DE10211951A1 - Method and device for processing components of a technical system - Google Patents

Abstract

For a reliable and particularly precise analysis of all components of a technical system, according to the invention, in a method for analyzing components (K1 to Kn) of the technical system (1), in particular a power plant system, the underlying dependencies of the components (K1 to Kn) are in a structural module (14), whereby for a component to be analyzed (K7, K1 to Kn) on the basis of the dependencies stored in the structure module (14), the other components (K1 to Kn) that are related to this component (K7, K1 to Kn) are identified and be issued.

Description

The invention relates to a method for processing Components of a technical system, especially for Analysis of relationships between the components of a Power plant. Furthermore, the invention relates to a Device for processing components of technical Investment.

To control technical systems, such as one Power plant, or a product such as one Vehicle, are now commonly used to automate the Plant process or product process Data processing equipment used. Depending on the complexity of the technical system the data processing system is comparatively large dimensioned or decentrally structured. I.e. Data processing systems can be found on different scales versatile application in the context of automation processes technical facility that is comparatively large dimensioned complex systems in the industrial area or in decentrally structured, comparatively small Components in the mobile area or in consumer products can be used. The Data processing system always on a number of components, which for proper operation of the system with each other linked and networked.

Due to increasing security and Information requirements takes the complexity of the technical system out of it resulting in networking and the number of Components of the system to be examined. This in turn leads during a functional test of the components based on the increasing complexity and structuring to an extreme difficult and very time consuming analysis of through the Networking-dependent dependencies and combinations of Components.

Therefore, to simplify such an analysis usually the plant process or the plant as those in a variety of small structograms divided which dependencies for the user and Combinations of components for sub-processes manageable do. The disadvantage here is that a plant or cross-product change is difficult to capture. In addition, the effects of a change are only in the relevant structure diagrams can be identified. Effects of There are errors in the overall system or the overall product not verifiable.

The invention is therefore based on the object of a method for processing components of a technical system indicate that a reliable and particularly precise All components of the technical system can be analyzed.

This object is achieved according to the invention for a method for processing components of a technical system, in particular a power plant in which the underlying Dependencies of the components in a structure module are stored, using for a component to be analyzed of the dependencies stored in the structure module this component is related to others Components are identified and output.

The invention is based on the consideration that the Description of each plant on an analysis of all Components based on at least one descriptive of these Criterion can be reduced. One should all Components common as well as particularly simple and large Criterion covering a large number of possible individual functions can be used for analysis. Furthermore, the Component are processed unchanged in an analysis. As The criteria are the system process, the system and / or links characterizing a product or Component dependencies determined and analyzed. To the components on which the components are based are preferred Dependencies stored in a structure module. By means of the Dependencies stored in the structural module will Interaction of all components of the entire system completely certainly. The characteristic of the Component viewed, which the dependencies of the Components from each other and thus their processing in the Entire plant describes. For example, as Interdependency logical operations of components viewed. With in other words, the Interdependency all dependencies of all components in the system, that is, the effect of each component on each other, especially neighboring component.

In a particularly advantageous embodiment, the dependency the linking of pairs of components is determined and im Structure module deposited. When changing one of the Components of the system are therefore those for description and analysis of the components required knowledge reduced to Analysis and processing of pairwise dependencies of the modified components.

It is expedient to use those to be analyzed Component related components chronologically and / or output in a logical sequence. To do this, the each with the component to be analyzed in Components that are related to each other based on their influence over time and / or logically analyzed and ordered accordingly. there the components are ordered in time and / or logically Output form of chains, particularly shown.

Advantageously, the interrelation between direction of action representing two components deposited. This ensures that starting from the analyzing component all dependencies of this on other components recursively and / or predictively identified and created. The analysis is carried out regardless of the type of component. That means as a component becomes, for example, a product, a module, a function and / or considered a signal.

In particular in the event of a fault Being able to analyze the effects over time is one of the Representing the causal relationship between two components Response time stored. In addition to the identification of the Order of effects of the disruption on others Components the temporal sequence, especially the temporal History analyzed and output. This enables quick Evaluation and recording of the current situation in the event of a malfunction, whereby reactions to the elimination of the accident also in their temporal relationships can be analyzed.

