No abstract

As science and technology improve, more and more complex global optimization difficulties arise in real-life situations. Finding the most perfect approximation and optimal solution using conventional numerical methods is intractable. Metaheuristic optimization approaches may be effective in achieving powerful global optimal solutions for these complex global optimization situations. Therefore, this paper proposes a new game-based algorithm called the Running City Game Optimizer (RCGO), which mimics the game participant's activity of playing the running city game. The RCGO is mathematically established by three newfangled search strategies: siege, defensive, and eliminated selection. The performance of the proposed RCGO algorithm in optimization is comprehensively evaluated on a set of seventy-six benchmark problems and eight engineering optimization scenarios. Statistical and comparative results show that RCGO is more competitive with other state-of-the-art competing approaches in terms of solution quality and convergence efficiency, which stems from a proper balance between exploration and exploitation. Additionally, in the case of engineering optimization scenarios, the proposed RCGO is able to deliver superior fitting and occasionally competitive outcomes in optimization applications. Thus, the proposed RCGO is a viable optimization tool to easily and efficiently handle various optimization problems.

Rolling element bearings constitute the key parts on rotating machinery, and their fault diagnosis is of great importance. In many real bearing fault diagnosis applications, the number of fault data is much less than the number of normal data, i.e. the data are imbalanced. Many traditional diagnosis methods will get low accuracy because they have a natural tendency to favor the majority class by assuming balanced class distribution or equal misclassification cost. To deal with imbalanced data, in this article, a novel two-step fault diagnosis framework is proposed to diagnose the status of rolling element bearings. Our proposed framework consists of two steps for fault diagnosis, where Step 1 makes use of weighted extreme learning machine in an effort to classify the normal or abnormal categories, and Step 2 further diagnoses the underlying anomaly in detail by using preliminary extreme learning machine. In addition, gravitational search algorithm is applied to further extract the significant features and determine the optimal parameters of the weighted extreme learning machine and extreme learning machine classifiers. The effectiveness of our proposed approach is testified on the raw data collected from the rolling element bearing experiments conducted in our Institute, and the empirical results show that our approach is really fast and can achieve the diagnosis accuracies more than 96%.

Seagull optimization algorithm (SOA) has the disadvantages of low convergence accuracy, weak population diversity, and tendency to fall into local optimum, especially for high dimensional and multimodal problems. To overcome these shortcomings, initially, in this study, a shared SOA (SSOA) is proposed based on the combination of a sharing multi-leader strategy with a self-adaptive mutation operator. In addition, seven new variants of the SSOA algorithm are proposed employing the Gaussian mutation operator, Cauchy mutation operator, Lévy flights mutation operator, improved Tent chaos mutation operator, neighborhood centroid opposition-based learning mutation operator, elite opposition-based learning mutation operator, and simulated annealing algorithm combined with other mutation operators, namely, GSSOA, CSSOA, LFSSOA, ITSSOA, ESSOA, NSSOA, and CMSSOA, respectively. Then, the performance of these variants was evaluated on 23 benchmark functions, and the various performances of the best variant were evaluated on a comprehensive set of 43 benchmark problems and three real-world problems compared to other optimizers. Experimental and statistical results demonstrate that the proposed CMSSOA algorithm outperforms other variants of the SSOA algorithm and competitor approaches.

No abstract