Damage detection in offshore structures using neural networks

This paper discusses the damage detection in offshore jacket platforms subjected to random loads using a combined method of random decrement signature and neural networks. The random decrement technique is used to extract the free decay of the structure from its online response while the structure is in service. The free decay and its time derivative are used as input for a neural network. The output of the neural network is used as an index for damage detection. It has been shown that function N is effective in damage detection in the members of an offshore structure. Experimental studies conducted on a reduced model for a real jacket structure with geometrical scale of 1:30 are used. The applied loads were random loads. Two different load spectra were used: White noise, and Pierson-Moskowitz.     

Non-trading,lexicographic and inconsistent behaviour in stated choice data

This paper discusses a number of issues relating to the pre-analysis and cleaning of stated choice data, where we look specifically at the problems caused by non-trading, lexicographic and inconsistent response patterns. We argue that this process is considerably more complex and challenging than many in the field have hitherto acknowledged, with the standard practice being the use of rather ad hoc procedures for the identification of these phenomena. A detailed analysis on four different stated choice datasets highlights the potential impacts of these methods on model estimation results.     

Seasonality of picophytoplankton chlorophyll a and biomass in the central Cantabrian Sea, southern Bay of Biscay

Seasonal changes in the abundance and biomass of cyanobacteria (Synechococcus and Prochlorococcus) and picoeukaryotes were studied by flow cytometry in the upper layers of the central Cantabrian Sea continental shelf, from April 2002 to April 2006. The study area displayed the typical hydrographic conditions of temperate coastal zones. A marked seasonality of the relative contribution of prokaryotes and eukaryotes was found. While cyanobacteria were generally more abundant for most of the year (up to 2.4 10⁵ cells mL^− 1), picoeukaryotes dominated the community (up to 10⁴ cells mL^− 1) from February to May. The disappearance of Prochlorococcus from spring through summer is likely related to shifts in the prevailing current regime. The maximum total abundance of picophytoplankton was consistently found in late summer–early autumn. Mean photic-layer picoplanktonic chlorophyll a ranged from 0.06 to 0.53 µg L^− 1 with a relatively high mean contribution to total values (33 ± 2% SE), showing maxima around autumn and minima in spring. Biomass (range 0.58–40.16 mg C m^− 3) was generally dominated by picoeukaryotes (mean ± SE, 4.28 ± 0.27 mg C m^− 3) with an average contribution of cyanobacteria of 30 ± 2%. Different seasonality of pigment and biomass values resulted in a clear temporal pattern of picophytoplanktonic carbon to chlorophyll a ratio, which ranged from 10 (winter) to 140 (summer). This study highlights the important contribution of picoplanktonic chlorophyll a and carbon biomass in this coastal ecosystem.     

Full vehicle simulation using thermomechanically coupled hybrid neural network shock absorber model

In the case of full vehicle models, the technique of multi-body simulation (MBS) is frequently used to study their highly non-linear dynamic behaviour. Many non-linearities in vehicle models are induced by force elements like springs, shock absorbers, bushings and tires. Commonly, spline functions are used to represent the force responses of these components. If the non-linear relationships are more complicated, the spline approximations are no more accurate. An alternative approach is based on empirical neural networks which are based on the mathematical approximation of measured data. It is well known that neural networks are able to represent and predict complex component responses accurately. The aim of this paper is to perform a dynamic full vehicle simulation using a thermomechanically coupled hybrid neural network shock absorber model. In this shock absorber model, the spline approach is combined with a temperature-dependent neural network. Based on a displacement-controlled excitation on a four post test rig in the ADAMS/Car MBS software, a rugged test track is simulated. In this way, the front and rear shock absorbers are dynamically loaded with comfort-relevant frequencies in the range of 0.75-30 Hz and velocity amplitudes up to 2 m/s. By the simulation, stability of the hybrid neural network model is demonstrated. Furthermore, the damping force, the vertical acceleration of the chassis and the required simulation times are compared. The standard spline approach is used as a reference.     

A hybrid model for better predicting capabilities under parallel wind conditions for NOx concentrations from highways in Ireland

Atmospheric dispersion models based on the Gaussian line source formulation tend to give poor results for low wind speed and parallel wind directions. Certain dispersion models like HIWAY-2 have an inbuilt dispersion algorithm to adjust in such situations but no correction exists for mathematical based Gaussian dispersion models. We evaluate the performance of a hybrid model a mathematical model, to assess its suitability as an alternative to the Gaussian based models.     

Assessment of diesel combustion noise overall level in transient operation

Combustion noise in passenger cars powered with direct injection (DI) diesel engines is frequently the main reason why end-users are reluctant to drive this type of vehicle. Thus, the great potential of diesel engines for environment preservation — due to their lower CO₂ emissions — could be missed. This situation worsens with the current design trends (engine downsizing) and the emerging new diesel combustion concepts (Homogeneous Charge Compression Ignition-HCCI, Premixed Charge Compression Ignition-PCCI, etc.), which are intrinsically noisy. This negative feature can be even more critical in transient operation due to the contribution of the temporal changes of both source and transmission path on engine noise. Therefore, combustion noise must be considered as an additional essential factor in engine development, together with performance, emissions and driveability. Thus, suitable evaluation procedures that can be integrated into the global engine development process in a timely and cost-effective manner are imperative. Regarding the evaluation procedures, most of the work available in the literature addressed combustion noise at steady operation. To surpass this limitation, two possible approaches — adapted from the classical and multiple regression methods — for the overall level assessment of combustion noise in transient conditions are evaluated in this paper.     

