DAMAGE CLASSIFICATION BY PROBABILISTIC NEURAL NETWORKS BASED ON LATENT COMPONENTS FOR TIME-VARYING SYSTEM

A new approach to damage classification for health monitoring of a time-varylng system is presented. The functional-series time-dependent auto regressive moving average （FS-TARMA） time series model is applied to the vibration signal observed in the time-varying system for estimating the TAR/TMA parameters and the innovation variance. These parameters are the functions of the time, represented by a group of projection coefficients on the certain functional subspace with specific basis functions. The estimated TAR/TMA parameters and the innovation variance are further used to calculate the latent components （LCs） as the more informative data for health monitoring evaluation, based on an eigenvalue decomposition technique. LCs are then combined and reduced to numerical values （NVs） as feature sets, which are input to a probabilistic neural network （PNN） for the damage classification. For the evaluation of the proposed method, numerical simulations of the damage classification for a tlme-varylng system are used, in which different classes of damage are modeled by the mass or stiffness reductions. It is demonstrated that the method can identify the damages in the course of operation and the change of parameters on the time-varying background of the system.

Computer vision-based limestone rock-type classification using probabilistic neural network

Proper quality planning of limestone raw materials is an essential job of maintaining desired feed in cement plant. Rock-type identification is an integrated part of quality planning for limestone mine. In this paper, a computer vision-based rock-type classification algorithm is proposed for fast and reliable identification without human intervention. A laboratory scale vision-based model was developed using probabilistic neural network（PNN） where color histogram features are used as input. The color image histogram-based features that include weighted mean, skewness and kurtosis features are extracted for all three color space red, green, and blue. A total nine features are used as input for the PNN classification model. The smoothing parameter for PNN model is selected judicially to develop an optimal or close to the optimum classification model. The developed PPN is validated using the test data set and results reveal that the proposed vision-based model can perform satisfactorily for classifying limestone rocktypes. Overall the error of mis-classification is below 6%. When compared with other three classification algorithms, it is observed that the proposed method performs substantially better than all three classification algorithms.

Remote Sensing Image Segmentation with Probabilistic Neural Networks

This paper focuses on the image segmentation with probabilistic neural networks (PNNs). Back propagation neural networks (BpNNs) and multi perceptron neural networks (MLPs) are also considered in this study. Especially, this paper investigates the implementation of PNNs in image segmentation and optimal processing of image segmentation with a PNN. The comparison between image segmentations with PNNs and with other neural networks is given. The experimental results show that PNNs can be successfully applied to image segmentation for good results.

EEG classification based on probabilistic neural network with supervised learning in brain computer interface

Aiming at the topic of electroencephalogram （EEG） pattern recognition in brain computer interface （BCI）, a classification method based on probabilistic neural network （PNN） with supervised learning is presented in this paper. It applies the recognition rate of training samples to the learning progress of network parameters. The learning vector quantization is employed to group training samples and the Genetic algorithm （GA） is used for training the network＇ s smoothing parameters and hidden central vector for detemlining hidden neurons. Utilizing the standard dataset I （a） of BCI Competition 2003 and comparing with other classification methods, the experiment results show that the best performance of pattern recognition Js got in this way, and the classification accuracy can reach to 93.8%, which improves over 5% compared with the best result （88.7 % ） of the competition. This technology provides an effective way to EEG classification in practical system of BCI.

Estimation of reservoir porosity using probabilistic neural network and seismic attributes

Porosity is one of the most important properties of oil and gas reservoirs. The porosity data that come from well log are only available at well points. It is necessary to use other method to estimate reservoir porosity.Seismic data contain abundant lithological information. Because there are inherent correlations between reservoir property and seismic data,it is possible to estimate reservoir porosity by using seismic data and attributes.Probabilistic neural network is a powerful tool to extract mathematical relation between two data sets. It has been used to extract the mathematical relation between porosity and seismic attributes. Firstly,a seismic impedance volume is calculated by seismic inversion. Secondly,several appropriate seismic attributes are extracted by using multi-regression analysis. Then a probabilistic neural network model is trained to obtain a mathematical relation between porosity and seismic attributes. Finally,this trained probabilistic neural network model is implemented to calculate a porosity data volume. This methodology could be utilized to find advantageous areas at the early stage of exploration. It is also helpful for the establishment of a reservoir model at the stage of reservoir development.

