• Title/Summary/Keyword: sensor prediction

Search Result 567, Processing Time 0.026 seconds

A Study on the Remaining Useful Life Prediction Performance Variation based on Identification and Selection by using SHAP (SHAP를 활용한 중요변수 파악 및 선택에 따른 잔여유효수명 예측 성능 변동에 대한 연구)

  • Yoon, Yeon Ah;Lee, Seung Hoon;Kim, Yong Soo
    • Journal of Korean Society of Industrial and Systems Engineering
    • /
    • v.44 no.4
    • /
    • pp.1-11
    • /
    • 2021
  • Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.

The study of blood glucose level prediction using photoplethysmography and machine learning (PPG와 기계학습을 활용한 혈당수치 예측 연구)

  • Cheol-Gu, Park;Sang-Ki, Choi
    • Journal of Digital Policy
    • /
    • v.1 no.2
    • /
    • pp.61-69
    • /
    • 2022
  • The paper is a study to develop and verify a blood glucose level prediction model based on biosignals obtained from photoplethysmography (PPG) sensors, ICT technology and data. Blood glucose prediction used the MLP architecture of machine learning. The input layer of the machine learning model consists of 10 input nodes and 5 hidden layers: heart rate, heart rate variability, age, gender, VLF, LF, HF, SDNN, RMSSD, and PNN50. The results of the predictive model are MSE=0.0724, MAE=1.1022 and RMSE=1.0285, and the coefficient of determination (R2) is 0.9985. A blood glucose prediction model using bio-signal data collected from digital devices and machine learning was established and verified. If research to standardize and increase accuracy of machine learning datasets for various digital devices continues, it could be an alternative method for individual blood glucose management.

Analysis on the Advanced Model for Solar Energy Harvesting (개선된 태양 에너지 하베스팅 모델에 대한 분석)

  • Nayantai, Bulganbat;Kong, In-Yeup
    • Journal of the Institute of Convergence Signal Processing
    • /
    • v.14 no.2
    • /
    • pp.99-104
    • /
    • 2013
  • Replacement of sensor nodes for monitoring a wide range area such as mountains and forests needs a lot of time and cost. Using new and renewable energy around them can maximize the lifetime of wireless sensor networks, in which solar energy is infinite energy source that is available in 365 days. To design these sensor networks, solar energy model is essential and to estimate and analyze the overall photovoltaic energy. Using this, we can figure out important data such as the size and performance of solar panel needed. However, existing researches for solar energy harvesting consider parts of many factors to influence the quantity of solar energy gathered. In this paper, we suggest advanced solar energy harvesting model considering angular loss (solar cell panel), overheat loss (solar cell), rechargeable battery heat and cooling for each monthly properties. From our experimental results according to outdoor temperature, panel angle and the surface temperature of solar panel, we show these impact factors are correctly configured.

Prediction of Fermentation Time of Korean Style Soybean Paste by using The Portable Electronic Nose (휴대용 전자코에 의한 된장의 숙성정도 예측)

  • Noh, Bong-Soo;Yang, Young-Min;Lee, Taik-Soo;Hong, Hyung-Ki;Kwon, Chul-Han;Sung, Yung-Kwon
    • Korean Journal of Food Science and Technology
    • /
    • v.30 no.2
    • /
    • pp.356-362
    • /
    • 1998
  • The study is to predict fermentation time of Korean style soybean paste by portable electronic nose that has six metal oxide sensors. Korean style soybean paste using Aspergillus oryzae was fermented at $15^{\circ}C,\;20^{\circ}C\;and\;25^{\circ}C$. The changes of sensitivity by electronic nose, amino nitrogen and reducing sugar were observed during fermentation. Sensitivities of six metal oxide sensor were decreased with increase of fermentation time while amino nitrogen was increased. Sensor #3 and #4 showed good correlation between sensitivities of the sensors and fermentation time $(r^2=0.71{\sim}0.95)$. And the good correlation between sensitivity by electronic nose and the produced amino nitrogen was shown until soybean paste was fermented. Portable electronic nose using metal oxide sensor (#3 and #4) could predict fermentation time of Korean style soybean paste.

