The 11th International Conference on Computer and IT Applications in the Maritime Industries (COMPIT) will be held from 16 to 18 April 2012 in the Belgian city of Liège.
First held in the year 2000, COMPIT has established itself as a leading conference in the information technology (IT) field[ds_preview] for the maritime industries, bringing together software developers and users. The conference covers the life cycle of ships, offshore structures and equipment, from design to operation. In addition, the conference serves as a contact platform for recruiting and the preparation of international R&D projects. Traditionally, most participants come from industry, reflecting the practical relevance of the event. Selected papers for COMPIT 2012 are discussed below, illustrating some of the general trends in IT for the maritime industries.
Main trends in short
The quest for energy efficiency goes on. This continuing factor is driving the development of a variety of IT applications.
In design, Computational Fluid Dynamics (CFD) is a key technology. Technology leaders couple CFD to formal optimization to create fuel efficient designs for hulls, propulsion improving devices and propellers. In operation, several companies are spearheading developments in monitoring and decision support tools, with a focus on fuel efficiency and emissions compliance.
The use of risk-based assessment in IMO regulations and classification rules, in terms of both design and operational procedures, grows. Corresponding IT tools facilitate the application of risk-based analyses.
Improvements in the autonomous operation of maritime robots occur. Advances in individual and swarm intelligence are opening new applications in surveying, hull cleaning and search tasks, in the offshore, oceanographic and naval fields.
Ship design without computers is no longer imaginable. IT supports the naval architect in every facet of the design process, from the initial concept to the detailed design. In the early design stages, fast and experience-based tools are needed, as little information is available and quick responses are required. At later design stages, more information is available and higher accuracy is required in evaluations. This constellation has promoted the development of different tools for different design stages.
The efficient sharing of information between the various tools has always been a problem, from the very beginnings of using computers in ship design. Here, there are basically two approaches: either one attempts to make the transition from one software to another largely automatic, for example by developing algorithms that extrude 2D models »intelligently« to 3D models, or one consolidates the software portfolio, opening the door to 3D modelling right from the start.
Over the decades, CAD (Computer Aided Design) has progressed from mere electronic drawing to a central design platform. This has gone hand in hand with the trend of employing 3D models of ship hulls and machinery. Modern product data models (PDMs) combine geometric information with other product data (such as material, tolerances, suppliers, etc.). PDMs are the starting point for a multitude of simulation options and support virtual prototyping. They also reflect the general trend towards distributed, concurrent work with frequent updates, where increasingly information is exchanged via the internet. All of the major CAD vendors support this trend and are presenting their latest innovations at COMPIT.
While PDMs are firmly established in shipbuilding, they have yet to develop in ship operation. Here, changes are being driven, in particular, by environmental compliance. It is expected that prototype applications will eventually also lead to the wider application of PDMs in ship operation. Some avant-garde applications give a glimpse of what the future may bring.
Risk-based assessment of designs and operational procedures are a general trend in IMO regulations and classification rules. The basic idea of risk-based design is that the designer is given the freedom to do »whatever he wants« as long as the resulting design meets a given safety level. Often, the safety level is indirectly prescribed by requiring at least an equivalent safety level as provided by conventional »standard« designs. For example, in cruise ship designs, larger fire zones (increasing risk) can be compensated for by better fire detection and fire fighting equipment (decreasing risk). Already widely used in other industries, most notably the aerospace and nuclear engineering, and offshore engineering, the risk-based approach is now playing an increasingly important role also in ship design and ship operation.
This approach is particularly important when we face completely new design or operational challenges. IT tools for risk-based assessment facilitate the application of risk-based analyses for design and operational procedures. Some examples of this approach at COMPIT 2012 include: Angus Grandison et al. (University College London) present »The Development of Modelling Methods and Interface Tools Supporting a Risk Based Approach to Fire Safety«, as applied in passenger vessels.
Two other papers focus on new operational challenges, namely offshore wind parks and shipping in the Northern Sea Route. Jean-David Caprace et al. (University of Liège) present progress »Toward a Risk Based Simulation for the Erection of an Offshore Windmill Field«, combining discrete event simulation, risk-based methods and formal optimization. Stein Ove Erikstad and Sören Ehlers, both professors at the NTNU in Trondheim, present
»A Risk Based Decision Support Framework for Exploiting Northern Sea Route Transport Opportunities«, which looks at the various uncertainties involved in arctic operation and deriving recommendations for two different scenarios (fixed schedule and fixed speed).
