Browsing by Author "Shu, Lei"

Now showing 1 - 3 of 3

A MapReduce-Based Ensemble Learning Method with Multiple Classifier Types and Diversity for Condition-Based Maintenance with Concept Drifts
(2017-11-01)
Lin, Chun-Cheng
;
Shu, Lei
;
Deng, Der-Jiunn
;
Yeh, Tzu-Lei
;
Chen, Yu-Hsiang
Condition-based maintenance in Industry 4.0 collects a huge amount of production datastreams continuously from the Internet of Things attached to machines to forecast the time when to maintain machines or replace components. However, as conditions of machines change dynamically with time owing to machine aging, malfunction, or replacement, the concept of capturing the forecasting pattern from the datastream could drift unpredictably, so it is hard to find a robust forecasting method with high precision. Therefore, this work proposes an ensemble learning method with multiple classifier types and diversity for condition-based maintenance in manufacturing industries, to address the bias problem when using only one base classifier type. Aside from manipulating data diversity, this method includes multiple classifier types, dynamic weight adjusting, and data-based adaption to concept drifts for offline learning models, to promote precision of the forecasting model and precisely detect and adapt to concept drifts. With these features, the proposed method requires powerful computing resources to efficiently respond to practical condition-based maintenance applications. Therefore, the implementation of this method based on the MapReduce framework is proposed to increase computational efficiency. Simulation results show that this method can detect and adapt to all concept drifts with a high precision rate.
On Lifetime Enhancement of Dynamic Wireless Sensor Networks with Energy-Harvesting Sensors
(2016)
Lin, Chun-Cheng
;
Deng, Der-Jiunn
;
Shu, Lei
;
Wang, Kun
;
Wang, Shang-Bin
This work investigates the problem of extending the lifetime of dynamic WSNs with energy-harvesting (EH) sensors to enhancing the total WSN lifetime. This problem is modeled as finding the maximal number of covers each of which can cover all targets to be monitored. Since the concerned problem is also NP-complete, this work proposes a novel harmony search algorithm. By simulation, the proposed algorithm is shown to be promising.
Router Node Placement With Service Priority in Wireless Mesh Networks Using Simulated Annealing With Momentum Terms
(2016-12)
Lin, Chun-Cheng
;
Shu, Lei
;
Deng, Der-Jiunn
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
In wireless mesh networks (WMNs), mesh clients communicate with each other via the gateway and bridging functions of mesh routers. The performance of a WMN is generally affected by its network connectivity and client coverage, both of which are determined by its router node placement (RNP) in the deployment area. For simplicity, previous works considered only the RNP where each mesh client is served as an equal. In practice, however, mesh clients should be served with different priorities owing to factors such as their importance and their different payments for the service access. To fulfil this requirement, by assuming that each mesh client is also associated with a service priority, this paper investigates an RNP problem with a service priority constraint in which the mesh clients with service priorities higher than a threshold must be served. Given that this problem inherited from the complexity of the original RNP problem is computationally intractable in general, this paper also develops a novel simulated annealing (SA) approach that takes into account momentum terms to improve the efficiency and accuracy of annealing schedules and prevent fluctuations in values of the acceptance probability function. Additionally, the time complexity of the proposed SA algorithm is analyzed. Furthermore, evaluation of different-size instances under various parameters and annealing schedules demonstrates the superiority of the proposed approach.