Temperature extraction for Brillouin optical fiber sensing system based on extreme learning machine

Drought is a complex and frequently occurring natural hazard in many parts of the world. Therefore, accurate drought forecasting is essential to mitigate its adverse impacts. This research has inferred the implication and the appropriateness of the extreme learning machine (ELM) algorithm for drought forecasting. For numerical evaluation, time series data of the Standardized Precipitating Temperature Index (SPTI) are used for nine meteorological stations located in various climatological zones of Pakistan. To assess the performance of ELM, this research includes parallel inferences of multilayer perceptron (MLP) and autoregressive integrated moving average (ARIMA) models. The performance of each model is assessed using root mean square error (RMSE), mean absolute error (MAE), mean absolute percent error (MAPE), Kling‐Gupta efficiency (KGE), Willmott index (WI), and Karl Pearson’s correlation coefficient. Generally, graphical results illustrated an excellent performance of the ELM algorithm over MLP and ARIMA models. For training data of SPTI‐1, ELM’s best performance has observed at Chitral station (RMSE = 0.374, KGE = 0.838, WI = 0.960, MAE = 0.272, MAPE = 259.59, R = 0.93). For SPTI‐1 at Astore station, the numerical results are (RMSE = 0.688, KGE = 0.988, WI = 0.997, MAE = 0.798, MAPE = 247.35). The overall results indicate that the ELM outperformed by producing the smallest RMSE, MAE, and MAPE values and maximum values for KGE, WI, and correlation coefficient values at almost all the selected meteorological stations for (1, 3, 6, 9, and 12) month time scales. In summary, this research endorses the use of ELM for accurate drought forecasting.

Distributed fiber-optic sensors employ the conventional fiber as the sensing medium and allow all locations along the optical fiber to be measured simultaneously. Thus the sensing scheme provides an efficient way of simultaneously achieving multi-points sensing with a convenient and low-cost configuration. One of the most popular technologies that enables distributed sensing is based on Brillouin scattering. Characteristics of Brillouin scattering are determined mostly by the material properties of the optical fiber and the Brillouin Frequency Shift (BFS) has linear dependence on both the temperature and strain of the fiber under test (FUT). Based on this principle, intensive research efforts have been made in distributed Brillouin fiber sensors during the past two decades. Among them, Brillouin Optical Time-Domain Analysis (BOTDA) has attracted intensive research interest due to its promising properties for temperature and strain monitoring. In BOTDA the local Brillouin Gain Spectrum (BGS) is reconstructed by scanning the frequency offset of a continuous wave and an optical pulse around BFS when they are counter-propagating inside FUT, and the local BFS and hence the temperature and/ or strain information are retrieved accordingly. Usually curve fitting techniques are employed in determining the BFSs of the measured BGSs along the FUT, in which the measured BGS is fitted using an ideal curve and the BFS is found to be the frequency with the peak gain on the fitted curve. Fitting techniques using different lineshape for ideal curves, e.g. Lorentzian curve fitting (LCF), Gaussian curve fitting (GCF), pseudo-Voigt curve fitting (pVCF), and quadratic curve fitting (QCF) have been used to find the BFS distribution along FUT. However, the curve fitting techniques require careful initialization of model parameters and as many data points as possible collected on the measured BGS to ensure the fitting accuracy, otherwise the accuracy in BFS determination will degrade significantly. In addition, long processing time is needed for the iteration of fitting algorithms to find the BFS, especially when the sensing distance is long with a large number of BGSs collected and processed. Recently we have utilized the nonlinear mapping capability of Artificial Neural Network (ANN) in a BOTDA system and has successfully extracted temperature distribution from the measured BGSs along the FUT. Before temperature extraction ANN is trained to acquire the knowledge about the BGS patterns under different temperatures, thus it allows better accuracy even if the data points on the collected BGS become fewer when large frequency scanning step is adopted during the acquisition of BGSs. Moreover, the processing speed of ANN for temperature extraction is much faster than that of curving fitting techniques. In this presentation, we will review our work on using ANN for temperature extraction in a BOTDA system. Different idea spectrum profiles are used for training ANN, and the corresponding performances are compared with those using conventional curve fitting techniques. Some potential applications using ANN in BOTDA are also discussed. We believe that ANN can be an attractive tool for direct temperature or strain extraction in BOTDA system at high accuracy and fast speed.

