WO2016163907A1

WO2016163907A1 - Monitoring of a mechanical device

Info

Publication number: WO2016163907A1
Application number: PCT/RU2015/000228
Authority: WO
Inventors: Mikhail Aleksandrovich KALINKIN; Alexander Vladimirovich LOGINOV; Jochen Luetche; Alexander Leonidovich PYAYT; Mark Schmitt; Michael Zidorn; Sergey Sergeyevich ZOBNIN
Original assignee: Siemens Aktiengesellschaft
Priority date: 2015-04-08
Filing date: 2015-04-08
Publication date: 2016-10-13

Abstract

The present patent application provides a monitoring system for use with a mechanical device, especially a gas turbine, comprising a number of sensors configured to measure physical data from the mechanical device, a data acquisition interface configured to acquire data from the number of sensors, a data processing unit configured to process the acquired data with complex event processing algorithms and/or stream processing algorithms, a data storage configured to store the acquired data and/or the processed data, and a user interface configured to receive configuration data from a user of the monitoring system and provide said configuration data to the data acquisition interface and/or the data processing unit and/or the data storage. Furthermore, the present invention provides a corresponding monitoring method.

Description

MONITORING OF A MECHANICAL DEVICE TECHNICAL FIELD

The invention relates to a monitoring system for use with a mechanical device, especially a gas turbine. Furthermore, the present invention relates to a corresponding method.

BACKGROUND

Although applicable to any system that has to be monitored, the present invention will be described in combination with mechanical or electromechanical systems like gas turbines.

Monitoring of modern mechanical or electromechanical systems is an important task to identify potential problems in time. This allows scheduling the maintenance of the respective system and reducing the overall down time of the system.

The most important tasks of condition monitoring e.g. of gas turbines are reduction of maintenance times and increase of the gas turbine availability. To sufficiently monitor a gas turbine it is necessary to collect measurement data from sensors installed in the monitored gas turbine and to fully understand and interpret the collected data.

The interpretation of the collected data is usually conducted by visual analytics. An operator of the gas turbine has to take decisions regarding the maintenance of the gas turbine based on real-time measurements.

Taking into account the huge amount of sensors and the high rate of measurements which leads to a huge amount of

measurement data, the task of data analysis becomes very complicated and requires a great amount of experience. Accordingly, there is a need for an improved monitoring of mechanical systems.

SUMMARY

This problem is solved by a monitoring system according to claim 1 and a method according to claim 12. Consequently, the present patent application provides a monitoring system for use with a mechanical device,

especially a gas turbine, comprising a number of sensors configured to measure physical data from the mechanical device, a data acquisition interface configured to acquire data from the number of sensors, a data processing unit configured to process the acquired data with complex event processing algorithms and/or stream processing algorithms, a data storage configured to store the acquired data and/or the processed data, and a user interface configured to receive configuration data from a user of the monitoring system and provide said configuration data to the data acquisition interface and/or the data processing unit and/or the data storage . Furthermore, the present invention provides a monitoring method for monitoring a mechanical device, especially a gas turbine, comprising the steps of measuring physical data from the mechanical device with a number of sensors, processing the measured data with complex event processing algorithms and/or stream processing algorithms in a data processing unit, storing the acquired data and/or the processed data in a data storage, receiving in a user interface configuration data from a user and provide said configuration data to the data acquisition interface and/or the data processing unit and/or the data storage, displaying the acquired data and/or the processed data in the user interface.

The present invention is based on the conclusion that the amount of data which is produced during the monitoring of mechanical devices like gas turbines is too big to be

effectively handled with traditional data processing like e.g. relational databases or the like. Therefore, the present invention uses this knowledge and combines modern complex event processing algorithms and modern stream processing algorithms to provide a flexible and adaptable monitoring of mechanical devices. A number, one or more, of sensors is used to measure physical data from the mechanical device. Typical sensor can for example be temperature sensors, pressure sensors, vibration sensors, current or voltage sensors, flow meters, speed sensors or tachometers, potentiometers or the like.

A data acquisition interface is provided to acquire the data from the sensors. The data acquisition interface can e.g. comprise a data converter, which converts the data from the different sensors into a standard data format, e.g. an XML- based data format or the like. This allows using different types of sensors or sensor data formats and provides for an easy extension of the monitoring system with new sensors.

