The water quality index (WQI) is a critical parameter that must be accurately predicted to ensure the sustainable management of water resources. Thus, our study develops the sine cosine optimization algorithm (SCOA)- ...
详细信息
The water quality index (WQI) is a critical parameter that must be accurately predicted to ensure the sustainable management of water resources. Thus, our study develops the sine cosine optimization algorithm (SCOA)- long short-term memory (LSTM) - Extreme gradient boosting (XGBoost), SCOA- LSTM - least square support vector machine (LSSVM), crow optimization algorithm (COA)- LSTM-XGBoost, and COA-LSTM-LSSVM models to predict WQI in Aidoghmoush river, Iran. First, COA and SCOA adjust the parameters of LSTM, LSSVM, and XGBoost. Then, LSTM captures temporal patterns in the time series data, which include water quality parameters. Finally, the LSSVM and XGBoost models use the captured patterns to make final predictions. Our results demonstrate that the SCOA-LSTM-XGBoost model achieves a Willmott's index (WI) of 0.96, an explained variance score (EVS) of 0.95, and a t-statistic (TS) of 0.021. The results of our paper show that SCOA-LSTM-XGBoost is a reliable model for predicting WQI.
Only a tiny amount of study has been done on fly ash-containing concrete to predict the hardened concrete properties. The features of self-compacting concrete (SCC), in its fresh and hardened forms, have hardly ever b...
详细信息
Only a tiny amount of study has been done on fly ash-containing concrete to predict the hardened concrete properties. The features of self-compacting concrete (SCC), in its fresh and hardened forms, have hardly ever been studied. It is proposed to build networks for predicting SCC's characteristics by support vector regression (SVR). The point of the study is to use salp swarm optimization (SSA) and flow direction algorithm (FDA) to discover the essential parameters of the SVR approach. The V-funnel test, slump flow, V-funnel test, and L-box test are all fresh-phase properties of SCC, whereas compressive strength is a hardened-phase trait. The regression analysis findings suggest a lot of potential for all attributes examined. Regarding developments and evaluations, it was obvious that the systems provided have suitable criteria values. In other words, it indicates that the association among practical and expected SCC features from hybrid models is adequate, indicating pinpoint efficiency in the development and simulation operation. Overall, the FDA-SVR analysis outperforms SSA-SVR, proving the method's capability to choose the most effective variables for the given framework.
The most important input parameter in all solar power generation forecasting systems is solar radiation. Your estimation is necessary for the development of any photovoltaic system project. However, this estimate depe...
详细信息
The most important input parameter in all solar power generation forecasting systems is solar radiation. Your estimation is necessary for the development of any photovoltaic system project. However, this estimate depends on expensive devices, namely pyranometers and pyrheliometers. Therefore, predicting such values through mathematical and computational models is an attractive approach where costs can be reduced. In particular, machine learning methods have been widely successfully applied to this task. The efficiency of a machine learning model depends on a suitable set of parameters. Evolutionary algorithms are helpful and widely used to optimize internal parameters and select the most relevant variables. In this context, machine learning models use evolutionary algorithms' search capability to improve forecasting performance. This work presents a study incorporating different evolutionary algorithms for parameter adjustment in machine learning models applied to solar radiation prediction. Two years of observation data from the Dar es Salaam weather station in Tanzania were used. The results show the presented framework's applicability to finding the best subset of variables, machine learning model and optimization algorithm combination. Although promising results have been obtained in the experiments, it should be clear that care must be taken to generalize the conclusions. The integration of machine learning model with optimization algorithms is limited to a defined data collection context, under specific local environmental conditions and only under data collection from a meteorological station.
The main objective in the one-dimensional cutting stock problem (1D-CSP) is to minimize material costs. In practice, it is useful to focus on auxiliary objectives, one of which is to reduce the number of different cut...
详细信息
The main objective in the one-dimensional cutting stock problem (1D-CSP) is to minimize material costs. In practice, it is useful to focus on auxiliary objectives, one of which is to reduce the number of different cutting patterns. This paper discusses the classical integer IDCSP, where only one type of stock object is included. Meanwhile, the demands of various items must be precisely satisfied in the constraints. In other words, no overproduction or underproduction is allowed. Therefore, to solve this issue, a variable-to-constant method based on a new mathematical model is proposed. In addition, we integrate the approach with two other representative methods to demonstrate its effectiveness. Both benchmark instances and real instances are used in the experiments, and the results show that the methodology is effective in reducing patterns. In particular, in terms of the solutions to the real-life instances, the proposed approach presents a 31.93 to 37.6% pattern reduction compared to other similar methods (including commercial software).
Online optimization applications require fast convergence without sacrificing accuracy. Although the gray wolf optimization (GWO) algorithm is showing good convergence performance, it still needs further improvement t...
详细信息
Online optimization applications require fast convergence without sacrificing accuracy. Although the gray wolf optimization (GWO) algorithm is showing good convergence performance, it still needs further improvement to achieve these requirements. Optimal determination of the GWO control parameters can substantially improve the converge performance. All studies in the literature introduced efforts in tuning these parameters on try-and-error bases which may not satisfy the requirements of the online applications. For this reason, a novel nested improved GWO (NEST-IGWO) is used to determine the optimal control parameters for the IGWO. This novel strategy substantially improved the convergence time and accuracy, especially with online control systems. This strategy is having two nested IGWO loops. The internal IGWO loop includes the target function needed to be optimized. Meanwhile, the external loop is used to optimally determine the control parameters of the internal one. The objective function of the external loop is the failure rate and convergence time of the internal one. The results obtained from the NEST-IGWO are compared to 10 existing optimization algorithms for 10 different benchmark functions. Moreover, these optimization algorithms were applied to determine the parameters of the PV-cell model as a real-world application. The results showed that NEST-IGWO outperformed the other 10 optimization algorithms for all benchmark functions understudy and the estimations of the PV-cell parameters in terms of failure rate and convergence time. With the use of the NEST-IGWO, the convergence time is reduced by 90% of the average convergence time for all other algorithms. Moreover, the failure rate is reduced to 0% which is not the case for other algorithms understudy. These outstanding results prove the superiority of the NEST-IGWO compared to the other algorithms, and it opens a new venue for determining optimal control parameters for all optimization algorithms.