An Advanced Probabilistic Neural Network for the Design of Breakwater Armor Blocks

In this study, an advanced probabilistic neural network （APNN） method is proposed to reflect the global probability density function （PDF） by summing up the heterogeneous local PDF which is automatically determined in the individual standard deviation of variables. The APNN is applied to predict the stability number of armor blocks of breakwaters using the experimental data of＇ van der Meet, and the estimated results of the APNN are compared with those of an empirical formula and a previous artificial neural network （ANN） model. The APNN shows better results in predicting the stability number of armor bilks of breakwater and it provided the promising probabilistic viewpoints by using the individual standard deviation in a variable.

Passenger Flow Status Evaluation in Subway Station Based on Probabilistic Neural Network

This paper select the escalator with large flow in the station as the object, analysing the correlation of the AFC data of the in and out gates and the passenger flow parameters by passenger flow density and the passing time acquired and calculated in the waiting area of the prediction escalator to select the gates related to the predicted the escalator. NARX neural network is used to predict the model of the passenger flow parameters of the escalator waiting area based on the related gates＇ AFC data, then a probabilistic neural network model was established by using the AFC data and predicted passenger flow parameters as input and the passenger flow status in the escalator waiting area of subway station as output.The result shows the predicting model can predict the passenger flow status of the escalator waiting area better by the AFC data in the subway station. Research result can provide decision basis for the operation management of the subway station.

Experiment Verification of Damage Detection for Offshore Platforms by Neural Networks

In the present work, damage detection for offshore platforms is divided into three steps. Firstly, the located direction of the damaged member is detemfined by the pmbabilistic neural network with input of the change rate of normalized medal frequency. Secondly, the profile and layer of the damaged member is also determined by the pmbabilistic neural network with input of the normalized damage-signal index. Finally, the damage extent is determined by the back propagation neural networks with input of the squared change rate of modal frequency. So the size of the network and the training time can be reduced greatly. All these networks are trained with simulated data obtained from the finite element model of an experiment model. Then these trained neural networks are examined with data obtained from impulse tests on the experiment model. The experiment results show that the trained neural networks are able to detect the damaged member with reasonable accuracy.

Fault prediction method for nuclear power machinery based on Bayesian PPCA recurrent neural network model

Early fault warning for nuclear power machinery is conducive to timely troubleshooting and reductions in safety risks and unnecessary costs. This paper presents a novel intelligent fault prediction method, integrated probabilistic principal component analysis(PPCA), multi-resolution wavelet analysis, Bayesian inference, and RNN model for nuclear power machinery that consider data uncertainty and chaotic time series. After denoising the source data, the Bayesian PPCA method is employed for dimensional reduction to obtain a refined data group. A recurrent neural network(RNN) prediction model is constructed, and a Bayesian statistical inference approach is developed to quantitatively assess the prediction reliability of the model. By modeling and analyzing the data collected on the steam turbine and components of a nuclear power plant, the results of the goodness of fit, mean square error distribution, and Bayesian confidence indicate that the proposed RNN model can implement early warning in the fault creep period. The accuracy and reliability of the proposed model are quantitatively verified.

Using Neural Networks to Predict Secondary Structure for Protein Folding

Protein Secondary Structure Prediction (PSSP) is considered as one of the major challenging tasks in bioinformatics, so many solutions have been proposed to solve that problem via trying to achieve more accurate prediction results. The goal of this paper is to develop and implement an intelligent based system to predict secondary structure of a protein from its primary amino acid sequence by using five models of Neural Network (NN). These models are Feed Forward Neural Network (FNN), Learning Vector Quantization (LVQ), Probabilistic Neural Network (PNN), Convolutional Neural Network (CNN), and CNN Fine Tuning for PSSP. To evaluate our approaches two datasets have been used. The first one contains 114 protein samples, and the second one contains 1845 protein samples.

Forecasting and optimal probabilistic scheduling of surplus gas systems in iron and steel industry

To make full use of the gas resource, stabilize the pipe network pressure, and obtain higher economic benefits in the iron and steel industry, the surplus gas prediction and scheduling models were proposed. Before applying the forecasting techniques, a support vector classifier was first used to classify the data, and then the filtering was used to create separate trend and volatility sequences. After forecasting, the Markov chain transition probability matrix was introduced to adjust the residual. Simulation results using surplus gas data from an iron and steel enterprise demonstrate that the constructed SVC-HP-ENN-LSSVM-MC prediction model prediction is accurate, and that the classification accuracy is high under different conditions. Based on this, the scheduling model was constructed for surplus gas operating, and it has been used to investigate the comprehensive measures for managing the operational probabilistic risk and optimize the economic benefit at various working conditions and implementations. It has extended the concepts of traditional surplus gas dispatching systems, and provides a method for enterprises to determine optimal schedules.

