SPE 118454 MS prodcution optimization in an oil production asset the BP azeri field

Among the main objectives of Pemex Exploracion and Produccion Region Sur is found the analysis of the operation, the optimizacion and the modernization of the systems of produccion, transportation and distribution of hydrocarbons, in order to provide better security conditions, efficiency and opportunity, without forgetting the proteccion to the environment and the society. For the foregoing, it results from great importance to count on a methodology for planning the behavior and development of the infrastructure of produccion and transportation to face current and future needs, bearing in mind the operational requirements and quality that demand best engineering practices. The utilization of Process and Transport Simulators to analyze the facilities in the current operation conditions, and to predict their behavior under different stages is a valuable tool that applied combined with the best practices of engineering allows to predict, with meaningful time savings, the necessary modifications to be adjusted to the new operative philosophy, without putting on risk the facilities. As added value, the results obtained from the application of this methodology also contributes to the construction of a data base, of the facilities of produccion as well of transportation, of great usefulness for activities related to the maintenance programs.

SPE 118454 Production Optimization in an Oil Producing Asset – The BP Azeri Field Optimizer Case N Jalilova, BP Azerbaijan; A Tautiyev, BP Azerbaijan; J Forcadell, Aspen Technology; J.C Rodríguez, Aspen Technology; and S Sama, Aspen Technology Copyright 2008, Society of Petroleum Engineers This paper was prepared for presentation at the SPE Gulf Coast Section 2008 Digital Energy Conference and Exhibition held in Houston, Texas USA, 20–21 May 2008 This paper was selected for presentation by a Gulf Coast Section Program Committee following review of information contained in an abstract submitted by the author(s) Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s) The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied The abstract must contain conspicuous acknowledgment of SPE copyright Abstract The Azeri Field Optimizer is a modeling and optimization platform that encompasses the requirements of offshore and onshore production teams in a single, unified decision support system The tool supports the following functionalities: data gathering and conditioning, model validation and calibration, economic process optimization and what-if analysis The Azeri Field Optimizer uses Aspen HYSYS, PROSPER, Microsoft Excel, Aspen IP.21 data historian and ad-hoc MS SQL databases The heart of the tool is a high fidelity asset-wide model linked to an optimization solver that receives plant data from a centralized data historian (IP.21) The model uses a combination of first principles and bespoke algorithms to provide an accurate representation of all pressure-flow relationships, material balances and energy equations from the well bores to product delivery points The tool has been designed around the existing asset workflows and processes to facilitate the adoption of the technology and it has been created in such a way that allows dynamic re-configuration of the model and optimization problem by automatically capturing equipment availability indicators This enables the asset optimization team to perform accurate dayto-day production analysis and optimization studies in a timely fashion The system has been tested in a number of real production situations where it has helped to: a) pinpoint hidden process behaviors that can be turned into an opportunity for economic improvement, b) identify process bottlenecks under various different production scenarios, c) co-relate offshore and onshore process constraints to find true overall optimums, and d) identify potential oil production increases averaging 3% This paper describes the technology and methodologies used for creating the Azeri Field Optimizer and discusses on the benefits achieved so far Introduction Oil and gas production systems represent an intricate combination of standard process engineering and specialized petroleum engineering and geology Although these disciplines base their knowledge in the same fundamental physical principles, they not necessarily tackle problems from the same angle and/or using similar modeling tools This dichotomy of backgrounds is directly observable in the interests and speeches of offshore and onshore production departments In multitude of occasions (if not all) what can be an opportunity of improvement for the platform engineers represents a real challenge for the onshore personnel and vice versa The impact of a well flow increase on the performance of the onshore flash gas compressors or the consequences of a temporary unavailability of one oil stabilization train on the platform production plans are common subjects that require the usage of a unified decision-making framework to be answered with engineering words 2 SPE 118454 The Azeri Field Optimizer (AFO) has been conceived to help the Azeri asset: • • • • Optimize current performance and oil production with available onshore and offshore equipment Define the best operating philosophy to maximize production in case of equipment outages Identify bottlenecks and plan additional investments Optimize short term gas lift distribution and longer term water and gas injection quotas The Azeri Asset The Azeri oil and gas production asset is within the top BP investments world-wide and possibly one of the most complex oil