KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/kmitl/1

Browse

Settings

Access type: UnknownAuthor: search.filters.author.Khunkitti, Pirat

Search Results

Now showing 1 - 3 of 3
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Label-Consistent Input Structuring Based on Second-Stage Sliding Window for PMSM Fault Diagnosis
    (2026-01-01)
    Nguyen, Thanh Son
    ;
    Khunkitti, Pirat
    ;
    Siritaratiwat, Apirat
    ;
    Seangwong, Pattasad
    Accurate fault diagnosis of electrical machines is essential for operational reliability and safety. Handcrafted features remain attractive in practice because of their interpretability and computational efficiency. However, this traditional approach relies on individual feature vectors and is limited in preserving inherent temporal dynamics. To address this issue, a label-consistent input structuring based on the second-stage sliding window (SSSW) method is proposed. This approach retains the benefits of handcrafted features and arranges feature vectors into temporally coherent sequences while ensuring label consistency. A hyperparameter optimization scheme is incorporated with a long short-term memory classifier to reduce manual tuning. Performance evaluations demonstrate the robustness of the proposed SSSW method across diverse operating conditions and input signal configurations. These include single-phase currents, multiphase currents, vibration, and fused current-vibration signals. Notably, the proposed approach achieves classification accuracy of up to 100% and demonstrates stable learning behavior. Experimental verification on a laboratory-scale permanent magnet synchronous motor testbed further validates the proposed method under realistic measurement noise and interference conditions. Finally, combining handcrafted features with the proposed SSSW input structuring method provides a practical, scalable solution for reliable electrical machine fault diagnosis.
      2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Improving Belief Propagation Decoding of Polar Codes using Bi-Directional Bit-Flipping Strategy
    (2022-07-01)
    Mitsungnern, Nuttawat
    ;
    Imtawil, Virasit
    ;
    Meesomboon, Anupap
    ;
    Khunkitti, Pirat
    ;
    Phakphisu, Watid
    It is difficult to achieve high throughput and low latency from the cyclic redundancy check (CRC-) aided successive cancellation list (CA-SCL) polar decoder employed in 5G new radio systems. To tackle this problem, the hardware-efficient belief propagation (BP) decoder has been extensively studied. Unfortunately, the performance of the BP decoding for polar codes is quite poor. To improve the performance of this decoder, a novel bi-directional bit-flipping (BF) strategy is proposed. The rule of generating the flip set based on the post log-likelihood ratio (LLR) is combined with the possibility of the two-way flipping. With the help of this idea, the decoding performance of BP decoder is identical to that of the CA-SCL decoder. Moreover, with the same complexity, our proposed BF strategy outperforms both the best known one-directional BF strategy and the previously found bi-directional BF strategy at BLER of 10<sup>-4</sup>.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Simulation of magnetic footprints for heat assisted magnetic recording
    (2017-05-01)
    Pituso, Kotchakorn
    ;
    Khunkitti, Pirat
    ;
    Kruesubthaworn, Anan
    ;
    Chooruang, Komkrit
    ;
    Tongsomporn, Damrongsak
    The heat assisted magnetic recording (HAMR) technology has been the promising candidate to overcome the thermal stability limitation at higher capacities of the hard disk drive. In this work, the characteristics of magnetic footprint of the medium written by HAMR were investigated through the micromagnetic simulations. The Voronoi granular media was firstly modeled, then the magnetic footprint technique was performed to observe the media behaviors at various linear densities. The results indicated that the pattern of magnetic footprint can be perverted at higher densities, which essentially causes a reduction of readback signal power. Also, the dependence of media grain size on the signal power shows that the larger grain size media can provide higher signal power at low linear density, while the smaller grain size media gives higher signal at high density. Thus, the magnetic footprint regarding the HAMR technology needs to be optimized to achieve the efficient recording system.
      1