Self-learning Model Research Articles

自适应光学系统的自学习控制模型及其验证

自适应光学系统的自学习控制模型及其验证

Self-learning model predictive control for dynamic activation of structural thermal mass in residential buildings

Changes in the electricity supply system induce the challenge of matching the highly fluctuating and unpredictable renewable energy generation with the yet inflexible electricity demand. This leads to an increasing demand for energy storage and demand-flexibility. Electrification of residential heating systems in combination with advanced controls utilizing dynamically the structural thermal mass (STM) of buildings as thermal storage could provide some of the required demand flexibility.In this work, a model predictive control (MPC) algorithm is developed and applied within a simulation framework to control dynamic heating operation as a measure of STM based residential load shifting (LS). The self-learning algorithm is functional without extensive measurement data or expert knowledge for parametrization. It optimizes heating operations required for LS according to a dynamic primary energy factor signal, while observing transient thermal comfort constraints. The implemented auto-regressive black-box model with explanatory variables predicts thermal conditions within the observed thermal zone with sufficient quality to support MPC. Based on that model, the control algorithm successfully activates STM as a measure of LS according to the given primary energy (PE) oriented utility function. For the observed system, the PE demand can be reduced by 3–7% while maintaining or even improving the thermal comfort.

Optimization of Rolling Force Self-learning Model in Unsteady Process of Hot Rolling

Optimization of Rolling Force Self-learning Model in Unsteady Process of Hot Rolling

A novel parallel object-tracking behavior algorithm based on dynamics for data clustering

Recently, many evolutionary algorithms (EAs) have been used to solve clustering problem. However, compared to K-means which is a simple and fast clustering algorithm, these EA-based clustering algorithms take too much computation time. In addition, the parameters of most EAs are fixed or dynamical adjustment by a simple method on different datasets, and it will cause that the performance of these algorithms is good on some datasets but bad on others. In order to overcome these disadvantages, a novel parallel object-tracking behavior algorithm (POTBA) based on dynamics is proposed in this paper. The proposed algorithm consists of three different models which are parallel object-tracking model, parameters self-learning model and energy model, respectively. First, the parallel object-tracking model is designed to accelerate the computation speed and avoid local minima. Second, the parameters of POTBA are self-adjusted by the parameters self-learning model. Third, the energy model is introduced to depict energy changes of POTBA during the evolutionary process. The correctness and convergence properties of POTBA are analyzed theoretically. Moreover, the effectiveness and parallelism of POTBA are evaluated through several standard datasets, and the experimental results demonstrate that POTBA exhibits superior overall performance than five other state-of-the-art algorithms. In the aspect of search performance, the results of POTBA are better than other comparison algorithms on most used datasets. In the aspect of time performance, the time overhead of POTBA is significantly reduced through parallel computing. When the number of processors increases to 32, the computation time of POTBA is less or close to K-means which is the fastest comparison algorithm.

A Self-learning Model to Detect Icing Alarm Based on Multi-agents for Wind Park

A Self-learning Model to Detect Icing Alarm Based on Multi-agents for Wind Park

Deadline-Aware MapReduce Job Scheduling with Dynamic Resource Availability

As MapReduce is becoming ubiquitous in large-scale data analysis, many recent studies have shown that the performance of MapReduce could be improved by different job scheduling approaches, e.g., Fair Scheduler and Capacity Scheduler. However, most exiting MapReduce job schedulers focus on the scenario that MapReduce cluster is stable and pay little attention to the MapReduce cluster with dynamic resource availability. In fact, MapReduce cluster resources may fluctuate as there is a growing number of Hadoop clusters deployed on hybrid systems, e.g., infrastructure powered by mix of traditional and renewable energy, and cloud platforms hosting heterogeneous workloads. Thus, there is a growing need for providing predictable services to users who have strict requirements on job completion times in such dynamic environments. In this paper, we propose, RDS , a Resource and Deadline-aware Hadoop job Scheduler that takes future resource availability into consideration when minimizing job deadline misses. We formulate the job scheduling problem as an online optimization problem and solve it using an efficient receding horizon control algorithm. To aid the control, we design a self-learning model to estimate job completion times. We further extend the design of RDS scheduler to support flexible performance goals in various dynamic clusters. In particular, we use flexible deadline time bounds instead of the single fixed job completion deadline. We have implemented RDS in the open-source Hadoop implementation and performed evaluations with various benchmark workloads. Experimental results show that RDS substantially reduces the penalty of deadline misses by at least 36 and 10 percent compared with Fair Scheduler and Earliest Deadline First (EDF) scheduler, respectively. In a Hadoop cluster running partially on renewable energy, the experimental result shows the green power based resource prediction approach can further reduce the penalty of deadline misses by 16 percent compared to Auto-Regressive Integrated Moving Average (ARIMA) prediction approach.

