Wavelet Neural Network Method Research Articles

A novel growing wavelet neural network algorithm for solving chemotaxis systems with blow‐up

In this study, we introduce a new growing neural network algorithm that is based on wavelet neural networks and call our algorithm a growing wavelet neural network (GWNN) method. We apply our proposed scheme to train a wavelet neural network to solve chemotaxis problems with blow‐up. These problems are highly nonlinear time‐dependent systems of partial differential equations, and it is a challenge to get the pattern of the solution accurately. The proposed structure is partial retraining of the network, which increases its capacity to catch the spiky pattern of the solution. Our neural network‐based algorithm allows us to solve the nonlinear chemotaxis problems without the use of linearization techniques and regularization techniques, most of which reduce the accuracy of the model. This mesh‐free‐based method can manage a variety of blow‐up models with curved boundaries without imposing an extra cost. By proving the consistency and stability of the method, we show the convergence of GWNN solutions to analytical solutions of the chemotaxis problem. Several illustrative examples and simulation results are provided to demonstrate the correctness of the results and the robust performance of the presented algorithm. Moreover, to illustrate the effectiveness of the GWNN method, we make a comparison with two other network‐based methods.

Scientific Research Management Helping the Development of Regional Cultural Industry from the Perspective of Artificial Intelligence

In order to solve the problem that traditional scientific research management is restricted by management methods and tools, and it is difficult to realize the development of cultural industry, this paper proposes a research method that scientific research management helps the development of the regional cultural industry from the perspective of artificial intelligence. Based on a wavelet neural network based on artificial intelligence, this paper establishes the goodwill value evaluation of cultural industry development under scientific research management. Firstly, the paper uses the hierarchical structure theory to construct the theoretical analysis framework and evaluation index system of cultural value and divides the influencing factors of cultural industry value into three levels: input layer network structure, hidden layer network structure, and output layer network structure. Then the wavelet neural network method is used to analyze, and the three-layer network structure is superimposed to reflect the hierarchical relationship between the index systems. Through the combination of normative analysis and empirical analysis, it verifies that the method has unique advantages and feasibility and realizes the promotion of scientific research management to the development of the regional cultural industry.

(Retracted) Deep graph convolutional network-based high-performance detection method for spectral domain gesture image stream

The use of vision-based high-performance detection technology has become an innovative technical methodology. Gestures can express more actions and even emotions, which is more in line with the design idea of large-scale integrated human–computer interaction software, and then assists the development of emotion recognition and other fields. The emerging deep graph convolutional network can capture the interdependence between instances, and then infer the complete information of the image based on specific features. The two-dimensional discrete wavelet and multi-resolution analysis are used to replace the traditional Fourier transform to realize the convolution operation, which improves the accuracy of the generated graph data. This work studies the spectral clustering method of Graph Wavelet Neural Network and adds the local correlation preserving support vector machine as a classifier. This classifier has a simplified structure compared with the cascade classifier and can achieve faster and stable classification results. On the test set, the average accuracy of the algorithm is 93.40%, the recall rate is 96.27%, and the average detection time per frame is 359 ms.

Wavelet neural network and complete ensemble empirical decomposition method to traffic control prediction

Traffic control prediction is one of the important issues of smart cities in that, by studying traffic parameters, there can be provided more peace and comfort in appropriate traffic routes. Combination of new and different technologies and scientific technical models for this complex prediction has always been paid attention to by researchers. In this paper, by presenting and improving one of the new methods of data collection with traffic congestion index, the appropriate models for predicting traffic control have been compared. Rapid and inexpensive collection of information and, the dynamics and momentary changes of traffic flows showed that the use of wavelet neural network was more accurate than other models of traffic control prediction. The application of combined Wavelet Neural Network with Complete Ensemble Empirical Mode Decompositionin traffic control prediction in this paper as CEEMD & WNN showed that the prediction accuracy increased compared to ARIMA, WNN, HYBRID ARIMA & WNN, TN methods and this new method has reasonable performance against the evaluation criteria to predict traffic control.

A novel adaptive noise reduction method for field natural gas pipeline defect detection signals

To suppress noise interference and improve defect identification accuracy, we propose a transfer learning-based adaptive wavelet neural network (AWNN) method for pipeline defect detection signal denoising. Simulated noisy signals and field detection signals are used to determine the network's initial structure and fine-tune its parameters. The Meyer wavelet basis function is chosen as an excitation function based on the defect signal waveform. The pretrained structure is determined by a search algorithm. The network parameters are adaptively fine-tuned according to the sample entropy of the noise. The new genetic beetle antennae search strategy (GBA) optimization algorithm is proposed to improve the AWNN to avoid local optima. Compared with traditional noise reduction methods, the AWNN has excellent ability to recovery the pure denoised signal with a higher SNR and smaller RMSE. The AWNN can realize adaptive noise reduction of different field signals, providing reliable input for subsequent defect diagnosis.

