Search

This paper presents a neural network predictive control scheme for studying the coagulation process of wastewater treatment in a paper mill. A multi-layer back-propagation neural network is employed to model the nonlinear relationships between the removal rates of pollutants and the chemical dosages, in order to adapt the system to a variety of operating conditions and acquire a more flexible learning ability. The system includes a neural network emulator of the reaction process, a neural network controller, and an optimization procedure based on a performance function that is used to identify desired control inputs. The gradient descent algorithm method is used to realize the optimization procedure. The results indicate that reasonable forecasting and control performances have been achieved through the developed system. [Copyright &y& Elsevier]

The Fokker–Planck equation has significant applications in dynamical systems. In recent years, some neural network methods have been used in combination with physical models to obtain its numerical solutions. However, it is also appealing if the analytical solution of the physical model can be obtained. This paper proposes a neural network-based method for the analytical solution of the FP equation. It relies on neural networks and uses their explicit model as the trial function for the FP equation. The trial function contains the weights and biases in the neural network. Therefore, the solving of the FP equation is converted into the calculation of the weights and biases. In the proposed method, the FP equations are first reduced to a set of easily solvable nonlinear algebraic equations using some trial functions, and then the corresponding weights and biases are determined using the method of pending coefficients. In this paper, linear and nonlinear numerical examples were used to verify the effectiveness of the proposed method. The results demonstrated that the proposed method can obtain the exact solution of the FP equations without data samples. Finally, the proposed method is compared in detail with physics-informed neural networks in terms of computational theory and computational effectiveness. [ABSTRACT FROM AUTHOR]

The online self-calibration is a required capability from an autonomous vehicle that should operate lifelong in a safe manner. This paper introduces relative weighting by a neural network into the batch Gauss–Newton calibration method of the wheel odometry model. A wheel odometry model with accurately estimated parameters could improve the motion estimation task of an autonomous vehicle, but the online parameter identification from only onboard measurements is a challenge due to the noises and the nonlinear behavior of the dynamic system. A possible solution to deal with the effect of noises is to calibrate the model with more segments at once forming a batch, but this can only reduce the distortion effect, not eliminate it. Our proposed algorithm improves this batch formulation by integrating relative weights for the segments to mitigate the distorting effect of noisy measurements. The method applies an AI-based tool to extract a proper weighting strategy from the previously recorded data that utilizes only online signals during operation. With the usage of the proposed architecture, the calibration accuracy significantly increased with the reduction of the distortion effect of faulty measurements, while the same amount of data is used as the raw batch estimation. The performance of the method is demonstrated with real measurements in a city driving with a passenger vehicle, where the calibration signals come from the equipped automotive-grade type of Global Navigation Satellite System and Inertial Measurement Unit. [ABSTRACT FROM AUTHOR]

Forecasting the ocean dynamics is a critical issue for a wide array of climate extremes and environmental crisis. The dynamic variations are traditionally approached by relying on numerical models with all the related physical processes identified beforehand. An efficient alternative forecasting approach is based on the data-driven models. Despite their potential ability in modeling spatio-temporal ocean data, they ignore the fact that the ocean variables in different spatial regions and time periods typically have ever changing influences on each other, thus cannot yield satisfactory prediction results. In this paper, we develop a novel attention based dynamic graph for the ocean variable forecasting problem, which captures both the spatial and temporal dependencies. Specifically, we employ joint self-attention to incorporate information from the spatial graph over the target region, and model the graph evolution across long-range time steps. The performance of the proposed prediction model has been examined in the Indian Ocean based on ocean grid data products datasets. Experimental results demonstrate that this model has significant forecasting capability within 12 months, compared with the numerical methods and the state-of-the-art spatio-temporal embedding baselines. [ABSTRACT FROM AUTHOR]

Monte Carlo Tree Search (MCTS) is a pronounced empirical search algorithm for agent decision-making, especially when enhanced by Deep Learning (DL), in mastering board games that were once thought to be unconquerable. However, it does not appear to be as equally successful in the domain of real-time video games, where the simulation time limit for exploration is a crucial factor, since they are generally designed to be played by human users and hence require a significant amount of resources for simulation. We in this paper propose a surrogate-assisted MCTS approach, specifically targeting commercial real-time video games by approximating the result of gameplay with a deep-learning-based surrogate model. The key contribution of our work is that we designed a modified MCTS for video games that are both commercial and processed in real-time. Since commercial video games include considerably more complex and dynamic gameplays to satisfy their market consumers, as opposed to their non-commercial analogs, our work can be regarded as having challenged the domain unattempted by precedent studies. We validated the performance of our method by conducting a comparative experiment with other algorithms, including the traditional MCTS, under the environment of a commercial real-time video game. [Display omitted] • A Monte Carlo tree search (MCTS) method applicable to commercial real-time video games is proposed. • The method integrates MCTS with a deep neural network structure to function efficiently in a real-time environment. • Performance verification of the proposed surrogate-assisted MCTS is done via two commercial real-time video games with distinct features. • Our surrogate model can be trained with a comparatively little amount of computational resources in real-time video games. [ABSTRACT FROM AUTHOR]

Soil water retention curves (SWRC) are constitutive relationships that are critical for predicting the hydro-mechanical behaviour of unsaturated soil in a wide range of applications in geoenvironmental and geotechnical engineering, hydrology, and agricultural and soil science. Directly measuring SWRC can be costly and time-consuming. Recent research has shown that machine learning, especially artificial neural networks (ANNs), can effectively predict SWRC using easy-to-measure soil properties as inputs. However, significant questions remain about prediction accuracy and what are the most effective input parameters to use, and only shallow neural networks with one hidden layer have been developed in the past for this purpose. In this study, effectiveness of deep neural networks (DNNs) - ANNs with more than one hidden layer - for SWRC prediction is explored for a range of models with different combinations that require easy-to-measure input parameters. Specifically, the study investigates whether switching from a shallow neural network (SNN) to a deep neural network (DNN) can enhance the accuracy of SWRC predictions for these models. Significant differences are found in seven out of thirteen cases when shifting from SNN to DNN. It was also found that models including soil texture, augmented by either dry density or soil porosity as input parameters, perform well. Specific gravity, soil depth, and organic content, on the other hand, do not add to the predictive capability of these models. Findings of this paper can help improve SWRC prediction using machine learning and can lead to more accurate modelling of unsaturated soil behaviour. [ABSTRACT FROM AUTHOR]

