ES2401880A1 - Procedure and system to optimize the maintenance operations in wind turbines and wind farms. (Machine-translation by Google Translate, not legally binding) - Google Patents

Abstract

System and maintenance procedure for wind turbines or wind farms that includes: an input connection to receive field parameters measured from a wind turbine or a wind farm system, a fault detection module that emits a fault detection signal, a module of troubleshooting which has storage means with hierarchical nodal structure data, a fault estimation or forecasting module having storage means with predetermined incipient failure parameters, and a maintenance programmer module for outputting a maintenance output signal said signal being based on the combination of said fault detection signal, said trouble finding signal and said estimated or predicted fault signal. (Machine-translation by Google Translate, not legally binding)

Description

PROCEDURE AND SYSTEM TO OPTIMIZE OPERATIONS

MAINTENANCE IN AEROGENERATORS AND WIND FARMS.

Object of the invention.

5

The present invention relates to a method and system for

optimize maintenance operations in wind turbines and parks

wind power

1st

Background of the invention.

A myriad of parameters must be taken into account when they are performed

the maintenance of complex wind turbine systems and

wind farms There can be an excess of confusing information and

fifteen

contradictory

Known repair systems are managed through the

human management with the help of spreadsheets, alarms or generation of

maintenance tasks based mainly on human intervention.

In such systems it happens that weather conditions

2nd

Unexpected may postpone or cancel maintenance operations.

It also happens that new components sometimes show a

poor performance or are damaged prematurely. Other events

they can be a sudden drop in production caused by a series of

cumulative breakdowns

2 5

It is the object of the present invention to present a system and

procedure that can accept a very large number of components and

parameters, as well as many possible locations and data types

different.

It is another object of the present invention to provide a method

3rd

simple and reliable to optimize or reschedule maintenance tasks

dynamically, considering that events may vary

quickly.

It is another object of the present invention to have a system that

can plan maintenance systems well in advance of

3 5

reliable way.

Description of the invention

The present invention relates to a method and system of

maintenance for wind turbines or wind farms. Said system has

5

Means of calculation comprising:

an input connection to receive measured field parameters

from a wind turbine or wind farm system, and

a fault detection module that has means of

storage with default parameters corresponding to

1st

default fault modes that emit a signal to detect

breakdowns, and

a troubleshooting module that has means of

storage with data in hierarchical nodal structure, in which said

Nodes correspond to wind turbine components and subsystems (WTG)

fifteen

or wind farm, said troubleshooting module emitting a signal

troubleshooting, and

a fault estimation or forecast module that has means of

storage with default incipient fault parameters,

issuing said failure forecast module at least one signal of

2nd

estimated or predicted breakdown for a component or subsystem, and

a maintenance programmer module to issue at least one

maintenance output signal, said signal based on the combination

of said fault detection signal, said location signal of

breakdowns and said estimated or predicted breakdown signal.

2 5

Preferably, the maintenance output signal is a signal of

maintenance time corresponding to maintenance operations

recommended and a repair date and time for a computer

maintenance. In addition, said maintenance output signal identifies

one or more components to replace at a specific location of the node

3rd

instance of the wind farm or the wind turbine.

Advantageously, the constraint parameters of the

maintenance. Such restriction parameters may be parameters of

weather conditions, availability parameters of parts of

spare or tools.

35

   In addition, said maintenance output signal is enabled,

disabled or enabled with a warning for a human operator,

selectively, based on values of said restriction parameters of the

maintenance. Said restriction parameters are preferably based

in meteorological parameters in the field near a wind turbine or a

5

wind farm and can be wind speed, field temperature,

the height of the waves, the thickness of the fog or lightning.

Said maintenance time signal can also be based on

a loss profile, in which said loss profile is defined by the

loss of output power of the wind turbine in standard operation

1st

or at maximum power generation.

Additionally, the maintenance system may include a matrix

of multi-dimensional applicability that contains true or false values

to correlate a given fault at a given location and said signal of

Maintenance output

fifteen

The system may also have data filtration means for

issue a selected number of parameters for a human operator, and

editing means that allow the human operator to edit the parameters

of failure, and an interface of the human being to admit or reject an event

of failure.

