PSO-ELM with Modified Acceleration Coefficients for Classifying the Active Compound

Background: The classification of active compounds based on their function using machine learning is essential for predicting the function of new active compounds quickly. These classification results are beneficial and accelerate the work of laboratory assistants in identifying the function of active compounds. In this study, an active compound is represented by the simplified molecular-input line-entry system (SMILES) code. Objective: This paper proposes a modified acceleration coefficient to improve the PSO-ELM performance for predicting the function of the SMILES code. Method: The research uses a machine-learning algorithm that is a combination of the Particle Swarm Optimization and Extreme Learning Machine (PSO-ELM). ELM is used to classify the SMILES code, while PSO is used to optimize ELM parameters, i.e., weight, bias, and the number of hidden neurons. The important parameters that significantly influence the PSO performance are acceleration coefficients. The acceleration coefficients, that is, modified sigmoid-based acceleration coefficient (SBAC), are introduced and compared with seven other acceleration coefficients. Results: The experimental results show that the sensitivity, specificity, accuracy, and the area under the curve (AUC) of the proposed acceleration coefficients outperforms all other acceleration coefficients. The increased accuracy of the proposed paper can reach up to 2.64%, 5.84%, 7.93%, 8.44%, and 16.29% for Support Vector Machine (SVM), decision tree, AdaBoost, MLP Classifier, and Gaussian Naïve Bayes algorithms, respectively. Conclusion: The acceleration coefficients affect the prediction accuracy of the SMILES code classification. The proposed acceleration coefficients improve the performance of the PSO-ELM for predicting the function of the SMILES code.

Optimal parameter values of PID controller for DC motor based on modified particle swarm optimization with adaptive inertia weight

A significant problem in the control field is the adjustment of PID controller parameters. Because of its high nonlinearity property, control of the DC motor system is difficult and mathematically repetitive. The particle swarm optimization PSO solution is a great optimization technique and a promising approach to address the problem of optimum PID controller results. In this paper, a modified particle swarm optimization PSO method with four inertia weight functions is suggested to find the global optimum parameters of the PID controller for speed and position control of the DC motor. Benchmark studies of inertia weight functions are described. Two scenarios have been suggested in order to modify PSO including the first scenario called M1-PSO and the second scenario called M2-PSO, as well as classical PSO algorithms. For the first scenario, the modification of the PSO was done based on changing the four inertia weight functions, social and personal acceleration coefficient, while in the second scenario, the four inertia weight functions have been changed but the social and personal acceleration coefficient stayed constant during the algorithm implementation. The comparison between the presented scenarios and traditional PID was carried out and satisfied simulation results have shown that the first scenario has rapid search speeds, and very effective and fast implementation compared to the second scenario and classical PSO and even improved PSO technique. Moreover, the proposed approach has a fast searching speed compared to classical PSO. However, it has been found that the classical PSO algorithm has a premature, inaccurate and local convergence process when solving complex optimization issues. The presented algorithm is proposed to increase the search speed of the original PSO.

The Expression of Seismic Surrounding Rock Pressure for a Shallow Tunnel Is Derived Using the Pseudostatic Method

Terzaghi developed a generalized expression of the vertical surrounding rock pressures of shallow tunnels by considering the limit equilibrium of soil masses. In this paper, based on the Terzaghi failure mode, the pseudostatic method is used to derive this expression under seismic loading conditions. The surrounding rock in the fractured zone of the tunnel side wall is analyzed as an isolated body using the limit equilibrium method to obtain the explicit calculation expressions of the horizontal surrounding rock pressures of a shallow tunnel under seismic loading. Case analysis indicates that the proposed method is feasible. In addition, the influence of the seismic acceleration coefficient on surrounding rock pressures is further discussed. The results show that the horizontal surrounding rock pressure decreases with the increase of seismic acceleration coefficients. The vertical surrounding rock pressure increases as the horizontal seismic acceleration coefficient increases, and it decreases with the increase of the vertical seismic acceleration coefficient, and the effect of the seismic acceleration coefficient on surrounding rock pressure is significant. The study results can provide reference for the seismic safety evaluation and structural seismic design of shallow tunnels.

Transient Pulse Test for Non-Darcy Flow Behaviors and Hydromechanical Coupling Effect of Fractured Limestone

In this paper, the transient pulse test is used to study the permeability and hydromechanical coupling effect of the fractured limestone. The permeability parameters (permeability, β factor of non-Darcy flow, and acceleration coefficient) of non-Darcy flow in fractured limestone are obtained by experimental data. The experimental results show that, in the process of transient seepage test of fractured Maokou limestone, the relationship between hydraulic pressure gradient and seepage velocity does not conform to Darcy’s law but meets Forchimer relationship. The relationship between hydraulic pressure difference and time can be fitted by quartic polynomial. The larger the confining pressure is, the more obvious the non-Darcy seepage effect of fractured rock seepage is. The seepage of rock fracture under high confining pressure is a highly nonlinear time-varying seepage. The permeability coefficient of rock decreases with the increase of volume stress. Under the action of low volume stress, the relationship between permeability coefficient and stress is more sensitive, while under the action of high volume stress, the relationship between permeability coefficient and volume stress is not significant. In the process of volume stress increasing, the β factor of non-Darcy flow appears negative. Under the action of low volume stress, the acceleration coefficient and β factor of non-Darcy flow increase, while under the action of high volume stress, the acceleration coefficient and β factor of non-Darcy flow decrease.