DEVELOPMENT OF AN OVERALL MODEL

Description

[0001] The invention relates to a method for controlling and / or monitoring a compressor system comprising one or more compressors and one or more peripheral devices, wherein the compressors and peripheral devices are arranged or connected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control-monitoring unit, and to a compressor system according to claim 15.

[0002] Compressor systems consist of a multitude of compressors and peripheral devices of various types, which are interconnected via an air piping network and, when heat recovery systems are used, a water piping network. Compressor systems are generally designed individually for the specific on-site conditions. A universally applicable structure for compressor systems does not exist. Therefore, the behavior of a specific compressor system can only be analyzed and evaluated to a limited extent without knowledge of the structure of the compressor system.

[0003] So-called higher-level station controllers are used to control and diagnose the compressors and peripheral devices in a compressor system. Their functions are comparable to control systems in process engineering. The key difference from control systems is that control systems usually work with a control and analysis procedure that was specifically developed for the process to be controlled. In the special development, the knowledge about the structure of the process to be controlled and the interrelationships of the individual components of the process is encoded. In station controllers, however, control and analysis procedures specifically developed for the compressor system to be controlled and monitored are the exception. Normally, a standard control and analysis procedure is used, which is simply adapted to the on-site conditions through parameterization.

[0004] Patent document US 2009 / 292514 A1 describes a method and system for automatically generating simulations for a distributed control system. A programmed process model generator automatically integrates a variety of process model data from predefined model libraries into the descriptions of the process equipment, including the control devices, to create simulation models with varying degrees of accuracy.

[0005] Patent document DE 10 2011 079732 A1 relates to a method and a device for controlling a fluid conveyor for conveying a fluid within a fluid line, wherein the fluid can be, in particular, gas or oil, and the fluid conveyor can be a compressor or a pump. Patent document DE 10 2008 064491 A1 relates to a method for controlling or regulating a compressed air station with at least a plurality of interconnected compressors, in particular of different technical design, and optionally further compressed air technology devices, which, in particular in control cycles, can effect both switching strategies via an electronic system control for influencing a quantity of compressed fluid in the compressed air station that is available at any time for one or more users of the compressed air station,and the quantity of compressed fluid available at any time for one or more users of the compressed air station can be adapted to future operating conditions of the compressed air station in accordance with the quantity of compressed fluid withdrawn from the compressed air station, whereby, before initiating a switching strategy, various switching strategies are tested in a pre-simulation process based on a model of the compressed air station and the relatively most favorable switching strategy is selected from the tested switching strategies based on at least one specified quality criterion and the selected switching strategy is forwarded to the system control system for initiation in the compressed air station.

[0006] Patent document US 2003 / 097243 A1 relates to a method and system for operating a hydrocarbon production plant. Specifically, patent document US 2003 / 097243 A1 relates to the method and system for optimizing the operation of a hydrocarbon production plant using a computer-aided process simulator comprising a linear solver and a nonlinear solver system.

[0007] The object of the present invention, in contrast, is to provide a method for controlling and / or monitoring a compressor system that takes the specific conditions of a specific compressor system into account even more precisely. Furthermore, a compressor system is to be proposed that is improved with regard to the problems described above.

[0008] This object is achieved from a process engineering perspective with a method for controlling and / or monitoring a compressor system according to the features of claim 1 and from a device engineering perspective with a compressor system according to the features of claim 15. Furthermore, a control and monitoring unit for controlling and / or monitoring a compressor system according to the features of claim 19 is proposed.

[0009] A core idea of the present invention is that based on one or more initial models M 1 , M 2 , ... of the compressor system, which is based, for example, on a P&I diagram, possibly including the specialization of the compressors and the peripheral devices, one or more derived models M̃ a , M̃ b ,...which take into account the causal relationships between the individual compressors and peripheral devices and, where applicable, dynamic processes, and that for the evaluation of recorded operating data, one or more derived models M̃ a , M̃ b , ... the subsequent control, monitoring, diagnostic or evaluation routines are used as a basis.

[0010] In the context of the proposed method, a derived model can, for example, result from exactly one initial model. However, it is also possible for a derived model to result from two or more initial models, or for two or more derived models to result from a single initial model. Finally, it is possible for two or more derived models to result from two or more initial models.

[0011] In the derived model M̃ a , M̃ b ,...it can be a final model that is used directly in subsequent control, monitoring, diagnostic, or evaluation routines or serves as the basis for these control, monitoring, diagnostic, or evaluation routines; however, it can also be an intermediate model from which the final model is developed in one or more steps, which is ultimately used in subsequent control, monitoring, diagnostic, or evaluation routines or serves as the basis for these control, monitoring, diagnostic, or evaluation routines.

[0012] A control, monitoring, diagnostic or evaluation routine should generally be understood to mean different control tasks, monitoring tasks, diagnostic tasks or evaluation tasks.

[0013] Whenever it is stated that the compressors and peripheral devices are arranged or connected in a predetermined configuration, this should be understood to include multiple alternating states, for example, an alternative configuration achievable by switching a valve or switch. A predetermined configuration is therefore the set of all conceivable configurations that the compressor system can assume in different operating states.

[0014] The configuration should also be comprehensively understood in the form of a P&I diagram and thus capture the functional relationships of the compressors and peripheral devices from different perspectives or in different domains, whereby capturing the functional relationships in one domain, from one perspective, is of course sufficient for the implementation of the invention. A possible domain orPossible perspectives to consider include, for example but not exclusively, the compressed air-related interactions that can be represented in a P&I diagram in the narrower sense, in particular a compressed air P&I diagram, the heat recovery-related interactions that can be represented in a P&I diagram in the narrower sense, in particular a heat recovery P&I diagram, the cooling water circuit-related interactions that can be represented in a P&I diagram in the narrower sense, in particular a cooling water circuit P&I diagram, and the power supply-related interactions that can be represented in an electrical circuit diagram.Furthermore, a P&ID diagram within the meaning of the present invention can be abstracted and limited to the basic causal relationships viewed from a single perspective / domain and therefore does not have to include all the details of a potentially otherwise conventional P&ID diagram. Instead of the term P&ID diagram, a graphical representation of the causal relationships from a specific perspective / in a specific domain can also be understood, such as a graphical representation of the compressed air-related causal relationships or a graphical representation of the heat recovery-related causal relationships. In this respect, it is a flow chart that represents the flow of energy and / or operating resources and / or compressed air between the individual compressors and the individual peripheral devices.