2nd

The maintenance programmer module can also comprise

a module for time triggered synchronization of said signal

maintenance output that has an update time

default, especially a daily update time, and a

module for synchronization triggered by an event of said

2 5

maintenance output signal.

A maintenance procedure for wind turbines or parks

Wind power is also part of the present invention. This one includes the

stages of:

receive measured field parameters from a wind turbine system or

3rd

a wind farm, and

detect fault parameters and emit at least one signal from

fault detection, and

Troubleshoot according to data in hierarchical nodal structure, in which

these nodes correspond to subsystems and components of the

3 5

   wind turbine or wind farm, issuing said location module of

faults a fault location signal, and

estimate or forecast incipient fault parameters by issuing

minus an estimated or predicted breakdown date signal for a

component or subsystem, and

5

emit at least one maintenance output signal, based

said signal in the combination of said fault detection signal, said

fault location signal and said estimated fault date signal.

Preferably the maintenance output signal is a signal of

maintenance time corresponding to maintenance operations

1st

recommended and a predicted date and time for a team of

maintenance.

In addition, said maintenance time signal is also based on

maintenance restriction parameters, said parameters being

restriction, weather conditions parameters,

fifteen

availability of spare parts or tools and personnel.

The meteorological restriction parameters in the field are

those near the wind turbine or wind farm and are preferably

wind speed, field temperature, wave height,

Thickness of fog or lightning.

2nd

Finally, the present invention can be implemented

preferably through a computer readable system program

computer.

Brief description of the drawings.

25

Figure 1 shows a modular representation of an embodiment

preferred according to the present invention.

Figure 2 shows an embodiment with a breakdown structure

hierarchical nodes that represent several subsystems.

3rd

Figure 3 shows a flow chart that refers to the detection

and the automatic forecast of fault parameters.

Figure 4 shows a flow chart that refers to a function

Fault recognition by a human operator.

Figure 5 shows a flow chart referring to the

3 5

   generation of maintenance tasks.

Figure 6 shows a Gantt chart with a program of

maintenance.

Figure 7 shows a comparative graph of the performance of a

wind turbine if a maintenance program is applied.

5

Figure 8 shows a flow chart with an embodiment for

Schedule maintenance

Figure 9 shows a flow chart with an embodiment for the

Maintenance execution

Figure 1O shows a flow chart with a nodal structure

1st

hierarchical

Figure 11 shows a flow chart of faults and maintenance.

Figure 12 shows embodiments with triggered synchronization

for an event and for time.

Figure 13 shows an embodiment of the system architecture

fifteen

according to the present invention.

Description of a preferred embodiment.

Figure 1 shows an embodiment with an integrated environment and

2nd

digitized to keep track of faults and to program parameters of

maintenance according to the present invention.

The maintenance system has an input connection (1) for

receive the measured parameters received from the systems in the field. He

system also has a fault detection module (2) that has

2 5

storage media with default parameters

corresponding to default fault modes and emitting a signal

fault detection, and a troubleshooting module that has

storage media with data in hierarchical nodal structure. Sayings

nodes (9, 1 O) correspond to wind turbine components and subsystems

3rd

(WTG) or wind farm (WF). The maintenance module also has a

estimation or forecast module (3) that has means of

storage with default incipient fault parameters and a

maintenance programmer module (6) to emit at least one signal from

maintenance output (7). That signal being based on the combination of

35

   said fault detection signal, said fault location signal

(5) and said estimated or predicted fault signal (4).

In one of the preferred embodiments, the output signal of

maintenance (7) is a maintenance time signal (8)

corresponding to recommended maintenance operations and a

5

Repair date for a maintenance team.

