Research & Development

The issues include:

• Non-ideal behaviour of CO2 at high pressures
• Impact of impurities from different sources and industries
• Equitable allocation of high compression power consumption
• The “Long Pipeline Effect”
• Calculation of uncertainties

The below presentation exploring the allocation concepts for CCS was delivered by Phil Stockton at NFOGM and Tekna Hydrocarbon Management Workshop in 2022:

The KTP began in 2019, and the goal was to develop a software product that can help clients identify measurement errors and/or value leakage in hydrocarbon processing and prevent subsequent errors in allocation.

The method can be integrated within existing systems and uses established data reconciliation techniques and machine learning to determine measurement quality of complex systems. The method is not limited to hydrocarbon processes and can be applied to, for example, water networks. 

This project led to a journal paper and several papers which have been presented at international conferences.

In collaboration with DP Diagnostics, Accord have developed an approach to reduce measurement uncertainties associated with differential pressure flow meters. This patent pending method uses the redundancy provided by an additional tap within the meter. As defined, the method can also be applied to other meter types and technologies where a redundant measurement is available e.g. ultrasonic meters.

In the 1960s, the Kalman filter was applied to navigation for the Apollo Project, which required estimates of the trajectories of manned spacecraft going to the Moon and back. The Kalman filter is an optimal estimation technique that uses dynamic mathematical models and physical measurements to obtain the most probable estimates of underlying variables as they vary with time. It incorporates both the measurement and also model uncertainties into its highly efficient estimation algorithm. Today, the use of the Kalman filter is extremely widespread throughout many science and engineering applications.

Accord have carried out feasibility studies on the application of the Kalman filter to the problem of estimating (and hence allocating) well production based on intermittent and possibly poor quality measurement data from well tests. The technique recognises the increasing uncertainty in the estimated well flows as the time since the last well test elapses. It also utilises the additional measurement of the wells’ daily, aggregate, commingled production as it exits a process. It can take advantage of process upsets in that when wells shut in, the commingled production of the remaining wells is measured and the drop in production is an estimate of the shut in wells. This information is smoothly incorporated into the well estimates and propagates forward in time.

A simple expression of uncertainty describes the probability of understatement or overstatement in a measured or allocated quantity. A common but overly simplistic approach is to multiply this by the throughput as a measure of the potential loss. However, using the normal distribution, we can derive a more representative and justifiable calculation of the exposure.

Accord have derived a method that expresses the risk of loss associated with measurement or allocation uncertainty in terms of a monetary value. This is utilised in cost benefit analyses to allow more expensive meters, or methods of allocation, with lower uncertainties, to be compared against cheaper alternatives. The approach has subsequently been cited in several Flow Measurement Workshop papers and appears in both the NORSOK Standard I-106 and the UK Measurement Guidelines.

There are two main approaches to systems of allocation that include long pipelines. The first accounts for each user’s hydrocarbons within the pipeline itself. The second method ignores the transit time in the pipeline and allocates the metered quantities exiting the pipeline based on the metered quantities input into the pipeline on the same day; using this approach parties will not be allocated precisely what they input to the pipeline on a day, but over a period of time there is an expectation that any daily gains and losses will even themselves out.

However, this is generally not true. In fact, parties will be systematically under and over allocated hydrocarbons due to the mathematics of the allocation agreement.