Sediment Identification for Geotechnics by Marine Acoustics

Informazioni relative al progetto

SIGMA

ID dell’accordo di sovvenzione: MAS3970100

Progetto chiuso

Data di avvio 1 Settembre 1997

Data di completamento 30 Novembre 2000

Finanziato da

• Specific programme of research and technological development in the field of marine science and technology, 1994-1998

Costo totale Nessun dato

Contributo UE Nessun dato

Coordinato da

Vrije Universiteit Brussel
Belgium

Risultati finali

Development and realization of a hydrophone streamer and positioning system

The hydrophone streamer itself has been designed and manufactured at the beginning of the project. Electronic equipment (amplifiers, filters, 48 channels) was required for processing the signals coming from the individual hydrophones before storage on-board the vessel. This equipment has been designed and manufactured in time for the 1st, 2nd and 3rd sea-trials. The positioning system is composed, on the one hand, of 3 transmitters (acoustical base, and the associated electronics) installed on the back of the tow-fish, and, on the other hand, of hardware and software required for processing the acoustical positioning signals. This equipment has been designed and manufactured in time for the 1st, 2nd and 3rd sea-trials. A. Geometry: 1) Generalities: The geometry (structure and spacing of the hydrophones) has been defined in taking into account the various parameters of both the SPA (Steerable Parametric Array) and the water depths considered for the experiments. The array is composed of 3 parts with different spatial sampling: i) Front part (16 m long) includes 9 hydrophones separated by 2 meters. ii) Middle part (16 m long) includes 32 hydrophones separated by 0.5 meter. iii) End part is identical to front part. This structure leads to a 48 meters array including 49 hydrophones. The 49-hydrophone signals will be digitised and stored on-board during the experiments. 2) Constant array offset thanks to SPA steering: The hydrophone array should be towed by the fish with variable offsets, but it is possible to maintain the offset constant by using the steering availability of the SPA. This steering will allow to make convergence of the bottom echoes onto the middle part of the hydrophone array whatever water depth. 3) Variation of grazing angle: In case of requirement of variation of grazing angle, front part and end part of the hydrophone array will be available for collecting echoes. B. Acoustical data: 1) Characteristics of the hydrophones: The 49 hydrophones are identical: -Wide bandwidth frequency response (10 Hz-40 kHz). -Receiving sensitivity-171 dB (re 1V/Pa) with 20 dB preamplifier. -Maximum working depth 300m. The 49 individual signals will be digitised and stored on board. 2) Obtaining a sampled array: The spatial sampling is optimised to provide a middle part being a properly sampled receiving array up to 1.5 kHz (wavelength about 1m, Nyquist criterion). The secondary frequencies of the SPA should be 3-6 kHz, so the middle part is not correctly sampled for this bandwidth. Nevertheless, some calculations have shown that ambiguous lobes will appear at 90 degrees (at 3 kHz) and at 30 and 90 degrees (at 6 kHz). The influence of these grating lobes should be reduced-using geometrical considerations- thanks to the SPA directivity; and by means of a time windowing processing. C. Positioning data: The positioning of the hydrophone array relative to the tow-fish will be achieved in real time on-board through a hybrid measurement:- short baseline acoustical positioning; depth measurement. The short baseline will be composed of 4 transmitters mounted on the tow-fish tail and several receivers which will be actually some of those of the hydrophone array. Several hydrophones will be used for both positioning and acoustical data. These hydrophones are not fixed in order to adjust their number and their position according to the requirements of navigation accuracy and real-time processing power. The position of each navigation hydrophone (relative to the transmitters on the tow-fish tail) will be processed using matched filtering and correlation. Furthermore, two depth sensors will be added to the end and front part of the hydrophone array. The depth information will be integrated into the hydrophone array positioning software. This should allow cancelling possible ambiguities. D. Streamer array positioning: The transmission/reception system used for the tests was the same than those used at sea. The transmission system is composed of 3 signals generators, followed by 3 linear power amplifiers. The source level has been decreased compared to sea trials, and is about 177 dB for each transmitter. The reception is achieved through a matched filtering stage (which can be bypassed), filters and amplifiers, and the echoes are digitised on a PC board. In addition to standard low frequency hydrophones, the array included high frequency hydrophones, which have been be used for acoustical positioning of the array. A special purpose device was developed which included two transmitters fixed on the ship and which transmitted suitable signals for the array positioning. The received high frequency signals have been processed in real time in order to provide with the angular displacement of the array in a horizontal plane. In addition to this device, several depth sensors provided with the immersion of the positioning hydrophones. The total system provided with an accurate estimation of the array distortions.

