Soft sensor of chemical processes with large numbers of input parameters using auto-associative hierarchical neural network

To explore the problems of monitoring chemical processes with large numbers of input parameters, a method based on Auto-associative Hierarchical Neural Network(AHNN) is proposed. AHNN focuses on dealing with datasets in high-dimension. AHNNs consist of two parts: groups of subnets based on well trained Autoassociative Neural Networks(AANNs) and a main net. The subnets play an important role on the performance of AHNN. A simple but effective method of designing the subnets is developed in this paper. In this method,the subnets are designed according to the classification of the data attributes. For getting the classification, an effective method called Extension Data Attributes Classification(EDAC) is adopted. Soft sensor using AHNN based on EDAC(EDAC-AHNN) is introduced. As a case study, the production data of Purified Terephthalic Acid(PTA) solvent system are selected to examine the proposed model. The results of the EDAC-AHNN model are compared with the experimental data extracted from the literature, which shows the efficiency of the proposed model.

4-CBA Soft Sensor Based on Fuzzy CMAC Neural Networks

Soft sensor is attractive in dealing with online product quality measurement by virtue of other easily measured variables. In AMOCO PTA (purified terephthalic acid) production process, the unavailability of real-time measurement of 4-CBA makes it impossible for timely adjustment and thereby influences the product quality and the plant economy benefit. In this paper, a kind of FCMAC (fuzzy cerebellar model articulation controller) method is presented to solve the online measurement problem. Different from the conventional CMAC (cerebellar model articulation controller) networks, which has inferior smoothing ability because of its table look-up based technology. Integrating fuzzy model into CMAC networks, it becomes more accurate in functional mapping without weakening its generalization ability. Numerical example and industrial application results show the method proposed here is satisfactory and feasible.

Event-based nonfragile state estimation for memristive recurrent neural networks with stochastic cyber-attacks and sensor saturations

This paper addresses the issue of nonfragile state estimation for memristive recurrent neural networks with proportional delay and sensor saturations. In practical engineering, numerous unnecessary signals are transmitted to the estimator through the networks, which increases the burden of communication bandwidth. A dynamic event-triggered mechanism,instead of a static event-triggered mechanism, is employed to select useful data. By constructing a meaningful Lyapunov–Krasovskii functional, a delay-dependent criterion is derived in terms of linear matrix inequalities for ensuring the global asymptotic stability of the augmented system. In the end, two numerical simulations are employed to illustrate the feasibility and validity of the proposed theoretical results.

A transfer learning enhanced physics-informed neural network for parameter identification in soft materials

Soft materials,with the sensitivity to various external stimuli,exhibit high flexibility and stretchability.Accurate prediction of their mechanical behaviors requires advanced hyperelastic constitutive models incorporating multiple parameters.However,identifying multiple parameters under complex deformations remains a challenge,especially with limited observed data.In this study,we develop a physics-informed neural network(PINN)framework to identify material parameters and predict mechanical fields,focusing on compressible Neo-Hookean materials and hydrogels.To improve accuracy,we utilize scaling techniques to normalize network outputs and material parameters.This framework effectively solves forward and inverse problems,extrapolating continuous mechanical fields from sparse boundary data and identifying unknown mechanical properties.We explore different approaches for imposing boundary conditions(BCs)to assess their impacts on accuracy.To enhance efficiency and generalization,we propose a transfer learning enhanced PINN(TL-PINN),allowing pre-trained networks to quickly adapt to new scenarios.The TL-PINN significantly reduces computational costs while maintaining accuracy.This work holds promise in addressing practical challenges in soft material science,and provides insights into soft material mechanics with state-of-the-art experimental methods.

Component Content Soft-sensor Based on Neural Networks in Rare-earth Countercurrent Extraction Process

Throught fusion of the mechanism modeling and the neural networks modeling,a compo- nent content soft-sensor,which is composed of the equilibrium calculation model for multi-component rare earth extraction and the error compensation model of fuzzy system,is proposed to solve the prob- lem that the component content in countercurrent rare-earth extraction process is hardly measured on-line.An industry experiment in the extraction Y process by HAB using this hybrid soft-sensor proves its effectiveness.