A Brillouin optical time domain reflectometer (BOTDR) is a fiber optic strain distributed sensing system which can measure strain along arbitrary regions of an optical fiber. We have used this system for structural health monitoring of composite structures, such as racing yachts. Although the BOTDR is useful to measure strain distributions of a large area efficiently, it is difficult to accurately measure strain changing sharply within the length of the spatial resolution. The resolution depends on the width of the pulse light emitted from the BOTDR and it is 2 m for the pulse width of 20 ns. In this paper, we propose a new technique to improve the spatial resolution of a commercially available BOTDR in strain measurements. This technique is based on analysis of the Brillouin gain spectrum along a sensing fiber. We confirmed experimentally that the BOTDR using 20 ns pulse could measure strain which has a rectangular distribution of 0.3 m in the sensing fiber.

In recent years, the demand for safe coalbed methane pipelines has dramatically increased in our society. With the superior advantages of large-scale monitoring, low cost, high sensitivity, quick response and immunity to electromagnetic interference, distributed optical fiber sensors provide an appropriate solution for large scale pipeline safety monitoring. Among them, Brillouin optical time domain reflectometer (BOTDR) and phase sensitive optical time domain reflectometer (Φ-OTDR) are fast developing scientific techniques and attract a significant amount of research attention. A multi-parameter measurement scheme based on BOTDR and Φ-OTDR is designed to monitor the strain and vibration conditions of the pipeline, respectively. In the BOTDR system, the pulse probe light modulated by an acoustic optical modulator (AOM) is launched into the sensing optical fibers and the beat frequency signal is obtained through coherent detection. Then, the Brillouin gain spectrum is figured out via heterodyne frequency sweeping. Ultimately the curve of Brillouin frequency shift versus sensing fiber length is plotted to realize the distributed strain measurement. In the Φ-OTDR system, light pulses from a laser with narrow linewidth and low frequency drift are gated into the fiber through an AOM, then amplified by erbium-doped fiber amplifier (EDFA), and finally coupled into the sensing fiber via a circulator. The Rayleigh backscattered light are monitored with a photodector (PD). The vibration position can be detected by subtracting the post-vibration Φ-OTDR trace from an earlier stored trace. The experimental results demonstrated that over a fiber sensing length more than 10 km, the accuracy of Brillouin frequency shift at the far end of the sensing fiber was ±0.88 MHz, which corresponds to the strain accuracy of 40 µe, and a vibration measurement with a positional accuracy less than 20m is also achieved, which can satisfy the practical requirements and shows considerable application potential for coalbed methane pipelines safety monitoring.

In this paper, an extreme learning machine (ELM) network based on an improved shuffled frog leaping algorithm (CCSFLA) is applied in early bearing fault diagnosis. ELM is a new type of single layer...

Robustness of Extreme Learning Machine in the prediction of hydrological flow series

An improved classification approach is proposed to solve the hot research problem of some complex multiclassification samples based on extreme learning machine (ELM). ELM was proposed based on the single‐hidden layer feed‐forward neural network (SLFNN). ELM is characterized by the easier parameter selection rules, the faster converge speed, the less human intervention, and so on. In order to further improve the classification precision of ELM, an improved generation method of the network structure of ELM is developed by dynamically adjusting the number of hidden nodes. The number change of the hidden nodes can serve as the computational updated step length of the ELM algorithm. In this paper, the improved algorithm can be called the variable step incremental extreme learning machine (VSI‐ELM). In order to verify the effect of the hidden layer nodes on the performance of ELM, an open‐source machine learning database (University of California, Irvine (UCI)) is provided by the performance test data sets. The regression and classification experiments are used to study the performance of the VSI‐ELM model, respectively. The experimental results show that the VSI‐ELM algorithm is valid. The classification of different degrees of broken wires is now still a problem in the nondestructive testing of hoisting wire rope. The magnetic flux leakage (MFL) method of wire rope is an efficient nondestructive method which plays an important role in safety evaluation. Identifying the proposed VSI‐ELM model is effective and reliable for actually applying data, and it is used to identify the classification problem of different types of samples from MFL signals. The final experimental results show that the VSI‐ELM algorithm is of faster classification speed and higher classification accuracy of different broken wires.