Often the combination of various software techniques opens the door for new applications or enhances existing software. There are several papers which illustrate this approach, referred to by a variety of different names, including »hybrid computing« and »synergistic computing«. Danilo Calcagni et al. (INSEAN) describe in »Automated Marine Propeller Design Combining Hydrodynamics Models and Neural Networks« how complex and expensive simulations for propellers can be condensed into simple »meta models« for fast early design loops. Traditionally, naval architects have resorted to standard series like the Wageningen B series. These series, however, are of limited value, as propeller shapes have evolved over time (for example towards larger skew) and special propellers have very different geometries. Today, tailored series can be created using numerical simulations. In the INSEAN approach, by systematic variation of selected blade shape parameters, a database of performance parameters can be generated through recursive calculations using an in-house propeller flow simulation code. This results in a virtual systematic series that can be tailored for a particular class of propellers and vessel. Artificial neural nets, a widely used technique for pattern recognition, then interpolate efficiently for new designs. This gives a quasi instant response, making the approach very interesting for early design.
Decisions in ship design, production and operation are increasingly based on simulations. As a consequence, a plethora of simulation methods have evolved. A common problem here is the effort required in model generation, which drives project time and costs. Ideally, model generation is fully automatic, allowing user-friendly integration in design systems and formal optimization. Finite element analysis (FEA) has started to replace simpler methods used traditionally for structural analysis in initial ship design. This has only become possible thanks to faster and more intelligent model generation, automating many steps that were previously performed manually. Ahmet Tasdemir and Serkan Nohut (Zirve University) show in »Practical Experience with Efficient Generation of Finite-Element Models of Ships using POSEIDON« that specific tools for ship structures allow much faster model generation than general purpose grid generation tools. In the reported cases (car carrier and multi-purpose vessel), POSEIDON enables finite-element models to be generated approximately ten times faster. George Korbetis and Dimitris Georgoulas (BETA VAE Systems) present in »Multi-Objective Design Optimization of a Rudder, using Automated CAE Model Set-Up with ANSA Pre-Processor« an application for the optimization of an appendage, in this case a spade rudder. The automated CAE model set-up capability of ANSA pre-processor is used to define CAE models for static, contact and CFD analyses. ANSA and the ETA post-processor, which is exploited for the extraction of responses from the solver results, are coupled to the modeFRONTIER optimising shell for the identification of the best rudder shape.
Virtual Reality (VR) offers a multitude of potential applications in ship design, ship construction and ship operation. At the bottom of Virtual Reality lies a 3D computer model of »the world« (for example the ship or shipyard). Most applications allow stereoscopic vision and interaction with »the world«, for example fly-through capability or the ability to move objects around. Audio and tactile elements are more advanced aspects of Virtual Reality and are often omitted in marine applications. Verónica Alonso and Tronstad Reidar (SENER) present the latest VR capabilities of the FORAN system, pointing out the »Advantages of using a Virtual Reality Tool in Shipbuilding«. Some of these advantages lie in the quick evaluation of modifications and design changes, checking ease of access in production, operation and maintenance, training of crews and last but not least commercial and marketing purposes. Many VR applications for ships involve objects moving in the virtual world, for example ships in a seaway or cranes. Here the challenge is to find the right compromise between realistic physical behaviour and an »instant« response for the user. Benjamin Mesing et al. (Fraunhofer Institute) illustrate this fundamental problem for a particular application, namely disembarking passengers with the help of davits in »Virtual Prototyping of Davits with Parameterized Simulation Models and Virtual Reality«. Hermann Lödding et al. (TU Hamburg-Harburg) are »Coupling Virtual Reality and Physics Engines for Ships« to create fast and appropriate behaviour in moving objects in their VR application. Physics engines are often used in computer games to generate realistic behaviour for coupling with Virtual Reality. Because of the connection to games, these engines have a focus on running simulations in real time, which is also very important for Virtual Reality, because the effects are needed dynamically during the session runtime. To achieve real-time computations, the accuracy is usually lower than in other multi-body simulations, but adequate for realistic behaviour in most of the examined scenarios.
