KMITL

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

Browse

Search Results

Now showing 1 - 9 of 9

Some of the metrics are blocked by your
consent settings
Two-Tier AI Surveillance: Enhancing Weapon Detection Through Edge and Cloud Collaboration
(2026-01-01)
Sun, Sanchai
;
Songsuwankit, Kanoknuch
;
Thamrongrongveerachart, Chawapon
;
Yanyong, Sarucha
;
Sri-on, Jiramate
This research presents a novel two-tier AI-based surveillance system designed for real-time weapon detection to enhance security and prevent potential robberies. The system leverages the computational capabilities of both edge and cloud resources, integrating a YOLOv5s model on a Jetson Nano for initial detection and a YOLOv8l model on a server equipped with an Nvidia A100 for refined analysis. The initial detection performed by the Jetson Nano rapidly identifies potential threats and forwards compressed images to the server, optimizing bandwidth usage and transmission speed. Upon receipt, the server applies image enhancement techniques to restore and upscale the images from 640 pixels back to 1280 pixels before further verification. Utilizing Python Django, the server processes the enhanced images with a more sophisticated model to ensure high accuracy in detection. The integration of edge computing optimizes the system’s performance by enabling real-time processing and reducing latency. This hybrid approach enhances the efficiency and scalability of the surveillance system, ensuring robust and timely detection. Upon confirming the presence of a weapon, the system sends immediate alerts via LINE Notify to both users and local law enforcement, thereby enabling prompt responses to potential security threats. Additionally, the Jetson Nano hosts a Flask application, allowing users to download previously recorded videos for further review and evidence collection. The proposed solution combines edge computing, cloud processing, and image optimization techniques to provide an efficient, scalable, and high-performance solution for real-time weapon detection which enhancing accuracy and reliability in real-world surveillance.
Some of the metrics are blocked by your
consent settings
Low-Cost System for Investigating a Small Motor Fault Classification Based on Current Signal
(2025-01-01)
Taweewat, Pat
;
Suwan-Ngam, Warachart
;
Songsuwankit, Kanoknuch
;
Konghuayrob, Poom
This research presents low-cost system for motor fault classification. This system uses a microcontroller with built-in ADC and communication capability. The two purposes of this article are to investigate the quality of the system for data acquisition and capability of the system for detecting early motor faults both by microcontroller on the system and personal computer. The embedded software on the system is designed to record current signals from a current sensor, compute FFT-based features and classify the fault based on tinyML method. Data communication between the system and the personal computer can be done by both serial port and TCP socket over Wi-Fi. The performances of the system and the personal computer are compared by the experiment as well as the quality of data recorded from the built-in ADC and a digital oscilloscope. The broken rotor bar and bearing fault in a 2.2kW induction motor are investigated. The classifier used in the experiment is a small feed forward neural network which can be implemented on both the proposed low-cost system and the personal computer. Although the recorded electrical current data by built-in ADC is contaminated with noise, the fault classification on the personal computer yield accuracy up to 90%.
Some of the metrics are blocked by your
consent settings
Comparison of Reduced-Length FFT-Based Feature for Induction Motor Fault Classification
(2025-01-01)
Taweewat, Pat
;
Suwan-Ngam, Warachart
;
Songsuwankit, Kanoknuch
;
Konghuayrob, Poom
This research presents a comparison of FFT-based features which can be used for classifying induction motor faults via neural network. In this paper, the misalignment and rotor bar damage faults are investigated by using stator current as input data only. As the length of the full FFT can include both informative data corresponding to the faults and uninformative data such as noise from environment or electrical supply, only relevant magnitude from FFT bins should be selected and used instead. This paper proposed to use threshold level determined from the magnitude of FFT bins in dataset as a criterion for the selection. From experimental results, an input feature vector created by proposed method can create short input feature vector length to be used by neural network efficiently. The trained neural network performs classification task at 99.98% in accuracy. Comparing to using dimension reduction by PCA, thresholding method needs basic computation, and yields result close to PCA method.