Numeral eddy current sensor modelling based on genetic neural network

This paper presents a method used to the numeral eddy current sensor modelling based on the genetic neural network to settle its nonlinear problem. The principle and algorithms of genetic neural network are introduced. In this method, the nonlinear model parameters of the numeral eddy current sensor are optimized by genetic neural network （GNN） according to measurement data. So the method remains both the global searching ability of genetic algorithm and the good local searching ability of neural network. The nonlinear model has the advantages of strong robustness, on-line modelling and high precision. The maximum nonlinearity error can be reduced to 0.037% by using GNN. However, the maximum nonlinearity error is 0.075% using the least square method.

A New Modeling Method Based on Genetic Neural Network for Numeral Eddy Current Sensor

In this paper,we present a method used to the numeral eddy current sensor modeling based on genetic neural network to settle its nonlinear problem.The principle and algorithms of genetic neural network are introduced.In this method, the nonlinear model parameters of the numeral eddy current sensor are optimized by genetic neural network (GNN) according to measurement data.So the method remains both the global searching ability of genetic algorithm and the good local searching ability of neural network.The nonlinear model has the advantages of strong robustness,on-line scaling and high precision.The maximum nonlinearity error can be reduced to 0.037% using GNN.However,the maximum nonlinearity error is 0.075% using least square method (LMS).

Calibration Method Based on RBF Neural Networks for Soil Moisture Content Sensor

Temporal and spatial variation of soil moisture content is significant for crop growth,climate change and the other fields.In order to overcome shortage of non-linear output voltage of TDR3 soil moisture content sensor and increase soil moisture content data collection and computational efficiency,this paper presents a RBF neural network calibration method of soil moisture content based on TDR3 soil moisture sensor and wireless sensor networks.Experiment results show that the calibration method is effective...

Soft sensor for ratio of soda to aluminate based on PCA-RBF multiple network

Based on principal component analysis, a multiple neural network was proposed. The principal component analysis was firstly used to reorganize the input variables and eliminate the correlativity. Then the reorganized variables were divided into 2 groups according to the original information and 2 corresponding neural networks were established. A radial basis function network was used to depict the relationship between the output variables and the first group input variables which contain main original information. An other single-layer neural network model was used to compensate the error between the output of radial basis function network and the actual output variables. At last, The multiple network was used as soft sensor for the ratio of soda to aluminate in the process of high-pressure digestion of alumina. Simulation of industry application data shows that the prediction error of the model is less than 3%, and the model has good generalization ability.

Sensor Fault Diagnosis and Reconstruction of Engine Control System Based on Autoassociative Neural Network

The topology and property of Autoassociative Neural Networks(AANN) and theAANN's application to sensor fault diagnosis and reconstruction of engine control system arestudied. The key feature of AANN is feature extract and noise filtering. Sensor fault detection isaccomplished by integrating the optimal estimation and fault detection logic. Digital simulationshows that the scheme can detect hard and soft failures of sensors at the absence of models forengines which have performance deteriorate in the service life, and can provide good analyticalredundancy.

Soft measurement model of ring's dimensions for vertical hot ring rolling process using neural networks optimized by genetic algorithm

Vertical hot ring rolling(VHRR) process has the characteristics of nonlinearity,time-variation and being susceptible to disturbance.Furthermore,the ring's growth is quite fast within a short time,and the rolled ring's position is asymmetrical.All of these cause that the ring's dimensions cannot be measured directly.Through analyzing the relationships among the dimensions of ring blanks,the positions of rolls and the ring's inner and outer diameter,the soft measurement model of ring's dimensions is established based on the radial basis function neural network(RBFNN).A mass of data samples are obtained from VHRR finite element(FE) simulations to train and test the soft measurement NN model,and the model's structure parameters are deduced and optimized by genetic algorithm(GA).Finally,the soft measurement system of ring's dimensions is established and validated by the VHRR experiments.The ring's dimensions were measured artificially and calculated by the soft measurement NN model.The results show that the calculation values of GA-RBFNN model are close to the artificial measurement data.In addition,the calculation accuracy of GA-RBFNN model is higher than that of RBFNN model.The research results suggest that the soft measurement NN model has high precision and flexibility.The research can provide practical methods and theoretical guidance for the accurate measurement of VHRR process.

