Skip to main content
European Commission logo print header
Contenuto archiviato il 2022-12-27

EXPERT SYSTEM FOR MONITORING/DIAGNOSTIC OF MULTIPHASE TRANSPORTATION PROCESSES

Obiettivo

The design, manufacturing and testing (in a real oil well) of an advanced system for monitoring and diagnostic of flow parameters in multiphase transportation systems, and for the process control. It is composed by several not intrusive sensors and measurement devices located in different positions in the pipeline, and by an expert system which collects and analyze all the process data, contains its own models for process diagnostic and control, and can communicate with a simulation code. The main innovative aspects are the integration of not intrusive measurement devices and advanced diagnostic technologies with a high capability expert system, which allows better accuracy in the process monitoring and control. It is expected a considerable energy saving due to a more profitable exploitation of the wells, particularly the " marginal" oil fields, a more efficient management of the transportation systems, and a better predictive maintenance with a decrease of plant shutdown.
The global expected results are summarized on the two following goals :
a) To achieve high accuracy, on line, not intrusive measurement systems. The increasing of the accuracy is obtained through the use of cross validation procedures based on the fusion of three different approaches: the direct measurements (by a spool piece), the flow modelling (based on process data analysis) and the statistical noise analysis (flow pattern identification and mutlivariate analysis of acoustical noise on a ckocke valve).
b) To achieve an integrated diagnostic tool (knowledge and rules based) for the global monitoring system. This feature provides to a wide possibility of application in multiphase processes where a complete knowledge (and monitoring) of the transportation system status is required. Information collected by measurements in some locations of the transport system are re-utilized by measurement systems on other locations and by the expert system in order drive control systems or monitoring the not directly measured zones of the process.
The first result obtained at this phase of the development is the conclusion of realization and laboratory calibration of all the instrumentation tools. They have been connected with the acquisition system and with the software modules for data processing in order to obtain a real time measurement.
Finally, all these tools (spool piece, flow pattern analyser, acoustic and fluidynamic ckoke valve analyser) have been installed at the Trecate oil field for qualify their performance in a real conditions industrial plant.
The second result obtained is an improvement of the accuracy of the measurements of flowrates due to the development of new tasks :
a) Cross correlation analysis between gamma and venturimeter signals: using an innovative tool for cross correlation calculation has been obtained an improvement of the accuracy from 17% to 10% on the maximum error on liquid flowrate.
b) Using a new design for the venturimeter and newcorrelations for liquid flowrate and slip a 10% maximum error has been obtained on flow rate measurement.
c) Using a fuzzy technique, the expert system has been able to decrease the errors of the 50% of their value (tested on field data previously recorded and also on laboratory data), increasing very much the robustness of the measurement in respect to instrumentation or algorithms breakdown or operative condition far for calibration tests.
d) The use of a very powerfull computer code, like OLGA, with a continuous assessment of the code parameters on the base of the measurement (the tuning procedure of the D-SPICE program) allows the knowledge of the flowrate along the pipeline with high level of accuracy and gives important information about the line diagnostic.
The need of more accuracy in process data measurement, the need of larger amount of data for models and computer codes improvement and assessment, and the need of higher levels of automation for the control of multiphase transportation systems, require the development of integrated systems, designed and assembled as a modular and flexible equipment.
To achieve these requirements it is needed together with advanced data analysis packages and phenomena simulation models - a knowledge based system, which has the following characteristics: a clear knowledge representation easy to modify; a field interface in real-time; the capability of carrying out temporal and asynchronous interfaces for communication between the operator and the field.
The monitoring/diagnostic system can be utilized in several ways, using a part of its actual measurement modules or the whole set of sensors; furthermore it can be easily expanded and improved, employing other sensors and instrumentation and developing the physical and statistical models, the data bases, and specially the knowledge base for process control.
The lack of knowledge and formalization on some transport phenomenologies is the main limitation to the designing and managing of lines. The acquisition of these knowledge requires the experimentation on large scale and the use of measurement stations that are highly qualified.
The slugging phenomena, that relate to slugging induced by the fluidymanic conditions or by line geometry, cause high vibrations and high pressure drops along the line, and furthermore causes operational problems to the separator. In order to reduce the operational problems due to the slug flow several alternatives are actually used, as slug catcher, chocking, production reducing, that however need simulation codes to predict flow patterns behaviour and to define the control strategies.
Most experimentations on slug flow have been carried out on air-water mixtures in small diameter and short pipelines. On the basis of this experimentation a large series of models and correlations have been developed. Recent data from field lines have been published that show a certain disagreement with the slug length and velocity and pressure drop calculated by such a correlations. Particularly these data show an evolution of the slugs along the line that is very difficult to predict.
The monitoring/diagnostic system represents a way to solve these kind of problems, for its innovative characteristics mainly related to the integration and to specific advanced methodologies for increasing the rangeability and the accuracy of the measurement systems. The use of a knowledge and rules based tool allows a wide flexibility to industrial evolution and integration with other supervision and control systems. The integration of the measurement systems allows a better accuracy obtained through cross-validation procedures, a larger rangeability because of redondant different instrumentation.

Invito a presentare proposte

Data not available

Meccanismo di finanziamento

DEM - Demonstration contracts

Coordinatore

Ente per le Nuove Tecnologie, l'Energia e l'Ambiente (ENEA)
Contributo UE
Nessun dato
Indirizzo
Via Anguillarese 301
00100 Roma
Italia

Mostra sulla mappa

Collegamenti
Costo totale
Nessun dato