Estimation of the Hydrophobicity of a Composite Insulator Based on an Improved Probabilistic Neural Network
Abstract
:1. Introduction
2. Principle of Probabilistic Neural Networks
2.1. Principle of the Probability Density Estimation
2.2. Structure of Probabilistic Neural Network
3. Hydrophobicity Image Processing
3.1. Image Pretreatment
3.2. Image Segmentation
3.3. Feature Extraction
3.4. Hydrophobicity Classification
4. Results & Discussion
5. Conclusions
- The combination of the 2-level wavelet denoising and the 3-scale Retinex enhancements not only filters out the noise, but it also enhances the local contrast. The image quality has been greatly improved in the information entropy, the contrast, and the signal-to-noise ratio. The use of an asymptotic semi-soft threshold function preserves the edge detail information and avoids a constant difference in the threshold; the two-dimensional adaptive Otsu’s method is demonstrated to be effective in segmentation for addressing the small difference between the water droplet edge and the composite insulator surface.
- The shape factor, the area ratio of the largest water droplet, and the coverage rate of the water droplet are selected as the feature parameters, and they are input into the improved PNN. The effect of the expansion speed on the learning network is compared and analyzed. The experimental results show that the proposed method is suitable for the superposition of feature parameters, and has a high recognition accuracy of up to 94.8% for a diversity of images. It is superior to the Improved Shape Factor Method, the Multifractal Method, and the RBF Neural Network.
- The proposed method can be applied to the area of image-based pattern recognition. It can be used for remote sensing image analysis, robot vision, image-based biometric verification, and so on.
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
Appendix A
References
- Chen, C.; Jia, Z.D.; Ye, W.A.; Guan, Z.C.; Li, Y.Z. Condition assessment strategies of composite insulator based on statistic methods. IEEE J. Mag. 2016, 23, 3231–3241. [Google Scholar] [CrossRef]
- Li, J.; Wei, Y.; Huang, Z.Y.; Wang, F.P.; Yan, X.Z.; Wu, Z.L. Electrohydrodynamic behavior of water droplets on a horizontal super hydrophobic surface and its self-cleaning application. Appl. Surf. Sci. 2017, 403, 133–140. [Google Scholar] [CrossRef]
- Hussain, M.M.; Farokhi, S.; McMeekin, S.G.; Farzaneh, M. Risk assessment of failure of outdoor high voltage polluted insulators under combined stresses near shoreline. Energies 2017, 10, 1661. [Google Scholar] [CrossRef]
- Mavrikakis, N.C.; Mikropoulos, P.N. Evaluation of field-ageing effects on insulating materials of composite suspension insulators. IEEE Trans. Dielectr. Electr. Insul. 2017, 24, 490–498. [Google Scholar] [CrossRef]
- Seyedmehdi, S.A.; Zhang, H.; Zhu, J. Superhydrophobic RTV silicone rubber insulator coatings. Appl. Surf. Sci. 2012, 258, 2972–2976. [Google Scholar] [CrossRef]
- Hillborg, H.; Gedde, U.W. Hydrophobicity changes in silicone rubbers. IEEE Trans. Dielectr. Electr. Insul. 1999, 6, 703–717. [Google Scholar] [CrossRef]
- Cao, H.W.; Yan, D.; Han, J.H.; Ren, H.; Lu, M.; Lv, Z.B. Investigation and corroboration of a novel method to estimate the hydrophobicity of composite insulators. IEEE Trans. Dielectr. Electr. Insul. 2012, 19, 2029–2036. [Google Scholar] [CrossRef]
- Berg, M.; Thottappillil, R.; Scuka, V. Hydrophobicity estimation of HV polymeric insulating materials development of a digital image processing method. IEEE Trans. Dielectr. Electr. Insul. 2001, 8, 1098–1107. [Google Scholar] [CrossRef]
- Peng, K.; Wang, Q.; Wang, X. Spray image analysis based measurement of hydrophobic of insulator surface. Insul. Mater. 2005, 7, 47–50. [Google Scholar] [CrossRef]
- Zhang, Z.Y.; Yan, K.; Wang, F.C.; Yang, S.J.; Li, N.C. Insulator hydrophobic identification method based on image feature extraction and BP neural network. High Volt. Eng. 2014, 40, 1446–1452. [Google Scholar] [CrossRef]
- Jarrar, I.; Assaleh, K.; El-Hag, A.H. Using a pattern recognition-based technique to assess the hydrophobicity class of silicone rubber materials. IEEE Trans. Dielectr. Electr. Insul. 2014, 21, 2611–2618. [Google Scholar] [CrossRef]
