Reduction of computational error by optimizing SVR kernel coefficients to simulate concrete compressive strength through the use of a human learning optimization algorithm

Engineering With Computers, Journal Year: 2021, Volume and Issue: 38(4), P. 3151 - 3168

Published: Feb. 24, 2021

Language: Английский

---

Assessment of distribution center locations using a multi-expert subjective–objective decision-making approach

Scientific Reports, Journal Year: 2021, Volume and Issue: 11(1)

Published: Sept. 30, 2021

Abstract Distribution is a strategic function of logistics in different companies. Establishing distribution centers (DCs) appropriate locations helps companies to reach long-term goals and have better relations with their customers. Assessment possible for opening new DCs can be considered as an MCDM (Multi-Criteria Decision-Making) problem. In this study, decision-making approach proposed assess DC locations. The based on Stepwise Weight Ratio Analysis II (SWARA II), Method the Removal Effects Criteria (MEREC), Weighted Aggregated Sum Product (WASPAS), simulation, assignment model. assessment process performed using subjective objective criteria weights determined multiple experts’ judgments. decision matrix, are modeled triangular probability alternatives. Then, simulation model, final aggregated results determined. A case addressed show applicability approach. comparative analysis also made verify results. analyses study that efficient dealing locations, congruent those existing methods.

Language: Английский

---

Assessment of the ground vibration during blasting in mining projects using different computational approaches

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: Oct. 30, 2023

The investigation compares the conventional, advanced machine, deep, and hybrid learning models to introduce an optimum computational model assess ground vibrations during blasting in mining projects. long short-term memory (LSTM), artificial neural network (ANN), least square support vector machine (LSSVM), ensemble tree (ET), decision (DT), Gaussian process regression (GPR), (SVM), multilinear (MLR) are employed using 162 data points. For first time, blackhole-optimized LSTM has been used predict blasting. Fifteen performance metrics have implemented measure prediction capabilities of models. study concludes that blackhole optimized-LSTM PPV11 is highly capable predicting vibration. Model assessed with RMSE = 0.0181 mm/s, MAE 0.0067 R 0.9951, a20 96.88, IOA 0.9719, IOS 0.0356 testing. Furthermore, this reveals accuracy less affected by multicollinearity because optimization algorithm. external cross-validation literature validation confirm PPV11. ANOVA Z tests reject null hypothesis for actual vibration, Anderson-Darling test rejects predicted This also GPR LSSVM overfit moderate problematic assessing vibration

Language: Английский

---

Mechanical Framework for Geopolymer Gels Construction: An Optimized LSTM Technique to Predict Compressive Strength of Fly Ash-Based Geopolymer Gels Concrete

Gels, Journal Year: 2024, Volume and Issue: 10(2), P. 148 - 148

Published: Feb. 16, 2024

As an environmentally responsible alternative to conventional concrete, geopolymer concrete recycles previously used resources prepare the cementitious component of product. The challenging issue with employing in building business is absence a standard mix design. According chemical composition its components, this work proposes thorough system or framework for estimating compressive strength fly ash-based (FAGC). It could be possible construct predicting FAGC using soft computing methods, thereby avoiding requirement time-consuming and expensive experimental tests. A complete database 162 datasets was gathered from research papers that were published between years 2000 2020 prepared develop proposed models. To address relationships inputs output variables, long short-term memory networks deployed. Notably, model examined several methods. modeling process incorporated 17 variables affect CSFAG, such as percentage SiO

Language: Английский

---

Predicting ground vibration during rock blasting using relevance vector machine improved with dual kernels and metaheuristic algorithms

