ABSTRACT. The presentation will collect and summarize a set results from a project on applied optimal control of wheel loaders where numerical optimal control has been used to evaluate and guide design changes in the vehicles. The first part introduces the tool chain with modeling, numerical optimal control and evaluations that allows us to analyze complex systems with many states and significant non-linearities. The results achieved from applying the toolchain that will be discussed are three innovations ranging from hardware changes, software changes, and driver instructions. One is the evaluation of torque converter selection, showing that a stiff torque converter is better than a weak. The other is the benefits of designing the mechatronics of the transmission and brake by wire system with a system called “intelligent braking”. The last result has impact on the instruction that are given as guidelines to drivers, while working on a site. All in all this illustrates that Optimal Control has evolved beyond a controls engineer tool and can now be used in the engineering design and innovation of complex mechatronic products.

ABSTRACT. The paper extends a mean value model of a parallel turbocharged internal combustion engine with a crank angle resolved cylinder model. The result is a 0D engine model that includes the pulsating flow from the intake and exhaust valves. The model captures variations in turbo speed and pressure, and therefore variations in the compressor operating point, during an engine cycle. The model is used to study the effect of the pulsating flow on mass flow balance and surge margin in parallel turbocharged engines, where two compressors are connected to a common intake manifold. This configuration is harder to control compared to single turbocharged systems, since the compressors interact and can work against each other, resulting in co-surge. Even with equal average compressor speed and flow, the engine pulsations introduce an oscillation in the turbo speeds and mass flow over the engine cycle. This simulation study use the developed model to investigates how the engine pulsations effect the in cycle variation in compressor operating point and the sensitivity to co-surge. It also shows how a short circuit pipe between the two exhaust manifolds could increase surge margin at the expense of less available turbine energy.

ABSTRACT. The most well-known technology for post combustion CO2 capture from exhaust gas is absorption in an amine-based solvent followed by desorption. The drawback of this method is the high heat consumption required for desorption. Reduction of the energy consumption can be achieved by using alternative configurations. In this work, the standard process, vapour recompression and vapour recompression combined with split-stream configurations have been simulated using Aspen HYSYS version 8.0 for 85% amine-based CO2 removal in search for optimum process. Energy optimisation has also been performed by varying the most important parameters. This study shows that it is possible to reduce energy consumption with both the vapour recompression and the vapour recompression combined with split-stream processes. The vapour recompression process has been calculated to be the energy optimum alternative among the configurations investigated.

ABSTRACT. Ukraine is an energy-dependent country and aims to reduce the import of natural gas, heat, and power in general. This implies extracting fuel from her own natural resources; one relevant and readily available energy carrier for such extraction is peat (bio mass). Currently, the operating regimes of the drying of peat are not energy efficient; these operating regime maps were developed in the 1970s and aimed only to get dry peat of required quality, without taking into account the cost of heat and electric energy in the drying process. The current quality of dried peat in the dryer does not always satisfy the necessary quality; e.g. parts of the peat may be insufficiently dried, or it may be over-dried. This affects the energy performance of peat briquette production.

To improve the quality and energy efficiency of the peat drying process, an analysis of the drying process is carried out: an empirical mathematical model of the drying process is developed using the GMDH principle [1, 2, 3], mapping input parameters and disturbances to output qualities based on available experimental data. Next, with known (measured) disturbances, optimal input parameters for the drying process are found. Changing the operational parameters too fast leads to insufficient drying or over-drying of parts of the peat. Thus, to avoid changing the operational conditions too fast, the operational conditions are classified into a number of classes corresponding to a certain range of values for the operational parameters. Finally, an iterative procedure for changing the input parameters from the past values to new values is introduced.

The resulting algorithm for finding the optimal operation of peat drying is based on mathematical models developed from experimental data, and aims to ensure improved quality and energy efficiency in the peat drying process.