Fault testing in the production line of automotive traction machines is essential to ensure the desired lifetime. Since repetitive partial discharges (PDs) caused by anomalies in the insulation system lead to premature breakdowns of electrical machines, a reliable PD detection is of great importance. This paper proposes deep learning (DL) methods to improve the discrimination of PD from background noise in comparison with the state-of-the-art amplitude based PD detection in the production line. First, a systematic data extraction and labeling procedure is introduced to obtain correctly labeled datasets from arbitrary PD measurements. In addition, datasets are enhanced with low signal-to-noise ratio PD pulses by applying a special data augmentation approach. 13 different convolutional, recurrent and fully connected neural networks are compared for various time-frequency representations of the input signals. Hyperparameters for input transform, network topology and solver are optimized for all 13 combinations to ensure a fair case study. As a result, the two-dimensional convolutional neural network with continuous wavelet transform achieves the best accuracy of around 99.76% on a test dataset of PD signals originating from previously not utilized test objects. All DL models considered in this comparison outperform the state-of-the-art threshold-based PD classification. Even for PD events with an amplitude close to the noise level, the detection rate is still around 95% for the best network. Furthermore, without applying the proposed data augmentation procedure, the DL models investigated are not able to distinguish small PD pulses from noise. • Deep learning to improve partial discharge classification of electrical machines. • Rise of partial discharge detection sensitivity and decrease in false alarm rate. • Data extraction, labeling and augmentation procedure to generate reliable datasets. • Comprehensive comparison of time-frequency transforms and network architectures. [ABSTRACT FROM AUTHOR]

Fault diagnosis of rotating machines is essential for the safe and efficient operation of maritime vessels. It prevents potential failures in rotating machines in maritime systems. Simultaneously developed faults severely damage the machines, leading to unprecedented incidents and higher maintenance costs. Data availability for developing multi-fault diagnosis solutions is limited, creating maintainability problems. The unavailability of data also causes further obstacles in the accurate diagnosis of machines. The efficiency of maritime machinery can be improved with intelligent diagnosis by advanced machine learning algorithms. This paper proposes a methodology to diagnose multi-faults with balanced and imbalanced acoustic, vibration and current signals datasets. After acquiring the data from an experimental setup under various speeds, a time-frequency-based Stepping Frame Synchrosqueezed Fourier Transform (SFSFT) is deployed in each segmented signal prior to feature extraction. The extracted features are used for training the Radial Basis Function Neural Network (RBFNN). Different balanced and imbalanced datasets are created to check the multi-fault classification ability of the proposed methodology. The results show that SFSFT-RBFNN is a generalised methodology, which performed tremendously well with a maximum classification accuracy of 100 % for balanced and imbalanced datasets in all three modality datasets. • Synchrosqueezed Fourier Transform is improved with stepping frame extraction method for multi-fault diagnosis. • Minute details of multi-fault signatures are extracted using SFSFT, and RBFNN is used for fault detection. • Performance of the proposed method is tested with balanced and different imbalanced datasets. • Methodology validation is carried out with acoustic, vibration and current datasets. [ABSTRACT FROM AUTHOR]

In this paper, an adaptive neural trajectory tracking controller with a bounded-input property is developed for cable-driven parallel robots (CDPRs). Due to the fact that the cables in these robotic systems should always remain in tension in a trajectory tracking task, a more precise tracking controller is needed for CDPRs comparing to the conventional rigid-link robotic systems. To achieve this objective, this paper proposes a new nonlinear controller with a learning ability for the robot dynamics. The controller includes an adaptive multi-layer neural network to compensate for the modeling uncertainties of the system, and utilizes an auxiliary dynamics to provide a priori bounded tension command for the cables. In addition to this novelty, a bounded-input controller is designed for the dynamics of the actuators, coupled with gearboxes, in order to follow the tensions, defined through the controller of robot dynamics. The boundedness feature of the controller facilitates considering the upper limit of the actuators in choosing the control gains. Stability of the whole system is well studied, and the uniformly ultimately bounded stability is guaranteed. The effectiveness of the proposed control scheme is validated through simulations on a 4-cable planar robot in both nominal and perturbed conditions. [ABSTRACT FROM AUTHOR]

Systems with a priori unknown and time-varying dynamic behavior pose a significant challenge in the field of Nonlinear Model Predictive Control (NMPC). When both the identification of the nonlinear system and the optimization of control inputs are done robustly and efficiently, NMPC may be applied to control such systems. This paper considers stable systems and presents a novel method for adaptive NMPC, called Adaptive Sampling Based Model Predictive Control (Adaptive SBMPC) , that combines a radial basis function neural network identification algorithm with a nonlinear optimization method based on graph search. Unlike other NMPC methods, it does not rely on linearizing the system or gradient based optimization. Instead, it discretizes the input space to the model via pseudo-random sampling and feeds the sampled inputs through the nonlinear model, producing a searchable graph. For this discretization, an optimal path is found using Lifelong Planning A ⁎ , an efficient graph search method. Adaptive SBMPC is used in simulation to identify and control a simple plant with clearly visualized nonlinear behavior. In these simulations, both fixed and time-varying dynamic systems are considered. Results are compared with an adaptive version of Neural GPC, an existing NMPC algorithm based on Newton–Raphson optimization and a back propagation neural network model. When the cost function exhibits many local minima, Adaptive SBMPC is successful in finding a low-cost solution that appears close globally optimal while Neural GPC converges to a solution that is only locally optimal. This paper presents the method, soundness and completeness theory, and two simulated NMPC examples. The first is a transparent single-input single-output example, and the second considers a more complex power plant combustion process with two inputs and three outputs. [ABSTRACT FROM AUTHOR]

Simultaneously extracting junctions and their corresponding line segments from images presents a promising approach to structural environment cognition. However, conventional methods employ square convolution for line feature extraction, resulting in the exclusion of long-range dependencies and the generation of suboptimal wireframe predictions. In this paper, we introduce an efficient and concise parsing method named Structural Asymmetric Convolution-based Wireframe Parser (SACWP). Taking advantage of the inherent similarities between structural asymmetric convolution and the predominant distribution of line segments in man-made environments, we propose a Structural Asymmetric Convolution module (SAC) that captures long-range contextual features while efficiently filtering out irrelevant information from neighboring pixels. Additionally, we introduce a feature aggregation module based on dilated convolution (DCFA) to seamlessly integrate contextual information from multiple receptive fields. We thoroughly evaluate our approach on the Wireframe and YorkUrban datasets, achieving preferable results of 69.3% and 29.7% msAP respectively. On the other hand, the promising results adequately demonstrate the effectiveness of SACWP to Wireframe Parsing task. [ABSTRACT FROM AUTHOR]