Figure 2 shows a hierarchical tree of nodes {WTG, 9, 1O) for

Identify the subsystem or component in which a fault originates. For

identify the set or part in which a fault occurred, must be defined

the broken down structure of wind turbines or wind farms. The

1st

broken down structure is defined as a tree that captures relationships

father-son of the parties. The elements of this broken down structure are

denominate as the nodes (WTG, 9, 10), and represent parts or assemblies

of the turbine (WTG). In the present invention, it is sufficient to represent the

parts that are relevant for maintenance purposes, specifically the

fifteen

wind turbines themselves and hierarchical turbine groups: parks

wind farms, wind farm groups maintained by the same regional office

and others.

To simplify the entry and generation of reports, each node is

assigned to a template. Templates are system configurations,

2nd

called type tags, which can be used to filter and group nodes

which refer to various operating states. Therefore the templates

they constitute trees that represent the natural father-son relationships between

system levels and default configuration subsystem.

The system also has fault storage media that

2 5

They include a library of defined faults. These means of

fault storage, also called the `` failure master '', is a

library of failures defined. The faults are characterized by their names

and preferably with additional parameters that are used by a

maintenance programmer

3rd

Some of these additional parameters or information may be,

preferably, the type of breakdown, which could be a real breakdown event,

a predicted or anticipated breakdown, a maintenance request or

periodic update.

Part of said additional parameters may preferably be

3 5

   a loss profile parameter, which is the property of the fault that

determines if the fault creates a loss in any operational condition or

simply at maximum when maximum power is generated.

Part of said additional parameters may preferably be

a parameter of production loss. Said loss parameter of

5

production is preferably a property of the fault that determines the

loss of production due to failure in percentage. They differ two

loss profiles: according to the first profile, a given percentage of

production in all operating conditions; according to the second profile, the

maximum output power is reduced by a given percentage, due to breakdown

1st

that does not occur at a lower performance. 1 00% means detention

complete.

In addition, part of said additional parameters may be,

preferably, an opportunity window parameter. Said parameter

refers to the time interval that provides a clue to the programmer

fifteen

how accurately can you schedule maintenance regarding

at the nominal date on which the fault occurred.

The system may also comprise storage media of

maintenance tasks, the so-called task master data

maintenance, which specify all the parameters and requirements of a

2nd

maintenance task This information is used by a programmer of

maintenance.

Such maintenance task parameters include,

preferably, weather conditions parameters that define

restrictions on measurable weather data, typically speed

2 5

of wind, temperature, waves, fog, snow, lightning, etc. Sayings

Weather parameters can enable or disable the execution of

a maintenance task A weather condition can be a

hard restriction, the automatic programmer will not place the task on that

interval, or a warning, where the task can be placed there, but a

3rd

Warning is sent to the human planner.

Such maintenance task parameters include

preferably service requirement parameters, typically a list

of resources such as a crane or a truck, which must be present in the

turbine during maintenance task execution.

35

   Such maintenance task parameters include,

preferably, service spare part requirements, which define

the requirements of spare parts of the maintenance operation.

Only the requirements whose availability, being present at the level of

stock or time of arrival, may impose a restriction on the date of

5

task execution.

The breakdowns and instance nodes of the wind turbine (WTG) and the

maintenance tasks are linked together by a

called applicability matrix. This matrix can be represented as a

three-dimensional matrix that contains TRUE or FALSE values. A

1st

TRUE value means that the maintenance task given can

be applied to repair the given fault of the wind turbine (WTG). They can

apply other means to correlate nodes and maintenance tasks.

Fault detection and forecasting are part of the system of

generation of maintenance tasks. Instances of failures include

fifteen

detected or actual failures and expected failures, as well as requests for

maintenance for maintenance or periodic updating. Each type of

fault can be entered into the system manually via the interface

of the system user. This can be achieved automatically by

a call, the so-called software agent, which includes calculation means

2nd

to simulate fault conditions. Figure 3 shows a flow chart

with the following stages:

select a wind turbine instance node (WTG) (turbine,

component, or installation of the park);

select a fault mode (only nodes are allowed

25

associated to the node); the system creates a fault instance by copying the

default fault parameters of the fault master record.

As can be seen in Figure 3, the fault parameters (loss

caused, date of the predicted fault, etc.) can be edited. Default,

entered faults receive a registered status that does not imply

3rd

some automated actions in the system. However, this may

Update on the fault input screen.