  • PDF

Development of Electronic Mapping System for N-fertilizer Dosage Using Real-time Soil Organic Matter Sensor (실시간 토양 유기물 센서와 DGPS를 이용한 질소 시비량 지도 작성 시스템 개발)

  • 조성인;최상현;김유용
    • Journal of Biosystems Engineering
    • /
    • v.27 no.3
    • /
    • pp.259-266
    • /
    • 2002
  • It is crucial to know spatial soil variability for precision farming. However, it is time-consuming, and difficult to measure spatial soil properties. Therefore, there are needs fur sensing technology to estimate spatial soil variability, and for electronic mapping technology to store, manipulate and process the sampled data. This research was conducted to develop a real-time soil organic matter sensor and an electronic mapping system. A soil organic matter sensor was developed with a spectrophotometer in the 900∼1,700 nm range. It was designed in a penetrator type to measure reflectance of soil at 15cm depth. The signal was calibrated with organic matter content (OMC) of the soil which was sampled in the field. The OMC was measured by the Walkeley-Black method. The soil OMCs were ranged from 0.07 to 7.96%. Statistical partial least square and principle component regression analyses were used as calibration methods. Coefficient of determination, standard error prediction and bias were 0.85 0.72 and -0.13, respectively. The electronic mapping system was consisted of the soil OMC sensor, a DGPS, a database and a makeshift vehicle. An algorithm was developed to acquire data on sampling position and its OMC and to store the data in the database. Fifty samples in fields were taken to make an N-fertilizer dosage map. Mean absolute error of these data was 0.59. The Kring method was used to interpolate data between sampling nodes. The interpolated data was used to make a soil OMC map. Also an N-fertilizer dosage map was drawn using the soil OMC map. The N-fertilizer dosage was determined by the fertilizing equation recommended by National Institute of Agricultural Science and Technology in Korea. Use of the N-fertilizer dosage map would increase precision fertilization up to 91% compared with conventional fertilization. Therefore, the developed electronic mapping system was feasible to not only precision determination of N-fertilizer dosage, but also reduction of environmental pollution.

Multi-Modal Wearable Sensor Integration for Daily Activity Pattern Analysis with Gated Multi-Modal Neural Networks (Gated Multi-Modal Neural Networks를 이용한 다중 웨어러블 센서 결합 방법 및 일상 행동 패턴 분석)

  • On, Kyoung-Woon;Kim, Eun-Sol;Zhang, Byoung-Tak
    • KIISE Transactions on Computing Practices
    • /
    • v.23 no.2
    • /
    • pp.104-109
    • /
    • 2017
  • We propose a new machine learning algorithm which analyzes daily activity patterns of users from multi-modal wearable sensor data. The proposed model learns and extracts activity patterns using input from wearable devices in real-time. Inspired by cue integration of human's property, we constructed gated multi-modal neural networks which integrate wearable sensor input data selectively by using gate modules. For the experiments, sensory data were collected by using multiple wearable devices in restaurant situations. As an experimental result, we first show that the proposed model performs well in terms of prediction accuracy. Then, the possibility to construct a knowledge schema automatically by analyzing the activation patterns in the middle layer of our proposed model is explained.

An Energy Efficient Cluster-based Scheduling Scheme for Environment Information Systems (환경정보 시스템에 적합한 클러스터 기반 에너지 효율적인 스케줄링 기법)

  • An, Sung-Hyun;Kim, Seung-Hoon
    • Journal of Korea Multimedia Society
    • /
    • v.11 no.5
    • /
    • pp.633-640
    • /
    • 2008
  • Sensor node clustering is one of the most popular research topics to reduce the energy of sensor nodes in wireless sensor networks. Previous researches, however, did not consider prediction effects of sensed environment information on TDMA scheduling of a cluster, resulting energy inefficiency. In this paper, we suggest an energy efficient cluster-based scheduling scheme that can be applied flexibly to many environment information systems. This scheme reflects the environment information obtained at the application layer to the MAC layer to set up the schedule of a cluster. The application layer information sets up the scheduling referring to the similarity of sensed data of cluster head. It determines the data transmission considering the result of similarity. We show that our scheme is more efficient than LEACH and LEACH-C in energy, which are popular clustering schemes, through simulation.