In this paper, a hybrid weight design and optimization mechanism is proposed to ensure a better application effect in the multi-level weighted evaluation/optimization process of intelligent neural network (NN) guidance algorithm. According to different target tracking scenarios, the tracking index are divided into different periods, a single period score is given according to a linear-nonlinear hybrid scoring mechanism. Furthermore, in a single index, the internal scores of different time periods are weighted and fused for the second time to obtain the index overall score. Finally, the third weighted fusion of the multi-index scores is used to obtain the comprehensive score of the algorithm. Design the algorithm parameter evaluation case sets, repeat the above compound weighting, obtain the case with the highest comprehensive score. Finally, the algorithm is optimized by the highest score case. In the hybrid weight design mechanism, the objective grey correlation first defines the weight range, then the subjective expert decision method with local extreme expert elimination and standard deviation global test for double optimization finally ensures the weight reliability of the hierarchical evaluation optimization. The evaluation optimization experiment using fuzzy NN radar seeker verifies the effectiveness of the method. [ABSTRACT FROM AUTHOR]

- In recent years, the aviation industry has witnessed a substantial integration of power electronics technology within Aircraft Microgrids (AMs). Consequently, the extension of electrical wiring networks has expanded, resulting in heightened intricacies within these systems. Therefore, the identification and location of faults in wiring networks have become an important topic in AMs to guarantee the safety of the electrical power systems. Time Domain Reflectometry (TDR) is widely used to locate and recognize electric wire faults. However, soft fault location on complex electrical networks using TDR is perplexing due to its weak effect on the reflected signal. Moreover, the existence of noise in the environment can worsen the TDR's performance. In this paper, a new approach based on TDR, along with the Subtractive Correlation Method (SCM) and Neural Network (NN), is proposed. The TDR response of the complex wiring network is determined using a newly proposed model based on the Finite Difference Time Domain (FDTD) method. The validity of the proposed model is established through experimentation including two distinct cable types also the introduced model notably enhances computational efficiency, a fact substantiated by our experimental findings and an extensive benchmarking against recent publications. These evaluations collectively underscore a significant reduction in computational time. Then, the reflected signal undergoes processing through SCM, a technique employed to amplify the subtle influence of the soft fault in two scenarios: one accompanied by noise and the other noise-free. Furthermore, NN is used to handle the inverse problem of localizing and characterizing the soft faults by their exact resistance values. Even within noisy environments, the proposed methodology excels in accurately locating and characterizing soft faults with a high degree of precision, all in real-time diagnostic scenarios. • Novel FDTD-based model for determining the TDR response of intricate wiring networks. • Empirically validated model enhancing computational efficiency and reducing time. • SCM approach amplifies subtle soft fault impacts, even in noisy environments. • Neural Networks excel in precisely localizing and characterizing soft faults by resistance values. • Real-time diagnostic capability for pinpointing and characterizing soft faults with exceptional precision. [ABSTRACT FROM AUTHOR]

This paper focuses on horizontal path following for an underactuated unmanned airship under output constraints and uncertainties. To handle the output constraints, we design a guidance loop based on a novel time-varying tan-type barrier Lyapunov function to generate the desired yaw angle. By using the backstepping technique, the controller is then designed to realize the desired yaw and velocity tracking. For uncertainties and disturbances, an adaptive radial basis function neural network is applied for estimation and compensation. Moreover, to avoid complex derivative calculations for virtual control and guarantee that the desired attitude is bounded, a command filter and an auxiliary system are combined in the system. Theoretical analysis shows that, despite the influence of disturbances, the designed controller ensures that the constraint of airship position is never violated while the closed loop signals are all uniformly ultimately bounded. Comparable simulations illustrate the performance of our controller. [ABSTRACT FROM AUTHOR]

Natural language processing is a technique to process data such as text and speech. Some fundamental research includes named-entity recognition, which recognizes name entities (i.e., persons, companies) from texts; semantic parsing, which is used to convert a natural language utterance to the representation of logical form; and co-reference resolution, which extracts nouns (including pronouns, noun phrases) pointing to the same reference body. In this paper, we mainly focus on the task of mention extraction, which extract and classify overlapping or nested structure mentions. We proposed a neural-encoded mention-hypergraph (NEMH) model to use hypergraph to model overlapping or nested structure mentions and use neural networks to extract features for hypergraph automatically. Unlike the existing approaches, our hypergraph model can effectively capture nested mention entities with unlimited lengths. Also, the proposed model is highly scalable and the time complexity of the proposed model is linear in the number of mention classes and the number of input words. Extensive experiments are conducted on several standard datasets to demonstrate the effectiveness of the proposed model. [ABSTRACT FROM AUTHOR]

Recently, facial expression recognition (FER) methods have achieved significant progress. However, FER is still challenged by factors such as uneven illumination and low-quality expression images. Exploring the potential of facial expression features to achieve robust FER is particularly important. Inspired by the Transformer's excellent performance in modeling long-range dependencies and in vision tasks, this paper proposes a Transformer Block Enhancement Module (TBEM) for enhancing the feature representation of facial expressions. The proposed module contains a Channel Enhancement (CE) block and a Spatial Enhancement (SE) block. The CE block can adaptively enhance the expression features on the channel dimension by effectively leveraging the channel dependency information, while the SE block enhances the expression features on the spatial dimension by integrating the spatial dependency information. TBEM can output a more robust expression representation by combining CE and SE. The proposed artificial intelligence learning module greatly improves the recognition accuracy of FER engineering tasks. To further illustrate the application of TBEM in real-world FER engineering, three engineering problems are used for verification. Extensive experiments demonstrate that the proposed method improves FER performance by focusing on more accurate decisional features and can be easily embedded into regular convolutional neural network models to help them improve the accuracy on FER tasks by about 2.64%–3.03%. The results show the proposed method achieves accuracies of 90.57% on FERPlus, 89.41% on RAFDB basic and 68.43% on RAFDB compound, respectively. The proposed method also provides a meaningful reference for further research on applying Transformer to other tasks. [ABSTRACT FROM AUTHOR]

