The paper reports application of recently developed adaptive control techniques based on neural networks to the induction motor control. This case study represents one of the more difficult control problems due to the complex, nonlinear, and time-varying dynamics of the motor and unavailability of full-state measurements. A partial solution is first presented based on a single input-single output (SISO) algorithm employing static multilayer perceptron (MLP) networks. A novel technique is subsequently described which is based on a recurrent neural network employed as a dynamical model of the plant. Recent stability results for this algorithm are reported. The technique is applied to multiinput-multioutput (MIMO) control of the motor. A simulation study of both methods is presented. It is argued that appropriately structured recurrent neural networks can provide conveniently parameterized dynamic models for many nonlinear systems for use in adaptive control.

The Hydrocarbon Field Planning Tool (HFPT), recently developed by Shell, provides capabilities for rigorous integrated subsurface-surface production forecasting in the medium to long term (1–30 years). HFPT can be used for gas, gas-condensate, oil and mixed gas-oil fields. HFPT models allow business optimisation by making more efficient use of the existing assets and by reducing investment costs in the new fields. In the simulation, HFPT uses a pressure-balanced solution of the integrated system: from the reservoir(s), through wells and surface facilities, to the delivery point. A wide range of fluid models is available, from simple gas-condensate and black oil PVT models to multi-component models with EOS flash calculations. HFPT provides optimisation functionality for maximising the returns in oil and gas fields, while accommodating operational preferences for production allocation and network constraints. It can also model injection networks and optimal lift gas distribution. Introduction The need for an integrated approach to dynamic field modelling has now been accepted by many players in the oil industry [1]. Issues which can be analysed in an integrated model, and which cannot be adequately addressed in a stand-alone reservoir model (or multiple stand-alone models), include:Pressure interaction between surface and subsurface.Pressure interference between different reservoirs and wells connected to a shared surface facility. An example is a high-pressure well backing out a low-pressure well.Mixing of dissimilar fluids from different reservoirs in the production network.Influence of facility constraints, e.g. separator limits, on a set of reservoirs connected to a shared facility.Production optimisation in the overall system against a set of common criteria. A number of field studies, performed using integrated subsurface-surface models, have already been reported, e.g. [2][3], showing benefits of such models. Over the past ten years, a wide range of applications from commercial vendors have appeared on the market which allow modelling of the subsurface and surface in an integrated way. Most of the available tools are designed either for a very simple reservoir description or for a simple surface description, or both. Hydrocarbon Planning Tool (HFPT) has been developed by Shell to fulfil the need for rigorous integrated subsurface-surface production modelling. It has been designed for accurate medium to long term forecasting, for optimising of production from existing fields, for analysing near-field potential in mature fields and for developing new fields. Currently, HFPT focuses mainly on medium to long term forecasting which is dominated by subsurface behaviour. However, surface and process facilities models can also be modelled in great detail, when needed. Requirements for Integrated modeling An integrated subsurface-surface model consists of the following main data modules, illustrated in Figure 1:PVT model.Subsurface model.Surface production system model.Processing facilities model.Overall integration and control, contracts, optimisation targets, development planning.

No abstract

No abstract

Gas Field Planning Tool (GFPT) was developed in 1990 by the Shell Group of Companies to fill the need for a tool for gas field planning and development using deterministic subsurface and surface models. Main initiators were Shell Canada, NAM (the Netherlands), Shell Expro (UK) and BSP (Shell Brunei), as these companies are major gas producers.Shell Companies now have several years experience with using the GFPT. Application ranges from simple single field models to corporate-level models with a large number of gas reservoirs and wells. Shell companies now using GFPT models are Shell Expro (UK), BSP (Brunei), SSB (Malaysia), Shell Canada, SPDC (Nigeria), SDA (Australia), Woodside (Australia), PDO (Oman), NAM (the Netherlands), New Business Development (e.g. Lunar Project) and in future also Shell Egypt.NAM currently has a GFPT model for the Anjum field in Friesland and for the Ten Arlo field in the north of Holland.GFPT is currently being migrated to an HFPT (Hydrocarbon Field Planning Tool), which can also be used for planning of condensate, oil and water developments and for control of hydrocarbon compositions in the network using PVT de-lumping at the well head (e.g. for LNG plants) and optimisation techniques (linear, non-linear or based on bean-back lists).