Application of dimensional analysis and multi-gene genetic programming to predict the performance of tunnel boring machines

Kazemi, Majid; Barati, Reza

doi:10.1016/j.asoc.2022.108997

Cited by 16 publications

(3 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These criteria are used to quantify the performance of a system and to make decisions based on data and evidence, rather than intuition or subjective judgment. Several performance evaluation criteria were developed to compare the results of different approaches (Alizadeh et al, 2017;Fuat Toprak & Savci, 2007;Hosseini et al, 2016;Kazemi & Barati, 2022;Luo & Xie, 2010;McCuen et al, 2006;Mohan, 1997;Toprak & Cigizoglu, 2008;Yoon & Padmanabhan, 1993).…”

Section: Statistical Performance Evaluation Criteriamentioning

confidence: 99%

River Flood Routing Using a Spatial Variable Exponent Parameter Nonlinear Muskingum Model and Sine Cosine Algorithm by Considering Lateral Flow

Atashi

Barati

Lim

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Due to advancements in optimization technology, numerous variable-parameter Muskingum models have been proposed in recent decades, aimed at enhancing the effectiveness of the Muskingum model. This study proposes a novel approach to river flood routing that employs a spatial variable exponent parameter nonlinear Muskingum model with lateral flow considerations. The proposed nonlinear Muskingum model with a variable exponent parameter considers spatial variations, in contrast to earlier studies that concentrated on modifying exponent parameters in response to variable inflow levels. The Muskingum parameters of the proposed model were estimated using an improved Sine Cosine Algorithm (SCA), which was applied to fit six previously reported flood hydrographs. The proposed method aims to minimize the Sum of Square Errors (SSE) between observed and routed outflows. The study demonstrates that by incorporating lateral inflows into the Muskingum model for Wilson, Linsley, and Viessman and Lewis flood data, two sub-reaches produce significantly better results. Conversely, for fitting flood data exhibited by Wye and Dinavar flood data, the three sub-reaches Muskingum model yields superior results. In the case of Lawler flood data, it is suggested that the traditional nonlinear model could be adequate, and there may be no need to incorporate sub-reaches, as the Sum of Square Errors (SSE) remains unchanged. Overall, the study provides a promising approach to addressing river flood routing problems.

show abstract

Section: Statistical Performance Evaluation Criteriamentioning

confidence: 99%

River Flood Routing Using a Spatial Variable Exponent Parameter Nonlinear Muskingum Model and Sine Cosine Algorithm by Considering Lateral Flow

Atashi

Barati

Lim

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…These indices are well-known for evaluating the reliability of hydrological models (Doycheva et al 2017). R 2 represents the fitness of observed and predicted data on 45 degrees reference line (Kazemi and Barati 2022). A higher coefficient indicates a better fit for the model.…”

Section: Performance Indicesmentioning

confidence: 99%

Application of Machine Learning-based Energy Use Forecasting for Inter-basin Water Transfer Project

Kondolf

Sandoval-Solís

et al. 2022

Water Resour Manage

View full text Add to dashboard Cite

Energy use forecasting is crucial in balancing the electricity supply and demand to reduce the uncertainty inherent in the inter-basin water transfer project. Energy use prediction supports the reliable water-energy supply and encourages cost-effective operation by improving generation scheduling. The objectives are to develop subsequent monthly energy use predictive models for the Mokelumne River Aqueduct in California, US. Partial objectives are to (a) compare the model performance of a baseline model (multiple linear regression (MLR)) to three machine learning-based models (random forest (RF), deep neural network (DNN), support vector regression (SVR)), (b) compare the model performance of the whole system to three subsystems (conveyance, treatment, distribution), and (c) conduct sensitivity analysis. We simulate a total of 64 cases (4 algorithms (MLR, RF, DNN, SVR) x 4 systems (whole, conveyance, treatment, distribution) x 4 scenarios (different combinations of independent variables). We concluded that the three machine learning algorithms showed better model performance than the baseline model as they reflected non-linear energy use characteristics for water transfer systems. Among the three machine learning algorithms, DNN models yielded higher model performance than RF and SVR models. Subsystems performed better than the whole system as the models more closely reflected the unique energy use characteristics of the subsystems. The best case was having water supply (t), water supply (t-1), precipitation (t), temperature (t), and population (y) as independent variables. These models can help water and energy utility managers to understand energy performance better and enhance the energy efficiency of their water transfer systems.

show abstract

“…The main reason why these methods are preferred in the study is that they can create nonlinear models of systems through automatic regression. Moreover, these methods are used in various fields to successfully solve engineering problems, such as predicting the performance of tunnel boring machines [8], inferring error severity from symbolic regressed inferential sensors [9], the artificial ant problem [10], automatic feature selection [11].…”

Section: Introductionmentioning

confidence: 99%

Investigation and Optimization of Biosorbent Capacities of Some Plants Used in Daily Life

Arslan,

Kütük

2024

Fırat Üniversitesi Mühendislik Bilimleri Dergisi

View full text Add to dashboard Cite

In this study, sage, chamomile, and tarragon leaves, which are used as spices and consumed as beverages in daily life, were considered as different biosorbents for water purification by biosorption. At the same time, the biosorption processes were modeled using evolutionary computation-based automatic programming (AP) methods. The effects of initial dye concentration (10-200 mg/L), temperature (20-50 ⁰C), and contact time (0-120 min) on the biosorption processes were investigated. The highest biosorption capacity (49 mg/g) was obtained in tarragon biosorbent at a dye concentration of 200 mg/L after 120 min of contact time. The biosorption mechanism was found to follow Freundlich isotherm and pseudo-second order (PSO) reaction kinetics for all biosorbents. AP methods used the parameters of initial dye concentration, temperature, and contact time to model the biosorption capacity obtained from the experiments. According to the simulation results, the methods successfully modeled the biosorption capacities with R^2 values as high as 0.99 for each biosorbent. This study demonstrated that the effects of biosorption processes can be studied not only by experiments but also by simulations.

show abstract

Application of dimensional analysis and multi-gene genetic programming to predict the performance of tunnel boring machines

Cited by 16 publications

References 40 publications

River Flood Routing Using a Spatial Variable Exponent Parameter Nonlinear Muskingum Model and Sine Cosine Algorithm by Considering Lateral Flow

River Flood Routing Using a Spatial Variable Exponent Parameter Nonlinear Muskingum Model and Sine Cosine Algorithm by Considering Lateral Flow

Application of Machine Learning-based Energy Use Forecasting for Inter-basin Water Transfer Project

Investigation and Optimization of Biosorbent Capacities of Some Plants Used in Daily Life

Contact Info

Product

Resources

About