scholarly journals Estimation of Arc Welding Pressure Pipeline Weld Peaking Parameters Based on Data Prediction

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Pu Liao ◽  
Guixiong Liu ◽  
Ningxiang Yang
Keyword(s):  
Base Metal ◽  
Short Term Memory ◽  
Estimation Method ◽  
Surface Profile ◽  
Prediction Method ◽  
Profile Data ◽  
Pressure Pipes ◽  
Data Prediction ◽  
Pressure Pipeline ◽  
Welding Pressure

Peaking parameter is the key content in the regular inspection of the pressure pipeline. Solving the problem of the peaking measurement method defined by a standard cannot be applied to a situation in which there exists a weld surface with reinforcement and misalignment. In this paper, a peaking estimation method based on data prediction was proposed to estimate the contour information of the base metal at the weld joint using the contour point set data of the base metal part of the weld. Herein, the longitudinal weld peaking estimation method based on a piecewise logistic regression (PLR) and the girth weld peaking estimation method based on a piecewise Bayesian linear regression (PBLR) were studied, and the midpoint of the two symmetrical points of the base metal on either side of the weld was used as a reference for calculating the peaking. Finally, we collected the surface profile data of longitudinal weld pressure pipes with diameters of 155 mm, 255 mm, 550 mm, and 600 mm and the surface profile data of girth weld pressure pipes with diameters of 120 mm, 130 mm, 140 mm, and 170 mm. This weld seam data used the data estimation method proposed in this article and traditional long short-term memory and fitting methods. The results showed that the proposed data prediction method could accurately predict the position of the base metal, and the theoretical mean absolute error of the peaking estimation based on the PBLR and PLR could attain 0.06 mm and 0.07 mm, respectively, which meets the parameter measurement requirements of related verification fields.

scholarly journals Interval prediction method based on Long-Short Term Memory networks for system integrated of hydro, wind and solar power

2019 ◽  
Vol 158 ◽  
pp. 6176-6182 ◽  
Author(s):  
Zhendong Zhang ◽  
Hui Qin ◽  
Liqiang Yao ◽  
Jiantao Lu ◽  
Liangge Cheng
Keyword(s):  
Solar Power ◽  
Short Term Memory ◽  
Prediction Method ◽  
Short Term ◽  
Term Memory ◽  
Interval Prediction ◽  
Long Short Term Memory

scholarly journals A Data Driven RUL Estimation Framework of Electric Motor Using Deep Electrical Feature Learning from Current Harmonics and Apparent Power

Energies ◽  
2021 ◽  
Vol 14 (11) ◽  
pp. 3156
Author(s):  
Tanvir Alam Shifat ◽  
Rubiya Yasmin ◽  
Jang-Wook Hur
Keyword(s):  
Reactive Power ◽  
Short Term Memory ◽  
Estimation Method ◽  
Feature Learning ◽  
Remaining Useful Life ◽  
Total Power ◽  
Bldc Motor ◽  
Current Harmonics ◽  
Power Monitoring ◽  
Brushless Dc

An effective remaining useful life (RUL) estimation method is of great concern in industrial machinery to ensure system reliability and reduce the risk of unexpected failures. Anticipation of an electric motor’s future state can improve the yield of a system and warrant the reuse of the industrial asset. In this paper, we present an effective RUL estimation framework of brushless DC (BLDC) motor using third harmonic analysis and output apparent power monitoring. In this work, the mechanical output of the BLDC motor is monitored through a coupled generator. To emphasize the total power generation, we have analyzed the trend of apparent power, which preserves the characteristics of real power and reactive power in an AC power system. A normalized modal current (NMC) is used to extract the current features from the BLDC motor. Fault characteristics of motor current and generator power are fused using a Kalman filter to estimate the RUL. Degradation patterns for the BLDC motor have been monitored for three different scenarios and for future predictions, an attention layer optimized bidirectional long short-term memory (ABLSTM) neural network model is trained. ABLSTM model’s performance is evaluated based on several metrics and compared with other state-of-the-art deep learning models.

Prediction Method for Plastic Collapse of Pipes Subjected to Combined Bending and Torsion Moments

2010 ◽  
Author(s):  
Yinsheng Li ◽  
Kunio Hasegawa ◽  
Phuong H. Hoang ◽  
Bostjan Bezensek
Keyword(s):  
Entire Range ◽  
Estimation Method ◽  
Prediction Method ◽  
Bending Moment ◽  
Combined Loading ◽  
Plastic Collapse ◽  
Finite Element Analyses ◽  
Pure Torsion ◽  
Stainless Steel Pipe ◽  
Service Inspection

When a crack is detected in a pipe during in-service inspection, the failure estimation method given in the codes such as ASME Boiler and Pressure Vessel Code Section XI non-mandatory Appendix C or JSME S NA-1-2008 Appendix E-8 can be applied to assess the integrity of the pipe. In the current editions of these codes, the failure estimation method is provided for bending moment and pressure. Torsion load is assumed to be relatively small and is not considered in the method. In this paper, finite element analyses are conducted for 24-inch stainless steel pipe with a circumferential surface crack subjected to the combined bending and torsion moments, focusing on large and pure torsion moments. Based on the analysis results, a prediction method for plastic collapse under the combined loading conditions of bending and torsion is proposed for the entire range of torsion moments.

scholarly journals Artificial Neural Networks, Sequence-to-Sequence LSTMs, and Exogenous Variables as Analytical Tools for NO2 (Air Pollution) Forecasting: A Case Study in the Bay of Algeciras (Spain)

Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1770
Author(s):  
Javier González-Enrique ◽  
Juan Jesús Ruiz-Aguilar ◽  
José Antonio Moscoso-López ◽  
Daniel Urda ◽  
Lipika Deka ◽  
...  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Short Term Memory ◽  
Prediction Method ◽  
Monitoring Network ◽  
Exogenous Variables ◽  
Prediction Horizon ◽  
Artificial Neural ◽  
Input Variables ◽  
Analytical Tools

This study aims to produce accurate predictions of the NO2 concentrations at a specific station of a monitoring network located in the Bay of Algeciras (Spain). Artificial neural networks (ANNs) and sequence-to-sequence long short-term memory networks (LSTMs) were used to create the forecasting models. Additionally, a new prediction method was proposed combining LSTMs using a rolling window scheme with a cross-validation procedure for time series (LSTM-CVT). Two different strategies were followed regarding the input variables: using NO2 from the station or employing NO2 and other pollutants data from any station of the network plus meteorological variables. The ANN and LSTM-CVT exogenous models used lagged datasets of different window sizes. Several feature ranking methods were used to select the top lagged variables and include them in the final exogenous datasets. Prediction horizons of t + 1, t + 4 and t + 8 were employed. The exogenous variables inclusion enhanced the model’s performance, especially for t + 4 (ρ ≈ 0.68 to ρ ≈ 0.74) and t + 8 (ρ ≈ 0.59 to ρ ≈ 0.66). The proposed LSTM-CVT method delivered promising results as the best performing models per prediction horizon employed this new methodology. Additionally, per each parameter combination, it obtained lower error values than ANNs in 85% of the cases.

Sign in / Sign up

Export Citation Format

Share Document