To further improve the usability of the structural model, is expediently the respective component Weighting factor assigned, which is stored in the structural model. The weighting factor of the respective component is included preferably chosen in such a way that a prioritization from this or hierarchical gradation of the relevant component is clearly derivable. Alternatively, the weighting factor represent a value of the component in question. Depending on the type and design, the weighting factor is preferred given as a numerical value. Alternatively or additionally the weighting factor can be in the form of a functional Relationship of pairs of components, especially as Characteristic curve.

For the analysis of the process on which the system is based expediently all components based on them underlying dependencies using a table or stored in a database in the structural model. Doing so for example, a list in the manner of an assignment table of all components that the individual components in sets a relationship to each other. Depending on the level of structuring The system uses the components Allocation table analyzed and processed in such a way that by means of Simple assignment is checked whether a component is part another component or is not. This represents the simplest form of assignment. In other words: The structure module is used to create two components considered to each other, again each component Effects over other components is considered so that an impact chain of interacting Components is generated. It is by means of such chains of effects which is particularly easy for the operating personnel to identify additional components, additional modules or additional products in Are related to the fault message to be analyzed. In this way there is signal tracking and thus one Fault diagnosis is particularly reliable and quick. A Improved form can also be found by analyzing the relevant interaction of two components representative direction of action or reaction time be achieved.

It is expedient for a diagnosis of the system status an interference signal is selected as a component. Depending on the type and Execution can also be used for traceability purposes archived or an online recorded signal become. This enables a quick determination of this Online captured signal correlated other process signals and thus possibly cause detection for the error message can be determined particularly quickly and reliably.

In another application of the analysis method expedient for commissioning the system as Component selected a process signal. Based on the Structure module is used for the process signal by means of Dependencies as a chain of effects a chain of executable Test steps generated. That means based on the given Process signal is a number of to the operating personnel Testing an operation or signal necessary technical test or procedural steps issued. alternative or additionally a hardware and / or can be used as a component Software module can be selected. In other words: better clarity when analyzing an accident or commissioning are based on a predetermined signal, interference signal or process signal, all of those Signals, modules and / or functions, output, in particular displayed which the signal in question directly or indirectly processed.

The second object is achieved according to the invention by a device for processing components of a technical system, in particular a power plant, comprising a data processing system with a recording module to determine all components of the system, one Structure module for depositing the components underlying Dependencies and an analysis module to choose one too analyzing component and for the identification of with the component to be analyzed in the context of action other component based on that in the structural model deposited dependencies. Starting from the one to be analyzed Component can by such a device the entire Structure of any plant or product be created in detail regardless of whether the Component the product itself, a module and / or a function is.

The advantages achieved with the invention exist especially in that by a suitable and particularly simple Description of the. based on two relationships Connection, in particular the causal link between Components, networked and arbitrarily within a short time structured processes and / or systems with regard to their Performance and functionality evaluated and can be analyzed. Especially through the formation of Relationships between two functions or components and their evaluation is a selective consideration of Allows combinations of a single component. An elaborate one Assessment of complex structure diagrams is surely avoided. Such a reduction thus enables Two-way relationships of components a data and structure compression of the relevant plant to be analyzed, for example on the basis of the generation of impact chains, which for the Operating personnel are particularly quickly and safely manageable. In addition, such a device in addition to Process analysis can also be used for process control.

Embodiments of the invention are based on a Drawing explained in more detail. In it show:

Fig. 1 is a functional scheme provided for implementing an analysis method for a technical plant components,

Fig. 2 to 7 schematically illustrate different embodiments of a structural module.

A power plant 1 is shown schematically as a technical installation in FIG. 1. The power plant 1 is designed as a gas and steam turbine plant 2 . The gas and steam turbine system 2 comprises a gas turbine 4 and a waste heat steam generator 6 connected downstream of this on the flue gas side, the heating surfaces of which are connected to the water-steam circuit 8 of a steam turbine 10 .