The acquired data can then be processed in the data

processing unit using complex event processing algorithms and/or stream processing algorithms. Such algorithms are specifically designed to work with huge amounts of data and can therefore provide an efficient way of handling the acquired data.

The data storage can store the raw acquired data as well as the processed data. This allows reusing the acquired data for later reviews or comparisons. The monitoring system has great flexibility because the configuration of the monitoring system can be done via the user interface. Consequently, the monitoring system is not statically configured for a specific task or to provide specific data processing steps. Rather, the present invention allows a user to configure especially the data processing unit to analyse the acquired data in specific ways, therefore providing a user with detailed information about the state of the mechanical device.

Further embodiments of the present invention are subject of the further subclaims and of the following description, referring to the drawings. In a possible embodiment the user interface is a web

technology based user interface displayed in a web browser and comprises a visual configuration utility for configuring the complex event processing algorithms and/or stream

processing algorithms. The web technology based web interface can e.g. be based on HTML, especially HTML 5, and JavaScript or the like. This provides a very flexible user interface without requiring a dedicated operation system specific application. In a further possible embodiment the data processing unit comprises a standard library which comprises a number of standard data processing functions. This allows the user to easily combine said standard functions to analyse the

acquired data. Possible standard function can for example be a Fourier-Transform- Function, especially a Fast-Fourier- Transform-Function, an inverse Fourier-Transform-Function, other transformation functions, statistical functions, which calculate statistical values of the acquired data, or the like. In one embodiment advanced algorithms can be used for data analysis, like e.g. a combination of spectrum analysis with one-side classification.

In another embodiment the data processing unit is configured to receive user defined libraries which comprise user defined data processing functions. This allows the user to extend the functionality of the monitoring system according to his needs. User defined functions can for example be provided as linkable libraries or as program code, e.g. Java code,

JavaScript code, C# code or the like. In one embodiment the visual configuration utility is

configured to configure at least one data processing chain of the complex event processing algorithms and/or stream

processing algorithms for the acquired data based on a user input, wherein the at least one data processing chain is configured to use at least one of the standard data

processing functions and/or at least one of the user defined data processing functions. The visual configuration utility can e.g. provide a block-based building environment, where a user can generate data processing chains by linking

predefined blocks, which execute certain algorithms on the input data, with each other. The visual configuration utility can e.g. provide a variety of input blocks, which represent the different sensors of the monitoring system or which represent specific physical data from the mechanical device. Furthermore, every standard data processing function can be represented by a specific block. Finally, data displays can also be represented by blocks.

In another possible embodiment the user interface comprises a data display configured to display the acquired data and/or the processed data.

A user could for example select an input block, which

represents vibration of a shaft of a gas turbine and couple said block with a Fast-Fourier-Transform-Block. The user could then couple the output of the Fast- Fourier-Transform- Block with a two axis or a three axis diagram.

As a result the user could view e.g. in his web browser the result of the data processing chain, that is vibration of the shaft in the frequency domain.

In one embodiment a user can combine already existing data processing chains into a new data processing chain. In another embodiment the complex event processing algorithms comprise user defined rules provided by a user via the user interface. Every user defined rule defines conditions for the selection of a data processing chain. The user defined rules can e.g. define upper and/or lower threshold values for sensor data. If a user defined rule is satisfied, the

respective data processing chain can be executed on the respective sensor data. Especially if multiple data processing chains are stored, a specific data processing chain can be automatically selected based on the user defined rules. The results of the selected data processing chain can then be automatically displayed on the data display.

In one embodiment the conditions of the user defined rules comprise an upper threshold value for the acquired data of at least one sensor, and/or lower threshold value for the acquired data of at least one sensor, and/or an upper limit and/or a lower limit for a mean value and/or a variance for the acquired data of at least one sensor, and/or an upper limit and/or a lower limit for a statistical value for the acquired data of at least one sensor. In one embodiment the monitoring system comprises a

scheduler, which is configured to schedule the allocation and/or the deployment and/or the execution of the complex event processing algorithms and/or stream processing

algorithms between the data processing unit and/or the data storage and/or the user interface.

The data processing unit can e.g. comprise a distributed processing framework distributed over a plurality of

computers, like e.g. a cloud computing framework.

