WO2013170860A1 - A system for providing a service priority index for a wind turbine - Google Patents

Abstract

A service management system for managing service on a group of wind turbines where an issue relevant for service for a wind turbine is categorized and a weight which identifies a level of importance for issues is assigned to the issue. To facilitate service planning, a priority index is calculated by a processor based on the categorized issues and the assigned weights. The invention further provides a method of managing service.

Description

A SYSTEM FOR PROVIDING A SERVICE PRIORITY INDEX FOR A WIND TURBINE

INTRODUCTION

The present invention relates to planning of service and maintenance work in general on wind turbines. Particularly, the invention is concerned with condition based maintenance.

BACKGROUN D OF THE INVENTION

Wind turbines operate in remote and partly unpredictable environments.

Typically, they require frequent scheduled maintenance to obtain a predictable deterioration and thereby predictable performance. In spite of scheduled maintenance, malfunction may occur and unscheduled maintenance may reduce the predictability in power production.

Systems for health monitoring exist for online or offline monitoring of selected wind turbine components. These systems, however, offer no complete maintenance planning but rather allow the owner to react timely on malfunction

SUMMARY OF THE INVENTION

It is an object of the invention to improve maintenance and to facilitate scheduling and prioritisation of maintenance for wind turbines.

According to this and other objects, the invention, in a first aspect, provides a service management system for managing service on a group of wind turbines. The system according to the invention comprises data storage space and a processor in communication therewith. The data storage according to the invention is capable of communicating with at least one data provider which can report an issue relevant for service for a wind turbine which is subject to management by the system .

An "issue" may refer to any observation or information which is relevant for service on the wind turbine. By way of examples, an issue may refer to a scheduled maintenance such as oil change, inspection of sensitive components, observations made by the owner or by service personnel and relating to maintenance e.g. observation of noise etc. An issue may also refer to statistic observations, e.g. that a large amount of similar turbines have experienced problems with a specific component etc. Moreover, an issue may refer to events which is encountered by monitoring systems, e.g. of the kind generally referred to as "Health monitoring". As an example, an issue may refer to expected remaining useful lifetime of a bearing or other components in the wind turbine. Moreover, an issue may refer to a monitored use of the wind turbine. As an example, an issue may refer to the fact that the wind turbine has reached a certain limit production value, e.g. in rounds of the rotor, in kilowatt hours or similar measurable variable, or the issue may refer to an incident which has happened with the wind turbine, e.g. that it has been subject to a certain climate condition etc. The data storage according to the invention comprises predefined categories of issues such that reported issues which are reported by data providers can be categorized. The data storage further comprising, for each predefined category, a weight which identifies a level of importance for issues in that category of issues. As an example, the data storage may comprise one group for issues related to the main bearing, one group for issues related to the main axle in the gear box, one group related to cooling circuit for power generator etc. Each group may have an assigned number, e.g. a number from 1 to 10 which identifies a level of importance with regards to service and/or repair of issues in the group in question. According to the invention, the processor is capable of assigning, based on the categorized issues and the predefined weights, a service priority index for each wind turbine managed by the system . The priority index could be a number, e.g. between 1 and 1000 which identifies a priority of service for one turbine relative to the other turbines in the group of turbines.

The priority index could be determined based on arithmetic manipulation of the weights and number of issues, e.g. by summing up the weights of the issues for a turbine or by any other arithmetic operation which may provide an index based on the issues and corresponding weights.

The weight for a category may, as described already, be a number which identifies the importance of service, e.g. on a level from 1 to 10. In one embodiment, the weight is determined as a function of variable which has impact on the need for service. In the following, such a function will be referred to as a weight function".

The variable in question could e.g. be selected from a group consisting of temporal variables, climate depending variables, usage depending variables or combination between temporal, climate and usage depending variables. A temporal variable could e.g. be a duration until a latest point in time where service is required, e.g. for contractual reasons, a time since last inspection, the age of the turbine etc.

Climate related variable could e.g. express a mean wind load, a peak wind load, a mean temperature or a peak temperature, humidity or other climate

conditions which could be important with regards to need for service.

The usage depending variables could e.g. be number of usage hours, produced effect (kWh), number of rounds of the drive shaft, number of starts or stopping of the turbine, number of reorientations of the nacelle or similar figures.