The quest for energy efficiency has fuelled interest in simulation methods for the main energy consumers on ships. Foremost, this concerns the resistance and propulsion
of ships. Computational Fluid Dynamics (CFD) is a key technology in this respect. The trend here is moving towards numerical propulsion tests at full scale conditions, which offers more realistic results than model tests, particularly for propulsion improvement devices such as nozzles, ducts and fins. The effect of propulsion improving devices is best assessed using CFD. Ganbo Deng et al. (Ecole Centrale de Nantes) and Benoit Mallol (Numeca) present one of the most advanced examples of a numerical propulsion test in »Computation of Free-Surface Viscous Flows around Self-Propelled Ships with the Help of Sliding Grids«. The key to success lies here in adaptive grids, which during the simulation get finer where needed (for accuracy) and coarser where possible (for computational efficiency). Such adaptive grid techniques are widely seen as a key technology to make CFD computations faster, which is already important for individual analyses, but vital for formal optimization. Sing-Kwan Lee et al. present in »Propeller Energy Loss Reutilizing for Full Form Ship Propulsion« results from a cooperation project between ABS and the Chinese ship design company SDARI. For a full hull form, first CFD simulations for the bare hull are performed. Through the detailed CFD flow field solution, not only can the energy loss due to propeller-induced axial and rotational velocities be evaluated, but the best potential energy recovering area for locating appropriate energy-saving device can also be indentified. This approach is still largely manual. Claus Abt el al. (Friendship System) present in »Domain Preparation for RANSE Simulations about Assemblies of Functional Surfaces« an approach that points towards the next level of design support. Their method supports the grid generation for complex
assemblies in ships, as illustrated for the assembly of a hull with a duct, a propeller, and a rudder. The method automatically copes with changes in shape by means of parametric modelling, which opens the door for formal optimization. Felipe Ruggeri et al. (University Sao Paulo) present an application that demonstrates how this approach may be more widely used in the future. In »Parametric Model and CFD Integrated Process to WED Optimization« they apply the Friendship Framework and the CFD code STAR ccm+ of CD-adapco to optimize a wake equalizing duct, considering cavitation and vibrations as constraints.
Hull optimization remains a key topic at COMPIT. There is less progress in optimization algorithms in this field, as evolutionary algorithms dominate the field – due to the ease with which they allow parallel computing. Alongside the »classic« genetic algorithms, swarm algorithms are now enjoying increasing popularity. Swarm algorithms mimic the strategies of bee or ant swarms. The current focus is on advanced industry applications, combining commercial optimization shells (most notably the Friendship Framework and modeFRONTIER), advanced simulation software and parametric CAD models. The fuel crunch has stimulated research into optimization of ship hulls and propellers. Karsten Hochkirch and Volker Bertram (FutureShip) give an overview of the historical development of ship hull optimization and extrapolate into the next decade in »Ship Hull Optimization – Past, Present and Future«. They point out that the term »optimization« is frequently used, but may denote different levels of sophistication and associated fuel saving potential. The paper discusses the state of the art in computers, optimization strategies (algorithms), CFD tools, and practical applications. Even for the technology leader, FutureShip, development is far from finished and further fuel savings may be realised by considering propeller, hull and appendages together in optimization projects, the added resistance in waves, and wider spectra of actual operational conditions (speed, load conditions, etc.). Fabian Pécot et al. (SIREHNA) illustrate the current state of the art in hull optimization in their paper »Shape Optimization of a Monohull Fishing Vessel«. The application combines simpler and faster CFD tools with more accurate, computationally expensive tools and looks at both resistance and seakeeping performance to obtain a design with significantly lower resistance reduction and reasonable seakeeping behaviour. Wei Qiu (Memorial University of Newfoundland) and Jan Land et al. (Friendship Systems) present a similar application, combining seakeeping and resistance aspects, albeit for an offshore supply vessel, and employing different software, in »Seakeeping Optimization for Design and Operation«.
