Wyniki wyszukiwania - BazTech

1

Research on cost prediction for construction project based on Boruta-SO-BP model

Cui Hao, Xia Junjie

Archives of Civil Engineering

|

2023

|

Vol. 69, nr 4

413--429

EN

Cost prediction for construction projects provides important information for project feasibility studies and design scheme selection. To improve the accuracy of early-stage cost estimation for construction projects, an improved neural network prediction model was proposed based on BP (back propagation) neural network and Snake Optimizer algorithm (SO). SO algorithm is adopted to optimize the initial weights and thresholds of the BP neural network. Cost data for 50 construction projects undertaken by Shandong Tianqi Real Estate Group in China was collected, and the data samples were clustered into three categories using cluster analysis. 18 engineering feature indicators were determined through a literature review and 10 feature indicators were selected using Boruta algorithm for the input set. Compared to BP neural network and PSO-BP neural network, the results show that the improved SO-BP model has higher prediction accuracy, stability, better generalization ability and applicability. Therefore, based on reasonable feature indicators, the method proposed in this paper has certain guiding significance for predicting engineering costs.

2

Lithology identification technology using BP neural network based on XRF

Wang Qingshan, Zhang Xiongjie, Tang Bin, Ma Yingjie, Xing Jisheng, Liu Longfeng

Acta Geophysica

|

2021

|

Vol. 69, no. 6

2231--2240

EN

The element content obtained by X-ray fluorescence (XRF) mud-logging is mainly used to determine mineral content and identify lithology. This work has been developed to identify dolomite, granitic gneiss, granite, limestone, trachyte, and rhyolite from two wells in Nei Mongol of China using back propagation neural network (BPNN) model based on the element content of drill cuttings by XRF analysis. Neural network evaluation system was constructed for objective performance judgment based on Accuracy, Kappa, Recall and training speed, and BPNN for lithology identification was established and optimized by limiting the number of nodes in the hidden layer to a small range. Meanwhile, six basic elements that can be used for fuzzy identification were determined by cross plot and four sensitive elements were proposed based on the existing research, both of which were combined to establish sixteen test schemes. A large number of tests are performed to explore the best element combination, and the result of experiments indicate that the improved combination has obvious advantages in identification performance and training speed. The author’s pioneer work has contributed to the neural network evaluation system for lithology identification and the optimization of input elements based on BPNN.

3

Control of the weld quality using welding parameters in a robotic welding process

Park M.-H., Jin B.-J., Yun T.-J., Son J.-S., Kim C.-G., Kim I. S.

Journal of Achievements in Materials and Manufacturing Engineering

|

2018

|

Vol. 87, nr 1

32--40

EN

Purpose: Since the welding automations have widely been required for industries and engineering, the development of the predicted model has become more important for the increased demands for the automatic welding systems where a poor welding quality becomes apparent if the welding parameters are not controlled. The automated welding system must be modelling and controlling the changes in weld characteristics and produced the output that is in some way related to the change being detected as welding quality. To be acceptable a weld quality must be positioned accurately with respect to the joints, have good appearance with sufficient penetration and reduce low porosity and inclusion content. Design/methodology/approach: To achieve the objectives, two intelligent models involving the use of a neural network algorithm in arc welding process with the help of a numerical analysis program MATLAB have been developed. Findings: The results represented that welding quality was fully capable of quantifying and qualifying the welding faults. Research limitations/implications: Welding parameters in the arc welding process should be well established and categorized for development of the automatic welding system. Furthermore, typical characteristics of welding quality are the bead geometry, composition, microstructure and appearance. However, an intelligent algorithm that predicts the optimal bead geometry and accomplishes the desired mechanical properties of the weldment in the robotic GMA (Gas Metal Arc) welding should be required. The developed algorithm should expand a wide range of material thicknesses and be applicable in all welding position for arc welding process. Furthermore, the model must be available in the form of mathematical equations for the automatic welding system. Practical implications: The neural network models which called BP (Back Propagation) and LM (Levenberg-Marquardt) neural networks to predict optimal welding parameters on the required bead reinforcement area in lab joint in the robotic GMA welding process have been developed. Experimental results have been employed to find the optimal algorithm to predict bead reinforcement area by BP and LM neural networks in lab joint in the robotic GMA welding. The developed intelligent models can be estimated the optimal welding parameters on the desired bead reinforcement area and weld criteria, establish guidelines and criteria for the most effective joint design for the robotic arc welding process. Originality/value: In this study, intelligent models, which employed the neural network algorithms, one of AI (Artificial Intelligence) technologies have been developed to study the effects of welding parameters on bead reinforcement area and to predict the optimal bead reinforcement area for lab joint in the robotic GMA welding process. BP (Back Propagation) and LM (Levenberg-Marquardt) neural network algorithm have been used to develop the intelligent model.

4

Prediction model of gas quantity emitted from coal face based on PCA-GA-BP neural network and its application

Qiu J. W., Liu Z. G., Zhou L., Qin R. X.