Abstract: There are much research effort in the literature using Monte Carlo simulation (MCS) which is a direct and simple numerical method, however, it can be computationally very expensive as the governing dynamic equations of the system to be simulated for each random sample using the MCS. In this paper, polynomial meta-models based on the evolved group method of data handling (GMDH) neural networks are obtained to simply calculate the probability of failure in the MCS, instead of direct solution of dynamic equation of system. In this way, some input–output data consisting of uncertain parameters of system and controller parameters as inputs and probability of failure of some cost functions as output are used for training such GMDH-type neural networks which replace the very time consuming direct solution of dynamical systems during the MCS. A multi-objective genetic algorithm is also used for Pareto optimal design of PI and PID controllers for both first and second order uncertain system with time delays using methodology of this paper. The objective functions that are considered for such Pareto multi-objective optimization are namely, probability of failure of settling time (P TS ) and probability of failure of overshoot (P OS ). The comparisons of the obtained results using the method of this paper with those obtained using direct method shows a significant reduction in computational time, whilst the accuracy is maintained. [Copyright &y& Elsevier]

Abstract: Improving the efficiency of the carbon dioxide (CO2) capture process requires a good understanding of the intricate relationships among parameters involved in the process. The objective of this paper is to study the relationships among the significant parameters impacting CO2 production. An enhanced understanding of the intricate relationships among the process parameters supports prediction and optimization, thereby improving efficiency of the CO2 capture process. Our modeling study used the 3-year operational data collected from the amine-based post combustion CO2 capture process system at the International Test Centre (ITC) of CO2 Capture located in Regina, Saskatchewan of Canada. This paper describes the data modeling process using the approaches of (1) neural network modeling combined with sensitivity analysis and (2) neuro-fuzzy modeling technique. The results from the two modeling processes were compared from the perspectives of predictive accuracy, inclusion of parameters, and support for explication of problem space. We conclude from the study that the neuro-fuzzy modeling technique was able to achieve higher accuracy in predicting the CO2 production rate than the combined approach of neural network modeling and sensitivity analysis. [Copyright &y& Elsevier]

The paper presents three new autonomous systems for the control of lateral-directional motion of airplane during landing. The geometry of airplane motion in lateral-directional plane takes into account the wind having different velocities and directions. All the three control systems contain an adaptive control subsystem based on the concepts of dynamic inversion and neural networks; the reference models of the adaptive control systems receive the signals provided by Pseudo Control Hedging blocks, as additional signals, to cancel the neural networks’ adapting difficulties in the case of nonlinear actuators with time delay and saturation zones with respect to the velocity and displacement. Two of the three automatic control systems have proportional–derivative/proportional–integral– derivative controllers for the control of the lateral deviation of airplane with respect to the runway. The new designed adaptive architectures have been software implemented and validated by complex numerical simulations; the obtained results prove the new architectures’ stability and small overshoots. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a performance enhancement scheme for the recently developed extreme learning machine (ELM) for multi-category sparse data classification problems. ELM is a single hidden layer neural network with good generalization capabilities and extremely fast learning capacity. In ELM, the input weights are randomly chosen and the output weights are analytically calculated. The generalization performance of the ELM algorithm for sparse data classification problem depends critically on three free parameters. They are, the number of hidden neurons, the input weights and the bias values which need to be optimally chosen. Selection of these parameters for the best performance of ELM involves a complex optimization problem. In this paper, we present a new, real-coded genetic algorithm approach called ‘RCGA-ELM’ to select the optimal number of hidden neurons, input weights and bias values which results in better performance. Two new genetic operators called ‘network based operator’ and ‘weight based operator’ are proposed to find a compact network with higher generalization performance. We also present an alternate and less computationally intensive approach called ‘sparse-ELM’. Sparse-ELM searches for the best parameters of ELM using K-fold validation. A multi-class human cancer classification problem using micro-array gene expression data (which is sparse), is used for evaluating the performance of the two schemes. Results indicate that the proposed RCGA-ELM and sparse-ELM significantly improve ELM performance for sparse multi-category classification problems. [ABSTRACT FROM AUTHOR]

Abstract: Economic evaluation of a new oil well is important for decision-making in the petroleum industry, and this evaluation is based on a good prediction on a well''s production. However, it is difficult to accurately predict a well''s production due to the complex subsurface conditions of reservoirs. The industrial standard approach is to use either curve-fitting methods or complex and time-consuming reservoir simulations. In this paper, an enhanced decision tree learning approach called neural-based decision tree (NDT) model is applied in an attempt to investigate its performance in predicting petroleum production. The primary strength of this model is that it can capture dependencies among attributes, and therefore, it is likely to provide an improved or more accurate prediction (). This paper presents an application of the NDT model for petroleum prediction. Our models were developed based on the five most significant parameters that affect oil production: permeability, porosity, first shut-in pressure, residual oil and saturation of water. The five parameters were used as input variables, and oil production is the output variable for modeling. Four different models were generated in the modeling process, and each involves a different combination of parameters. First, an overall oil production model is developed using the three geoscience parameters of permeability, porosity and first shut-in pressure. Secondly, two different models, with different input parameters, were developed to predict production in the post-water flooding stage only. The results of the above models indicate that data-driven models may not be effective for classifying the data set. Hence, a trend model was developed in an attempt to improve the effectiveness and accuracy of the predictive model. The result shows that the trend model can provide an improved performance, and its performance is comparable to that of the artificial neural network. [Copyright &y& Elsevier]

Abstract: This paper deals with the problem of robust tire/road friction force estimation. Availability of actual value of the friction force generated in contact between the tire and the road has significant importance for active safety systems in modern cars, e.g. anti-lock brake systems, traction control systems, vehicle dynamic systems, etc. Since state estimators are usually based on the process model, they are sensitive to model inaccuracy. In this paper we propose a new neural network based estimation scheme, which makes friction force estimation insensitive to modelling inaccuracies. The neural network is added to the estimator in order to compensate effects of the friction model uncertainties to the estimation quality. An adaptation law for the neural network parameters is derived using Lyapunov stability analysis. The proposed state estimator provides accurate estimation of the tire/road friction force when friction characteristic is only approximately known or even completely unknown. Quality of the estimation is examined through simulation using one wheel friction model. Simulation results suggest very fast friction force estimation and compensation of the changes of the model parameters even when they vary in wide range. [Copyright &y& Elsevier]

This paper proposes a new passive robust fault detection scheme using non-linear models that include parameter uncertainty. The non-linear model considered here is described by a group method of data handling (GMDH) neural network. The problem of passive robust fault detection using models including parameter uncertainty has been mainly addressed by checking if the measured behaviour is inside the region of possible behaviours based on the so-called forward test since it bounds the direct image of an interval function. The main contribution of this paper is to propose a new backward test, based on the inverse image of an interval function, that allows checking if there exists a parameter in the uncertain parameter set that is consistent with the measured system behaviour. This test is implemented using interval constraint satisfaction algorithms which can perform efficiently in deciding if the measured system state is consistent with the GMDH model and its associated uncertainty. Finally, this approach is tested on the servoactuator being a FDI benchmark in the European Project DAMADICS. [Copyright &y& Elsevier]

