DE19742448C1 - Diagnostic module for electric automation circuits for overall system diagnosis - Google Patents

Abstract

The diagnosis is based on models of the overall system and has arrangements for setting up a reduction diagram with a main diagram whose nodes have only fixed or measurable values, for detecting the measurable values required for the diagnosis, for interpreting the measurable parameters in the main diagram, for forming the interpretation conducted in the main diagram on the plane of fundamental parts and for forming the interpretation on the plane of the components. The last arrangement yields the diagnosis results in the form of components suspected of being faulty and faulty components. The first arrangement divides the systems into components and fundamental parts and analyses their interconnections, states, measurable parameters and influence.

Description

The invention relates to a diagnostic module for creating a diagnosis for electrical controllable systems arranged within an overall system.

Furthermore, the invention relates to a diagnostic device for an overall system with electrically controllable and arranged within the overall system Systems.

For a variety of applications from electrically controllable components standing complex networked overall systems such. B. from the field of Automation technology, it is process control engineering, circuit development or vehicle electronics necessary for fault diagnosis, the normal and faulty behavior of the individual Components, the effects of this behavior on other components and in particular to know exactly the entire system.

Through data material about the overall systems, e.g. B. from their development phase, it is possible to describe the behavior of the components using models and by simulation knowledge data about the behavior of the individual components to generate controlled. Known simulation methods allow the after-picture of individual components and subsystems, but fail very often if the Simulations are to be extended to the entire system. This is essential Lich if the effects of the behavior of individual components or the behavior of Subgroups to be examined for the overall system. More declarative Description procedures take the diagnostic knowledge from libraries or specifically exploit knowledge of experience.  

DE 41 24 542 C2 describes a fault diagnosis device for determining a Cause of error in a tested device with one detection device and one Storage device described, wherein the detection device, the parameters of the tested device detected. In the storage device are a search tree with Kno which correspond to the respective subunits of the tested device, as well as the Nodes each assigned test tables, in each of which at least one of the Detection device to be detected parameters and a related test are specified, an error probability table corresponding to the Results of tests according to the at least one test condition and name of Daughter nodes stored in advance, being in a test table that a node with is assigned to at least three daughter nodes, additionally at least two to detek parameters and test conditions are given. In addition, in the A search / interference device, which ent selects nodes along the search tree and evaluates the associated test tables, doing the node selection based on the result of the evaluation of the test tables makes.

US Pat. No. 5,099,436 describes a method and an apparatus for Perform a system fault diagnosis described on a hybrid Knowledge representation of the system to be diagnosed based. During the System runtime data is recorded with an event-based system display compared, which includes a variety of predefined events. An event will recognized when the captured data with the critical parameters of the event to match. The detected event and an associated ambiguous sentence effects group effects, which identify components that correspond to a assigned sort effect in an ambiguity group should be analyzed. You can also use a symptom defect model and a Nonfunctional model to be analyzed to show the symptom error relationships and the Determine the type of non-functions that are applicable to the system run. Each Applicable symptom error relationship and any type of nonfunction will also associated with a set of ambiguity group effects that the ambiguity group sorted again. Starting with those components in the ambiguity group whose failure is most likely to become a structural model analyzed, and as a result of the analysis, repair proposals are made on System to run tests issued.

From the patent US 5,661,668 is the use of graphene and their Simplification for complex systems, including diagnostic devices len, removable. The correlation of observed events to specific pro blemen is split into two activities: the creation of efficient groups pen of error events and in decrypting the event stream.

Redundant data is generated when the groups of error events are generated eliminated, which leads to a greatly reduced amount of data. The groups of Feh Learning events are compared with observed symptoms, which is the necessary Computational complexity significantly reduced.  

Structural principles of a computer-aided fault diagnosis device for a Motor vehicle are in the publications N. Waleschkowski et al., "A knowledge based vehicle diagnostic system for use in the vehicle workshop "Grund and applications of artificial intelligence, Springer-Verlag, 1993, page 277, as well as N. Waleschkowski et al., "Knowledge modeling and knowledge acquisition on Example of vehicle diagnosis ", Artificial Intelligence magazine KI 1/95, page 55, described. This facility includes a diagnostic workflow stage a knowledge base that provides a structural model of the hierarchical structure of the technical system from individual subsystems, an impact model on the Relationships between the individual subsystems and a diagnosis  process-defining error model that includes the relationships between Error causes and their effects as well as suitable test procedures and repairs representations.