PSOWNNs-CNN: A Computational Radiology for Breast Cancer Diagnosis Improvement Based on Image Processing Using Machine Learning Methods.

Early diagnosis of breast cancer is an important component of breast cancer therapy. A variety of diagnostic platforms can provide valuable information regarding breast cancer patients, including image-based diagnostic techniques. However, breast abnormalities are not always easy to identify. Mammography, ultrasound, and thermography are some of the technologies developed to detect breast cancer. Using image processing and artificial intelligence techniques, the computer enables radiologists to identify chest problems more accurately. The purpose of this article was to review various approaches to detecting breast cancer using artificial intelligence and image processing. The authors present an innovative approach for identifying breast cancer using machine learning methods. Compared to current approaches, such as CNN, our particle swarm optimized wavelet neural network (PSOWNN) method appears to be relatively superior. The use of machine learning methods is clearly beneficial in terms of improved performance, efficiency, and quality of images, which are crucial to the most innovative medical applications. According to a comparison of the process's 905 images to those of other illnesses, 98.6% of the disorders are correctly identified. In summary, PSOWNNs, therefore, have a specificity of 98.8%. Furthermore, PSOWNNs have a precision of 98.6%, which means that, despite the high number of women diagnosed with breast cancer, only 830 (95.2%) are diagnosed. In other words, 95.2% of images are correctly classified. PSOWNNs are more accurate than other machine learning algorithms, SVM, KNN, and CNN.

Internal model control for rocket launcher position servo system based on improved wavelet neural network

This paper proposes an improved wavelet neural network-internal model controller (WNN-IMC) for the rocket launcher position servo system. Due to complex nonlinearities and uncertainties of external disturbances in the rocket launcher position servo system, it is vitally challenging to establish its accurate model by the mechanical modeling technique. A wavelet neural network (WNN) identification method is proposed to determine the system mathematical model through test datum, which optimized by the hybrid algorithm of differential evolution (DE) and particle swarm optimization (PSO). Then, the proposed method is applied to identify the semi-physical simulation platform of the rocket launcher velocity servo system. The results demonstrate that the validity of the DEPSO-WNN method is better than that of the WNN and PSO-WNN methods. Finally, compared with the WNN-IMC controller and the ADRC controller, the effectiveness of the improved WNN-IMC controller is verified by the semi-physical simulation experiments.

Early Fluid Resuscitation of Burn Patients Based on High-Precision Weighing System

Correct and continuous weight measurement of a burn patient is important for fluid resuscitation. Generally, a body weight bed uses four sensors to obtain signals. However, owing to signals dynamic characteristics, the air fluidized therapy bed has a large error in this mode, which is difficult to reduce. In this study, we optimized the sensor configuration scheme and developed a set of signal processing and methods to evaluate noise reduction. Additionally, the wavelet neural network was used to improve the weighing accuracy. Furthermore, medical tools are located on different body parts of a patient to acquire weight characteristics to assist with the rehydration of patient. The average error of 183.33g is acceptable from a medical perspective. An average error of 250.21g in dynamic weight after using wavelet de-noising and wavelet neural network methods was obtained.

Research on improved convolutional wavelet neural network

Artificial neural networks (ANN) which include deep learning neural networks (DNN) have problems such as the local minimal problem of Back propagation neural network (BPNN), the unstable problem of Radial basis function neural network (RBFNN) and the limited maximum precision problem of Convolutional neural network (CNN). Performance (training speed, precision, etc.) of BPNN, RBFNN and CNN are expected to be improved. Main works are as follows: Firstly, based on existing BPNN and RBFNN, Wavelet neural network (WNN) is implemented in order to get better performance for further improving CNN. WNN adopts the network structure of BPNN in order to get faster training speed. WNN adopts the wavelet function as an activation function, whose form is similar to the radial basis function of RBFNN, in order to solve the local minimum problem. Secondly, WNN-based Convolutional wavelet neural network (CWNN) method is proposed, in which the fully connected layers (FCL) of CNN is replaced by WNN. Thirdly, comparative simulations based on MNIST and CIFAR-10 datasets among the discussed methods of BPNN, RBFNN, CNN and CWNN are implemented and analyzed. Fourthly, the wavelet-based Convolutional Neural Network (WCNN) is proposed, where the wavelet transformation is adopted as the activation function in Convolutional Pool Neural Network (CPNN) of CNN. Fifthly, simulations based on CWNN are implemented and analyzed on the MNIST dataset. Effects are as follows: Firstly, WNN can solve the problems of BPNN and RBFNN and have better performance. Secondly, the proposed CWNN can reduce the mean square error and the error rate of CNN, which means CWNN has better maximum precision than CNN. Thirdly, the proposed WCNN can reduce the mean square error and the error rate of CWNN, which means WCNN has better maximum precision than CWNN.

