Number Of Nodes In Hidden Layer Research Articles

Mathematical modeling for total organic carbon content prediction with logging parameters by neural networks: A case study of shale gas well in South China

The lower Cambrian Niutitang Formation in northwestern Hunan, South China, has already reached its high or over matured stage and is formed with hydrothermal activity and deposition. Thus, it is extremely difficult to predict the total organic carbon (TOC) content accurately by common methods with well-logging data. To solve this problem, we use artificial neural networks for predicting the TOC of the black shales in our study case. We got the input vectors through principal component analysis and based on the relationships and the logging response mechanism between TOC and logging data. In the back-propagation algorithm, some important parameters including the sample databases, the number of hidden layer nodes, transfer function, and weight value adjustment were all optimized correctly in the networks. Then, we built the mathematical model through a large number of learning sample datum and the error function between the actual and expected outputs, and we found that the results are good according to many performance indicators. In testing samples, mean absolute and relative errors are all reduced probably due to the datum ranges and features being focused, but the accuracy also drops when the numbers of participating samples are reduced. Through redefining the [Formula: see text] sample database, we gained more accurate values for the high-TOC source rock. Finally, we think that the results suggest that the method is suitable for shale gas resource exploration under similar geologic conditions and data characteristics.

Dynamic modeling of NOX emission in a 660 MW coal-fired boiler with long short-term memory

With the rapid development of renewables, increasing demands for the participation of coal-fired power plants in peak load regulation is expected. Frequent transients result in continuous, wide variations in NOX emission at the furnace exit, which represents a substantial challenge to the operation of SCR systems. A precise NOX emission prediction model under both steady and transient states is critical for solving this issue. In this study, a deep learning algorithm referred to as long short-term memory (LSTM) was introduced to predict the dynamics of NOX emission in a 660 MW tangentially coal-fired boiler. A total of 10000 samples from the real power plant, covering 7 days of operation, were employed to train and test the model. The learning rate, look-back time steps, and number of hidden layer nodes were meticulously optimized. The results indicate that the LSTM model has excellent accuracy and generalizability. The root mean square errors of the training data and test data are only 7.6 mg/Nm3 and 12.2 mg/Nm3, respectively. The mean absolute percentage errors are within 3%. Additionally, a comparative study between the LSTM and the widely used support vector machine (SVM) was conducted, and the result indicates that the LSTM outperforms the SVM.

Improve Efficiency of Extreme Learning Machine based on Chaos Particle Swarm Optimization Method

Extreme Learning Machine (ELM) and Regularized Extreme Learning Machine (RELM) have advantages of fast training speed and good generalization. However, ELM/RELM often needs numerous number of hidden layer nodes to get better performance. The superabundant nodes in hidden layer maybe lead to low running speed. Thus it is not feasible to use ELM in some fields that require high speed algorithms. Therefore, in this paper, we propose an Improved ELM/RELM Optimized based on Chaos Particle Swarm Optimization (CPSO-ELM/RELM) to reduce the number of hidden layer nodes, but still maintain a desirable accuracy. At the same time, it lowers the running speed compared with other algorithms. To verify the application of this method, we design numerous experiments for ELM and RRELM. Their simulation shows that the approach improves the speed of the algorithms, and the accuracy is still high. This makes it possible to use improved CPSO-ELM/RELM in some system with high real-time requirements.

Back propagation neutral network based signal acquisition for Brillouin distributed optical fiber sensors.

This manuscript proposes a method based on back propagation (BP) neural network and the spectral subtraction method to quickly obtain sensing information in Brillouin fiber optics sensors. BP neural network's characteristics which can realize any complex nonlinear mapping help to determine the frequency shift section(s) information. The training function, transfer function and number of hidden layer nodes of BP neural network are determined with experimental data. The experimental results show that comparing with traditional Lorentz fitting algorithm and edge detection with Sobel operator, the BP neural network is about 1/12 in terms of time complexity with the Lorentz algorithm, about 1/9 with the edge detection based on Sobel operator; while the respective accuracy on determine the frequency shifted section(s) has improved by 79.4% and 27.9%.