[0015] In one possible embodiment, the derived model M̃ a , M̃ b ,... an aspect-specific model AM, which is created using an aspect-specific analysis algorithm from one or more initial models M 1 , M 2 , ... or from one or more intermediate models M 1 ', M 2 ', AAM 1 , AAM 2 . An aspect-specific model is to be understood as a model that illuminates the compressor system with regard to a specific question. Conceivable aspects that could give rise to an aspect-specific model include, but are not limited to: air humidity, pressure drop, compressed air quality, pressure grade, pressure behavior, energy efficiency, energy consumption, energy balance, temperatures, volume flows / mass flows, costs, reserve level and / or reliability.

[0016] In a preferred embodiment of the method according to the invention, the information about the aspect-specific behavior of a compressor or a peripheral device is contained in one or more component models KM to be taken into account in the aspect output model AAM and / or in the analysis algorithm itself.

[0017] The aspect-specific analysis algorithms encode how the output models are to be interpreted with respect to one or more aspects. Likewise, the aspect-specific analysis algorithms can include knowledge about the dynamic behavior of compressors and / or peripherals with respect to the aspect(s) under consideration. Optionally, the aspect-specific analysis algorithm uses aspect-specific component models to create the derived models.

[0018] In a preferred embodiment of the method according to the invention, an associated analysis algorithm is provided for each aspect-specific model AM to be created. An analysis algorithm is typically developed specifically for a problem that is to be answered with the aspect-specific model. For related problems, the same or a slightly modified analysis algorithm may be used. As already mentioned, however, each aspect-specific problem or each aspect-specific model is assigned a specific analysis algorithm.

[0019] Other derived models, especially intermediate models, can also be developed from analysis algorithms that, in individual cases, are the same or similar to the derived aspect-specific analysis algorithms.

[0020] In a preferred embodiment, it can be provided that the derived model M̃ a , M̃ b , ... - regardless of whether it is a final or intermediate model - is stored in the control / monitoring unit or is stored externally at the request of the control / monitoring unit.

[0021] In a possible further training it may be provided that the derived model(s) M̃ a , M̃ b , ... are derived and / or stored and / or applied in an external system. In this case, one or more activities related to the derived models M̃ a , M̃ b , ... externally and possibly not even under the direct or indirect control of the control / monitoring unit.

[0022] Specifically, the following process would be conceivable: 1. Operating parameters are recorded in the compressor system and stored in the control / monitoring unit. 2. These operating parameters are evaluated using appropriate models on an external server. This is not part of the control / monitoring unit or the compressor system. 3. The method step mentioned under 2. is also understood as a method according to the invention for controlling and / or monitoring a compressor system, even though it is not carried out within the compressor system itself and even though the control / monitoring unit is not involved in it. 4. The result of the above-mentioned monitoring / evaluation method can be, for example: a) The calculation of the next maintenance date b) Optimization of control or regulation parameters (for example the calculation of a reduced demand pressure) 5. With the methods mentioned under 4.Based on the results mentioned above, a) a compressor system manufacturer or operator can then carry out (automatic) maintenance planning to increase the availability of the compressor system through timely maintenance. b) The energy efficiency of compressed air generation can be improved by reconfiguring (manually or automatically) the compressor system control system to the calculated optimal demand pressure.

[0023] In a preferred embodiment, it can be provided that the derived model(s) M̃ a , M̃ b , ... be reviewed continuously, cyclically, or event-based and, if necessary, automatically adjusted. For example, an improved data basis may arise during operation of the compressor system, which can then be the reason for adjusting the derived model(s).

[0024] In a further optional embodiment, it can be provided that if the initial model(s) M 1 , M 2 , ... are changed or if one or more component models KM are changed, for example because structural changes have been made to the compressor system, the derived model(s) are also adapted.

[0025] In one possible embodiment, different domain-specific initial models can be considered for the creation of one or more derived models, whether as intermediate or final models. Domain-specific initial models can include, in particular, compressed air-specific initial models, power supply-specific initial models, cooling water circuit-related initial models, or heat recovery-related initial models, and thus two or more models from different domains can be combined or considered. Interactions between the different domains can also be captured and mapped.

[0026] In a possible embodiment of the present invention, an intermediate model is considered to be a derived model which, compared to the initial model(s), also takes into account a dynamic behavior or different operating states of the compressor system.

[0027] An alternative or additional intermediate model, which - if additional - can be hierarchically positioned before or after the previously discussed intermediate model, consists in being adapted to specific aspects and, if necessary, simplified compared to one or more initial models or compared to one or more intermediate models. For example, if the question of which compressor is linked to a downstream dryer in a compressor system is being investigated, filters arranged in between play no role and can be ignored in an intermediate model from the perspective of this question, allowing a simplified intermediate model to be defined.

[0028] In a preferred embodiment, it can be provided that the initial model(s) M 1 , M 2 is based on the concretely given configuration of the compressor system in the form of a P&I diagram, which is entered via an editor, in particular after the compressor system has been created.

[0029] Although the causal relationships in a P&ID diagram are illustrated graphically, it is not absolutely necessary to enter or process them graphically via the editor, although this represents a possible embodiment of the present invention. Rather, it is also possible to enter the causal relationships, in whole or in part, textually via the editor. Finally, other input forms are also conceivable, for example, verbal input or input via automatic image recognition. In this respect, the editor can also be viewed as an input interface.

[0030] The editor and / or the storage section can also be part of the control / monitoring unit or integrated there.

[0031] While it is generally preferred to enter the specific configuration in the form of a P&I diagram only after the compressor system has been created, it is also conceivable to make this entry before or during the creation of the compressor system, provided that the specific configuration or the specific causal relationship has already been determined.

[0032] A prerequisite for creating the initial models is knowledge of the causal relationships within the compressor system. It is not absolutely necessary for the compressor system to exist. One or more initial models, a P&ID, or multiple domain-specific P&IDs can be created during the planning phase of a compressor system. It is also not necessary for the initial models to be created immediately, or for the P&IDs to be entered immediately, once the structure of the compressor system has been established. The initial models or P&IDs must be defined or created at the latest when the P&IDs or initial models are to be used.

[0033] It goes without saying that the method according to the invention can also be carried out repeatedly, for example if something has changed in the specific configuration, i.e. in the actual structure of the compressor system, or if an error has occurred in the previous input of the P&I diagram.

[0034] The control / monitoring unit can be designed as a unit that performs both controlling and monitoring functions or only controlling or only monitoring functions.

[0035] A useful further development of the method provides that after the specific configuration has been entered, it is saved at the request of the control / monitoring unit.

