| Gear transmission system is one of the key components of the wind power generation equipment. Different from the common gear drive, the wind power gear drive has three significant features:firstly, the input load is in characteristic of low speed and large torque; secondly, unsteady gear load caused by the fluctuation of wind speed; thirdly, very scarce effective gear samples for the statistical inference of reliability. So, the research on its dynamic reliability is of great theoretical value and practical significance. For this reason, this dissertation takes the gear transmission system in a wind turbine for an example. In order to form a complete research system and carry out further explorer with consideration of their own characteristics, reliability of the transmission gear is focused on and research results of other transmission parts are combinedthe on basis of domestic and foreign researches. Here are the main works:1) For the characteristic of variable wind speed, a probability model is presented for the gear stress under the condition of variable wind speed. The probability model of the blade load is studied. In addition, the probability relationship between the fluctuation characteristic of gear stress and the disturbance of wind speed is deduced under the condition of non-constant external load. This model fully considers the random feature of mutational wind speed and provides basis for the calculating of the prior data of gear stress.2) For the characteristic that the prior data of gear strength used for data fusing is lacking under the conditions of small samples, a model of multi-source data translation is presented. The gear strength data of4similar factors (parameters, structure, working condition, material) is selected from the historical test data. The gray correlation theory is improved and the weighing of each factor is taken into account. This model overcomes the bottleneck that the statistical inference of reliability can't be carried out on the condition of small samples and provides basis for the calculating of the prior data of gear strength.3) In order to establish the time-varying paths for wind power gear stress/strength, a Bayes data fuse model is presented. In the premise that prior values and theoretical values are acquired, the Bayes data fuse model is used to estimate the parameters of time-varying path for gear stress/strength. On this basis, a stress-strength interference model is established and the dynamic reliability curve of wind power gear is solved. The data fuse model combines the prior and observed information. It realizes update and expansion of the gear data. At the same time, the hypothesis testing guideline ensures equivalence of the prior data.4) In order to solve dynamic reliability of the whole gear transmission system, dynamic reliability researches of this dissertation and those of other transmission parts are combined to establish the dynamic fault tree model. For the purpose of simplifying the dynamic fault tree model, it is divided into several subunits and basic feature events are defined to generalize the probability characteristics of all basic events. So, the solution strategy is fixed on and the dynamic reliability curve of gear transmission system is solved. Meanwhile, the objectivity of this solution strategy is verified by comparing.5) In the end, in order to prove the objectivity, scientific, systematic of this work, an engineering case of1.5MW wind gear transmission system is selected to apply all the theory and methods of this dissertation. Establishing the statistic and analysis model for wind speed; establishing the data fusion model; establishing the dynamic fault tree model and solving it. Finally get the dynamic reliability curve of gear transmission system and make a deep analysis. The engineering case indicates that the methodology can not only provide the design reference early in the design, but also dynamically predict and track the reliability of gear transmission system.
|
PDF Full Text Request