Some of the metrics are blocked by your
consent settings
Linear Variable Differential Transformer Signal Conditioning Circuit Based on Phase-Locked Loop
(2024-01-01)
Songsuwankit, Kanoknuch
;
Petchmaneelumka, Wandee
;
Riewruja, Vanchai
;
Rerkratn, Apinai
The purpose of this paper is to propose a novel technique for extracting the position signal from an inductive displacement transducer named a linear variable differential transformer (LVDT). In general, the movement of the LVDT core causes its primary inductance change in linear form. The primary winding of the LVDT is used as a time-dependent element for the triangular and square wave generator, which can be called self-oscillation, to generate frequency. The advantage of the proposed technique is that it can measure the displacement using the LVDT without an external oscillator. The change in primary inductance causes the frequency deviation generated by the oscillator. The deviated frequency is captured and converted into a voltage signal using the principle of the phase-locked loop. All the components used in this study are commercially available. The merits of this proposed technique are simple configuration, small size, and low cost. Moreover, the operating range of the LVDT can be extended without the limitation of the nonlinear transfer characteristic. The performance of the proposed technique is discussed in detail and confirmed by experimental implementation. Experimental results show that the maximum error from the proposed technique is about 0.42% and the operating range of the LVDT can be extended to more than 200%. It can be seen that the proposed technique is suitable for embedded measurement in small or micro robots.
Some of the metrics are blocked by your
consent settings
A Small Deep Learning Model for Fault Detection of a Broken Rotor Bar of an Induction Motor
(2024-01-01)
Taweewat, Pat
;
Suwan-Ngam, Warachart
;
Songsuwankit, Kanoknuch
;
Konghuayrob, Poom
In this paper, we present an investigation of a small deep learning model applied to the detection of a broken rotor bar of an induction motor. The motor current spectrum analysis is the base method for fault detection. This proposed method focuses on the analysis of the modification of the input vector and model configuration. This method was implemented and it showed that the feature length and size of the model are reduced compared with the existing method. The experimental results showed that only feature extraction using the spectral-based method and limit range of its coefficient are adequate to provide accuracy of small deep learning comparable to that of the parallel-layer deep learning model. Likewise, at the same accuracy level, based on the deep learning model, a shorter sampling duration than that required by the reference model is needed.
Some of the metrics are blocked by your
consent settings
Extension of Linear Operating Range for Linear Variable Differential Transformer Using Its Inverse Transfer Characteristic
(2023-01-01)
Petchmaneelumka, Wandee
;
Songsuwankit, Kanoknuch
;
Rerkratn, Apinai
;
Gullayanon, Rutchanee
;
Riewruja, Vanchai
An analog circuit technique to realize an inverse transfer characteristic of a linear variable differential transformer (LVDT) is presented in this paper. Practically, the structure of the LVDT causes a narrow linear operating range compared with its full stroke range. However, a large linear operating range requires a huge structure for the LVDT, making it unsuitable for a small or compact measurement system. The proposed technique can be used in a commercial LVDT to extend the linear operating range to its full stroke range. The technique utilizes an inherent behavior of an operational transconductance amplifier (OTA) to emulate the LVDT transfer characteristic. The LVDT transfer characteristic generated by the OTA is used as a feedback path of the inverting amplifier formed by an operational amplifier (opamp) to realize the inverse transfer characteristic. The residual error due to the OTA behavior is very small and can be neglected without adversely affecting the performance of the proposed technique. All devices used in the proposed scheme are commercially available. The attractive features of the proposed technique are its simple configuration, small size, low cost, and high accuracy. The performance of the proposed technique is discussed in detail and confirmed by its experimental implementation. Measurement results demonstrate that the linear operating range of the commercial LVDT used in this study can be extended by a factor of more than 2.4, and a fullscale percentage error of about 0.068% was obtained.
Some of the metrics are blocked by your
consent settings
A temperature-compensation technique for improving resolver accuracy
(2021-09-01)
Petchmaneelumka, Wandee
;
Riewruja, Vanchai
