| MQ250 transmission has good power performance and fuel economy. It is a transverse-style manual gearbox, and It is widely applied in the Bora1.8T,Sagitar1.8T,Sagitar2.0L cars. There will be a noise on-line detection after the assembly, and the gear noise detection is an important part in the transmission noise tests. With the help of transmission noise detection test bench and gear automatic detection test bench, this dissertation has researched on gear noise and gear deviations. This dissertation analyzed the relationship of gear deviations and gear noise, through data analysis and test methods. The main content of this dissertation is as follows:1. Gear vibration and noise of the research status and significance.It researched on the transmission vibration and noise of the present situation. Analyzed gear vibration and noise of the present situation in domestic and foreign countries, and gear noise calculation and evaluation methods. Proposed the significance of research on gear noise. The text based on the actual situation of the adopted transmission,only researching the relationship of gear processing errors and noise.2. Study the vibration and noise of gear mechanismGear is an important component of transmission, and gears vibration and noise is the main source of transmission noise. The detection of gear noise is an important part in transmission noise on-line detection. This dissertation has studied the vibration and noise of gear mechanism, and put various specific gear processing errors as a key. Stiffness excitation related to contact ratio and gear's material. The selected material and contact ratio is basically determined in the actual production. Gear deviations related to processing technology. Gear deviations as a point on the research of gear noise is more important to reduce the noise of gearbox.3. Transmission gears noise test bench researchThis dissertation has introduced the component of transmission noise test bench and working principle in detail. This test bench can real-time monitor vibration of each shaft of gearbox, and show the change of gear noise. When the gearbox did not past the noise detection test, the test bench would alarm and show fault diagnosis report. All the data are automatically stored in the database after testing. Using these data can generate a three-dimensional order map which is used to analyzed transmission noise. And the map can display the relationship among order, speed and noise value. Furthermore, using these data can generate a two-dimensional map between the speed and the noise, or get a spectrum diagram between a rotating speed and order. In this paper, a standard of the noise detection limit was set to judge the actual measured noise whether accord with the requirement.Referred to standard chart of transmission speed in the type of HJK and HDR which was made by Germany experts in Volkswagen Co., Ltd. Noise limit standards of MQ250 JQP 1.8L transmission has adopted in this dissertation , which was used to extract the data of gear noise values in specific speed.Gear parameters can be detected automatically with the gear test bench. Test results will generate statements, in which test results can be used to numerical analysis. This article focused on the influence of gear noise parameters, such as profile slope deviation, helix slope deviation, profile form deviation, helix form deviation, single pitch deviation, radial jumped error, base pitch error, total profile deviation, total helix deviation, pitch accumulated deviation.4. Single gear deviation factor and gear noise experimental researchIn order to research the relationship between the single gear deviation factor and gear noise, this dissertation has introduced "standard gear" and "gear with errors" in the meshing. The so-called "standard gear"is a gear with the small errors whose processing precision is high. The so-called " gear with errors"is a gear with the larger error value which error is subject research, and other error parameters value control in a small range. This approximation consider that other parameters were unchanged, choosing different gears to study a single gear deviation factor how to impact the noise.Through the experimental research, analyzed the factors of the influence of gear noise, such as profile slope deviation, helix slope deviation, profile form deviation, helix form deviation, single pitch deviation, radial jumped error, base pitch error and pitch accumulated deviation. Experimental results displayed how much these deviation parameters effected gear noise. And researched on the mechanism about these gear deviation parameters how to generate noise.5. Analysis of gear noise prediction model including gear deviationCarry on gear noise multivariate linear regression analysis with gear deviations, such as profile slope deviation, helix slope deviation, profile form deviation, helix form deviation, single pitch deviation, radial jumped error, base pitch error, pitch accumulated deviation. Based on the test of the statistics model, it was found that the noise prediction model of gear has existed multiple linear relationships, and radial jumped error and single pitch deviation were negative to affect noise. This phenomenon did not conform to the real situation. Therefore consider to keep the original variables, and re-establish a prediction model between gear deviations and noise value.There are many kinds of methods to eliminate the linear relationship between gear deviations. In this paper, using ridge regression analysis method to solve the least squares estimates of linear error estimation by dependent coefficient. The re-established prediction model not only eliminated multiple linear problems, but also reserved all the gear error parameters. The ridge regression gear model passed the statistics test, and the prediction model conformed to the real situation.The coefficient of each parameter reflects the impact of the deviation on the noise in the prediction model. Profile slope deviation and helix slope deviation have the biggest influence on gear noise; Profile form deviation, helix form deviation, base pitch error and pitch accumulated deviation have also a great influence. These conclusions are in accordance with the experimental results.The prediction model can be used to prediction of gear noise. The forecast results compared with the experimental results, the results showed that prediction value fluctuated at a small range of the true value. The accuracy of the model was verified. It means that the model can truly reflect the value of gear noise.6. ConclusionHas given the full text main conclusion, and has carried on the forecast to the next research work of the relationship between gear deviation and noise, proposed present's research direction.
|
PDF Full Text Request