Single upper limb functional movements decoding from motor imagery EEG signals using wavelet neural network

In this study, we propose a wavelet neural network (WNN) to improve the performance of movements decoding from motor imagery (MI) EEG signals. Fifteen healthy subjects performed six functional movements of single upper limb: forearm pronation/supination, hand open/close, and elbow flexion/extension. The wavelet packet decomposition (WPD) was used to decompose MI EEG into sub-bands, and the statistical features were extracted from the sub-bands to code the functional movements. Subsequently, the principal component analysis (PCA) was used to select the best feature vector. The factors of mother wavelet, multidimensional wavelet function, number of channels, and the time of imaging segment were optimized through experiments. As a result, the best accuracy of 86.27 ± 6.98% was achieved with the optimized coif1 mother wavelet with a six-level decomposition, Mexican Hat wavelet as a hyperparameter, the time of imaging segment set as 0.7 s, and 61 channels data. In addition, to further verify the efficiency of the WNN classifier, the comparative experiments were conducted with the support vector machine (SVM), linear discriminant analysis (LDA), k-nearest neighbor (KNN), and single hidden layer feedforward neural network (ANN). The results show that the accuracy of WNN is improved by about 15 ~ 40%, which proved the effectiveness of the WNN method. This study provides a new way to decode from motor imagery EEG in multi-movement classification.

Joint angle estimation with wavelet neural networks

This paper presents a wavelet neural network (WNN) based method to reduce reliance on wearable kinematic sensors in gait analysis. Wearable kinematic sensors hinder real-time outdoor gait monitoring applications due to drawbacks caused by multiple sensor placements and sensor offset errors. The proposed WNN method uses vertical Ground Reaction Forces (vGRFs) measured from foot kinetic sensors as inputs to estimate ankle, knee, and hip joint angles. Salient vGRF inputs are extracted from primary gait event intervals. These selected gait inputs facilitate future integration with smart insoles for real-time outdoor gait studies. The proposed concept potentially reduces the number of body-mounted kinematics sensors used in gait analysis applications, hence leading to a simplified sensor placement and control circuitry without deteriorating the overall performance.

A novel approach to estimate reservoir permeability using machine learning method

Reservoir permeability in upstream of petroleum engineering plays an essential role in crude oil production. Due to the high cost and difficulty in direct measurement of the permeability, having a robust model of this parameter based on the openhole logs and data is preferred. The sonic, volumetric density, gamma ray, total porosity, neutron porosity logs are available in the time of logging and have the highest correlation with reservoir permeability characteristics. To estimate the permeability of the reservoir based on these available data, a new intelligent method of Genetic Algorithm (GA) and Wavelet Neural Network (WNN) is derived. In the developed model, a new objective function has been introduced. For avoiding more complexity of the objective function, the initializing weights of neural network has been done by GA. Then, the training levenberg marquardt algorithm is utilized to update the optimal weighting. In other words, wavelet as activation function of neural network enhances exploitation search abilities of the algorithm and leads to a robust model. In the following, a sample reservoir as a source of data in this field is selected to evaluate the effectiveness of the proposed algorithm in the permeability estimation. For the sake of comparison, two algorithms of BP-ANN and GA-BP, which have been already presented in the literature, are applied for the same data sets and the superiority of developed model in estimation has been illustrated.

A New Hybrid Wavelet-Neural Network Approach for Forecasting Electricity

This study investigates the performance of a novel neural network technique in the problem of price forecasting. To improve the prediction accuracy using each model’s unique features, this research proposes a hybrid approach that combines the -factor GARMA process, empirical wavelet transform and the local linear wavelet neural network (LLWNN) methods, to form the GARMA-WLLWNN process. In order to verify the validity of the model and the algorithm, the performance of the proposed model is evaluated using data from Polish electricity markets, and it is compared with the dual generalized long memory -factor GARMA-G-GARCH model and the individual WLLWNN. The empirical results demonstrated the proposed hybrid model can achieve a better predicting performance and prove that is the most suitable electricity market forecasting technique.