systems around the world The present offshore facilities consist of production platforms located in the Caspian Sea, named Central, East and West Azeri and a gas processing platform (CWP) Each production platform is provided with oil separation (HP and LP), flash gas compression and gas dehydration CWP is equipped with parallel compression trains driven by 25 MW gas turbines, which supply HP gas for reservoir injection and well lifting CWP also contains machinery for water injection and gas compression pressure support The onshore terminal is provided with oil stabilization trains equipped with stages electrical driven flash gas compressors, oil and gas receiving facilities and a gas dew-pointing unit The associated gas is delivered through pipeline to the local gas consumers Critical Success Factors The importance of addressing People, Process and Technology issues in order to ensure successful adoption of new technologies has been widely documented (Mochizuki et al 2006; Truschinger 2004) Taking a simplistic approach and focusing on only one or two of the three areas has been identified as the single most important reason for failed technology adoption initiatives In the case of Oil & Gas producing assets this is even more important since the industry is under tremendous pressure to keep the assets producing at full capacity (or agreed-upon rates) which leads to most operators being risk-averse These three factors are briefly described before entering in detail into BP Azeri Field Optimizer specific matters People People issues are tremendously wide in nature and clearly covering them even on their surface is beyond the scope of this document There are several “people” issues of particular importance and whose overlook may lead to implementation failure: • “What’s in it for me?” Often the adoption of new technology, perceived by the management as a means to improve productivity, reduce risks, improve HSE compliance, can be perceived by other levels of the organization as extra work or even as a threat to job security This is often difficult to identify and may lead to complete failure in case the project lead’s attitude is that of reluctantly accepting the new technology • “Why have we not done this before?” The adoption of new technology may point out ways to operate better an asset There is the temptation of blaming staff for not having found before those ways to operate better It happens more often, in fact, that there is a fear of management thinking so Pedagogy to explain to the various stakeholders in the project that those improvements could not possibly be found without the newly available technology is needed in order to mitigate this risk Processes Usually business processes have been designed with a certain framework in mind Technological changes may deem existing processes inadequate or obsolete There is a risk that the need for re-engineering business processes not be identified in technology adoption initiatives, with the net result of unachieved performance expectations as the processes seriously constraint the usefulness of the newly available technology During an initiative of technology adoption, the existing processes (workflows) need to be assessed and then need for their re-engineering identified It is equally important to plan the transition from “As-is” to “To-be.” A common mistake is to design the new workflows but not plan how the transition from the current way of working to the new one will be achieved SPE 118454 Successful transition usually requires a serious investment in training and technology transfer so that staff can cope successfully with the new environment (of technology and processes) A period of “hand-holding” whereby staff is helped achieve their daily objectives in the new environment can work miracles in this area Technology It is usually the case that technology works in a highly synergistic manner: a single technology is often of little use unless other enabling technologies are also available and with the adequate level of maturity In the area of field optimization, successful implementation requires the synergistic collaboration of the following technologies: • • • • • • • • • Fast running modeling and simulation tools Fast running optimization tools Digitally enabled metering systems Communication systems Real-time databases/process historians Modeling and simulation tools Optimization tools User interface infrastructure Workflow-handling tools The above considerations clearly make implementation methodology a key art and science when introducing new technology into the field of Oil & Gas operations (see Stenhouse, B and Goodwin, S 2004; Stenhouse, B 2006) A methodology that would disregard people and processes factors, and that, focus entirely on delivering technology is deemed for failure It is well-known that often implementation methodologies are too shy to bring people and process issues to the surface (they are often left for the Oil & Gas organization to be dealt with at a later time, which is usually an unrealistic plan) Azeri Field Optimizer Overview The Azeri Field Optimizer (AFO) is an offline advisory system conceived to find out the optimum values of key process settings in a number of business scenarios These scenarios cover from the typical asset revenue maximization objective to more specific business challenges, such as maximization of injected gas, maximization of exported gas and others ( HYSYS ) Validation Material Balance Calibration Optimisation Figure – AFO architecture SPE 118454 Figure – AFO user interface snapshot AFO consists of the following software components: • • • • • Well models (PROSPER) Facilities simulation package (Aspen HYSYS) User interface (named AFOI and coded in Excel) Data historian (Aspen IP.21) Well test database (Microsoft SQL) The user interface acts as the main controller to transfer information between system components (see Figure 1) Thus the AFOI performs: • • • • • • • Batch runs of the individual PROSPER well simulation models to create simplified Pressure-Flow relationships for each individual well performance that can be later on utilized for optimization Transferring of these relationships to the HYSYS simulation environment, where they will be solved together with HYSYS facilities models Handling of user inputs for the optimization Retrieval of field data variables to be used as inputs for model validation and calibration Data conditioning of field variables Execution of optimizer in its various modes: 1) model initialization, 2) mass balance, 3) model calibration and 4) optimization Presentation of results Design Principles There are a number of basic considerations that have been applied in the AFO design: • • • • • • • • • Optimizer must provide fast response time All optimizer underlying models must be able to be validated against key plant data and provide the means to be easily tuned when necessary The models have to be dynamically reconfigurable to represent the actual state of asset equipment at all times The user interface must be friendly and intuitive Built-in field data validation and visualization mechanisms shall be in place Clear ability to display/visualize model results in a way that can be effectively communicated among team members The tool must deliver the optimized well line up and facilities operating point in a reasonable computing time The user should be able to specify / modify well and key equipment data via the interface prior to triggering an optimization or calibration run Provide online diagnostics and advice in the case of any model failure SPE 118454 • Optimization logic in the tool could be implemented in different ways but it needs to reflect actual process flow logic and direction in order to mirror all process configurations and allow adequate optimization One Model – Multiple Objectives Approach The heart of AFO is a multi-purpose high fidelity model which is provided with auto-calibration mechanisms The internal model logic takes care of re-configuring the problem to solve (mass balance reconciliation, model calibration, and different optimization scenarios) The advantage of this approach is that the same model which is tuned to the actual plant conditions can be extrapolated to new operating conditions with confidence The tuned model can be used to monitor the performance of the plant equipment, performing off-line studies (for potential upgrading schemes or studies for different feed or economic conditions) and for optimization studies Typical optimization tools allow varying process parameters, but have the optimization problem pre-configured and hardcoded AFO provides the flexibility to define the scope of the optimization problem to solve without the need of a manual reconfiguration Optimization scenarios can be set to look at the entire asset performance or can be restricted to certain subsystems, e.g offshore optimization, onshore optimization, etc Figure – AFO model organization Strict rules have been applied during model construction in what concerns the hierarchy of systems and subsystems so that activation/deactivation of parts of the asset model can be properly cascaded down to the individual pieces of equipment (see Figure 3) Model configuration is performed through “custom” macros which are controlled from the graphical frontend, and whose execution is launched through a series of configuration flags available to the users Modeling Philosophy One of the most common mistakes when creating a model based optimization system is isolating the modeling and optimization activities with the assumption that a good model that it is valid for simulation studies can be also used for optimization purposes Although that premise is in essence true, the level of robustness required for an optimization model is not comparable to the one required for standard process simulation Other differences that are easy to overlook are: • Individual pieces of equipment need to be protected against infeasible operating regions and/or be provided with extrapolation mechanisms to avoid model failures during the optimization process This may require not only a significant amount of time pondering but also engineering background, especially when high fidelity rated models have to be utilized • Model needs to be constructed with a clear idea of which are the true decision variables (process set points) Most process simulators nowadays have a lot of flexibility to specify the model in several ways, not necessarily compatible with actual instrumentation installed in the process • A considerable testing effort will have to be invested before claiming the model is prepared to support process optimization The small inefficiencies and short-cuts that will never manifest in a standard simulation model tend to pop-up rather quickly when the model is left to be controlled by the optimization engine 6 SPE 118454 There is no general rule or magic recipe that indicates the steps and pre-cautions that need to be taken into account to create a model suitable for optimization; nevertheless there are a number of common areas that shall be given sufficient attention: • Repetitive calculations (in particular material recycles) which are solved in an iterative manner in sequential modular flowsheeters shall be