These known methods for generating diagnostic knowledge are are declarative procedures, d. H. Errors are only determined on the basis of known and previously defined or set events. Certain previously known ones Results chains are also saved and allow a description of the System behavior by the symptoms found during diagnosis by Ver certain faults can be assigned directly with the given effect chains. Errors can thus be indicated by the known symptoms recognized pattern recognition processes.

So z. B. the specification of normal behavior in a certain loading driving state and the comparison with reality the recognition and localization of Mistakes. In practice, limit values are often used, which in the case of over- or Indicate a shortfall if there is an error.

Declarative procedures are not suitable where the ones to be diagnosed are Change sub-systems or entire systems continuously and sometimes at very short notice and further develop and use several different variants at the same time are. The declarative diagnostic systems then require a high and maintenance effort.

The object of the invention is to provide a diagnostic module and a diagnostic to provide direction for making a diagnosis, the disadvantages mentioned do not have.

According to the invention, this object is achieved by the diagnostic module with the Features of claim 1 and the diagnostic device according to claim 7.

The diagnostic module creates a slide based on models of an overall system for electrically controllable and arranged within the overall system Systems. It contains means for drawing up a reduction graph that has a head graph with nodes that have only fixed or measurable sizes.

The means for setting up the reduction graph comprise means for decomposing the electrically controllable systems in components that have one or more basic comprise building blocks, means for detecting the electrical connections between the basic building blocks and between the components, means for assigning discrete electrical status values for the basic building blocks, means for definition ren of the operating states and the possible component states of the individual Components in relation to the electrical state values of a compo nent basic building blocks, means for determining the necessary for the diagnosis agile and measurable sizes, means on the basic building blocks and components for determining basic building blocks based on the specified measurable sizes  have no influence and means to summarize and eliminate the mitit telten uninfluenced basic building blocks.

The diagnostic module further contains means for detecting the need for the diagnosis agile measurable sizes, for interpreting the measurable sizes in the head graph, for mapping the interpretation carried out in the header graph onto the level the basic building blocks and for mapping those depicted on the basic building blocks Interpretation at the level of the components, the latter being suspect and / or specify faulty components as the diagnosis result.

The diagnostic module takes into account both the structure of the system components and also the relationships between the components, the components and the Basic building blocks and component behavior. The overall system with its electrically controllable systems is thus about the direct use of its Circuit diagram data is displayed and the huge amounts of data are meaningfully compressed lubricated.

The error message is generated using a simple model that is based on nodes with fixed or measurable sizes reduced head graph, analyzed. The error analysis takes place thus only at the time of diagnosis. This offers the advantage up to the diagnosis time additional knowledge about the entire system or about subsystems in to include the error analysis.

So z. B. also during the life of the system to be diagnosed due to aging processes, specific user behavior or environmental influences changing default probabilities can be taken into account.

The knowledge generation required for diagnosis takes place directly from the con structural documents or circuit diagrams of the overall system to be examined by means of the reduction process. To do this, the design documents must Component information with the electrical status values and notes included on the measurable sizes. With this type of diagnostic knowledge generation the otherwise required time-consuming simulations that and failure of individual components to be transferred to the overall system.

The diagnostic device provides an error statement about the overall system. she contains, in addition to the diagnostic module for creating the models of an overall system-supported diagnosis for the electrically controllable and within the Systems arranged overall means for creating a functio on the correlations between the individual electrically controllable systems supported diagnosis for the entire system. With these means one is made up of two parts existing graph can be created, the first subgraph showing the structure of the Components and the second subgraph show the relationships between the total system functions and the functions of the individual components. The Diagnostic device further contains means for generating error statements about the entire system as well as statements on the impairment of use of Ge overall system functions.  

The diagnostic device not only delivers defective components but also hits furthermore error statements about the overall system, whereby it for the It is particularly advantageous for users of the overall system to recognize whether and where one there is a faulty function.

Another advantage for the user is if the diagnostic device is additional Means for suggesting appropriate remedial measures include, as in sub claim 12 executed.

