Bachmann Monitoring GmbH

Bachmann Monitoring GmbH

Conditioning Monitoring System

(CMS)

Dr. Steffen BiehlM.Sc. Peter Collmann 09.08.2018

Who is AGF Energia & Neo Wind

Since 2016 Neo Wind and AGF Energia have worked together to offer to the market the best solutions.

Neo Wind's technical expertise in renewable projects allied with AGF Energia's multidisciplinary project implementation and O&M experience provides the best solutions and innovations for its customers, always seeking quality, efficiency and cost reduction in its services.

RJPR

CE

Headquarter: Rudolstadt (Federal State of Thuringia)

Established: July 1998 (as µ-Sen GmbH) 20 years of experience in wind energy

Single source: CMS Hardware & Software

Remote Monitoring ServicesTraining

Sensor technologies

Staff: 62

Owner: 100% Bachmann electronic GmbH Feldkirch since 01.01.2011

Who is Bachmann Monitoring GmbH (BAM)

Knowledge Based Maintenance OptimisationBachmann Monitoring GmbH

Maintenance Optimisation

Run to Failure

Maintenance Optimisation

Breakdown Maintenance

Preventive Maintenance

Reactive Maintenance

Predictive Maintenance

Time-based Maintenance

Total Productive Maintenance

Scheduled Maintenance

Firefighting

Reliability Centred Maintenance

On Condition Maintenance

…

Maintenance Strategies

Only Three Strategies:

Reactive Maintenance (Run to Failure)

Run the machine until it fails – high repair costs, poor availability

Predictive Maintenance (Condition Based)

Monitoring of selected parameters to assess the condition of the machine so that maintenance can be planned.

Preventive Maintenance (Time-based)

Run the plant for a pre-determined period then overhaul


Knowledge based maintenanceChoose your Strategy


Choose repair time based on interval

Good partsreplaced

Maintenance induced problems

Some parts will failearly – CM neededfor critical items

Can be augmentedwith equivalentoperating hours

Reactive Maintenance Predictive maintenance Optimised Maintenance

Repair item on failure

Unplanneddowntime

Consequentialdamage

Fine forconsumables in non critical plant

Requires ConditionIndication

Failure indicatormust give sufficientlead time

Detectionprobability must behigh

Plant must haveopportunities formaintenance

Strategy based on plant item

Reducesunnecessary work

Reduced unplannedunavailability

Supports planningfor major items

Inputs from Big Data methods toimproveprognostics




Operating Informationnecessary

Knowledge ofreliabilty generated

Fewer Actions, but some unnecessaryand most too early

Reactive Maintenance Predictive maintenance Optimised Maintenance

Little Informationneeded

Little Knowledgegenerated

Lots of Actionsrequired

Large amount ofInformationgenerated

Expert analysis toconvert toKnowledge

Actions at optimumtime, but not applicable to all plant items

Collects and Collates multiple Information sources

Knowledge usedand updatedcontinuously

Minimum amountof Action tomaintain plant health

Cost Risk



Preventive MaintenanceReactive Maintenance Predictive maintenance Optimised Maintenance

Data

Information

Knowledge

Action

Data

Information

Knowledge

Action

Data

Information

Knowledge

Action

Data

Information

Knowledge

Action


CMS – On Line Vibration Monitoring

Vibration of turbine bearings; speed marker; process parameters

Converted into signals by sensors and fed to CMS hardware

Order related data; CVs trends; history; machine build

Analysed daily for long term trends

Current state of machine; cause of any anomalies; recommendations

Transmitted via weblog e-mails; stored in logs; work orders; reports

Data Information Knowledge

Knowledge based maintenance


Diagnosis

Process of converting the information to knowledge

Commonly via a Fault/Symptom approach

Pattern of information associated with a particular fault – symptoms

Individual symptoms can indicate multiple faults

Identify patterns of multiple symptoms and corroborating information

Draws on previous experience

A good diagnosis will include

All fault modes considered possible and why

Which faults are considered most likely, and relative confidence

Reference to previous similar faults

A recommendation as to remedial action

Diagnosis turns information into knowledge

Idealised flow of Data, Information and Knowledge

Knowledge based maintenance


Optimum Position of Managed Risk

Condition Monitoring

Risk Averse


Condition Monitoring Monitoring of selected parameters to assess

the condition of the machine.

