Delft Hydraulics Laboratory Research Articles

Experimental investigation of TARMAX model for modeling of hydrodynamic forces on cylinder-like structures

A new approach that models lift and drag hydrodynamic force signals operating over cylindrical structures was developed and validated. This approach is based on stochastic auto regressive moving average with exogenous (ARMAX) input and its time-varying form, TARMAX. Model structure selection and parameter estimation were discussed while considering the validation stage. In this paper, the cylindrical structure was considered as a dynamic system with an incoming water wave and resulting forces as the input and outputs, respectively. The experimental data, used in this study, were collected from a full-scale rough vertical cylinder at the Delft Hydraulics Laboratory. The practicality of the proposed method and also its efficiency in structural modeling were demonstrated through applying two hydrodynamic force components. For this purpose, an ARMAX model is first used to capture the dynamics of the process, relating in-line forces provided by water waves; secondly, the TARMAX model was applied to modeling and analysis of the lift forces on the cylinder. The evaluation of the lift force by the TARMAX model shows the model is successful in modeling the force from the surface elevation.

Simulating Wave Reflection Using Radiation Boundary

A finite element wave propagation model for linear periodic waves in a coastal sea region is developed. The model includes refraction, diffraction, and reflection of gravity waves on water over arbitrary bathymetry. The authors discuss the use of the computer model in simulating waves using a number of classical examples involving a circular pile, submerged shoal, and breakwater. The method of solution involves complex potential theory. The equation used by Berkhoff (1976; Mathematical Models for Simple Harmonic Linear Water Waves—Wave Diffraction and Refraction. Delft: Delft Hydraulics Laboratory, Delft University of Technology, p. 111) and the authors within the domain is elliptic in type allowing wave trains to cross, thereby producing amphidromic points. An amphidromic point is the two-dimensional version of a node in a one-dimensional standing wave caused by imperfect reflection or wave train interference. Radiating and partially reflecting boundaries are modelled by the authors, using a parabolic equation developed in different ways by Radder (1979; On the parabolic equation method for water-wave propagation, Journal of Fluid Mechanics, 95, 159–176) and Booij (1981; Gravity waves on water with non-uniform depth and current. Delft: Delft Hydraulics Laboratory, Delft University of Technology, p. 127), allowing the passage of energy through a boundary over arbitrary bathymetry. Radder (1979) and Booij (1981) develop this equation in the domain as an alternative to the elliptic equation. Berkhoff (1976) uses a downstream radiation boundary condition based on Hankel functions for the shoal problem, valid only in constant depth. The upstream boundary condition of Berkhoff (1976) for the same shoal problem is derived using the wave ray method. The limitation of the wave ray method is that for general purposes the rays frequently cross, resulting in no solution. The method used by the authors has the advantage of simplicity in that the boundary conditions are very simple to implement but none of the physical features are lost.

Cross-shore distribution of longshore sediment transport: comparison between predictive formulas and field measurements