Measured values MW recorded by sensors (not shown) and message signals MS issued by signaling elements (not shown) are fed to an automation system 12 . The measured values MW and the message signals MS are processed by means of the automation system 12 . Optionally, control signals SI are emitted to components of the gas and steam turbine system 2 . The gas and steam turbine system 2 is automatically controlled and monitored by the processes running within the automation system 12 . According to its structure, the plant process of the gas and steam turbine plant 2 and the automation process of the automation system 12 are divided into a number of components K1 to Kn.

Components K1 to Kn are both software and hardware components, in particular products. For example, a component K1 to Kn can represent a product Pa to Pz, a module Ma to Mz, a function Fa to Fz and / or a signal Sa to Sz. Depending on the type and structure of the gas and steam turbine system 2 defined as component K1, this includes, for example in the case of a software product in the manner of a subroutine of a computer program, a number of modules Ma to Mz with associated functions Fa to Fz that run in technical process steps, which in turn, if necessary Process signals Sa to Sz, which can be determined and processed as continuous components K2 to Kn.

For example, the component K1 comprises the gas and steam turbine system 2 as the product Pa, the modules Ma, Mb, Mc and Md, that is to say the gas turbine 4 (= module Ma), the waste heat steam generator 6 (= module Mb), the water-steam cycle 8 (= module Mc) and the steam turbine 10 (= module Md). Furthermore, the individual modules Ma, Mb, Mc and Md as further components Kn can in turn comprise functions Fa, Fb, Fc, Fd to Fz. For example, the module Ma, that is to say the gas turbine 4 , comprises as a function Fa: "emergency shutdown of the pump". This function Fa "emergency shutdown of the pump" is described by a combination of process steps, which generate control signals SI as signals Sa to Sz, as follows: "shut off the pump power supply" (signal Sa), "shut off the supply and discharge lines of the pump" ( Activate signal Sb) and "Reserve pump (signal Sc).

Thus, for the component K1 - the gas and steam turbine system 2 - the product Pa is a number of modules Ma to Mz, which in turn contains a number of functions Fa to Fz, which in turn comprise a number of signals Sa to Sz. For example, in the alternative gas and steam turbine system 2 defined as component K2 and product Pb, only one gas turbine 4 is processed as module Ma.

Depending on the function of the gas and steam turbine plant 2 in question , the components K1 to Kn are combined and structured accordingly.

A structure module 14 is provided for an analysis of the structure of the plant process of the gas and steam turbine plant 2 and of the automation process of the automation system 12 , in which the dependencies underlying the components K1 to Kn are stored as information I. By means of the structure module 14 , the components K1 to Kn are formed on the basis of a comprehensive list of all components K1 to Kn, that is to say all products Pa to Pz, all modules Ma to Mz and all functions Fa to Fz, by means of the correlations or interdependencies on which they are based , In other words, the dependencies of all components K1 to Kn are determined and assigned, for example, using a table. Alternatively, a database can be used instead of a table.

Depending on the type and structure of the structural module 14 , the products Pa to Pz representing the components K1 to Kn, modules Ma to Mz, functions Fa to Fz and / or signals Sa to Sz are determined chronologically or hierarchically and stored. Alternatively, an associated structure module 14 can be provided for each component K1 to Kn. For example, only products Pa to Pz and their dependencies are stored in a structure module 14 .

For an analysis of a type of gas and steam turbine system 2 , the product Pa is selected as the component K1 to be analyzed. For this purpose, an analysis module 16 is provided, which uses information I stored in the structure module 14 to identify and output the other components K2 to Kn that are functionally related to this component K1. Depending on the type and design of the analysis module 16 , the analysis method can be carried out recursively or with foresight. That is to say by means of the analysis module 16 , starting from the component K1 to be analyzed, the combinations or dependencies with other components K2 to Kn that follow in the process flow or automation process are identified and analyzed. The analysis method is therefore suitable for any analysis starting from any point or feature, that is to say starting from any product Pa to Pz, module Ma to Mz or function Fa to Fz.

