On The FarmNews, Products and Solutions for Farming

Operator Maintenance

Reliability Centered Maintenance Combined With Condition Monitoring Systems In Wind Power Systems

Introduction

            In recent times, the wind power industry has experienced tremendous growth, focusing largely on better economic condition, growing market and development of large wind turbines and offshore farms. Higher availability of wind turbines is not because of good reliability or maintenance management in the wind turbines, it is because of the fast and frequent service. Condition Monitoring System (CMS) is better than traditional maintenance management system for increasing the reliability of the wind turbines. CMS monitors the performance of wind turbine parts e.g., transformer, gearbox, generator etc., and precise maintenance work time is determined with the help of CMS. Corrective maintenance is the maintenance which will take place after the problem arises and preventive maintenance is the maintenance carried out before any problem occurs. Reliability Centered Maintenance (RCM) is the structured approach to find the balance between corrective maintenance and preventive maintenance and to determine plans focusing on the reliability aspects.

                        Reliability Centered Maintenance (RCM) has not yet been implemented as a tool in wind power in Sweden, nor Denmark. US is the world's largest wind power producer. This is however a feasible solution for the wind power industry. In Norway a RCM has been carried out by Vestas. And German Nordex already has reliability oriented maintenance as part of their CMS systems. The application of RCM is a cost-effective planned maintenance program combined with an effective equipment-monitoring plan for the wind power system. This study focuses on the condition monitoring like vibration analysis and oil analysis for the wind power industries, wide-ranging knowledge about the wind turbine, how CMS could be applied to Reliability Centred Maintenance (RCM) and how this has been performed earlier in e.g. hydropower. 

               Few terms are needed to understand the maintenance and RCM as a possible tool in wind power. They are: 

Failure rate: Failure rate is inversely proportional to Mean Time Between Failure (MTBF).

Reliability: Capacity of an item to carry out a required task under given conditions for a given time interval.  

Types of Maintenance

  There are two types of maintenance. They are corrective and preventive maintenance.

1) Corrective Maintenance: It is carried out after the component fails. No maintenance process takes place before the failure of the component. The component which has failed can be repaired or replaced. It can be replaced with the new technology component or the same as the one that was available before it failed. 

2) Preventive Maintenance: It is a proactive maintenance done before the system fails. It reduces the probability of failure of the components. It is a planned and periodical maintenance and carried out in regular intervals to prevent occurrence of failures.

      a. Scheduled maintenance: Preventive maintenance carried out in accordance with a recognized time schedule or established number of units of use.

      b. Condition based maintenance: Preventive maintenance based on performance and/or parameter monitoring. The need for maintenance is predicted with the statistical methods, it is about when, how and why the component has failed.  

Maintenance optimization

             Minimizing total costs or maximizing resources of maintenance is the main goal of maintenance optimization. Maintenance should be carried out so that the tools have high safety and the equipment have a long life and its quality level is preserved. It is difficult to find the relationship between the corrective and preventive maintenance and determine which corrective maintenance cost is associated with a particular amount of preventive maintenance. 

Wind Turbines

             Wind power generation is becoming more prevalent and wind turbines and farms are increasing, so, there must be new ways to minimize downtime and maximize availability and profit. To transform the kinetic power in the wind into electricity wind turbine, machines are used. The main parts are bearings, gearbox, generator, brakes, control system, a part that balances the electricity and rotor and hub, this rotor and hub design can vary, but the most common is the horizontal axis. With two or three blades, the axis of rotation rotates parallel with the ground. Gearbox is to speed up the rotation from a low speed, generator demand constant speed and it can be cooled in two days with air or water. Wind turbine has two brakes; one brakes the rotor and another one for emergency purpose. Control system is to maximize the energy production, nacelle is a part which controls different parts of the wind turbine and a small motor which runs a gearwheel so that the nacelle is always in the direction of the wind.

              Two vibration frequencies in wind turbines are gear frequency and bearing frequency, in typical machineries there are two gear frequencies and there are several axes that are supported by bearings that produce four different frequencies. The technique has become old in wind power and insurance companies complain about large damages and introduce maintenance restrictions. Monitoring is a solution to avoid damages. Wind Farm Monitoring that periodically inspects the whole machinery is the common monitoring procedure for wind power today. Though damages can then be discovered, these inspections do not give answers to when and how the damage has occurred. 