The skill of six well-known formulas developed for calculating the longshore sediment transport rate was evaluated in the present study. Formulas proposed by Bijker [Bijker, E.W., 1967. Some considerations about scales for coastal models with movable bed. Delft Hydraulics Laboratory, Publication 50, Delft, The Netherlands; Journal of the Waterways, Harbors and Coastal Engineering Division, 97 (4) (1971) 687.], Engelund–Hansen [Engelund, F., Hansen, E., 1967. A Monograph On Sediment Transport in Alluvial Streams. Teknisk Forlag, Copenhagen, Denmark], Ackers–White [Journal of Hydraulics Division, 99 (1) (1973) 2041], Bailard–Inman [Journal of Geophysical Research, 86 (C3) (1981) 2035], Van Rijn [Journal of Hydraulic Division, 110 (10) (1984) 1431; 110(11) (1984) 1613; 110(12) (1984) 1733], and Watanabe [Watanabe, A., 1992. Total rate and distribution of longshore sand transport. Proceedings of the 23rd Coastal Engineering Conference, ASCE, 2528–2541] were investigated because they are commonly employed in engineering studies to calculate the time-averaged net sediment transport rate in the surf zone. The predictive capability of these six formulas was examined by comparison to detailed, high-quality data on hydrodynamics and sediment transport from Duck, NC, collected during the DUCK85, SUPERDUCK, and SANDYDUCK field data collection projects. Measured hydrodynamics were employed as much as possible to reduce uncertainties in the calculations, and all formulas were applied with standard coefficient values without calibration to the data sets. Overall, the Van Rijn formula was found to yield the most reliable predictions over the range of swell and storm conditions covered by the field data set. The Engelund–Hansen formula worked reasonably well, although with large scatter for the storm cases, whereas the Bailard–Inman formula systematically overestimated the swell cases and underestimated the storm cases. The formulas by Watanabe and Ackers–White produced satisfactory results for most cases, although the former overestimated the transport rates for swell cases and the latter yielded considerable scatter for storm cases. Finally, the Bijker formula systematically overestimated the transport rates for all cases. It should be pointed out that the coefficient values in most of the employed formulas were based primarily on data from the laboratory or from the river environment. Thus, re-calibration of the coefficient values by reference to field data from the surf zone is expected to improve their predictive capability, although the limited amount of high-quality field data available at present makes it difficult to obtain values that would be applicable to a wide range of wave and beach conditions.

Free and bound capillary waves as microwave scatterers: laboratory studies

Laboratory measurements of the fine space-time structure of short gravity-capillary waves, as well as Ku-band scattering at grazing and moderate incidence from wind waves in the large Delft Hydraulics Laboratory wind-wave channel are presented. This study was stimulated by the need to verify the processes that significantly contribute to scattering at grazing and moderate incidence. A scanning laser slope gauge was used for measuring capillary waves from 2-mm to 2-cm wavelengths and frequencies ranging up to 100 Hz. A dual-polarized (vertical, VV, and horizontal, HH), coherent, pulsed Ku-band scatterometer with good temporal resolution (3 ns) was used to obtain simultaneous Doppler spectra and the absolute cross section of scattered signals for grazing angles 6 and 25/spl deg/ and for winds in the range 2.5-12.5 m/s. Two-dimensional (2D) filtering and bispectral analyses were used to separate and study the influence of free and bound surface waves. The results of this study demonstrate that the frequency-wavenumber spectra of capillary waves consist of two parts. The first corresponds to free capillary waves, which satisfy the dispersion relationship. The second corresponds to bound parasitic capillary waves, which are located near the crests of steep wind waves. The phase velocity of these capillary waves is approximately equal to the phase velocity of the steep waves. Measurements of the Doppler frequency of the scattered signals show that the Doppler spectra also have a bimodal structure. While the first low-frequency part of the spectrum corresponds to the Bragg scattering from the free capillary waves, the highfrequency part is associated with Bragg scattering from the bound capillary waves on the crests of the steep waves. This type of scattering is predominant for the upwind direction of illumination (especially for HH-polarization).

Numerical model of the turbulent wave boundary layer induced by finite amplitude water waves

The characteristics of the mean flow are presented by considering the closure scheme based on the one equation model in the turbulent boundary layer. Stokes' second order irrotational velocity profile is adopted at the upper boundary as the boundary condition along with Batchelor's condition for the streaming; a one-layer time varying viscosity profile is applied to give a good prediction of the velocity distributions in the turbulent wave boundary layer in comparison with the experimental data of Van Doorn (Rep. M1, 423, Delft Hydraulics Laboratory, 1981). Finally, the calculation is carried out based on various modes of velocity and turbulent energy. The mean velocity is found to differ from condition to condition of the relative depth and roughness. The mean velocity is found to differ from condition to condition of the relative depth and roughness. The mean velocity of the laminar boundary layer is always positive but the mean velocity of the turbulent boundary layer played a very complicated role.