Computerised processing of medical images can ease the search of the representative features in the images. The endoscopic images possess rich information expressed by texture and regions affected by diseases, such as ulcer or coli, may have different texture features. In this paper schemes have been developed to extract features from the texture spectra in the chromatic and achromatic domains for a selected region of interest from each colour component histogram of images acquired by the M2A Swallowable Imaging Capsule. The implementation of neural network schemes and the concept of fusion of multiple classifiers have been also adopted in this paper. The preliminary test results support the feasibility of the proposed method. [Copyright &y& Elsevier]

The pneumatic muscle actuator (PMA) is one of the most promising actuators especially for the applications that require greater proximity between the humans and the robots. Fast and precise control of PMA, however, is difficult to achieve due to the compressibility of the air and the elasticity of the PMA. In order to achieve accurate and consistent tracking performance of a one axis PMA actuated manipulator over considerably wide ranges of frequency and stroke, an intelligent adaptive control algorithm is developed in this paper. The adaptive learning is enabled by a neural network in which the control gains to a fuzzy sliding mode controller (FSMC) and an integrator are adjusted to minimize the tracking error. Experimental results show that the proposed control strategy achieves fast, accurate and consistent performance tracking sinusoidal reference trajectories up to 1 Hz in frequency and close to the extreme stroke of the PMA actuated manipulator with the compressed air regulated to 4 bar. Results also show that the proposed control strategy, with a more aggressive learning for the control gain to the FSMC, achieves satisfactory performance tracking a trapezoidal reference trajectory. [ABSTRACT FROM AUTHOR]

This paper presents a new general methodological approach to imbalanced learning as one of the challenging problems in pattern classification. The presented method is founded on maximization of the sample entropy. The method involves detection of distributive properties of ideally balanced regular lattice sample and acceptable transfer of these properties to an arbitrary imbalanced sample increasing its representativeness. The proposed procedure assumes undersampling applied on areas of high probability density in the sample space combined with oversampling in the areas of low density. The main achievement of this method is the increased sample class entropy which reduces the inductive learner’s tendency to favor prominent class, or cluster. In addition to class balancing, this method can be useful for function approximation, clustering, and sample dimension reduction. The high degree of generality of the method implies its applicability on data of various complexity and imbalance. The presented theoretical foundation of the method was verified on a set of proper synthetic samples. The method’s practical usability is confirmed by a comparative classification of a large set of databases including speech signal samples. [ABSTRACT FROM AUTHOR]

Autonomous robotic grasping is an essential skill for service robots to perform specified tasks in unstructured scenarios. Previous work focus on simple pick-and-place tasks, and it is not satisfactory for real-world scenes that have requirements for manipulation. In this paper, we present a modular intelligent robot architecture via multi-task convolutional neural network which can be used for specific object grasping and manipulation in a stacked and cluttered environment. Firstly, an end-to-end, multi-task semantic grasping convolutional neural network (MSG-ConvNet) that simultaneously outputs the results of grasp detection and semantic segmentation is proposed to recognize the affiliations between objects and grasps in cluttered scenarios. Secondly, we propose a post-processing method which allows the robot to select an optimal grasping area in an active perception way through simply reasoning on the multi-modal information output by the proposed model. The proposed multi-task network has a great improvement in both recognition accuracy and detection speed on the public multi-object dataset GraspNet-1Billion compared with the benchmark. The proposed grasp detection method also yields state-of-the-art performance with accuracies of 95.06% and 98.6% on the public single-object Jacquard Dataset and Cornell Dataset, respectively. In addition, the experiments in a real-world scene demonstrate that our proposed method has stronger robustness and adaptability than the simple direct grasping strategy in the environment with higher mutual occlusion. [ABSTRACT FROM AUTHOR]

TV program recommendation is important for users in the face of a huge amount of information data. The existing TV program recommendation mainly relies on a collaborative filtering method to recommend through interactive data between users and programs. Although some methods utilize auxiliary information to enrich semantic features, most of them only use a single data type, which cannot capture a more diverse feature representation of the user and program. In this paper, we propose a neural TV program recommendation model with multi-source heterogeneous data, which makes full use of the multi-source heterogeneous auxiliary information. Specifically, we combine heterogeneous features derived from auxiliary information to learn a deep program representation in the program encoder module. To more accurately capture user preferences, we further utilize the personalized attention mechanism to determine the importance of different programs to the user representation based on the interaction between users and programs in the user encoder module. Extensive experiments on a real dataset of the Chinese capital show that our model can effectively improve the performance of TV program recommendations compared to the existing models. [ABSTRACT FROM AUTHOR]

Cavitation flow is a typical complex flow phenomenon, which involves many flow mechanisms. At present, the main approach to reconstruct the cavitation flow field based on the experimental results is numerical simulation, which has the defects of low computational efficiency and is difficult to effectively use the experimental data. In this paper, a chain-style physics-informed neural network (chain-style PINN) is developed to solve the reconstruction problem of cavitation flow field. On the basis of decoupling the governing equations, our method solves the physical quantities of interest serially by introducing multiple serial PINNs. A physics-informed loss function is defined to realize the assimilation of experimental data and physical mechanism. The prediction for a 3D NACA66 hydrofoil case is validated by comparing with Direct Numerical Simulation (DNS), which demonstrates that the calculation time is reduced by about 70% while the relative L 2 errors of pressure and liquid volume fraction fields are only 0.0030 and 0.0035. While comparing with the existing method Hidden Fluid Mechanics (i.e., baseline PINN), the results show the validity of our method. To the best of our knowledge, this is the first theoretical work that applies PINN to cavitation flow. [ABSTRACT FROM AUTHOR]

The exchange rate forecasting plays an important role in the economic and financial fields. Oil price fluctuations have a great impact on the country's economic activity. Based on the theory of exchange rate determination, this paper combines quantitative and qualitative research methods to study the impact of three major international benchmark crude oil price changes on the real effective exchange rate forecast of the RMB. In this paper, the correlation between the three international benchmark oil prices and the real exchange rate is discussed through the hybrid Copula function, and then the bivariate neural network model is constructed. Brent crude oil price having the greatest degree of correlation with China's real exchange rate, the Kendall correlation coefficient and the Spearman correlation coefficient are 0.4327 and 0.5792. MAPE values of the three Brent variable models in the one-step prediction reached 0.54%, 0.51%, and 0.54%. The results of the experiments and discussions shows that the bivariate model has excellent forecasting performance, and indicates that the continuous fluctuations in oil prices have a great impact on the exchange rate, and that oil price information has provided effective help for China's real exchange rate forecasting. • A hybrid Copula function is successfully developed to calculate correlations. • A data preprocessing strategy is employed to process the raw data sequence. • A bivariate forecasting model is proposed to achieve superior performance. • A scientific and effective evaluation system is established. [ABSTRACT FROM AUTHOR]