avoided wherever possible, as they are a sink of simulation time and a source of noise for the calculation of the Jacobian and Hessian matrices Although the natural tendency is to place recycle loops where they naturally exist in the process, this may not be completely necessary • Sometimes the same unit operation block can be solved in different ways, being this dependent on the process specifications provided to the model Caution shall be applied to always using the least time consuming option (direct calculations preferred to iterative loops inside individual units) If the optimization algorithm is flexible enough, it will allow relieving the flowsheet solver from most iterative calculations, as some balances are equally solvable as part of the optimizer problem configuration Re-casting the model to allow infeasible path optimization maybe that most effective option in terms of computational speed and system robustness • In other situations, it may be convenient to lump several unit operation blocks into a single model that is solved analytically Thus, for instance, the pressure-flow coupling of a gas injection unit served by various parallel compression trains delivering to a bundle of gas injectors can be solved more efficiently by re-organizing the engineering equations governing the performance of the compressors and the injectors and solve them analytically than letting the flowsheet to iterate at the boundary to converge on pressure and flow rate Workflow Description The Azeri Field Optimizer is provided with a centralized graphical environment (AFOI) that is responsible for collecting all data required for initializing and calibrating the asset model from data historians, interface with PROSPER well models and depict the well head hydraulic behavior, handle the traffic of information to/from the model and execute the selected optimization runs AFOI (AFO Interface) has been designed to help the end-user to: • • • • • Control the transfer of information between the different system components including the well models Perform plant data validation and pre-processing Do well model calibration against well test data retrieved from existing well test databases Execute the model/optimizer in its various modes: o Model initialization for validation – The Model is configured with validated plant data corresponding to the particular day of study o Mass balance - Key oil, gas and water flow-rates retrieved from the plant data historian are used by the HYSYS optimizer to reconcile well flow-rates calculated by the well models o Model calibration - The HYSYS optimizer is used to calculate key equipment performance tuning factors After a model calibration (or model tuning), first principle model equations continue being valid, but their result are shifted with an offset calculated to mirror reality o Optimization - It is essential that what is being optimized accurately reflects the current state of the plant/equipment, i.e., a tuned model must be the starting point for the optimization Present results Figure – AFO execution modes and well calibration SPE 118454 Data Validation It is commonly known that field data in the oil and gas sector tends to be scarce and when available sometimes of bad quality, as a result of the nature of the process (e.g multi-phase flow with frequent disruptions) The wise old dictum of garbage in, garbage out (GIGO) is particularly important in the context of this type of systems The quality of the data used for model initialization and calibration determines the level of confidence of model and optimizer results It is therefore essential that the optimization system is provided with robust, reliable and intuitive data validation mechanisms All field instruments have a fixed error or offset and a random error This is especially true for flow-meters which rely on an assumed fluid density which will almost certainly be different from the actual fluid density The instrument offset can be estimated from the current plant performance and is assumed to remain the same over a fairly large time range Figure – Gross error detection and manual data correction mechanisms The gross error detection engines check a piece of data against its corresponding high or low scale instrument calibrated range and alarms when out of ranges are found (see Figure 5) Bad data must be detected and replaced with an alternate piece of data to prevent the model from stopping or become misleading This alternate piece of data may come from an alternate sensor (i.e in the case of redundant sensor installations), or from a calculation Ad-hoc data quality analysis engines are placed around critical pieces of equipment These components look at the overall consistency of a group of field measurements and can pinpoint suspicious data points where the gross error detection engines saw no problem or instrument failure Thus, for instance, suction flow rates in parallel compression trains shall be consistent with the corresponding train status indicators, as well as with the power consumption readings, discharge pressure readings, etc Well Models Tuning Wells are simulated with the help of individual PROSPER well models These models are updated and calibrated regularly to reflect actual well performance The well models are critical to the predictive capability of the asset model in that they define how each well responds to changes in system pressure and which wells provide the most oil recovery To speed up simulations and optimizations, well models are transformed into well head pressure-flow empirical expressions, which are developed by best fitting data simulated with the