Reference

[1] Ivakhnenko, A.G. (1971). “Polynomial Theory of Complex Systems”. IEEE Transactions on Systems, Man, and Cybernetics. Vol. SMC-1, No. 4, October 1971, pp. 364-378. [2] Farlow, S.J. (1981). “The GMDH Algorithm of Ivakhnenko”. The American Statistician, Vol. 35, No. 4 (Nov. 1981), pp. 210-215. [3] Anastasakis, L., and Mort, N. (2001). “The Development of Self-organization Techniques in Modelling: a Review of the Group Method of Data Handling (GMDH)”. Research Report No. 813, October 2001, Department of Automatic Control & Systems Engineering, The University of Sheffield, UK.

ABSTRACT. Bitumen is used as anticorrosion to protect steel armor wires of subsea cables and umbilicals. Because bitumen's viscoelastic behavior influences the mechanical properties of cables, it is desirable to include bitumen in mechanical cable analyses. For this purpose, the author and his colleagues have in a previous scientific paper presented a newly developed laboratory instrument for identification of bitumen's viscoelastic properties. The present paper derives an improved mathematical model for how to establish bitumen's viscoelastic properties from the sensor measurements of the laboratory instrument. It is proved mathematically that the model of the previous paper and the model of the present paper give identical results when bitumen has shear modulus of small magnitude. The difference between the two models grows with increasing magnitude.

ABSTRACT. Biogas is one interesting fuel for the decentralized energy production as it can be originated from many different sources. The purification of biogas to the point where it can also be used as a substitute for natural gas requires several purification steps. This paper focuses on the separation of carbon dioxide. A common solver using the discretized Gauss-Seidel method was applied for three different biogas purification process models describing MEA absorption, pressurized water scrubbing and multistage membrane separation. Simulations based on experimental designs were conducted in order to evaluate the applicability of those purification processes for different sources of biogas. The models used were rather simplified descriptions of the processes aimed to describe the directions of the changes in a general manner and indicate the controllability of the processes.

ABSTRACT. Recovery of heavy oil is associated with challenges in terms of early water or gas breakthrough caused by the fingering phenomenon. This study is related to Californian heavy oil production and includes 2D simulations of heavy oil with viscosity 300cP in horizontal wells with autonomous inflow control devices (AICDs). The simulations were carried out using ANSYS/Fluent as the Computational Fluid Dynamics (CFD) software. Volume of Fluid (VOF) is used as the multiphase model. Preliminary studies are performed and a model for heavy oil production has been developed and validated against experimental data found in literature. Grid resolution tests and time step dependency tests are performed. 2D simulations were carried to study reservoir flow, annular flow and flow through the AICDs into the production pipe. Before breakthrough the analysis of data gives quite realistic results which can be theoretically justified. The simulations clearly stated the fingering phenomenon and the effect on breakthrough and oil production. Higher pressure drop is observed near the well bore (or across AICD) as the finger growth takes place. Simulation of two pipe sections in a homogeneous reservoir does not reflect the function of AICDs adequately. This is due to negligible frictional pressure drop in the pipe which breakthrough at approximately the same time in the two sections. This study has /identified the formation of fingering in heavy oil reservoir with water drive and according to the simulated results, oil production is highly affected by fingering behaviour, since it enables the early water breakthrough while most of the oil is left unproduced.

ABSTRACT. In calcium looping (CL), calcium oxide (CaO) is used as a sorbent for carbon dioxide (CO2). The CO2 reacts with CaO to produce calcium carbonate (CaCO3) in a carbonator operating at about 650 °C. The CaCO3 is then sent to another reactor, a calciner operating at about 900 °C, where the CaCO3 is calcined, producing CaO (which is returned to the carbonator for another cycle) and more or less pure CO2, which is removed from the system. The calcination is an endothermic process, so a significant flow of thermal energy must be supplied to the calciner for the reaction to occur. In conventional CL the heat is transferred directly by oxy-combustion in the calciner; pure oxygen is required as the oxidizer to avoid mixing CO2 with nitrogen. Even though most of the thermal energy supplied in the calciner can be recuperated in the carbonator, the oxy-combustion gives an unwanted energy penalty of the CL technology. However, if the heat could be transferred indirectly to the calciner, then then energy penalty associated with oxy-combustion could be avoided, and this would make CL a much more attractive alternative for the thermal power industry. The low energy penalty of CL with indirect calciner heat transfer is due to high-temperature integration between the CO2 capture plant and the power plant. In this work, Aspen Plus is used to simulate the CL process with indirect heat transfer. The results confirm that such a scheme could give an energy penalty lower than for example amine scrubbing or oxy-combustion.