New Monitoring Ways   

            Condition Monitoring is implemented in a widespread manner to avoid preventive replacement. Condition Monitoring is a steady monitoring of each condition. It is the use of advanced technology for the purpose of monitoring a machinery condition and predicting failures in each condition. In practice, there are two ways of condition monitoring, one is continuous monitoring and another one is periodical mechanical diagnosis. In continuous monitoring, suitable sensors are connected with the control system that raises an alarm as soon as data is changed e.g. failures of the gearbox can be found at an early stage by comparing the current frequency of the gearbox with the original frequency. In periodical mechanical diagnosis, inspections of machinery are carried out periodically with measuring tools and compared with the latest measurement.

            The cost of maintenance of a wind turbine is economically high; condition monitoring should be introduced to discover failures early to reduce costs for corrective maintenance and repairs could be better planned and this would give shorter down times. Some of the functions of control systems in wind turbines, come close to the concept of condition monitoring. For large wind turbines the adjustment system is a module of the security system and it is one of the most susceptible systems in a wind turbine. One of the cost effective methods of condition monitoring is visual inspection of machineries, but it is limited to stationary equipments.

Two techniques of condition monitoring explained in this study are,

1. Vibration analysis and

2. Oil analysis.

1. Vibration Analysis

Vibration analysis is the dominant technique used for condition monitoring, especially for rotating equipments, so in wind turbines wheels and bearings in the gearbox, bearings in the generator and main bearing can be monitored adequately. Vibration analysis is not new. In early 80s, the instrumentations and systematic skills required for noise or vibration based condition monitoring were completely developed.

Bearing condition in wind power is indicated by noise; higher rotation per minute (rpm) and higher stress wear out the bearing soon. Few years ago computer based equipment which can measure the sound and evaluate the condition of the bearing was available; it didn't need a person to go out to the wind turbines and climb the tower to listen to the bearings.

There were no products in the market for monitoring of bearings that were suitable for wind turbines in the beginning of 2000. But later nine vibration sensors were assembled to monitor bearings on the main axis, gearbox and generator in the turbine Elida on Risholmen in the archipelago of Gothenburg. This system is called SKF's CMS system WindCon.

This analysis is based on two factors, they are;

• All common failure modes have distinctive vibration frequency components which can be identified and isolated and
• The amplitude of the frequency remains constant, unless there is a change in the operation dynamics of the machine.

  Vibration and noise is normal in machines, so it is therefore important to remember that:

• Every machine will have a normal level of vibration and noise.
• Operating problem is usually the reason for increase in noise and vibration.
• Own unique way of vibration and noise is generated by each operating problem.

The benefits, if problem can be detected and analyzed early, are as follows:

• Convenient time for shutdown of the systems for repairs can be planned.
• Extensive damage is minimized.
• Work schedule, requirements for manpower, tools and replacements parts can be equipped before shutdown.
• Reduce machinery downtime.

Periodic monitoring will not suit high performance machines. For these machines, on-line, continuous, automatic monitoring is required using vibration sensors located at critical points on the machine, when the vibration exceeds pre-set levels, an alarm or the machine would shut down involuntarily.

 Techniques Used in Vibration Analysis for Machine Condition Monitoring

            Following are a few techniques of machine conditioning monitoring,

1. Waveform analysis

2. Indices

3. Synchronous averaging

4. Orbit

5. Statistical analysis

6. Digital fast Fourier analysis.

Waveform analysis records the history of time of the events on a storage oscilloscope or a real time analyzer and in wind power. It is easy to identify the damage occurring in gears and bearings, such as broken teeth and cracks.

Indices used to quantify the time signals with the help of peak level and the RMS (Root Mean Square) level is not reliable in detecting damages in continuously operating systems.

Synchronous averaging is useful in gear-vibration diagnosis where multiple shafts are present; it not only removes the noise, but also periodic events not synchronous with the machine being monitored. Bearing wear, shaft misalignment, shaft difference, lubrication instabilities in hydrodynamic bearings and shaft rub are indicated by Orbit.

Statistical analysis derives curves from machinery vibration signals which can be used in monitoring machine condition. Digital fast Fourier analysis derives the frequency domain signal and the signature spectrum obtained can afford important information with regards to machine condition.

2. Oil Analysis       

Oil failure in wind power is sometimes not detected and it is too late when it is detected. Continuous measurement is needed to the find the defects in the oil. Oil analysis is used to determine the condition and state of the lubricants commonly used in turbines and equipments. The design and operating dynamics of the bearing, lubrication and rotor support structure of machinery is commonly referred as tripology. Lubricating oil analysis is one of the tribology technique used in condition monitoring; it evaluates the condition of lubrication oil in mechanical and electrical equipments. Comparison of the trace metals in oil samples indicates the pattern of oil wetted parts which will provide an indication of the risk of machine failure. Microprocessor based systems are available for lubricating oil analysis.