Condition monitoring of wooden railway sleepers

Wooden railway sleeper inspections in Sweden are currently performed manually by a human operator; such inspections are to large extent based on visual analysis. In this paper a machine vision based approach has been considered to emulate the visual abilities of the human operator to enable automation of the process. Digital images from either ends (left and right) of the sleepers have been acquired. A pattern recognition approach has been adopted to classify the condition of the sleeper into classes (good or bad) and thereby achieve automation. Appropriate image analysis techniques were applied and relevant features such as the number of cracks on a sleeper, average length and width of the crack and the condition of the metal plate were determined. Feature fusion has been proposed in order to integrate the features obtained from each end for the classification task which follows. The effect of using classifiers like multi-layer perceptron and support vector machines has been tested and compared. Results obtained from the experiments show that multi-layer perceptron and support vector machines have achieved encouraging results, with a classification accuracy of 90%; thereby exhibiting a competitive performance when compared to a human operator.     

Adapting the light · biomass (BZI) models of phytoplankton primary production to shallow marine ecosystems

Several authors have reported a strong linear relationship between daily phytoplankton production and the product of chlorophyll biomass, photic depth, and incident irradiance for a variety of estuaries. This “light · biomass” (BZ_pI_o) formulation has been proposed as an alternative to traditional mechanistic approaches for computing phytoplankton production in numerical estuarine models. One limitation to their application in shallow systems is that the BZ_pI_o models have been developed in relatively deep estuaries where light does not reach the bottom. We propose a nonlinear correction factor to adapt the BZ_pI_o relationship to shallow systems where light does reach the bottom. Our function takes into account variations in incident irradiance, attenuation coefficient for light, photosynthetic efficiency, and maximum rate of photosynthesis. A series of correction polynomials are proposed for various ranges of incident irradiance, and are integrated into a single multiple polynomial which applies across all irradiance levels. Our new correction factor was tested against a ¹⁴C-based productivity dataset from shallow stations in Narragansett Bay, RI and an O₂-based dataset from shallow (1.1 m) lagoon mesocosms at the University of Rhode Island. Results showed that our polynomials accurately correct BZ_pI_o-predicted rates of production in shallow water columns. Application of our correction factor to a series of shallow water productivity datasets from the literature together with theoretical calculations show how significant the shallow water correction can be, especially in very shallow water columns with low turbidity.     

Parking policies in large cities in Germany

This article deals with highly motorized large West German cities of about 200,000 inhabitants and more, which usually provide reasonable public transport systems. Illegal parking with shares of about 40 to 50% of the total parking is widespread in the parking problem areas of those cities, especially in the inner-city residential and mixed-use areas. Parking spaces are demanded by residents, employees, customers and visitors, and by delivery and service traffic. The different characteristics of parking demands by different user groups are discussed. The total parking supply consists of public and private spaces. The share of private spaces is about 40 to 50% of the total parking spaces in German cities. The amount of car traffic generated by a parking space depends on parking duration and parking turnover, as well as on search traffic. So the change of a space from long-duration use of an employee to short-duration of customers — as often discussed in parking concepts — generates at least five-fold car traffic. The measurements and effects of parking control of public spaces as well as the parking regulations in zoning ordinances, restrictions on the construction of new private parking spaces and park-and-ride are discussed. Finally, a parking concept methodology — using the example of Frankfurt am Main — is discussed.     

Biologically induced differences in erodibility and aggregation of subtidal and intertidal sediments: a possible cause for seasonal changes in sediment deposition

This study was carried out to describe the difference in erodibility and aggregation in a tidal basin including both subtidal and intertidal study sites and to use these results to explain the shifting erosion/deposition cycles at the sites. Erosion thresholds, erosion rates and settling velocities of the eroded material were measured at a mudflat transect and at sediment cores taken from a nearby tidal channel during surveys made in May 2000 and March 2002. Surface samples were analysed for grain-size, chl. a content, faecal pellet content, dry bulk density and organic content. Additionally, surface samples were taken at eight occasions in the period January 2002 to May 2003 from shallow tidal channels in the area. These samples were analysed for mud content and showed that major shifts in sediment distribution occurred in the period. The erodibility of the mudflat was generally high due to pelletization by the mudsnail Hydrobia ulvae but close to the salt marsh much lower erodibility was found, probably due to stabilisation by microphytobenthos. In contrast, the erodibility of the channel bed seemed to be very little influenced by biological activity and the relatively low erodibility found here was caused by physical characteristics of the sediment. The sediment eroded from the mudflat was generally strongly pelletized and showed high settling velocities whereas less aggregation and lower settling velocities were found for the channel bed sediments. Temporal variations of the mudflat stability and hydrodynamics resulted in temporal variations of deposition and erosion and the changing stability at the mudflat is likely to be one of the main reasons for a general transport of fine-grained sediment from the mudflat to the channel in the cold seasons and vice versa during the rest of the year.