Journal of Power Technologies

|

2017

|

Vol. 97, nr 3

169--178

EN

Gas has always been a serious hidden danger in coal mining. The quantity of gas emitted from the coal face is affected by many factors. To overcome the difficulty in accurately predicting the quantity of emission, a novel predictive model (PCA-GABP) based on principal component analysis (PCA), genetic algorithm (GA) and back propagation (BP) neural network was proposed. The model was tested and applied in different coal seams at Panbei Coal Mine in Huainan, China, involving sixteen training samples and four predicting samples. Results showed that: Gas emission quantity was significantly correlated with burial depth, gas content in the mining layer, gas content in the adjacent layer, and layer spacing. The correlations among these variables exceeded 60%. Linear regression analysis using the optimized model was affected by sample size and discreteness. The correlation coefficient (R) and maximum relative error (MRE) of the PCA-GA-BP model were 0.9988 and 3.02%, respectively. The MRE of the optimized model was 70.2% and 53.2% smaller than that of the BP and GA-BP models, respectively. The conclusions obtained in the study provide technical support for the prediction of gas quantity emitted from coal face, and the proposed method can be used in other engineering fields.

5

Day-ahead Electricity Price Prediction Based on Improved ANN Information Fusion

Li K.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 3b

78--80

EN

A novel information fusion method is proposed based on the characters of day-ahead electricity price. An improved BPNN is used for its better performance as the core algorithm of information fusion. Using the information fusion ideas, a new modelling approach is proposed to establish the prediction model. The day-ahead electricity price prediction model is tested by the real data. The experiments demonstrate that the new prediction model established by improved BPNN information fusion method has better performance.

PL

W artykule przedstawiono nową metodę fuzji danych w oparciu o charakterystyki cenowe elektryczności z dnia poprzedniego. Algorytm oparto na sieci neuronowej BPNN. Jego działanie poddano badaniom, bazując na prawdziwych danych, których wyniki wskazują na skuteczność działania proponowanego rozwiązania.

6

A New Intrusion Detection Model Based on Data Mining and Neural Network

Dong Y., Qi B., Zhu W., Gao W.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 1b

88--90

EN

Today, we often apply the intrusion detection to aid the firewall to maintain the network security. But now network intrusion detection have problem of higher false alarm rate, we apply the data warehouse and the data mining in intrusion detection and the technology of network traffic monitoring and analysis. After network data is processed by data mining, we will get the certain data and the uncertain data. Then we process the data by the BP neural network, which based on the genetic algorithm, again. Finally, we propose a new model of intrusion detection based on the data warehouse, the data mining and the BP neural network. The experimental result indicates this model can find effectively many kinds behavior of network intrusion and have higher intelligence and environment accommodation.

PL

Obecnie, w celu utrzymania bezpieczeństwa sieci, stosuje się wykrywanie ataków przy pomocy zapory ogniowej, co często powoduje za wysoki poziom fałszywych ataków. W proponowanym rozwiązaniu proponuje się wykorzystanie magazynowania i pozyskiwania danych oraz analizę monitoringu ruchu sieci. Przetwarzanie danych polegało dotychczas na ustaleniu danych pewnych i niepewnych; obecnie proponujemy wykorzystanie genetycznego algorytmu sieci neuronowych BP. Ostatecznie, wprowadzono nowy model detekcji ataków bazujący na magazynowaniu i pozyskiwaniu danych oraz neuronowych sieciach BP. Badania eksperymentalne wykazują, że zaprezentowany model pozwala na znalezienie wielu rodzajów zachowań ataków sieci, jest bardziej inteligentny, zapewnia wyższy standard obsługi środowiska.

7

An improved neural networks for stereo-camera calibration

Xing Y. J., Xing J., Sun J., Hu L.

Journal of Achievements in Materials and Manufacturing Engineering

|

2007

|

Vol. 20, nr 1-2

315-318

EN

Purpose: Improve the generalization capability and speed of back-propagation neural network (BPNN). Design/methodology/approach: In this paper, CCD cameras are calibrated implicitly using BP neural network by means of its ability to fit the complicated nonlinear mapping relation. Conventional BP algorithms easily fall into part-infinitesimal, slowing speed of convergence and exorbitance training that will influence the training result, delay convergence time and debase generalization capability. During our experiments, dense sample data are acquired by using high precisely numerical control platform, and the variances error (PVE) is adopted during training the neural network. Findings: Experiments indicate that the neural network used PVE has great generalization. The error percentages obtained from our set-up are limitedly better than those obtained through Mean Square Error (MSE). The system is generalization enough for most machine-vision applications and the calibrated system can reach acceptable precision of 3D measurement standard. Research limitations/implications: The value needs to be decided by experiments, and the reconstruction images will be distorted if the value is more than 6. Originality/value: The variances error is be adopted in BPNN first.