ABSTRACT. Due to complexity in interactions between solid particles and between particles and carrier fluid, modeling of biomass gasification kinetics has been shown to be very challenging. Some commercial packages such as Barracuda VR describe the process using a complicated Eulerian-Lagrangian approach of modeling gas - solid multiphase flow. Since chemical reactions play the major role in the gasification process, this paper focuses on a simplified model that describes the influence of reaction kinetics on the process in a bubbling fluidized bed reactor with circulating bed material. The developed model assumes that the system is ideally mixed with a constant bed temperature. Two different set of reaction rate constants are applied to test the model, and their results are compared with that from Barracuda. The model is used to study the effect of steam-biomass ratio on char conversion, and the results show that the conversion of char increases with steam-to-biomass ratio. Sensitivity analysis shows that the output of the model strongly depends on temperature and slightly depends on the bed material particles size.

ABSTRACT. Uncertainty Quantification (UQ) is vital to ensure credibility in simulation results and to justify model-based design decisions – especially in early development phases when system level measurement data for traditional model validation purposes are scarce. Central UQ challenges in industrial applications are computational cost and availability of information and resources for uncertainty characterization. In an attempt to meet these challenges, this paper proposes a framework for early and approximate UQ intended for large simulation models of dynamical systems. A Modelica simulation model of an aircraft environmental control system including a liquid cooling circuit is used to evaluate the industrial applicability of the proposed framework.

ABSTRACT. This paper summarizes the continuation of the work performed in one of the work-package of the FP7 \textit{\textbf{i}Tesla} project. This work consisted in the development of power system component models of PSS/E models in Modelica. The PSS/E models selected for this work are used in different dynamic models of the Norwegian grid.

The development of the Modelica models was done in progressive stages to validate all the models. The performance of the Modelica models was validated by comparing several simple grid models that were implemented in both PSS/E and Modelica. The grid models were built to integrate different kinds of perturbation to accurately validate the models.

ABSTRACT. The Modelica language is one of the most important languages for representing a large class of systems, ranging from vehicles to climate control systems in buildings. Component libraries, containing components like valves, motors, pumps, etc., have been built to facilitate the construction of complex systems, but at present model construction and parameter estimation are entirely manual. We have developed software for (i) automating the process of constructing models that optimise a range of metrics (e.g., model-simulation accuracy or diagnostics accuracy), and (ii) dynamically updating the model parameters due to dynamic changes in the observed data and/or health status of the modelled system.

We assume that in these component libraries a component may be modeled at multiple levels of fidelity, e.g., as a non-linear system (high-fidelity model), linear system, or a qualitative system (low-fidelity model). Choosing the right component model for system simulation is a difficult task and requires a search in the space of all possible component type combinations. In this paper we propose a method that automates this task and computes a system model that optimizes a set of metrics in a set of simulation scenarios. We describe initial experimental results showing the trade-offs of accuracy and inference time.

This software has the potential to revolutionise how industry uses Modelica, i.e., changing the use from an expensive manual process to a fully automated process that is adaptive to changing external conditions.

ABSTRACT. This paper presents a software architecture for a collaborative virtual environment (CVE) for simulation and visualisation based on the Functional Mock-up Interface (FMI) for co-simulation and web technologies. FMI has been chosen in order to have a standardised and independent interface to models created in different modelling tools. The user interface has been implemented using web technologies, which enables a very high degree of flexibility. The Web Graphics Library (WebGL) is used for interactive 3D visualisations, enabling native cross-platform rendering directly in the browser without the need of installing any additional plug-ins. Employing the bi-directional communication capabilities of the WebSocket protocol, multiple users can interact with the same simulation models simultaneously. A software prototype has been developed in order to demonstrate the applicability of the proposed architecture. As a case study, we have considered virtual prototyping of marine cranes, to illustrate the use on real world problems.

