Nonlinear Finite Impulse Response Model Research Articles

Fouling Detection in Industrial Heat Exchanger Using Number of Transfer Units Method, Neural Network, and Nonlinear Finite Impulse Response Models

Due to reduced energy efficiency and productivity loss caused by fouled heat exchangers in industrial plants there is an obvious need for detection of fouling formation. Based on continuously measured temperatures and flow rates collected from the refinery plant history database, a semi-empirical number of transfer units (NTU) model and neural network-based models are developed. In order to confirm the reliability of proposed fouling factor calculation, the entire procedure was performed by developing a dynamic nonlinear finite impulse response (NFIR) model. Developed models are intended for fouling detection for industrial shell and tube heat exchangers. The performance criteria of developed models together with residual monitoring indicate that not only neural networks but the NTU and NFIR models effectively detect fouling formation.

Modelling the Influence of Soil Properties on Crop Yields Using a Non-Linear NFIR Model and Laboratory Data

This paper introduces a new non-linear correlation analysis method based on a non-linear finite impulse response (NFIR) model to study and quantify the effects of ten soil properties on crop yield. Two versions of the NFIR model were implemented: NFIR-LN, accounting for both the linear and non-linear variability in the system, and NFIR-L, accounting for linear variability only. The performance of the NFIR models was compared with a non-linear random forest (RF) model, to predict oilseed rape (2013) and wheat (2014) yields in one field at Premslin, Germany. The ten soil properties were used as system inputs, whereas crop yield was the system output. Results demonstrated that the individual and total contribution of the soil properties on crop yield varied throughout the different cropping seasons, weather conditions, and crops. Both the NFIR-LN and RF models outperformed the NFIR-L model and explained up to 55.62% and 50.66% of the yield variation for years 2013 and 2014, respectively. The NFIR-LN and RF models performed equally during yield prediction, although the NFIR-LN model provided more consistent results through the two cropping seasons. Higher phosphorus and potassium contributions to the yield were calculated with the NFIR-LN model, suggesting this method outperforms the RF model.

Nonlinear Finite Impulse Response Estimation using Regularized Neural Networks

Abstract This work presents a new regularization scheme for identifying nonlinear finite impulse response (NFIR) models using artificial neural networks (ANN). Prior knowledge, such as the exponentially decaying nature of an impulse response, is included during the identification using a regularization approach inspired on the well-known regularized linear finite impulse response identification literature. More specifically the sensitivity of the modeled output with respect to the delayed input of the NFIR model is penalized to provide an exponentially decaying prior. The proposed method is illustrated and compared to other ANN regularization schemes on a simulation example.

Optimization-Oriented High Fidelity NFIR Models for Estimating Indicated Torque in Diesel Engines

In this paper, optimization-oriented high fidelity indicated torque models which cover the whole operating regions under both steady-state and transient cycles for heavy-duty vehicles are developed. Two different experiments are performed and their data are merged to be utilized in the training of the models. In the first experiment, all combustion input channels are excited by quadratic chirp signals with different sweeps in their frequency profiles. Different from the first experiment, the engine speed is excited by ramp-hold signals in the second experiment. The estimations of friction, pumping and inertia torques in addition to the torque measured from the engine dynamometer are utilized in the indicated torque calculations. In order to model the calculated indicated torque, a nonlinear finite impulse response (NFIR) model with a single layer sigmoid neural network has been designed. A sensitivity analysis is performed by generating several models with different number of input regressors and neurons. Experimental results show that the majority of the models in a selected wide range of the model parameters are validated with fit accuracies higher than 90 % and 85 % on the World Harmonized Stationary Cycle (WHSC) and the World Harmonic Transient Cycle (WHTC), respectively.

Imaging of Nonlinear and Dynamic Functional Brain Connectivity Based on EEG Recordings With the Application on the Diagnosis of Alzheimer's Disease.

Since age is the most significant risk factor for the development of Alzheimer's disease (AD), it is important to understand the effect of normal ageing on brain network characteristics before we can accurately diagnose the condition based on information derived from resting state electroencephalogram (EEG) recordings, aiming to detect brain network disruption. This article proposes a novel brain functional connectivity imaging method, particularly targeting the contribution of nonlinear dynamics of functional connectivity, on distinguishing participants with AD from healthy controls (HC). We describe a parametric method established upon a Nonlinear Finite Impulse Response model, and a revised orthogonal least squares algorithm used to estimate the linear, nonlinear and combined connectivity between any two EEG channels without fitting a full model. This approach, where linear and non-linear interactions and their spatial distribution and dynamics can be estimated independently, offered us the means to dissect the dynamic brain network disruption in AD from a new perspective and to gain some insight into the dynamic behaviour of brain networks in two age groups (above and below 70) with normal cognitive function. Although linear and stationary connectivity dominates the classification contributions, quantitative results have demonstrated that nonlinear and dynamic connectivity can significantly improve the classification accuracy, barring the group of participants below the age of 70, for resting state EEG recorded during eyes open. The developed approach is generic and can be used as a powerful tool to examine brain network characteristics and disruption in a user friendly and systematic way.

Low-order Nonlinear Finite-Impulse Response Soft Sensors for Ionic Electroactive Actuators Based on Deep Learning

This paper introduces a soft sensor (SS) for the estimation of the deflection of a polymeric mechanical actuator. The actuator is based on ionic polymer-metal composites (IPMCs). Applications of IPMCs have been proposed in fields such as robotics, surgery, and aerospace, to mention the most interesting ones. In such application fields, both the complexity and the size of the actuating system are of chief importance. An SS can be, therefore, preferred to hardware measuring the actuator output, for estimating the actuator motion. Also, low-order models are of interest to limit the computational load, which can be a constraint in real-time applications. To this aim, several data-driven nonlinear finite-impulse response (NFIR) models have been investigated. Data, used for the model identification, have been acquired, in controlled environmental conditions, by using swept signals as the input to the IPMC actuator. Linear and nonlinear models, based on principal component analysis, shallow, and deep neural networks (NNs), have been investigated, for different model orders. The best results have been obtained by an SS based on a fifth-order NFIR model, implemented by a deep belief NN.