The system and procedure preferably include the

Fault recognition by an expert or human operator. The system

advantageously provides support to human experts to understand the

3 5

   nature of the failures presenting measurements in the user interface.

This preferably occurs in a SCADA environment. To avoid overloading

to the user with data, the system advantageously filters in search of

relevant, important and correct measurements. The system can present

the corresponding SCADA measurements for each fault instance of

5

the following ways:

a list of fault instances is displayed, filtered for a

registered status The list contains basic data on the faults (mode of

fault, wind turbines (WTG) affected, date, etc.); Y

a group recognition, in which the human expert selects

1st

the fault instances to be recognized or the status of all faults

selected changes to quot; admittedquot ;; Y

admissions of individual failures, in which the human expert

Check and potentially edit the fault parameters detailed in the

fault details screen or, if you think the fault is valid, support

fifteen

the fault, otherwise rejects it.

Figure 4 shows a maintenance task generator

Automated In this flowchart, they are recommended and generated

maintenance tasks for admitted failures. The user can cancel the

system suggestion. In addition, multiple tasks can be assigned to one

2nd

breakdown (although by default, the system only recommends one), and a

task can repair multiple faults, in addition the system recognizes if a

task can cover multiple faults). When it is automatically suggested

a maintenance task, the system considers all tasks

Candidates assigned to the given fault and the instance node of the

25

wind turbine (WTG) in the applicability matrix. Between homework

Candidates, select a suggested task based on cost analysis,

considering the expected period of inactivity, availability of requirements

of spare parts and services. The system lists the faults for which

A maintenance task must be generated. These are the failures admitted

3rd

You don't have an open task assigned. To get a suggestion

Automatic maintenance tasks, the user selects the fault for

the one that wants to generate maintenance tasks, the generator engine of

maintenance creates a suggested task assignment, and the suggestion is

shows in the fault list. To assign maintenance tasks for

35

   manually, or cancel the system suggestion, the user selects a

breakdown, and goes to the details of the breakdown, where an instance of

fault admitted.

Apart from showing the tasks currently assigned to the fault, this

page provides multiple ways to select a different task,

5

specifically:

generate a new task instance based on the array of

Applicability for the fault instance and WTG given, and

generate a new task instance, selecting and applying a

arbitrary task template (filtering possibilities are provided);

1st

assigning an existing task instance, which has been generated for a

different fault of the same turbine. Homework instances just

generated are added to the database.

As can be seen in Figure 5 and Figure 6, the tasks of

generated maintenance are subsequently assigned to equipment

fifteen

Maintenance and are scheduled. The programming takes into account the

equipment capacity restrictions, availability of requirements for

spare parts and service and weather forecast, among others.

Similarly to the generation of maintenance tasks, the system

Automatically generates and optimizes a program. This program can be

2nd

canceled, however, by human experts.

The programming stage entry is the task list of

open maintenance, that is, the tasks that have been generated, but still

They have not been executed (completely). The output is a program of

maintenance, which consists of an allocation of equipment and a

2 5

Timing (8) of tasks. The program will be shown in a diagram of

Gantt as depicted in Figure 6, and the expected performance of the

turbine over time, in a different diagram.

The Gantt chart is represented in Figure 6. In this program,

tasks are divided into task segments, that is fractions that are

3rd

They run without interruption. A short task, such as one that has

a maximum of 2 hours, the value can be edited in the user interface,

will correspond to a single segment, while long tasks (above

2 hours, for example, 3 work days for a generator change)

They will divide into multiple segments. The programmer adopts a low level of

35

   conservatism, specifically following the previous version of

program: the tasks whose execution has started will not move in the

time will not be reallocated to a different team; the tasks that have been

accepted by a team will be assigned to the same team.

Figure 7 shows a temporal diagram of the energy produced by

5

A wind turbine Its maximum production will not be reached if the system has

incipient breakdowns or loss failures occur. Therefore, when

schedule repair actions, sometimes it is necessary to consider the

Loss of performance.