Further advantageous refinements are specified in the subclaims.

Preferred embodiments of the invention will be be based on drawings wrote, of which show:

Fig. 1 shows a reduction graph

Fig. 2 shows the procedure for creating the diagnosis based on system models

Fig. 3 shows the starting situation for the analysis of a measurable node a in the head graph

Fig. 4 the conclusion about a star reduction.

The main components in the periphery of automation devices occur, are resistors, switches, lines, plugs, lamps, bundles of lines, Sensors, relays, fuses, solenoid valves, motors and signal transmitters. A part the components can be summarized in aggregate A. The aggregatin forma Structural relationships are usually processed in a data processor tional form available. This information is already during the product development phases e.g. B. present in CAD systems or can model-based programs are derived.

The periphery of an overall system can be modeled as a graph, in which the reason building blocks that represent edges. Ver connection points necessary, which appear as nodes in the graph.

The entire behavior of the electrical periphery is characterized by the elementary peripheral graph G _ep .

In an elementary peripheral graph G _ep , the node potentials P _i or the currents I _i can be measured by a basic module. The measurability of the node potential P _i or the branch current I _i depend on the local position and the structure of the entire system and can take the values measurable or not measurable.

Starting from the elementary peripheral graph G _ep , the basic operations known from network analysis can be used to reduce the graph, which contains the essential behavioral properties with regard to the observable physical quantities.

As shown in FIG. 1, a reduction graph G _{r can} consist of several subgraphs. These include the elementary peripheral graph G _ep , several intermediate graphs G _t and a head graph G _h . The nodes and edges of the individual subgraphs are connected to each other and represent the reduction rule.

The electrical ones contained and specified in the elementary peripheral graph Properties are measurable by the reduction process Sizes simplified. The behavioral properties of the electrical components are clearly mapped to the elements in the header graph due to the reduction. There against is a clear mapping of the elements based on the header graph the component level is not ensured. This property is due to the limited information in relation to the causative components. Practically this means that a particular fault symptom has multiple causes of fault characterized.

The reduction takes place with the basic operations. After each reduction run, the number of nodes N _i decreases. The elementary basic graph G _ep is processed until no node or edge can be removed from the graph.

At the end of the reduction process, the head graph consists of nodes with fixed or measurable potentials. All nodes with unknown potentials are removed in the head graph G _h . The header graph now contains all the discrete states specified in the elementary periphery graph.

Especially for complex overall systems with a large number of components the time required for the creation of the reduction graph is considerable.

It may therefore be advantageous to have one before the actual reduction process To carry out clustering, which takes into account the knowledge that not all Be drive and faulty conditions of components affect all system sizes. It Rather, there are spatially independent areas in which certain states exist of components, and areas that are not affected by all components be influenced.

The clustering allows even very complex networked overall systems to be included search because the subsequent diagnosis is reduced to a local problem.

In FIG. 2, the process sequence is shown for creating based on system models diagnosis schematically.

First, a reduction graph is to be drawn up using the steps described above len, which forms the starting point for model-based diagnosis. For that are the electrically controllable systems in components with one or more Disassemble basic building blocks.  

Then the electrical connections between the basic modules and detected between the components and the basic building blocks discrete electrical State values assigned. In relation to the electrical state values of the Basic components belonging to a component are the operating states and the possible component states of the individual components are defined and the necessary for diagnosis and on the basic building blocks and components measurable sizes set. Finally, basic building blocks are identified that are based on the specified measurable quantities have no influence. The determined influence loose basic building blocks are summarized and eliminated.

The reduction graph set up in this way has a head graph, whose nodes either have a fixed potential or whose potential can be measured at the node. After recording the measurable quantities necessary for the diagnosis the measurable sizes interpreted in the head graph.

In addition, means for detecting error messages can be provided. A Incoming error message is detected by the electrically measurable quantities are monitored and available as input information for diagnosis stand.

Known error detection methods can be used to distinguish between normal and incorrect behavior of a subsystem. The resulting symptoms or error messages are interpreted for the subsequent diagnosis. The interpretation takes place in the header graph G _h taking into account all information necessary for the diagnosis in the form of measurable physical quantities.