Comparing Models of Spontaneous Variations, Maneuvers and Indexes to Assess Dynamic Cerebral Autoregulation.

We analyzed the performance of linear and nonlinear models to assess dynamic cerebral autoregulation (dCA) from spontaneous variations in healthy subjects and compared it with the use of two known maneuvers to abruptly change arterial blood pressure (BP): thigh cuffs and sit-to-stand. Cerebral blood flow velocity and BP were measured simultaneously at rest and while the maneuvers were performed in 20 healthy subjects. To analyze the spontaneous variations, we implemented two types of models using support vector machine (SVM): linear and nonlinear finite impulse response models. The classic autoregulation index (ARI) and the more recently proposed model-free ARI (mfARI) were used as measures of dCA. An ANOVA analysis was applied to compare the different methods and the coefficient of variation was calculated to evaluate their variability. There are differences between indexes, but not between models and maneuvers. The mfARI index with the sit-to-stand maneuver shows the least variability. Support vector machine modeling of spontaneous variation with the mfARI index could be used for the assessment of dCA as an alternative to maneuvers to introduce large BP fluctuations.

Tracking Nonlinear Correlation for Complex Dynamic Systems Using a Windowed Error Reduction Ratio Method

Studying complex dynamic systems is usually very challenging due to limited prior knowledge and high complexity of relationships between interconnected components. Current methods either are like a “black box” that is difficult to understand and relate back to the underlying system or have limited universality and applicability due to too many assumptions. This paper proposes a time-varying Nonlinear Finite Impulse Response model to estimate the multiple features of correlation among measurements including direction, strength, significance, latency, correlation type, and nonlinearity. The dynamic behaviours of correlation are tracked through a sliding window approach based on the Blackman window rather than the simple truncation by a Rectangular window. This method is particularly useful for a system that has very little prior knowledge and the interaction between measurements is nonlinear, time-varying, rapidly changing, or of short duration. Simulation results suggest that the proposed tracking approach significantly reduces the sensitivity of correlation estimation against the window size. Such a method will improve the applicability and robustness of correlation analysis for complex systems. A real application to environmental changing data demonstrates the potential of the proposed method by revealing and characterising hidden information contained within measurements, which is usually “invisible” for conventional methods.

Fault detection in multi-sensor networks based on multivariate time-series models and orthogonal transformations

We introduce the usage of multivariate orthogonal space transformations and vectorized time-series models in combination with data-driven system identification models to achieve an enhanced performance of residual-based fault detection in condition monitoring systems equipped with multi-sensor networks. Neither time-consuming annotated samples nor fault patterns/models need to be available, as our approach is solely based on on-line recorded data streams. The system identification step acts as a fusion operation by searching for relations and dependencies between sensor channels measuring the state of system variables. We therefore apply three different vectorized time-series variants: (i) non-linear finite impulse response models (NFIR) relying only on the lagged input variables, (ii) non-linear output error models (NOE), also including the lags of the own predictions and (iii) non-linear Box–Jenkins models (NBJ) which include the lags of the predictions errors as well. The use of multivariate orthogonal space transformations allows to produce more compact and accurate models due to an integrated dimensionality (noise) reduction step. Fault detection is conducted based on finding anomalies (untypical occurrences) in the temporal residual signal in incremental manner. Our experimental results achieved on four real-world condition monitoring scenarios employing multi-sensor network systems demonstrate that the Receiver Operating Characteristic (ROC) curves are improved over those ones achieved with native static models (w/o lags, w/o transformations) by about 20–30%.

Dynamic gray‐box modeling for on‐line monitoring of polymer extrusion viscosity

AbstractMelt viscosity is a key indicator of product quality in polymer extrusion processes. However, real time monitoring and control of viscosity is difficult to achieve. In this article, a novel “soft sensor” approach based on dynamic gray‐box modeling is proposed. The soft sensor involves a nonlinear finite impulse response model with adaptable linear parameters for real‐time prediction of the melt viscosity based on the process inputs; the model output is then used as an input of a model with a simple‐fixed structure to predict the barrel pressure which can be measured online. Finally, the predicted pressure is compared to the measured value and the corresponding error is used as a feedback signal to correct the viscosity estimate. This novel feedback structure enables the online adaptability of the viscosity model in response to modeling errors and disturbances, hence producing a reliable viscosity estimate. The experimental results on different material/die/extruder confirm the effectiveness of the proposed “soft sensor” method based on dynamic gray‐box modeling for real‐time monitoring and control of polymer extrusion processes. POLYM. ENG. SCI., 2012. © 2012 Society of Plastics Engineers

Model Order Selection of N-FIR Models by the Analysis of Variance Method

Identification of nonlinear finite impulse response (N-FIR) models is studied. In particular the selection of model structure, Le., to find the contributing input time lags, is examined. A common method, exhaustive search among models with all possible combinations of the input time lags, has some undesired drawbacks, as a tendency that the minimization algorithm gets stuck in local minima and heavy computations. To avoid these drawbacks it is needed to know the model structure prior to identifying a model.In this contribution it is shown by experiment that a statistical method, the analysis of variance, is a good alternative to exhaustive search in the identification of the structure of nonlinear FIR-models.It is possible to reduce the risks of getting an erroneous model structure due to the non-convexity of the minimization problems, reduce the computation time needed and also get a good estimate of how far the fit of the desired model can be enhanced.