As can be seen in Figure 8, program in this embodiment

1st

It comprises the following stages:

show the set of open maintenance tasks. When you open one

details view of individual tasks, the user can edit their

parameters, be these duration, requirements, etc. When the user clicks

in the save button, the user verifies the edited program, and displays it;

fifteen

Y

generate warning messages for any violation of

restriction (unavailability of spare parts or requirements, conditions

inadequate weather, etc.). The user can decide whether to solve these

problems or save the program as presented. After saving

2nd

the program, the edited task will arrive will go to manual mode, which means

that manual changes will be retained even in the following stages

of automatic programming.

In addition to the previous representation, technicians can also

access the maintenance program in PDA, where the program of the

2 5

Specific equipment will be presented in a list format.

The maintenance execution is shown in figure 9. The equipment

Maintenance can access your daily schedule, and description

detailed maintenance tasks to execute. The details page of

the task also provides links to documents related to the

3rd

given maintenance task, for example, maintenance manuals. He

maintenance team provides feedback in three ways

different. The team selects some of its task segments

assigned, and accepts or rejects them. The team selects the segment of

task that begins to execute, and changes its status to quot; execution startedquot ;. He

3 5

   team selects the task segment that you completed, and provides

Feedback as follows:

set the status of the task segment to 'closed';

update the status of the task (closed / failed, could not execute the

unsuccessful task 1 - task executed, but the fault persists 1 to rehearse) and

5

possibly the remaining duration;

update the fault status accordingly (closed)

rehearse / admitted-means that the fault continues to persist), and

Provides feedback on the actual use of parts.

The maintenance execution information screen will be

1st

Available in both PC and PDA formats, where technicians

You can use this last format to enter quasi-online information

About the execution.

The system has additional data aggregation capabilities,

preferably implemented in SCADA. SCADA collects large

fifteen

amounts of data, for example in an average of 1 O minutes, data

operating wind turbines (WTG). This data can be accessed

in the application through different diagrams and reports. Further,

The application also analyzes operational, availability and data

Long-term production through various reports. The requirements of

2nd

efficiency and storage make it necessary to store data long

term aggregate. The system performs daily data aggregation, and

compute the following operational data for each day. One realization

Preferred for each wind turbine (WTG) includes:

technical time available (minutes, sum of the following times:

2 5

Progressive coding, Preparation, Execution);

technical time not available (minutes, sum of the following

times: Pause, Stop, Emergency);

available operating time (minutes, sum of the following times:

Preparation, Execution);

30

operating time not available (minutes, sum of the following

Times: Pause, Progressive Coding, Stop, Emergency, Manual),

total energy produced,

total expected energy (estimate based on wind speed and

production function).

3 5

   In addition, the system computes and stores alarm data for each

day and each wind turbine (WTG). Among other data, the

following data: alarm code, number of alarms, total duration

of the alarm.

To eliminate any distortion caused by repetitive appearance

5

of alarms, or false alarms due to operating conditions

special, for example stop or start up of the wind turbine

(WTG), alarm filtering will be applied.

Preferably based on SCADA measurements and data from

breakdown and maintenance, the system provides various types of

1st

reports. The following reports can be advantageously implemented:

availability analysis: the report provides information on the

technical and operational availability

turbine availability (tabular format);

reliability analysis: the report addresses the reliability of WTG

fifteen

individual or WF, as well as the effect of the properties of

WFJWTG / WTG components (for example, WTG product number,

component supplier, customer, or geographic location) in indicators

reliability (tabular format);

Ease of maintenance analysis: the report contains a

2nd

statistical evaluation of the maintenance facility of WTG, WF,

maintenance related to specific failure modes, or tasks of

defined maintenance The reports are based on historical data on

maintenance tasks executed (tabular format);

alarm analysis: this report aims to analyze the frequencies of

2 5

appearance and characteristics of alarms (tabular format);

Energy production analysis: the report presents statistics

on the energy produced by WTG / WF given during a period of

given time (tabular format);

maintenance cost analysis: this report analyzes the structure

3rd

of maintenance cost based on maintenance tasks

completed, including direct and indirect costs (also loss of

production, in tabular format);

graphical display of operational data and alarms: the signals of

SCADA and alarms can be displayed and analyzed graphically. Be

35

   They will implement two types of configurable diagrams: signals and alarms at

over time; and two SCADA signals against each other.

