The hottest oil pump design research based on know

Research on oil pump design based on knowledge and visualization technology (Part 2)

(3) select the corresponding discovery methods and algorithms, such as classification, regression, etc

(4) implement the discovery algorithm to obtain the association relationship and reliability between the above databases

(5) manage and maintain knowledge, such as the conflict of inspection knowledge

knowledge discovery in the process of oil pump design is an attribute oriented operation using statistical technology and database technology. It is an activity of understanding and discovering the relationship between attributes. Among them, there are mainly methods such as cutting branches, merging branches, finding cause and effect and finding correlation. The other is the preparation of urea modified phenolic foam. Pruning is to prune the attribute domain that has no contribution to the discovery task or has a very low contribution rate; Merging is to carry out principal component analysis and synthesize similar attributes; Finding cause and effect is to conduct regression analysis on the attributes represented by numerical values; Finding correlation is to conduct factor analysis to find dependencies between attributes with disordered and discrete values. In the research and Prospect of raw plastic modified concrete in data mining of A-type pump design process, it is found that the fuel injection flow coefficient is related to the main engine supporting factory. The coefficient of a factory in Shandong is lower than the average value, while that of a factory in Dalian is higher than the average value. This coefficient is closely related to the manufacturing accuracy of the orifice diameter of the supporting factory. It is the comprehensive embodiment of the manufacturing method, manufacturing process and manufacturing equipment of the factory

4 injection performance simulation

the injection process of the oil pump has complex dynamic effects, which reflects the generation, propagation, reflection and superposition of pressure waves in the high-pressure oil circuit, so the injection performance is affected by many factors. If we rely on the data of observing the calculation results of injection performance, it is difficult to find the accurate law. Therefore, using animation technology, the spray performance results under various factors are reproduced, and the spray process is reflected by curve description and animation display, which is conducive to the further understanding of the spray law by the test personnel after years of development, and is conducive to the determination of the parameters of the design scheme by the designers, so as to reduce the manufacturing of physical prototype

some simplifications are made in the mathematical model of injection process, such as ignoring the elastic deformation of parts under the action of oil pressure or impact; It is considered that fuel density, flow coefficient and pressure wave propagation velocity are constant values; The influence of oil flow inertia, leakage at the precision coupling and spring vibration is not considered [4,5]. The model is expressed as a set of differential equations

the three links in the visualization process are: ① in the simulation of oil pump performance, data manipulation is mainly to select different time steps to solve the differential equations and obtain various laws of injection performance; ② Visual mapping is to convert the calculation results into displayed numerical quantities, such as the displacement of the plunger, outlet valve and needle valve relative to the cam angle, and the quantity of the fuel injection law, pump end pressure, nozzle end pressure, total fuel supply, total fuel injection, fuel supply law, and the cam angle relative to the flow section of the outlet valve; ③ These quantities are represented in the form of curves and animation [6]

by solving the differential equations, many relationships reflecting the injection performance can be obtained. These performance calculation results can also be represented by animation simulation results (as shown in Figure 5). It can animate the up and down movement of the plunger caused by the rotation of the cam (omitted in the figure), resulting in oil pressure changes, outlet valve lift changes, needle valve lift changes, starting and ending points of fuel injection, and five main fuel injection performance curves

Figure 5 fuel injection performance simulation animation interface

5 fuzzy evaluation and fuzzy redesign of injection performance

whether caxd (computer aided x-application design), DFX (Design for X-Function), constraint based design method or case-based reasoning technology, design scheme evaluation, multi scheme sequencing and redesign are finally required. In the design stage of oil pump scheme, the injection performance is evaluated

the fuzzy evaluation process of injection performance is:

(1) the factor set of the evaluation object? U = {u? 1, u? 2, u? N}, here, u? i(i=1,2,n)? They are pump end pressure, nozzle end pressure, injection rate, injection starting point and injection duration respectively

(2) evaluation set? V={v?1,v?2,v?m}。 The evaluation set is a set of grades. Here, v? i(i=1,2,m)? They are very good, good, average, poor and very poor respectively

(3) the factor weight set corresponding to the index is? A＝{a?1,a?2,a?n}。 Here, a? I (i=1,2, n) are u? I the importance of this factor set, and the weight set is generally given by experts. With R?? ij? Indicates u? I rated v? J, the fuzzy comprehensive evaluation model of the design scheme can be obtained

(4) after the weight reset and synthesis operator are determined, the design scheme can be evaluated. If the evaluation result is not satisfactory, redesign is required

the key of redesign is how to use failure information and related knowledge to feedback and complete local and global redesign tasks. The essence of redesign is to solve three problems: the selection of modified parameters, the determination of modified parameter values, and the decision-making of backtracking levels. Due to the complexity of the injection model, it is difficult to implement the redesign strategy of sensitivity and relevance. However, the engineering community has accumulated a lot of experience about the relationship between performance and structural parameters, which is conducive to the application of fuzzy reasoning in redesign

the fuzzy reasoning of oil pump redesign is multidimensional and multiple. Here, the generalized first search strategy is used for forward reasoning. According to the evidence that the uncertainty of the system input is actually not inferior to the past, using the uncertain knowledge in the fuzzy knowledge base and fuzzy database, the parameter modification suggestions with a certain degree of reliability are given according to a certain fuzzy reasoning strategy. The fuzzy redesign expert system uses the relational database structure to describe the modification properties of the designable parameter Xi, the performance evaluation factor UJ and the rule reliability eij. Its inference rule base is represented by relational database

considering the correlation between rules, let the synthesis operator be DC, the propagation operator be DT, the activation threshold of each rule be j, the acceptance threshold of each parameter modification conclusion be I, and let E J be the reliability of each premise. During fuzzy redesign, the reasoning of parameter Xi in each rule is:

set Ti as the final conclusion of the change of parameter Xi, and the reliability is:

its reasoning rule is that when the indicator UJ is triggered, the record j+1 in the rule base works (represented by RJ = 1, otherwise RJ = 0)

from this, the change trend of each modified parameter and the reliability of the modified conclusion can be obtained. When the reliability of several reasoning conclusions is greater than the adoption threshold, the control principles of modifying parameter selection should be considered, such as the principle of conclusion reliability, the principle of manufacturing resources, the hierarchical principle of modifying parameters, the principle of validity, the principle of continuous selection restriction and the principle of negative effect [7]

the process of fuzzy redesign is: ① the selection of modified parameters can be determined by fuzzy reasoning and selection principles; ② The modified parameter value is determined by the reliability of the modified conclusion and by referring to the negative effect principle; ③ The backtracking level of modified parameters is determined by the hierarchy of parameters. The pre check principle can also be used in the redesign process, that is, before the design parameters are traced back, the constraint pre check and target pre check related to the parameters can be carried out. In case of constraint conflict or target dissatisfaction, the redesign parameters can be re selected