Furthermore, the data storage can comprise a non-relational database, which is configured to handle big amounts of data which rapidly changes . The advantage of this approach is that it greatly extends the scalability and elasticity of the monitoring system. For example it becomes possible to move computations to web browser which the user uses to interact with the monitoring system. Scheduling specific data processing algorithms to run in the web browser allows decreasing the computational load on back-end. The advantage of moving computations to the data storage, like a NoSQL database, where data analysis can be carried out directly in data storage, is that data analysis latency is reduced. In case of a very huge amount of data this also decreases network load because only the results need to be transferred to the user's web browser.

To simplify the distribution and scheduling of the different algorithms and data processing functions the algorithms and data processing functions can be provided in a scripting language like JavaScript. In one embodiment the data

processing unit, the data storage and/or the user interface can be configured to execute a common scripting language.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings are included to provide a further understanding of the present invention and are incorporated in and constitute a part of this specification. The drawings illustrate the embodiments of the present invention and together with the description serve to explain the principles of the invention. Other embodiments of the present invention and many of the intended advantages of the present invention will be readily appreciated as they become better understood by reference to the following detailed description. The elements of the drawings are not necessarily drawn to scale relative to each other. Like reference numerals designate corresponding similar parts.

For a more complete understanding of the present invention and advantages thereof, reference is now made to the

following description taken in conjunction with the accompanying drawings. The invention is explained in more detail below using exemplary embodiments which are specified in the schematic figures of the drawings, in which: shows a block diagram of an embodiment of a

monitoring system according to the present invention; shows a flow diagram of an embodiment of a monitoring method according to the present invention; shows a block diagram of another embodiment of a monitoring system according to the present invention; shows a block diagram of another embodiment of a monitoring system according to the present invention; and

Fig. 5 shows a block diagram of another embodiment of a

monitoring system according to the present invention.

In the figures like reference signs denote like elements unless stated otherwise.

DETAILED DESCRIPTION OF THE DRAWINGS

Fig. 1 shows a block diagram of an embodiment of a monitoring system 1-1 according to the present invention.

In Fig. 1 a mechanical device 2 is equipped with a plurality of sensors 3-1 - 3-n. The sensors 3-1 - 3-n can e.g. sense any physical data from the mechanical device 2 that is useful when assessing or monitoring the mechanical device 2. Typical sensors 3-1 - 3-n comprise temperature sensors, pressure sensors, vibration sensors, current or voltage sensors, flow meters, speed sensors or tachometers, potentiometers or the like. Furthermore, complex sensors equipped with a processing unit, like e.g. camera and image recognition systems or the like can be used as sensors 3-1 - 3-n. In general a sensor 3- 1 - 3-n can be any data source which delivers physical raw data or pre-processed data about the mechanical device 2.

The sensors 3-1 - 3-n provide data 4-1 - 4-n about the measurements to the data acquisition interface 5. In Fig. 1 the data acquisition interface 5 is displayed as a simple block, which forwards the data 4-1 - 4-n to the data

processing unit 6-1. In one embodiment the data acquisition interface 5 can comprise data conversion functionality which is capable of converting data 4-1 - 4-n which is provided in different formats by the sensors 3-1 - 3-n to standard formats, like e.g. XML-based data formats.

In one embodiment the data acquisition interface 5 comprises digital interfaces to communicate with sensors 3-1 - 3-n which also comprise a digital interface. Such digital

interfaces can be field bus interfaces, serial interfaces, parallel interfaces, USB- interfaces , Ethernet-interfaces or the like.

Furthermore, the data acquisition interface 5 can comprise analogue interfaces, like analogue-to-digital converters, which can be used to read data 4-1 - 4-n from sensors which provide the data 4-1 4-n analogue form.

In the data processing unit 6-1 the data 4-1 - 4-n is

processed by a combination of complex event processing algorithms 7-1 and stream processing algorithms 8-1. A data storage 9-1 is provided to store the data 4-1 - 4-n in raw and/or processed form.