Reports can be parameterized in various ways. They can

present with different aspects of parameters (for example, period of

time, breakdown, maintenance task, component -where these are

5

relevant to the given report), its granularity can be set (for example

daily / weekly / monthly / annual granularity for the temporal aspect), and

they can be filtered according to the relevant objects (for example, date, turbine,

breakdown, maintenance, etc.).

The present system and procedure preferably treats all

1st

information related to breakdowns and maintenance along with

fault. A fault mode defines and describes the problem, for example, a

clogged oil filter. To solve a problem defined by the mode of

fault, the place where the fault occurred (for example, filter

of hydraulic oil from a specific turbine in a specific wind farm)

fifteen

and the maintenance operation (for example, changing the oil filter).

The three main aspects of breakdowns are as follows:

fault mode;

the wind turbine node instance (WTG), specifically the place

in which the fault occurred; Y

2nd

the maintenance operations that are necessary.

The system works from a failure mode library and

related maintenance operations. These libraries are called

template data, which provide a type of selection quot; a la cartequot; of the

possible instances of breakdowns and maintenance. Template data is

2 5

consider master data that must be defined when the system,

Normally implemented by a computer program, it is configured.

These three processes of entering master data are detailed in the

following paragraphs.

Wind turbine node instance input (WTG). He

3rd

flowchart for the input of node instance master data is

shown in figure 1 O. The main stages of the process are:

definition of the node template tree, which contains the tags

of type of instance nodes;

instance tree definition, specifying the parent-relationship

3 5

   child between instance nodes,

definition of the node properties of the node instances, and

addition of travel times to the nodes of the wind farm.

The parameters corresponding to the service and the spare parts

required are defined in the service master entry. Information

5

maintenance teacher refers to the service master records and

of parts, therefore these are preferably defined before they are

Create the maintenance master. The service master record can

advantageously contain the following data:

description, name of the service,

1st

cost in hourly rate of the use of a specific service,

delivery time, preferably in days.

The spare part master record defines the parts of

spare used in the maintenance operation. A record contains the

Following data:

fifteen

logistic control number,

description, name of the part,

price, price of a unit,

UOM, Unit of measurement, and

delivery time of the supplier, specifically the delivery time if

2nd

the supplier was requested, preferably in days.

Figure 11 shows a flow chart of the master inputs of

parts and service that show weather restrictions. The data

fault and maintenance masters are sets of information

closely related. This information represents the entrance to the

2 5

system. The main stages of the process flow are:

specify a fault mode,

associate the failure mode with instance nodes

define the set of possible maintenance tasks that could

used to repair a given fault mode.

3rd

For each maintenance task, the following are defined

restrictions: weather restriction, service requirements parameters

of spare parts.

Figure 12 shows synchronization of transactional data with

synchronization triggered by an event and data. The data

35

   transactional data are considered data that change dynamically in

contrast with the master data, which are considered static. The data

Transactional are:

parts inventories,

service calendars,

5

team calendars, and

weather forecasts.

Parts inventories and weather forecast have to

synchronize, preferably daily through a process

automatic. The service and equipment calendars are, rather,

1st

triggered by event. This is represented in figure 12.

Other features of the system and procedure according to this

invention are:

coverage of various types of wind turbines (WTG). The system is

able to manage different types of WTG, even within it

fifteen

eolico Park. It is also able to model different WTG that

build from different components, have different faults and

fault characteristics, and may require maintenance of different

ways; Y

The system is capable of managing all types of maintenance,

2nd

including corrective, preventive, and predictive maintenance, as well as

update activities and client tasks.