ABSTRACT. Brazil has special attention in the world oil scenario due its recent findings of oil reservoirs. Most of these reservoirs are heavy oil. The increasing energy consumption and discovery of heavy oil reservoirs require more studies about the extraction of high viscous oil and the related challenges. Besides, the choice for horizontal drilling has being more and more applied in the fields in Brazil. Horizontal wells have shown higher recovery rate and are very suitable for thin oil reservoirs. However, the early water and/or gas breakthrough due to heterogeneities and the heel-to-toe effect is a big challenge. The problem with early water and/or gas breakthrough to the horizontal well can be avoided with new inflow control technology. One of the newest technologies is Autonomous Inflow Control Valve (AICV). The AICV autonomously closes for unwanted fluids as water and or gas, and stay open for oil. In this work, a computational study of a heavy oil (228cP) reservoir in the Santos Basin with 115 m thickness is modeled using Rocx coupled with OLGA. The horizontal well is simulated with both AICV (300 days) completion and openhole with gravel pack completion (200 days). The well has a length of 650m, diameter of 0.24m and is divided into 10 sections. The differential pressure is 20bar. The reservoir is considered heterogeneous with permeability ranging from 5D to 50D. Using AICV completion the valves closes first in the highest permeability zone and then following the heel-to-toe effect. The results confirm that the water breakthrough to the well can be avoided by using AICV completion. The technology will contribute to extend the lifetime of a well and result in reduced costs regarding oil-water separation.

ABSTRACT. Heat pump technology provides an efficient and sustainable solution for both heating and cooling. A traditional heat pump can be defined as a mechanical-compression cycle refrigeration system powered by electricity. Traditional refrigerants used in heat pumps are ammonia or chlorinated and fluorinated hydrocarbons. Because many of these chlorofluorohydrocarbons (CFC’s) are ozone-depleting components, evaluation of more environmentally friendly refrigerants like pure hydrocarbons is important. The efficiency of a heat pump is traditionally measured by the ratio of delivered heat at a high temperature divided by the electricity (or work) input. This is defined as the coefficient of performance (COP).

There are several simulation tools available for the simulation of heat pumps. Traditional process simulation tools like Aspen HYSYS or Aspen Plus are useful because they have data for several components inside the program, and many thermodynamic models like equations of states available. It is of interest to calculate the efficiency of a heat pump system when varying temperature conditions, refrigerants, simulation programs and thermodynamic models. There are few references to such comparisons in the open literature.

The circulation medium (refrigerant) alternates by the help of a compressor and an expansion valve between the temperatures 22 and 7, 24 and 5 or 24 and -15 °C. The lowest temperature is the evaporation temperature and the highest is the condensing temperature. The pressures were specified as the saturation pressures at the given temperatures. The evaluated components are ammonia, R-11, R-12, R-22 and propane. The equations of state Peng-Robinson and Soave-Redlich-Kwong (SRK) and the activity model Non-Random-Two-Liquid (NRTL) were used in the process simulation programs Aspen HYSYS and Aspen Plus.

COP values have been calculated to values between 3 and 9. The highest COP was calculated for the lowest temperature difference. The components giving the highest COP value between the temperatures 22 and 7 °C were ammonia and R-12, and R-22 gave the highest COP between -15 to 22 °C. Propane (which is not a CFC) gave slightly lower COP values than the other components.

The differences between the thermodynamic models and the different programs were normally low. However, some differences between Aspen Plus and Aspen HYSYS for the same model were calculated. The calculated deviations between the same models using different programs are difficult to explain. Different model parameters in different programs may explain differences between the same models in different programs.

Aspen Plus and Aspen HYSYS are evaluated to be powerful tools for heat pump calculations. The calculated differences between heat pump efficiencies with different components at different conditions are thought to be reasonable.

ABSTRACT. Biomass is an attractive source of energy. It is possible to utilize the energy through the thermo-chemical conversion processes of pyrolysis and gasification. The Aspen Plus simulation tool is useable to simulate the biomass gasification processes. The most common way is to simulate the gasification reactor using Gibbs reactor. The Gibbs reactor applies Gibbs free energy minimization with phase splitting to calculate equilibrium. The reactions in the gasification process are complex and for the Gibbs reactor it is not necessary to specify the stoichiometry or the reaction rates. A set of reactions that describes the major conversion rates in a gasifier reactor can be extracted from the literature. Using these stoichiometric chemical equations and their reaction rates in Aspen Plus it is possible to simulate the gasification process also with continuous stirred tank reactor (CSTR). The aim is to compare the composition of produced gas based on simulation with Gibbs reactor and continuous stirred tank reactor. The influence of parameters like reactor temperature, reaction residence time and steam flow are examined in sensitivity analysis. The results show significant variation in the produced gas composition for the Gibbs reactor and the continuous stirred tank reactor.