Complex event processing implements techniques used to process and monitor in a large amount, especially in streams, of data the occurrence of certain events and derive specific conclusions from said events. A complex event refers to events which can only hardly be detected by a single base event. For example in the mechanical device 2 a vibration which exceeds a predefined threshold value, event 1, as a single event cannot serve to analyse the status of the mechanical device 2 because the vibration could happen due to a load change in the mechanical device 2 or any other

external source. Therefore using complex event processing algorithms the occurrence of a plurality of events can be monitored in the streams of data 4-1 - 4-n and a possible malfunction in the mechanical device 2 can be derived with more complete understanding of the actual situation in the mechanical device 2. The stream processing algorithms 8-1 refer to algorithms which allow performing parallel operations on huge amounts of data 4-1 - 4-n, which is advantageously provided as data streams. For example stream processing algorithms 8-1 can apply a predefined function or a number of functions on all or specific elements of the data stream.

By combining complex event processing algorithms 7-1 and stream processing algorithms 8-1 the data processing unit 6-1 allows e.g. activating specific stream processing pipelines only when the complex event processing algorithms 7-1 detects a predefined complex event.

By combining complex event processing algorithms 7-1 with stream processing algorithms 8-1 the present invention allows capturing a huge amount of data 4-1 - 4-n which could not possibly be processed by a human user 12. Instead the huge amount of data 4-1 - 4-n can be pre-processed and analysed by the complex event processing algorithms 7-1 and the stream processing algorithms 8-1.

The processed data 4-1 - 4-n can be provided to the user 12 via a user interface 10-1 of the monitoring system 1-1.

In one embodiment the data processing unit 6-1 comprises a distributed processing framework distributed over a plurality of computers. Such a distributed processing framework can e.g. be a cloud based processing framework, a server farm, or the like, in which the processing tasks can be distributed over a plurality of processing elements. Furthermore, at least part of the data acquisition interface 5, e.g. the data conversion functions, can also be implemented in the

distributed processing framework. The user interface 10-1 can comprise a back-end, which can be provided separate to or in the distributed processing

framework and can e.g. comprise a single or a plurality of webservers, or the like. A second element of the user

interface 10-1 can be a web page which can be loaded from the webservers using e.g. a web browser on a user's 12 computer. Using a web based user interface 10-1 allows providing a simple to use, flexible, scalable user interface which can be easily connected to the data processing unit 6-1 or the data storage 9-1.

The data- storage 9-1 can comprise data bases, which are optimized for working with huge amounts of data and frequent changes in data. This requirements cannot or only with difficulties be met by relational databases. Therefore, in one embodiment the data storage 9-1 comprises a so called NoSQL, not only SQL, database which doesn't rely on a

relational data model.

To connect the user interface 10-1 or the web browser of the user 12 to the data processing unit 6-1 and the data storage

9-1 a plurality of possibilities exist. In one example the work load is performed by the aforementioned web server which requests the data 4-1 - 4-n from the data processing unit 6-1 and the data storage 9-1 and the prepared data is provided to the user's 12 web browser. In another embodiment a JavaScript application in the user's 12 web browser receives from the web server the unprepared data 4-1 - 4-n and prepares the data 4-1 - 4-n for displaying to the user 12. The communication between the web server and the web browser can be based on any interface that can be used to transfer data from a web server to a browser. For example a REST-API can be provided by the web server and the web browser can request data 4-1 - 4-n via said REST-API. Other communication standards like SOAP, JSON, XML etc. can also be used.

In one embodiment the complex event processing algorithms 7-1 can be used on the data 4-1 - 4-n to detect complex events, which indicate a malfunction or a maintenance need of the mechanical device 2. In one embodiment the mechanical device 2 can be automatically deactivated if a malfunction or a maintenance need is detected. In addition or as an

alternative maintenance of the mechanical device 2 can be automatically scheduled or requested.

Fig. 2 shows a flow diagram of an embodiment of a monitoring method according to the present invention, which can be used for monitoring a mechanical device 2, especially a gas turbine .

To provide a detailed view of the mechanical device the method comprises measuring SI physical data 4-1 - 4-n from the mechanical device 2 with a number of sensors 3-1 - 3-n e.g. via a data acquisition interface 5 as explained in conjunction with Fig. 1.

To extract usefully information for monitoring the mechanical device the acquired data 4-1 - 4-n is then processed, S2, with complex event processing algorithms 7-1 - 7-3 and/or stream processing algorithms 8-1 - 8-3. This can e.g. be done in a data processing unit 6-1 as explained in conjunction with Fig. 1.

