Test Scenarios and Validation Results of POP Methodology

Autonomous robots are unsuccessful at operating in complex, unconstrained environments. They lack the ability to learn about the physical behavior of different objects through the use of vision. We combine Bayesian networks and qualitative spatial representation to learn general physical behavior by visual observation. We input training scenarios that allow the system to observe and learn normal physical behavior. The position and velocity of the visible objects are represented as qualitative states. Transitions between these states over time are entered as evidence into a Bayesian network. The network provides probabilities of future transitions to produce predictions of future physical behavior. We use test scenarios to determine how well the approach discriminates between normal and abnormal physical behavior and actively predicts future behavior. We examine the ability of the system to learn three naive physical concepts, "no action at a distance", "solidity" and "movement on continuous paths". We conclude that the combination of qualitative spatial representations and Bayesian network techniques is capable of learning these three rules of naive physics.

Download Full-text

Finding Load Inducing Test Scenarios Using Genetic Algorithms and Tree Based Encoding

Proceedings of the IEEE/ACM 42nd International Conference on Software Engineering Workshops ◽

10.1145/3387940.3392216 ◽

2020 ◽

Author(s):

Ege Apak ◽

Ayse Tosun

Keyword(s):

Genetic Algorithms ◽

Test Scenarios

Download Full-text

Product Inspection Methodology via Deep Learning: An Overview

Sensors ◽

10.3390/s21155039 ◽

2021 ◽

Vol 21 (15) ◽

pp. 5039

Author(s):

Tae-Hyun Kim ◽

Hye-Rin Kim ◽

Yeong-Jun Cho

Keyword(s):

Deep Learning ◽

Product Quality ◽

Inspection System ◽

Quality Inspection ◽

Learning Models ◽

Unified Framework ◽

Learning Techniques ◽

Inspection Systems ◽

System Maintenance ◽

Test Scenarios

In this study, we present a framework for product quality inspection based on deep learning techniques. First, we categorize several deep learning models that can be applied to product inspection systems. In addition, we explain the steps for building a deep-learning-based inspection system in detail. Second, we address connection schemes that efficiently link deep learning models to product inspection systems. Finally, we propose an effective method that can maintain and enhance a product inspection system according to improvement goals of the existing product inspection systems. The proposed system is observed to possess good system maintenance and stability owing to the proposed methods. All the proposed methods are integrated into a unified framework and we provide detailed explanations of each proposed method. In order to verify the effectiveness of the proposed system, we compare and analyze the performance of the methods in various test scenarios. We expect that our study will provide useful guidelines to readers who desire to implement deep-learning-based systems for product inspection.

Download Full-text

Design and Implementation of Maiden Dual-Level Controller for Ameliorating Frequency Control in a Hybrid Microgrid

Energies ◽

10.3390/en14092418 ◽

2021 ◽

Vol 14 (9) ◽

pp. 2418

Author(s):

Abdul Latif ◽

S. M. Suhail Hussain ◽

Dulal Chandra Das ◽

Taha Selim Ustun

Keyword(s):

Frequency Control ◽

Electric Heater ◽

Maximum Frequency ◽

Stabilization Time ◽

Control Approach ◽

Proportional Integral ◽

Hybrid Microgrid ◽

Performance Results ◽

System Frequency ◽

Test Scenarios

It is known that keeping the power balance between generation and demand is crucial in containing the system frequency within acceptable limits. This is especially important for renewable based distributed hybrid microgrid (DHμG) systems where deviations are more likely to occur. In order to address these issues, this article develops a prominent dual-level “proportional-integral-one plus double derivative {PI−(1 + DD)} controller” as a new controller for frequency control (FC) of DHμG system. The proposed control approach has been tested in DHμG system that consists of wind, tide and biodiesel generators as well as hybrid plug-in electric vehicle and an electric heater. The performance of the modified controller is tested by comparing it with standard proportional-integral (PI) and classical PID (CPID) controllers considering two test scenarios. Further, a recently developed mine blast technique (MBA) is utilized to optimize the parameters of the newly designed {PI − (1 + DD)} controller. The controller’s performance results are compared with cases where particle swarm optimization (PSO) and firefly (FF) techniques are used as benchmarks. The superiority of the MBA-{PI − (1 + DD)} controller in comparison to other two strategies is illustrated by comparing performance parameters such as maximum frequency overshoot, maximum frequency undershoot and stabilization time. The displayed comparative objective function (J) and JFOD index also shows the supremacy of the proposed controller. With this MBA optimized {PI − (1 + DD)} controller, frequency deviations can be kept within acceptable limits even with high renewable energy penetration.

Download Full-text

Investigating the vibrational behaviour of a rotating two-blade propeller by using a self-tracking method

Proceedings of the Institution of Mechanical Engineers Part C Journal of Mechanical Engineering Science ◽

10.1177/0954406218764514 ◽

2018 ◽

Vol 233 (3) ◽

pp. 835-847 ◽

Cited By ~ 1

Author(s):

Mohammad-Reza Ashory ◽

Farhad Talebi ◽

Heydar R Ghadikolaei ◽

Morad Karimpour

Keyword(s):

Natural Frequency ◽

Measurement Errors ◽

Mode Shapes ◽

Model Parameters ◽

Test Results ◽

Modal Assurance Criterion ◽

Tracking Method ◽

Strong Resemblance ◽

Test Scenarios ◽

Good Agreement

This study investigated the vibrational behaviour of a rotating two-blade propeller at different rotational speeds by using self-tracking laser Doppler vibrometry. Given that a self-tracking method necessitates the accurate adjustment of test setups to reduce measurement errors, a test table with sufficient rigidity was designed and built to enable the adjustment and repair of test components. The results of the self-tracking test on the rotating propeller indicated an increase in natural frequency and a decrease in the amplitude of normalized mode shapes as rotational speed increases. To assess the test results, a numerical model created in ABAQUS was used. The model parameters were tuned in such a way that the natural frequency and associated mode shapes were in good agreement with those derived using a hammer test on a stationary propeller. The mode shapes obtained from the hammer test and the numerical (ABAQUS) modelling were compared using the modal assurance criterion. The examination indicated a strong resemblance between the hammer test results and the numerical findings. Hence, the model can be employed to determine the other mechanical properties of two-blade propellers in test scenarios.

Download Full-text