- Yang, L.; Bi, J.K.; Hao, Y.P.; Nian, L.P.; Zhou, Z.J.; Li, L.C.; Liao, Y.F.; Zhang, F.Z. A recognition method of the hydrophobicity class of composite insulators based on features optimization and experimental verification. Energies 2018, 11, 765. [Google Scholar] [CrossRef]
- Yin, Q.G.; Yang, Z.K.; Tan, Z. Pattern Recognition and Neural Networks; Machinery Industry Press: Beijing, China, 2003. [Google Scholar]
- Kowalski, P.A.; Kusy, M. Sensitivity analysis for probabilistic neural network structure reduction. IEEE Trans. Neural Netw. Learn. Syst. 2018, 29, 1919–1932. [Google Scholar] [CrossRef] [PubMed]
- Wang, G.D.; Dong, Q.; Pan, Z.K.; Zhang, W.Z.; Duan, J.M.; Bai, L.; Zhang, J.P. Retinex theory based active contour model for segmentation of inhomogeneous images. Digit. Signal Process. 2016, 50, 43–50. [Google Scholar] [CrossRef]
- Gonzalez, C.I.; Melin, P.; Castro, J.R.; Mendoza, O.; Castillo, O. An improved Sobel edge detection method based on generalized type-2 fuzzy logic. Soft Comput. 2016, 20, 773–784. [Google Scholar] [CrossRef]
- Chen, Q.; Zhao, L.; Lu, J.; Kuang, G.; Wang, N. Modified two-dimensional Otsu image segmentation algorithm and fast realization. IET Image Process. 2012, 6, 426–433. [Google Scholar] [CrossRef]
Metrics | Con | E | SNR |
---|---|---|---|
b | 17.87 | 4.07 | - |
c | 18.58 | 4.01 | 63.85 |
d | 27.01 | 4.61 | 69.56 |
HC Level | The Hydrophobic Image | Canny Operator | The Proposed Method |
---|---|---|---|
HC1 | |||
HC2 | |||
HC3 | |||
HC4 | |||
HC5 | |||
HC6 | |||
HC7 |
Number | Maximum Area Ratio | Feature Factor | Coverage | Expected Output |
---|---|---|---|---|
1 | 0.0081 | 0.9022 | 0.2101 | 1 |
2 | 0.0253 | 0.8513 | 0.2121 | 2 |
3 | 0.0603 | 0.5214 | 0.3527 | 3 |
4 | 0.5027 | 0.1450 | 0.3146 | 4 |
5 | 0.5803 | 0.5124 | 0.7353 | 5 |
6 | 0.9052 | 0.7023 | 0.9156 | 6 |
7 | 0.9908 | 0.7547 | 0.9623 | 7 |
8 | 0.0101 | 0.8638 | 0.1141 | 1 |
9 | 0.0180 | 0.4781 | 0.1779 | 2 |
10 | 0.0435 | 0.5482 | 0.2207 | 3 |
11 | 0.0404 | 0.1586 | 0.2749 | 4 |
12 | 0.7539 | 0.6021 | 0.8853 | 5 |
13 | 0.8952 | 0.7030 | 0.9006 | 6 |
14 | 0.9633 | 0.7517 | 0.9754 | 7 |
Training Errors | Spread = 1.0 | Spread = 0.8 | Spread = 0.5 | Spread = 0.2 | Spread = 0.1 |
---|---|---|---|---|---|
HC1 | 0 | 0 | 0 | 0 | 5 |
HC2 | 20 | 20 | 20 | 2 | 2 |
HC3 | 4 | 0 | 0 | 0 | 0 |
HC4 | 0 | 0 | 0 | 0 | 0 |
HC5 | 1 | 1 | 1 | 1 | 1 |
HC6 | 0 | 0 | 0 | 0 | 0 |
HC7 | 0 | 0 | 0 | 0 | 0 |
Total Number of Errors | 25 | 21 | 21 | 3 | 8 |
Error Rate | 17.85% | 15.00% | 15.00% | 2.14% | 5.71% |
Hydrophobic Grade | Improved Shape Factor Method | Multi-Fractal Method | RBF Neural Network | The Proposed Method |
---|---|---|---|---|
Misjudgment/Total Number of Samples/Correct Rate | Misjudgment/Total Number of Samples/Correct Rate | Misjudgment/Total Number of Samples/Correct Rate | Misjudgment/Total Number of Samples/Correct Rate | |
HC1 | 30/30/100% | 26/30/87% | 28/30/93% | 29/30/97% |
HC2 | 29/30/97% | 21/30/70% | 28/30/93% | 26/30/87% |
HC3 | 28/30/93% | 23/30/77% | 27/30/90% | 29/30/97% |
HC4 | 26/30/87% | 28/30/93% | 26/30/87% | 30/30/100% |
HC5 | 26/30/87% | 28/30/93% | 26/30/87% | 28/30/93% |
HC6 | 27/30/90% | 30/30/100% | 28/30/93% | 27/30/90% |
HC7 | 27/30/90% | 30/30/100% | 27/30/90% | 30/30/100% |
Total Recognition Rate | 92.0% | 88.6% | 90.0% | 94.8% |
© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Sun, Q.; Lin, F.; Yan, W.; Wang, F.; Chen, S.; Zhong, L. Estimation of the Hydrophobicity of a Composite Insulator Based on an Improved Probabilistic Neural Network. Energies 2018, 11, 2459. https://doi.org/10.3390/en11092459
Sun Q, Lin F, Yan W, Wang F, Chen S, Zhong L. Estimation of the Hydrophobicity of a Composite Insulator Based on an Improved Probabilistic Neural Network. Energies. 2018; 11(9):2459. https://doi.org/10.3390/en11092459
Chicago/Turabian StyleSun, Qiuqin, Fei Lin, Weitao Yan, Feng Wang, She Chen, and Lipeng Zhong. 2018. "Estimation of the Hydrophobicity of a Composite Insulator Based on an Improved Probabilistic Neural Network" Energies 11, no. 9: 2459. https://doi.org/10.3390/en11092459