Nonlinear Process Monitoring Based on Global Preserving Unsupervised Kernel Extreme Learning Machine

Recently, the unsupervised extreme learning machine (UELM) technique as a nonlinear data mining approach has been employed to diagnose nonlinear process faults. However, during the dimensionality reduction of process observation data, UELM only aims at mining the local structure feature information of process data and lacks of the ability of preserving the global structure analysis of process data, which would perform unsatisfactorily for monitoring nonlinear process. Furthermore, the choice of the optimal number of hidden layer nodes in UELM method is still a challenging problem. To handle these two tough problems, a new fault detection approach on the basis of global preserving unsupervised kernel extreme learning machine (GUKELM) technique is proposed to monitor nonlinear process effectively. In GUKELM technique, the data global structure preserving framework is naturally incorporated into standard UELM to capture the local structure information as well as to preserve the global structure feature information of nonlinear process observation data during the dimensionality reduction. Meanwhile, to tackle the strong nonlinearity of process observation data, the kernel trick is utilized to successfully solve the challenging issue of setting the optimal number of hidden layer nodes. Based on extracted process data low dimensional feature information from GUKELM method, support vector data description algorithm is adopted to construct the monitoring statistic to detect process fault. Finally, the feasibility and effectiveness of the proposed process monitoring strategy is illustrated through a numerical nonlinear system and the continuous stirred tank reactor process which is a typical nonlinear process.

Particle Swarm Optimization-Based Deep Neural Network for Digital Modulation Recognition

Modulation recognition is a major task in many wireless communication systems including cognitive radio and signal reconnaissance. The diversification of modulation schemes and the increased complexity of the channel environment put higher requirements on the correct identification of modulated signals. Deep learning (DL) is considered as a potential solution to solve these problems due to the superior big data processing and classification capabilities. This paper proposes an efficient digital modulation recognition method based on deep neural network (DNN) model. Furthermore, we present the particle swarm optimization (PSO) algorithm to optimize the number of hidden layer nodes of the DNN so as to solve the problem that the traditional DNN is trapped in local minimum values and the number of hidden layer nodes needs selecting manually. In this paper, we utilize the proposed PSO-DNN method to learn characteristics extracted from the modulated signal added by additive white Gaussian noise (AWGN) and to train the network, which can improve the performance of recognition under the condition of low signal-to-noise ratio (SNR). The experimental results demonstrate that the recognition rate on this algorithm has improved by 9.4% and 8.8% compared with methods that adopt conventional DNN and support vector machine (SVM) when SNR equals 0 and 1 dB, respectively. Besides, another experiment compared with the genetic algorithm (GA) also proves that our proposed algorithm is more effective in optimizing the DNN. The proposed method is easy to be implemented so that it has a broad development prospect in modulation recognition.

Neural-network based boost pressure prediction for two-stage turbocharging system of diesel engine

Abstract A boost pressure prediction research based on neural network is carried out for a diesel engine with two-stage adjustable turbocharging system. First, based on the co-simulation using GT-POWER and MATLAB/Simulink model, the dynamic influence of cycle fuel injection quantity and bypass valve opening of the turbine on boost pressure of diesel engine is studied. The boost pressure shows strong nonlinear relationship with the two aforementioned affecting parameters. Second, a nonlinear autoregressive neural network prediction model (NARXNN) with external input is designed to predict the boost pressure of the diesel engine. The cycle fuel injection quantity, bypass valve opening of turbine, and current boost pressure are used as the inputs of the neural network in order to predict the future boost pressure. Then the neural network prediction model is identified based on the mean absolute percentage error (MAPE). The identification results show that the minimum MAPE (3.52%) is achieved when the number of hidden layer nodes is 15. At last, the neural network prediction model for boost pressure is verified by simulation. The one-step prediction simulation results of boost pressure show that the established prediction model can accurately predict the boost pressure, and the MAPE of the verification data is 3.98%. And the multistep prediction simulation results show that the pressure prediction error in the 50-step prediction domain is within 0.05bar, which indicates that the prediction model has high accuracy.

Selecting the Optimal Hidden Layer of Extreme Learning Machine Using Multiple Kernel Learning

Extreme learning machine (ELM) is emerging as a powerful machine learning method in a variety of application scenarios due to its promising advantages of high accuracy, fast learning speed and easy of implementation. However, how to select the optimal hidden layer of ELM is still an open question in the ELM community. Basically, the number of hidden layer nodes is a sensitive hyperparameter that significantly affects the performance of ELM. To address this challenging problem, we propose to adopt multiple kernel learning (MKL) to design a multi-hidden-layer-kernel ELM (MHLK-ELM). Specifically, we first integrate kernel functions with random feature mapping of ELM to design a hidden-layer-kernel ELM (HLK-ELM), which serves as the base of MHLK-ELM. Then, we utilize the MKL method to propose two versions of MHLK-ELMs, called sparse and non-sparse MHLK-ELMs. Both two types of MHLK-ELMs can effectively find out the optimal linear combination of multiple HLK-ELMs for different classification and regression problems. Experimental results on seven data sets, among which three data sets are relevant to classification and four ones are relevant to regression, demonstrate that the proposed MHLK-ELM achieves superior performance compared with conventional ELM and basic HLK-ELM.

