Gestión de las operaciones de importación: Flujo grama

Design for reliability

The following general principles should be observed.

Element selection. Only elements with well-established failure rate data/models

should be used. Furthermore some technologies are inherently more reliable than others. Thus an inductive LVDT displacement sensor (Chapter 8) is inherently more reliable than a resistive potentiometer; the latter involves a contact sliding over a wire track, which will eventually become worn. A vortex flowmeter (Chapter 12) involves no moving parts and is therefore likely to be more reliable than a turbine flowmeter which incorporates a rotor assembly.

Environment. The environment in which the element is to be located should first

be defined and the element should consist of components and elements which are capable of withstanding that environment. Thus the diaphragm of a differential pres- sure transmitter on a sulphuric acid duty should be made from a special alloy, e.g. Hastelloy C, which is resistant to corrosion.

Minimum complexity. We saw above that, for a series system, the system failure

rate is the sum of the individual component/element failure rates. Thus the number of components/elements in the system should be the minimum required for the system to perform its function.

Redundancy. We also saw that the use of several identical elements/systems con-

nected in parallel increases the reliability of the overall system. Redundancy should be considered in situations where either the complete system or certain elements of the system have too high a failure rate.

Diversity. In practice faults can occur which cause either more than one element

in a given system, or a given element in each of several identical systems, to fail simultaneously. These are referred to as common mode failures and can be caused by incorrect design, defective materials and components, faults in the manufacturing process, or incorrect installation. One common example is an electronic system where several of the constituent circuits share a common electrical power supply; fail- ure of the power supply causes all of the circuits to fail. This problem can be solved using diversity; here a given function is carried out by two systems in parallel, but each system is made up of different elements with different operating principles. One example is a temperature measurement system made up of two subsystems in parallel, one electronic and one pneumatic.

Maintenance

The mean down time, MDT, for a number of items of a repairable element has been defined as the mean time between the occurrence of the failure and the repaired element being put back into normal operation. It is important that MDT is as small as possible in order to minimise the financial loss caused by the element being out of action.

There are two main types of maintenance strategy used with measurement system elements. Breakdown maintenance simply involves repairing or replacing the element when it fails. Here MDT or mean repair time, TR, is the sum of the times taken for a number of different activities. These include realisation that a fault has occurred, access to the equipment, fault diagnosis, assembly of repair equip- ment, components and personnel, active repair/replacement and finally checkout.

Preventive maintenance is the servicing of equipment and /or replacement of

components at regular fixed intervals; the corresponding maintenance frequency is m times per year. Here MDT or mean maintenance time, TM, is the sum of times for access, service/replacement and checkout activities and therefore should be significantly less than mean repair time with breakdown maintenance.

7.2

Choice of measurement systems

The methods to be used and problems involved in choosing the most appropriate mea- surement system for a given application can be illustrated by a specific example. The example used will be the choice of the best system to measure the volume flow rate of a clean liquid hydrocarbon, range 0 to 100 m3_h−1_{, in a 0.15 m (6 inch) diameter}

pipe. The measured value of flow rate must be presented to the observer in the form of a continuous trend on a chart recorder. The first step is to draw up a specification for the required flow measurement system. This will be a list of all important par- ameters for the complete system such as measurement error, reliability and cost, each with a desired value or range of values. The first two columns of Table 7.4 are an example of such a ‘job specification’. As explained in Chapter 3, system measure- ment error in the steady state can be quantified in terms of the mean + and standard deviation σEof the error probability distribution p(E). These quantities depend on the

imperfections, e.g. non-linearity and repeatability, of every element in the system. System failure rate λ and repair time TRwere defined in Section 7.1. Initial cost CI

is the cost of purchase, delivery, installation and commissioning of the complete system. CRis the average cost of materials for each repair.

Parameter Job System 1 System 2 System 3 System 4 specification Orifice plate Vortex Turbine Electromagnetic Measurement +≤ 0.25 0.2 0.1 0.03

error σE≤ 0.8 0.7 0.3 0.1

(at 50 m3_h−1_{) m}3_h−1

Initial cost CI≤ £4000 3500 3000 4200 _{Not technically}

Annual failure rate λ ≤ 2.0 1.8 1.0 2.0 feasible failures yr−1

Average repair time TR≤ 8 h 6 5 7

Material repair cost CR≤ £200 100 100 300 5 4 4 4 4 6 4 4 4 4 7 Table 7.4 Comparison

table for selection of flow measurement system.