Fig. 2 shows a possible embodiment for the structure of module 14 in the form of a table. All components K1 to Kn of technical system 1 are recorded, for example, in the form of a list. The components K1 to Kn then form both the rows and the columns of a matrix 18 . The interdependencies of the individual components K1 to Kn are arranged in the matrix 18 . For example, this is identified by the capital letters X and N. Components K1 to Kn, which are dependent on one another, are identified by X. Components K1 to Kn, which must not be linked, are identified by the capital letter N. Using structure module 14 , the entire system process or the entire automation process is thus analyzed and structured on the basis of simple two-way relationships. The simplest form is the table form. Alternatively, the relationship between two components K1 to Kn can also be described and recorded using a function diagram or another form of representation for a combination structure.

When analyzing the gas and steam turbine system 2 , the dependencies of the components K1 to Kn stored in the structure module 14 as information I are analyzed by means of the analysis module 16 . For this purpose, starting from a component K1 to Kn to be analyzed, for example from component K7, the further components K1 and K4 or K8, K13 and K15, which are functionally related to this component K7, are determined chronologically and / or in a logical sequence. The component K7 to be analyzed is defined as the basic component of a first level E1. The further dependent components K1, K4, K8, K13 and K15 determined by means of the structure module 14 are assigned to a further second level E2. The components K1, K4, K8, K13 and K15 assigned to the second level E2 are then again examined for dependencies on further components K1 to Kn. Further components K1 and K4 identified and dependent on components K1, K4, K8, K13 and K15 of level E 2 are assigned to a further level E3. This algorithm is carried out by means of the analysis module 16 until no dependencies between components K1 to Kn are identified. Such formation of levels E1 to En thus describes the branching or nesting of the dependencies of components K1 to Kn in the manner of a tree structure. The sum of all basic features of level E1 represents the entire system structure or the entire automation structure based on the component K7 to be analyzed in the form of branches or effect chains Wn. Such a branch or an effect chain Wn is shown for example in FIG. 3.

FIG. 3 shows the effect chains Wn for the components K1, K4, K8, K13 and K15 which are in the operative connection with the component K7 to be analyzed. Depending on the type and design of the analysis module 16 , these effect chains Wn are determined and output chronologically, that is to say in chronological order and / or in a logical sequence. A sub-branch represents a partial structure of system 1 or the automation process in the form of partial chains.

Alternatively, depending on the specification, the respective effect chain Wn according to FIG. 3 can be output in a forward-looking and / or retrospective manner. For this purpose, an analysis direction 16 is used to determine and analyze an action direction which is stored as information I in the structure module 14 and which represents the action relationship between two components K1 to Kn. In a further preferred embodiment, in addition to analyzing the direction of action of the interrelationships of several components K1 to Kn, a reaction time representing the interrelationship between components K1 to Kn can be determined and analyzed.

Such generation and output of effect chains Wn for the plant process or the automation process enables, for example, a particularly simple test routine which supports the operating personnel when the gas and steam turbine plant 2 is started up . For better clarity, starting from the specified component K7, all those modules Ma to Mz and functions Fa to Fz are processed and output as components K1 to Kn that process component K7 directly or indirectly.

In a further application, a change in the plant process or in the automation process can be identified on the basis of the comparison of effect chains Wn. An example of this is described in FIG. 4 on the basis of the expansion of the component K10, that is to say the expansion of the function Fj by further dependencies on the module Mb and the functions Fg, Fh and Fi (identified by the letter X in bold).

In FIG. 5, the force resulting from the function expansion changes of the plant structure are shown. FIG. 5 shows the generated on the basis of the analysis module 16 by means of the structural module 14 results chain Wn, wherein both the generated before the change of the function Fj results chains W and the generated according to the change of the function Fj results chains W'n are shown. Here, the effect chains W'n determined as new were identified by means of the analysis module 16 on the basis of a comparison of previous and current effect chains Wn before and after the change.