The acquired data 4-1 - 4-n and/or the processed data 4-1 - 4-n can be stored, S3, in a data storage 9-1 - 9-2, or displayed, S5, to a user 12. To provide flexibility in monitoring the mechanical device configuration data 11-1 - 11-2 can be received, S4 , from a user 12 and said configuration data 11-1 - 11-2 can be provided to the data acquisition interface 5 and/or the data processing unit 6-1 - 6-4 and/or the data storage 9-1 - 9-2. The configuration data 11-1 - 11-2 can especially be used to configure the complex event processing algorithms 7-1 - 7-3 and/or stream processing algorithms 8-1 - 8-3.

In one embodiment at least one data processing chain 18-1 - 18-n consisting of the complex event processing algorithms 7- 1 - 7-3 and/or stream processing algorithms 8-1 - 8-3 can be defined for the acquired data 4-1 - 4-n based on a user input.

The data processing chains 18-1 - 18-n can be configured to use standard data processing functions 15-1 - 15 -n, which can be provided by a monitoring system on which the method is executed. Furthermore, the data processing chains 18-1 - 18-n can be configured to use user defined data processing

functions 17-1 - 17-n.

The complex event processing algorithms 7-1 - 7-3 can

comprise user defined rules provided by a user 12, wherein the user defined rules define conditions based on the

acquired data 4-1 - 4-n and a respective data 4-1 - 4-n processing chain which is to be used if the respective condition is satisfied.

In one embodiment multiple data processing chains 18-1 - 18-n can be stored. A specific data processing chain 18-1 - 18-n can automatically be selected based on the user defined rules and the results from the selected data processing chain 18-1 - 18-n can automatically be displayed, e.g. on a data display 19.

The conditions which can be set by the user 12 comprise at least the following:

· an upper threshold value for the acquired data 4-1 - 4-n of at least one sensor 3-1 - 3-n

• a lower threshold value for the acquired data 4-1 - 4-n of at least one sensor 3-1 - 3-n • an upper limit and/or a lower limit for a mean value and/or a variance for the acquired data 4-1 - 4-n of at least one sensor 3-1 - 3-n

• an upper limit and/or a lower limit for a statistical value for the acquired data 4-1 - 4-n of at least one sensor 3-1 - 3-n

In one embodiment the method comprises scheduling the

allocation and/or the deployment and/or the execution of the complex event processing algorithms 7-1 - 7-3 and/or stream processing algorithms 8-1 - 8-3 between the data processing unit 6-1 - 6-4 and/or the data storage 9-1 - 9-2 and/or the user interface 10-1 - 10-3. This allows performing the necessary calculation where the most computational resources are available or where the least latency is generated, or the like.

Fig. 3 shows a block diagram of another embodiment of a monitoring system 1-2 according to the present invention which is based on the monitoring system 1-1 of Fig. 1.

For ease of understanding the block diagram of Fig. 3 only comprises the data processing unit 6-2 which receives data 4- 2 and provides the processed data 4-2 to a data display 19 of the user interface 10-2.

The data processing unit 6-2 is depicted as a cloud 6-2 which shall make clear that the data processing unit 6-2 is a distributed processing framework, like a cloud processing framework.

The data processing unit 6-2 comprises complex event

processing algorithms 7-2 which analyse the incoming data 4-2 for the occurrence of predefined or user defined complex events.

Furthermore, the data processing unit 6-2 comprises stream processing algorithms 8-2. In Fig. 3 the stream processing algorithms 8-2 comprise a standard library 14 with standard data processing functions 15-1 - 15 -n. The standard data processing functions 15-1 - 15-n can comprise functions like Fourier-Transformations, inverse Fourier-Transformations, other transformation functions, statistical functions, which calculate statistical values of the acquired data, or the like. In one embodiment advanced algorithms can be used for data analysis, like e.g. a combination of spectrum analysis with one-side classification. The stream processing algorithms 8-2 also comprise user defined libraries 16-1 - 16-n. The user defined libraries 16- 1 - 16-n can be provided by a user 12 and can comprise any user defined function 17-1 - 17-n the user 12 provides. The user defined libraries 16-1 - 16-n can e.g. be provided as linkable libraries, or as source code libraries, e.g.

JavaScript source code or the like.