To be of use it must: Produce a measurable, non catastrophic

effect

Give sufficient warning time before failure

Be reproducible

Give a signature from which a diagnosis can be made

The ultimate aim: Provide a maintenance solution for the

plant which gives minimum cost and managed risk

Condition MonitoringWhy do we do it?


Early detection of damage indications

More time to plan

Targets necessary maintenance

Avoids unnecessary maintenance

Informs customers about their machines

Allows OEM to be challenged

Enables operation under fault conditions

Reduces risk

SAVES MONEY

Condition MonitoringWhy we do it


Defines the time between first possible detection (P) and functional failure (F)

Different for each machine and for each failure Mode

Essential part of the process definition – if PF is too short there is no point monitoring

Remember, P is where an incipient fault is detectable

Some conditions can be reset by maintenance, but most are the first sign of an impending failure

PF CurveDo we get enough warning?


Temperature – either process parameters or

thermographic images

Pressures – either static or dynamic

Efficiency

Process Information

Vibration

Oil Analysis

Electrical Testing

Acoustic Emission

NDT

Not forgetting routine maintenance

inspections or whatever else is appropriate…

Suitable Parameters Include:

Condition Monitoring


Development of a fault to failure

Trend Diagnosis

• High frequency energy only

• Limited riskStage 1

• Frequency spreads

• Bearing swap requiredStage 2

• Loss of shaft control

• Damage to journal/housing likelyStage 3

• No bearing related vibration energyNormal


Trend DiagnosisIncreasing trend of an inner ring roll over frequency intermediate shaft bearing

April 20140,045g

March 20130,010g

June 20120,007g


Cost Benefit Analysis

Electrical components may show a higher

failure rate but cause usually little

downtimes only

Failures on the mechanical components of

the drive train usually cause higher

downtimes per failure

Forwarning provides more planning time and allows maintenance to be optimised

Statistical failure rate vs. downtime rate (IWES)


Versus Preventive Maintenance

Uses full life of parts – no replacement ofhealthy components

Reduced maintenance requirements

Reduced risk of maintenance inducedfailure

Fewer replacement parts

Versus Reactive Maintenance

Reduced downtime through planning

Reduced lost time due to weather

Parts ordered at right time, not kept on stock

Repair cheaper than replacement

Refurbishment of large items usuallypossible

No consequential damage

Cost Benefit AnalysisCondition Based Maintenance

Fraunhofer IWES, „Condition Monitoring of Wind Turbines“ 2015


Analyse all possible failure modes

In each case calculate cost of:

undetected failure

routine replacement

planned replacement due to CM

Considering

Probability of failure

Probability of detection

Probability of maintenance induced failure

Aggregate across farm

Subtract monitoring costs (opex + capex)

Savings depend on assumptions

Most companies reluctant to share values

Academic studies suggest payback in between 1 and 3 years

Cost Benefit Analysis Theory

May, McMillan and Thöns, „Economic Analysis of CMS for offshore wind“ DTU at AWEA 2014


Individual savings can be spectacular, but don‘t just take our word for it.

Sometimes system manufacturersoverstate the benefits

Operators and manufacturers don‘t like totalk about the actual savings, due tocommercial considerations (cost ofdowntime and cost of spares can becalculated)

Overall figures are hard to come by, but Almost all mayor WTG manufacturers have

adopted CMS as standard

Several operators now insist on CMS beingfitted.

Do you think it is worth it?

