, the principle of preflight can also be used, that is, before the design parameters are backdated, the constraint pre-check and the target pre-check related to the parameter are performed. If the constraint conflict or the target is not satisfied, the re-design parameters may be reselected.
6 Constraint-based product family design
Constraint-based product family design is an effective method in the development of variant products for oil pumps, especially in detailed design phases. It takes advantage of modularity, configurability, serialization of products, and economies of scale and scale. Constraints of product families can be divided into multi-functional group constraints, multi-functional group constraints, designers belonging to the same division of labor, and constraints corresponding to individual designers. Constraints are handled through a constraint management system in the product development process that includes editing functions such as creating, deleting, reading, modifying, and writing constraints, as well as grammar and rationality checks on constraints. Oil pump product assembly and gene parts design is carried out on the Pro / E, product family design is developed in VB. In order to ensure seamless integration between variant part design and assembly design, the design drawings and dimensions of the product family are all from the product modeling in Pro / E. In particular, the characteristic labeling code exactly matches the gene part in Pro / E. The grammar check is to check if the constraint's edits match the algebraic expression, and the variables in the constraint must be the parameter codes for the genetic part in Pro / E. The plausibility check is to check if all editorial constraints are accepted by the size of the gene part.
Constraints create principles are:
(1) Design specifications. Product variants to meet the standards of product design, series and other requirements;
(2) product performance requirements. Variants of the combination of parameters to meet product performance and performance requirements;
(3) production resources. Product variants to consider the company's production resources for the production of flexible processes and flexible organizations accepted;
(4) design habits. Product variants to consider product design habits and product development strategy developers;
(5) gene products. The establishment of the constraint must be approved by the gene product.
Product family design flow can be described as:
① characterize the needs of users;
② according to the functional structure of the mapping relationship to establish product classification;
③ by the company's design capabilities, manufacturing capacity and production organization to establish product family;
④ configuration rules to determine the basic building blocks;
⑤ constraints by the product family to determine the characteristics and parameters of part variants.
The rules for configuration are: Compatibility in different building blocks, for example, interference between building blocks; Mapping between user requirements and design parameters.
7 Based on product family assembly simulation
Simulation-based assembly analysis is the test of the ease of assembly, LL, of assembly in a design and is a technique that optimizes the design of the product to achieve the lowest assembly cost. In the product design stage, the assembly performance of the product is evaluated by analyzing various factors that affect the assemblability of the product.
According to the three aspects of the visualization process, the data manipulation of assembly simulation is to determine the variation characteristics and parameters, and these parameters and characteristics are obtained from the product design and performance verification. The visual mapping maps these parameters and characteristics into the variation characteristic file, Parts; Drawing is the use of software in the assembly, the display function to reflect the modified parts after the assembly, and calculate the amount of interference in the assembly [6] .
Product variant features and dimensions are determined by the product design expert system, performance evaluation subsystem, and redesign expert system. Variations feature files are transmitted to the assembly simulation subsystem via the network for assembly simulation. Product family design and assembly simulation process shown in Figure 6.
![Illustration eight]()
Figure 6 product family design and assembly simulation process
8 Conclusion
This paper presents a design method of oil pump based on knowledge and visualization technology. In the stage of program design, the example-based design method or the model-based design method can be used to establish the injection model in animation form. The fuzzy evaluation technology is used to evaluate the program design. The redesign is assisted by the fuzzy redesign expert system. A detailed design is supported using a constraint-based product family design methodology and evaluated with assembly simulations. The CBD method uses a large number of program design case bases to support the design of the main parameters. KDD technology excavates design knowledge from previous design tasks, product structure and commissioning evaluation databases to find out the rule of selection of design experience coefficients to support model-based design methods .
The CBD method and KDD technology facilitate the expression, accumulation and dissemination of design experience. The injection process simulation and assembly simulation can reflect the complicated injection process and assembly structure, which is beneficial to the further understanding of the staffs of the pump products on the injection law, and facilitates the digital pre-assembly for the designers to reduce the manufacturing of the physical prototype. These key technologies and the corresponding software developed have been applied in the factory.
(CIM Institute, Shanghai Jiaotong University, Shanghai 200030, China)
references:
[1] Tsatsoulis C, Kashyap RL L. Case-based reasoning and learning in manufacturing with the TOLEC planner [J]. IEEE Transactions on Systems, Man, and cybernetics, 1993, 23 (4).
[2] Pwarce M. Case-based design support, a case study in architectural design [J]. IEEE Expert, 1992, 10: 14-20.
[3] Fayyad U, Uthurusamy R. Datamining and knowledge [J]. IEEE expert, 1996,10: 20-25.
Gu Hongzhong diesel engine working process [M] Beijing: National Defense Industry Press, 1987.
[5] Jiang Deming. Internal combustion engine principle [M] Beijing: China Agricultural Machinery Press, 1988.
[6] Jiang Zuhua. Virtual product development and its key technologies [D]. Shanghai: Shanghai Jiaotong University (Ph.D. dissertation), 1998.9.
[7] Jiang Zuhua, Yan Junqi, Fu Shiwei, etc. Research and development of fuzzy redesign expert system development tools [J]. China Mechanical Engineering, 1999,10 (3): 295-297.
Deep Groove Ball Bearings
623zz Bearing,Micro Bearings,Wheel Bearing
Angular Contact Ball Bearing,Tapered Roller Bearing,Thrust Ball Bearing Co., Ltd. , http://www.nsballbearing.com