The explanations for the symptom that occurred in the head graph G _{h represent} a further processing step on the basic building blocks of the elementary peripheral graph G _ep . This enables conclusions to be drawn in the reduction graph at the basic building block level.

The interpretation depicted on the basic building blocks is then mapped to the component level. The transfer of calculated discrete values of the elementary peripheral graph G _ep to the component level makes it possible to determine possible faulty and / or suspicious components.

In systems that can only be described by two measurable quantities or potentials, possible error messages can represent the state of an interruption of the control circuit, short circuit to ground or short circuit to V _cc .

In the case of a large number of different potentials in the system, any error message due to a potential or current error caused by the individual star resistances at the measurable node must be described. The following prerequisites must be observed so that the faulty component can be located:

• - There is a reduction graph with node N _i whose potentials P _{i are} all known in the diagnostic cluster.
• - There are possible discrete values for the individual basic building blocks U _i .
• - For the localization of voltage errors all nodes have to potentials of the node to be analyzed and the corresponding one Knowing neighboring nodes.
• - For the localization of current errors, at least one current I _i through a basic module U _i , the potentials of the node to be analyzed and the corresponding neighboring nodes must be known in the node N _a to be analyzed.

Only one measurable node a is taken into account in the analysis in the head graph. The outer nodes i have a fixed potential for this consideration or it is assumed to be fixed. The initial situation is shown in Fig. 3.

The following applies to a potential deviation:

The measured potential should be explained by the discrete resistance values and form all possible combinations of the individual values from the header graph the candidate set. The result of all possible combinations setting node potentials determined. Is the hypothetically determined potential correct coincide with the measured potential, the existing in the combination form discrete values an explanation for the current observations.

The following applies to a current fault:

The calculation of the current error requires knowledge of the voltage potential at measuring node a and at outer node i. All measurable currents I _{aj are included} in the calculation. The arrows are all directed to node a.

Here, too, all combinations of the individual values of the header graph form the possible explanations for the current error. If the potential between an outer node i and the inner node a is the same, the equation provides no information about the corresponding star resistance value R _i .

If there is an undefined potential at the measuring node, you can directly without knowing the possible discrete values of the header graph be true.

In the event of measurement inaccuracies, the values determined for the fault localization are already assigned predefined values. A relatively high current is compared to a short circuit with I _i = ∞ for example in comparison to normal current. Measurement inaccuracies can be eliminated in this way.

Another possibility is the hypothetically calculated value of a Be to judge to what extent it corresponds to reality. Leave with it weight the individual results and prioritize them in further processing ren.

The result of the analysis in the header graph is a set of discrete values of the one individual elements that explain the observed symptom.

The analysis carried out in the header graph is now on the level of the basic building blocks pictured. To do this, superordinate discrete values must be in the reduction graph can be explained by subordinate values. Some subordinate discrete values can be specified directly. So with all graph elements, there is only one contain discrete value or several identical values, the subordinate value un already known depending on the higher value. The subordinate values can be determined directly from the superordinate substitute values without a subordinate value is available for the calculation. Generally considered the result is always ambiguous, since the reduction is an abstract tion of the system and a summary of several subordinate a higher value. For certain special cases arise despite this general procedure the subordinate values directly from the abstract onsoperation and the superordinate values.

For all graph elements in the lower level with alternative discrete values a conclusion in the reduction graph is necessary.

One way to do this is to take the values from the one below Calculate graphs for the next higher one and for agreement with the check known values.  

It is also possible, directly from the known higher values and under Be taking into account the type of reduction operation on the discrete values of the under to close ordered graph elements.

A search with targeted calculation assumes a known one Worth a systematic search and there can be conditions that differ through the Result in reduction operation, flow into the search. With the following Given the given conditions, impossible combinations of values are possible from the start exclude.

Conclusion about series reduction:

The following boundary conditions apply to the series reduction with the basic modules of the type resistors R _k and voltage sources Q _v :
for resistors: V = R
for voltage source with the same counting direction: V = U

Conclusion on parallel reduction:

The calculation of the individual causative values for the basic modules of the type resistors R _k and current sources Q _{i are} simplified as follows:
for resistors: R = 1 / V
for current sources with the same counting direction: QI = V

In contrast to known reduction processes, series and parallel processes take place reduction not only with two components, but can be any order exhibit. This results in advantages in terms of compact and effi more focused representation of the overall system.