Hydrodynamic forces on vertical cylinders and the lighthill correction

In the keynote address to the 1979 Behaviour of Offshore Structures (BOSS) Conference, Sir James Lighthill pointed out the absence of a second-order term of potential origin from the Morison description of the hydrodynamic force on a vertical cylinder. This term, referred to as the Lighthill correction, is due to the nonlinear interaction between the flow velocity and its horizontal gradient. As noted by Lighthill, if this term is omitted, the estimated drag force in the Morison equation in equal in effect to the actual drag force plus the Lighthill correction. Thus, it would appear that in cases where hydrodynamic damping plays an important role and should therefore be estimated as accurately as possible, corrections of the Lighthill type might have to be added to the Morison expression for the hydrodynamic force. (One such case is the dynamic amplification of wind-induced fluctuating motions of tension leg platforms.) In particular, it might be expected that the estimation of the damping force would be more strongly affected in situations involving low Keulegan-Carpenter numbers, and therefore relatively low damping forces. It is thus of interest to examine the effect of the Lighthill correction quantitatively. In this work, the expression for the Lighthill correction was derived for finite water depths. Measurements obtained in periodic wave flow at the Naval Civil Engineering Laboratory and in random wave flow at the Delft Hydraulics Laboratory were subjected to an extensive analysis. The results of the analysis showed that for both the periodic and random wave conditions the addition of the Lighthill correction (1) did not improve the Morison equation significantly, and (2) had no significant effect on the estimation of the drag force, including the drag force corresponding to very low Keulegan-Carpenter numbers.

Evaluation and improvement of hyrological concepts used in PAWN

In the years 1976–1983 the PAWN-study (Police Analysis for the Water-management in The Netherlands) was carried out in order to assess data for a national water management policy as reported in the Nota Waterhuishouding 1984, submitted to the Dutch parliament in the framework of the preparation of the Water Management Act. Among the mathematical models used in PAWN, DEMGEN (DEMand GENerator) represents agriculture, by far the greatest interest in terms of water demand and economic importance. By simulating the hydrological cycle in agricultural areas the model produces data on sprinkling demand, total water demand, agricultural damage and the changes in groundwater levels due to water suppletion. Furthermore the model evaluates sprinkling costs, which enables cost/benefit analyses for water supply plans. After the PAWN exercises for various reasons the need was felt for a critical evaluation of DEMGEN, specifically with regard tro future applications. This also implied an updating of DEMGEN on basis of recent scientific developments. In the period 1982–1986 investigations were carried the following fields: • - evaluation and improvement of implemented physical concepts; • - development and implementation of new concepts; • - updating of input files and facilities with respect to physical characteristics; • - updating of the areal schematization of The Netherlands and the development of relevant standard procedures for this subdivision The paper deals with the results with respect to the follwoing physical concepts in DEMGEN: • - Unsaturated flow • - Evapotranspiration • - Drainage • - Irrigation Contributions were made by Rijkswaterstaat, the Delft Hydraulics Laboratory (DHL) and the Institute for Land and Water Management Research (ICW). Most of the studies dealt with separate concepts, whereas Rijkswaterstaat also carried out an integral sensitivity analysis and a validation of DEMGEN on basis of data from two experimental catchments in The Netherlands. In all investigations the sensitivity of certain assumptions and the estimation of reliable model parameters was the major topic, rather than the assesment of the highest possible degree of accuracy in the simulation of physical processes.

Breakwater Engineering

Breakwaters have been built for thousands of years to protect harbours and coasts against the waves. Because of their very nature, breakwaters are exposed to strong wave attack. During recent years a number of newly constructed breakwaters have been severely damaged: Sines in Portugal in the winters of 1978 and 1979; Arzew-el-Djedid in Algeria in 1980; Tripoli, Libya, in beginning of 1981; Gioja Tauro in Italy in 1980; San Ciprian, Spain, during the winter of 1980. These recent examples of breakwater failure are not unique: there have frequently been reports of storm damage. One can even say that a rubble-mound breakwater is designed to absorb a certain amount of damage during its lifetime. What now alarmed the world of designers and builders of such structures, however, was the unprecedented increase in the number of failures, the fact that the breakwaters were either under construction or had been just completed and also the extent of damage in each of the above cases, the total damage being estimated as a thousand million US dollars.So far, no overall picture has been presented. All failures were individually investigated. The investigation reports left few issues unexamined, but remain confidential themselves. The Delft Hydraulics Laboratory participated in some of the damage investigations and was entrusted with the hydraulic investigations for the rehabilitation of all five breakwaters. A unique chance, therefore, existed for the development of an overall picture. Without affecting the confidentiality of the individual cases, the most important deficiencies in the design of rubble-mound breakwaters are treated in this paper. A plan is made for a design approach in which hydraulic, structural and geotechnical aspects are integrated. The state of the art of each of these different aspects is presented globally, emphasizing the interactions between them and the consequences of an integral approach for the design methods.