  • PDF

Color Interpolation Algorithm for Pixel Resolution Modus of Image Sensor (영상센서의 출력 해상도 모드를 고려한 색상 보간 알고리즘)

  • Kim, Bu-Gong;Kim, Moon-Cheol
    • Journal of the Institute of Electronics and Information Engineers
    • /
    • v.51 no.9
    • /
    • pp.129-138
    • /
    • 2014
  • Various interpolations for digital imaging devices with a single image sensor have proposed. However, conventional methods did not consider the resolution modus of image sensor using periodic sampling. Therefore, the resulting images have problems such as quality degradation and color artifacts(color moire, zipper). In this paper, we propose a color interpolation algorithm for pixel resolution modus of image sensor. The proposed algorithm consisted of an initial step to compensate edge prediction effectively and refinement step using minimum directions for pixel resolution modus. To analyze a result of the proposed algorithm with conventional methods, we evaluated subjectively using images quality comparison and objectively using PSNR(Peak Signal to Noise Ratio). Experimental results showed that the proposed algorithm was more successful in eliminating the color artifacts than conventional methods judged by both objective and subjective criteria.

Preliminary Study for Non-destructive Measurement of Stress Tensor on H-beam in Tunnel Support System using a Magnetic Anisotropy Sensor (자기 이방성 응력측정법을 활용한 터널 지보 구조물의 비파괴계측에 관한 기초적 연구)

  • Lee, Sang-Won;Akutagawa, Shinichi;Kim, Young-Su;Jin, Guang-Ri;Jeng, Ii-Han
    • Proceedings of the Korean Geotechical Society Conference
    • /
    • 2008.03a
    • /
    • pp.766-777
    • /
    • 2008
  • Currently in increasing number of urban tunnels with small overburden are excavated according to the principle of the New Austrian Tunneling Method (NATM). Successful design, construction and maintenance of NATM tunnel demands prediction, control and monitoring of ground displacement and support stress high accuracy. A magnetic anisotropy sensor is used for nondestructive measurement of stress on surfaces of a ferromagnetic material, such as steel. The sensor is built on the principle of the magneto-strictive effect in which changes in magnetic permeability due to deformation of a ferromagnetic material is measured in a nondestructive manner, which then can be translated into the absolute values of stresses existing on the surface of the material. This technique was applied to measure stresses of H-beams, used as tunnel support structures, to confirm expected measurement accuracy with reading error of about 10 to 20 MPa, which was confirmed by monitoring strains released during cutting tests The results show that this method could be one of the promising technologies for non-destructive stress measurement for safe construction and maintenance of underground rock structures encountered in civil and mining engineering.

  • PDF

Non-destructive Measurement of H-beam in Support System using a Magnetic Anisotropy Sensor (자기이방성 응력측정법을 이용한 강아치 지보구조물의 비파괴 계측)

  • Yoo, Ji-Hyeung;Moon, Hong-Deuk;Lee, Jae-Ho;Kim, Dae-Sung;Kim, Hyuk
    • Proceedings of the Korean Geotechical Society Conference
    • /
    • 2010.03a
    • /
    • pp.1392-1397
    • /
    • 2010
  • Currently in increasing number of urban tunnels with small overburden are excavated according to the principle of the New Austrian Tunneling Method(NATM). Successful design, construction and maintenance of NATM tunnel demands prediction, control and monitoring of ground displacement and support stress high accuracy. A magnetic anisotropy sensor is used for non-destructive measurement of stress on surfaces of a ferromagnetic material, such as steel. The sensor is built on the principle of the magneto-strictive effect in which changes in magnetic permeability due to deformation of a ferromagnetic material is measured in a non-destructive manner, which then can be translated into the absolute values of stresses existing on the surface of the material. This technique was applied to measure stresses of H-beams, used as tunnel support structures, to confirm expected measurement accuracy with reading error of about 10 to 20MPa, which was confirmed by monitoring strains released during cutting tests The results show that this method could be one of the promising technologies for non-destructive stress measurement for safe construction and maintenance of underground rock structures encountered in civil and mining engineering.

  • PDF