In Fig. 4 the conclusion about a star reduction is shown.

In order to calculate the corresponding star values from the values of the polygon, the relationship between the two leg _resistances R _ν0 , R _µ0 and the equivalent resistance R _{νµ applies} :

R _νµ ≧ R _ν0 + R _µ0

The individual discrete values obtained from the elementary peripheral graphs can now still transfer to the corresponding states of the components will.

Due to the ambiguous inference process, contradictions in the discrete states occur. The component states can be divided into the three Categories "defective", "OK" and classify "possibly defective". Contradictory Statements following the conclusion between "O. K." and any one Error status results in the status "possibly defective".

There are a lot of possible explanations for each observation. The individual explanations can be prioritized based on a simple assumption. Assuming that the failure probability P of a component z. B. an exponential distribution of the shape

P (t) = 1 - e ^- λ ^t

and the failure probability of all components for all arbitrary times t _{x has} the same size, the probability of an n-fold error results from the following relationship:

on condition

This applies on the assumption that there are no probability dependencies between the components. From this it can be deduced that the higher the order of the error, the less likely it is to occur, or formally expressed:

P _n-1-fold (t) <P _n-fold (t)
with P _n-1-fold (t) ∈ [0.1]

The diagnostic result is output with a prioritization of the number of defective components per diagnostic result. With existing or during the The runtime of the overall system can determine the probability of failure of the diagnostic result even after a corresponding prioritization leads.

During the conclusion in the reduction graph or when mapping the slide Results on individual components, it is possible to find inconsistencies between the knowledge gained from the measurements and the model detect. For example, are the calculated ones in a subordinate graph does not contain discrete values, it is an inconsistency.

The cause of the inconsistency can be found on the one hand in the measurements carried out or occur in the diagnostic knowledge, the model. Error in the meas solutions caused by external influences or faulty measuring equipment cause incorrect measurement results, which in turn cause inconsistency.

With the described procedures, the complex overall becomes system taking into account the component structure and using ge own clustering procedure to the information necessary for the diagnosis compiled. The model generation is limited to a necessary one and thus limited number of components of the overall system, which leads to a considerable acceleration of the diagnostic process.

An inadequate model, i. H. a model with insufficient detail degree to the description of the error case, forms the second cause for the incon sistence. The second case is due to error models with different levels of detail degrees to eliminate. This means that the underlying failure model in the case an inconsistency is expanded by defining a new level of detail and the diagnostic process starts again. The expansion will be according to the on probabilities of the failure per component.  

The diagnostic device according to the invention contains a diagnostic module that follows the detailed procedure for every electrically controllable system made a diagnosis. The determined results of the individual system diagnoses are hierarchically structured, the functional relationships between the overall system structure taking into account the electrically controllable systems made available.

The hierarchical structure can be represented by a directional graph consisting of two subgraphs. The first subgraph describes the physical structure of the individual components. The second subgraph characterizes the graded relationships between the functions of the overall system and the individual components. Both graphs consist of nodes N _i and edges E _j .

The edges N _i represent the relations between the individual nodes. The nodes of the graph include components c _i , functions f _j and connection operations o _k . Each node has additional information that is stored in attributes.

The first subgraph contains the components installed in the system as nodes and is therefore referred to as a compositional hierarchy. To the attributes the node belongs to the information whether the respective component is a smallest represents interchangeable unit. An attribute also indicates whether a node has an over crowd forms. The excess quantities allow the individual components to be better closed structure. For example, the excess of automation devices all individual automation devices. The connection of the individual components with each other with a conjunctive link, because the individual Systems must be clearly assigned to an over component. The direction of the Edges refers to the supercomponent above it.

The second subgraph describes the function hierarchy. In this subgraph are the individual functions are listed as nodes. Every node that has a function represents, as an attribute, the level of output and that of criticality. The The output level is used to differentiate between error messages for different Liche user groups of the overall system, such. B. for the user or for the Service engineer.