[0036] In a possible, optional embodiment, the input of the P&ID diagram exclusively includes the configuration or interconnection of the compressors and peripheral devices already known to the control / monitoring unit. In this case, the control / monitoring unit may already be aware of the existence of a specific compressor or peripheral device as such, or the control / monitoring unit may already be aware of the specific specifications of the existing compressors and peripheral devices in addition to the mere existence of specific compressors or peripheral devices. In an alternative, also optional embodiment of the method, the input of the P&ID diagram includes not only the configuration or interconnection of the compressors and peripheral devices but also the specific specifications of the existing compressors and peripheral devices.

[0037] In a particularly preferred embodiment of the method, the P&ID diagram is entered by the user. Since users often combine compressors and / or peripheral devices from different manufacturers to create a specific compressor system, recording the causal relationships on the user side is a preferred possible embodiment.

[0038] In a possible refinement of the method according to the invention, it can be provided that, when the P&ID diagram is entered, the editor predefines or selects the compressors and peripheral devices as corresponding graphical symbols. Likewise, in a possible embodiment, it can be provided that, when the P&ID diagram is entered, the editor predefines or selects the connections between the compressors and peripheral devices as corresponding graphical symbols.

[0039] Furthermore, it can be provided that when entering the P&ID diagram, the editor can specify or select various possible specifications of the compressors and / or the peripheral devices.

[0040] In one possible embodiment, it can be provided that the presence of one or more compressors and one or more peripheral devices and / or the specification of individual or all compressors and / or individual or all peripheral devices is transmitted externally, in particular by uploading corresponding information, for example by uploading a file provided by the plant manufacturer.

[0041] A possible embodiment of the method further provides for individual or all compressors and / or peripheral devices to register themselves independently with the control / monitoring unit and, preferably, also transmit their specifications independently. This registration or transmission can be carried out via a cable connection, in particular via a wire or fiber optic connection, or via radio waves.

[0042] As an alternative to entering a P&ID diagram into the control / monitoring unit using an editor, knowledge about the causal relationships can also be entered into the control / monitoring unit in other ways: Import of a description / model into the control / monitoring unit. Automatic detection and, if necessary, recording of the specifications of individual or all compressors or individual or all peripheral devices of a compressor system. For example, it is conceivable that the control / monitoring unit queries the compressors and / or peripheral devices for their properties through appropriate communication. Alternatively, the compressors and / or peripheral devices report to the control / monitoring unit, preferably by providing their specifications. Finally, automatic detection of parts of the compressor system could also be considered, for example, through operating parameter analysis and / or image recognition systems. This can be directed towards the following partial detection: Detection of the interconnection of compressors and / or peripheral devices Detection of compressors and / or peripheral devices Tuning of model parameters

[0043] In one possible concrete embodiment of the method according to the invention, the selection of graphic symbols for representing compressors or peripheral devices and / or the selection of specific specifications and / or the selection of special links can be performed by marking them in a suggestion list, whereby the selected object or the selected information can then be transferred to the P&ID diagram to be created in the editor. However, this is a very concrete, possible embodiment. Numerous other variations are conceivable, particularly in connection with partial or exclusive textual input, input via voice recognition, or input via image recognition.

[0044] As already mentioned, the invention can provide for a causal relationship or a configuration in the form of a P&I diagram to be entered for only one domain / perspective. In one possible embodiment, respective P&I diagrams or respective configurations or causal relationships are entered for two or more different domains / perspectives, for example, a Compressed air P&I diagram and / or a heat recovery P&I diagram and / or a cooling circuit-related P&I diagram and / or a power supply-related functional relationship, in particular an electrical circuit diagram.

[0045] In general, it should be noted that the control / monitoring unit can be implemented in one or more servers that are in communication with each other or in a virtual computer.

[0046] If this approach is taken, ie an aspect-specific initial model is defined, an aspect-specific model can preferably be created using an aspect-specific analysis algorithm. In the selected example, the aspect-specific initial model AAM, in which, for example, the filters were left aside as peripheral devices not to be considered in the specific question, can be used to answer the question of which dryer can be supplied by which compressor using a suitable analysis algorithm.

[0047] In a possibly alternative or additional embodiment of the method according to the invention, a method for monitoring a compressor system comprising one or more compressors and one or more peripheral devices is proposed, wherein the compressors and peripheral devices are arranged or connected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control / monitoring unit, wherein the method creates a forecast for the next maintenance date of the compressor system or individual compressors or individual peripheral devices, namely according to claim 1, characterized in that that, in particular after the compressor system has been created, the concretely given configuration is entered in the form of a P&I diagram via an editor (23) and this input forms the basis for one or more initial models, that based on these initial models (M 1 , M 2 , ...) one or more derived models ( M̃ a , M̃ b , ...), the causal relationships between the individual compressors (11, 12, 13) and the peripheral devices (14 to 21), possibly also taking dynamic processes into account, and taking into account standardized operating data of the compressor system using the derived model(s) ( M̃ a , M̃ b , ...) a forecast for the next maintenance appointment is created.

[0048] In a possibly alternative or additional embodiment of the method according to the invention, a method for monitoring a compressor system comprising one or more compressors and one or more peripheral devices is further proposed, wherein the compressors and peripheral devices are arranged or connected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control / monitoring unit, wherein the method is designed as a diagnostic method for diagnosing the compressor system or individual compressors or individual peripheral devices, namely according to claim 1, characterized in that that, in particular after the compressor system has been created, the concretely given configuration is entered in the form of a P&I diagram via an editor (23) and this input forms the basis for one or more initial models, that based on these initial models (M 1 , M 2 , ...) one or more derived models ( M̃ a , M̃ b , ...), the causal relationships between the individual compressors (11, 12, 13) and the peripheral devices (14 to 21), possibly also taking dynamic processes into account, and taking into account standardized operating data of the compressor system using the derived model(s) ( M̃ a , M̃ b , ...) an error diagnosis is carried out.

[0049] According to the invention, a compressor system comprising one or more compressors and one or more peripheral devices and a control / monitoring unit is also claimed, wherein the compressors and peripheral devices are arranged or connected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control / monitoring unit, which is characterized in that the control / monitoring unit is designed and configured in such a way that it can be used for evaluating recorded operating data in control, monitoring, diagnostic or evaluation routines on one or more derived models M̃ a , M̃ b , ... of the compressor system, which is created based on one or more initial models M 1 , M 2 , ... of the compressor system, but also takes into account the causal relationships between the individual compressors and the peripheral devices as well as, if necessary, dynamic processes.

[0050] In a possible further development of the compressor system, it can be provided that the control / monitoring unit for storing the derived model(s) M̃ a , M̃ b , ...cares.

