Background: The recent progress and achievements in the advanced, accurate, and rigorously evaluated algorithms has revolutionized different aspects of the predictive microbiology including bacterial growth.Objectives: In this study, attempts were made to develop a more accurate hybrid algorithm for predicting the bacterial growth curve which can also be applicable in predictive microbiology studies.Materials and Methods: Sigmoid functions, including Logistic and Gompertz, as well as least square support vector machine (LSSVM) based algorithms were employed to model the bacterial growth of the two important strains comprising Listeria monocytogenes and Escherichia coli. Even though cross-validation is generally used for tuning the parameters in LSSVM, in this study, parameters tuning (i.e.,‘c’ and ‘σ’) of the LSSVM were optimized using non-dominated sorting genetic algorithm-II (NSGA-II), named as NSGA-II-LSSVM. Then, the results of each approach were compared with the mean absolute error (MAE) as well as the mean absolute percentage error (MAPE).Results: Applying LSSVM, it was resulted in a precise bacterial growth modeling compared to the sigmoid functions. Moreover, our results have indicated that NSGA-II-LSSVM was more accurate in terms of prediction than LSSVM method.Conclusion: Application of the NSGA-II-LSSVM hybrid algorithm to predict precise values of ‘c’ and ‘σ’ parameters in the bacterial growth modeling resulted in a better growth prediction. In fact, the power of NSGA-II for estimating optimal coefficients led to a better disclosure of the predictive potential of the LSSVM.

One of the challenging problems in the oil and gas industry is accurate and reliable multiphase flow rate measurement in a three-phase flow. The application of methods with minimized uncertainty is required in the industry. Previously developed correlations for two-phase flow are complex and not capable of three-phase flow. Hence phase behavior identification in different conditions of designing and modeling of three-phase flow is important. Numerous laboratory and theoretical studies have been done to describe the Venturi multiphase flow meter in both horizontal and vertical flow. However, it is not possible to select the measurement devices for all similar conditions. In this study, a new venturi model is developed to implement in Simulink/Matlab for predicting the mass flow rate of gas, water, and oil. This model is simple and semi-linear. Several classified configurations of three-phase flow are simulated using computational fluid dynamics analysis to get hydrodynamics parameters of the flows to use as inputs of the model. The obtained data is used as a test and train data in the least squares support vector machine (LSSVM) algorithm. The pressure drop and mass flow rate of gas, oil, and water are calculated with the LSSVM method. Two tuning parameters of LSSVM, namely γ and 𝜎 2, are obtained as 1150954 and 0. 4384, 53. 9199 and 0. 18163, 8. 8714 and 0. 14424, and 1003913. 2214 and 0. 74742 for the pressure drop, the mass flow rate of oil, gas mass flow rate, and the water mass flow rate, respectively. Developed models are found to have an average relative error of 5. 81%, 6. 31%, and 2. 58% for gas, oil, and water, respectively.

This paper concentrates on a new procedure which experimentally recognizes gears and bearings faults of a typical gearbox system using a least square support vector machine (LSSVM). Two wavelet selection criteria Maximum Energy to Shannon Entropy ratio and Maximum Relative Wavelet Energy are used and compared to select an appropriate wavelet for feature extraction. The fault diagnosis method consists of three steps, firstly the six different base wavelets are considered. Out of these six wavelets, the base wavelet is selected based on wavelet selection criterion to extract statistical features from wavelet coefficients of raw vibration signals. Based on wavelet selection criterion, Daubechies wavelet and Meyer are selected as the best base wavelet among the other wavelets considered from the Maximum Relative Energy and Maximum Energy to Shannon Entropy criteria respectively. Finally, the gearbox faults are classified using these statistical features as input to LSSVM technique. The optimal decomposition level of wavelet is selected based on the Maximum Energy to Shannon Entropy ratio criteria. In addition to this, Energy and Shannon Entropy of the wavelet coefficients are used as two new features along with other statistical parameters as input of the classifier. Some kernel functions and multi kernel function as a new method are used with three strategies for multi classification of gearboxes. The results of fault classification demonstrate that the LSSVM identified the fault categories of gearbox more accurately with multi kernel and OAOT strategy.

Evaporation is one of the most important and influential processes in the water cycle. Evaporation results in the loss of more than half of the precipitation in arid areas. Evaporation pan is used as an indicator for determining the evaporation of lakes and reservoirs due to the ease of interpreting its data around the world. On the other hand, the study on evaporation from the pan and the rate of evapotranspiration of the reference plant shows that there is a linear and direct relation between evaporation from the pan and evapotranspiration of the reference plant. Therefore, by correctly recording the amount of evaporation from the bath, the evapotranspiration of the reference plant can be estimated. ­The empirical relationships presented for estimating evaporation from free surfaces, considering meteorological parameters as inputs, are highly diverse. The accuracy of empirical relationships varies in different regions and needs calibration in each area. Also, it does not have high accuracy and access to all input parameters is difficult or time consuming. The aim of this study was to evaluate the efficiency of backup vector machine and least squares support vector machine for estimating evaporation from free water level in Golestan province. In this research, three synoptic stations (Kelaleh, Gorgan and Bandar-Turkman) were used for daily weather data for 17 years (1997-2015). The results showed that the input patterns with relative humidity input parameters, maximum relative humidity, wind speed and sunshine hours with the highest R2 and the lowest RMSE and MBE.