The artificial intelligence (AI) revolution offers significant opportunities to capitalise on the growth of digitalisation and has the potential to enable the 'system of systems' approach required in increasingly complex infrastructure systems. This paper reviews the extent to which research in economic infrastructure sectors has engaged with fields of AI, to investigate the specific AI methods chosen and the purposes to which they have been applied both within and across sectors. Machine learning is found to dominate the research in this field, with methods such as artificial neural networks, support vector machines, and random forests among the most popular. The automated reasoning technique of fuzzy logic has also seen widespread use, due to its ability to incorporate uncertainties in input variables. Across the infrastructure sectors of energy, water and wastewater, transport, and telecommunications, the main purposes to which AI has been applied are network provision, forecasting, routing, maintenance and security, and network quality management. The data-driven nature of AI offers significant flexibility, and work has been conducted across a range of network sizes and at different temporal and geographic scales. However, there remains a lack of integration of planning and policy concerns, such as stakeholder engagement and quantitative feasibility assessment, and the majority of research focuses on a specific type of infrastructure, with an absence of work beyond individual economic sectors. To enable solutions to be implemented into real-world infrastructure systems, research will need to move away from a siloed perspective and adopt a more interdisciplinary perspective that considers the increasing interconnectedness of these systems. • Literature on the use of AI in infrastructure sectors is reviewed. • Types of AI, trends across sectors and purposes within infrastructure are identified. • Findings have been evaluated using an analysis framework. • The future of AI and the 'Survey PaperInternet of Things' in infrastructure are discussed. [ABSTRACT FROM AUTHOR]

Abstract: Skin colour is considered to be a useful and discriminating spatial feature for many skin detection-related applications, but it is not sufficiently robust to address complex image environments because of light-changing conditions, skin-like colours and reflective glass or water. These factors can create major difficulties in face pixel-based skin detectors when the colour feature is used. Thus, this paper proposes a multi-agent learning method that combines the Bayesian method with a grouping histogram (GH) technique and the back-propagation neural network with a segment adjacent-nested (SAN) technique based on the YCbCr and RGB colour spaces, respectively, to improve skin detection performance. The findings from this study have shown that the proposed multi-agent learning for skin detector has produced significant true positive (TP) and true negative (TN) average rates (i.e. 98.44% and 99.86% respectively). In addition, it has achieved a significantly lower average rate for the false negative (FN) and false positive (FP) (i.e. only 1.56% and 0.14% respectively). The experimental results show that multi-agent learning in the skin detector is more efficient than other approaches. [Copyright &y& Elsevier]

Abstract: This paper presents a capable neural network for solving strictly convex quadratic programming (SCQP) problems with general linear constraints. The proposed neural network model is stable in the sense of Lyapunov and can converge to an exact optimal solution of the original problem. A block diagram of the proposed model is also given. Several applicable examples further show the correctness of the results and the good performance of the model. [Copyright &y& Elsevier]

Abstract: Neural networks with fixed input length are not able to train and test data with variable lengths in one network size. This issue is very crucial when the neural networks need to deal with signals of variable lengths, such as speech. Though various methods have been proposed in segmentation and feature extraction to deal with variable lengths of the data, the size of the input data to the neural networks still has to be fixed. A novel Self-Adjustable Neural Network (SANN) is presented in this paper, to enable the network to adjust itself according to different data input sizes. The proposed method is applied to the speech recognition of Malay vowels and TIMIT isolated words. SANN is benchmarked against the standard and state-of-the-art recogniser, Hidden Markov Model (HMM). The results showed that SANN was better than HMM in recognizing the Malay vowels. However, HMM outperformed SANN in recognising the TIMIT isolated words. [Copyright &y& Elsevier]

Abstract: In this paper, a neural network model is constructed on the basis of the duality theory, optimization theory, convex analysis theory, Lyapunov stability theory and LaSalle invariance principle to solve general convex nonlinear programming (GCNLP) problems. Based on the Saddle point theorem, the equilibrium point of the proposed neural network is proved to be equivalent to the optimal solution of the GCNLP problem. By employing Lyapunov function approach, it is also shown that the proposed neural network model is stable in the sense of Lyapunov and it is globally convergent to an exact optimal solution of the original problem. The simulation results also show that the proposed neural network is feasible and efficient. [Copyright &y& Elsevier]

Abstract: This paper shows how temporal difference learning can be used to build a signalized junction controller that will learn its own strategies through experience. Simulation tests detailed here show that the learned strategies can have high performance. This work builds upon previous work where a neural network based junction controller that can learn strategies from a human expert was developed (Box and Waterson, 2012). In the simulations presented, vehicles are assumed to be broadcasting their position over WiFi giving the junction controller rich information. The vehicle''s position data are pre-processed to describe a simplified state. The state-space is classified into regions associated with junction control decisions using a neural network. This classification is the strategy and is parametrized by the weights of the neural network. The weights can be learned either through supervised learning with a human trainer or reinforcement learning by temporal difference (TD). Tests on a model of an isolated T junction show an average delay of 14.12s and 14.36s respectively for the human trained and TD trained networks. Tests on a model of a pair of closely spaced junctions show 17.44s and 20.82s respectively. Both methods of training produced strategies that were approximately equivalent in their equitable treatment of vehicles, defined here as the variance over the journey time distributions. [Copyright &y& Elsevier]

Abstract: A type of optimized neural networks with limited precision weights (LPWNN) is presented in this paper. Such neural networks, which require less memory for storing the weights and less expensive floating point units in order to perform the computations involved, are better suited for embedded systems implementation than the real weight ones. Based on analyzing the learning capability of LPWNN, Quantize Back-propagation Step-by-Step (QBPSS) algorithm is proposed for such neural networks to overcome the effects of limited precision. Methods of designing and training LPNN are represented, including the quantization of non-linear activation function and the selection of learning rate, network architecture and weights precision. The optimized LPWNN performance has been evaluated by comparing to conventional neural networks with double-precision floating-point weights on road recognition of image for intelligent vehicle in ARM 9 embedded systems, and the results show the optimized LPWNN has 7 times faster than the conventional ones. [Copyright &y& Elsevier]

