Adaptive Strategies Based on Differential Evolutionary Algorithm for Many-Objective Optimization

The decomposition-based algorithm, for example, multiobjective evolutionary algorithm based on decomposition (MOEA/D), has been proved effective and useful in a variety of multiobjective optimization problems (MOPs). On the basis of MOEA/D, the MOEA/D-DE replaces the simulated binary crossover (SBX) operator with differential evolution (DE) operator, which is used to enhance the diversity of the solutions more effectively. However, the amplification factor and the crossover probability are fixed in MOEA/D-DE, which would lead to a low convergence rate and be more likely to fall into local optimum. To overcome such a prematurity problem, this paper proposes three different adaptive operators in DE with crossover probability and amplification factors to adjust the parameter settings adaptively. We incorporate these three adaptive operators in MOEA/D-DE and MOEA/D-PaS to solve MOPs and many-objective optimization problems (MaOPs), respectively. This paper also designs a sensitive experiment for the changeable parameter η in the proposed adaptive operators to explore how η would affect the convergence of the proposed algorithms. These adaptive algorithms are tested on many benchmark problems, including ZDT, DTLZ, WFG, and MaF test suites. The experimental results illustrate that the three proposed adaptive algorithms have better performance on most benchmark problems.

Organism State Identification and Forecast Based on Comprehensive Sensor Measurement Processing with the Adaptive Algorithms

Many medical devices are using photoplethysmography (PPG) signals to estimate cardiac rate (CR), respiratory rate (RR), blood pressure (BP) and blood oxygen (SpO2). Photoplethysmography demonstrated its great potential in non-invasive monitoring of the human organism state [17], but application of this method with wearable devices is extremely difficult due to its vulnerability to motion artifacts. This paper presents implementation of a photoplethysmography device on the Raspberry Pi 3 B+ single-board computer. The work uses adaptive algorithms to study the cardiovascular system state in severe device operating conditions degrading the evaluation accuracy of CR rate and other parameters of the heart rate. Selection of the device component base and component parts was made based on their availability and multi-functionality. The manufactured mockup made it possible to carry out research to determine the most effective algorithms for digital processing of signals received from sensors. Methods of digital signal processing based on adaptive algorithms are proposed: Wiener algorithms, algorithms based on the method of least squares (MLS) and algorithms based on the Kalman filtering. In the course of measurements taken on simulation objects and volunteers invited to participate in the study, analysis of the results of various measurement processing algorithms operation was carried out. A method is proposed for assessing the accuracy of calculating the CR and analyzing effectiveness of the external noise filtering with adaptive filters. Processing the sensor measurements made it possible to monitor the heart rate with the given accuracy, as well as to predict the human body state.

PPG System Development for the Organism Physiological Parameters Monitoring with Artificial Intelligence Technologies

Monitoring such health parameters as cardiac rate (CR), respiration rate (RR), blood pressure (BP), degree of oxygen in blood (SpO2), body temperature and other requires careful approach to design and development of medical devices. New non-invasive methods introduced in measuring human physiological parameters based on photoplethysmography (PPG) demonstrated their significant potential in monitoring the state of an organism, but their use in wearable devices is largely hampered by exposure to motion artifacts. This article presents a device for photoplethysmographic studies using various adaptive algorithms for processing the registered signals. The work uses artificial intelligence technologies to monitor the heart rate exposed to external mechanical and electrical interference worsening accuracy characteristics of the system. Besides, system architecture was developed, and a device model was manufactured, which made it possible to measure the optimal algorithm for digital signal processing. When using the PPG system, methods of adaptive signal processing based on Wiener filters, filters on the method of least squares (MLS) and Kalman filtering were used. To ensure heart rate monitoring with the given accuracy, studies were performed with participation of volunteers, and analysis was carried out of the results of various signal processing algorithms operation. In the course of experimental studies, a method was proposed to estimate the heart rate calculation accuracy and to analyze the external noise filtering efficiency by adaptive algorithms. PPG designed and developed system made it possible to monitor the heart rate with the given accuracy, control the organism current state and could be used as a means of cardiovascular disease diagnostics.

On the bias and stability of the results of comparative judgment

Comparative judgment is a method that allows measurement of a competence by comparison of items with other items. In educational measurement, where comparative judgment is becoming an increasingly popular assessment method, items are mostly students’ responses to an assignment or an examination. For assessments using comparative judgment, the Scale Separation Reliability (SSR) is used to estimate the reliability of the measurement. Previous research has shown that the SSR may overestimate reliability when the pairs to be compared are selected with certain adaptive algorithms, when raters use different underlying models/truths, or when the true variance of the item parameters is below one. This research investigated bias and stability of the components of the SSR in relation to the number of comparisons per item to increase understanding of the SSR. We showed that many comparisons are required to obtain an accurate estimate of the item variance, but that the SSR can be useful even when the variance of the items is overestimated. Lastly, we recommend adjusting the general guideline for the required number of comparisons per item to 41 comparisons per item. This recommendation partly depends on the number of items and the true variance in our simulation study and needs further investigation.

Hardware in the loop simulation technology evaluation method for power steering systems

In the paper, the method of measuring and assessing of steering systems with the help of test benches with HILS technology implementation is justified in part of the adaptive algorithms of the regulation of the assistant steering torque. In the article, overall principal of creating virtual – physical system using as a physical part the rack and pinion steering system with electromechanical assistant is described. Interaction of the physical and virtual parts and fields of usage are described. Development, calibration and tweaking of the newly designed ECU and adaptive algorithms suppose a lot of testing. Time consumption and difficultness can be reduced only by the way of carrying out bench tests with usage of HILS technology.