Delineation of Alteration Zones Based on Wavelet Neural Network (WNN) and Concentration–Volume (C-V) Fractal Methods in the Hypogene Zone of Porphyry Copper Deposit, Shahr-e-Babak District, SE Iran

In this paper, we aim to achieve two specific objectives. The first one is to examine the applicability of wavelet neural network (WNN) technique in ore grade estimation, which is based on integration between wavelet theory and Artificial Neural Network (ANN). Different wavelets are applied as activation functions to estimate Cu grade of borehole data in the hypogene zone of porphyry ore deposit, Shahr-e-Babak district, SE Iran. WNN parameters such as dilation and translation are fixed and only the weights of the network are optimized during its learning process. The efficacy of this type of network in function learning and estimation is compared with Ordinary Kriging (OK). Secondly, we aim to delineate the potassic and phyllic alteration regions in the hypogene zone of Cu porphyry deposit based on the estimation obtained of WNN and OK methods, and utilize Concentration–Volume (C–V) fractal model. In this regard, at first C–V log–log plots are generated based on the results of OK and WNN. The plots then are used to determine the Cu threshold values of the alteration zones. To investigate the correlation between geological model and C-V fractal results, the log ratio matrix is applied. The results showed that, Cu values less than 1.1% from WNN have more overlapped voxels with phyllic alteration zone by overall accuracy (OA) of 0.74. Spatial correlation between the potassic alteration zones resulted from 3D geological modeling and high concentration zones in C-V fractal model showed that the alteration zone has Cu values between 1.1% and 2.2% with OA of 0.72 and finally have an appropriate overlap with Cu values greater than 2.2% with OA of 0.7. Generally, the results showed that the WNN (Morlet activation function) with OA greater than OK can be can be a suitable and robust tool for quantitative modeling of alteration zones, instead of qualitative methods.

Study on soft sensor modeling method for sign of contaminated fermentation broth in Chlortetracycline fermentation process

During Chlortetracycline fermentation, contamination of fermentation broth by non-target bacteria is an unavoidable problem. There is no online analytical instrument to determine whether the fermentation broth has been contaminated. Only the results of manual sampling analysis can be used to determine whether the fermentation broth is contaminated. This analysis process usually takes several hours. In order to predict online whether the fermentation broth is contaminated by non-target bacteria, a soft sensor modeling method for the signs of contamination in Chlortetracycline fermentation broth was proposed in this paper. Based on recursive wavelet neural network (RWNN) and Gaussian process regression (GPR) method, the soft sensor model of online measurable parameters and total sugar content of fermentation broth was established. By deeply analyzing the correlation between the total sugar content (it is a parameter that is difficult to measure online) of fermentation broth and the signs of bacterial contamination during fermentation, a soft sensor model was established combining with the correlation between the total sugar content of fermentation broth and the symptoms of bacterial infection, and the symptoms of non-target bacterial infection of fermentation broth were predicted. Based on the field data of the fermentation process, the different signs of Chlortetracycline fermentation were predicted for the fermentation broth uninfected with non-target bacteria, infected with bacilli and infected with phages. The experimental results showed that the proposed soft sensor model could be used to predict the occurrence of contamination during Chlortetracycline fermentation. Based on the field data, the validity of the modeling method is verified. The proposed soft sensor model of signs of bacterial contamination can be used to predict the occurrence of bacterial contamination in Chlortetracycline, Penicillin and related biological fermentation processes. So that the site operators can take effective measures in time to reduce production losses to a minimum.

OP Performance Prediction for Complex Mobile Multiuser Networks Based on Extreme Learning Machine

Due to the complex and variable environments of mobile communication, the mobile multiuser networks become a hot topic. To process active complex event in mobile multiuser networks, it is important to predict the system performance. In this work, the authors consider the multiuser networks which utilizes transmit antenna selection (TAS). We derive novel closed-form expressions for the outage probability (OP) in terms of the Meijer's G-function. Then, a extreme learning machine (ELM)-based OP performance prediction algorithm is proposed. We use the theoretical results to generate training data. We test back-propagation (BP) neural network, locally weighted linear regression (LWLR), wavelet neural network, ELM, and support vector machine (SVM) methods. Compared with wavelet neural network, SVM, BP neural network, and LWLR methods, the Monte-Carlo results shows that the proposed prediction algorithm can consistently achieve higher OP performance prediction results.