Not surprisingly, we also see very dynamic software developments for more efficient ship operation. One of the challenges in this field is that modern ships generate a flood of data for assorted specific applications, so much that we need computers to handle the flood of data created by other computers and filter the information intelligently for us. To get to grips with this, Wiegand Grafe and Sven Matho (Germanischer Lloyd) present »One Model to Run Them All«, a ship-specific product lifecycle management concept which compiles and manages all of the ship information generated during a vessel’s lifecycle. Germanischer Lloyd has developed a generic and modularized data model, which is combined with a central service bus that contains all of the business logic that was previously part of assorted software tools. Maritime software tools for various technical and operational purposes work on this central product data model. All of the data received by one application, as well as the business logic, are open to all other applications. The underlying principles are described by demonstrating GL MachineryManager (software tool for condition based maintenance) and GL EmissionManager (tool for Energy Efficiency Operational
Indicator determination).
There is a multitude of fuel saving options in ship operation. Diesel-electric configurations, waste heat recovery, trim optimization, operational profiles, and alternative fuels represent some of the many different possibilities which are open to the operator. This flexibility and the wide range of potential options to choose from does come with a downside however, growing system complexity. Jukka Ignatius (ABB) presents in »EMMA – Ship Energy Manager« a holistic energy management and optimization solution for ship operation. The integrated solution presented gives decision support to both the users onboard and also the management ashore. Progress in sensor technology, data collection systems and satellite internet connections have opened new possibilities for collecting extensive datasets of performance-related information from vessels. Analyzing this performance-related data is challenging because the performance of a ship is affected by many complex factors such as wind, waves, currents, and shallow water. Often the ship’s operating condition can have a significant impact on performance as well, for example varied displacements in operation. Risto Kariranta and Esa Henttinen (NAPA) discuss these problems and IT solutions for performance monitoring for ships in »Utilization of Performance Data Collection and Normalization for Ship Operations and Design«. The results from performance monitoring systems are often ignored, both due to data contamination by environmental conditions and because too much superfluous data and too little pertinent information is displayed. Only smart filtering and visualization allows insight into complex systems and large amounts of data. Pekka Autere and Aatos Heikkinen (Eniram) present in »Data Presentation and Reporting Techniques for Decision Making« a performance monitoring system based on real-time data collection onboard the ship.
Hull condition assessment (surveying) and hull cleaning are tasks that traditionally rely on the intervention of human experts. Could robots partially, largely, or completely perform such tasks? The safe answer is »partially – for the time being«, but we can speculate as to how far technology might progress in the decades to come. Robots are in high demand for offshore, oceanographic and military applications and technology is rapidly advancing in this field. These advances are taking place along two broad fronts, on the one, the cognitive skills (the »intelligence«) of robots is developing, while on the other researchers focus on using swarms of robots, which can perform tasks in parallel. Andrea Caiti (University of Pisa) is among the leading authorities on this field in Europe. In »Underwater Robots: Past, Present and Future« Caiti et al. give an overview of the development of marine robotics. The paper highlights some success stories of the recent past, but also points out difficulties that explain why progress in underwater robotics has been slower than in many other fields of robotics. Current challenges include underwater communication, data fusion, autonomous decision and distributed cooperation.
The European MINOAS project focuses on Marine Inspection Robotic Assistant System(s) and proposes the re-engineering of the overall ship inspection methodology. The concept integrates human personnel with highly mobile robots, effectively »tele-porting« the human inspector from the vessel’s hold to a control room with virtual reality properties. Swarms of robots can then investigate different parts of the ship in parallel, reporting problematic findings to the human inspector for further processing. Within the project, various prototypes of mobile robots have been developed. Two papers, Marco Bibuli et al. (ISSIA Institute of Intelligent Systems for Automation from the National Research Council CNR of Italy): »MARC – Magnetic Autonomous Robotic Crawler Development and Exploitation in the MINOAS Project«, and Kristin Fondahl et al. (German Research Center for Artificial Intelligence): »A Magnetic Climbing Robot for Marine Inspection Services«, describe robots that stick to steel structures (like bulkheads), using electro-magnets. Yannis Koveos et al. (Glafcos Marine) focus in »Robotic Arm Development for Ultrasonic Inspection in the MINOAS Project« on a particular task within robotic inspection, namely the preparation of a (corroded) plate and the positioning of an ultrasonic measuring device by a robotic arm that would be installed on the robotic crawler. Alberto Ortiz et al. (University of Balearic Islands) describe »A Micro Aerial Vehicle for Vessel Visual Inspection Assistance«, i.e. a mini helicopter that can rapidly move around high up in a cargo hold and take videos (or images) of the structures present.
