In the paper, it is proposed to develop a machine learning based intelligent defect detection system for metal products. The common machine vision system has the surface (stain, shallow pit, shallow tumor, scratches, Edge defects, pattern defects) detection, or for the processing of the size, diameter, diameter, eccentricity, height, thickness and other parts of the non-contact numerical parameters of detection. Considering the quality of the work piece and the defects of the standard, so for the quality of customized testing requirements, the study is the development of machine vision and machine learning metal products defect detection system, mainly composed of three procedures: Image preprocessing, training procedures and testing procedures. The system architecture consists of three parts: (1) Image preprocessing: we first use the machine vision. OPENCV to carry out the image pre-processing part of the product before the detection. (2) Training procedures: The algorithm of the machine learning includes the convolution neural network (CNN), chunk-max pooling is used to train the program, and the generative adversarial network (GAN) based architecture is used to solve the problem of small datasets for surface defects. (3) Testing procedures：The Python language is used to write the program and implement the testing procedures with the GPU-Based embedded hardware In industries, collecting training dataset is usually costly and related methods are highly dataset-dependent. So most companies cannot provide Big-data to be analyzed or applied. By the experimental results, the recognition accuracy can be obviously improved as increasing data augmentation by GAN-Based samples maker. Manual inspection is labor intensive, costly and less in efficiency. Therefore, this study will contribute to technological innovation, industry, national development and other applications. (1) The use of intelligent machine learning technology will make the industry 4.0 technology more sophisticated. (2) It will make the development of equipment industry be better by the machine learning applications. (3) It will increase the economics and productivity of countries for the aging of the population by machine learning.

EVWUDFW²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² IX]]\ VOLGLQJ PRGH FRQWURO SURSRUWLRQDOLQWHJUDO 3, ZLQG WXUELQH JHQHUDWRU :7* SHUPDQHQW PDJQHW V\QFKURQRXV JHQHUDWRU 306*

No abstract

Physical fitness and health of white collar business person is getting worse and worse in recent years. Therefore, it is necessary to develop a system which can enhance physical fitness and health for people. Although the exercise prescription can be generated after diagnosing for customized physical fitness and healthcare. It is hard to meet individual execution needs for general scheduling of physical fitness and healthcare system. So the main purpose of this research is to develop an intelligent scheduling of execution for customized physical fitness and healthcare system. The results of diagnosis and prescription for customized physical fitness and healthcare system will be generated by fuzzy logic Inference. Then the results of diagnosis and prescription for customized physical fitness and healthcare system will be scheduled and executed by intelligent computing. The scheduling of execution is generated by using genetic algorithm method. It will improve traditional scheduling of exercise prescription for physical fitness and healthcare. Finally, we will demonstrate the advantages of the intelligent scheduling of execution for customized physical fitness and healthcare system.

This paper presents an intelligent solar charging system with fuzzy logic control method. With the scarce energy source and the worsening environmental pollution, how to create and use a clean and never exhausted energy is becoming very important day by day. This solar charging system is composed of a solar cell, a charger, batteries, a buck converter and a digital signal processor. In the meantime, it also combines the fuzzy logic method with the tactics of charging to improve the efficiency of charging, suppress the abnormal battery temperature rise, lengthen the battery's life, and reduce the waste used. Finally, experimental and simulation results are shown to demonstrate the effectiveness and validity of the system.

Knowledge graphs are useful sources for various AI applications, however the basic paradigm to support pilot training is still unclear. In the paper, It is proposed to generate the customized knowledge graph of flight trainings using machine learning method for the flight training program. In order to provide the successful key to the further understanding of the learning problems between the students and the instructors. In this research, we collected data from an aeronautical academic in Taiwan that students were trained for Recreation Pilot License Program. We performed a test on 24 students at the first of each training course, 16 data of collected been used on building the module, 8 of them used to exam the module. There are 12 courses in the training program, and 30 hours total time were suggested by academic. The score which we applied on test were based on LCG method which is the sum of Maneuver and SRM Grades. For the indicators of course component in Learner Centered Grading, namely (a) CCS1: Operation & Effect of Controls; (b) CCS2: Straight & Level; (c) CCS3: Climbing & Descending; (d) CCS4: Turning; (e) CCS5: Stalling; (f) CCS6: Revision; (g) CCS7: Circuits; (h) CCS8: Cross-Wind Training; (i) CCS9: Circuit Emergency; (j) CCS10: Solo Circuit; (k) CCS11: Forced Landing; and (l) CCS12: Precautionary & Searching Landing. Through the method of Knowledge Graph, we deduct and predict the number of hours that need to be added for each student’s learning. Using the dynamic knowledge graph to display the key issues of the course learning continuously, and make follow-up decisions for the students, instructors and airliners.

In the paper, a cloud-dust based intelligent maximum power analysis system for photovoltaic is proposed. In order to resolve NP problem for photovoltaic, factors of photovoltaic are integrated to cloud-dust based intelligent maximum power analysis system for computing. This study is the development of the maximum power analysis system for photovoltaic, to improve the solar panels effects of the different region and enable them to get maximum efficiency of the power generation. The design methodology of this study includes: (1) The monitoring and control Module (2) The prediction and evaluation module (3) The performance diagnosis module (4) The maintenance prescription module. At last, we can find the advantages of the cloud-dust based intelligent maximum power analysis system for photovoltaic. It increases overall competitive performance of products, reduces cost of products and consummation rates of human resources.

In this paper, an intelligent solar panel cleaning system that monitors the output of solar panels is designed. The output voltage of the solar panel is used to decide if the solar panel needs to clean or not. The control system is developed using Lab-VIEW. The direction and position of the system is set by the light sensor, which is parallel to the direction of sunlight. The data from the light sensors, along with the fuzzy logic control software developed using Lab-View determines the control commands for the stepper motors controlling the cleaning process. The commands are stop, forward or reverse and the cleaning process is repeated until the generated power output of the solar panels is sufficient. The cleaning process is performed in real-time to maintain the power generating capacity of the solar cells.