By way of example, the weight could be specified as a certain initial value at a certain starting point and to increase in accordance with a function depending on duration in hours, days or usage of the wind turbine, etc. As an example, the weight could start with the value zero on day one and increase by a fixed number for each day which has passed since day one. Alternatively, the weight may have an initial value, e.g. zero when the number of rounds of the drive shaft has reached a certain initial value and increase with the number of rounds of the drive shaft exceeding the initial value.

The weight could e.g. be specified in form of a ramp function between two values forming upper and lower end values for the weight, e.g. a linear function or a parabolic function.

To enable an easy access to service scheduling data, the system may comprise an interface capable of extracting lists of turbines within the group. The lists could e.g. be extracted based on the assigned priority indexes such that the service attendant can start from the top of the list with that turbine having the highest index and which is therefore considered most suitable for immediate service.

Once a turbine is selected for service, the system may further facilitate the service attendants work by listing issues which are registered for the turbine in question, e.g. depending on the weight for each issue.

In addition to receiving issues from the data provider, the system may be capable of auto-generating issues for turbines, e.g. based on knowledge about need for service on other similar turbines. Accordingly, the system may be adapted to generate issues based on statistic information related to a turbine, a group of turbines, or based on statistic information related to turbine parts or information related to suppliers of parts for the turbine. This may be relevant when other related turbines or parts have shown a tendency of defects or includes a known error.

The system may also generate issues based on knowledge about service attendants. An example of the last mentioned issue could be when knowledge is received about a specific service attendant having generally made an error in a service procedure. In that case, the system may auto-generate an issue for any involved turbine. The system may further be adapted to assign weights to categories based on similar information about known related turbines, components or service attendants, i .e. the system may use statistic information related to turbines to define the weights for a category of issues. Generally, the issues may be received from any kind of data provider. By way of examples, the system may communicate directly with a SCADA system or generally with sensors provided in or near the turbine and capable of providing service related information, e.g. related to the use, loading or malfunctioning of the turbine or parts in the turbine. The system may also include an interface enabling import of data from a resource planning systems. Such information may e.g. relate to contractual obligations to carry out inspection or service or any other resource planning related information which may have impact on the need for service on the turbine. The system may also comprise a user interface for manual entering of data related to the need for service.

The system may also allow user input for entering or modifying categories, or weights or weight functions for the defined categories of issues.

In a second aspect, the invention provides a method of managing service on a group of wind turbines, the method comprising the steps of: defining categories of issues; defining for each category a weight which identifies a level of importance for issues in that category, receiving an issue relevant for service for a wind turbine in the group of wind turbines; categorizing the received issue in one of the predefined categories of issues; and

— assigning, based on the categorized issues and the predefined weights, a service priority index for each wind turbine in the group of wind turbines.

The method may include any further step relating to those features described relative to the first aspect of the invention.

DETAILED DESCRIPTION

In the following, an embodiment of the invention will be described by way of example with reference to the drawing in which :

Fig. 1 illustrates system architecture of a system according to the invention;

Fig. 2 illustrates data requirements for the Service Priority Index (SPI) server;

Fig. 3 illustrates levels of authorization for a system according to the invention;

Fig. 4 illustrates further process details of the system; and Figs. 5 and 6 illustrate further details.

Further scope of applicability of the present invention will become apparent from the following detailed description and specific examples. However, it should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration only, since various changes and modifications within the scope of the invention will become apparent to those skilled in the art from this detailed description.

The purpose of the service management system according to the invention (in the following shortly referred to as "the system") is to optimize maintenance planning - the system may therefore be seen as a system for providing priority index for planning purposes. Thus the system may not only reflect the condition of the turbine from a technical point of view but also instances of overdue scheduled service or other required service activities related to a wind turbine.

The system may e.g. be used by wind turbine owners, by companies servicing wind turbines, or by makers of wind turbines.

Fig. 1 illustrates a system architecture which could be suitable for the system.

