Nonlinear black-box modeling in system identification: a unified overview

A nonlinear black-box structure for a dynamical system is a model structure that is prepared to describe virtually any nonlinear dynamics. There has been considerable recent interest in this area, with structures based on neural networks, radial basis networks, wavelet networks and hinging hyperplanes, as well as wavelet-transform-based methods and models based on fuzzy sets and fuzzy rules. This paper describes all these approaches in a common framework, from a user's perspective. It focuses on what are the common features in the different approaches, the choices that have to be made and what considerations are relevant for a successful system-identification application of these techniques. It is pointed out that the nonlinear structures can be seen as a concatenation of a mapping form observed data to a regression vector and a nonlinear mapping from the regressor space to the output space. These mappings are discussed separately. The latter mapping is usually formed as a basis function expansion. The basis functions are typically formed from one simple scalar function, which is modified in terms of scale and location. The expansion from the scalar argument to the regressor space is achieved by a radial- or a ridge-type approach. Basic techniques for estimating the parameters in the structures are criterion minimization, as well as two-step procedures, where first the relevant basis functions are determined, using data, and then a linear least-squares step to determine the coordinates of the function approximation. A particular problem is to deal with the large number of potentially necessary parameters. This is handled by making the number of ‘used’ parameters considerably less than the number of ‘offered’ parameters, by regularization, shrinking, pruning or regressor selection.     

Variance analysis of identified linear MISO models having spatially correlated inputs,with application to parallel Hammerstein models

This contribution concerns variance analysis of linear multi-input single-output models when the inputs are temporally white but where different inputs may be correlated. An expression is provided for the variance of a linearly parametrized estimate of the frequency response function from one block, i.e. from one input to the output. In particular, this expression reveals that the variance increases in one block when the number of estimated parameters in another block is increased, but levels off when the number of parameters in the other block reaches the number of parameters in the block in question. It also quantifies exactly how correlation between inputs affects the resulting accuracy and a graphical representation is provided for this purpose. The results are applicable to parallel MISO Hammerstein models when the nonlinearities are known and generalize an existing variance expression for this type of model.     

Optimally robust system identification of systems subject toamplitude-bounded stochastic disturbances

In this paper it is shown that log cos(πx/(2C)) is the optimally robust criterion function for prediction error methods with respect to amplitude-bounded stochastic disturbances. This criterion function minimizes the maximum asymptotic covariance matrix of the parameter estimates for the family of innovations of the systems which are amplitude bounded by the constant C. Furthermore, the stochastic worst case performance of the estimate corresponding to the criterion function log cos(πx/(2C)) is better than the worst case performance of the least squares estimate even if the constant C is chosen larger than the actual amplitude bound on the innovations. In addition to its favorable properties in a stochastic setting, this criterion function also generates estimates which are unfalsified in a deterministic framework     

Hyperphagia modifies FA profiles of plasma phospholipids,plasma FFA,and adipose tissue TAG

Hyperphagia was achieved by continuous intracerebroventricular infusion of a melanocortin receptor antagonist (HS024; Neosystem, Strasbourg, France) in rats. The effects of hyperphagia on FA composition and concentration of plasma phospholipids (PL), plasma FFA, and adipose tissue TAG were studied in rats for 8 d [short-term hyperphagia (STH); n=8], or 28 d [longterm hyperphagia (LTH); n=9]. The control rats were treated with artificial cerebrospinal fluid for 8 d (n=8) or 28 d (n=10). The rats were fed the same regular diet. In STH rats the plasma PL and fasting plasma FFA contained higher concentrations of saturated FA (SFA) and monounsaturated FA (MUFA), and plasma FFA contained lower n−6 PUFA than in the control rats. In LTH rats the plasma PL contained higher concentrations of SFA, MUFA, and n−3 PUFA and higher proportions of 16∶1n−7 and 18∶1n−9 at the expense of 18∶2n−6 than in the control rats. In LTH rats the abundant dietary intake of 18∶2n−6 did not enrich 18∶2n−6 of the plasma PL or adipose tissue TAG. In LTH rats the fasting plasma FFA contained more than twofold higher concentrations of SFA and MUFA, and higher proportions of 16∶1n−7 and 18∶1n−9 at the expense of 18∶2n−6 than in the control rats. This animal obesity model shows that LTH affects the FA composition and concentration of plasma PL, plasma FFA, and adipose tissue TAG, a result consistent with changes associated with increased risk of various diseases in humans. These results also demonstrate that LTH alters the FA composition of plasma PL and adipose tissue TAG in a way that does not reflect the FA composition of dietary fat.     