The system also enables the integration of fault detection and

prognostic agents. The system provides appropriate interfaces for the

fault detection and forecasting agents that recognize faults in

2 5

based on field data (for example, S EACH), and report failures to

through a standard fault entry process. In addition, the system

makes it possible for expert users to introduce breakdowns in a way

manual, or edit fault parameters recognized by agents

automated

3rd

In addition, the system displays SCADA data in diagrams

configurable, to allow experts to analyze the circumstances in the

that failures or alarms occurred.

The systems provide decision making with initiative

mixed. You must ensure that expert users can benefit from the

3 5

   integration of human and automated decision making in the

Maintenance planning Specifically, the system optimizes the

maintenance program using advanced mathematical techniques in

based on all the information modeled within the system, but also leaves

that the expert user tailor the program to your needs

5

real and to a broader scope.

Another advantage of the present invention is the consideration of all

important aspects of maintenance. When tasks are generated from

maintenance and program, the system must consider all

important aspects of maintenance: availability of equipment

1st

maintenance, spare parts and service requirements and parameters

of availability, weather conditions, loss of production

caused by different breakdowns, etc.

In addition, maintenance cost minimization is achieved

total. Decision support functions (especially generation and

fifteen

maintenance task scheduling) address minimization of

total cost related to maintenance, including spare parts,

maintenance requirements, and travel costs, as well as the

lost of production.

The system also issues configurable reports. The system

2nd

provides various configurable reports to analyze availability,

reliability and ease of maintenance of WTG from different

aspects.

The system supports web-based applications for

Multiple users It is implemented as an application based on a

2 5

Web page to allow multiple users (for example, engineers from

maintenance, service engineers, technicians, etc.) cooperate in the management

of wind farms (WF). For this purpose, read access rights

and user writing should be defined according to their role in the company

wind farm operator.

3rd

In addition, the system allows the establishment of interfaces with others

information systems. The system is able to import data from

various computer systems of the wind farm operator, including

SCADA (potentially different instances of SCADA for different

wind farms), ERP (Enterprise Resource Planning), and others

3 5

   maintenance management systems.

Presentation capabilities on both PC and PDA. Although that

most users will connect to the system from their computers

personnel (PC), technicians can consult the program of

maintenance and provide feedback through their PDAs. From

5

done, the pages frequently used by technicians (for example,

maintenance schedule, execution report) must be available

in low resolution format, also optimized for presentation in

PDA

Figure 13 shows a preferred embodiment showing the

1st

System architecture. This one is built as an application for

Multiple users based on a Web page, which is integrated with the

Existing computer systems of wind turbine operators. The

application is preferably a multi-layer business application

(J2EE). It stores data in a dedicated database. The logic

fifteen

enterprise is implemented in Enterprise JavaBeans. User interface

It is done as a quot; lightweight client; that runs in web browsers and

communications with the server through authenticated lines.

The system establishes an interface with various computer systems

of the wind turbine operator, as well as external services:

2nd

SCADA systems (imports SCADA measurements, mainly to

via JDBC connection), ERP system (imports WTG master data and

transactional data, for example, stocks and equipment calendars; through

XML file sharing - for example, with SAP XI, if the system

ERP is an SAP),

2 5

maintenance management system (look for requests from

periodic maintenance and update), and

   weather forecast service.

Claims (11)

1.-Maintenance system for wind turbines or wind farms,

said maintenance system having digital calculation means that

5

include:

at least one input connection (1) to receive parameters from

measured field of a wind turbine (WTG) or a wind farm system,

Y

a fault detection module (2) having means of

1st

storage with default parameters corresponding to

default fault modes and it emits a detection signal from

breakdowns, and

a troubleshooting module that has means of

storage with data in hierarchical nodal structure, in which said

fifteen

nodes correspond to subsystems or components of a wind turbine

(WTG) or a wind farm, issuing said location module for

faults a fault location signal (5),

characterized by

a fault estimation or forecast module (3) that has means

2nd

storage with default incipient failure parameters,

issuing said failure forecast module at least one signal of

estimated or predicted breakdown (4) for a component or subsystem, and

a maintenance programmer module (6) to emit at least

a maintenance output signal (7), said signal based on the

2 5

combination of said fault detection signal, said signal of

fault location (5) and said estimated or predicted fault signal

(4).

