Prediction of the height of a water-flowing fracture zone(WFFZ)is the foundation for evaluating water bursting conditions on roof coal.By taking the Binchang mining area as the study area and conducting an in-depth st...Prediction of the height of a water-flowing fracture zone(WFFZ)is the foundation for evaluating water bursting conditions on roof coal.By taking the Binchang mining area as the study area and conducting an in-depth study of the influence of coal seam thickness,burial depth,working face length,and roof category on the height of a WFFZ,we proposed that the proportion of hard rock in different roof ranges should be used to characterise the influence of roof category on WFFZ height.Based on data of WFFZ height and its influence index obtained from field observations,a prediction model is established for WFFZ height using a combination of a genetic algorithm and a support-vector machine.The reliability and superiority of the prediction model were verified by a comparative study and an engineering application.The results show that the main factors affecting WFFZ height in the study area are coal seam thickness,burial depth,working face length,and roof category.Compared with multiple-linear-regression and back-propagation neural-network approaches,the height-prediction model of the WFFZ based on a genetic-algorithm support-vector-machine method has higher training and prediction accuracy and is more suitable for WFFZ prediction in the mining area.展开更多
Diagnosing intermittent fault is an important approach to reduce built-in test（BIT） false alarms. Aiming at solving the shortcoming of the present diagnostic method of intermittent fault, and according to the merit ...Diagnosing intermittent fault is an important approach to reduce built-in test（BIT） false alarms. Aiming at solving the shortcoming of the present diagnostic method of intermittent fault, and according to the merit of support vector machines （ SVM） which can be trained with a small-sample, an SVM-based diagnostic model of 3 states that include OK state, intermittent state and faulty state is presented. With the features based on the reflection coefficients of an alarm rate （ AR ） model extracted from small vibration samples, these models are trained to diagnose intermittent faults. The experimental results show that this method can diagnose multiple intermittent faults accurately with small training samples and BIT false alarms are reduced.展开更多
文摘Prediction of the height of a water-flowing fracture zone(WFFZ)is the foundation for evaluating water bursting conditions on roof coal.By taking the Binchang mining area as the study area and conducting an in-depth study of the influence of coal seam thickness,burial depth,working face length,and roof category on the height of a WFFZ,we proposed that the proportion of hard rock in different roof ranges should be used to characterise the influence of roof category on WFFZ height.Based on data of WFFZ height and its influence index obtained from field observations,a prediction model is established for WFFZ height using a combination of a genetic algorithm and a support-vector machine.The reliability and superiority of the prediction model were verified by a comparative study and an engineering application.The results show that the main factors affecting WFFZ height in the study area are coal seam thickness,burial depth,working face length,and roof category.Compared with multiple-linear-regression and back-propagation neural-network approaches,the height-prediction model of the WFFZ based on a genetic-algorithm support-vector-machine method has higher training and prediction accuracy and is more suitable for WFFZ prediction in the mining area.
文摘Diagnosing intermittent fault is an important approach to reduce built-in test（BIT） false alarms. Aiming at solving the shortcoming of the present diagnostic method of intermittent fault, and according to the merit of support vector machines （ SVM） which can be trained with a small-sample, an SVM-based diagnostic model of 3 states that include OK state, intermittent state and faulty state is presented. With the features based on the reflection coefficients of an alarm rate （ AR ） model extracted from small vibration samples, these models are trained to diagnose intermittent faults. The experimental results show that this method can diagnose multiple intermittent faults accurately with small training samples and BIT false alarms are reduced.