[0051] In a further possible embodiment, the control / monitoring unit can be designed such that it is implemented in one or more servers that are operatively connected to one another or in a virtual computer.

[0052] Finally, in a possible embodiment, it can also be provided that the control / monitoring unit also comprises an editor and the editor is designed and intended for entering the concretely given configuration of the compressor system in the form of a P&I diagram and the editor is furthermore operatively connected to the control / monitoring unit in such a way that the entered P&I diagram is transferred to the control / monitoring unit and there it serves as the starting basis for creating one or more output models M 1 , M 2 , ...

[0053] Finally, a control / monitoring unit for controlling and / or monitoring a compressor system comprising one or more compressors and one or more peripheral devices is proposed, wherein the compressors and peripheral devices are arranged or connected in a predetermined configuration and the control / monitoring unit effects the control and / or monitoring of the compressor system, characterized in that based on one or more output models M 1 , M 2 , ... of the compressor system, which is based, for example, on a P&I diagram, possibly including the specification of the compressors and the peripheral devices, one or more derived models are created which take into account the causal relationships between the individual compressors and peripheral devices and possibly also dynamic processes, and that for an evaluation of recorded operating data, the one or more derived models M̃ a , M̃ b ,... be used as a basis for subsequent control, monitoring, diagnostic or evaluation routines.

[0054] Preferably, the compressor system may also comprise an editor which is designed and intended for entering the specific configuration in the form of a P&I diagram and which is operatively connected to the control / monitoring unit in such a way that the entered P&I diagram is transmitted to the control / monitoring unit in order to form the basis for subsequent control, monitoring, diagnostic or evaluation routines.

[0055] In a specific further development, it can be provided that the control / monitoring unit is responsible for storing the entered P&ID diagram(s). Furthermore, in a possible embodiment, it can be provided that specifications of the compressors or peripheral devices used in the compressor system are stored in the control / monitoring unit in such a way that the input of the P&ID diagram exclusively covers the configuration or interconnection of the compressors and peripheral devices. Alternatively, however, it can also be provided, as already described with regard to the method, that when the P&ID diagram is entered, not only the configuration or interconnection of the compressors and peripheral devices is specified, but also initially the presence of one or more compressors and one or more peripheral devices and / or preferably also specifications of at least some of the compressors or peripheral devices are entered.

[0056] In one possible embodiment, the control / monitoring unit can store lists of specifications of possible compressors or peripheral devices and / or graphical symbols representing the compressors or peripheral devices used and / or graphical symbols representing possible connections in a memory section for selection by providing a P&ID diagram in the editor. The selection of the corresponding information or graphical symbols can be made using various input methods, such as cursor or mouse control, text control, voice control, etc. Similarly, the P&ID diagram can be created in the editor based on the selected symbols and / or information.

[0057] It goes without saying that the aspects, advantageous embodiments and advantages addressed with regard to the method according to the invention can also be transferred to the compressor system and the control / monitoring unit in a correspondingly adapted manner.

[0058] The invention will be explained in more detail below with regard to further features and advantages using various embodiments and with reference to the following figures. Figure 1 shows a purely exemplary configuration of a real compressor system that interacts with a control / monitoring unit according to the invention. Figure 2 shows a model of the compressor system according to Figure 1 according to the state of the art. Figure 3 shows an initial model that represents the compressor system in its actual configuration in the form of a P&I diagram. Figure 4 shows a P&I diagram according to Figure 3Model M' derived as a starting model, which takes into account different operating conditions. Figure 5a from the P&I diagram according to Figure 3 A model derived as a starting model, which can be regarded as a simplified aspect-specific starting model AAM. Figure 6 shows an aspect-specific model AM, which is derived from the aspect-specific starting model AAM according to Figure 5was developed. Figure 7 shows an overview of the possible paths for developing an aspect-specific model or an aspect-specific final model from an initial model. Figure 8 shows a schematic overview of the individual process steps and the resulting advantages when, on the one hand, the causal relationships between the compressors and peripheral devices of a compressor system are taken into account in the form of a P&I diagram as the basis for one or more initial models and, on the other hand, derived models are developed from these initial models. Figure 9 shows an example of an optimization routine in which the demand pressure p demand of a compressor system is optimized using models. Figure 10 shows an example illustrating how the effective buffer volume of a compressor system can be determined using a quantitative model of a compressor system. Figure 11 shows a model from the structure according to Figure 10derived aspect-specific output model AAM. Figure 12 Illustration of the change in the pressure gradient at the time of compressor switching.

[0059] In Figure 1An exemplary structure of a real compressor system is illustrated. The exemplary compressor system is to comprise three compressors 11, 12, 13 arranged in series. Each compressor 11, 12, 13 can be uniquely assigned a filter 14, 15, 16, which is arranged downstream of the associated compressor 11, 12, 13. Two dryers 19, 20 are arranged downstream of the filters 14, 15, 16. The compressed air downstream of the first filter 14 should always flow through the first dryer 19. The compressed air downstream of the second filter 15 can be directed via two valves 17, 18 either through the first dryer 19 or through the second dryer 20. The two valves 17, 18 are designed or controlled in such a way that they are never opened at the same time, i.e. when the first valve 17 is opened the second valve 18 remains closed or when the second valve 18 is opened the first valve 17 remains closed.

[0060] A compressed air reservoir 21 is arranged downstream of the two dryers 19, 20. A pressure sensor 26 is arranged downstream of the compressed air reservoir 21 to detect the operating pressure there.

[0061] To control and / or monitor the compressor system, a control / monitoring unit 22 is provided, which is operatively connected to the compressors 11, 12, 13, as well as the filters 14, 15, 16, the valves 17, 18, the dryers 19, 20, the compressed air reservoir 21, and the pressure sensor 26. The filters 14, 15, 16, the valves 17, 18, the dryers 19, 20, the compressed air reservoir 21, and the pressure sensor 26 form peripheral devices of the compressor system. The control / monitoring unit 22 is also operatively connected to a memory section 24 and an editor 23. The control / monitoring unit 22 can perform control functions, monitoring functions, or both control and monitoring functions. In this case, monitoring is understood to mean any form of evaluation, i.e. in addition to monitoring for malfunctions, unusual operating conditions, alarm situations, etc.also a diagnosis, especially if an error message is already present, an evaluation with a view to optimization or an evaluation to forecast the next maintenance appointment (predictive maintenance).

