KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 10

Some of the metrics are blocked by your
consent settings
Automated Assembly Machine Stopping System Using Machine Learning for Electrical Failure Prediction in Hard Disk Drive Manufacturing
(2025-01-01)
Panjasamanwong, Tanakrit
;
Thanasopon, Bundit
The assembly of diode lasers and sliders is a critical process in hard disk drive (HDD) manufacturing, directly affecting product electrical performance. Delays in detecting problematic assembly machines increase scrap rates, reduce yield, and raise production costs. This research presents the development of supervised machine learning models to predict electrical testing results using key parameters from assembly machines, for example, diode laser alignment position after assembly and laser intensity measurement during assembly. Dataset was preprocessed through outlier removal and feature selection using correlation analysis and permutation importance before applying SMOTE technique to balance the minority class in the training set, followed by normalization before modeling and evaluating performance. Multiple models were developed and evaluated to select the best performing model. Evaluation results showed that the AdaBoost model combined with Random Forest achieved the highest performance, with an AUC of 92.36 %. The selected model was integrated into monitoring, alerting and machine auto stopping system when the predicted failure rate exceeded a predefined threshold. The implemented system reduces the detection time for problematic assembly machines, minimizes defective parts, and improves overall manufacturing efficiency.
Some of the metrics are blocked by your
consent settings
Magnetic Recording Reliability Evaluation
(2022-01-01)
Phonyiam, Ekkaphon
;
Netisopakul, Ponrudee
Magnetic recording heads in Hard Disk Drive (HDD) integrates of two main components included of writer and reader devices for injecting and reading the pattern on recording media. Both features were fabricated in a single piece and become a multi-layer structure. Reliability test for magnetic recording head has become a challenge that could introduce complex problems when failure occurred. This paper aims to conceptualize testing workflow and measurement for magnetic recording head reliability evaluation. The testing process lets head performing cycles of writing and reading operations in designated ambient until it meets stopping criteria. And the reliability analysis procedure focuses on lifetime analysis with Weibull distribution on different failure modes and characters extracting from time-series bit-error-rate profile which was obtained from reliability test.
Some of the metrics are blocked by your
consent settings
Non-linear fuzzy-PID controller design for track following of hard disk drive servo system
(2017-12-13)
Boonpranchoo, V.
;
Numsomran, A.
;
Chaiyasith, P.
;
Chaoraingern, J.
;
Tipsuwanporn, V.
This paper presents a non-linear control design scheme for track following control of hard disk drive using Fuzzy-PID controller. The proposed scheme, using a parallel structure combined of non-linear Fuzzy PI controller and Fuzzy PD controller, controls the response of a track following control of hard disk drive system to decrease total of non-repeatable position error (NRPE) and position error signal (PES). A performance verification of the proposed controls was also conducted using HDD benchmark simulation system. Results showed that the non-linear Fuzzy PID controller satisfies an acceptable performance in hard disk drive head track following control and effectively robust to disturbance and process parameter perturbations.
Some of the metrics are blocked by your
consent settings
Head Gimbal Assembly circuit with vision technique and Fuzzy C-Means Clustering
(2015-12-23)
Praserttaweelap, Rawinun
;
Kiatwanidvilai, Somyot
In Hard Disk Drive (HDD) Industry, the automation system is the one key for manufacturing process. Head Gimbal Assembly (HGA) is a part of HDD which have the reader and writer circuit. The HGA circuit is very important for read/write process. This research proposes the new vision technique and clustering by Fuzzy C-Means algorithm for HGA circuit inspection in 3 groups. HGA circuits in 3 groups are good, bridging, and missing group. The bridging and missing groups are the defect group. Blob analysis is the one of vision technique that it can measure the properties of image. The measurement properties from blob analysis are used in clustering technique. Fuzzy C-Means Clustering is the clustering technique which is grouped the measurement data into the cluster group based on the natural grouping of data. From the experiment results of this research, the clustering performance from Fuzzy C-Means Clustering is 99.11% accuracy based on the measurement properties in blob analysis with 225 samples.
Some of the metrics are blocked by your
consent settings
Vision inspection with k-means clustering on head gimbal assembly defect
(2015-01-01)
Praserttaweelap, Rawinun
;
Kiatwanidvilai, Somyot
Head Gimbal Assembly (HGA) is an important feature of read and write process in a Hard Disk Drive (HDD). Currently, HGA circuit inspections are done using human operators under microscope; the vision processing for inspection in automated systems is required. This research work proposes an algorithm for detection the HGA circuit defect by using the blob detection, and then analysis the properties of blob tool. By the measurement properties of the blob tool, the K-Means Clustering can specify the data in each group in 95.45% accuracy with 110 samples.