In the present study, precipitation in six stations of Karun3 basin is downscaled by using the hybrid of least squares support vector machine and whale optimization algorithm (LSSVM-WOA), K nearest neighbor (KNN), and artificial neural network (ANN). For downscaling precipitation, first, the days of year are classified into wet and dry days by using MARS and M5 algorithms. Then, the amount of precipitation for wet days is estimated by using each of LSSVM-WOA, KNN and ANN methods. Based on the findings, MARS algorithm is superior over M5 algorithm. Based on the mean precipitation in the six stations, ANN is a little bit better than LSSVM-WOA (0.5 percent more accurate). While, by regarding the mean of standard deviations, the Nash-Sutcliff for Ann is up to 5.04 percent more accurate than LSSVM-WOA. Eventually, the amount of precipitation is predicted based on the CanESM2 model under RCP2.6, RCP4.5 and RCP8.5 scenarios for 2020-2040 and 2070-2100 periods. Based on the results of applying LSSVM-WOA, the precipitation in each three scenarios is decreased compared to the base period. Maximum decrease of precipitation (18%) is calculated by RCP8.5 for 2070-2100 period. Minimum decrease of precipitation (1%) is related to RCP2.6 scenario for 2020-2040 future period. But, the precipitation variation amount that is predicted by ANN is between -43 and 72 percent. Therefore, the results of LSSVM-WOA are more reliable and less uncertain

Monitoring network optimization in water resources, a decision-making process is available for having the best combination in between stations. The study of particle swarm algorithm or PSO algorithm were used for determination of the location and number of network observation wells. The first using the least squares support vector machine model and input parameters coordinates, evaporation, precipitation last two months, one month before groundwater table with RBF kernel function was simulated groundwater table. Then the link LSSVM models and PSO model proper location well under two scenarios were determined. In the first scenario was to determine the location of fixed piezometers 42. In the second scenario was considered as a variable number of piezometers. The results of this study showed, Given that our objective function is to minimize the difference between the observed and the simulated, in the first scenario is the least amount of difference in repeat 280 with the objective function 0. 9865. The results of the second scenario shows that the number was 28 piezometers Which represents a decrease of 55 percent compared to the initial state is the number of piezometers. In both scenarios, the distribution of points in the southern parts has increased due to the increase in the hydraulic slope and has decreased in the northern parts. In this scenario, the lowest error was repeated 338 with the aim of 0. 9145. This optimization at various stages with several stages of trial and error, show the degree of importance and superiority of the latter scenario than the first scenario.

One of the landscape management approaches is the construction of bridges along the rivers. On the other hand, the bridge scouring is a serious damage to river engineering as the main source of water and sustaining planet life. Accordingly, in this research, using field data, the accuracy of empirical methods, genetic algorithm (GA), least squares support vector machine (LSSVM), and combined method were compared in estimating scour depth of simple bridge piers. In the GA method, a number of empirical relationships were modified and the results of these modified relationships were compared with the measured scour values. In the LSSVM method, through the input of different independent parameters, model training was performed, and scour depth was predicted. In the combined method, using the LSSVM model from combining the results of different individual relations, the scour depth of the bridge piers was estimated. The results showed that modified relationships by genetic algorithm and LSSVM model have higher accuracy than empirical methods. Also, if only the parameters used in the empirical relationships are included as input parameters to the LSSVM model, the modified relationships have less error than the LSSVM model. The statistical evaluation criteria of RMSE, E, R2, and NSE for the best state of the combined method were 0. 4 m, 49%, 0. 88, and 0. 58 respectively in the training stage and 0. 52 m, 50%, 0. 7, and 0. 38 respectively in the test stage. In general, the combined method estimates scouring depth with higher accuracy than other methods.

This study focuses on the investigation of intelligent form-finding and vibration analysis of a triangular polyhedral tensegrity that is enclosed within a sphere and subjected to external loads. The nonlinear dynamic equations of the system are derived using the Lagrangian approach and the finite element method. The proposed form-finding approach, which is based on a basic genetic algorithm, can determine regular or irregular tensegrity shapes without dimensional constraints. Stable tensegrity structures are generated from random configurations and based on defined constraints (nodes located on the sphere, parallelism, and area of upper and lower surfaces), and shape finding is performed using the fitness function of the genetic algorithm and multi-objective optimization goals. The genetic algorithm's efficacy in determining the shape of structures with unpredictable configurations is evaluated in two distinct scenarios: one involving a known connection matrix and the other involving fixed or random member positions (struts and cables). The shapes obtained from the algorithm suggested in this study are validated using the force density approach, and their vibration characteristics are examined. The findings of the comparative study demonstrate the efficacy of the proposed methodology in determining the vibrational behavior of tensegrity structures through the utilization of intelligent shape seeking techniques.

In smart grids, customers will be enabled to change their strategies by electricity prices. In fact, in smart grid, we obvious a great correlation between price and load signals which show the market participants will have complex model in their decisions to maximize their profit. Many pervious-studies forecasted load or price independently. But they were not suitable for smart grid environment. To overcome this shortage, we present Multi-Input Multi-Output based Least Squares Support Vector Machine (MIMO-LSSVM) forecasted engine which can consider the correlation between price and load signals in simultaneous model. In other words, this paper presents a new hybrid algorithm to forecast day-ahead price and load in the electricity market. It consists of four stages known as a Discrete Wavelet Transform (DWT) to make valuable subsets, fuzzy mutual information (FMI) to select best input candidate and LSSVM-MIMO model. Finally, the LSSVM-MIMO parameters are optimized by a novel Improved Artificial Bee Colony (IABC) algorithm. Some forecasting indexes based on error factor are considered to shows the forecasting accuracy. Simulation results are examined on New England and New South Wales (NSW) Zone in Australia’s electricity markets. The numerical simulation results show that the proposed hybrid algorithm has good potential for forecasting simultaneous loap/price problems.