Prediction of 2‐D Bed Topography in Rivers

Mathematical models are being developed at the Delft Hydraulics Laboratory for the computation of time‐dependent, two‐dimensional bed deformation in alluvial rivers. A most promising verification of the predictive capacity of one such model has been obtained for the Waal River in the Netherlands. This model was developed for rivers with dominant bed load transport without significant grain‐sorting effects and with approximately constant width. Also, some results of computations with variable boundary conditions are given which simulate the effect of the river regime and the extraction and return of cooling water on the bed topography.

Johannes Theodoor Thijsse 1893–1984

On April 30, 1984, Professor Doctor Engineer J.T. Thijsse, professor emeritus of the Delft University of Technology, the Netherlands, passed away at the age of 91.Johannes T. Thijsse was born in Amsterdam, the Netherlands, in 1893, the son of a well‐known Dutch biologist. He graduated from the Delft University of Technology in 1917 with a degree in civil engineering. In 1927, he became the first director of the newly formed Delft Hydraulics Laboratory, a function he continued until 1960. In 1936, Thijsse received a lectureship in hydraulics at the Delft University, to be followed in 1938 by his appointment as Professor Extraordinary and in 1946 as ordinary professor.

The cyclone pump

Pumping abrasive solids such as sand always leads to damage to the pumping unit. As a contribution to the solution of this problem a modified jet pump is introduced here based on the hydrocyclone principle. For this reason the device has been called the “Cyclone Pump”. The drive is achieved by an exterior pump. The output of the cyclone pump with respect to the driving system is rather low (20-30%). Consequently application is only recommended for cases in which energy costs are not significant, for instance when the pump is a small part of a large system. In the Delft Hydraulics Laboratory the cyclone pump has been successfully applied to the continuous pumping of sand-water mixtures for a sand flume and river scale models with movable bed. In this paper the flow pattern in the cyclone pump and its separation power i.e., its ability to separate solids from the flow, are treated qualitatively. In addition, the results of some preliminary experiments are given.

EXPERIMENTS ON NATURAL AND CHEMICAL DISPERSION OF OIL IN LABORATORY AND FIELD CIRCUMSTANCES

ABSTRACT At the Delft Hydraulics Laboratory a laboratory flume has been constructed to aid research into natural and chemically induced dispersion processes as well as to test the effectiveness of dispersant in specific conditions. The flume allows the selection of variables and conditions, for instance, the generation of nonbreaking and breaking waves and currents, the variation of temperature, salinity, oil layer thickness, dispersant spray droplet size, and the droplet impact on the slick surface. The flume has been verified with empirical data gathered from an extensive sea survey on the natural and chemical dispersion of a number of oil slicks. The field experiments on natural dispersion can be modeled satisfactorily in the flume with respect to the formation of oil droplets from the oil slick and the initial intrusion in the water column. The further mixing of oil droplets in the water mass are to be calculated from diffusion theories and other transport processes. Field experiments on the natural and chemical dispersion of 10 artificial oil spills in the North Sea led to the remarkable conclusion that spraying an oil slick with chemical dispersant did not enhance the (natural) dispersion process, while a premixed dispersant in oil was very effective. The ineffectiveness of sprayed dispersant could not be explained from a limited series of experiments performed in the laboratory flume on the effects of evaporation, photo-oxidation, emulsification, and layer thickness on natural and chemical dispersion.