Wireless Sensor Networks Routing Attacks Prevention with Blockchain and Deep Neural Network

Routing is a key function inWireless Sensor Networks(WSNs)since it facilitates data transfer to base stations.Routing attacks have the potential to destroy and degrade the functionality ofWSNs.A trustworthy routing system is essential for routing security andWSN efficiency.Numerous methods have been implemented to build trust between routing nodes,including the use of cryptographic methods and centralized routing.Nonetheless,the majority of routing techniques are unworkable in reality due to the difficulty of properly identifying untrusted routing node activities.At the moment,there is no effective way to avoid malicious node attacks.As a consequence of these concerns,this paper proposes a trusted routing technique that combines blockchain infrastructure,deep neural networks,and Markov Decision Processes(MDPs)to improve the security and efficiency of WSN routing.To authenticate the transmission process,the suggested methodology makes use of a Proof of Authority(PoA)mechanism inside the blockchain network.The validation group required for proofing is chosen using a deep learning approach that prioritizes each node’s characteristics.MDPs are then utilized to determine the suitable next-hop as a forwarding node capable of securely transmitting messages.According to testing data,our routing system outperforms current routing algorithms in a 50%malicious node routing scenario.

Forest Fire Detection Using Artificial Neural Network Algorithm Implemented in Wireless Sensor Networks

A forest fire is a severe threat to forest resources and human life, In this paper, we propose a forest-fire detection system that has an artificial neural network algorithm implemented in a wireless sensor network （WSN）. The proposed detection system mitigates the threat of forest fires by provide accurate fire alarm with low maintenance cost. The accuracy is increased by the novel multi- criteria detection, referred to as an alarm decision depends on multiple attributes of a forest fire. The multi-criteria detection is implemented by the artificial neural network algorithm. Meanwhile, we have developed a prototype of the proposed system consisting of the solar batter module, the fire detection module and the user interface module.

Nonlinear Correction of Pressure Sensor Based on Depth Neural Network

With the global climate change,the high-altitude detection is more and more important in the climate prediction,and the input-output characteristic curve of the air pressure sensor is offset due to the interference of the tested object and the environment under test,and the nonlinear error is generated.Aiming at the difficulty of nonlinear correction of pressure sensor and the low accuracy of correction results,depth neural network model was established based on wavelet function,and Levenberg-Marquardt algorithm is used to update network parameters to realize the nonlinear correction of pressure sensor.The experimental results show that compared with the traditional neural network model,the improved depth neural network not only accelerates the convergence rate,but also improves the correction accuracy,meets the error requirements of upper-air detection,and has a good generalization ability,which can be extended to the nonlinear correction of similar sensors.

Sensor Registration Based on Neural Network in Data Fusion

The contents of sensor registration in the multi-sensor data fusion system are introduced, and some existing methods are analyzed. Then, one approach to sensor registration based on BP neural network is proposed. Here the measurements from radar are transformed from the polar coordinate system to the Cartesian coordinate through a BP neural network. With this approach, the systematic errors are removed as well as the coordinate is transformed. The efficiency of this method is demonstrated by simulation, and the result show that this approach could remove the systematic errors effectively and the DAR are closer to real position than DBR.

Registration algorithm for sensor alignment based on stochastic fuzzy neural network

Multiple sensor registration is an important link in multi-sensors data fusion. The existed algorithm is all based on the assumption that system errors come from a fixed deviation set. But there are many other factors, which can result system errors. So traditional registration algorithms have limitation. This paper presents a registration algorithm for sensor alignment based on stochastic fuzzy neural network (SNFF), and utilized fuzzy clustering algorithm obtaining the number of fuzzy rules. Finally, the simulative result illuminate that this way could gain a satisfing result.

Improved training framework in a neural network model for disruption prediction and its application on EXL-50

A neural network model with a classical annotation method has been used on the EXL-50tokamak to predict impending disruption.However,the results revealed issues of overfitting and overconfidence in predictions caused by inaccurate labeling.To mitigate these issues,an improved training framework has been proposed.In this approach,soft labels from previous training serve as teachers to supervise the further learning process;this has lead to a significant improvement in predictive model performance.Notably,this enhancement is primarily attributed to the coupling effect of the soft labels and correction mechanism.This improved training framework introduces an instance-specific label smoothing method,which reflects a more nuanced model assessment on the likelihood of a disruption.It presents a possible solution to effectively address the challenges associated with accurate labeling across different machines.