THE STUDENTS’ CRITICAL THINKING SKILLS ON SCIENCE LEARNING SUBJECT AT 5TH GRADE IN INQUIRY LEARNING

Abstra ct : This study are aims to determine the differences in students' critical thinking skills using inquiry learning model. The methods that used in this research is experimental design with Nonequivalent Control Group Design (pre-test and post test in experimental group and control group). The samples are taken using simple random sampling technique with 5 th grade of A class (30) and B class (31). This study using test and documentation as the data collection techniques which will be processed to answer the specified hypothesis. The technique of data analysis using t-test result of critical thinking skills in experimental group and control group with level of 0,05 are a proved result of hypothesis-table ≤ ttable ≥ ttable that is -2,001 ≤ 4,096 ≥ 2,001 then Ha is accepted, means that both group are not the same. This indicates that there is a differences of students' critical thinking skills between students using inquiry models with students using conventional learning model (lectures and assignment). It can be concluded that the inquiry self-learning models can influence the students' critical thinking skills with the percentage score of the critical thinking skills are 55% in the experimental group and the control group are 27%. Based on the results of these research, the suggestions that researchers can convey that inquiry learning models can be used as an alternative in science learning. Keywords : Inquiry Learning Model, Critical Thinking Skill

Deep reinforcement learning enabled self-learning control for energy efficient driving

To address the air pollution problems and reduce greenhouse gas emissions (GHG), plug-in hybrid electric vehicles (PHEV) have been developed to achieve higher fuel efficiency. The Energy Management System (EMS) is a very important component of a PHEV in achieving better fuel economy and it is a very active research area. So far, most of the existing EMS strategies are either just simply following predefined rules that are not adaptive to changing driving conditions; or heavily relying on accurate prediction of future traffic conditions. Deep learning algorithms have been successfully applied to many complex problems and proved to even outperform human beings in some tasks (e.g., play chess) in recent years, which shows the great potential of such methods in practical engineering problems. In this study, a deep reinforcement learning (Deep Q-network or DQN) based PHEV energy management system is designed to autonomously learn the optimal fuel/electricity splits from interactions between the vehicle and the traffic environment. It is a fully data-driven and self-learning model that does not rely on any prediction, predefined rules or even prior human knowledge. The experiment results show that the proposed model is capable of achieving 16.3% energy savings (with the designed PHEV simulation model) on a typical commute trip, compared to conventional binary control strategies. In addition, a dueling Deep Q-network with dueling structure (DDQN) is also implemented and compared with single DQN in particular with respect to the convergence rate in the training process.

A Co-Point Mapping-Based Approach to Drivable Area Detection for Self-Driving Cars

The randomness and complexity of urban traffic scenes make it a difficult task for self-driving cars to detect drivable areas. Inspired by human driving behaviors, we propose a novel method of drivable area detection for self-driving cars based on fusing pixel information from a monocular camera with spatial information from a light detection and ranging (LIDAR) scanner. Similar to the bijection of collineation, a new concept called co-point mapping, which is a bijection that maps points from the LIDAR scanner to points on the edge of the image segmentation, is introduced in the proposed method. Our method positions candidate drivable areas through self-learning models based on the initial drivable areas that are obtained by fusing obstacle information with superpixels. In addition, a fusion of four features is applied in order to achieve a more robust performance. In particular, a feature called drivable degree (DD) is proposed to characterize the drivable degree of the LIDAR points. After the initial drivable area is characterized by the features obtained through self-learning, a Bayesian framework is utilized to calculate the final probability map of the drivable area. Our approach introduces no common hypothesis and requires no training steps; yet it yields a state-of-art performance when tested on the ROAD-KITTI benchmark. Experimental results demonstrate that the proposed method is a general and efficient approach for detecting drivable area.

Unsupervised Deep Learning Features for Lung Cancer Overall Survival Analysis.

Lung cancer overall survival analysis using computed tomography (CT) images plays an important role in treatment planning. Most current analysis methods involve hand-crafted image features for survival time prediction. However, hand-crafted features require domain knowledge and may lack specificity to lung cancer. Advanced self-learning models such as deep learning have showed superior performance in many medical image tasks, but they require large amount of data which is difficult to collect for survival analysis because of the long follow-up time. Although data with survival time is difficult to acquire, it is relatively easy to collect lung cancer patients without survival time. In this paper, we proposed an unsupervised deep learning method to take advantage of the unlabeled data for survival analysis, and demonstrated better performance than using hand-crafted features. We proposed a residual convolutional auto encoder and trained the model using images from 274 patients without survival time. Afterwards, we extracted deep learning features through the encoder model, and constructed a Cox proportional hazards model on 129 patients with survival time. The experiment results showed that our unsupervised deep learning feature gained better performance (C-Index = 0.70) than using hand-crafted features (C-Index = 0.62). Furthermore, we divided the patients into two groups according to their Cox hazard value. Kaplan-Meier analysis indicated that our model can divide patients into high and low risk groups and the survival time of these two groups had significant difference (p < 0.01).