In addition, the second subgraph contains connection operations. The connection operations consist of conjunctive and disjunctive connecting elements like from negations. To connect the individual nodes contains the hierarchical Function graph directed edges, each on the next parent Reference function. One or more connections can be made between two functions operation, which are the logical dependencies between the over and Represent sub-functions. Individual functions, each based on the normal relate to the function of the individual subsystems, each form the basis for several over-functions. Connection operas are therefore for multiple use included in the subgraph.  

To connect the two subgraphs, the overall graph contains edges in which the respective elementary basic components to the elementary subfunctions be formed.

The hierarchical structure includes that belonging to a faulty component smallest interchangeable unit associated with a given function or component can be averaged.

The information contained in this hierarchical system representation can can be used in different ways. One way is from a defective component, the functions affected in the ent to determine the speaking level. Another option is determined from components causing a defective function. A mixture of the two procedures, each in a certain number of defective components and failed functions is known and from this all unknown states the components and functions are determined is also conceivable.

To conclude in the hierarchical representation, dynamic data are necessary agile, which re the current status of the individual components and functions present. These dynamic data include the three states "error-free Component / function "," unknown state "and" faulty component / radio tion ".

To from known subordinate states by means of the connection operations The three states Relatio are used to infer superordinate states NEN, each with an initial value of a connection operation or put two input states in relation.

According to this scheme, a conclusion can also be drawn from a parent Level to a subordinate level.

To determine the smallest exchangeable unit of a component with the state "defective component / function" is the subgraph that contains the component in the direction of the edges until the search procedure is based on a com component that matches the attribute's smallest interchangeable unit and state contains "faulty component / function". The states of the parent com components result from relations.

If the states of the individual components are known, a search procedure is carried out first the states of the functions in the lowest level of the second subgraph. Then all functions are in this graph in the direction of the directed edges searched for the state "faulty component / function" with the previously fixed output level. The state of each element in the graph results from the established relations during the search process. The mitit The function nodes provide the information about the failed ones Functions with the corresponding criticality.  

If the failure of a function is known, the causing components can be be determined. Based on the function with the status "faulty compo element / function ", the method determines all subordinate function states to to the lowest level with the help of the established relations. The states of the Basic functions represent the component states and can get to the bottom components are mapped. They can be made small from the basic components determine the most exchangeable units. Only the functions and are included in the analysis Components of interest that have the status "faulty component / function" or "unknown condition". An occurring disjunctive connection operation creates an alternative path and thus an alternative set causing components.

For complex networked overall systems that consist of a multitude of components exist, the function and components are created as automatically as possible hierarchy expedient for economic reasons, especially if additional A large number of versions and variants of the systems occur. An analysis the underlying knowledge base shows that the functional hierarchy increases with it the degree of accuracy of an ever increasing specialization by the actual one Implementation is subject. Functions at a very high level of abstraction are hardly or not at all characterized by the implementation. Functions at the lowest level, on the other hand, have a very strong one individual components characterized by implementation dependency. The top part The functions can be fixed regarding the version and variants of a system are assumed, on the other hand, is the lower part of the functional hierarchy shaped by the individual implementation.

The component hierarchy can essentially be found in a library. The individual attributes of the components can be found in the component library be placed. Data from the development process chain or from production Documents provide the exact number of units installed in the overall system. The compo can also group the individual components library. The reason can also be found in the component library functions are stored.

The creation effort for the fixed part of the function hierarchy with the individual Attributes are a one-time process because there are only minor changes conditions, for example in the further development of the overall system. Therefore the fixed part of the function hierarchy also from an entire system library be taken.

The implementation-dependent part of the function hierarchy is filled with data material over the entire system, e.g. B. with data from the development process of the system or with construction drawings or circuit diagrams. This is what the occurring signal, energy and information flows as input data. Everyone one individual signal, energy or information flow contains a certain fixed function in the overall system. So z. B. a signal flow in the overall system of a vehicle  available with the task of displaying the speed in the display instrument. Enable all components with the corresponding component functions the speed display. The individual connection operations of the second Partial graphs result directly from the signal flow.

