BP Neural Network Theory Research Articles

The Establishment of Land Appraisal Model in Xiamen Based on BP Neural Network

Establish the land evaluation model of Xiamen by means of BP neural network theory, taking 2007-2009 land evaluation cases of Xiamen as examples. Through statistical analysis, we find that the neural network which has 9 net work hidden layer nodes and 19% of maximal error index is more suitable for Xiamen land price assessment than others. Empirical analysis shows that the model has a good generalization ability, which can be used for land evaluation practices. The results indicates that the properties of autonomous learning of BP network can reduce the subjective factors of appraiser in land evaluation , also, the network has the advantage of simple and quick calculation.

Predication of GRT Fiber-Rubberized Haydite Concrete Bend Strength Based on Multiple Regression Analysis and BP Neural Network

Experiment with intensity level for the LC30 ceramsite concrete as the research object, changing the content of cement, GRT fiber, rubber powder by the orthogonal test to configure GRT fiber—rubberized haydite concrete samples, maintenance samples 7d and 28d in standard conditions and respectively testing their bend strength. Through the analysis of the test data, using multiple regression analysis established the GRT fiber—rubberized haydite concrete 7d and 28d bend strength regression formulas.By means of BP neural network theory combine MATLAB programme established GRT fiber—rubberized haydite concrete 7d and 28d bend strength neural network model.Finally using 3 groups new test data to compare the value of multiple regression equations and BP neural network’s predicted value.The results indicate that the multiple regression equations of 28d’s and 28d’s BP neural network model are availabled.But because of the water and cement which in the GRT fiber—rubberized haydite concrete can not hydration reaction sufficiently during the 7d’s,so the multiple regression equations of 7d’s is unavailabled.

Predicition of GRT Fiber-Rubberized Haydite Concrete Compressive Strength Based on Multiple Regreeion Analysis and BP Neural Network

Experiment with intensity level for the LC30 ceramsite concrete as the research object, changing the content of cement, GRT fiber, rubber powder by the orthogonal test to configure GRT fiber—rubberized haydite concrete samples, maintenance samples 7d and 28d in standard conditions and respectively testing their standard compressive strength. Through the analysis of the test data, using multiple regression analysis established the GRT fiber—rubberized haydite concrete 7d and 28d standard compressive strength regression formulas.By means of BP neural network theory combine MATLAB programme established GRT fiber—rubberized haydite concrete 7d and 28d standard compressive strength neural network model.Finally using 3 groups new test data to compare the value of multiple regression equations and BP neural network’s predicted value.The results indicate that the multiple regression equations and BP neural network model are availabled.

A New Dynamic Assessment for Multi-parameters Information of Water Inrush in Coal Mine*

The coal mine flood is one of five nature disasters in the mine construction and production process. During the past 20 years, more than 250 mines were flooded, and the direct economic losses reached as high as more than 350 hundred million Yuan. In this article, the regional arrangement of sensors in the face of the program, considering the water level roadway, aquifer water levels, the top and bottom side pressure and the amount of water before driving and other factors, in combination with hydro-geological information and monitoring data, based on BP neural network and DS theory of levels of evidence coal face integration evaluation model of water bursting. It has been carried on the example analysis in the ore 3# coal bed in Shanxi, the results of prediction in line with the actual results, and different conditions for many parameters of the mine statistics, storage and evaluation.

Reverse Design of Cam Contour Curve Based on BP Neural Network

To improve precision of reverse design in cam contour curve, method of using BP neural network was presented. The theory and algorithm of BP neural network were introduced, the principle and process of BP neural network modeling used in the reverse design of cam contour curve were expounded, and the actualize method of reverse design with modeled network was afforded. Experiment indicates that this method is suitable to reverse the contour curve of cam, and has high approach precision. At last, disadvantages of the method were discussed.

Fault Monitoring of Transformer Based on FTIR

Transformer is important power transmission equipment, and its working condition directly affects the safety level. Using the Fourier transformer infrared spectrometer for the qualitative analysis of the fault gases, using the theory of BP neural network for the quantitative analysis of the characteristics of the fault gases, can determine the operational status of transformer. This method monitors the function of the transformer effectively, judges the potential failure or hidden dangers accurately.

Shear Bearing Capacity Forecast of Reinforced Concrete Frame Abnormal Node Based on the BP Neural Network Theory

In order to forecast shear bearing capacity of reinforced concrete frame abnormal node on the basis of analyzing feedforward neural network theory, the complex nonlinear relationship, which is between shear bearing capacity of reinforced concrete frame abnormal node under the low cyclic loading and each major influence factor, is simulated by BP neural network, and the neural network prediction model is built. The comparison with the result of experiment has brought satisfying result. The method of shear bearing capacity forecast of reinforced concrete frame abnormal node based on the BP neural network theory is proved feasible and applicable.

A Relative Colorimetric Rending Mapping Method Based on BP Neural Network

There are very complicated nonlinear relations between input value and presentation color of digital prepress devices, as well as in the gamut mapping among different devices. Not all the relations can be described by linear model properly. In recent years, neural network and black box theory are introduced to describe the relations. In this research, the printer gamut is divided into micro space sections, a relative colorimetric rending mapping method is proposed based on BP neural network theory to complete the gamut mapping from monitor gamut to color printer. Experimental results show that the model reduces the color error of color blocks between the two devices, and the color rendering method can be used for gamut mapping of image with less continuous tone.

Tool Wear Prediction when Maching Ti6Al4V Using Different Tool Materials

The tool life is different when using different tool materials machine Ti6Al4V under different cutting conditions. Based on the experiments of cutting Ti6Al4V using different tool materials, the tool wear prediction model is established with BP neural network theory. The wear prediction model is verified by further experiments. The results show that there is a valid prediction interval when using BP neural network predicts tool wear. When the cutting speed is in the vicinity of the training sample data, the relative error of the tool wear prediction and experimental results is less than 10%, which can meet the actual production. The tool wear prediction model is helpful for cutting tool material selection, cutting speed optimization and predomination the time of changing tool accurately when cutting titanium alloy.

A Preliminary Study on Targets Association Algorithm of Radar and AIS Using BP Neural Network

Targets association judgment is one of kernels in the processing of multi-targets information fusion. An association algorithm of the target information of the marine radar and AIS was proposed based on BP neural network theory. First, the design of the network structure was discussed in detail, then, the computer simulation based on Matlab was carried out. The results indicated that this method can achieve the association judgment effectively for different targets’ information.

RAILWAY PASSENGER TRAFFIC VOLUME PREDICTION BASED ON NEURAL NETWORK

In order to overcome the shortcomings of the methods in railway passenger traffic volume prediction, an improved BP neural network method is adopted to predict railway passenger traffic volume. By analyzing the improved BP neural network theory, the neural network prediction model of railway passenger traffic volume is set up. The network learning and training simulation experiment is carried out on the data of railway passenger traffic volume from 1980–1998. Comparing with standard BP neural network, the improved BP neural network presents more accurate and reliable prediction results and faster learning speed. The improved BP neural network provides a new and feasible thought to predict railway passenger traffic volume, which can also be used to predict other relative problems.