Some Features of Neural Networks as Nonlinearly Parameterized Models of Unknown Systems Using an Online Learning Algorithm

This paper deals with deriving the properties of updated neural network model that is exploited to identify an unknown nonlinear system via the standard gradient learning algorithm. The convergence of this algorithm for online training the three-layer neural networks in stochastic environment is studied. A special case where an unknown nonlinearity can exactly be approximated by some neural network with a nonlinear activation function for its output layer is considered. To analyze the asymptotic behavior of the learning processes, the so-called Lyapunov-like approach is utilized. As the Lyapunov function, the expected value of the square of approximation error depending on network parameters is chosen. Within this approach, sufficient conditions guaranteeing the convergence of learning algorithm with probability 1 are derived. Simulation results are presented to support the theoretical analysis.

Uplifting the complexity of analysis for probabilistic security of electricity supply assessments using artificial neural networks

The energy sector faces rapid decarbonisation and decision-makers demand reliable assessments of the security of electricity supply. For this, detailed simulation models with a high temporal and technological resolution are required. When confronted with increasing weather-dependent renewable energy generation, probabilistic simulation models have proven. The significant computational costs of calculating a scenario, however, limit the complexity of further analysis. Advances in code optimization as well as the use of computing clusters still lead to runtimes of up to eight hours per scenario. However ongoing research highlights that tailor-made approximations are potentially the key factor in further reducing computing time. Consequently, current research aims to provide a method for the rapid prediction of widely varying scenarios. In this work artificial neural networks (ANN) are trained and compared to approximate the system behavior of the probabilistic simulation model. To do so, information needs to be sampled from the probabilistic simulation in an efficient way. Because only a limited space in the whole design space of the 16 independent variables is of interest, a classification is developed. Finally it required only around 35 min to create the regression models, including sampling the design space, simulating the training data and training the ANNs. The resulting ANNs are able to predict all scenarios within the validity range of the regression model with a coefficient of determination of over 0.9998 for independent test data (1.051.200 data points). They need only a few milliseconds to predict one scenario, enabling in-depth analysis in a brief period of time.

基于贝叶斯图注意力Transformer的航空发动机剩余使用寿命概率预测

航空发动机作为飞机的心脏,其健康状态对飞机的安全飞行至关重要.深度学习强大的数据挖掘能力,为通过海量历史数据预测航空发动机的剩余使用寿命提供了新方法.然而,传统基于深度学习的方法大都关注于挖掘数据在时间上的关联,而忽略了多个传感器监测数据之间复杂的非欧氏空间关系.此外,少有研究考虑数据或者预测过程本身具有的不确定性,缺乏对预测结果可靠性的评估.为解决上述问题,本文提出了一种基于贝叶斯网络和图注意力Transformer的航空发动机剩余使用寿命概率预测方法.将图注意力机制融入Transformer的时间多头注意力模块,结合图注意力网络在空间特征提取上的优势和Transformer模型在时间特征提取的优势,实现数据特征时空关系的联合提取.同时,利用改进的贝叶斯网络度量预测不确定性,在得到剩余使用寿命预测点值的同时给出相应的置信区间.最后,通过在公开航空发动机数据集上的实验,证明了所提模型的有效性和先进性.

Probabilistic Residential Load Forecasting with Sequence-to-sequence Adversarial Domain Adaptation Networks

Lately,the power demand of consumers is increasing in distribution networks,while renewable power generation keeps penetrating into the distribution networks.Insufficient data make it hard to accurately predict the new residential load or newly built apartments with volatile and changing time-series characteristics in terms of frequency and magnitude.Hence,this paper proposes a short-term probabilistic residential load forecasting scheme based on transfer learning and deep learning techniques.First,we formulate the short-term probabilistic residential load forecasting problem.Then,we propose a sequence-to-sequence(Seq2Seq)adversarial domain adaptation network and its joint training strategy to transfer generic features from the source domain(with massive consumption records of regular loads)to the target domain(with limited observations of new residential loads)and simultaneously minimize the domain difference and forecasting errors when solving the forecasting problem.For implementation,the dominant techniques or elements are used as the submodules of the Seq2Seq adversarial domain adaptation network,including the Seq2Seq recurrent neural networks(RNNs)composed of a long short-term memory(LSTM)encoder and an LSTM decoder,and quantile loss.Finally,this study conducts the case studies via multiple evaluation indices,comparative methods of classic machine learning and advanced deep learning,and various available data of the new residentical loads and other regular loads.The experimental results validate the effectiveness and stability of the proposed scheme.