For electrical components in the periphery of automation devices also the normal states of the elements appearing in the head graph to determine the components belonging to a function. For this purpose, the electrical components present in the periphery are included their normal state according to the reduction process described above depicted in a header graph. For each input / output of the automation device devices are fixed functions from the specification of the automation device knows. These functions can be assigned to the individual elements in the header graph arrange that are at the corresponding input / output of the automation device. From the individual elements of the header graph with the corresponding functions the components involved must now be determined. Between the components there is always a conjunctive connection. This makes it possible to Functions of the individual electrical components present in the periphery automatically determined and finally entered in a function hierarchy.

Compared to known and z. B. in Wiedmann, Hans: "Object-oriented knowledge presentation for model-based diagnosis at production facilities ", Berlin, Springer-Verlag 1993 detailed functional system descriptions is the hierarchical structure in any logical with the present approach Dependencies can be represented. The connection of the function and component hierarchy chie takes place only on the lowest levels and thus simplifies the automati The creation and consistency of the two levels of observation, functions and Components, significantly.

The functions involved can be differentiated from the components Determine levels of abstraction. Likewise, from the faulty functions involved components emerge.

The overall system structure thus constructed can take into account the suspect components error statements about the entire system will. The described structure of the overall system structure also offers the Possibility of making statements about the impairment of functions of the whole systems to meet.

When diagnosing highly complex overall systems, it can be advantageous to find errors in localize cross-system networks. There is a parent for this necessary model based on a functional description of the effects. This allows the signal, energy and information flows that occur write. A Ver is used to localize errors between the subsystems drive, which on the in J. Sticklen et al., "Troubleshooting based on a Functional Device Representation: Diagnosing Faust in the External Active Thermal Control System of Space Station FRREDOM ", AAAI Workshop Reasoning About Function,  Washington D.C., July 11, 1993, pp.149-156, described the FR-Dx algorithm builds and develops it further.

With the functional description of effects, the real system becomes a more targeted one Function graph created in which the individual functions of the components with the system sizes to be observed are shown. The edges of the function graph reproduce the individual functions of the components and characterize the nodes system sizes. For edges and nodes there are binary statements about their Condition, normal condition or fault condition, possible.

For an adequate representation of the real system there are connecting elements between the individual information flows required. Should both effects with Nor painting behavior as well as effects in case of faulty behavior are shown In addition to conjunctive and disjunctive system dependencies. In addition to the fasteners, the function graph requires branches that provide the individual system sizes for various other dependencies can.

The individual Ab are obtained by means of a graph transfer process dependencies between system sizes and functions. Starting from each individual system size, the graph is against the directed edges run through and the individual relations are determined from this.

The overall system description takes place with a disjunctive normal form in which a system size through several disjunctively and conjunctively linked functions is shown.

The effect relationships can be represented in a three-dimensional function matrix, where in one level the columns are the system sizes and the rows represent the function of the components. The elements are within a level conjunctively linked and the connection between the individual levels happens disjunctive.

The creation of the function graph and the function matrix depends on the structure of the individual sources of information. In addition to the z. B. from the development process chain information that can be generated in the overall system can also include declarative relationships stored between the system sizes and the components as a model will be, because the provided description elements be capturing allow any dependencies between system sizes and components. In the Diagnosis can also include declarative relationships.

The functional impact descriptions can be easily created because the Method does not require exact mathematical relationships, only that Relations between the input and output variables recorded.

The method takes into account not only faulty functions but also System sizes that have no errors. As an input variable, additional also information about defective components from subsystem diagnostics be viewed and included in the processing.  

This extended FR-Dx procedure allows the display and diagnosis of com complex networked systems with different types of components, such as B. of electronic, mechanical and / or hydraulic components.

As with all model-based diagnostic procedures, in contrast to the dekla ratative procedure, with the extended FR-Dx procedure also multiple errors recognizable.

It is essential in this process that primarily not the defective components, the faulty functions make up the result of the diagnosis.

The diagnostic device advantageously contains a data memory so that the determined diagnostic results and other data relevant to the diagnosis can be saved.

A dialog control unit is advantageous which enables the dialog with the users of the whole systems controls. For example, if the overall system is one Vehicle, the driver or workshop personnel must be specified as the user. The dialogue is made easier for them if the error statements and the statements on Impairment of use are prepared audiovisually. For this you can on the slide appropriate means are provided.