ABSTRACT. Heavy oil and bitumen represent a massive world resource more than twice the size of global reserves of light or conventional oil. In reservoirs with extra heavy oil and bitumen, thermal methods are used to reduce the oil viscosity, in order to extract the oil. Steam-Assisted Gravity Drainage (SAGD) is a thermal method where continuous steam injection is used. In this method, two horizontal wells are placed in parallel. Different types of inflow control technologies are developed to increase oil recovery by avoiding production of unwanted fluids, and the newest on the market is an autonomous inflow control valve (AICV). In the SAGD processes, it is important that the residence time for steam is high enough to ensure that the injected steam condenses in order to utilize all the latent heat. The AICV technology has the ability avoid steam to be produced together with the oil. The valves will locally shut off the zones with steam breakthrough, and simultaneously produce oil from the other zones along the well. The technology eliminates the steam breakthrough problems and contributes to optimize the SAGD process. Reservoir simulations with Rocx in combination with OLGA are performed to study the potential of increased oil recovery using SAGD in combination with AICV completion. TechPlot is used to study the reservoir conditions. The results are discussed regarding oil recovery, residual oil saturation and optimization of steam injection.

ABSTRACT. Changing operating condition may require updates for the dynamic models. Recursive updates are needed when there are not sufficient information about the new situations. In machine diagnostics and prognostics, the analysis starts from good conditions and new phenomena, which activate with time, may change considerably the model. In biological wastewater treatment processes, the condition of the biomass changes drastically the dynamic operation of the treatment process. Direct measurements of the biomass condition are under development. Recursive modelling is clearly needed in these situations. The usual approachis to modify the model equations. However, the interactions do not necessarily change if the meanings of the variables are modified. This paper keeps the the model equations constant and modifies the nonlinear scaling of the variables by extending the data-driven scaling to recursive approach. The recursive methodology is tested in two applications: machine diagnostics and wastewater treatment.

ABSTRACT. The sliding mode approach is recognized as an efficient tool for treating the chattering behavior in hybrid systems. However, the amplitude of chattering, by its nature, is proportional to magnitude of discontinuous control. A possible scenario is that the solution trajectories may successively enter and exit as well as slide on switching manifolds of different dimension. Naturally, this arises in dynamical systems and control applications whenever there are multiple discontinuous control variables. The main aim of this paper is to provide a robust computational framework for the most general way to extend a flow map on the intersection of p transversally intersected (n-1)-dimensional switching manifolds in at least p dimensions for any finite (positive) integer p. We explore a new formulation to which we can define unique solutions for such particular behavior in hybrid systems. We illustrate the concepts with examples throughout the paper.

ABSTRACT. In many organizations the utilization of available computer power is very low. If it could be harnessed for parallel simulation and optimization, valuable time could be saved. A framework monitoring available computer resources and running distributed simulations is proposed. Users build their models locally, and then let a job scheduler determine how the simulation work should be divided among remote computers providing simulation services. Typical applications include sensitivity analysis, co-simulation and design optimization. The latter is used to demonstrate the framework. Optimizations can be parallelized either across the algorithm or across the model. An algorithm for finding the optimal distribution of the different levels of parallelism is proposed. An initial implementation of the framework, using the Hopsan simulation tool, is presented. Three parallel optimization algorithms have been used to verify the method and a thorough examination of their parallel speed-up is included.