Ship roll motion prediction based on ℓ1 regularized extreme learning machine.

In this paper, a new method is proposed for prediction of ship roll motion based on extreme learning machine (ELM). To improve the prediction accuracy and avoid over or under fitting, two techniques are adopted to select the appropriate structure of ELM. First, the inputs of the ELM are selected from the roll motion time series using Lipschitz quotient method. Second, the number of hidden layer nodes is determined via ℓ1 regularized technique. Finally, the ℓ1 regularized ELM is solved by least angle regression (LAR) algorithm. The effectiveness of the proposed method is demonstrated by ship roll motion prediction experiments based on the real measured ship roll motion time series.

Short-term wind speed prediction based on improved PSO algorithm optimized EM-ELM

ABSTRACTShort-term wind speed prediction is of importance for power grids. It can mitigate the disadvantageous impacts of wind farms on power systems and enhance the competitiveness of wind power in electricity markets. A short-term wind speed prediction model is proposed. Many useless neurons of incremental extreme learning machine have little influences on the final output, at the same time, reduce the efficiency of the algorithm. In order to solve this problem, based on error minimized extreme learning machine, an improved particle swarm optimization algorithm is proposed to decrease the number of useless neurons, achieve the goal of reducing the network complexity and improving the efficiency of the algorithm. The stability and convergence of the algorithm are proved. The actual short-term wind speed time series is used as the research object. Multistep prediction simulation of short-term wind speed is performed out. Compared with the other prediction models, the simulation results show that the prediction model proposed in this paper reduces the training time of the model and decreases the number of hidden layer nodes. The prediction model has higher prediction accuracy and reliability, meanwhile improve the prediction performance indicators.

Intrusion Detection of Data Platform Based on Extreme Learning Machine in Civil and Military Integration

With the deepening development of military and civilian integration, the explosive growth of internal information is guaranteed and the advantages of big data in ensuring system efficiency are increasingly obvious. However, many hidden dangers are exposed. To solve the problem of ELM, which is overly dependent on the single hidden layer feedforward neural network (SLFN) with many hidden nodes, this paper proposes an SVM-based ELM (SVM-ELM) Intrusion detection algorithm. This algorithm reduces the number of hidden layer nodes significantly and uses SVM weight to optimize the weights and offsets of each node, which improves the decision-making level of nodes and significantly improves the decision performance of ELM. Verification using KDD99 dataset shows that the generalized performance of SLFN with only 5 nodes constructed by SVM-ELM has obvious advantages compared with the original ELM. Compared with other algorithms such as BP, the SVM-ELM algorithm can quickly complete the training and has a higher detection accuracy.

Production prediction and energy-saving model based on Extreme Learning Machine integrated ISM-AHP: Application in complex chemical processes

One of the key issues of the sustainable development in all countries is industrial productivity improvement, especially the production capacity improvement and energy-saving of complex chemical processes. Therefore, this paper proposes a production prediction and energy-saving model based on Extreme Learning Machine (ELM) integrated Interpretative Structural Modeling (ISM) and Analytic Hierarchy Process (AHP). The factors that affect the productivity are divided into different levers by the ISM. And then the attributes of each layer are fused by the AHP based on the entropy weight, which greatly reduces the complexity of the input attributes. Moreover, the production prediction and energy-saving model is established based on the ELM. Compared with the traditional ELM, the validity and the practicability of the proposed method are verified by University of California Irvine (UCI) datasets. Finally, the proposed method is applied in the production capacity prediction and energy-saving of ethylene production systems and Purified Terephthalic Acid (PTA) production systems. The experimental results show the proposed method could reduce the number of hidden layer nodes and improve the training time of the ELM. Furthermore, the prediction accuracy of the ethylene production and the PTA production reaches about 99% to improve the energy efficiency of complex chemical processes.