[0062] In order to fulfill one or more of these functions, models of the functional relationships of the compressors 11, 12, 13 and the peripheral devices 14 to 21 must be recorded and made accessible to the control / monitoring unit 22. For this purpose, on the one hand, the concretely given configuration is preferably entered in the form of a P&I diagram via an editor 23, which is operatively connected to the control / monitoring unit 22. On the other hand, one or more output models M 1 , M 2 , ... are created by the control / monitoring unit 22 or externally from this input configuration, which encodes the functional relationship between the compressors 11, 12, 13 and the peripheral devices 14 to 21, and one or more derived models based on these output models M 1 , M 2 , ... M̃ a , M̃ b , ...which take into account the causal relationships between the individual compressors 11, 12, 13 and peripheral devices 14 to 21.

[0063] In contrast to this, the state of the art, as shown by Figure 2 As illustrated, the interaction between the individual compressors 11, 12, 13 and the peripheral devices 14 to 21, specifically in the present example the filters 14, 15, 16, the dryers 19, 20 as well as the valves 17, 18 and the compressed air reservoir 21, is not taken into account. In methods for controlling and / or monitoring a compressor system according to the state of the art, the peripheral devices of the compressed air treatment and the interconnection of these peripheral devices are, from a structural perspective, nothing more than a black box, which in this respect does not accurately represent the compressor system.

[0064] The present invention therefore provides that in order to create a model of the compressor system, in particular after the compressor system has been created, the concretely given configuration is entered in the form of a P&I diagram via the aforementioned editor 23. A model for the compressor system according to Figure 1 representative P&I diagram is in Figure 3reproduced. It should be noted that many variants are conceivable for the input of this P&I diagram. The input of the P&I diagram should at least include the step of determining the functional relationships between the compressors 11, 12, 13 and the peripheral devices 14 to 21, but can preferably also include a preliminary step, namely the recording of the presence of the compressors 11, 12, 13 and the individual peripheral devices 14 to 21, and preferably also a further third step, namely the input of the specifications of the compressors 11, 12, 13 and the peripheral devices 14 to 21. It is conceivable that the recording of the presence, the functional relationships and the specifications takes place in different ways, for example the functional relationships are entered graphically via the editor 23, the other information of the control / monitoring unit 22 is transmitted in another way, for example by uploading afile provided by the plant manufacturer. A wide variety of input options are also conceivable for input via Editor 23, as already explained in the introductory section of the description.

[0065] In Figure 4 A derived model M' is illustrated, which already represents different possible operating states of the initial model given as a P&I scheme according to Figure 3 While the model for Figure 3 the information about which dryer 19, 20 is supplied by which compressor 11, 12, 13 is not yet included, takes into account the Figure 4 Model M' illustrated the different circumstances by making the case distinction, "if first valve 17 and second valve 18 are closed, then ..., otherwise...". V1 designates the first valve 17, V2 the second valve 18. T1 designates the first dryer 19, T2 the second dryer 20, K1 designates the first compressor 11, K2 the second compressor 12 and K13 the third compressor 13.

[0066] In Figure 5 is a P&I diagram according to Figure 3A model derived as a starting model is illustrated, which can be viewed as a simplified aspect-specific starting model AAM. If, for example, within the compressed air domain, the aspect of which dryer 19, 20 can be supplied by which compressor 11, 12, 13 is examined in order to develop an aspect-specific model AM that illuminates this question, an associated analysis algorithm and, if necessary, component models are used. The associated analysis algorithm can be stored in a corresponding location, in particular within the access area of the control and monitoring unit 22. Aspect-specific component models can, for example, be integrated into a database.

[0067] In the present example, the analysis algorithm knows that for the question of which dryer 19, 20 is supplied by which compressor 11, 12, 13, the output model according to Figure 3The remaining filters 14, 15, 16 can be disregarded. This results in the question being Figure 5 presented aspect-specific initial model AAM, which compared to the initial model according to Figure 3 is simplified in that filters 14, 15, 16 are disregarded.

[0068] In Figure 6 The aspect-specific model AM is now shown, which is derived from the aspect-specific initial model according to Figure 5 for the question of which dryer 19, 20 is supplied by which compressor 11, 12, 13. In this respect, the aspect-specific model was Figure 6 the aspect-specific initial model AAM was used, which can be regarded as an intermediate model and is based on the initial model M according to Figure 3 is based on.

[0069] In Figure 7various paths for developing an aspect-specific model AM (or an aspect-specific final model) from an initial model M are presented. In this respect, it is conceivable to develop one or more aspect-specific models AM directly from one or more initial models M. However, it is also conceivable to develop one or more aspect-specific models AM via intermediate models from one or more initial models M. Either models M' or models AAM can be considered as intermediate models, whereby the models M' generalize for a domain / perspective and define one (or possibly several) models that, for example, define different operating states of one or more initial models. Aspect-specific initial models AAM form the basis for arriving at an aspect-specific model AM with the aid of an analysis algorithm and / or taking component models into account.In this respect, it is conceivable to reach the aspect-specific model AM via the path M, M', AAM. Alternatively, it is also possible to proceed from the initial model M via AAM and then M' to the initial model AM. Finally, it is also possible to proceed from the initial model M via M' to the aspect-specific model AM. The example in . Figure 6 has shown that it is also possible to proceed from the initial model M via AAM to the aspect-specific model AM. All models M', AAM, AM are derived models M , which are developed from an initial model M. However, the aspect-specific model AM can be regarded as a final model that is suitable for answering specific questions and can therefore be taken into account in subsequent control, monitoring, diagnostic or evaluation routines.

[0070] In Figure 8A schematic overview of the individual process steps and the resulting advantages is illustrated if, on the one hand, the causal relationships between the compressors and peripheral devices of a compressor system are taken into account in the form of a P&I diagram as the basis for one or more initial models and, on the other hand, derived models are developed from these initial models.

[0071] In this respect, Figure 8 illustrates the interrelationships between the individual applications of models. The causal relationships depicted in the figure can be interpreted both cumulatively and alternatively. The basis for data processing is data standardization, meaning that each individual piece of data is assigned a well-defined meaning.

[0072] Based on standardized data and the domain-specific initial models M 1 , M 2 , ..., aspect-specific models AM can be derived, which can then be used for numerous other applications. Concrete applications include, for example, open-loop and closed-loop control, optimization of the boundary conditions under which a specific compressor system is operated, data analysis, monitoring, diagnostics, and predictive maintenance. With regard to the application of open-loop and closed-loop control, it can be stated that the operation of a compressor system can be improved by using models to determine and implement control actions for the compressors and / or peripheral devices under given boundary conditions (e.g. demand pressure to be maintained) of the compressor system. This is an optimization that is applied in real time (online application).