Extreme Learning Machine (ELM) is a learning algorithm proposed recently to train single hidden layer feed forward networks (SLFN). It has many attractive properties that include better generalization performance and very fast learning. ELM starts by assigning random values to the input weights and hidden biases and then in one step it determines the output weights using Moore-Penrose generalized inverse. Despite the aforementioned advantages, ELM performance might be affected by the random initialization of weights and biases or by the large generated network which might contain unnecessary number of neurons. In order to increase the generalization performance and to produce more compact networks, a hybrid model that combines ELM with competitive swarm optimizer (CSO) is proposed in this paper. The proposed model (CSONN-ELM) optimizes the weights and biases and dynamically determines the most appropriate number of neurons. To evaluate the effectiveness of the CSONN-ELM, it is experimented using 23 benchmark datasets, and compared to a set of static rules extracted from literature that are used to determine the number of neurons of SLFN. Moreover, it is compared to two dynamic methods that are used to enhance the performance of ELM, that are Optimally pruned ELM (OP-ELM) and metaheuristic based ELMs (Particle Swarm Optimization-ELM and Differential Evolution-ELM). The obtained results show that the proposed method enhances the generalization performance of ELM and overcomes the static and dynamic methods.

Among the different computational intelligence techniques avalaible for hyperspectral data classification, support vector machines (SVMs) have played a dominant role. Recently, a new learning algorithm for single-layer feedforward neural networks called the extreme learning machine (ELM) was proposed. This technique is competitive with SVMs in terms of accuracy, learning speed, and computational scalability. In this article, we propose and evaluate the use of ELM for land-cover classification from hyperspectral images. In addition, we consider two ELM-based techniques integrating spectral and spatial information of the image. The first is a scheme that uses a majority vote approach in order to combine the results of a pixel-wise spectral classification by ELM and a segmentation map obtained by a watershed algorithm. The second introduces spatial information from a small spatial neighbourhood after the classification by ELM. We show the usefulness of spatial–spectral ELM-based classification techniques in hyperspectral imaging. The results are compared to those obtained by similar SVM-based techniques and show improved classification results and much lower execution time. These simple and computationally cheap methods can be combined with others traditionally applied to hyperspectral images.

The increasing use of artificial neural network in the prediction of yarn quality properties calls for constant improvement of the models. This research work reports the use of a novel training algorithm christened extreme learning machines (ELM) to prediction yarn tensile strength (strength). ELM was compared to the Backpropagation (BP) and a hybrid algorithm composed of differential evolution and ELM and named DE-ELM. The three yarn strength prediction models were trained up to a mean squared error (mse) of 0.001. This is an arbitrary level of mse that was selected to enable a comparative study of the performance of the three algorithms. According to the results obtained in this research work, the BP model needed more time for training, while the ELM model recorded the shortest training time. The DE-ELM model was in between the two models. The correlation coefficient (R2) of the BP model was lower than that of ELM model. In comparison to the other two models the DE-ELM model gave the highest R2 value.

We propose and experimentally demonstrate the use of principal component analysis (PCA) based pattern recognition to extract temperature distribution from the measured Brillouin gain spectra (BGSs) along the fiber under test (FUT) obtained by Brillouin optical time domain analysis (BOTDA) system. The proposed scheme employs a reference database consisting of relevant ideal BGSs with known temperature attributes. PCA is then applied to the BGSs in the reference database as well as to the measured BGSs so as to reduce their size by extracting their most significant features. Now, for each feature vector of the measured BGS, we determine its best match in the reference database comprised of numerous reduced-size feature vectors of the ideal BGSs. The known temperature attribute corresponding to the best-matched BGS in the reference database is then taken as the extracted temperature of the measured BGS. We analyzed the performance of PCA-based pattern recognition algorithm in detail and compared it with that of curve fitting method. The experimental results validate that the proposed technique can provide better accuracy, faster processing speed and larger noise tolerance for the measured BGSs. Therefore, the proposed PCA-based pattern recognition algorithm can be considered as an attractive method for extracting temperature distributions along the fiber in BOTDA sensors.