Cost Benefit AnalysisPractice

RES Website Case Study

Uniper Presentation at VGB Wind turbine Conference 2016

Bachmann CMS – An overall concept

Certified remote monitoring service in Europe , North America and Asia

Interface and installation through AGF Energia & Neo

Wind

Highest quality and customized CMS hardware based on Bachmann-own

production and development

Cabinet dimensions: 380x380x210mm

Direct mounting or installation via fixed bases ormagnets

Product line: Stand-Alone CMS

Mounting in control cabinet, controller-independent

Low expenses for hardware and low installation effort

Product line: Top-box CMS

BA

M1

00

• Sensitivity100 mV/g

• Output IEPE-compatible

• Measuring range0,5 Hz – 14 kHz

• recommendedfor monitoringfast rotatingcomponents

BA

M5

00

• Sensitivity500 mV/g


• Measuring range0,2 Hz – 14 kHz

• recommendedfor slow rotatingcomponents

µ-b

rid

ge

• Sensitivity0,7 V/N


• Measuring range0,05 - 1.000 Hz

• recommendedfor very slowrotatingcomponents

Product line: Sensors

Robust design

18 channels of ICP enabled vibration channels

1 speed signal input

3 Analogue inputs for process parameters

Browser-based configuration

Remote access

Set of sensors and cables available as an accessory pack

Based on same hardware as installed systems

Product line: Portable CMS

Best quality and availability by 100% test coverage and 48 h Run-In

Robustness even under extreme environmental conditions by using „ColdClimate“- modules

Open programmable standards IEC 61131-3, C/C++, Java, HTML5, Matlab/Simulink®

Long term availability by compatibility

Partnership relied on service and local support

Reduced Time-to-Market by system solutions

Designed for 20+ years life time

Committed to „Quality first“

Quality first

Bachmann CMS – Production

https://www.youtube.com/watch?v=VxW8uEVYI7s

All Installation Instructions

Step by Step Process

Documented using built in camera

Certain fields mandatory

Automatically generatesEnd of Installation Protocol and e-mails it

Installation Tools

Several Forms Input & Choice Boxes Picture taken by Camera

Signature Save Mandatory EntriesInstant Data Transfer

Smartphone Application

HD-StartLauncher.exe

Browser Based

Universally available

Access to tickets

Access to currentstatus

Access to simple Graphical Overview

Inclusive withMonitoring Service

Weblog

Software Capabilities

Client Based

Licensed Install

Access to databaseServer

Designed to Support Monitoring Process

Advanced Analysis Tools

ComprehensiveConfiguration Toolset

Weblog Expert


Highly Configurable:

Module

Channel

Acquisition

Classification

FFT Settings

Time Series

Filters

Thresholds

Logging

Weblog Expert - Configuration



Blade rotor unbalance

Mass unbalance occurs if centre of mass and centre of rotation are not in the same place, eg due to a different Centre of mass between blades

Aerodynamic unbalance occurs if blades are not performing equally, eg due to pitch errors, or damage to the surface

Why is it important?

Blade mass centres cannot be perfectly matched, there will be a residual unbalance which needs correcting

Generally no balance quality is specified, or maybe G16 if any

In practice much better balance can be achieved, but is it?

Unbalance causes 1/ rev vibration which affects the entire structure ofthe turbine

Potential to shorten life of structural components and main bearings

Plug In to Bachmann CMS

Blade Unbalance Calculator




Blade rotor unbalance calculation on operating systems

No teams needed on the wind turbine

No need to interrupt productivity

Evaluates mass and aerodynamic unbalance

What we need?

Tower geometry and mass distribution

Tight rotor speed ranges for optimal results

10 to 20min measure time

RPM- and acceleration sensor (2D MEMS)

Optional: Angle of rotor axis for unbalance phase


Blade rotor unbalance calculation on operating systems

Results generated on line during periods of stable operation at a specified speed range

Distinguishes between aerodynamic and mass unbalance

Calculates the result in kgm (distance from the centre is also important)

Calculates the phase if additional marker fitted

Benefits?

Assessment of residual unbalance on a regular basis

Early detection of out of tolerance machines before extreme values occur

On site focus can be on machines with problems

Answer for weight fitment requires no further trial run (with option)

Also detects pitch system errors




27 Datasets27.07.17-6.11.17

Unbalance Phase

Estimated by weight addition (3rd party)

376.1kgm 276°

Blade Unbalance CalculatorMean 377.6kgm 266°

First Results


Polar histogram of phase

Also successfully detected aerodynamic unbalance on a test turbine

Case Studies Bachmann Monitoring GmbH

Case Study: Main Bearing Damage

Damage initiation seen at start of winter period

Damage progressed through winter

Under close monitoring

Repair in spring

Customer acted on knowledge

Advance warning provided

Ran on under close monitoring

Repair planned well in advance

Findings via µ-bridge sensor

First Warning

Repair

Progression


Case Study: Planetary Stage Tooth Damage

Early Detection of planetary stage damage

Trend in CV relating to TM frequency.