Some of the metrics are blocked by your
consent settings
A model-reference sliding mode for dual-stage actuator servo control in HDD
(2014-01-01)
Sonkham, Siwaphon
;
Pinsopon, Unnat
;
Chatlatanagulchai, Withit
This paper presents a method of sliding mode control (SMC) designing and developing for the servo system in a dual-stage actuator (DSA) hard disk drive. Mathematical model of hard disk drive actuators is obtained, extracted from measuring frequency response of the voice-coil motor (VCM) and micro-actuator separately. The Matlab software tools are used for mathematical model estimation and also for controller design and simulation. A model-reference approach for tracking requirement is selected as a proposed technique. The simulation results show that performance of the model-reference SMC controller in DSA servo control can be satisfied in the tracking error, as well as keeping the positioning of the head within the boundary of +/-5% of track width under the presence of disturbance. The overall results of model-reference SMC design in DSA are met per requirement specifications and significant reduction in %off track is found when compared to the singlestate actuator (SSA). © 2014 IEEE.
Some of the metrics are blocked by your
consent settings
The design of load/unload algorithm for 2.5'' hard disk drive in SSTW process
(2012-12-01)
Nakharin, Watsamon
;
Boonpranchoo, Vichitpon
;
Tipsuwanporn, Vittaya
;
Numsomran, Arjin
To prevent striking of head to disc in servo track writing process, head velocity is controlled by ramp loading procedure for moving read/write head from ramp to disc surface properly. The author presents the approach to design load/unload algorithm for 2.5'' hard disk drive in SSTW process. The 2 Degree of Freedom PID Control Structure has been applied instead of conventional PID for VCM velocity control. Furthermore; we use Characteristic Ratio Assignment Technique for adjusting PID parameters in order to obtain less overshoot in output response. The experimental results tested with self-servo track writer PH4LC unit have been shown that 2DOF PID algorithm can achieve the performance specification requirement and reducing HDI problem in load/unload control of 2.5'' hard disk drive, efficiently. © 2012 IEEE.
Some of the metrics are blocked by your
consent settings
The design of PID controller for track following control of hard disk drive using Coefficient Diagram Method
(2011-12-01)
Chaoraingern, Jutarut
;
Vaidee, Worapon
;
Trisuwannawat, Thanit
;
Numsomran, Arjin
;
Tipsuwanporn, Vittaya
This paper presents the design of PID controller for track following control of hard disk drive using Coefficient Diagram Method (CDM). For this reason, Track following control of hard disk drive process has many disturbances from internal and external sources, so determining satisfactory controller parameters to stabilize control system and satisfying response specification is difficult. The proposed PID controller design using Coefficient Diagram Method (CDM) based on proper choosing the coefficients of the characteristic transfer function of the closed loop control system in accordance with stability index and equivalent time constant is applying an uncomplicated and convenience technique to controller design of Track following control in hard disk drive process. Matlab simulation results shown that CDM design is fairly stable and robust whilst giving the convenience in controller's parameters adjustment. © 2011 ICROS.
Some of the metrics are blocked by your
consent settings
Customer failure modes prediction for hard disk drive using neural networks rank-level fusion
(2011-08-12)
Tepin, Waraporn
;
Kidjaidure, Yuttana
The Prediction of Customer Failure Modes in Hard Disk Drive (HDD) is proposed using Neural Networks Rank-Level Fusion applied on key parameters measured in the manufacturing process of a HDD. In our methods, Neural Networks, Discriminant Analysis, Bayesian Networks, Support Vector Machines are applied to classified data which was obtained from Principal Component Analysis. The output of the classifiers is further aggregated using Neural Networks Rank-Level Fusion to form the final prediction model. The resultant of the model is a highly accurate prediction superior to Borda Count, Logistic Regression Fusion Methods and beyond current known reliability predictors of HDD failures. © 2011 IEEE.
Some of the metrics are blocked by your
consent settings
The design of PID controller for track following control of hard disk drive using coefficient diagram method
(2011-01-01)
Chaoraingern, Jutarut
;
Vaidee, Worapon
;
Trisuwannawat, Thanit
;
Tipsuwanporn, Vittaya
;
Numsomran, Arjin
This paper presents the design of PID controller for track following control of hard disk drive using Coefficient Diagram Method (CDM). For this reason, Track following control of hard disk drive process has many disturbances from internal and external sources, so determining satisfactory controller parameters to stabilize control system and satisfy response specification is difficult. The proposed PID controller design using Coefficient Diagram Method (CDM) based on proper choosing the coefficients of the characteristic transfer function of the closed loop control system in accordance with stability index and equivalent time constant is an uncomplicated and convenience technique for applying to Track following control in hard disk drive process. Matlab simulation results shown that PID controller using CDM technique is fairly stable and robust whilst give the convenience in controller's parameters adjustment. © 2011 SICE.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results