In Fig. 3 a visual configuration utility 13 can be provided to the user 12. This visual configuration utility 13 can be a browser based visual configuration utility 13 where the user 12 can visually build data processing chains of the complex event processing algorithms 7-2, the standard data processing functions 15-1 - 15-n and the user defined data processing functions 17-1 - 17-n. The visual configuration utility 13 can e.g. provide visual blocks which represent the event processing algorithms 7-2, the standard data processing functions 15-1 - 15-n and the user defined data processing functions 17-1 - 17-n. The user 12 can then place the

necessary blocks in the visual configuration utility 13 and model data flows via connections between the respective blocks .

The results of the data processing can then be displayed in the data display 19, which can also be provided in a web browser of the user 12.

Fig. 4 shows a block diagram of another embodiment of a monitoring system 1-3 according to the present invention which is based on the monitoring system 1-1 of Fig. 1. For ease of understanding only the data processing unit 6-3 is shown in Fig. 4. The data processing unit 6-3 comprises complex event

processing algorithms 7-3. In the complex event processing algorithms 7-3 a number of user defined rules 20-1 - 20-n, or complex events, are shown, which define a data processing chain 18-1 - 18 -n which should be used for processing the incoming data 4-1 - 4-n if the specific rule 20-1 - 20-n is fulfilled.

The data processing chains 18-1 - 18 -n comprise building blocks 22-1 - 22-4 and 23-1 - 23-5, where the building blocks 22-1 - 22-4 represent data sources, e.g. single sensors 3-1 - 3-n. The building blocks 23-1 - 23-5 represent algorithms which should be used to process the data 4-1 - 4-n of the respective sources 22-1 - 22-4. In one specific example the data processing chains 18-1 - 18- n in the stream processing algorithms 8-3 can be built up of user defined rules, standard data processing functions, user defined data processing functions, and the like. Fig. 5 shows a block diagram of an embodiment of a monitoring system 1-4 according to the present invention which is based on the monitoring system 1-1 of Fig. 1.

For ease of understanding in Fig. 5 only the data processing unit 6-4, the data storage 9-2 and the user interface 10-3 are shown. Furthermore, in Fig. 5 a scheduler 21 is added to the monitoring system 1-4.

Data processing is shown by gears in the data processing unit 6-4, the data storage 9-2 and the user interface 10-3.

Furthermore, the data processing unit 6-4 comprises a data processing chain 18-2 consisting of sources 22-5 - 22-6 and algorithms 23-6 - 23-10. The scheduler 21 in Fig. 5 exemplarily assigned the

processing of the data processing chain 18-2 to the data processing unit 6-4. But, according to the task complexities, the required processing power, the location of the necessary data 4-1 - 4-n, and the like the scheduler 21 can also schedule processing tasks to run in the data storage 9-2, e.g. directly in a database system, or in the user interface 10-3, e.g. in a browser or on a web-server of the user interface 10-3.

Consequently, the processing functions, tasks, libraries and the like can be provided in one example in a data format, programming language or the like, which can be executed in the data processing unit 6-4, the data storage 9-2 and the user interface 10-3.

The present invention provides with all the above described features a monitoring 1-1 - 1-4 system which is easily scalable and can e.g. also be used to monitor a plurality of mechanical devices 2. Furthermore, the monitoring system 1-1 - 1-4 is easily extensible with new functionality via user defined libraries 16-1 - 16-n.

The web based user interface which can not only be used to view the collected data 4-1 - 4-n but can also be used to configure the monitoring system 1-1 - 1-4 allows using the monitoring system 1-1 - 1-4 with a plurality of different user devices, like PCs, Tablet-PCs, Smartphones, or the like. Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that a variety of alternate and/or

equivalent implementations exist. It should be appreciated that the exemplary embodiment or exemplary embodiments are only examples, and are not intended to limit the scope, applicability, or configuration in any way. Rather, the foregoing summary and detailed description will provide those skilled in the art with a convenient road map for

implementing at least one exemplary embodiment, it being understood that various changes may be made in the function and arrangement of elements described in an exemplary- embodiment without departing from the scope as set forth in the appended claims and their legal equivalents. Generally, this application is intended to cover any adaptations or variations of the specific embodiments discussed herein.