A wastewater treatment plant (WWTP) is an essential part of the urban water cycle, which reduces concentration of pollutants in the river. For monitoring and control of WWTPs, researchers develop different models and ...
详细信息
A wastewater treatment plant (WWTP) is an essential part of the urban water cycle, which reduces concentration of pollutants in the river. For monitoring and control of WWTPs, researchers develop different models and systems. This study introduces a new deep learning model for predicting effluent quality parameters (EQPs) of a WWTP. A method that couples a convolutional neural network (CNN) with a novel version of radial basis function neural network (RBFNN) is proposed to simultaneously predict and estimate uncertainty of data. The multi-kernel RBFNN (MKRBFNN) uses two activation functions to improve the efficiency of the RBFNN model. The salp swarm algorithm is utilized to set the MKRBFNN and CNN parameters. The main advantage of the CNN-MKRBFNN-salp swarm algorithm (SSA) is to automatically extract features from data points. In this study, influent parameters (if) are used as inputs. Biological oxygen demand (BODif), chemical oxygen demand (CODif), total suspended solids (TSSif), volatile suspended solids (VSSif), and sediment (SEDef) are used to predict EQPs, including CODef, BODef, and TSSef. At the testing level, the Nash-Sutcliffe efficiencies of CNN-MKRBFNN-SSA are 0.98, 0.97, and 0.98 for predicting CODef, BODef, and TSSef. Results indicate that the CNN-MKRBFNN-SSA is a robust model for simulating complex phenomena.
This paper introduces a Julia package for tackling linear Diophantine systems and related optimization problems using the Polyhedral Omega algorithm. The package integrates partition analysis and polyhedral geometry t...
详细信息
There are several benefits to constructing a lightweight vision system that is implemented directly on limited hardware devices. Most deep learning-based computer vision systems, such as YOLO (You Only Look Once), use...
详细信息
There are several benefits to constructing a lightweight vision system that is implemented directly on limited hardware devices. Most deep learning-based computer vision systems, such as YOLO (You Only Look Once), use computationally expensive backbone feature extractor networks, such as ResNet and Inception network. To address the issue of network complexity, researchers created SqueezeNet, an alternative compressed and diminutive network. However, SqueezeNet was trained to recognize 1000 unique objects as a broad classification system. This work integrates a two-layer particle swarm optimizer (TLPSO) into YOLO to reduce the contribution of SqueezeNet convolutional filters that have contributed less to human action recognition. In short, this work introduces a lightweight vision system with an optimized SqueezeNet backbone feature extraction network. Secondly, it does so without sacrificing accuracy. This is because that the high-dimensional SqueezeNet convolutional filter selection is supported by the efficient TLPSO algorithm. The proposed vision system has been used to the recognition of human behaviors from drone-mounted camera images. This study focused on two separate motions, namely walking and running. As a consequence, a total of 300 pictures were taken at various places, angles, and weather conditions, with 100 shots capturing running and 200 images capturing walking. The TLPSO technique lowered SqueezeNet's convolutional filters by 52%, resulting in a sevenfold boost in detection speed. With an F1 score of 94.65% and an inference time of 0.061 milliseconds, the suggested system beat earlier vision systems in terms of human recognition from drone-based photographs. In addition, the performance assessment of TLPSO in comparison to other related optimizers found that TLPSO had a better convergence curve and achieved a higher fitness value. In statistical comparisons, TLPSO surpassed PSO and RLMPSO by a wide margin.
An online modeling algorithm is derived from a generic stochastic dual averaging (DA) method. It employs a negative entropy as a distance-generating function and the Volterra series expansion as a dictionary. Assuming...
详细信息
An online modeling algorithm is derived from a generic stochastic dual averaging (DA) method. It employs a negative entropy as a distance-generating function and the Volterra series expansion as a dictionary. Assuming that the measurement data are not i.i.d. but generated by a nonlinear dynamical system with an infinite, exponentially fading memory, the error bounds are established for both the generic DA method and for the proposed modeling algorithm. The experiments performed on a set of benchmark systems confirm the applicability of the algorithm in real-world scenarios and demonstrate its low computational complexity.
In this article, a stochastic incremental subgradient algorithm for the minimization of a sum of convex functions is introduced. The method sequentially uses partial subgradient information, and the sequence of partia...
详细信息
In this article, a stochastic incremental subgradient algorithm for the minimization of a sum of convex functions is introduced. The method sequentially uses partial subgradient information, and the sequence of partial subgradients is determined by a general Markov chain. This makes it suitable to be used in networks, where the path of information flow is stochastically selected. We prove convergence of the algorithm to a weighted objective function, where the weights are given by the Cesaro limiting probability distribution of the Markov chain. Unlike previous works in the literature, the Cesaro limiting distribution is general (not necessarily uniform), allowing for general weighted objective functions and flexibility in the method.
暂无评论