Comparison of enhancement techniques based on neural networks for attenuated voice signal captured by flexible vibration sensors on throats

Wearable flexible sensors attached on the neck have been developed to measure the vibration of vocal cords during speech.However,highfrequency attenuation caused by the frequency response of the flexible sensors and absorption of high-frequency sound by the skin are obstacles to the practical application of these sensors in speech capture based on bone conduction.In this paper,speech enhancement techniques for enhancing the intelligibility of sensor signals are developed and compared.Four kinds of speech enhancement algorithms based on a fully connected neural network(FCNN),a long short-term memory(LSTM),a bidirectional long short-term memory(BLSTM),and a convolutional-recurrent neural network(CRNN)are adopted to enhance the sensor signals,and their performance after deployment on four kinds of edge and cloud platforms is also investigated.Experimental results show that the BLSTM performs best in improving speech quality,but is poorest with regard to hardware deployment.It improves short-time objective intelligibility(STOI)by 0.18 to nearly 0.80,which corresponds to a good intelligibility level,but it introduces latency as well as being a large model.The CRNN,which improves STOI to about 0.75,ranks second among the four neural networks.It is also the only model that is able to achieves real-time processing with all four hardware platforms,demonstrating its great potential for deployment on mobile platforms.To the best of our knowledge,this is one of the first trials to systematically and specifically develop processing techniques for bone-conduction speed signals captured by flexible sensors.The results demonstrate the possibility of realizing a wearable lightweight speech collection system based on flexible vibration sensors and real-time speech enhancement to compensate for high-frequency attenuation.

An Efficient and Robust Fall Detection System Using Wireless Gait Analysis Sensor with Artificial Neural Network (ANN) and Support Vector Machine (SVM) Algorithms

In this work, a total of 322 tests were taken on young volunteers by performing 10 different falls, 6 different Activities of Daily Living (ADL) and 7 Dynamic Gait Index (DGI) tests using a custom-designed Wireless Gait Analysis Sensor (WGAS). In order to perform automatic fall detection, we used Back Propagation Artificial Neural Network (BP-ANN) and Support Vector Machine (SVM) based on the 6 features extracted from the raw data. The WGAS, which includes a tri-axial accelerometer, 2 gyroscopes, and a MSP430 microcontroller, is worn by the subjects at either T4 (at back) or as a belt-clip in front of the waist during the various tests. The raw data is wirelessly transmitted from the WGAS to a near-by PC for real-time fall classification. The BP ANN is optimized by varying the training, testing and validation data sets and training the network with different learning schemes. SVM is optimized by using three different kernels and selecting the kernel for best classification rate. The overall accuracy of BP ANN is obtained as 98.20% with LM and RPROP training from the T4 data, while from the data taken at the belt, we achieved 98.70% with LM and SCG learning. The overall accuracy using SVM was 98.80% and 98.71% with RBF kernel from the T4 and belt position data, respectively.

Radial Basis Function Neural Networks-Based Modeling of the Membrane Separation Process: Hydrogen Recovery from Refinery Gases

Membrane technology has found wide applications in the petrochemical industry, mainly in the purification and recovery of the hydrogen resources. Accurate prediction of the membrane separation performance plays an important role in carrying out advanced process control （APC）. For the first time, a soft-sensor model for the membrane separation process has been established based on the radial basis function （RBF） neural networks. The main performance parameters, i.e, permeate hydrogen concentration, permeate gas flux, and residue hydrogen concentration, are estimated quantitatively by measuring the operating temperature, feed-side pressure, permeate-side pressure, residue-side pressure, feed-gas flux, and feed-hydrogen concentration excluding flow structure, membrane parameters, and other compositions. The predicted results can gain the desired effects. The effectiveness of this novel approach lays a foundation for integrating control technology and optimizing the operation of the gas membrane separation process.