[0073] For the control or regulation of a compressor system, the following examples are given of how an improvement in the operating process of a compressor system can be achieved in a specific aspect using models and, if necessary, derived models: a) Energy efficiency aspect: For example, by considering the causal relationships of the compressor system, i.e., the corresponding P&ID diagram, and possibly considering other derived models, it can be achieved that dryers in a compressor system are only operated when there is a need to dry compressed air. During periods when there is no need to dry compressed air, dryers are not operated, thus saving energy for covering "thermal leaks." b) Compressed air quality aspect: With knowledge of the causal relationships between the compressors and peripheral devices of a compressor system, the response to the failure of a dryer can be regulated as follows: If a dryer fails, the compressors assigned to the dryer are only operated when the delivery volume of the other compressors is insufficient to cover the compressed air demand.If the compressed air piping allows, the compressed air from the compressors assigned to the failed dryer is distributed to other dryers.

[0074] As an example of data analysis using models within the meaning of the present invention, aspect-specific models can be created. In principle, quantitative or qualitative statements can be considered for most conceivable applications, not only in data analysis, but also in monitoring, diagnostics, etc. For the aspect of the compressor system's reliability, for example, a quantitative statement can be made in the sense of a mean-time-to-failure quantification, for example, 10,000 hours. However, a statement illuminating the compressor system's reliability can also be made qualitatively, for example as follows: The compressor system's reliability is rated "high," "medium," or "low."

[0075] An example of optimization could be the determination of the demand pressure parameter. This optimization can be performed both offline and during ongoing operation of the compressor system. In this regard, reference is made to the illustration in Figure 9 which makes the problem of optimising the demand pressure to the actually necessary pressure more understandable.

[0076] The key requirement for a control / monitoring unit in a compressor system is the minimum pressure (demand pressure) that must be present at the point of transfer to the customer network. The control / monitoring unit 22 attempts to control the compressors 11, 12, 13 in such a way that the demand pressure (p demand ) is always maintained and the electrical energy required to generate compressed air is minimized. A sudden increase in compressed air consumption can result in the control / monitoring unit issuing a switch-on command to a compressor 11, 12, 13 too late, resulting in the pressure falling below the required pressure. For this reason, the demand pressure (p demand ) in the control / monitoring unit is always set somewhat higher than the pressure that the customer actually needs (p necessary ). The difference between the set demand pressure p demand and the actually required pressure p necessary is a safety buffer.However, the higher demand pressure p demand increases the energy required to generate compressed air, as the electrical power consumption of compressors 11, 12, 13 increases with the demand pressure p demand. It is therefore desirable to set the demand pressure as low as possible, but still high enough so that the pressure actually required is not undercut in the event of sudden surges in consumption (see . Figure 9 ).

[0077] However, simply analyzing the historically recorded demand pressure (p demand) over time is not sufficient to optimize the demand pressure (p demand) because changes in the demand pressure (p demand) affect the behavior of the compressor system's control system. This results in different switching actions, which result in a different pressure curve.

[0078] By applying a simulation model of the compressor system, derived from the P&I diagram of the compressor system, the minimum demand pressure can be determined based on historically recorded pressure profiles, at which the actual required pressure will no longer be exceeded. Using such a simulation model (further explanations can be found, for example, in WO 2010 / 072803 A1), it is also possible to determine how much energy can be saved by optimizing the demand pressure p demand.

[0079] Models can also be used to monitor compressor systems. By comparing the behavior of the real process with the model of the real process, it is possible to detect behavior in the real process that was unexpected (at least based on the model). If reality and the model diverge, a warning or fault is triggered, for example, an alarm signal is triggered or an alert email is sent to the plant manager.

[0080] In the area of diagnostics, models can be used to narrow down or determine the cause of a malfunction. For example, various error scenarios are simulated on the model and compared with the data observed in the real process when the error occurred. The scenario that best matches reality provides an indication of the cause of the error.

[0081] Using predictive simulations, it is possible to estimate the next maintenance date for the compressor system(s) or peripheral devices. Assuming a compressed air flow profile (e.g., observed in the past in the affected compressor system), it is determined how the individual compressors and peripheral devices of a compressor system are expected to be operated and stressed in the coming weeks or months. Based on the operating status history of the compressors and peripheral devices and a model for the wear of the maintenance-relevant components / fluids, the date at which the wear limit (service life) of the component / fluid is reached can be determined.

[0082] The following is another example of how the effective buffer volume of a compressor system can be determined using a quantitative model of a compressor system. This is illustrated using the compressor system as shown in Figure 10 The compressor system consists of the three compressors 11, 12, 13, the two dryers 19, 20, the two filters 14, 15 and the compressed air reservoir 21. The task of the control / monitoring unit 22 is to communicate the effective buffer volume of the compressed air reservoir 21 (volume possibly known) together with the piping network (volume usually unknown). The information about the effective buffer volume is used, for example, to calculate the current compressed air consumption, from which the times for the switching operations of compressors are derived. To determine the effective buffer volume, the compressor system is Figure 10initially represented in a model that only contains the components relevant for determining the buffer volume. This results in a simplified AAM model with regard to the buffer volume, as described in Figure 11 Dryers 19, 20 and filters 14, 15 are not relevant for calculating the effective buffer volume and are therefore not taken into account in the aspect-specific output model AAM. Compressors 11, 12, 13, however, are relevant because the switching operations on compressors are used to determine the effective buffer volume via the change in the gradient of the pressure sensor 26 attached to the compressed air reservoir 21. The buffer volume is calculated by comparing the pressure gradient before the switching operation with the pressure gradient after the switching operation. The switching operation of a compressor leads - with constant compressed air consumption - to a change in the pressure gradient, as can be seen from Figure 12The calculation is based on several assumptions: The "real" effective buffer volume does not change around the time of the switching operation. Compressed air consumption remains constant around the time of the switching operation. The temperature of the compressed air in the piping system and in the buffer tank remains constant.

[0083] Assuming that the delivery volume of the switching compressor is known, e.g. because the delivery volume is stored in a database accessible to the control and monitoring unit 22, the effective buffer volume V eff can be calculated from the change in the pressure gradient and the change in the delivery volume of the compressors ΔFAD (corresponds to the delivery volume of the individual compressor) and the ambient pressure p amb: V eff = Δ FAD dp N dt 2 − dp N dt 2 ∗ p amb

[0084] In particular, the assumption that compressed air consumption remains constant around the time of the switching operation will not be maintained in practice for every switching operation. Therefore, it is possible and advantageous to offset the individual estimate of the effective buffer volume against previous estimates using filtering (e.g., averaging). For further processing, the average of the last 20 estimates is then used. Under the very realistic assumption that the change in consumption during the switching process is equally likely to increase or decrease, the average consumption changes during filtering will cancel each other out.