Oil analysis examines the oil and indicates the elemental chemical composition of the debris present. The major elements present in oil samples are the ingress of extraneous matter such as dust and lubricant with its associate additives.

Atomic absorption, infrared spectro analysis, atomic emissions, coupled plasma, X-ray fluorescence and energy dispersive X-ray analysis are the examples of oil analysis techniques.  Atomic absorption analysis is for wear debris analysis, infrared spectro analysis detects and measures the molecular compounds, and atomic emission is used for debris analysis. Coupled plasma and X-ray fluorescence are used to analyze wear debris and oil additive elements and energy dispersive X-ray analysis is used for the analysis of dry wear debris.

 Gearbox

            Monitoring gearboxes is quite a difficult process as they include many inner components. Wind turbines have three stage gearboxes which have several bearings and gear wheels; it is difficult to find the source of the failure.  Gearbox is monitored with sensors and four or six accelerometers. Since it is working in a changing environment, injury may not be detected until the future. Gearbox wear and failure is due to the load carrying elements such as shafts, gears and bearings. Surface cracks of these parts are due to contact between metals during operation and cyclic loading and this leads ultimately to failure.

            Vibration analysis investigates the gearbox vibration and it is divided into two categories; spectral and feature analysis. Spectral analysis includes Fast Fourier Transform (FFT) which plots amplitude of the vibration signal as a function of frequency.

 Reliability Centered Maintenance (RCM)

             Reliability Centered Maintenance (RCM) is the structured approach to find the balance between corrective maintenance and preventive maintenance and to determine plans focusing on the reliability aspects. It determines the maintenance plan by e.g. prioritizing critical components and through the choice of maintenance tasks and chooses right preventive maintenance activities for the right component at the right time to reach the most cost proficient solution. Civil aircraft industry is the place of origin for reliability centered maintenance (RCM) and first description came in 1978 by Nowlan and introduction in nuclear power came in 1980 and in hydro power in 1990. RCM is characterized by:

1. System function maintenance.

2. Failure mode identification

3. Function prioritizing

4. Choosing efficient maintenance measure

 Descriptions of the process to define a RCM plan are,

1. Seven steps of smith 1993

2. Steps for a first RCM plan - Nowlan

3. Seven question of Moubray.

  Seven steps of Smith     

              Theoretical analysis models and decision tree analysis is the base for smith method. The     seven steps of smith are,

1. Information collection and choice of system

2. System boundary making

3. Description of system and function block diagram

4. Functions of system and losses of function

5. Failure Effect and Function Analysis (FMEA)

6. Decision tree analysis

7. Maintenance measurement estimation.

       Failure Effect and Function Analysis   

                 In reliability analysis, failure effect and function analysis FMEA is used and it can determine the connection between possible failure modes for a construction and the failure effects. Aircraft manufacturer Boeing introduced this method in 1957. To find all the ways in which a product can fail is the purpose of this method. The three questions answered are,

1. What failures/events could appear?

2. What are the effects of the failures/events?

3. What are the causes of the failures/events?

With the answers of these three questions failure frequency is indicated with a number between e.g. 1 and 10. These numbers indicate the seriousness of the consequence and the probability for discovery. These numbers are multiplied into a combined index number, for which a higher value indicates a worse failure and it is called risk priority number. Using the size of the risk priority number, estimation of the seriousness of the failures can be determined and then a measurement can be formulated.

 Steps for a first RCM plan - Nowlan

1. Identify the items that require intensive study by partitioning the equipment into object categories.

2. Identify significant items that have essential safety or economic consequences.

3. Identify hidden functions that require scheduled maintenance.

4. Selecting only the tasks that will satisfy the maintenance requirements by evaluating the maintenance requirement for each significant item and hidden functions in terms of the failure consequences.

5. Identifying items for which no applicable or effective task can be found.

6. For each of the included tasks, select conservative initial intervals and group the tasks in maintenance packages for application

7. To provide the factual information necessary to revise initial decisions establishing an age-exploration program.

 Seven question of Moubray's RCM II

               Identifying the system items that ought to be analyzed is the first step to analyze the maintenance of a system. After this, according to Moubray we should answer seven questions,

1. What are the functions and performances required?

2. In what ways can each function fail?

3. What causes each functional failure?

4. What are the effects of each failure?

5. What are the consequences of each failure?

6. How can each failure be prevented?

7. How does one proceed if no preventive activity is possible?

Functions are what an asset is expected to perform and when the function is specified it is important to indicate a certain level that the unit should meet. There are two functions, primary and secondary; primary function is the main purpose of the asset and secondary function is additional features the asset should meet. The cause of the failure is then described and the consequences of a failure are divided into three categories,

1. Safety- and environmental consequences

2. Operational consequences

3. Non-operational consequences

If the failure cause an environmental law to be broken it is classified as Safety- and environmental consequences. Environmental consequences and operational consequences have an effect on costs regarding production and operation and non operational consequences only gives cost in the form of operations. The decision tree determines which maintenance should be carried out depending on the consequence of the failure.