A further relief for the user is if he doesn't just make error statements about the overall system, but if it is based on the existing system If you have knowledge of the overall system structure, additional remedial measures are available be beaten.

The structure and operation of the used in the diagnostic device Diagnostic modules not only allow the diagnostic device to be diagnosed purposes if an error message is received. The diagnostic facility can rather during the normal use of the overall system, in the example of the vehicle while driving, without impairing use action can be used.

Claims (14)

1. Diagnostic module for creating a diagnosis for electrically controllable systems arranged within an overall system, the diagnosis being based on models of the overall system, with the features:

• 1. Means for setting up a reduction graph with a head graph whose nodes have only fixed or measurable sizes, the means comprising: means for decomposing the electrically controllable systems into components which comprise one or more basic components, means for detecting the electrical connections between the Basic building blocks and between the components, means for assigning discrete electrical state values to the basic building blocks, means for defining the operating states and the possible component states of the individual components in relation to the electrical state values of the basic building blocks belonging to a component, means for specifying those for the Diagnostics necessary and measurable on the basic building blocks and components, means for determining basic building blocks that have no influence on the defined measurable sizes and means for summarizing and eliminating the determined unaffected basic building blocks;
• 2. means for recording the measurable quantities necessary for the diagnosis;
• 3. Means for interpreting the measurable quantities in the head graph;
• 4. Means for mapping the interpretation carried out in the header graph onto the level of the basic building blocks and
• 5. Means for mapping the interpretation mapped onto the basic building blocks at the level of the components, these means indicating suspect and / or defective components as the diagnostic result.

2. Diagnostic module according to claim 1, characterized in that for the diagnosis Different models of the overall system can be used and the models are in distinguish the accuracy of the overall system description. 3. Diagnostic module according to claim 1 or 2, characterized in that the model of the overall system can be taken directly from the design documents.   4. Diagnostic module according to one of claims 1 to 3, characterized in that in the diagnostic module has means for detecting an error message and an error message by means of interpreting the measurable quantities in the Head graph is interpretable. 5. Diagnostic module according to one of claims 1 to 4, characterized in that the suspect and / or defective components according to their probability of failure are probably sortable. 6. Diagnostic module according to claim 5, characterized in that in the calculation the probability of failure all up to the generation of error statements about the Whole system existing information about the whole system. 7. Diagnostic device for an overall system with electrically controllable systems arranged within the overall system, with the features:

• 1. Diagnostic module for creating a diagnosis based on system models for each electrically controllable system that can be broken down into components according to one of claims 1 to 6, wherein the diagnostic module indicates suspect and / or defective components as the diagnostic result;
• 2. Means for creating a diagnosis for the overall system based on the functional relationships between the individual electrically controllable systems, with the means being able to create a graph consisting of two subgraphs and the first subgraph representing the structure of the components and the second subgraph representing the relationships between represents the overall system functions and the functions of the individual components; and
• 3. Means for generating error statements about the overall system and statements about the impairment of use of overall system functions.

8. Diagnostic device according to claim 7, characterized in that in the Diagnostic device Means for creating diagnoses for subsystems are available are and with these means a function graph can be created that the function of Components as well as system sizes describing the overall system.   9. Diagnostic device according to claim 7 or 8, characterized in that in the Diagnostic device Means for controlling the dialogue with the users of the slide are available. 10. Diagnostic device according to one of claims 7 to 9, characterized in that that means in the diagnostic device for storing data, in particular of Diagnostic results and error statements are available. 11. Diagnostic device according to one of claims 7 to 10, characterized in that that the means for creating a diagnosis for the overall system are designed that information about system sizes that the normal functioning of the overall system describe information that indicates misconduct and information that which are stored in the means for storing data, can be recorded. 12. Diagnostic device according to one of claims 7 to 11, characterized in that the means generating the error statements provide further means to suggest of remedial action include and the proposals on the means of taxation of the dialogue are accessible to the users of the diagnostic device. 13. Diagnostic device according to one of claims 7 to 12, characterized in that that means for the audiovisual input and output of Information is available. 14. Diagnostic device according to one of claims 7 to 13, characterized in that the diagnostic device during normal operation of the overall system can be used and over the entire service life of the overall system continuous monitoring can be ensured.