[0085] An example of analyzing the behavior of a compressor system is analyzing the reserve level: A measure of the reliability of a compressor system is the reserve level. The reserve level is primarily determined by whether the compressed air consumption for the time interval under consideration exceeded the available delivery quantity (taking into account the compressed air stored in the compressed air reservoir) or would have exceeded it if a compressor had failed. For this analysis routine, it is useful to use the configuration of the compressor system entered as a P&ID diagram and models derived from it. The calculation of the reserve level can be viewed as both an analysis and a monitoring exercise.If the reserve level analysis is used for monitoring purposes, the compressor system operator can react and shut down compressed air consumers before the minimum required pressure is reached or equip the compressor system with additional compressors. The degree of overload could be determined by the failure of a compressor: Worst degree of overload: Did pressure dropouts occur during the period under review even though all compressors were available to generate compressed air? Severe degree of overload: Would pressure dropouts have occurred if the smallest compressor had failed? Moderate degree of overload: Would pressure dropouts have occurred if a medium-sized compressor had failed? Mild degree of overload: Would pressure dropouts have occurred if a large compressor had failed?

[0086] The analysis of whether, for example, a severe degree of overload occurred during the period under consideration is carried out by simulating the compressed air station using the derived models, with the specification that the smallest compressor may not be used for compressed air supply. The simulation itself can be carried out, for example, as described in WO 2010 / 072803 A1.

[0087] By applying derived models of the behavior of the components of a compressor system, a forecast can be made for the next maintenance date for the compressor system, for individual compressors or individual peripheral devices. Derived models can be used for this purpose for predictive simulation. Using a derived model for predictive simulation, it is possible to predict how the individual compressors or peripheral devices of the compressor system will behave over time for a given compressed air consumption profile. To carry out a predictive simulation, the switching commands to the compressors are calculated within the framework of model accuracy and taking into account the control algorithm in the control / monitoring device, as they would result in the real compressor system for the given compressed air consumption profile. The running behavior of the compressors can be derived from the switching commands to the compressors.The running behavior of the compressors describes in which operating state a compressor is at a given time.

[0088] If a model exists for the maintenance-relevant components or operating materials that allows the wear condition of the maintenance-relevant component / operating material to be deduced from the operating condition of the compressor system, the results of the advance simulation can be used to determine when a wear condition will be reached that requires maintenance. For compressors, the wear condition of a maintenance-relevant component / operating material is currently determined based on the running hours of a compressor. For example, an oil change is necessary every 3,000 operating hours. In the future, it will be possible to determine the wear condition of a maintenance-relevant component / operating material not only based on the operating hours, but also on the ambient / operating conditions of the compressor system. If the models for the advance simulation take into account the effects relevant to the determination of the wear condition (e.g.If the parameters (e.g. compression temperature, pressure in the oil separator tank, particle load of the intake air, ambient temperature) are mapped with sufficient accuracy, forecasts for maintenance measures can also be made if the wear condition of the affected component / operating fluid cannot be determined solely from the running hours.

[0089] The accuracy of the forecast for the date of the next maintenance measure naturally depends on the extent to which the compressed air consumption trend assumed in the forecast simulation also occurs in reality.

[0090] An advantage of predicting the next maintenance date based on advance simulations compared to the trivial method of extrapolating running hours is that a forecast is possible even if the composition of the compressed air station changes (e.g. adding or removing a compressor) or if the compressed air control system is re-parameterized (e.g. changing the switching on and off sequence of compressors).

[0091] The forecast for the next maintenance task is repeated regularly (e.g., once a day), with the compressor condition history observed in the current compressor system since the last forecast being used to create a new forecast. This makes the forecast for the next maintenance date increasingly accurate over time, as the proportion of actually observed wear is incorporated into the forecast, thus reducing the proportion of wear still to occur until the next maintenance task (as an uncertainty in the model).

[0092] The method according to the invention is characterized in that, starting from the acquisition of operating data to the evaluation of the operating data for the purpose of Control and / or regulation Monitoring Diagnosis Optimization Forecast of a maintenance date (predictive maintenance) individual procedural steps must be carried out. The individual procedural steps can be defined as follows. The causal relationships in the compressor system to be analyzed must be defined and, if necessary, entered. The compressor system's operating data must be standardized in an appropriate form. Based on the compressor system's causal relationships, one or more initial models and derived models are created. Aspect-specific models of the compressor system are used to answer specific questions.

[0093] The four process steps described are decoupled from each other both spatially and temporally. There is only a temporal connection (before-after connection) between the process steps, meaning that some process steps must be executed before other process steps, and their results must be made available before other process steps that use the results from previous process steps can run. However, the process steps can be distributed across different systems (but do not have to be). If the process steps run on different systems, the possibility for information exchange (at least unidirectional) must be provided.

[0094] Although the invention has been described using a compressor system, i.e. for overpressure, all principles can also be applied to a vacuum system in which pumps work together instead of compressors.

[0095] Furthermore, compressors were referred to generally in this context without specifying the specific type of compressor. In one embodiment, all compressors can be designed as positive displacement compressors, for example, although this is only to be considered a specific embodiment and is not generally mandatory. List of reference symbols

[0096] 11, 12, 13Compressors 14, 15, 16Filters 17, 18Valves 19, 20Dryers 21Compressed air reservoir 22Control / monitoring unit 23Editor 24Storage section 25Data acquisition unit 26Pressure sensor

Claims

1. Method for controlling and / or monitoring a compressor system comprising one or more compressors (11, 12, 13) and one or more peripheral devices (14 to 21), wherein the compressors (11, 12, 13) and peripheral devices (14 to 21) are arranged or interconnected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control / monitoring unit (22), wherein, based on one or more base models (M1, M2, ....) of the compressor system, which is based, for example, on a P&I diagram, optionally including the specification of the compressors (11, 12, 13) and the peripheral devices (14 to 21), one or more derived models (M̃a, M̃b, ...), which take into account the operative relationships between the individual compressors (11, 12, 13) and peripheral devices (14 to 21) and optionally also dynamic processes, are created, characterized in that the one or more derived models (M̃a, M̃b, ...) are used as a basis in subsequent control, monitoring, diagnostic or evaluation routines for an evaluation of recorded operating data.

2. Method according to claim 1, characterized in that the derived model (M̃a, M̃b, ...) is an aspect-specific model (AM) which is created using an aspect-specific analysis algorithm from the one or more base models (M1, M2, ...) or from one or more intermediate models (M1', M2', ..., AAM1, AAM2, ...).

