Grey artificial neural network combination model based method for predicting height of water-flowing fractured zone

Abstract

The invention discloses a grey artificial neural network combination model based method for predicting height of a water-flowing fractured zone. The method includes the steps of S1, collecting an influencing factor index of the height of the mine water-flowing fractured zone and corresponding height of the water-flowing fractured zone so as to obtain a sample data set; S2, by means of a weakening buffer operator in the grey system theory, weakening extreme changes of the sample data set; S3, normalizing the sample data set weakened, and dividing the normalized sample data set into a training sample and a testing sample; S4, by means of a Matlab artificial neural network toolbox, entering the training sample into a compiled program to establish a water-flowing fractured zone height prediction model; S5, comparing a predicted value of the testing sample and its actual value, if a relative error is less than or equal to 10%, determining that the model is effective, if not, determining that the model is ineffective, changing programming parameters, and re-constructing the model.

Description

technical field [0001] The invention relates to the application of gray system theory and artificial neural network in the technical field of safe mining of coal resources, especially the prediction of the combined model of buffer operator and BP network in the mine water-conducting fissure zone height. Background technique [0002] Effectively controlling mine floods, predicting the height of water-conducting fissures, and improving the utilization rate of coal resources in my country have become the focus of domestic scholars. The prediction of the height of the water-conducting fracture zone (referred to as "conduction height") is a key parameter to determine whether coal mining under water can be carried out safely. However, due to the complexity, diversity, ambiguity and uncertainty of the rock (soil) medium, the estimation of the height of the water-conducting fracture zone in China has remained at the stage of combining experience and theory for many years. The empir...

AI Technical Summary

Problems solved by technology

These methods have their own shortcomings, such as high prediction cost and cumbersome operation.

[0004] A patent application titled "A Method for Predicting the Height of Mine Water Conducting Fracture Zones" filed by Liaoning University of Engineering Technology on October 31, 2013, and a patent application titled "A Method Based on Sensitivity The patent application of the method for predicting the height of the fissure zone based on the analysis of the analysis, and the patent application of Shandong University of Science and Technology on September 28, 2014, entitled "A Method for Predicting the Height of the Water-conducting Fracture Zone", each proposed a water-conducting fracture zone height Forecasting methods, but the above methods do not weaken extreme changes, which is not conducive to revealing the hidden laws of the data, and the calculation speed of the method is slow, and the relative error is large

Images