Application of online SVR on the dynamic liquid level soft sensing

Bai, Shan; Jiang, Zijian; Wang, Tong; Haozhe, Lai

doi:10.1109/ccdc.2013.6561460

Cited by 5 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The learning and training error rate were used for testing in the limited data error. ANN and SVR are the methods that could be run data to find the best model with their data's characteristic [5]. The model will represent the [6].…”

Section: Introductionmentioning

confidence: 99%

Initial Optimal Parameters of Artificial Neural Network and Support Vector Regression

Fradinata¹,

Suthummanon²,

Suntiamorntut³

2018

IJECE

View full text Add to dashboard Cite

This paper presents architecture of backpropagation Artificial Neural Network (ANN) and Support Vector Regression (SVR) models in supervised learning process for cement demand dataset. This study aims to identify the effectiveness of each parameter of mean square error (MSE) indicators for time series dataset. The study varies different random sample in each demand parameter in the network of ANN and support vector function as well. The variations of percent datasets from activation function, learning rate of sigmoid and purelin, hidden layer, neurons, and training function should be applied for ANN. Furthermore, SVR is varied in kernel function, lost function and insensitivity to obtain the best result from its simulation. The best results of this study for ANN activation function is Sigmoid. The amount of data input is 100% or 96 of data, 150 learning rates, one hidden layer, trinlm training function, 15 neurons and 3 total layers. The best results for SVR are six variables that run in optimal condition, kernel function is linear, loss function is ౬-insensitive, and insensitivity was 1. The better results for both methods are six variables. The contribution of this study is to obtain the optimal parameters for specific variables of ANN and SVR.

show abstract

Section: Introductionmentioning

confidence: 99%

Initial Optimal Parameters of Artificial Neural Network and Support Vector Regression

Fradinata¹,

Suthummanon²,

Suntiamorntut³

2018

IJECE

View full text Add to dashboard Cite

show abstract

“…For example, the online support vector regression algorithm is presented to be a solution to measure the dynamic liquid level. This method not only has good performance, but also can update the forecast model by online [24]. The optical method, through the analytical expressions that include the relationship between optical intensity and liquid surface sloshing wave and the expressions of the wavelength as well as amplitude, can obtain the optical patterns corresponding to static and sloshing liquid surface.…”

mentioning

confidence: 99%

An Accurate Detection for Dynamic Liquid Level Based on MIMO Ultrasonic Transducer Array

Cai

Shen

et al. 2015

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

In many circumstances, the conventional ultrasonic liquid-level detection presents the unreliable estimations due to the dynamically changed liquid level. In addition, there are circumstances where the level change involves not only the fluctuation but also the rise or fall of liquid level. To improve the measuring accuracy of liquid level using the ultrasonic method in dynamically changed level case, an attractive ultrasonic method, named the liquid-level detection based on the multiple-input multiple-output ultrasonic transducer array, is proposed in this paper. This method is different from the early ultrasonic liquidlevel detections, including those that utilize transducer array. Based on the virtual element technology, the method employs the multitransducer array to achieve the reduction of system complexity and cost, and then applies the synthetic aperture technology to realize rapid samples of liquid level. Besides, one adaptive searching scheme of focused position in beamforming of synthetic aperture in each scanning direction is optimized to get the high precise samples of liquid level. The proposed method is verified by simulation and a real system, and compared with the conventional single-channel approach. We still proposed a simulation method of ultrasonic echo signal from liquid level based on the boundary-layer theory and the ultrasonic scattering theory. The simulated and actual measurement results demonstrate that the proposed method is obviously superior to the conventional approach. Meanwhile, the factors influencing on the proposed method are investigated by simulation also. The investigations reveal that the focus position in beamforming, the SNR of echo signal, and the wave size of liquid level have impact on the suggested method. However, it is indicated that the good performance of the proposed method remains, provided that the focus in beamforming is properly set in every scanning direction and the noise of echo signal is effectively controlled.Index Terms-Acoustics, arrays, level measurement, measurement errors, multiple-input multiple-output (MIMO) systems, transducers.

show abstract

Efficient Regional Traffic Signal Control Scheme Based on a SVR Traffic State Forecasting Model

You

Chen

2015

Internet of Vehicles - Safe and Intelligent Mobility

View full text Add to dashboard Cite

Application of online SVR on the dynamic liquid level soft sensing

Cited by 5 publications

References 3 publications

Initial Optimal Parameters of Artificial Neural Network and Support Vector Regression

Initial Optimal Parameters of Artificial Neural Network and Support Vector Regression

An Accurate Detection for Dynamic Liquid Level Based on MIMO Ultrasonic Transducer Array

Efficient Regional Traffic Signal Control Scheme Based on a SVR Traffic State Forecasting Model

Contact Info

Product

Resources

About