GRNN Prediction Model for Temperature-Induced Deformation of CRTS II Unballasted Slab Track

The General Regression Neural Network (GRNN) is one of the algorithms of artificial neural networks (ANN) that receives much attention in prediction applications. This research used the GRNN to predict the temperatureinduced deformation of unballasted track structures based on experimental data considering external weather conditions, such as sunshine duration, rain conditions, daily maximum temperature, daily minimum temperature, and daily average wind speed. The GRNN network predicts the average absolute error of the prediction results (0.0318 ℃), the maximum absolute error (1.7729 ℃), and the GRNN prediction sample mean squared error (0.070701). The average relative error is 0.32%. The finding of this study shows that the GRNN prediction method has good accuracy and robustness. Furthermore, it can promote the research of unballasted track temperature fields that are related to concrete structures.

A Three-Dimensional Nonlinear Dynamic Numerical Optimization of the Risks of Stope Blasting Based on FOA-GRNN

The fruit fly optimization algorithm-general regression neural network (FOA-GRNN) coupled model and the Finite Element Method-Smoothed Particle Hydrodynamics (FEM-SPH) numerical calculation method are comprehensively used. The control problem of blasting vibration in the process of mining hidden resources under complex environmental conditions has been studied. Taking a lead-zinc mine as the engineering background, the development of hidden resources in the collapse area due to unreasonable mining was studied. Based on the establishment of the first mining stope and its mining method in this area, biosimulation and generalized neural networks were introduced to solve this problem, the coupling of blasting parameters was analyzed, and the 3D nonlinear dynamic coupling model was constructed for numerical simulation. The results show that the blasting parameters of deep-hole mining were optimized, including the values of six output quantities: hole distance, row spacing, side hole distance, explosive unit consumption, minimum resistance line, and interval ratio (the Root Mean Squared Error value is only 0.051). The error between the network optimization parameters and the empirically obtained values was controlled to within 0.05; five possible edge-hole charge structures were designed (the interval ratio is 0.696), and the vibration velocity peak and pressure peak variations with time after detonation are reflected by the simulation results. The dynamic evolution law of the rock mass velocity vector and the damage of the rock damage was revealed. According to the analysis in this paper, the smallest and optimal edge-hole charge structure of the surrounding rock was obtained.

Investigation of Magneto-/Radio-Metric Behavior in Order to Identify an Estimator Model Using K-Means Clustering and Artificial Neural Network (ANN) (Iron Ore Deposit, Yazd, IRAN)

The study area is located near Toot village in the Yazd province of Iran, which is considered in terms of its iron mineralization potential. In this area, due to radioactivity, radiometric surveys were performed in a part of the area where magnetometric studies have also been performed. According to geological studies, the presence of magnetic anomalies can have a complex relationship with the intensity of radioactivity of radioactive elements. Using the K-means clustering method, the centers of the clusters were calculated with and without considering the coordinates of radiometric points. Finally, the behavior of the two variables of magnetic field strength and radioactivity of radioactive elements relative to each other was studied, and a mathematical relationship was presented to analyze the behavior of these two variables relative to each other. On the other hand, the increasing and then decreasing behavior of the intensity of the Earth’s magnetic field relative to the intensity of radioactivity of radioactive elements shows that it is possible to generalize the results of magnetometric surveys to radiometry without radiometric re-sampling in this region and neighboring areas. For this purpose, using the general regression neural network and backpropagation neural network (BPNN) methods, radiometric data were estimated with very good accuracy. The general regression neural network (GRNN) method, with more precision in estimation, was used as a model for estimating the radiation intensity of radioactive elements in other neighboring areas.

INTELLIGENT HEALTH STATUS DETECTION METHOD FOR LOCOMOTIVE FUEL CELL BASED ON DATA-DRIVEN TECHNIQUES

Main drawbacks of fuel cell systems, namely, high cost, poor reliability, and short lifespan, limit the large-scale commercial application of fuel cell systems. The health status detection of fuel cell systems for locomotives is of great significance to the safe and stable operation of locomotives. To identify the failure modes of the fuel cell system accurately and quickly, this study proposed an intelligent health status detection method for locomotive fuel cells based on data-driven techniques. In this study, the actual test data of a 150-kW fuel cell system for locomotives was analyzed. The t-distributed stochastic neighbor embedding (t-SNE) algorithm was combined with the general regression neural network (GRNN) to intelligently detect the health status of the fuel cell system for locomotives. Specifically, t-SNE was used to process the high-dimensionality and strong coupling raw data of health status, enabling the dimensional reduction of the raw data to reflect essential features. Then, GRNN was used to identify the feature data to achieve the fast and accurate detection of the health status of the fuel cell system. Results show that the proposed method can effectively detect four health conditions, namely, normal state, high inlet coolant temperature, low air pressure, and low spray pump pressure, with a diagnostic accuracy of 98.75%. This study is applicable to the analysis of the actual measurement data of high-power level fuel cell systems and provides a reference for the health status detection of fuel cell systems for locomotives. Keywords: fuel cell system for locomotive; data-driven; general regression neural network; t-distributed stochastic neighbor embedding; health status detection

Wind power short-term power forecasting based on Improved Grey Wolf algorithm and Optimized Generalized Regression Neural Network

Aiming at the problem of wind turbine output prediction, a wind power prediction method based on Improved Gray Wolf algorithm and optimized generalized regression neural network is proposed in this paper. Firstly, according to the daily similarity of wind speed and wind power, cluster analysis is used to classify the data. Considering that the degree of each factor affecting wind power output changes, based on the selection of similar days, an improved gray wolf algorithm is introduced to optimize the weight of each influencing factor. The two models of the first mock exam are selected to input the radial single mode function RBF and the back propagation (BP) network to predict the output of the wind turbine separately. The prediction results of the two models are input to the generalized regression neural network optimized by the Wolf Wolf algorithm and the nonlinear combination forecasting is carried out. The basis models are used to predict the output of the wind turbine. The example analysis of an area shows that the model can be closer to the real value in the peak and valley of the prediction curve and has higher prediction accuracy than the combined prediction model of single BP, RBF and non optimized general regression neural network (GRNN).