The parameters chosen for adjustment determine what the match means. Adjusting parameters that are genuinely uncertain, such as aquifer strength, fault transmissibility, vertical communication and relative permeability endpoints, produces a model that has learned something about those quantities. Adjusting parameters that are well constrained by independent measurement, such as porosity in a well-logged reservoir or fluid properties from good PVT, produces a model that has absorbed error from elsewhere. The first is history matching; the second is curve fitting with a reservoir simulator.

Mismatch has diagnostic signatures. A model that under-predicts pressure everywhere suggests too little connected volume or too little support. A model that matches pressure but breaks through water too early suggests a permeability contrast or a barrier issue rather than a volume issue. A single well that will not match while its neighbours do suggests a local completion, damage or allocation problem rather than a reservoir description problem. Reading these signatures before adjusting anything is what makes the process converge rather than oscillate.

Non-uniqueness is not a defect to be eliminated. Many parameter combinations will match a given history, and which one is correct cannot be determined from the history alone. The appropriate response is to retain several matched models that differ in the parameters that matter for prediction, and to carry them forward so that the forecast is a range rather than a line. A single matched model presented as the answer conceals uncertainty that the data cannot resolve.

Finally, the history match should be tested rather than declared. Holding back the most recent period of production, matching to the remainder and then predicting the held-back period is a blind test that reveals whether the model predicts or only reproduces. Models that pass this test forecast with credibility; models that have never been tested this way frequently do not.

By the end of this training, participants will be able to:

• Prepare and quality control production, pressure and surveillance data for history matching
• Define misfit functions and weighting schemes and understand their effect on the match
• Select history matching parameters based on genuine uncertainty rather than convenience
• Parameterise the model appropriately including regional, zonal and multiplier approaches
• Sequence a history match from field scale through region to individual wells
• Diagnose mismatch by its signature and identify the likely physical cause
• Manage non-uniqueness by retaining multiple matched models
• Test predictive capability using blind periods and independent data
• Assess prediction confidence and communicate forecast uncertainty from a matched model
• Document a history match so that it can be reviewed, audited and updated

The training is worked on reservoir models with real production histories. Participants prepare data, define misfit, select parameters, and carry out matching in sequence, with the effect of each decision visible in the result. Poorly conducted matches are examined alongside well conducted ones for the same field, and the resulting forecasts compared, which demonstrates the consequence of undisciplined parameter adjustment more effectively than argument. Blind period testing is applied to matched models. Mismatch diagnosis is practised on models exhibiting characteristic failure signatures.

Organisations sending participants to this training will:

• Produce matched models whose forecasts are more reliable
• Reduce time spent on history matching through better sequencing and diagnosis
• Reduce the incidence of models matched by parameter adjustments that cannot be defended
• Improve forecast uncertainty representation through multiple matched models
• Strengthen review of history matching work performed internally and by contractors
• Improve the value obtained from surveillance data by using it in the match

Participants will:

• Approach a history match systematically rather than by trial and error
• Select and defend the parameters adjusted in a match
• Diagnose mismatch and identify its physical cause
• Retain and use multiple matched models to represent uncertainty
• Test whether a matched model actually predicts
• Build a capability central to reservoir simulation practice

• Reservoir simulation engineers
• Reservoir engineers using simulation output for forecasting and development planning
• Reservoir modellers and geoscientists supporting dynamic modelling
• Reserves and evaluation engineers relying on simulation forecasts
• Technical staff reviewing simulation studies
• Consultants delivering reservoir modelling work
• Graduate engineers entering simulation roles

Upon successful completion of this training course, delegates will be awarded an official Certificate of Completion issued by the Petroleum Engineers Association (PEA), an ISO 9001:2015 certified training organization. The certificate carries 10 Credits and formally records the total learning hours completed.

Each certificate is signed by the Course Facilitator and the CEO of the Petroleum Engineers Association, and serves as verifiable proof of professional training that delegates can present to employers and professional bodies worldwide.