In the foregoing detailed description, various features are grouped together in one or more examples or examples for the purpose of streamlining the disclosure. It is understood that the above description is intended to be illustrative, and not restrictive. It is intended to cover all alternatives, modifications and equivalents as may be included within the scope of the invention. Many other examples will be apparent to one skilled in the art upon reviewing the above

specification .

Specific nomenclature used in the foregoing specification is used to provide a thorough understanding of the invention. However, it will be apparent to one skilled in the art in light of the specification provided herein that the specific details are not required in order to practice the invention. Thus, the foregoing descriptions of specific embodiments of the present invention are presented for purposes of

illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed; obviously many modifications and variations are possible in view of the above teachings. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, to thereby enable others skilled in the art to best utilize the

invention and various embodiments with various modifications as are suited to the particular use contemplated. Throughout the specification, the terms "including" and "in which" are used as the plain-English equivalents of the respective terms "comprising" and "wherein," respectively. Moreover, the terms "first," "second," and "third," etc., are used merely as labels, and are not intended to impose numerical requirements on or to establish a certain ranking of importance of their obj ects .

Claims

Patent claims

1. Monitoring system (1-1 - 1-4) for use with a mechanical device (2) , especially a gas turbine, comprising:

a number of sensors (3-1 - 3-n) configured to measure

physical data (4-1 - 4-n) from the mechanical device (2) ; a data acquisition interface (5) configured to acquire data (4-1 - 4-n) from the number of sensors (3-1 - 3-n);

a data processing unit (6-1 - 6-4) configured to process the acquired data (4-1 - 4-n) with complex event processing algorithms (7-1 - 7-3) and/or stream processing algorithms (8-1 - 8-3) ;

a data storage (9-1 - 9-2) configured to store the acquired data (4-1 - 4-n) and/or the processed data (4-1 - 4-n);

a user interface (10-1 - 10-3) configured to receive

configuration data (11-1 - 11-2) from a user (12) of the monitoring system (1-1 - 1-4) and provide said configuration data (11-1 - 11-2) to the data acquisition interface (5) and/or to the data processing unit (6-1 - 6-4) and/or to the data storage (9-1 - 9-2) .

2. Monitoring system (1-1 - 1-4) according to claim 1, wherein the user interface (10-1 - 10-3) is a web technology based user interface (10-1 - 10-3) displayed in a web browser and comprises a visual configuration utility (13) for

configuring the complex event processing algorithms (7-1 - 7- 3) and/or stream processing algorithms (8-1 - 8-3).

3. Monitoring system (1-1 - 1-4) according to claim 2, wherein the data processing unit (6-1 - 6-4) comprises a standard library (14) which comprises a number of standard data processing functions (15-1 - 15-n) .

4. Monitoring system (1-1 - 1-4) according to any one of claims 1 to 3 ,

wherein the data processing unit (6-1 - 6-4) is configured to receive user defined libraries (16-1 - 16-n) which comprise user defined data processing functions (17-1 - 17-n) .

5. Monitoring system (1-1 - 1-4) according to any one of claims 3 and 4,

wherein the visual configuration utility (13) is configured to configure at least one data processing chain (18-1 - 18 -n) of the complex event processing algorithms (7-1 - 7-3) and/or stream processing algorithms (8-1 - 8-3) for the acquired data (4-1 - 4-n) based on a user input;

wherein the at least one data processing chain (18-1 - 18 -n) is configured to use at least one of the standard data processing functions (15-1 - 15-n) and/or at least one of the user defined data processing functions (17-1 - 17-n) .

6. Monitoring system (1-1 - 1-4) according to any one of claims 1 to 5 ,

wherein the user interface (10-1 - 10-3) comprises a data display (19) configured to display the acquired data (4-1 - 4-n) and/or the processed data (4-1 - 4-n) .

7. Monitoring system (1-1 - 1-4) according to claims 5 and 6,

wherein the complex event processing algorithms (7-1 - 7-3) comprise user defined rules provided by a user (12) via the user interface (10-1 - 10-3), wherein the user defined rules define conditions based on the acquired data (4-1 - 4-n) and a respective data processing chain (18-1 - 18-n) which is to be used if the respective condition is satisfied;

wherein when multiple data processing chains (18-1 - 18-n) are stored a specific data processing chain (18-1 - 18-n) is automatically selected based on the user defined rules and the results from the selected data processing chain (18-1 - 18-n) are automatically displayed on the data display (19) .

