Anomaly Detection in Automotive Industry Using Clustering Methods—A Case Study

Intelligent Monitoring Model for Fall Risks of Hospitalized Elderly Patients

Early detection of high fall risk is an important process of fall prevention in hospitalized elderly patients. Hospitalized elderly patients can face several falling risks. Monitoring systems can be utilized to protect health and lives, and monitoring models can be less effective if the alarm is not invoked in real time. Therefore, in this paper we propose a monitoring prediction system that incorporates artificial intelligence. The proposed system utilizes a scalable clustering technique, namely the Catboost method, for binary classification. These techniques are executed on the Snowflake platform to rapidly predict safe and risky incidence for hospitalized elderly patients. A later stage employs a deep learning model (DNN) that is based on a convolutional neural network (CNN). Risky incidences are further classified into various monitoring alert types (falls, falls with broken bones, falls that lead to death). At this phase, the model employs adaptive sampling techniques to elucidate the unbalanced overfitting in the datasets. A performance study utilizes the benchmarks datasets, namely SERV-112 and SV-S2017 of the image sequences for assessing accuracy. The simulation depicts that the system has higher true positive counts in case of all health-related risk incidences. The proposed system depicts real-time classification speed with lower training time. The performance of the proposed multi-risk prediction is high at 87.4% in the SERV-112 dataset and 98.71% in the SV-S2017 dataset.

Metaheuristics-Based Optimization of a Robust GAPID Adaptive Control Applied to a DC Motor-Driven Rotating Beam with Variable Load

This work aims to analyze two metaheuristics optimization techniques, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), with six variations each, and compare them regarding their convergence, quality, and dispersion of solutions. The optimization target is the Gaussian Adaptive PID control (GAPID) to find the best parameters to achieve enhanced performance and robustness to load variations related to the traditional PID. The adaptive rule of GAPID is based on a Gaussian function that has as adjustment parameters its concavity and the lower and upper bound of the gains. It is a smooth function with smooth derivatives. As a result, it helps avoid problems related to abrupt increases transition, commonly found in other adaptive methods. Because there is no mathematical methodology to set these parameters, this work used bio-inspired optimization algorithms. The test plant is a DC motor with a beam with a variable load. Results obtained by load and gain sweep tests prove the GAPID presents fast responses with very low overshoot and good robustness to load changes, with minimal variations, which is impossible to achieve when using the linear PID.

Migration-Based Moth-Flame Optimization Algorithm

Moth–flame optimization (MFO) is a prominent swarm intelligence algorithm that demonstrates sufficient efficiency in tackling various optimization tasks. However, MFO cannot provide competitive results for complex optimization problems. The algorithm sinks into the local optimum due to the rapid dropping of population diversity and poor exploration. Hence, in this article, a migration-based moth–flame optimization (M-MFO) algorithm is proposed to address the mentioned issues. In M-MFO, the main focus is on improving the position of unlucky moths by migrating them stochastically in the early iterations using a random migration (RM) operator, maintaining the solution diversification by storing new qualified solutions separately in a guiding archive, and, finally, exploiting around the positions saved in the guiding archive using a guided migration (GM) operator. The dimensionally aware switch between these two operators guarantees the convergence of the population toward the promising zones. The proposed M-MFO was evaluated on the CEC 2018 benchmark suite on dimension 30 and compared against seven well-known variants of MFO, including LMFO, WCMFO, CMFO, CLSGMFO, LGCMFO, SMFO, and ODSFMFO. Then, the top four latest high-performing variants were considered for the main experiments with different dimensions, 30, 50, and 100. The experimental evaluations proved that the M-MFO provides sufficient exploration ability and population diversity maintenance by employing migration strategy and guiding archive. In addition, the statistical results analyzed by the Friedman test proved that the M-MFO demonstrates competitive performance compared to the contender algorithms used in the experiments.