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Aug. 28, 2024

The ground vibration caused by rock blasting is an extremely hazardous outcome of the operation. Blasting activity has detrimental effects on both ecology and human population living in proximity to area. Evaluating magnitude vibrations requires careful evaluation peak particle velocity (PPV) as a fundamental essential parameter for quantifying velocity. Therefore, this study employs models using relevance vector machine (RVM) approach predicting PPV resulting from quarry blasting. This investigation utilized conventional optimized RVM first time prediction. work compares thirty-three choose most efficient performance model. following conclusions have been mapped outcomes several analyses. each model demonstrates achieved more than 0.85 during testing phase, there was strong correlation observed between actual predicted ones. analysis metrics (RMSE = 21.2999 mm/s, 16.2272 R 0.9175, PI 1.59, IOA 0.8239, IOS 0.2541), score (= 93), REC curve 6.85E-03, close actual, i.e., 0), fitting 1.05 best fit, 1), AD test 11.607 9.790), Wilcoxon 95%), Uncertainty (WCB 0.0134), computational cost 0.0180) demonstrate that PSO_DRVM MD29 outperformed better other phase. will help mining civil engineers experts select kernel function its hyperparameters estimating project. In context industry, application offers significant potential enhancing safety protocols optimizing operational efficiency.

Language: Английский

---

Predicting the Permeability of Pervious Concrete Based on the Beetle Antennae Search Algorithm and Random Forest Model

Advances in Civil Engineering, Journal Year: 2020, Volume and Issue: 2020(1)

Published: Jan. 1, 2020

Pervious concrete is an environmentally friendly material that improves water permeability, skid resistance, and sound absorption characteristics. Permeability the most important functional performance for pervious while limited studies have been conducted to predict permeability based on mix‐design parameters. This study proposed a method combine beetle antennae search (BAS) random forest (RF) algorithm of concrete. Based 36 samples designed in laboratory 4 key influencing variables, can be obtained by varying parameters RF. BAS was used tune hyperparameters RF, which were then verified so‐called 10‐fold cross‐validation. Furthermore, model RF validated correlation The results showed tuned efficiently; conventional construct evolved concrete; cement/aggregate ratio significant variable determine followed coarse aggregate proportions.

Language: Английский

---

A novel approach for classification of soils based on laboratory tests using Adaboost, Tree and ANN modeling

Transportation Geotechnics, Journal Year: 2020, Volume and Issue: 27, P. 100508 - 100508

Published: Dec. 31, 2020

Language: Английский

---

Tensile strength prediction of rock material using non-destructive tests: A comparative intelligent study

Transportation Geotechnics, Journal Year: 2021, Volume and Issue: 31, P. 100652 - 100652

Published: Sept. 9, 2021

Language: Английский

---

Novel hybrid model based on echo state neural network applied to the prediction of stock price return volatility

Expert Systems with Applications, Journal Year: 2021, Volume and Issue: 184, P. 115490 - 115490

Published: June 29, 2021

Language: Английский

---

Prediction of ground vibration due to mine blasting in a surface lead–zinc mine using machine learning ensemble techniques

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: April 21, 2023

Abstract Ground vibration due to blasting is identified as a challenging issue in mining and civil activities. Peak particle velocity (PPV) one of the undesirable consequences, which resulted during emission blasted bench. This study focuses on PPV prediction surface mines. In this regard, two ensemble systems, i.e., artificial neural networks extreme gradient boosting (EXGBoosts) were developed for largest lead–zinc open-pit mines Middle East. For modeling, several ANN XGBoost base models separately designed with different architectures. Then, validation indices such coefficient determination (R 2 ), root mean square error (RMSE), absolute (MAE), variance accounted (VAF), Accuracy used evaluate performance models. The five top high accuracy selected construct an model each methods, ANNs XGBoosts. To combine outputs achieve single result stacked generalization technique, was employed. Findings showed increase predicting comparison best individual EXGBoosts superior method PPV, obtained values R , RMSE, MAE, VAF, corresponding (0.990, 0.391, 0.257, 99.013(%), 98.216), (0.968, 0.295, 0.427, 96.674(%), 96.059), training testing datasets, respectively. However, sensitivity analysis indicated that spacing (r = 0.917) number blast-holes 0.839) had highest lowest impact intensity,

Language: Английский

---