Thermographic phosphors for thermometry: A survey of combustion applications

Temperature is a fundamental thermodynamic parameter used to describe physical, chemical and biological processes. In combustion as in many other applications, knowledge about temperature plays a substantial role in helping to maintain an efficient and clean environment. Being able to measure temperature accurately in combustion and in fire-related applications is important for giving a better understanding of heat transfer phenomena and improving existing models.In the present review paper a method based on the spectroscopy of inorganic luminescent materials is described and exemplified in experiments related to combustion. The method involves the use of thermographic phosphors which enable remote temperature diagnostics to be performed with a high degree of sensitivity and accuracy. The technique is superior to those based on thermocouples and pyrometry, particularly in the vicinity of flames and when the measured surface is subjected to random movements. Several phosphor materials suitable for temperature probing are described. The application of thermographic phosphors to temperature measurements in one-point and in two-dimensions in flame spread scenarios, and in pyrolysis experiments involving different construction materials and polymers are described. Many thermographic phosphors have the property of being insensitive to variations in pressure up to 1 GPa. This property extends the use and development of thermographic thermometry to other domains, such as internal combustion engines. The temperature has been measured in a point and in two-dimensions inside the combustion chamber. The complex procedures required to implement the use of thermocouples on moving objects inside an engine make thermocouples an expensive choice. It also limits the possibilities of altering the measurement locations and thereby also complicating the investigation of different engine geometries and components.Thermographic phosphors have also been employed in gas turbine applications. Temperature probing in the afterburner of a full-size aircraft engine is described with the aim to study the effects of various engine loads on the wall temperature. Furthermore, the application of thermographic phosphors to study the temperature of droplets in relation to sprays is described. In spray dynamics, temperature is a crucial parameter for gaining an understanding of atomisation, evaporation and heat convection from the surrounding gases. Finally the application of thermographic phosphors for gas temperature measurement by seeding the particles into a gas flow is described together with the challenges associated with seeding the particles for in-situ flame measurements.     

Seasonal effects on the physicochemical characteristics of fish sauce made from capelin (Mallotus villosus)

Fresh capelin (Mallotus villosus) was harvested from the North Atlantic during both summer and winter fishing seasons. Reaction conditions for fish sauce processing were optimized with respect to temperature, salt concentration and reaction time, using a response surface methodology (RSM) experimental design. Whole capelin was minced and samples were ground with increasing salt concentrations. RSM optimizations were conducted, ranging from 5% to 30% salt, and incubating at 5° intervals from 0 to 65 °C. Autolytic activity was estimated by extracting the liquid formed by the mixture with trichloroacetic acid and estimating protein content by the Lowry method. Samples for fish sauce production were then prepared under optimized conditions by mixing ground capelin with 10% salt and incubating at 50 °C for up to 270 days for the summer capelin and up to 360 days for the winter capelin. Samples were collected at regular intervals and analyzed for liquid yield, moisture, protein, soluble solids, specific gravity, pH, colour and amino acid content. Kjeldahl protein content in the fish sauce from summer capelin was 2.03% after 250 days of fermentation and twice as high as that in winter capelin fish sauce. Moisture content and pH were lower in the summer capelin fish sauce, but Brix and density were higher than those in fish sauce from winter capelin. Brown colour formation was very rapid in the summer capelin fish sauce but slow in the winter capelin fish sauce. Summer capelin may be successfully utilized for the production of fish sauce without added enzymes.     