Abstract: This research investigates the performance of dynamic modelling using non-parametric techniques for identification of a flexible structure system for development of active vibration control. In this paper, the implementation details are described and the experimental studies conducted in this research are analysed. The input–output data of the system were first acquired through the experimental studies using National Instruments (NI) data acquisition system. A sinusoidal force was applied to excite the flexible plate and the dynamic response of the system was then investigated. Non-parametric modelling of the system were developed using several artificial intelligent methodologies namely Adaptive Elman Neural Networks (ENN), Backpropagation Multi-layer Perceptron Neural Networks (MLPNN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). The performance of all these methodologies were compared and discussed. Finally, validation and verification of the obtained model was conducted using One Step Ahead (OSA) prediction, mean squared error (MSE) and correlation tests. The prediction ability of the model was further observed with unseen data. The results verified that the MLPNN converge to an optimum solution faster and the dynamic model obtained described the flexible plate structure very well. The non-parametric models of the flexible plate structure thus developed and validated will be used as the representation of the transfer function of the system in subsequent investigations for the development of active vibration control strategies for vibration suppression in flexible structures. [Copyright &y& Elsevier]

Abstract: This paper presents an original software implementation of the elitist non-dominated sorting genetic algorithm (NSGA-II) applied and adapted to the multi-objective optimization of a polysiloxane synthesis process. An optimized feed-forward neural network, modeling the variation in time of the main parameters of the process, was used to calculate the vectorial objective function of NSGA-II, as an enhancement to the multi-objective optimization procedure. An original technique was utilized in order to find the most appropriate parameters for maximizing the performance of NSGA-II. The algorithm provided the optimum reaction conditions (reaction temperature, reaction time, amount of catalyst, and amount of co-catalyst), which maximize the reaction conversion and minimize the difference between the obtained viscometric molecular weight and the desired molecular weight. The algorithm has proven to be able to find the entire non-dominated Pareto front and to quickly evolve optimal solutions as an acceptable compromise between objectives competing with each other. The use of the neural network makes it also suitable to the multi-objective optimization of processes for which the amount of knowledge is limited. [Copyright &y& Elsevier]

Abstract: Measuring the quality parameters of materials at mines is difficult and a costly job. In this paper, an image analysis-based method is proposed efficiently and cost effectively that determines the quality parameters of material. The image features are extracted from the samples collected from a mine and modeled using neural networks against the actual grade values of the samples generated by chemical analysis. The dimensions of the image features are reduced by applying the genetic algorithm. The results showed that only 39 features out of 189 features are sufficient to model the quality parameter. The model was tested with the testing data set and the result revealed that the estimated grade values are in good agreement with the real grade values (R 2=0.77). The developed method was then applied to a case study mine of iron ore. The case study results show that proposed image-based algorithm can be a good alternative for estimating quality parameters of materials at a mine site. The effectiveness of the proposed method was verified by applying it on a limestone deposit and the results revealed that the method performed equally well for the limestone deposit. [Copyright &y& Elsevier]

Abstract: This paper investigates the potential of support vector machines based regression approach to model the local scour around bridge piers using field data. A dataset of consisting of 232 pier scour measurements taken from BSDMS were used for this analysis. Results obtained by using radial basis function and polynomial kernel based Support vector regression were compared with four empirical relation as well as with a backpropagation neural network and generalized regression neural network. A total of 154 data were used for training different algorithms whereas remaining 78 data were used to test the created model. A coefficient of determination value of 0.897 (root mean square error=0.356) was achieved by radial basis kernel based support vector regression in comparison to 0.880 and 0.835 (root mean square error=0.388 and 0.438) by backpropagation neural and generalized regression neural network. Comparisons of results with four predictive equations suggest an improved performance by support vector regression. Results with dimensionless data using all three algorithms suggest a better performance by dimensional data with this dataset. Sensitivity analysis suggests the importance of depth of flow and pier width in predicting the scour depth when using support vector regression based modeling approach. [Copyright &y& Elsevier]

Abstract: Uncertain ground conditions represent the primary source of risk in underground tunnel construction. However, this problem can be solved by developing an accurate, probabilistic description of the geology. This paper presents a general model for probability based determination of tunnel geology that can be used as a basis for developing more effective decision support systems for tunneling design and construction. The proposed model is based on a Hidden Markov Model (HMM) and a neural network (NN). An approximate inference technique – a Particle Filter (PF) Algorithm – is used to simulate the geological parameters. This model overcomes the deficiencies of existing models by readily incorporating all available geologic information and updating geologic predictions based on observations given by the neural network. In order to validate the proposed model, the “Drainage Water Tunnel Project” at Zhong-He, Taipei, Taiwan was used. The results showed that the Neural-HMM model provides high accuracy in geological prediction. [Copyright &y& Elsevier]

Abstract: Neural networks have been employed in a multitude of transportation engineering applications because of their powerful capabilities to replicate patterns in field data. Predictions are always subject to uncertainty arising from two sources: model structure and training data. For each prediction point, the former can be quantified by a confidence interval, whereas total prediction uncertainty can be represented by constructing a prediction interval. While confidence intervals are well known in the transportation engineering context, very little attention has been paid to construction of prediction intervals for neural networks. The proposed methodology in this paper provides a foundation for constructing prediction intervals for neural networks and quantifying the extent that each source of uncertainty contributes to total prediction uncertainty. The application of the proposed methodology to predict bus travel time over four bus route sections in Melbourne, Australia, leads to quantitative decomposition of total prediction uncertainty into the component sources. Overall, the results demonstrate the capability of the proposed method to provide robust prediction intervals. [Copyright &y& Elsevier]

Abstract: Improving the efficiency of the carbon dioxide (CO2) capture process requires a good understanding of the intricate relationships among parameters involved in the process. The objective of this research is to study the nature of relationships among the key parameters using the approaches of artificial neural network and statistical analysis. Our modeling study used the three-year operational data collected from the amine-based post-combustion CO2 capture process at the International Test Centre of CO2 Capture (ITC) located in Regina, Saskatchewan of Canada. The goal of CO2 capture is to capture and remove CO2 from industrial gas streams before they are released into the atmosphere. The amine solution is used at ITC for absorbing CO2 from the industrial flue gas, and then the CO2 is separated from the amine solution. The amine solution recycles for further CO2 capture and the CO2 stream can be stored or used for other industrial purposes. This paper describes the data modeling process using the approaches of: (1) statistical analysis and (2) neural network modeling combined with sensitivity analysis. The results from the two modeling process were compared from the perspectives of predictive accuracy, inclusion of parameters, support for exploration and explication of problem space, modeling uncertainty, and involvement of experts. It was observed that the approach of neural network modeling combined with sensitivity analysis achieved much higher accuracy on predicting CO2 production rate than the statistical study. [ABSTRACT FROM AUTHOR]