2. A maintenance system according to claim 1, wherein said

maintenance output signal (7) is a time signal of

30

maintenance (8) corresponding to maintenance operations

recommended and a repair date for a maintenance team.

3. A maintenance system according to claim 2, wherein said

maintenance output signal (7) identifies at least one component a

replace at a specific location of a park instance node

35

wind 0fVF) or the wind turbine (WTG) (9, 10).

4. A maintenance system according to claim 2, wherein said

Maintenance time signal (8) is also based on parameters of

maintenance restriction (11), said restriction parameters being,

weather conditions parameters, availability parameters of

5

spare parts or tools.

5. A maintenance system according to claim 4, wherein said

Maintenance output signal (7) is enabled, disabled or enabled

with a warning for a human operator, selectively, based on

values of said maintenance restriction parameters (11).

10

6. A maintenance system according to claim 4, wherein said

Maintenance restriction parameters (11) are based on parameters

Weather in the countryside near a wind turbine (WTG) or park

wind (WF).

7. A maintenance system according to claim 5, wherein said

fifteen

Weather parameters are wind speed, temperature

field, the height of the waves, the thickness of the fog or lightning.

8. A maintenance system according to claim 2, wherein said

Maintenance time signal (8) is also based on a profile of

loss, in which said loss profile is defined by the loss of

2nd

output power of the wind turbine in standard operation or in the

Maximum power generation.

9. Maintenance system according to claim 1, wherein said

storage means of said fault location module

they store hierarchical nodes (9, 1O) of each wind turbine (WTG) or

2 5

wind farm (WF), in which the node that is the highest in the hierarchy

corresponds to a wind turbine system (WTG), the second highest (9)

corresponds to its subsystems, especially the source module

Power, rotor module, gondola module and system

control and communication

30

10. Maintenance system according to claim 1, wherein said

maintenance module includes a multi-applicability matrix

dimensional that contains true or false values to correlate a

breakdown given at a given location and said maintenance output signal

(7).

35

11.-Maintenance system according to claim 1, wherein said

fault detection module comprises:

data filtration means for issuing a selected number of

parameters for a human operator, and

editing media that allow the human operator (13) to edit

5

fault parameters, and

an interface of the human being (12) to admit or reject an event of

fault.

12. Maintenance system according to claim 1, wherein said

maintenance programmer module (6) also includes:

10

a module for time-triggered synchronization (14) of

said maintenance output signal (7) having a time of

default update, especially an update time

daily, and

a module for synchronization triggered by an event (15)

fifteen

of said maintenance output signal (7).

13.-Maintenance procedure for wind turbines (WTG) or

wind farms (WF), which includes the stages of:

receive measured field parameters from a wind turbine system or

a wind farm, and

2nd

detect fault parameters and emit at least one signal from

fault detection, and

Troubleshoot according to data in hierarchical nodal structure, in which

these nodes correspond to subsystems and components of the

wind turbine or wind farm, issuing said location module of

2 5

breakdowns a fault location signal,

characterized by

estimate or forecast incipient fault parameters by issuing

minus an estimated or predicted breakdown date signal for a

component or subsystem, and

3rd

emit at least one maintenance output signal, based

said signal in the combination of said fault detection signal, said

fault location signal and said estimated fault date signal.

14. A maintenance procedure according to claim 13, in the

that said maintenance output signal (7) is a time signal of

35

maintenance (8) corresponding to maintenance operations

recommended and a repair date for a maintenance team. 15. A maintenance system according to claim 14, wherein said maintenance time signal (8) is also based on maintenance restriction parameters (11), said restriction parameters being 5 parameters of weather conditions, availability parameters of spare parts or tools. 16. A maintenance system according to claim 15, wherein said maintenance restriction parameters (11) are based on meteorological parameters in the field near a wind turbine (WTG) or 10 a wind farm (YVF). 17. A maintenance procedure according to claim 16, wherein said meteorological parameters are wind speed, field temperature, wave height, fog thickness or lightning. 15. A computer-readable program for implementing the maintenance procedure according to claim 13.