FFT indicated damage on the sunwheel

Levels and trend suggest progession likely

Customer informed of finding

Customer inspected gearbox

The sun gear was found to be „broken“

Gearbox exchange planned

Two weeks Advance Warning

Planetary Stage Catastrophic Failure avoided

Refurbishment of replacement of gearbox

Saving of €50k

Alert Raised 13.11.


Case Study: Helical stage gearbox

Sudden change in tooth mesh frequency

Customer Inspected gears

Early warning of tooth damage

Gear could be replaced

No consequential damage


Case Study: HSS Inner Ring Defect

• A HSS defect was noted

• Damage initiation possibly present when monitoring began

• First indication after the warning level was exceeded in February 2018

• Allowed the customer to plan a repair

• Intervention at ideal time

• Damage beginning to trend, but at this point minimal consequential damage

• Advance warning meant that the teams were ready to make the exchange with minimal loss of production when the damage began to increase

First Warning

Reminder

Repair


Case Study: Generator bearing

Bachman identified ball spin frequency in signal

Diagnosed roller damage

Progression relatively slow

Repair made when progression accelerated

Customer scheduled exchange when convenient

Spall from single ball

Can result from water in grease

Original material defect also possible

First Warning

Reminder

Repair


Case Study: Rotor Unbalance (Aerodynamic)

All aerodynamic unbalances corrected

Ongoing trial of mass unbalance system

Aerodynamic unbalance re-appeared after a few months

Customer requested more exact time for change

We identified a 6 hour window

Technicians on site applying loctite to pitch system

Rapid warning provided

Reduced loss of earnings

Customer: “That system rocks the party!“

Highest quality and customized hardware based on Bachmann-own production and development

Stand-alone solutions for any turbine make & model

Cost efficient integrated CMS for WTG with Bachmann controls

Portable CMS for offline diagnosis

PROVEN EXCELLENT QUALITY.

20 years of experience, more than 25 make and 80 models of wind turbines

Remote monitoring locations in China, Europe and USA

Cutting edge diagnostics from certified vibration analysts

Sophisticated software for efficient monitoring

Integration of data into SCADA

EXPERIENCE AND EFFICENCY.

Unique Quality, Experience, Efficiency out of one hand

Service by Bachmann CMS in WTG world wide distributeddata acquisition, analysing, threshold monitoring,automated data delivery to remote monitoring centre

customer, operator, maintenance team

CM REMOTE MONITORING CENTRE

evaluation, analysing, interpretation

reports with information for maintenance & logistics

data archiving, building a object database for monitored WTG for use of

LCC – optimized maintenance

support,

maintenance

and logistic

related

information

for use of

condition

based

maintenance

qualified support of CMS (HW, SW)

secured communication to world wide distributed CMS

Bro

wser

Ala

rm

DNV-GL-certified monitoring centre (20+ staff)

ISO 18436-2 certified vibration analysts

(CAT II, CAT III, CAT IV)

References Wind

600 kW – 8 MW (On- and Offshore)

Large and small O&M companies

Large and small utilities

Large and small OEM’s

Wind projects all over the world(from 2 to >1000 WTG)

Marine applications

Effektivwert

20 years experience

on WTG

More than

8 GW in CMS-

Monitoring

Verified on > 9000

WTG

Experience on 85

different gearbox

types

WTG

600kW to 8MW

Portfolio covers

more than 80 different WTG types

Onshore and

Offshore WTG

WTG portfolio covers 25

OEM

Experience on 10 drive train setups

>500 wind farms (from

2 WTG to 200 WTG)

World Nr. 1 CMS Solution Provider

Bachmann CMS – Experience counts

Logo Bachmann

Endereço Rio de Janeiro - RJRua Senador Dantas, 117, Salas 1742

Rio de Janeiro - RJ – BrasilTelefone: +55 21 3549 – 1669

[email protected]

Endereço Fortaleza - CERua Pinto Madeira 1500 - Sala 02

Fortaleza - CE / BrasilTelefone: +55 85 3033-0072

[email protected]

Bachmann Monitoring GmbH

Documents

Transcript of Bachmann Monitoring GmbH