Abstract: This paper develops an evolutionary fuzzy hybrid neural network (EFHNN) to enhance project cash flow management. The developed EFHNN combines neural networks (NN) and high order neural networks (HONN) into a hybrid neural network (HNN), which acts as the major inference engine and operates with alternating linear and nonlinear NN layer connections. Fuzzy logic is employed to sandwich the HNN between a fuzzification and defuzzification layer. The authors developed and applied the EFHNN to sequential cash flow trend problems by fusing HNN, FL, and GA. Results show that the proposed EFHNN can be deployed effectively to sequential cash flow estimation. The performance of linear and nonlinear (high order) neuron layer connectors in the EFHNN was significantly better than the performance achieved by previous models that used singular linear NN. Trained results were used for the prediction and strategic management of project cash flow. The proposed strategy can assist project managers to control project cash flows within the banana envelope of the S-curve to enhance project success. [Copyright &y& Elsevier]

Abstract: Analysis of safety in surface coal mines represents a very complex process. Published studies on mine safety analysis are usually based on research related to accidents statistics and hazard identification with risk assessment within the mining industry. Discussion in this paper is focused on the application of AI methods in the analysis of safety in mining environment. Complexity of the subject matter requires a high level of expert knowledge and great experience. The solution was found in the creation of a hybrid system PROTECTOR, whose knowledge base represents a formalization of the expert knowledge in the mine safety field. The main goal of the system is the estimation of mining environment as one of the significant components of general safety state in a mine. This global goal is subdivided into a hierarchical structure of subgoals where each subgoal can be viewed as the estimation of a set of parameters (gas, dust, climate, noise, vibration, illumination, geotechnical hazard) which determine the general mine safety state and category of hazard in mining environment. Both the hybrid nature of the system and the possibilities it offers are illustrated through a case study using field data related to an existing Serbian surface coal mine. [Copyright &y& Elsevier]

Abstract: In this paper two energy dispatch controllers for use in a grid-independent photovoltaic (PV) system are presented. The first, an optimal energy dispatch controller, is based on a class of Adaptive Critic Designs (ACDs) called Action Dependent Heuristic Dynamic Programming (ADHDP). This class of ACDs uses two neural networks to evolve an optimal control strategy over time. The first neural network or “Action” network dispenses the actual control signals while the second network or “Critic” network uses these control signals along with the system states to provide feedback to the action network, measuring performance using a utility function. This feedback loop allows the action network to improve behavior over time. The optimal energy dispatcher places emphasis on always meeting the critical load, followed by keeping the charge of the battery as high as possible so as to be able to power the critical load in cases of extended low output from the PV array, and lastly to power the non-critical load in so far as to not interfere with the first two objectives. The second energy dispatch controller is a smart energy dispatch controller and is built using knowledge from an expert, codified into a series of static rules. This smart energy dispatch controller is called the “PV-priority 2” controller. These energy dispatchers are compared with a static scheme called the “PV-priority 1”. The PV-priority 1 controller represents the standard control strategy. Results show that the ADHDP-based optimal energy dispatcher (or controller) outperforms the standard PV-priority 1 energy dispatcher in meeting the stated objectives, but trails the PV-priority 2 energy dispatcher. However, the major advantage of the ADHDP controller is that no expert is required for designing the controller, whereas for a rule-based controller such as the PV-priority 2 controller, an expert is always required. [Copyright &y& Elsevier]

Abstract: This paper proposes a reinforced iterative formalism to learn from intentional human errors called barrier removal and from uncertainty on human-error parameters. Barrier removal consists in misusing a safety barrier that human operators are supposed to respect. The iterative learning formalism is based on human action formalism that interprets the barrier removal in terms of consequences, i.e. benefits, costs and potential dangers or deficits. Two functions are required: the similarity function to search a known case closed to the input case for which the human action has to be predicted and a reinforcement function to reinforce links between similar known cases. This reinforced iterative formalism is applied to a railway simulation from which the prediction of barrier removal is based on subjective data. [Copyright &y& Elsevier]

Abstract: This paper presents an off-line (finite time interval) and on-line learning direct adaptive neural controller for an unstable helicopter. The neural controller is designed to track pitch rate command signal generated using the reference model. A helicopter having a soft inplane four-bladed hingeless main rotor and a four-bladed tail rotor with conventional mechanical controls is used for the simulation studies. For the simulation study, a linearized helicopter model at different straight and level flight conditions is considered. A neural network with a linear filter architecture trained using backpropagation through time is used to approximate the control law. The controller network parameters are adapted using updated rules Lyapunov synthesis. The off-line trained (for finite time interval) network provides the necessary stability and tracking performance. The on-line learning is used to adapt the network under varying flight conditions. The on-line learning ability is demonstrated through parameter uncertainties. The performance of the proposed direct adaptive neural controller (DANC) is compared with feedback error learning neural controller (FENC). [Copyright &y& Elsevier]

Abstract: This paper deals with fluidized bed coating of metal substrates with high performance thermoplastic powders (polyftalamide, PPA). Two different experimental scenarios were investigated: the conventional hot dipping fluidized bed (CHDFB) process and the electrostatic fluidized bed (EFB) coating process. The preliminary experimental plan was scheduled employing design of experiment (DOE) technique. Three experimental factors and operative ranges large enough for practical purposes were considered in both of the examined scenarios. In particular, coating time and airflow rate were chosen as experimental factors in both CHDFB and EFB. The third factor was the preheating temperature of metal substrates in CHDFB and the applied voltage in EFB. A general linear model based upon analysis of variance (ANOVA) was used to evaluate the significance on coating processes of each experimental factor. Main effect plots (MEPs) and interaction plots (IPs) of coating thickness were drawn. Trends consistent with the settings of the operative parameters were displayed. The experimental trends were first modelled by numerical regression of the experimental data and, subsequently, by using artificial neural network. The reliability of the neural network solution and of the built ad hoc regression models was comparatively evaluated. Multi-layer perceptron (MLP) neural network trained with back propagation (BP) algorithm was found to be the most valuable in fitting the coating thicknesses trends for both the coating processes. Examining the developed models outside the operative ranges they were designed for, the good generalization capability and high flexibility of the neural network solution was definitely stated. [Copyright &y& Elsevier]