Artificial bee colony algorithm–optimized error minimized extreme learning machine and its application in short-term wind speed prediction

In order to improve the prediction accuracy of short-term wind speed, a short-term wind speed prediction model based on artificial bee colony algorithm optimized error minimized extreme learning machine model is proposed. The extreme learning machine has the advantages of fast learning speed and strong generalization ability. But many useless neurons of incremental extreme learning machine have little influences on the final output and, at the same time, reduce the efficiency of the algorithm. The optimal parameters of the hidden layer nodes will make network output error of incremental extreme learning machine decrease with fast speed. Based on the error minimized extreme learning machine, artificial bee colony algorithm is introduced to optimize the parameters of the hidden layer nodes, decrease the number of useless neurons, reduce training and prediction error, achieve the goal of reducing the network complexity, and improve the efficiency of the algorithm. The error minimized extreme learning machine prediction model is constructed with the obtained optimal parameters. The stability and convergence property of artificial bee colony algorithm optimized error minimized extreme learning machine model are proved. The practical short-term wind speed time series is used as the research object and to verify the validity of the prediction model. Multi-step prediction simulation of short-term wind speed is carried out. Compared with other prediction models, simulation results show that the prediction model proposed in this article reduces the training time of the prediction model and decreases the number of hidden layer nodes. The prediction model has higher prediction accuracy and reliability performance, meanwhile improves the performance indicators.

Ultrasound–assisted extraction optimization of phenolic compounds from Psidium guajava L. using artificial neural network-genetic algorithm

Artificial Neural Network-Genetic Algorithm (ANN-GA) was used in this work to model the extraction parameters using Ultrasound Assisted Extraction (UAE) in producing phenolic compounds of guava (Psidium guajava L.) leaves. The ANN-GA model was used to evaluate and predict the effect of frequency, temperature, time, and their interaction under optimum conditions. The controlled parameters were frequency (20-40 kHz), temperature (2535 °C), and sonication time (20-40 minutes). Furthermore, the input and output parameters were modeled by LM-back propagation ANN. The optimization variable was then determined by GA. ANN trial and error method was used to determine the bias, weight, and number of hidden layer nodes. The fitness function for optimization with GA used the bias and weight which were obtained from ANN. The prediction of total phenolic on guava leaves using the ANN model shows the mean absolute precentage error value is 13.07. Based on the experiment the optimum conditions were obtained at 40 kHz, 35°C,30 minutes and resulted 589.02 ppm. While ANN-GA model obtained optimum results at 39.79 kHz, 34.83°C, 20 min, then the mean absolute error optimization was 3.46. It can be concluded that ANN-GA model can be used for optimizing the phenolic yields of guava leaves.

Machine learning-based thermal response time ahead energy demand prediction for building heating systems

Energy demand prediction of building heating is conducive to optimal control, fault detection and diagnosis and building intelligentization. In this study, energy demand prediction models are developed through machine learning methods, including extreme learning machine, multiple linear regression, support vector regression and backpropagation neural network. Seven different meteorological parameters, operating parameters, time and indoor temperature parameters are used as feature variables of the model. Correlation analysis method is utilized to optimize the feature sets. Moreover, this paper proposes a strategy for obtaining the thermal response time of building, which is used as the time ahead of prediction models. The prediction performances of extreme learning machine models with various hidden layer nodes are analyzed and contrasted. Actual data of building heating using a ground source heat pump system are collected and used to test the performances of the models. Results show that the thermal response time of the building is approximately 40 min. Four feature sets are obtained, and the performances of the models with feature set 4 are better. For different machine learning methods, the performances of extreme learning machine models are better than others. In addition, the optimal number of hidden layer nodes is 11 for the extreme learning machine model with feature set 4.

Product Family Configuration Price Prediction Based On Neural Network

In order to quickly and effectively respond to customer enquiry in the dynamic changeable market, a new method for product family configuration price prediction is proposed based on neural network with adjustable number of hidden layer nodes. Based on product family configuration history of various product variants and the corresponding prices, the relationship model between the product configuration and price is established using neural network, which is utilized to predict the market price of the newly configured product variant for some random order.

