Fault Analysis And Failure Rate Prediction Method For Bridge Crane

Bridge crane is to achieve the production process of mechanization and modernization as a special equipment, there is a wide range of applications in modern industrial, considering of its huge structure、complex composition、frequently using and more relevant operating workers, often caused great loss when facing an accident, therefore analysis of the bridge crane failure and prediction of the failure probability has great significance to prevent the occurrence of a fault and thus prevent accidents.Fault Tree Analysis is one of the most widely used method for failure analysis, the fault displayed clearly and concisely when using fault tree analysis for fault analysis on bridge crane failure by the systemic summary of fault type and overall summarization of the reason causing the fault, conduct a qualitative analysis of the fault. Analytic Hierarchy Process is a comprehensive evaluation method, which is used for failure analysis is based on fault tree analysis, draw the corresponding hierarchical model of the fault according to the fault tree, use the experience of experts to establish judgment matrix and calculate the heavy weight of impact of each factors. Use AHP quantitative analysis, both to avoid the powerful workload by the establishment of minimal cut sets of fault tree but also reduces the subjections in division in minimal cut set, to solve the uncertain of a logical relation in the fault tree, and determine the degree of importance on both sides of the issue in the face of logic gates. Calculate importance degree of factors in each layer relative to the above factors using the analytic hierarchy process, rather than calculate the weight of the factor relative only to the target layer.After analyzing and classifying of bridge crane failure, predict the probability of failure by consolidating statistical data according to the probability of each type of failure in 24 years, use the self-learning and self-adaptive neural network to predict the probability of failure of the next year, so as to prevent failure and prepare parts and provide a reasonable time to arrange the maintenance and work plans. This article uses two neural networks to predict failure rates, and analyze the prediction accuracy of the two methods. First, use genetic algorithm to optimize BP neural network, genetic algorithm optimization BP neural network is used to predict as the most widely used neural network, using this method has more reliable prediction. Secondly, use Elman neural network to predict, Elman neural network is a dynamic neural network, commonly used in the power load forecasting, this article will be used to predict failure rates.This paper describes the structure of bridge crane and the form of failure, studies the combination prediction method of fault tree analysis and analytic hierarchy analysis, draws the fault tree and hierarchical model of the main fault of bridge crane, establishes the judgment matrix, and calculates the weight of each factor. Used two neural networks to predict the failure rate, studied the weights of the neural network, mapped the forecast result and the prediction error, and analyzed the results, studied the availability by using neural network to predict the failure rate.