The wind farm 1 comprises a group of wind turbines which are managed by the system. Each turbine comprises sensors which monitors power production, loading of blades, drive train etc., or which measures other parameters which may influence the need for service. These data are transmitted, e.g. by wireless communication, to the SCADA system 2 where the data is stored and/or treated statistically or in other ways transformed into useful maintenance information. The information from the SCADA system, herein referred to as "turbine data" \s communicated to a data warehouse server 3 which collects data from different sources. An ERP system 4 provides data related to planned maintenance, contractual obligations, service attendant availability etc. The planning data is send to the data warehouse. Other sources 5 may further provide data to the system . This may e.g. be manually entered data relevant for service on one or more turbines. The data warehouse server 3 communicates with the front-end-server system 6. The front end server system may also communicate with other data providers. This is visualized by the arrow 7. As an example, categories of issues may be defined manually or weights assigned to categories may be changed manually at the front end server. The front-end server 6 forwards the information to a service priority index server, herein referred to as the "SPI" server. The SPI defines, based on the received information, a number of issues and assigns categories to the issues. Based on predefined weights or weight functions for each category, the SPI server calculates a service priority index for each turbine in the managed group of turbines. For each turbine, the SPI server further establishes a list of issues with corresponding weights. This list could be seen as a prioritization of the issues for each individual turbine. Accordingly, two levels of priorities can be considered, i .e. the priority of one turbine relative to the other turbines in a group of turbines, and the priority of one issue relative to the other issues for an individual turbine.

Data from the SPI server can be exported to other applications visualized by the server 9. These applications may include EPR systems such as a SAP planning system and other systems which may benefit from the prioritization of the service tasks. Fig. 2 illustrates data requirements for the SPI server including, without being limited to:

• Parameter metadata,

• data parameters

• Alert weight/severity (default and user specific) · Alert Acknowledge Status

• User information

• Turbine information

Parameter Metadata : The parameter metadata is the information regarding an individual input parameter. An example of parameter metadata could be a turbine monitor ID and description. Each parameter has a type e.g. "continuous improvement management" or "wind turbine monitor". The parameter metadata also contains information about how to convert raw data to normalized data which can be used for system input. The parameter metadata also contains the mapping of input parameters to turbines. Data Parameters: The data parameters are the raw data that goes into the calculation of the system . An example of a data parameter could be a specific (current) wind turbine monitor alert level.

Turbine Information : a data presentation server will provide all relevant information such as turbine serial number, park name, sales business unit, turbine type, mark version, controller type, software release etc.

User Information : It may be possible for individual authorized users to modify the weights of the different categories and thus issues. In addition it could be possible for users to define a selection of turbines and to store this information. Authorization may exist in 3 levels as indicated in Fig. 3 namely:

• Global change access: The user may be able to set the default values of all input parameters.

• SBU (Sales business unit) change access: The may be able to modify the default values of all input parameters on SBU level if the SBU wishes to overrule the default values.

• Turbine specific change access: The user may be able to modify values of all input parameters on individual turbine level if the user wishes to overrule the default values and the SBU defined values. Access is required from an SBU administrator. · Read access, internal : The user may be able to read all the system information but not change or acknowledge anything. This access level is default for all users in an active directory for the system (AD).

• Read access, external turbine owners or operators: The user may be able to read the system information for the selection of turbines that user has access to, but without being able to change or acknowledge issues. The user may in other words be permitted a read access through the system . • Global administrator: Administrators may be able to handle all security settings.

• SBU administrator: Administrator may be able to create internal users and assign change access to an internal user for turbines in the specific SBU. SBU administrators are created by the global administrators.

The alert weight/severity defines how to weight issues in the different predefined categories relative to each other, e.g. how severe is overdue scheduled service compared to a red wind turbine monitor system alert. In one embodiment, a default definition of weights is defined, and individual users may be permitted an access to change the weights. Weights should probably only be changeable by authorized users.

The raw data is provided by the date warehouse. This data can arrive in any format (typically dates, Booleans, numbers, or text strings). The data could be converted into a normalized system input format which is a single number that is equivalent to a priority level. The system will produce an output which is a priority number for the given turbine. The parameter metadata contains information about how to normalize the particular type of data as well as descriptions. If the parameter metadata is not placed in data warehouse, it must be possible to import it from the data warehouse since information from the wind turbine monitors is stored in the data warehouse.

The turbine information contains information about the individual turbines including the turbine hierarchy, e.g. sales business unit > country > wind farm > turbine. The process is illustrated in Fig. 4. A relevant part of the system is the normalization procedure of the raw data. The system could in general rank according to the severity of the "worst" of the underlying input parameters, i.e. "a single severe problem or issue is worse than two less severe issues".