3. Method according to claim 2 or 3, characterized in that the information about the aspect-specific behavior of a compressor (11, 12, 13) or a peripheral device (14 to 21) is contained in one or more component models (KM) to be taken into account in the aspect base model (AAM) and / or in the analysis algorithm itself.

4. Method according to claim 2 or 3, characterized in that an associated analysis algorithm is provided for each aspect-specific model AM to be created.

5. Method according to one of claims 1 to 4, characterized in that the derived model (M̃a, M̃b, ...) is stored in the central control / monitoring unit (22).

6. Method according to one of claims 1 to 5, characterized in that the derived model or models (M̃a, M̃b, ...) can be derived and / or stored and / or applied in an external system.

7. Method according to one of claims 1 to 6, characterized in that the derived model or models (M̃a, M̃b, ...) are checked continuously or cyclically or event-based and, optionally, automatically adjusted.

8. Method according to one of claims 1 to 7, characterized in that when the base model or models (M1, M2, ...) are changed and / or when a component model (KM) is changed, for example because structural changes have been made to the compressor system, the derived model or models are also adapted.

9. Method according to one of claims 1 to 8, characterized in that different domain-specific base models (M1, M2, ...), in particular one or more compressed air-specific base models, one or more power supply-specific base models, one or more base models related to the cooling water circuit or one or more base models related to the heat recovery are taken into account.

10. Method according to one of claims 1 to 9, characterized in that the derived model (M̃a, M̃b, ...) is an intermediate model (M1', M2', ...) which, compared with the base model or models, also takes into account a dynamic behavior or different operating states of the compressor system.

11. Method according to one of claims 1 to 10, characterized in that the derived model (M̃a, M̃b, ...) is an intermediate model (AAM1, AAM2, ...) which is adapted and optionally simplified with respect to one or more base models (M1, M2, ...) or one or more other intermediate models (M1', M2',...) in an aspect-specific manner.

12. Method according to one of claims 1 to 11, characterized in that the base model or models (M1, M2, ...) is or are based on the specifically given configuration of the compressor system in the form of a P&I diagram, which is entered via an editor (23) in particular after the compressor system has been created.

13. Method for monitoring a compressor system comprising one or more compressors (11, 12, 13) and one or more peripheral devices (14 to 21), wherein the compressors (11, 12, 13) and peripheral devices (14 to 21) are arranged or interconnected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control / monitoring unit (22), characterized in - that the method generates a forecast for the next maintenance date of the compressor system or individual compressors or individual peripheral devices, specifically according to claim 1, wherein, in particular after the compressor system has been created, the specifically given configuration in the form of a P&I diagram is entered via an editor (23) and this input forms the basis for one or more base models, - that based on these base models (M1, M2, ...) one or more derived models (M̃a, M̃b, ...), which take into account the operative relationships between the individual compressors (11, 12, 13) and the peripheral devices (14 to 21), optionally including dynamic processes, are created, and - a forecast for the next maintenance date is created using the derived model(s) (M̃a, M̃b, ...), taking into account standardized operating data of the compressor unit.

14. Method for monitoring a compressor system comprising one or more compressors (11, 12, 13) and one or more peripheral devices (14 to 21), wherein the compressors (11, 12, 13) and peripheral devices (14 to 21) are arranged or interconnected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control / monitoring unit (22), characterized in - that the method is designed as a diagnostic method for diagnosing the compressor system or individual compressors or individual peripheral devices, specifically according to claim 1, wherein, in particular after the compressor system has been created, the specifically given configuration is entered in the form of a P&I diagram via an editor (23) and this input forms the basis for one or more base models, - that based on these base models (M1, M2, ....), one or more derived models (M̃a, M̃b, ...), which take into account the operative relationships between the individual compressors (11, 12, 13) and the peripheral devices (14 to 21), optionally including dynamic processes, are created and - a fault diagnosis is carried out using the derived model(s) (M̃a, M̃b, ...), taking into account standardized operating data of the compressor unit.

15. Compressor system comprising one or more compressors (11, 12, 13) and one or more peripheral devices (14 to 21) and a control / monitoring unit (22), wherein the compressors (11, 12, 13) and peripheral devices (14 to 21) are arranged or interconnected in a predetermined configuration, wherein the compressor system is controlled and / or monitored via a control / monitoring unit (22), wherein one or more derived models (M̃1, M̃2, ...) are created based on one or more base models (M1, M2, ...) of the compressor system, but also taking into account operative relationships between the individual compressors (11, 12, 13) and the peripheral devices (14 to 21) and optionally including dynamic processes, characterized in that the control / monitoring unit (22) is designed and set up in such a way that it accesses the one or more derived models (M̃1, M̃2, ...) of the compressor system for an evaluation of recorded operating data in control, monitoring, diagnostic or evaluation routines.

16. Compressor system according to claim 15, characterized in that the control / monitoring unit ensures the storing of the derived model or models (M̃1, M̃2, ...).

17. Compressor system according to claim 15 or 16, characterized in that the control / monitoring unit (22) is implemented in whole or in part in one or more operatively interconnected servers or in one or more virtual computers.

18. Compressor system according to one of claims 15 to 17, which in addition to the control / monitoring unit also comprises an editor (23), characterized in that the editor (23) is designed and intended for entering the specifically given configuration of the compressor system in the form of a P&I diagram and the editor (23) is operatively connected to the control / monitoring unit (22) in such a way that the entered P&I diagram is transmitted to the control / monitoring unit (22) where it serves as the starting basis for creating one or more base models (M1, M2, ...).

19. Control / monitoring unit for controlling and / or monitoring a compressor system comprising one or more compressors (11, 12, 13) and one or more peripheral devices (14 to 21), wherein the compressors (11, 12, 13) and peripheral devices (14 to 21) are arranged or interconnected in a predetermined configuration and the control / monitoring unit effects the control and / or monitoring of the compressor system, wherein, based on one or more base models (M1, M2, ...) of the compressor system, which is based, for example, on a P&I diagram, optionally including the specification of the compressors (11, 12, 13) and the peripheral devices (14 to 21), one or more derived models (M̃a, M̃b, ...), which takes / take into account the operative relationships between the individual compressors (11, 12, 13) and peripheral devices (14 to 21) and, optionally including dynamic processes, is / are created, characterized in that the one or more derived models (M̃a, M̃b, ...) are used as a basis in subsequent control, monitoring, diagnostic or evaluation routines for an evaluation of recorded operating data.