Table of Contents ECNDT '98 Session: Chemical, Petrochemical

Making NDT predictive G. COCO, A. MARINO PROCONTROL S.R.L. VIA INAMA, 21 , 20133-MILANO, ITALY TEL.: +39 (02) 70101050 FAX: +39 (02) 70100504 Corresponding Author Contact: Email: pcontrol@procontrol-ndt.com

TABLE OF CONTENTS

Foreword Inspection plan Piping and Equipment thickness gauging Data entry coherence Thickness gauging survey procedure Customizing Hardware

A study by the Italian NDT Group Procol Procontrol to manage ultrasounds thickness surveys made over pipelines and vessels in chemical and petrochemical plants. The goal of this study is to forecast the residual life of plants by means of an appropriate software called Life Extension.

Foreword

In Chemical and Petrochemical Industries dangerous items such as Raw and End Products, Additives and Residuum after process are treated and processed; this may results in a strong hazard for man and his environment.

Moreover, these agents have an aggressive impact on the machinery and equipment which process them. They produce different kinds of flaws (Cracks, Blisters, Stress_Corrosion), and make machinery and equipment loose part of their original thickness. As a consequence failures, undesired leaks of products, and other potentially dangerous situations may occur (explosions, pollution, ...)..

Manufacturing conditions tend to enhance these hazards because of the Heavy Conditions at which plants have to work:

• Continuous cycle processing
• High Pressures
• High Temperatures
• On-Off Transient status

Other risks are represented by Processing, Stock Transfers, Storage, Thermal and Mechanical stress.
Consequently all these elements contribute to make the whole processing system RISKY.

Inspection plan

In order to have all these threats correctly monitored, a proper Machinery and Equipment Inspection Plan is crucial and must be the basis for a serious predictive analysis that should ensure:
• Better reliability and safety of plants.
• Functionality and value of plants over the years.
• General reduction of maintenance costs.
• Forecasting of the residual life for each plant circuit.

Such an Inspection Plan should also include:

• Visual Inspections.
• Thermograph Inspections.
• Expansion and lining checks.
• Vibration, noise and overheating checks.
• Alarm and blocking systems checks.
• Seal system Periodical checks.
• Piping and Equipment thickness Gauging.

Piping and Equipment thickness gauging

Knowing the residual thickness of machines and equipment means toforecast their residual life and thereof to be able to replace them before they crash.

This way it will be possible to avoid Leacks of products and dangerous substances as well as undesired and unexpected plant stops.
An increase in the number of spot surveys doesn't necessarily mean a more complete answer: "The more I do, the more I know" has always been the philosophy underlying this type of controls.
Instead better results can be achieved adopting a robust procedure that leads to the best control distribution.

This can be obtained if controls are planned and performed following these three fundamental principles :

• Selection Criteria
• Accuracy of Survey's
• Data Analysis

Selection Criteria include locating the spots to be checked and defining the extent and frequency of surveys; they will be determined considering the following aspects:

• Project characteristics, (such as pressure, temperature, working hours, year of construction and expected life extension)
• Type of fluid
• Direction of fluid
• Highly critical areas
• Dangerous elements for enviroment and operations
• Results of previous surveys

As far as theAccuracy of Survey is concerned, the equipment currently used for this kind of measurements is based on accurate and sophisticated Technologies.
On the other hand, technicians must be aware of the philosophy underlying surveys. They have to know why surveys are carried out. To this purpose the following aspects should be considered:

• Specific skill
• Training
• Strict application of codes and procedures
• Thorough knowledge of testing and analysis philosophy

Data Analysis

This is the real innovative aspect that we have been successfully developing in recent years and that has produced the LIFE_EXTENSION software with the idea to help all technicians and inspectors involved in internal corrosion monitoring of pipelines and vessels.
We obviously do not want, and surely would not be able, to take the place of specialists who manage plants and whose knowledge and experience are the most valuable asset of companies.
However this software can provide a supporting tool an aid for a better management of the thousands of thickness surveys usually performed for corrosion monitoring, and tipically printed on paper and left on a shelter to gather dust.

The philosophy of this software is, on the contrary, to capitalise the survey results and make them ready to be used and compared in the years to follow.

The goals of our Life Extension software can be summarised as follows:

• To forecast the expected residual life of machines and equipment for each plant circuit.
• To plan in advance maintenance standstill times for the different lines or parts thereof that will be replaced during a standstill.
• To Analyse, for each plant circuit, the corrosion rates of each line in order to make the necessary changes to plant machines and equipment, design and running technique.