ABSTRACT. The Environmental Control System (ECS) of the Saab Gripen fighter provides a number of vital functions, such as provision of coolant air to the avionics, comfort air to the cockpit, and pressurization of the aircraft fuel system. To support system design, a detailed simulation model has been developed in the Modelica-based tool Dymola. The model needs to be a “good system representation”, during both steady-state operation and relevant dynamic events, if reliable predictions are to be made regarding cooling performance, static loads in terms of pressure and temperature, and various other types of system analyses. A framework for semi-automatic validation of the ECS model against measurements is developed and described in this paper. Applied methods for validating the model in steady-state operation and during relevant dynamic events are presented in detail. The developed framework includes automatic filtering of measurement points defined as steady-state operation and visualization techniques applied on validation experiments conducted in the previously mentioned points. The proposed framework both simplify continuous validation throughout the system development process and enables a smooth transition towards a more independent verification and validation process.

ABSTRACT. Due to increased burden on the environment caused by human activities, focus on industrial ecology designs are gaining more attention. In that perspective an environmentally effective integration of bionergy and agriculture systems has significant potential. This work introduces a modeling approach that builds on Life Cycle Inventory and carries out Life Cycle Impact Assessment for a consequential Life Cycle Assessment on integrated bioenergy and agriculture systems. The model framework is built in Python which connects various freely available software that handle different aspects of the overall model. C-TOOL and Yasso07 are used in the carbon balance of agriculture, Dynamic Network Analysis is used for the energy simulation and Brightway2 is used to build a Life Cycle Inventory compatible database and processes it for various impacts assessment methods. The model is successfully demonstrated using a manure utilization case study where the manure is used to produce biogas and then heat and power, whereas its digestate is used as an organic fertilizer to a wheat field. The case study is compared with direct manure to wheat field application.

ABSTRACT. Insufficient level of the energy efficiency in Ukraine requires an immediate reformation of the Ukrainian power sector in order to stimulate power enterprises and final consumers to decreasing of fossil fuel and power resources unit costs and reduction of harmful emissions. Taking into account current introduction of the liberalized electricity market in Ukraine it is advisable to use the experience of the best energy efficiency managing practices of other energy markets for determining and developing the instruments and mechanisms that could stimulate Ukrainian market players for efficient resource usage. The world markets experience is based on the offering of saved electrical energy or reduced capacity as a resource in the market. Thus it is important to determine and quantify accurately the expected and actual demand reduction and amount of saved electrical energy. The article discusses and compares the usage of existing methods and approaches for determining necessary characteristics and data for solving the mentioned tasks. The methods for modeling the energy consumption of the objects which implement energy saving measures are investigated. The paper contains the comparative analysis of the methods for development of the energy consumption models which based on the measured data.

ABSTRACT. Design of optimal operation and control of a run-of-river hydro power plant depends on good models for the elements of the plant. River reaches are often considered to be shallow channels with free surface flow. A typical model for such reaches thus use the Saint Venant model, which is a 1D model based on the mass and momentum balances. This combination of free surface and momentum balance makes the problem numerically challenging to solve.

Here, the finite volume method with staggered grid is used to illustrate the dynamics of the river upstream from the Grønvollfoss run-of-river power plant in Telemark, Norway, operated by Skagerak Energi. A model of the same river in the Grønvollfoss power plant has been studied previously, but here the geometry of the river is changed due to new information from Skagerak Energi. The numerical scheme for solving the model has been further developed.

In addition, the behavior of the dynamic model is compared to data form experiments, carried out on the Grønvollfoss run-of-river power plant. The essence of the experiments is to consider the time taken from an increase in the input volumetric flow, to a measured change in level in front of the dam at Grønvollfoss.

The results of the improved model (numerically, tuned to experiments), is a model that can be further used for control synthesis and analysis.

ABSTRACT. Near well computational fluid dynamic simulation for extra heavy oil production using steam assisted gravity drainage was studied with ANSYS Fluent software. Computational fluid dynamic simulation can predict the multiphase flow behavior in the well annulus and the base pipe when the well involves an Autonomous Inflow Control Valve which is a promising approach for enhanced oil recovery. Volume of Fluid multiphase model was used to simulate the fluid behavior. Three different case studies were carried out by changing the volume fraction and the orientation of the valve. The mixing region of the two immiscible fluids is increased if the water volume fraction of the inlet from the reservoir to wellbore is increased. Simulation results showed that the valve orientation has an impact on oil/water separation.