Bed deformation in curved alluvial channels

Recent research activities at the Delft Hydraulics Laboratory and the Delft University of Technology have increased the understanding of the large-scale bed deformation in alluvial rivers with relatively stable banks. In this paper the most important results of these activities are given. First bed deformation in river bends is explained in the light of the results of some laboratory experiments in curved flumes with fixed banks. The deformation is then explained in terms of wave length and damping with a linear analysis of the water and sediment motion for the steady state. From this analysis the conclusion is drawn that the point bar height and pool depth in bends cannot be predicted solely from local conditions. A significant part of the lateral bed slope is due to an overshoot effect induced by the redistribution of the water and sediment motion in the first part of the bend. Finally the results of computations with a non-linear two-dimensional model, simulating three curved flume experiments, are dis...

THE COMPUTATION OF BED SHEAR IN A NUMERICAL MODEL

Obstructions located in coastal and offshore waters usually disturb the natural flow pattern. This disturbed flow will, in general, cause local morphological changes in the position of the erodable boundary. Often these changes should not be allowed to exceed certain limits, for example, when local scour around an offshore construction may endanger foundations. Local morphological changes result from changes in the local sediment balance, brought about by the flow disturbance. In the present paper a mathematical model is described which gives the bottom shear stresses and the configuration of the seabed around an obstruction using a computation of the two dimensional turbulent flow field. The obstruction considered is a submarine pipeline laid uncovered on a seabed consisting of non-cohesive sediment. A research project on the local scour near submarine pipelines is being carried out at the Delft University of Technology. Part of the project is the application and extension of an advanced numerical flow model for scour development near pipelines on the seabed exposed to current action. This work is being carried out in cooperation with the Delft Hydraulics Laboratory. The code of the flow model has been developed in a joint venture between the Delft Hydraulics Laboratory and the Laboratoire National d"Hydraulique in France. The turbulent flow field is computed taking into account the influence of turbulence generated at the bed and by the pipe. The bed shear stresses are assumed to play the key role in the interaction between the flow and the seabed. In the computer model the bed shear is related to the flow through the "law of the wall". The model operation is schematized in the diagram below in which the first loop represents the evolution of the velocity field through a series of hydraulic time steps. After the velocity field is stabilized, in the second loop one morphological time step can be used for the computation of the local seabed changes. In this second loop the computed bed shear is applied together with a sediment transport formula. After the morphological time step a new bed topography is obtained and a new grid is generated for the next flow computation.

LOW FREQUENCY OSCILLATIONS ON THE DUTCH COAST

During several months in 1981 and 1982/1983 two extensive programs of field measurements were carried out at Egmond on the Dutch North Sea coast, one during the spring of 1981 and the other during the winter of 1982/83. In these programs researchers of the Dutch "Rijkswaterstaat", of The Delft University of Technology and of the Delft Hydraulics Laboratory cooperated in the planning and execution of the measurements and are now involved in data analysis and evaluation of results. During the year 1983 the senior author spent a sabbatical year with the Rijkswaterstaat in the Netherlands and joined the team effort in analyzing the data. Reference is made to other papers presented at this conference, highlighting some aspect of this research program: Van Heteren and Stive (1984), Derks and Stive (1984). The present paper deals with low frequency oscillations in the surfzone and their potential impact on certain aspects of coastal morphology. The conclusions of this paper have to be considered as tentative and should be followed by further study. Its main intention is to point to certain possible relationships regarding coastal morphology so that in future measurements the program can be designed in such a way that certain relationships can clearly and unmistakenly be identified. The programs of measurements carried out in 1981 and 1982/83 were primarily designed to evaluate the usefulness and accuracy of various types of instruments in the surfzone. The measurements were never designed for the use we have made of them and results may therefore not be fully conclusive and thus subject to discussion. Nevertheless the authors found their tentative findings of sufficient interest to share them with their colleagues.