Embedding based quantile regression neural network for probabilistic load forecasting

Compared to traditional point load forecasting,probabilistic load forecasting(PLF) has great significance in advanced system scheduling and planning with higher reliability. Medium term probabilistic load forecasting with a resolution to an hour has turned out to be practical especially in medium term energy trading and can enhance the performance of forecasting compared to those only utilizing daily information. Two main uncertainties exist when PLF is implemented: the first is the temperature fluctuation at the same time of each year; the second is the load variation which means that even if observed indicators are fixed since other observed external indicators can be responsible for the variation. Therefore, we propose a hybrid model considering both temperature uncertainty and load variation to generate medium term probabilistic forecasting with hourly resolution. An innovative quantile regression neural network with parameter embedding is established to capture the load variation, and a temperature scenario based technique is utilized to generate temperatureforecasting in a probabilistic manner. It turns out that the proposed method overrides commonly used benchmark models in the case study.

Neural decoding based on probabilistic neural network

Brain-machine interface (BMI) has been developed due to its possibility to cure severe body paralysis. This technology has been used to realize the direct control of prosthetic devices,such as robot arms,computer cursors,and paralyzed muscles. A variety of neural decoding algorithms have been designed to explore relationships between neural activities and movements of the limbs. In this paper,two novel neural decoding methods based on probabilistic neural network (PNN) in rats were introduced,the PNN decoder and the modified PNN (MPNN) decoder. In the ex-periment,rats were trained to obtain water by pressing a lever over a pressure threshold. Microelectrode array was implanted in the motor cortex to record neural activity,and pressure was recorded by a pressure sensor synchronously. After training,the pressure values were estimated from the neural signals by PNN and MPNN decoders. Their per-formances were evaluated by a correlation coefficient (CC) and a mean square error (MSE). The results show that the MPNN decoder,with a CC of 0.8657 and an MSE of 0.2563,outperformed the traditionally-used Wiener filter (WF) and Kalman filter (KF) decoders. It was also observed that the discretization level did not affect the MPNN performance,indicating that the MPNN decoder can handle different tasks in BMI system,including the detection of movement states and estimation of continuous kinematic parameters.

Prediction of chilling damage risk in maize growth period based on probabilistic neural network approach

Low temperature chilling damage is one of the most serious disasters in maize production,which is a typical non-linear complex issue with numerous influencing factors and strong uncertainty.How to predict it is not only a hot theoretical research topic,but also an urgent practical problem to be solved.However,most of the current researches are focusing on post-disaster static descriptive assessment rather than pre-disaster dynamic predictive analysis,resulting in the problems such as no indicative result and low accuracy.In this study,the satisfaction rate of environmental accumulated temperature for maize production was used to measure the chilling damage risk,and a model for maize chilling damage risk prediction based on probabilistic neural network was constructed.The model was composed of input layer,pattern layer,summation layer and output layer.The obtained results showed that the prediction accuracy for the most serious risk level was as high as 0.91,and the rates of the Type I Error and Type II Error made by the model were 0.1 and 0.09,respectively.This indicated that the model employed was promising with good performance.The results of this research are of both theoretical significance for providing a new reference method of pre-disaster prediction to study maize chilling disaster risk and practical significance for reducing maize production risk and ensuring yield safety.

Predicting the shrinkage of thermal insulation mortar by probabilistic neural networks

This study explored the potential of using probabilistic neural networks (PNN) to predict shrinkage of thermal insulation mortar.Probabilistic results were obtained from the PNN model with the aid of Parzen non-parametric estimator of the probability density functions (PDF).Five variables,water-cementitious materials ratio,content of cement,fly ash,aggregate and plasticizer,were employed for input variables,while a category of 56-d shrinkage of mortar was used for the output variable.A total of 192 groups of experimental data from 64 mixtures designed using JMP7.0 software were collected,of which 120 groups of data were used for training the model and the other 72 groups of data for testing.The simulation results showed that the PNN model with an optimal smoothing parameter determined by the curves of the mean square error (MSE) and the number of unrecognized probability densities (UPDs) exhibited a promising capability of predicting shrinkage of mortar.