Development of and computer code for acoustical inversion technique based on system identification

Within the framework of determining the characteristics of sediments on an acoustic base, a global system identification approach was used for an experimental validation of the wave propagation through sediments and for the determination of its acoustical parameters. In order to achieve this goal, the Maximum Likelihood Estimator in the frequency domain was put to use in several SISO and MIMO representations. Measurements were carried out on fine and medium sorted sands, on viscoelastic materials simulating the seafloor and on a Silicone containing glass particles. For the measurements on real sediments, two configurations were utilized: a water-Plexiglas-Sand-Plexiglas-water and a water-sediment-Plexiglas configuration. The propagation of the multiples in the sediment, and the calibration method, which is incorporated in the global system identification approach, are important assets that are not exploited when carrying out direct measurements of dispersion or absorption with transducers buried in the sediment. The smallest model errors were found with the viscoelastic rational form model, followed by the Buckinghams and viscoelastic CQ-model, which were giving acceptable errors too. This confirms the presence of an intrinsic attenuation almost linear in the frequency for the fine sediments. On the other hand, bigger model errors were found with Biots model, which can be explained by the many parameters that are still unknown and from which some of them need to be fixed in the inverse procedure. For the medium sorted sand the dynamic multi-scattering model of Waterman-Truell was applied. The high-frequency set-up at sea was simulated. The measurements on fine sands in normal and oblique incidence were combined in MIMO sense to estimate the RMS-roughness of the surface, the longitudinal parameters and the frequency independent shear velocity.

Realisation of an improvement high resolution system

The result relates to the use of a chirp source vibrator. Such a source lead to a full control of the transmitted signal and to its spectral content. It was mainly used for P wave generation. The chirp generation was associated with real-time pulse compression in order to validate the concept during the trials. The task improved the receiving side of the seismics first by increasing the number of the receivers in the array for both P and S waves and then by investigating beam forming and array processing in order to increase the system penetration and accuracy.Two different set-ups for high resolution seismic on the seabottom have been tested and operated during the first and second sea trials respetively. The elements of the seismic system operated during the sea trials is given here below: 1)Underwater elements: -1 x streamer. -Dragged on the sea bottom, 8 x 3 receivers (see description above). -1 x airgun, Bolt airgun, volume range 10-80 cubic inches. -1 x sled, weight in water approx. 1 ton. 2)Hosting: -Seismic source, winch to winch in and winch out the streamer. -Hydraulic power pack to power the winch. -Electronics to handle the hydraulic power pack, the seabed-to-surface connections. -Electronics to handle the winch in / winch out / shot / record sequence. -Pressure sensor checking the compression of the air in the airgun. 3) Seabed-to-surface connections: -1 x 500m long umbilical comprising 1x air hose, data lines (upward and downward flows), power lines to the hydraulic power pack (600V, 3-10 Amps). -1 winch fed with umbilical. -1 traction cable to the sled (fed on 1 of the Belgicas winch) 4) Surface equipment: -1 x seismic recorder with QC software. -1 x firing box. -1 x power pack and power handling system. -1 x Shark software to handle the acquisition sequence and visualise all sensors other than seismic sensors. -1 x compressor and 24 x compressed air bottles to continuously feed the airgun. The acquisition sequence was defined as follows: -Step 1: Winch streamer out during x seconds until stabilisation of gimbal mounted geophones. -Step 2: Shot and record before end of winch out phase. -Step 3: End of record and winch in phase. -Step 4: End of winch in phase, Sailing, all instruments quiet, back to winch streamer out phase.