Multivariate Molten Iron Quality Modeling Based on Improved Incremental Random Vector Functional-link Networks

Aiming at the problems that the model structure is complex and the training time is too long for traditional incremental random vector functional-link networks (I-RVFLNs), this paper proposes an improved incremental random vector functional-link networks (Improved I-RVFLNs). Compared with the traditional I-RVFLNs, this algorithm changes the mode of constructing network. First, the number of hidden layer nodes is fixed in advance, and then the output weights of all nodes are changed in sequence. Finally, based on the actual data of a large blast furnace in Southern China, a nonlinear auto-regressive exogenous (NARX) model of multivariate molten iron quality (MIQ) is established by using the improved I-RVFLNs, and then it is compared with other modeling algorithms. The results show that compared with other quality modeling algorithms, the proposed method not only has higher model accuracy and shorter test time, but also solves the problem that the number of optimal hidden layer nodes in the conventional RVFLNs algorithm is difficult to choose, and the network is too complicated due to too many hidden nodes in the traditional I-RVFLNs.

The Forecasting Model of Stock Price Based on PCA and BP Neural Network

Based on Principal Component Analysis (PCA) and Back Propagation neural network, this paper establishes stock forecast model, and takes the Yunnan Baiyao (000538) as example, 29 indicators are selected from stocks technical analysis, and the neural network is input after dimension reduction and further confirms number of hidden layer nodes, learning rate, activation function and training function of the network in accordance with comparison and analysis of Mean Square Error (MSE) and Mean Absolute Error (MAE) in different parameter data experiments. Lastly, the model with steadiness and accuracy is obtained.

Determination of zinc oxide content of mineral medicine calamine using near-infrared spectroscopy based on MIV and BP-ANN algorithm

Near-infrared (NIR) spectroscopy has been widely used in the analysis fields of traditional Chinese medicine. It has the advantages of fast analysis, no damage to samples and no pollution. In this research, a fast quantitative model for zinc oxide (ZnO) content in mineral medicine calamine was explored based on NIR spectroscopy. NIR spectra of 57 batches of calamine samples were collected and the first derivative (FD) method was adopted for conducting spectral pretreatment. The content of ZnO in calamine sample was determined using ethylenediaminetetraacetic acid (EDTA) titration and taken as reference value of NIR spectroscopy. 57 batches of calamine samples were categorized into calibration and prediction set using the Kennard-Stone (K-S) algorithm. Firstly, in the calibration set, to calculate the correlation coefficient (r) between the absorbance value and the ZnO content of corresponding samples at each wave number. Next, according to the square correlation coefficient (r2) value to obtain the top 50 wave numbers to compose the characteristic spectral bands (4081.8–4096.3, 4188.9–4274.7, 4335.4, 4763.6,4794.4–4802.1, 4809.9, 4817.6–4875.4cm−1), which were used to establish the quantitative model of ZnO content using back propagation artificial neural network (BP-ANN) algorithm. Then, the 50 wave numbers were operated by the mean impact value (MIV) algorithm to choose wave numbers whose absolute value of MIV greater than or equal to 25, to obtain the optimal characteristic spectral bands (4875.4–4836.9, 4223.6–4080.9cm−1). And then, both internal cross and external validation were used to screen the number of hidden layer nodes of BP-ANN. Finally, the number 4 of hidden layer nodes was chosen as the best. At last, the BP-ANN model was found to enjoy a high accuracy and strong forecasting capacity for analyzing ZnO content in calamine samples ranging within 42.05–69.98%, with relative mean square error of cross validation (RMSECV) of 1.66% and coefficient of determination (R2) of 95.75% in internal cross and relative mean square error of prediction (RMSEP) of 1.98%, R2 of 97.94% and ratio of performance to deviation (RPD) of 6.11 in external validation.

A thermal response time ahead energy demand prediction strategy for building heating system using machine learning methods

Energy demand prediction of building heating is conducive to optimal control, fault detection and diagnosis and building intelligent. In this paper, the prediction models are developed using machine learning methods including extreme learning machine (ELM), multiple linear regression, support vector regression and BP neural network. The feature variable sets are optimized through correlation analysis and supplementing indoor temperature. Besides, this paper proposed a strategy to determine the time ahead of prediction model. The thermal response time of building is used as the prediction time step of model. The prediction performances of ELM models with different hidden layer nodes are analyzed and contrasted. The actual data of the building heating using ground source heat pump system are collected and used to test the performances of the models. The results show that the thermal response time of the building is about 40 minutes. Four feature sets are obtained and performances of models with FS4 are better. For different machine learning methods, the performances of ELM models are better than others. In addition, the optimal number of hidden layer nodes is 11 for the ELM model with FS4.