Study On Gradual And Dynamic Reliability Analysis Method For Key Components Of A Mine Hoist Based On The Linear Moment

The key components of a hoist mainly include the main shaft,the main bearing and the reducer,whose design reliability directly affect the safe and efficient operation of the hoisting system.The failure modes of the main shaft,the main bearing and the reducer are various,and their corresponding reliability problems belong to gradual reliability problem,dynamic reliability problem and gradual and dynamic coupling reliability problem.The difficulty of gradual reliability analysis for the main shaft is the static reliability of the main shaft under a limited number of samples.That is the static reliability of the main shaft for a hoist at a potential risk value when the damage of the main shaft of a hoist degenerates to a potential risk value and only a limited number of samples of variables can be obtained.The difficulty of dynamic reliability analysis for the main shaft is probability space partition under a limited number of samples.That is the partition for probability space composed of a limited number of samples.Therefore,it is crucial for the key components of a hoist to study the robust gradual and dynamic reliability analysis methods for a limited number of samples.Based on the linear moment and the linear comoment,the thesis proposes a highorder linear moment-based static reliability method and a high-order linear momentbased probability space partition method,and analyzes the gradual and dynamic reliability of the key components of a hoist.The study work mainly includes:(1)For the trouble of reliability analysis under the lack of probability information,a high-order linear moment-based static reliability method and a high-order linear moment-based probability space partition method are proposed.The probability distribution modeling method based on the linear moment and the linear comoment is studied.According to the relationship between the linear comoment correlation coefficient and the Pearson linear correlation coefficient,the completely monotonic expression of Pearson linear correlation coefficient is derived.The probability distribution model based on the linear moment and the linear comoment and the completely monotonic expression of Pearson linear correlation coefficient are integrated into the FORM,and then a high-order linear moment-based static reliability method is proposed.The probability distribution model based on the linear moment and the linear comoment and the completely monotonic expression of Pearson linear correlation coefficient are integrated into the probability space partition method with complete information,and a high-order linear moment-based probability space partition method is proposed.The robustness and accuracy of the high-order linear momentbased static reliability method are verified by using three examples of the linear limit state equation,the strongly nonlinear limit state equation and the highly dimensional limit state equation respectively.The effectiveness of the high-order linear momentbased probability space partition method is verified by using two examples of the nonnormal probability space and the highly dimensional probability space respectively.(2)Study on gradual reliability analysis method for main shaft of a hoist based on the linear moment is carried out.The random response modeling of the main shaft for a hoist is performed using Latin stratified sampling based on the linear moment.Considering the influence of average stress effect and stress loading sequence on structural fatigue damage,the probabilistic model of residual strength of the main shaft for a hoist is constructed.Based on the NARX neural network and the active learning Kriging,a high-fidelity adaptive two-level dynamic metamodel of the main shaft is established.Using the high-order linear moment-based static reliability method,the gradual reliability analysis of the main shaft is conducted,and the conclusion is drawn that "the ratio of the reliable service life of the main shaft for a hoist under the different shifts and the ratio of daily working time under the different shifts are reciprocal".The highly reliable shift of the multi-layer winding hoist is compiled.The usage of highly reliable shift is described in detail through an example.(3)Study on dynamic reliability analysis method for main bearing of a hoist based on the linear moment is carried out.A new technical strategy of twin data-driven design reliability analysis of random vibration system for main bearing of a hoist is put forward,which involves dynamics model of multi-scale coupling system for rolling element bearing-probability density evolution of vibration acceleration-stochastic process modeling of probability density evolution path-probability distribution of vibration power spectral density-calculation of design reliability based on conditional probability.Based on the macroscale system dynamics equation of rolling element bearing and the mesoscale sliding wear equation of the material,a dynamics model of macroscalemesoscale coupling system of the vibration state evolution of the main bearing for a hoist is established.By adopting the high-order linear moment-based probability space partition method and the condition data to drive the dynamics model of macroscalemesoscale coupling system of the main bearing for a hoist,the probability density evolution of vibration acceleration of the main bearing for a hoist is studied.Based on the probability density evolution of vibration acceleration of the main bearing for a hoist and Karhunen-Loève expansion,a modeling approach of non-stationary random process for vibration acceleration of the main bearing for a hoist is proposed,and then the twin data of random sequence of vibration acceleration of the main bearing for a hoist is obtained.The probability distribution of vibration power spectral density for main bearing of a hoist is studied,and the reliability index of main bearing for a hoist within the service time is calculated.(4)Study on reliability analysis method of the gradual and dynamic coupling system for the reducer of a hoist based on the linear comoment is carried out.The statistical characteristics of random response of the planetary drive system for a reducer is studied.The probabilistic model of fatigue damage accumulation of the tooth root strength for the planetary gear is established.Gradual reliability of the first exceeding failure of the tooth root strength of the planetary gear for a reducer is evaluated.The bending-torsion coupling dynamics model of the planetary drive system for a reducer is constructed.The probability density evolution of the tooth root bending stress of the planetary gear for a reducer under random parameters is studied.Dynamic reliability of the first exceeding failure of the tooth root strength of the planetary gear for a reducer is evaluated.The gradual and dynamic coupling reliability of the reducer for a hoist is evaluated based on linear comoment.