corresponding PROSPER well model When well tests are available, these can be compared with model predictions graphically and if deviations are found to be acceptable the well head performance relationship can be adjusted to match well test data (water cut, gas/oil loading -GORas well as main flow rate volumes) Well models calibration is controlled from the AFOI Excel 2D and 3D graphical capabilities can be used to check well model results against well tests data and if necessary to adjust well models If the well model is far from current observed performance, the appropriate platform production engineer must be contacted to schedule a new well test Wells composition varies as a result of well aging; this is translated into different water cuts and different GOR values As part of the well calibration process, the system uses a phase re-combination mechanism to adjust the proportions of water and coning gas to best match the available well test measurements 8 SPE 118454 System Mass Balance Having a correct mass balance throughout a complex oil asset may sometimes be problematic Offshore flow meters are seldom reliable with the exception of fiscal meters located at the platforms delivery or re-delivery points Intra-platform meters tend to be noisy due to the multiphase and dynamic nature of the fluids passing through Bypasses and cross-overs between production manifolds can experience counter current phases flow In the most simple situations, material make-ups or withdraws can be configured in the model to best match reliable instrumentation readings (e.g fiscal meters) In other cases a combination of several instrument readings have to be juggled to come up with a methodology that gives the most sensible distribution of flows (parallel trains loading); this may imply re-combination of fluid phases locally Model Calibration It is well-known that no engineering calculation can be better than the assumptions used as a basis to carry out the calculations Any model of a real item of equipment has an intrinsic error resulting from the assumptions and simplifications embedded in the model or from the fact that certain real behaviors cannot be modeled in a form suitable for optimization (e.g dynamic wax deposition in heavy oil pipelines) In order to represent the process as accurately as possible, equipment models are provided with ad-hoc parameters that can be tuned to match the actual process performance These parameters are assumed to be fixed over a wide range of operating conditions After a model calibration (or model tuning), first principle model equations continue being valid, but their results are shifted with an offset calculated to mirror reality The values of these “calibration factors” have themselves an intrinsic value as they are an indication of how far the operation is from the physical principles that govern its behavior An oscillatory pressure drop offset in a gas pipeline, may indicate the presence of significant pipeline dynamic effects A continuously decreasing compressor head offset will indicate compressor fouling and/or mechanical degradation In general, a model calibration problem is a least squares minimization problem that can be solved asset-wide using the same mathematical algorithm that will be employed for optimization This avoids the usage of additional side models or model configurations and facilitates model organization and maintenance Optimization Optimization is a very generic term to refer to the processes at finding the “optimum” or more favorable conditions to carry out a certain activity of interest (e.g process) The term optimum has probably as many interpretations as individuals using it In the oil and gas sector sometimes optimization refers to simply finding the mechanisms to run a compression train with maximum delivery flows at all times; other the optimization process scope expands to cover the performance of a hundreds or thousands well systems The term optimum can be interpreted as that point in which the benefit of the operation calculated as the differences between revenues and processing costs reaches its maximum Nowadays, with the current oil and gas market prices most operators will interpret that optimizing means maximizing production, and likely short term production Regardless of the level of model and optimization problem complexity there are a number of aspects which are recurrent and that need to be taken into consideration: • Process decision variables – a typical optimization problem will contain “obvious” true decision variables and “indirect” decision variables An obvious decision variable is the pressure set point of an HP oil separator This is a controllable variable whose set point can be adjusted to the values indicated by an optimizer An “indirect” decision variable may be the speed of a centrifugal compressor This variable in most cases will be a controller manipulated variable Indirect decision variables are sometimes a convenient (efficient) way of solving the mathematical problem Indirect variables shall be carefully analyzed as their manipulation may require changing the configuration of corresponding process controllers • Constraints – these are as important as decision variables Some constraints are easily identifiable from the operating personnel experience and/or from the set points of field alarms Others are “fuzzy” constraints (e.g the capacity of an oil/gas separator, the maximum allowed flow rate in a gas pipeline) whose values are established based on existing design and/or best operating practices Sometimes these “fuzzy” constraints happen not to be SPE 118454 real hard constraints, and