Pendidikan Nilai Keberagaman dalam Membangun Kepribadian Sehat (Studi Deskriptif terhadap Siswa Madrasah Aliyah Darul Arqam Muhammadiyah Garut)

&#x0D; Abstract: Potency of the high absorption obtained if learning in effective. Effective learning occurs when students are placed as individual active and direct contact with the subject matter. This research aims to reduce the limitations of the tool (a computer&#x0D; or laptop) and maximizing existing facilities (hand phone) with the aim of achieving effective learning that puts students as subjects of learning. This study is a research field for conducting comparative academic performance of two models of learning. The first learning model is simulation learning and the second model is self-learning via mobile&#x0D; facility. Self-learning materials in this research is the material in the form of video 3GP and transferred to the student’s mobile. The research population is 85 students and a sample taken by 70 students. The data in this study is the performance of students from simulation learning model and self- learning model based 3GP video. Data analysis using&#x0D; Inferential statistical, namely the t-test. Data analysis was performed after the fulfillment of the requirements for normality of data. The results of hypothesis testing obtained the results as following: The value t count bigger than t table (5.957&gt; 2.025). Thus Ha is&#x0D; received and Ho is rejected (significance below or equal to 0.05 so Ha is received). Means that there are significant differences between simulation learning model and self-learning model based 3GP video.&#x0D; Keywords: Learning Media, 3GP Video.&#x0D;

Pembelajaran Mandri Berbasis Video 3GP

&#x0D; Abstract: Potency of the high absorption obtained if learning in effective. Effective learning occurs when students are placed as individual active and direct contact with the subject matter. This research aims to reduce the limitations of the tool (a computer or laptop) and maximizing existing facilities (hand phone) with the aim of achieving effective learning that puts students as subjects of learning.&#x0D; This study is a research field for conducting comparative academic performance of two models of learning. The first learning model is simulation learning and the second model is self-learning via mobile facility. Self-learning materials in this research is the material in the form of video 3GP and transferred to the student’s mobile. The research population is 85 students and a sample taken by 70 students. The data in this study is the performance of students from simulation learning model and self- learning model based 3GP video. Data analysis using inferential statistical, namely the t-test. Data analysis was performed after the fulfillment of the requirements for normality of data.&#x0D; The results of hypothesis testing obtained the results as following: The value t count bigger than t table (5.957&gt; 2.025). Thus Ha is received and Ho is rejected (significance below or equal to 0.05 so Ha is received). Means that there are significant differences between simulation learning model and self-learning model based 3GP video.&#x0D; Keywords: Learning Media, 3GP Video.&#x0D;

Outlier Detection for Control Process Data Based on Improved ARHMM

In view of the difficulty of accurate online detection for massive data collecting real-timely in a strong noise environment during control process, an order self-learning Autoregressive Hidden Markov Model (ARHMM) algorithm is proposed to carry out online outlier detection in industrial control process. The algorithm utilizes AR model to fit the time series and makes use of HMM as basic detection tool, which can avoid the deficiency of presetting the threshold in traditional detection methods. In order to update parameters of ARHMM online, the structure of traditional Brockwell–Dahlhaus–Trindade (BDT) algorithm is improved to be a double-iterative structure in which iterative calculation from both time and order is applied respectively. With the purpose of reducing the influence of outlier on parameter update of ARHMM, the strategies of detection-before-update and detection-based-update are adopted, which also improve the robustness of algorithm. Subsequent simulation by model data and practical application verify the accuracy, robustness and property of online detection of the algorithm. According to the result, it is obvious that new algorithm proposed in this paper is more suitable for outlier detection of control process data in process industry.

The Durban Holocaust Centre Educators and the Origins of their Historical and Pedagogical Knowledge

ABSTRACTEducation at Holocaust museums worldwide often falls to volunteer museum educators. The Durban Holocaust Centre in South Africa is no different. We set out to understand who the educators at the Durban Holocaust Centre were, where their historical and pedagogical knowledge came from, and to examine the connection between the two. The study revealed the diverse nature of the museum educators’ biographies as well as their motivations for guiding. Their knowledge acquisition was generally a blend of formal objectivist and informal constructivist methods. It emerged that the self-learning model was successful as the educators were highly professional and sufficiently motivated.