8. Monitoring system (1-1 - 1-4) according to claim 7, wherein the conditions comprise one or more of:

an upper threshold value for the acquired data (4-1 - 4-n) of at least one sensor (3-1 - 3-n); and/or

a lower threshold value for the acquired data (4-1 - 4-n) of at least one sensor (3-1 - 3-n) ; and/or an upper limit and/or a lower limit for a mean value and/or a variance for the acquired data (4-1 - 4-n) of at least one sensor (3-1 - 3-n); and/or

an upper limit and/or a lower limit for a statistical value for the acquired data (4-1 - 4-n) of at least one sensor (3-1 - 3-n) .

9. Monitoring system (1-1 - 1-4) according to any one of claims 1 to 8 ,

comprising a scheduler, which is configured to schedule the allocation and/or the deployment and/or the execution of the complex event processing algorithms (7-1 - 7-3) and/or stream processing algorithms (8-1 - 8-3) between the data processing unit (6-1 - 6-4) and/or the data storage (9-1 - 9-2) and/or the user interface (10-1 - 10-3) .

10. Monitoring system (1-1 - 1-4) according to any one of claims 1 to 9 ,

wherein the data processing unit (6-1 - 6-4) comprises a distributed processing framework distributed over a plurality of computers .

11. Monitoring system (1-1 - 1-4) according to any one of claims 1 to 10,

wherein the data storage (9-1 - 9-2) comprises a nonrelational database.

12. Monitoring method for monitoring a mechanical device

(2), especially a gas turbine, comprising the steps of: measuring (SI) physical data (4-1 - 4-n) from the mechanical device (2) with a number of sensors (3-1 - 3-n) via a data acquisition interface (5);

processing (S2) the acquired data (4-1 - 4-n) with complex event processing algorithms (7-1 - 7-3) and/or stream

processing algorithms (8-1 - 8-3) in a data processing unit (6-1 - 6-4) ;

storing (S3) the acquired data (4-1 - 4-n) and/or the

processed data (4-1 - 4-n) in a data storage (9-1 - 9-2) ; receiving (S4) in a user interface (10-1 - 10-3) configuration data (11-1 - 11-2) from a user (12) and provide said configuration data (11-1 - 11-2) to the data acquisition interface (5) and/or the data processing unit (6-1 - 6-4) and/or the data storage (9-1 - 9-2) ;

displaying (S5) the acquired data (4-1 - 4-n) and/or the processed data (4-1 - 4-n) in the user interface (10-1 - 10- 3) .

13. Monitoring method according to claim 12, further

comprising :

configuring at least one data processing chain (18-1 - 18-n) of the complex event processing algorithms (7-1 - 7-3) and/or stream processing algorithms (8-1 - 8-3) for the acquired data (4-1 - 4-n) based on a user input;

wherein the at least one data processing chain (18-1 - 18-n) is configured to use standard data processing functions (15-1 - 15-n) and/or user defined data processing functions (17-1 - 17-n) ;

wherein the complex event processing algorithms (7-1 - 7-3) comprise user defined rules provided by a user via the user interface (10-1 - 10-3) , wherein the user defined rules define conditions based on the acquired data (4-1 - 4-n) and a respective data (4-1 - 4-n) processing chain which is to be used if the respective condition is satisfied;

14. Monitoring method according to claim 13,

wherein the conditions comprise

an upper threshold value for the acquired data (4-1 - 4-n) of at least one sensor (3-1 - 3-n) ; and/or

a lower threshold value for the acquired data (4-1 - 4-n) of at least one sensor (3-1 - 3-n) ; and/or an upper limit and/or a lower limit for a mean value and/or a variance for the acquired data (4-1 - 4-n) of at least one sensor (3-1 - 3-n) ; and/or

15. Monitoring method according to any one of claims 12 to 14, further comprising:

scheduling the allocation and/or the deployment and/or the execution of the complex event processing algorithms (7-1 - 7-3) and/or stream processing algorithms (8-1 - 8-3) between the data processing unit (6-1 - 6-4) and/or the data storage (9-1 - 9-2) and/or the user interface (10-1 - 10-3) .