they can be challenged Significant gains can sometimes be achieved by challenging these constraints in a step-wise manner • Objective function – sometimes the mathematical formulation of the objective function is coincident with the desired business objective, others it is a combination of the process economics and a number of weighted (penalized) key process constraints re-cast inside the objective function to better solve the mathematical problem A proper formulation of the objective function requires extensive system testing a suite of relevant business scenarios covering the span of process situations that the algorithm will have to face Optimization provides brand new ideas about how to operate or revamp an existing process, instead of simply repeating a schema that has been used before It provides a solid engineering baseline to benchmark individual process engineer’s estimates, which often are strongly dependent on their particular experience and preferences Moreover, the optimization leads to a significant rationalization of engineering efforts, time and engineering costs for competitive plants Beyond Optimization Integration of process engineering with business processes is vital for E&P companies looking to optimize the recovery from what they already have It provides greater transparency and accuracy in decision-making Significant benefits can be achieved by transforming asset data into knowledge, from control room to the boardroom, and to concurrently drive consistent decision-making in engineering, operations and business Data-centric drives efficiency gains, accuracy and collaboration across different groups Equipment Condition Monitoring As mentioned before, the main objective of a modeling/optimization package is to evaluate the actual operation of the plant equipment to the optimum operation of that equipment In order to have meaningful results, the Model needs to be calibrated periodically to reflect the loss in performance due to fouling or degradation and to approximate its answers to reality The equipment condition monitoring functionality is a by-product of the model calibration (i.e model tuning or parameter estimation) feature of the system The economic penalty ($/d) of using those new parameters can be traced and analyzed each time the model calibration system updates the equation coefficients (tuning factors) - to reflect changes in equipment performance due to fouling or degradation This information may be used by the plant engineers and management to schedule maintenance cycles and/or equipment replacement What-If Studies What-if studies can be used to evaluate the impact of operating the asset equipment at significantly different operating conditions than current These differences can range from changes in input data to different model parameters to different equipment Examples of studies and model uses are: • • • Studies to evaluate the impact of major changes on specific set points Evaluate the impact of starting idled equipment or stopping some currently operating equipment Constraint sensitivity analysis (what is the value of eliminating a constraint) The initial data for a what-if study could come from a plant data tuned system, an optimized system or an existing, previously saved, what-if study Results The rigorous model of the complete asset has been constructed and spot checked against plant data taken over a period of months The results of these runs show that the model is in good agreement with the measured plant data and most of the instruments can be used for model calibration Off-line runs of the optimizer, using the validated plant model, indicate that the expected level of benefit can be achieved 10 SPE 118454 A major benefit from the process of producing and calibrating any detailed model is to show which instruments and equipment are working reliably and areas of the plant which require attention or adjustment The project has identified some meters which had shown the incorrect readings Some of these readings are current active optimization variables so their accuracy directly impacts the plant economics For example, ambient temperature needs to be correctly specified as gas turbine power output is affected by it High ambient air temperature would result in lower power The ambient air temperature exceeds 22 ºC on CA-CWP for about three months of the year During this period gas compression capacity is reduced For the remainder of the year, gas compression capacity can at least be maintained and may even exceed design capacity The Azeri system exhibits a rather complex network of pressure flow interactions The gas and the oil compression, pumping and transport systems are tightly coupled The high variability in the climate conditions during the year in the Sangachal area plays an important role in the performance (and capacity) of the gas conditioning and oil stabilization facilities Capacity constraints therefore can be observed anywhere from the well heads to the terminal delivery points The specific business requirements of a particular day and/or the equipment availability strongly affect the operating philosophy to follow in order to best utilize the facilities Usage Scenario: Maximize System Capacity In most cases the ability of the facilities to utilize the associated gas either by injecting it into the reservoir or by transporting it onshore and dew-pointing it to the pipeline specifications for distribution inlands constraint the maximum oil production that can be attained Four factors influence that ability and therefore affect oil handling capacity: • • • • The performance of the gas injection compressors: o number of trains in service o performance of the gas turbine drivers (affected by ambient conditions) The pressure setting of the CWP gas dehydration column CWP receives all produced gas streams The lower the pressure the higher the drawdown from the wells, but also the lower the capacity of the CWP to Sangachal gas export pipeline and also lower the gas injection compressors discharge pressure resulting in less volumes been delivered to the ground This trade-off is among the most important factors determining the maximum production quotas that can be attained The performance of the propane refrigeration circuits installed in the Sangachal Gas Dew Pointing unit The propane condensers are seriously affected by the ambient conditions The pressure setting of the LP separators in CA, EA and WA This affects the amount of generated flash gas and hence the capacity utilization of the highly loaded CWP gas pipeline Controlling the extent of oil stabilization can serve to best use available transport capacity In a typical case where the optimizer was run to predict the maximum attainable oil production for a particular day of interest, it was found that the pressure set point of the CWP dehydration pressure could be lowered by around %, which would allow lowering the local EA and WA dehydrators pressure set points by ca 2-3 bars and consequently to increase production an average of 3% The extra associated gas is handled by: • • Reducing the amount of flash gas generated in EA and WA by increasing the LP separators operating pressure, and Lowering the Sangachal slug catcher pressure to its minimum value of 41 bar in order to maximize export volumes The usage of the Azeri Field Optimizer has helped the asset to better understand the overall performance of the facilities and to propose refined operating strategies The optimizer has been executed for a number of different asset situations, such as: platform shut-down scenarios, specific business requirements as maximizing the amount of exported or injected gas, varying ambient conditions, et cetera In all cases the optimizer has provided sound answers and revealed interesting aspects of the asset behavior that are presently being explored These and other business decisions can be analyzed and justified by executing appropriate AFO case studies Technology Adoption SPE 118454 11 The successful implementation of AFO relies on having high-level management support for the technology This is essential to ensure that, within the asset, sufficient people resources and funds are available to support and nurture the technology Delivered tool/technology along with associated business process was presented to the high level management which was well received and supported Request from management was to convert optimizer results into extra production as soon as possible in order to release required additional funds and resources which will allow to progress further with AFO embedment in to Asset organization First optimization case was successfully prepared and put forward for implementation utilizing MOC (Management Of Change) process This case suggests that in current operational conditions export pressure at CWP could be lowered around 10% and still maintain required gas injection and export volumes with production increase up to 15 mbd However, during preparation for the implementation was found out that pressure reduction will decrease the performance of the separators, resulting in a liquid carry over into the gas line with consequent problems for the dehydration system Such performance decrease is not modeled by steady state simulation utilized in the AFO indicating the need for further technology enhancement All of that prompted the Asset to start project for the separation capacities review at different operating pressures which is currently ongoing In addition to the first case, several “what if” cases were successfully ran One of them is the conversion of the wells from HP (High Pressure) to LP (Low Pressure) manifolds Results had good correlations with the results achieved after modification References Mochizuki, S et al 2007 Real-Time Optimization: Classification and Assessment Paper SPE 90213 first presented at the 2004 SPE Annual Technical Conference and Exhibition, Houston, 26-29 September Revised paper published in Nov 2006 SPE Production & Operations Mullick, S and Strathman, M 2007 Operating Excellence – Modeling for Profits Upstream Technology Magazine, December 2007 Stenhouse, B 2000 The BP Harding Daily Optimizer Presented at the Hyprotech 2000 User’s Conference, Amsterdam, November 2000 Stenhouse, B and Goodwin, S 2004 Barriers to Delivering Value from Model Based Gas Field Production Optimisation Paper presented at the Gas Processing Association Conference, Dublin, 20 May 2004 Stenhouse, B 2006 Learnings on Sustainable Model-Based Optimization – The Vallhall Optimizer Field Trial Paper SPE 99828, April 2006 Truschinger, J 2004 Focus on People Paper presented at the SPE Digital Energy Conference, Houston April 2004  ... and gas injection quotas The Azeri Asset The Azeri oil and gas production asset is within the top BP investments world-wide and possibly one of the most complex oil systems around the world The. .. gas injectors can be solved more efficiently by re-organizing the engineering equations governing the performance of the compressors and the injectors and solve them analytically than letting the. .. year in the Sangachal area plays an important role in the performance (and capacity) of the gas conditioning and oil stabilization facilities Capacity constraints therefore can be observed anywhere