CALIBRATION AND ADJUSTMENT PROCEDURES FOR THE RHINE-MEUSE ESTUARY SCALE MODEL

At the Delft Hydraulics Laboratory the hydraulic model of the Rhine- Meuse estuary has been rebuilt and extended after being destroyed by fire in 1979. The new model has been operational since the summer of 1982 for water quantity and water quality research. The present paper deals with the calibration and adjustment procedures for this hydraulic model. Successively the (physical) backgrounds for the calibration process, the techniques used for provision of boundary conditions and the procedures for adjusting flows and salinity distribution are discussed. Finally, the results of calibration and verification and the experiences gained on the employed techniques will be analysed.

WATERQUANTITY AND -QUALITY RESEARCH FOR THE RHINE MEUSE ESTUARY

In commission of the Dutch Government (Ministry of Public Works), the Delft Hydraulics Laboratory carries out a long term research programme on the water quantity and water quality of the Rhine-Meuse estuary. The research programme has to provide the necessary hydraulic data as Input for a management decision-making process. The programme is based on an "integrated approach philosophy" with respect to the various subjects to be studied and the research tools to be used (data from nature, hydraulic model, mathematical models and fundamental research). The paper presents the essentials and backgrounds of the research programme and discusses some first results.

WAVE FORCES AND IMPACTS ON A CIRCULAR AND SQUARE CAISSON

The existing breakwater at Civitavecchia Harbour, Italy, will be extended with a part of caissons. In order to investigate wave forces for this caisson, the Ministero del Lavori Public! commissioned the Delft Hydraulics Laboratory to perform a model investigation on the proposed extension. In the first part of the study quasi-static wave forces on the entire caisson were measured by means of a force measuring frame. In the second part wave impacts were recorded by pressure cells at 22 positions of the caisson. The investigation contained two main items which are discussed in this paper: Comparison of quasi-static wave forces with maximum forces by wave impacts. Comparison of forces and impacts for a circular caisson with those measured for a square caisson.

Decomposition of the mass transport in narrow estuaries

The mass transport as measured in the Rotterdam Waterway Estuary and in the tidal salinity flume of the Delft Hydraulics Laboratory is analysed to obtain insight into the magnitude of the dispersive mass transport in narrow estuaries. It is found that the magnitude of the dispersive mass transport in the real time one-dimensional advection-diffusion equation varies considerably over the tidal cycle, that the ratio between the magnitude of the dispersive mass transport by the net vertical circulation and by the vertical oscillatory shear in the tidally averaged one-dimensional advection-diffusion equation decreases when the degree of stratification in the flume is decreased, and that the dispersive mass transport in the Rotterdam Waterway and in the tidal salinity flume is affected significantly by the mixing processes at sea near the estuary mouth.

A simulation model of production, seasonal changes in biomass and distribution of eelgrass (Zostera marina) in Lake Grevelingen

A simulation model has been described, based on data from Lake Grevelingen, The Netherlands, as a predictive tool for lake management. The model has been developed as part of a large-scale aquatic modelling effort in Lake Grevelingen, carried out in close cooperation with the Delft Hydraulics Laboratory and the Delta Department, Environmental Research Division of the Ministry of Transport and Public Works. Available data on growth rates per unit eelgrass biomass, obtained with the leaf-marking technique, and on above- en below ground biomass and shoot density changes per unit area have been used. A space limitation depending on density of the above ground biomass and a growth limitation due to shortage in below ground biomass have been introduced. The seasonal changes in eelgrass production, both above and below ground, have been simulated as functions of the external forcing variables light, water temperature, wind generated water movements and of the internal control variables due to aging of the plant material. The vertical distribution of eelgrass can be partly explained from the modelling results on space, light and below ground biomass limitations. From the shore down to about 1 m waterdepth the above ground eelgrass biomass suffers from space limitation. Between 1 and 2 m production and biomass reach maximum values. Between 2 and 3 m waterdepth the above ground eelgrass growth is limited by the availability of below ground biomass. Between 3 and 5 m waterdepth both below ground biomass and light are the growth-limiting factors. Below 5 m waterdepth light is not sufficient to sustain net growth of eelgrass from rhizomes. Together with additional data — not used in the model — on seed production and growth of eelgrass shoots from seeds the vertical and horizontal distribution of the dominant macrophyte in the lake can be explained.