Neural Network Self-Learning Model for Complex Assessment of Drinking Water Safety for Consumers

We need to take into complex assessment a set of influencing factors of drinking water safety. This raises the task of developing an integrated methodology assessing the safety of drinking water that reaches the consumers. For the integrated assessment of the safety of drinking water, the method of clustering was chosen, namely, the neural network method of Kohonen self-organizing maps. Zones were separated by the method of cluster neural network analysis. The zones are characterized by different content of metal cations in drinking water, levels of carcinogenic and non-carcinogenic risk to the health of the child population, and the probability of the receipt of metal cations with potable water to consumers.

The Integrated Maintenance Information System of Marine Nuclear Power Platform Based on Data Cloud

Marine Nuclear Power Platform (MNPP) is far away from land, and the repair work on sea is inconvenient and expensive, which greatly affects its stability and economy. On the basis of the deep application of ship information, an integrated maintenance information system for MNPP is designed aiming at the operation and maintenance of the whole life cycle. The integrated maintenance information system, taking the data cloud as the basic means, horizontally includes the design, construction, installation, operation, maintenance and decommissioning stage of MNPP, and contains longitudinally equipment source and basic data, system self-learning model and platform integrated model. The integrated maintenance information system would eventually form a unified data set and model set of MNPP, which could provide intelligence operation and maintenance, fault assessment, repair decision and other technique supports for MNPP, and shed great light on the long-term development of MNPP.

Research and Practice of a Self-Learning Based Interactive Teaching Model for Electronic Circuit Course

This paper studies the construction of self-learning model based on network environment. The application in the electronic circuit course is based on the characteristics and development of advanced education and teaching mode at home and abroad. By making full use of the network environment, applying the network’ Interactive Q & A and record common problems, teachers can grasp students’ learning status all the time. Teachers preparing teaching on the learning state of students will lead to a substantial increase in classroom teaching efficiency and teaching effectiveness. The learners will have a comprehensive grasp of the basic knowledge of the course and basic ideas. Through the course teaching practice, the model can fully mobilize the enthusiasm of students and cultivate learners' active thinking and self-learning ability, significantly improving the quality of teaching.

Physical and mathematical modeling of pollutant emissions when burning peat

The article presents an original neural network model of CO dispersion around the experimentally simulated peat fire. It is a self-learning model considering both the measured CO concentrations in the smoke cloud and the refined coefficients of the main equation. The method is recommended for the development of air quality control and forecasting systems.

Developing a self-learning model based on open-ended questions to increase the students' creativity in calculus

This article reports on research that is aimed at developing and producing a self-learning model based on open-ended questions or the PMBST (accronym form Indonesian) model to increase students’ creativity in a calculus subject. The main product generated is a book on the PMBST model. It consists of: a) introduction; b) theoretical rationality; c) empirical rationality; d) supporting theories; e) components of the model; f) instructions for the model; and g) bibliography. Furthermore, to assess the quality of this model, the authors used product quality assessment criteria created by Nieveen, which covers hypothetical validity and practicality. The instrument, which is developed for supporting the PMBST model development cover: an assessment sheet of the PMBST model, and a validation format of the PMBST model assessment sheet. The subjects of the PMBST model in this research are Master’s programme students in the Mathematics Department at Universitas Negeri Makassar who have finished or are learning the calculus subject. The result of this research is the PMBST model developed with high theoretical rationality as indicated by experts’ assessments.

A novel ensemble learning approach to unsupervised record linkage

Record linkage is a process of identifying records that refer to the same real-world entity. Many existing approaches to record linkage apply supervised machine learning techniques to generate a classification model that classifies a pair of records as either match or non-match. The main requirement of such an approach is a labelled training dataset. In many real-world applications no labelled dataset is available hence manual labelling is required to create a sufficiently sized training dataset for a supervised machine learning algorithm. Semi-supervised machine learning techniques, such as self-learning or active learning, which require only a small manually labelled training dataset have been applied to record linkage. These techniques reduce the requirement on the manual labelling of the training dataset. However, they have yet to achieve a level of accuracy similar to that of supervised learning techniques. In this paper we propose a new approach to unsupervised record linkage based on a combination of ensemble learning and enhanced automatic self-learning. In the proposed approach an ensemble of automatic self-learning models is generated with different similarity measure schemes. In order to further improve the automatic self-learning process we incorporate field weighting into the automatic seed selection for each of the self-learning models. We propose an unsupervised diversity measure to ensure that there is high diversity among the selected self-learning models. Finally, we propose to use the contribution ratios of self-learning models to remove those with poor accuracy from the ensemble. We have evaluated our approach on 4 publicly available datasets which are commonly used in the record linkage community. Our experimental results show that our proposed approach has advantages over the state-of-the-art semi-supervised and unsupervised record linkage techniques. In 3 out of 4 datasets it also achieves comparable results to those of the supervised approaches.