White Electroluminescence by Supramolecular Control of Energy Transfer in Blends of Organic‐Soluble Encapsulated Polyfluorenes

Here, it is demonstrated that energy transfer in a blend of semiconducting polymers can be strongly reduced by non‐covalent encapsulation of one constituent, ensured by threading of the conjugated strands into functionalized cyclodextrins. Such macrocycles control the minimum intermolecular distance of chromophores with similar alignment, at the nanoscale, and therefore the relevant energy transfer rates, thus enabling fabrication of white‐light‐emitting diodes (CIE coordinates: x = 0.282, y = 0.336). In particular, white electroluminescence in a binary blend of a blue‐emitting, organic‐soluble rotaxane based on a polyfluorene derivative and the green‐emitting poly(9,9‐dioctylfluorene‐alt‐benzothiadiazole ( F8BT ) is achieved. Morphological and structural analyses by atomic force microscopy, fluorescence mapping, µ‐Raman, and fluorescence lifetime microscopy are used to complement optical and electroluminescence characterization, and to enable a deeper insight into the properties of the novel blend.     

Processing of Aluminum Dross: The Birth of a Closed Industrial Process

This is the history of a modern aluminum dross recycling company, from its beginnings in the last years of the twentieth century to the present day. The vision of the founders was to build a local recycling plant and take full responsibility for sensitive environmental issues by recycling aluminum dross locally rather than shipping it abroad. The paper tells the history of the company from the environmental perspective, and gives an overview of some of the challenges and the decisions that followed from this vision, for instance the selection of technology. The company developed a closed industrial process for the recycling of aluminum dross, and the paper discusses some of their laboratory experiments and industrial trials. An important milestone has now been reached as the process in its present form is recognized by the environmental authorities in the country. Furthermore, it seems realistic that in the near future the final product from this process will be comparable to the product delivered in the processing of salt cake in specialized chemical plants, but at a fraction of the cost.     

On the Influence of Crosslinker on Template Complexation in Molecularly Imprinted Polymers: A Computational Study of Prepolymerization Mixture Events with Correlations to Template-Polymer Recognition Behavior and NMR Spectroscopic Studies

Aspects of the molecular-level basis for the function of ethylene glycol dimethacrylate and trimethylolproprane trimethacrylate crosslinked methacrylic acid copolymers molecularly imprinted with (S)-propranolol have been studied using a series of all-component and all-atom molecular dynamics studies of the corresponding prepolymerization systems. The crosslinking agents were observed to contribute to template complexation, and the results were contrasted with previously reported template-recognition behavior of the corresponding polymers. Differences in the extent to which the two crosslinkers interacted with the functional monomer were identified, and correlations were made to polymer-ligand recognition behavior and the results of nuclear magnetic resonance spectroscopic studies studies. This study demonstrates the importance of considering the functional monomer–crosslinker interaction when designing molecularly imprinted polymers, and highlights the often neglected general contribution of crosslinker to determining the nature of molecularly imprinted polymer-template selectivity.     

Composite modeling of transfer functions

The problem under consideration is how to estimate the frequency function of a system and the associated estimation error when a set of possible model structures is given and then one of them is known to contain the true system. The “classical” solution to this problem is to, first, use a consistent model structure selection criterion to discard all but one single structure, second, estimate a model in this structure and, third, conditioned on the assumption that the chosen structure contains the true system, compute an estimate of the estimation error. For a finite data set, however, one cannot guarantee that the correct structure is chosen, and this “structural” uncertainty is lost in the previously mentioned approach. In this contribution a method is developed that combines the frequency function estimates and the estimation errors from all possible structures into a joint estimate and estimation error. Hence, this approach bypasses the structure selection problem. This is accomplished by employing a Bayesian setting. Special attention is given to the choice of priors. With this approach it is possible to benefit from a priori information about the frequency function even though the model structure is unknown