(Modifid)condition m0nitoring of longwall face supportd

Condition monitoring of longwall face support

RAVIKANT KUMAR 111MN0472

1. Introduction

2. Condition monitoring needs

3. Present status

4. Method of condition monitoring

5. Applicability

6. DATA processing and Interpretation

7. Conclusion

8. Refrences

TOPICS TO BE COVERED

INTRODUCTION

Despite the most refined maintenance strategies, equipment failures do occur.

The degree to which an industrial process or system is affected by these depends on the severity of the faults /

failures, the time required to identify the faults and the time required to rectify the faults.

Real-time fault detection and identification (FDI) and other methods offer maintenance personnel the ability to minimize,

and potentially eliminate one or more of these factors, thereby facilitating greater equipment utilization and

increased system availability.

What is condition monitoring ?

Condition monitoring (or, colloquially, CM) is the process of monitoring a parameter of condition in machinery (vibration,

temperature etc), in order to identify a significant change which is indicative of a developing fault.

It is a major component of predictive maintenance.

The use of conditional monitoring allows maintenance to be scheduled, or other actions to be taken to prevent failure and avoid

its consequences.

Condition monitoring has a unique benefit in that conditions that would shorten normal lifespan can be addressed before they

develop into a major failure. 

What is to be done?

Understand the failure we want

to predict.

Identify a strategy that is appropriate to

the failure mode.

Identify a reliable indicator.

Action and review result.

When it is wrong to use CM?

When it doesn’t tell anything useful.

When we don’t understand the result.

When it adds cost to the cost of operation for no

operational improvement.

When we don’t use the result.

Longwall Maintenance-Current practices

As with all forms of mining, production downtime represents massive losses in potential revenue.

An effective maintenance program is seen in the industry as the key to a profitable and sustainable

longwall.

Maintenance is broadly categorized as either planned or reactive.

CONT…..

Reactive (or breakdown)

maintenance refers to the repair or

replacement of failed or faulty equipment

which interrupts longwall production.

Typically, one shift per week is assigned to

routine planned maintenance, with major overhauls of

machinery occurring at the end of each panel.

• It is these actions which are reported in the maintenance log of all unscheduled downtime.

• This maintenance is attended to by longwall operators, as each team typically includes specialists in both mechanical and electrical repairs.

Longwall Monitoring

Equipment monitoring is playing an increasingly significant role in the modern longwall.

By monitoring it is meant both the determination of the overall state of plant as well as measurement of individual properties; for example, the running

temperature of gearbox.

Condition monitoring of longwall equipment takes on two main forms: on-line and off-line.

On-line monitoring includes all sensor measurements recorded and transmitted to the surface using a PLC-driven, SCADA network.

All other measurement and monitoring, including regular maintenance inspections are classed as off-line.

On-line Monitoring

The on-line monitoring systems at longwall mines consist of a network of sensors and switches, which may be monitored from the surface or by operators

underground.

The interface allows users to view the overall operation, as well as monitor individual pieces of

machinery.

The system is designed to indicate simple faults such as an over-temperature trip by displaying

messages and changing the colour of the particular machine at fault.

ONLINE MONITORING

• SCADA• P-F Interval or Fault Detection Threshold(FDT)

Some popular methods of online monitorin

g:

•RATE OF MOVEMENT

•MICROSEISMIC MONITORING

OTHER IMPORTA

NT METHOD

1) SCADA-Supervisory Control And Data Acquisition

A SCADA system

performs four functions:

1. Data acquisition

2. Networked data

communication 3. Data

presentation

4. Control

Application of SCADA in longwall monitoring

This system is very expensive, both to purchase and maintain.

At present, the display is only looked at when production personnel indicate that the longwall

has stopped due to an unknown fault.

Data that is recorded, is rarely used, and undergoes no more than a visual analysis.

Over 300 analogue sensors are typically monitored by these SCADA systems.

Cont…..

In the past, data storage was considered one of the major pitfalls of archiving

all that was sampled.Today, however, this has been mitigated by

the inexpensive and very large capacity computer hard drives.

SCADA data is very very significant for Fault Detection AND Identification(FDI)

2) P-F Interval or FDT Method

A common curve that illustrates the behavior of equipment as it approaches

failure is the P-F curve.

The curve shows that as a failure starts manifesting, the equipment deteriorates to the point at which it can possibly be

detected (P).

If the failure is not detected and mitigated, it continues until a "hard" failure occurs

(F).

Cont…….

It is the window of opportunity during which an inspection can possibly detect the imminent failure

and address it.

P-F intervals can be measured in any unit associated with the

exposure to the stress (running time, cycles,

miles, etc).

Example

If the P-F Interval = 200 days and The item will fail at 1000 days,Then, the approaching failure begins to be detectable at 800 days.

Significance of P-F Interval

Reducing Equipment Downtime: The case for FDI

Available techniques to minimise equipment/process downtime are examined first.

After all, any maintenance program has the principle

goal of achieving the greatest possible equipment

utilization.

Available Techniques

1) Preventative Maintenance

The most common way that industries attempt is to improve machine availability is through preventative maintenance.

This consists of regular inspections, regular replacement of lubricant, filters etc. and occasional reconditioning of parts.

Preventative maintenance has proven very successful in prolonging the life of machinery, as well as showing

improvements in performance, such as the mean time between failures (MTTF) and mean time to repair (MTTR).

Cont…

2) Design Modification • It is possible to redesign machinery to alter

it’s inherent reliability.• That is, based on industry experience of

certain faults, redesign parts or entire machinery to reduce or stop the occurrence of the fault.

• Typically, machine design is the responsibility of the manufacturer, and is a very time consuming and costly operation.

3) Failure Prediction

Similar to weather forecasting, the knowledge of an imminent failure of a piece of machinery would allow machine operators to cease or change operation to minimize and potentially avoid any subsequent downtime.

There is an inherent risk in such a concept in that false or misleading predictions could cause additional downtime, and quickly lose operator confidence.

For a predictive system to be effective it must produce not only a prediction of an imminent fault, but provide sensible reason as to the prediction, so that operators may make an informed decision whether to act or ignore the suggestion.

Data processing and Interpretation

Since the data-driven approach does not assume an understanding of individual faults on data properties, large amounts of data must be collected.

If possible, the data should span an operational time for which a large number of example faults occur and exhibit an ordinary proportion of faults/failures per unit time.

The first concern in processing data is however the estimation of missing entries.

Cont….

In order to generate a list of classifications, one must attempt to determine the ‘state’ of the longwall at each observation.The only record of the activity of the longwall is the maintenance log, which details all unscheduled downtime at the longwall face.

Table 1 is an excerpt of the maintenance log corresponding to the condition monitoring data discussed earlier.

It records the time that the delay began and the duration of downtime experienced.

Conclusion

All longwall mines employ preventative maintenance.

It is one of the largest sub operations at any mine, and proves effective in that when less attention is paid to maintenance, more faults occur.

The optimum level of planned maintenance is difficult to determine because not all failures are age-related.

In fact, many failures follow an exponential distribution with a uniform failure rate that is not related to age.

Refrences

• Willsky, A S, (1976) A survey of design methods for failure detection in dynamic systems, Automatica, 12:601-61

• Bongers, D, (2004) Development of a Classification System for Fault Detection in Longwall Systems, PhD Thesis, The University of Queensland .

• Frank, P M, (1990) Fault diagnosis in dynamic systems using analytical and knowledge- based redundancy - a survey and some new results, Automatica, 26(3): 459-474 .