Figs. 5 and 6 illustrate definitions of the weights of the wind turbine. The items listed below are examples of input sources for defining issues, categorizing the issues, determining weights of the issues, and finally for determining a service priority index. The list may be expanded.

Input from wind turbine monitor system Data Source could be a wind turbine monitor data Base on the SCADA system. The Turbine Monitor is an application that monitors various turbine failure modes. The system issues an alert if a problem or issue is detected. A typical system input from wind turbine monitor system would be an alert level for a given turbine. An example could be a vibration monitoring system . Input from an Overdue scheduled service

Data Source could be an ERP system. Typically, turbines have scheduled service performed on a regular basis within a certain window. If the service provider for this task fails to fulfill these requirements the provider could be charged with penalties. The system input in this case could be days to next scheduled service. Input from Pending CIM case implementations

Data Source could be an ERP system . This relates to planned improvements for selected turbines e.g. due to safety issues. These must typically be implemented within a certain time frame.

Input from Component Inspection Reports Data Source could be an inspection reporting system associated with the Service priority index system according to the invention. Often, component inspection reports are made for selected turbines. These reports contain information about component inspections. Information from these reports is valuable since it enables components to be changed or fixed on a proactive basis.

Input from calls to service desk Data Source could be an ERP system or any similar reporting system . This may enable customer calls to be recorded e.g. from a technical support desk.

Examples of issues which could be recorded in this way are unusually loud noise, vibrations in the tower, and visually detectable faults in general . Input from recent faults and warnings

Data Source could be an ERP system or any similar reporting system. Faults and warnings are usually a sign of problems and recent alarms and warnings may therefore advantageously be taken into account and issues may possible be recorded based on such previous faults.

Claims

1. A service management system for managing service on a group of wind turbines, the system comprising a data storage and a processor,

— the data storage being capable of communicating with at least one data provider which can report an issue relevant for service for a wind turbine in the group of wind turbines, the data storage comprising predefined categories of issues such that reported issues can be categorized, the data storage further comprising, for each predefined category, a weight which identifies a level of importance for issues in that category of issues, and

— the processor being capable of assigning, based on the categorized issues and the predefined weights, a service priority index for each wind turbine in the group of wind turbines.

2. A system according to claim 1, where the weight for at least one category of issues is defined as a function of a variable.

3. A system according to claim 2, where the variable is selected from a

group consisting of temporal variables, climate depending variables, usage depending variables and combinations between temporal, climate and usage depending variables.

4. A system according to claim 2 or 3, where the weight is specified in form of a ramp function between two values forming upper and lower end values for the weight.

5. A system according to any of the preceding claims, where the processor is capable of listing turbines based on the assigned priority indexes.

6. A system according to any of the preceding claims, where the processor is capable of listing issues for a selected turbine based on the weights of categories in which the issues are categorized.

7. A system according to any of the preceding claims, where the system is adapted to define issues based on statistic information related to a turbine, a group of turbines, or a turbine parts.

8. A system according to any of the preceding claims, where the system is adapted to assign weights to categories based on statistic information related to turbines.

9. A system according to any of the preceding claim, where the system is adapted to communicate with a sensor providing sensible information related to a turbine and thereby constitutes a data provider for the system .

10. A system according to any of the preceding claims, further comprising an interface enabling import of data from a resource planning systems containing planning information related to the turbine and which thereby constitutes a data provider for the system .

11. A system according to claim 10, where the planning information relates to scheduled maintenance.

12. A system according to any of the preceding claims, where the planning information relates to scheduled replacement of parts in the wind turbine.

13. A system according to any of the preceding claims, where the system

comprises user input means adapted for entering input parameters which are relevant for maintenance of the at least one wind turbine, the input means thereby constitutes a data provider for the system .

14. A system according to any of the preceding claims, where the system

comprises user input means adapted for entering or modifying weights or functions for weights for the categories of issues.

15. A system according to any of the preceding claims, where the system comprises user input means adapted for entering or modifying categories of issues

16. A method of managing service on a group of wind turbines, the method comprising the steps of: defining categories of issues; defining for each category a weight which identifies a level of importance for issues in that category, receiving an issue relevant for service for a wind turbine in the group of wind turbines; categorizing the received issue in one of the predefined categories of issues; and

— assigning, based on the categorized issues and the predefined weights, a service priority index for each wind turbine in the group of wind turbines.