To achieve these goals, our software is able to perform the following functions:

• Storage of up to 15 surveys in active archives.
• Display of 6 consecutive survey on the screen
• Unlimited Transfers of active archives to hystorical archives .
• Adding and/or cancelling surveys - even intermediate ones - with automatic re-numbering and updating of all calculations (corrosion rate and residual life)
• Data comparison between two surveys at any date and immediate calculation of the following parameters:
  • Yearly relative spot corrosion rate.
  • Mean corrosion rate at each survey.
  • Mean square deviation.
  • Percent of lost overthickness.
  • Residual spot life.
  • Residual minimum life.
  • Percent of comparable spots.
  • Percent of controlled spots.
  • Minimum absolute thickness set.

Example1:

Section Parameters		Surveys
Rated Thickness	6,3 mm	1st Survey - May 1995	6,2 mm
Rated Corrosion Overthick.	1,3 mm	2nd Survey - May 1996	6,0 mm
Rejection Thickness	5,0 mm	Thickness loss	0,2 mm
Time period	1 Year	Mean Corrosion Rate	0,2 mm/Year

If the same processing condition are kept, this section will reach the rejection thickness in 5 years from May 1996, that is to say May 2001

Should a comparison between the two last surveys not be possible, refrence can be made to previous surveys or even to the design data.

Data entry coherence

It is very important that data entered in the data bases of the software are correct and are not misunderstood.

That's why it is possible to insert customised filters; INPUT TABLES not only refer to operational parameters (such as, for instance, Code of ultrasound operator, Material code, Piping class, Survey frequency code, Rated diameters, Rated thicknesses, Corrosion overthicknesses, Rejection thicknesses, Survey position code), but they can also provide information regarding the access to pipelines (Unfeasible spot code, Line code, Insulation code, Scaffolding code, Junction code). This allows to anticipate the estimate of costs related to the survey.

Thickness gauging survey procedure

The use of the Life Extension program requires a thickness gauging survey procedure that must not be changed between two consecutive surveys: it would be otherwise impossible to perform a survey of the same spot years later and this would result in removing the accuracy of calculations.

The procedure will also account for:

• Encoding.
• Procedure.
• Timing.

Encoding

Diversification of survey spots for every piping section is required.

Some sections are more critical because of:

• Direction of fluid
• Orientation
• Exposure

It is useful to consider that, with the same fluid flooding, some areas are more critical because of:

• Extrados, instead of intrados of a curve
• After the centre line of the curve in the direction of fluid rather than before

Examples:

Encoding for a vertical 90° curve

Encoding for a vertical "TEE" insertion

Operational procedure

Surveys must always be made in the same position, in order to assure the accuracy of all thickness measurements even after years. The main reason why this procedure must be followed is that corrosion rate calculations must be based on the comparison between surveys of the same spot at different times and not on statistical calculations.
(A Thickness survey in the same spot even after years is absolutely feasible and reliable if the right procedures are applied)

Timing of surveys

It shall be fixed according to:

• The criticalities of the line.
• The corrosion rate recorded in the recent past.
• The comparison of corrosion rates of different lines - even located in different plants - but where the same process conditions can be envisaged.
• If a change in local operating conditions occurs the timing must be changed accordingly .

Customizing

The software uses customizable algorithms for:
• Cross-analysis between surveys carried out at any date.
• Getting information even if the comparison between two surveys should be impossible.
• Establishing the minimum residual thickness under which no other analysis will be allowed.

Example 2:

Section Parameters		Surveys
Rated Thickness	6,3 mm	1st Survey - May 1995	6,2 mm
Rated Corrosion Overthick.	1,3 mm	2nd Survey - May 1996	6,2 mm
Rejection Thickness	5,0 mm	Thickness loss	0,0 mm
Time period	1 Year	Mean Corrosion Rate	0,0 mm/Year

If a corrective coefficient is not used, the information given should be: Never ending Life.
The case shown in Example 2 could still be accepted, as the rejection thickness is still far from being reached.
But let's see Example 3:

Section Parameters		Surveys
Rated Thickness	6,3 mm	1st Survey - May 1995	5,0 mm
Rated Corrosion Overthick.	1,3 mm	2nd Survey - May 1996	5,0 mm
Rejection Thickness	5,0 mm	Thickness loss	0,0 mm
Time period	1 Year	Mean Corrosion Rate	0,0 mm/Year

If a corrective coefficient is not used, the information given should be: Never ending Life, while the section is already at the rejection thickness.

Correction procedures have been studied in The Life Extension program in order to always give an end of life date, automatically providing a relative corrosion rate equal to the mean of the whole pipeline (if it exists, it is reliable, because it is derived from the comparison of at least 80% of the spots on the pipelines) or at least 0,1 mm/Year.

Other corrections are activated in case the % of comparable spots is less than 100%; no results is given if the % of comparable spots is below 80%.

Hardware

The Hardware configuration depends on the amount of data to be treated. On the basis of the experience made by those sites using the Life_Extension Software since 1989, we know that 20 Mbyte should be enough for each plant, supposing 100 pipelines for plant and 100 spots for pipelines (DOS version).

The Windows 95^Ò version of the software, which is supposed to be ready by March '98, will obviously need more powerful hardware.