This paper develops a state-of-charge (SOC) estimation model for a lithium-ion battery using an improved extreme learning machine (ELM) algorithm. ELM is suitable for an SOC estimation since the ELM algorithm has fast estimation speed, good generalization performance, and high accuracy. However, the performance of ELM is highly dependent on training accuracy and the number of neurons in a hidden layer. Hence, a gravitational search algorithm (GSA) is applied to improve the ELM computational intelligence by searching for the optimal value hidden layer neurons. The optimal ELM-based GSA model does not require internal battery knowledge and mathematical model for an SOC estimation. The model robustness is validated at different temperatures using different electric vehicle drive cycles. The performance of the ELM-GSA model is verified with two popular neural network methods: back-propagation neural network (BPNN) and radial basis function neural network (RBFNN). The results are evaluated using different error rates and computation costs. The results demonstrate that the ELM-based GSA model offers a higher accuracy and lower SOC error rate than those of BPNN-based GSA and RBFNN-based GSA models. Furthermore, a detailed comparative study between the proposed model and existing SOC strategies is conducted, which also demonstrates the superiority of the proposed model.

Extreme learning machine (ELM) is a fast and quick learning algorithm with better generalization performance. However, the randomness of input weight and hidden layer bias may affect the overall performance of ELM. This paper proposed a new approach to determine the optimized values of input weight and hidden layer bias for ELM using whale optimization algorithm (WOA), which we call WOA-ELM. An online gearbox vibration signals is used in this study. Empirical mode decomposition (EMD) and complementary mode decomposition (CEEMD) are used to decompose the signals into sub-signals known as intrinsic mode functions (IMFs). Then, statistical features are extracted from selected IMFs. WOA-ELM is used for classification of healthy and faulty condition of gearbox. The result shows that WOA-ELM provide better classification result as compared with conventional ELM. Therefore, this study provide a new diagnosis approach for gearbox application.

We consider the problem of intrusion detection in a computer network, and investigate the use of extreme learning machines (ELMs) to classify and detect the intrusions. With increasing connectivity between networks, the risk of information systems to external attacks or intrusions has increased tremendously. Machine learning methods like support vector machines (SVMs) and neural networks have been widely used for intrusion detection. These methods generally suffer from long training times, require parameter tuning, or do not perform well in multi-class classification. We propose a basic ELM method based on random features, and a kernel based ELM method for classification. We compare our methods with commonly used SVM techniques in both binary and multi-class classifications. Simulation results show that the proposed basic ELM approach outperforms SVM in training and testing speed, while the proposed kernel based ELM achieves higher detection accuracy than SVM in multi-class classification case.

We report on an approach for reducing the effects of temperature in a fiber optic distributed sensor. This technique employs a sensing fiber and a Brillouin optical time domain reflectometer (BOTDR). The BOTDR has been proposed for measuring both strain and optical loss distribution along optical fibers by accessing only one end of the fiber. The BOTDR analyzes changes in the Brillouin frequency shift caused by strain. This device can measure distributed strain with an accuracy of better than plus or minus 60 X 10-6 and a high spatial resolution of up to 1 m over a 10 km long fiber. However, temperature fluctuations have an adverse effect on the accuracy with which the Brillouin frequency shift can be measured because the shift changes with temperature as well as with strain. This has meant that both spatial and temporal fluctuations in temperature must be compensated for when a fiber optic distributed sensor is used for continuous strain measurements in massive civil structures. We describe a method for the simultaneous determination of distributed strain and temperature which separates strain and temperature in a fiber optic sensor.© (1999) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.