;
Songsuwankit, Kanoknuch
;
Rerkratn, Apinai
Variation in the ambient temperature deteriorates the accuracy of a resolver. In this paper, a temperature-compensation technique is introduced to improve resolver accuracy. The ambient temperature causes deviations in the resolver signal; therefore, the disturbed signal is investigated through the change in current in the primary winding of the resolver. For the proposed technique, the primary winding of the resolver is driven by a class-AB output stage of an operational amplifier (opamp), where the primary winding current forms part of the supply current of the opamp. The opamp supply-current sensing technique is used to extract the primary winding current. The error of the resolver signal due to temperature variations is directly evaluated from the supply current of the opamp. Therefore, the proposed technique does not require a temperature-sensitive device. Using the proposed technique, the error of the resolver signal when the ambient temperature increases to 70 °C can be minimized from 1.463% without temperature compensation to 0.017% with temperature compensation. The performance of the proposed technique is discussed in detail and is confirmed by experimental implementation using commercial devices. The results show that the proposed circuit can compensate for wide variations in ambient temperature.
Some of the metrics are blocked by your
consent settings
Temperature Compensation for Transformer-type Transducer
(2021-01-01)
Songsuwankit, Kanoknuch
;
Riewruja, Vanchai
;
Watanachaturaporn, Pakorn
;
Rerkratn, Apinai
;
Petchmaneelumka, Wandee
A novel technique to compensate the temperature effect of a transformer-type transducer is proposed in this paper. The effect of the ambient temperature on the transformer-type transducer is investigated from a primary-winding current. The advantage of the proposed technique is that the temperature effect is compensated without requiring a temperature sensor, making it suitable for applications in robotic and automation systems operated in harsh environments. The primary-winding current of the transducer is generated using a second-generation current conveyor (CCII). The excitation signal of the transformer-type transducer is driven by the CCII and the current flowing through the primary winding is transferred to an output signal of the CCII. The deviation of the primary-winding current due to the temperature effect is evaluated from the output signal of the CCII. The temperature effect on the transducer is manipulated by a closed-loop principle using a subtract-and-sum action instead of a traditional proportional-plusintegral action to eliminate the deviation of the primary-winding current. Therefore, the temperature effect on the transducer is compensated. A linear variable differential transformer (LVDT) is used to demonstrate the proposed technique, whose performance is discussed in detail and confirmed experimentally. All devices used in this experiment are commercially available. Experimental results show that the measured error of the output signal from the LVDT at 70 C can be reduced from 6.2% without temperature compensation to 0.06% by using the proposed technique, which has the advantages of a low cost, simple configuration, and high performance.
Some of the metrics are blocked by your
consent settings
Linear-range extension for linear variable differential transformer using binomial series
(2020-01-01)
Petchmaneelumka, Wandee
;
Songsuwankit, Kanoknuch
;
Tongcharoen, Jakkapun
;
Riewruja, Vanchai
The linear-range extension technique for a linear variable differential transformer (LVDT) is described in this paper. Generally, the LVDT has a narrow linear operating range caused by its nonlinear transfer characteristic. To extend the linear operating range, the nonlinear behavior of the LVDT must be adjusted. In this paper, the circuit building block providing the LVDT inverse transfer characteristic using binomial series approximation is proposed for linearizing the nonlinear behavior of the LVDT. The third-order inverse transfer characteristic of the LVDT is synthesized from analog multipliers and a difference amplifier comprising an operational amplifier (opamp). All active devices used in this study are commercially available. Therefore, the attraction of the proposed technique is in the simple configuration and low cost, making it suitable for an embedded measurement system. The performance of the proposed technique is discussed in detail. Simulation and experimental results confirming the performance are also included. As a result, the linear range of the commercial LVDT used in this study can be extended more than 500%. The full scale error of the measured value is about 0.23% over the entire operating range.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results