Instead of using effect chains Wn and comparing them, the dependency of components K1 to Kn on one another can be described by means of a weighting factor G. Exemplary is a weighting factor G in the form of a priority 1-5 stored in the Fig. 6 for the respective component K1 to Kn as an information I in the structure of module 14.

FIG. 7 shows an analysis of the weighting factors G carried out by means of the analysis module 16 , starting from the function Fj. On the basis of the weighting factors G stored in the structure module 14 , an average is formed for the components K1 to Kn forming an individual effect chain Wn. The value that is determined for the relevant chain of effects Wn represents the degree of correlation of the relevant components K1 to Kn of a single chain of effects Wn. A particularly low mean value of 1.5 for the weighting factors G of the relevant effect chain W6 represents a particularly high degree of correlation of the relevant components K1 to Kn.

The analysis method described here can also be used for process control. In this case, chronological processes are simulated using simple descriptions of relationships between various components K1 to Kn of the gas and steam turbine system 2 : "Switch on pump A", "Pressure A increases", "Open sampling valve B", "Pressure A falls". When analyzing an accident, the analysis signal 16 is then identified and analyzed using the structure module 14 as component K1 to Kn, the signal MS "a pipe is leaking".

On the basis of the effect chains Wn generated by means of the structure module 14 , the effects of the accident are identified before the actual subsequent messages detected by the plant process occur, as a result of which corresponding control signals SI can be generated. For example, the structure module 14 detects that the pressure drops. On the basis of the analysis of the dependencies by means of the structure module 14 , the reserve pump is then automatically switched on as a possible reaction for eliminating the malfunction before the actual signal MS "pressure drops" is detected. Alternatively, removal can be reduced. Depending on the type and design of the structure module 14 , with additional information I stored in the structure module 14 , such as, for example, stored reaction times or stored operating values, the operation of the system under the influence of the identified fault is predicted and described on the basis of the generation of effect chains. The analysis method described here can therefore not only be used for process control, but rather also for process analysis, simulation or forecasting of a technical system.

Claims (11)

1. Method for processing components (K1 to Kn) of a technical system ( 1 ), in particular a power plant, in which the underlying dependencies of the components (K1 to Kn) are stored in a structure module 14 , whereby for a component to be analyzed (K7, K1 to Kn) on the basis of the dependencies stored in the structure module ( 14 ), the other components (K1 to Kn) that are related to this component (K7, K1 to Kn) are identified and output. 2. The method of claim 1, wherein the with the analyzing component (K1 to Kn) in the operative context standing components (K1 to Kn) chronologically and / or in logical sequence. 3. The method as claimed in claim 2 or 3, in which an action direction representing the action relationship between two components (K1 to Kn) is stored in the structure module ( 14 ). 4. The method according to any one of claims 1 to 3, in which a reaction time representing the functional relationship between two components (K1 to Kn) is stored in the structure module ( 14 ). 5. The method according to any one of claims 1 to 4, in which the respective components (K1 to Kn) is assigned a weighting factor (G) which is stored in the structure module ( 14 ). 6. The method as claimed in one of claims 1 to 5, in which all components (K1 to Kn) are stored in the structure module ( 14 ) using a table or a database on the basis of the dependencies on which they are based. 7. The method according to any one of claims 1 to 6, in which as Component (K1 to Kn) selected an interference signal (Sa to Sz) becomes. 8. The method according to any one of claims 1 to 7, in which as Component (K1 to Kn) a process signal (Sa to Sz) is selected. 9. The method according to any one of claims 1 to 8, in which as Component (K1 to Kn) a hardware or software module (Ma to Mz) is selected. 10. Device for processing components (K1 to Kn) of a technical system ( 1 ), in particular a power plant, comprising a data processing system ( 12 ) with a structure module ( 14 ) for storing dependencies on which the components (K1 to Kn) are based and an analysis module ( 16 ) for selecting a component to be analyzed (K1 to Kn) and for identifying other components (K1 to Kn) which are related to the component to be analyzed (K1 to Kn) on the basis of the dependencies stored in the structure module ( 14 ). 11. The device according to claim 10, wherein a detection module ( 18 ) is provided for determining all components of the system.