Fault diagnosis and location method for electrical power supply and distribution of buildings

This paper presents a new fault diagnosis and location method for electrical power supply and distribution of buildings using Bayesian and wavelet neural network (WNN). Aiming at the complex trunk-type power supply and distribution structure of buildings, wavelet transform (WT) is adopted to process the current, voltage and phase data of each branch to extract the features that can distinguish faults effectively. And the fault diagnosis model based on Bayesian network is established by the above features. In order to improve the accuracy of WNN in fault location in buildings, a WNN method optimised by dragonfly algorithm (DA) is proposed to obtain better thresholds and weights, which are utilised to enhance the prediction ability. A simulation study was made with MATLAB/Simulink to verify the performance of the proposed method on a power supply and distribution model.

Single phase fault diagnosis and location in active distribution network using synchronized voltage measurement

To ensure the safe operation of the active distribution network (ADN), accurate fault diagnosis and location are crucial to improve the reliability indices and reduce the outage time. This paper proposes a characteristic model-based method for the single phase to earth fault in the ADN system. Firstly, the characteristic model of the fault factor extracts the phasor distribution characteristics of the voltage and current along the distribution feeder line and estimates the current contribution of DG units to the fault point. Based on the minimum entropy theory, the solution of nonlinear characteristic model is transformed to a single-objective optimization of the characteristic entropy and the diagnosis criteria is formulated. Then, the two-stage fault location scheme for single phase fault is proposed. The fault diagnosis stage estimates the suspicious fault section to reduce the search area and the fault location stage locates the exact fault distance. The Fibonacci search algorithm is utilized for fault location to the optimize iterative and minimize the respond time. The proposed scheme is general for all DG types and overcomes the requirement of its individual model parameters. The proposed method is validated in the IEEE 34-bus distribution test system using the phasor measurement unit (PMU). Test results of the model-based diagnosis and location method can reveal accurate fault location and rapid response at different fault impedance and small time delay comparing with the radial basis function neural network (RBF), and wavelet neural network (WNN) method.

An adaptive hybrid model for short-term urban traffic flow prediction

With the rapid increase in car ownership, urban transport systems are challenged by the overwhelming traffic demand and congestion. Dynamic prediction of traffic flows is of considerable significance for congestion mitigation and demand management. Real-time and precise prediction models are capable of analyzing traffic flow characteristics, predicting traffic flow trends, and motivating reasonable inductive actions. Considering the periodicity and variability of traffic flow and limitations of single prediction models, an adaptive hybrid model for predicting short-term traffic flow was proposed in this study. Firstly, the linear Autoregressive Integrated Moving Average (ARIMA) method and non-linear Wavelet Neural Network (WNN) method were used to predict traffic flow. Then, outputs of the two individual models were analyzed and combined by fuzzy logic and the weighted result was regarded as the final predicted traffic volume of the hybrid model. The results indicate that the hybrid model can offer better performance in predicting short-term traffic flow than the two single models either in stable or in fluctuating conditions. The relative error is within ±10%, showing that the proposed hybrid model is both accurate and reliable.

A water cycle optimized wavelet neural network algorithm for demand prediction in cloud computing

Nowadays, using cloud computing services is becoming very popular among corporations and dependent users because of its flexibility and diverse facilities. So, demand for its infrastructures is increasing exponentially and with this huge growth of demands, it is getting more challenging for cloud providers to serve all the user requests in a way that quality of service is high enough and service level agreement is also met. Along years, researchers found out that for an optimized resource allocation which provides required user quality of service it is needed to know the amount of future workload in advance so resources can be prepared. There are many methods introduced for cloud workload prediction but excessive changes in the number of requests in cloud workloads cause a reduction in their prediction accuracy. To solve this problem, we have proposed a novel hybrid wavelet neural network method which can efficiently model excessive changes in workload and predict the upcoming requests. For training the proposed method accurately, we have used two heuristic algorithms, Artificial Immune System and Water Cycle Algorithm. The two mentioned heuristic algorithms are used for finding optimized parameters (such as bias and weight) for the wavelet neural network algorithm. The proposed wavelet neural network trained with heuristic algorithms is tested on real and standard cloud workloads. At last, the accuracy of the proposed method is thoroughly surveyed statistically and also compared with other state-of-the-art prediction methods. Simulation results show that our method improves MAPE error at least 10% in comparison to other rival prediction methods.