 Benefits of RCM  

1. The amount of preventive maintenance often can be reduced or replaced with corrective maintenance through the cautious analysis of the failures

2. Decrease in spare parts carriage

3. At an early stage bad construction methods are revealed. 

Reliability centered asset management (RCAM)

   This method is developed from RCM principle attempting to relate preventive maintenance to the total maintenance cost and system reliability. To see the effect on a component level of preventive maintenance on system reliability, with the quantitative methods, is the aim of this technique. In this method, the identified critical components for the system reliability are studied and the relationship between reliability and maintenance has been established by relating the effect of preventive maintenance to the causes of failures for the part being assessed. Main stages of RACM approach are,

1. System reliability analysis

2. Component reliability modeling

3. System reliability and cost/benefit analysis.

 Vattenfall Vattenkraft RCM (VVK RCM) in hydro power

              Vattenfall Vattenkraft developed and implemented a RCM model for its hydro power station; it is similar to Moubray's RCM II in many aspects. This model follows the same steps in Moubray's RCM II but differs in maintenance strategy.

1. Define the functions

2. Determine the function failure

3. Determine the failure modes

4. Risk analysis performance

5. Determine the possible maintenance tasks

6. Analyze and decide on maintenance strategies

Each function in the standard analysis is studied and functional failures are added or removed if needed. Using the standard analysis as the reference failure mode all reasonable causes for each functional failure are listed.

Eight different consequence categories for each failure are,

 1. Cost incurred due to break down

2. Overall costs

3. Environment in which work is taking place

4. Dam flow

5. Damages take place in the environment

6. Personal need

7. Production ceases

8. Efficiency

Risk analysis is performed for every failure mode and how often a failure occurs is being approximately estimated. Using risk matrix, we can get a number between 0 and 5 that describes the risk of the failure, if the number is three or greater than three, preventive maintenance could be used, otherwise corrective maintenance should be used.

The data, like the cost of personnel to perform maintenance, stop time and costs due to loss of production, cost of new parts for damaged equipment and other costs in connection with the maintenance tasks and breakdowns are needed to economically decide the kind of maintenance that has to be used.

The main difference between VVK RCM and RCM II is, in VVK RCM II with the help of a risk and cost analysis the best maintenance strategy is picked but RCM II has a decision diagram where certain maintenance strategies are preferred and if feasible, they are then chosen.

 Conclusion

           Planning maintenance with RCM as a system that collects data is needed for wind turbine industry.RCM combined with CMS helps the wind power industry to monitor the equipment which can't be continuously monitored manually. For any industry, not only the maintenance but reliability is also an important factor. RCM provides better maintenance management system for increasing the reliability of the wind turbines. RCM provides the amount of preventive maintenance that can often be reduced or replaced with corrective maintenance through the cautious analysis of the failures. Co-operation between operators of wind farms, owners and original equipment manufacturers (OEMs) is very important to design the maintenance process based on the maintenance experience in wind power industry.

 

            

 

 

 

 

 

Combination Probability Question?

Having a huge amount of trouble working this stuff out!?!

In a plant there are 23 workers, 15 are operators, 8 are maintenance.

How many ways can you select 6 workers in which:

i) none are maintenance
ii) one is maintenance
ii) at least one is maintenance

Also, find the probability that there will be at least one maintenance person in the selection.

Cheers!

Formula you need to use to answer these questions is nCr = n!/(r!*(n-r)!) n is the number of options, and r is the number you are choosing. So...

i. If none are maintenance, you are just choosing the six out of 15 operators. n=15, and 6=r. 15C6 = 15!/(6!*9!) = 5005
ii. If one out of 6 must be maintenance, you pick 5 out of operator group (15C5), and one out of maintenance group (8C1). (15C5)*(8C1) = (15!/(5!*10!))*(8!/7!)=3003*8 = 24024
iii. Having at least one maintenance is the opposite of (i), which was having zero maintenance. So we take all possible combinations, which is picking 6 workers out of 23 workers. This is 23C6 = 100947. We take out the optionof having no maintenance, which is 5005. So 100947-5005=95942.
iv. Possibility of having at least one maintenance is (iii) divided by all possibilities. So, 95942/100947 = 95.042%