Designing and development of the SIGMA database and Geographic Information System (GIS)

The SIGMA relational database developed in this project provided the necessary facilities to perform correlation between the collected data, obtained from the different marine instruments. Attention was paid to the selection of the platform and the software for the database and GIS used, as well as to the shells, which were developed on top of the core of the database, to enable fast and easy storage and access of the parameters. Compatibility with the existing firmware of the partners was reached as well as transportability of the database. The navigation data of the vessels used in the sea trials as well the position of the acoustic transmitter-receiver elements have been integrated in the SIGMA database and GIS to assure accurate estimates of the positions where investigations of the sea bed were made. The same holds for the geophysical measurements. The physical recovered sea bed parameters (from coring and acoustic analysis) have been used to represent graphically the nature of the investigated marine environments. The thematic maps derived from spatial analysis in the GIS are benefits for advanced visualization and decision-support systems, for research, monitoring and sustainable management activities of the marine ecosystems. Furthermore the structure of the SIGMA database and GIS (i.e. empty with respect to data inputs, but with the structure of the database included) could be used in future by third parties for other marine survey programs. Designing and development of the SIGMA database and Geographic Information System (GIS): 1) The SIGMA relational database was needed for storing all non-graphic information directed by the necessary to reach the goal in the project. It implies: i) Determining data sources (e.g.: historical data, derived from previous cruises; doctoral thesis and scientific publications;- Measured data from the estimated bottom parameters, sea trials, results from tank experiments; coring results). ii) Identification and definition of parameters that are the most useful for sea bed identification and characterisation; their normalisation (type, format, dimensions, and interrelations of the data to be stored , and also elimination of ambiguous attribute and repeating groups of attributes, avoiding anomalies), their storage, and access. iii) Acquisition or collection of all pertinent data relative to the test sites. iv) Their storage and access. The SIGMA database objects such as Tables, Queries, sub-queries, forms and sub-forms, and reports, have been implemented during the development of the database in defining attributes and manipulating the attribute values in conjunction with the G.I.S. software packages, the Intergraph Modular GIS Environment (MGE). Two database management systems (DBMS) were chosen and implemented for use in this project: the Oracle relational database and the Microsoft Access database. Both software systems run on Windows NT operating systems. These software allowed easy storage and access of the historical and measured data from sea experiments. Before storing all these data in the designed SIGMA database, quality control and formatting processes were carried out. Flat file data or RDBMS files other than the selected RDMS were converted to Oracle or Microsoft Access tables. 2) The SIGMA GISOne of the major components of the SIGMA project was the development of a geographic information system (GIS) to analyse and integrate a wide range of acoustic, geophysical, geotechnical, geological, and sedimentological data related to seafloor environment in order to determine the sea bed characteristics. The surveys, conducted in selected representative areas have lead to the constitution of a well controlled database on sea bottom sediment characteristics. The system aided also in the acquisition, management, verification, analysis, synthesis, and interpretation of information produced and used in the project. The current state of the SIGMA GIS allows to implement a complete representation of the graphic information by means of the following associations: i) Point feature when measured data are associated to a sampling point or station in space. ii) Linear (line) feature when values are associated to a line on the x,y plane. iii) Area or polygon feature when measured data are associated to a geographical area. These graphic elements contain at least two linkages, one linkage for the feature record (feature name and feature characteristics-feature type, and feature symbology) in the feature table and one linkage for the corresponding attribute record in the SIGMA relational database attribute table. A variety of spatial queries can be performed to explore possible relationships between the different parameters identified as having a significant effect on sea bed identification and characterisation, and to model their effects in the area of interest by selecting one or many layers interactively. An understanding of such relationship is essential for interpreting marine environmental impact. The results of these queries can then be displayed graphically as thematic maps so that one can observe the exact location of the specific sea beds. Indeed, a set of thematic maps on marine sea bed characteristics can be carried out based mainly on the above analytical operations. Thus, modelling operations of the area of interest may be carried out either directly on thematic map layers (using the class values) or indirectly on tables of attributes linked to spatial objects, or on both combinations of both maps and their attributes.