Method for producing an analysis device by means of a sliding window simulation
Patent Information
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- SIEMENS AG
- Filing Date
- 2024-09-06
- Publication Date
- 2026-04-29
Smart Images
Figure EP2024074903_13032025_PF_FP_ABST
Abstract
Description
[0001] Description
[0002] Manufacturing method for an analysis device using a
[0003] Sliding window simulation
[0004] The invention relates to a method for producing an analysis device and to a computer program product by means of which the manufacturing method can be partially carried out. Furthermore, the invention relates to a digital twin of an analysis device, based on which the claimed method can be implemented. Likewise, the invention relates to an analysis device produced by the claimed method and to the use of a sliding window solver.
[0005] Patent specification KR 102511137 Bl discloses a method for the similarity analysis of chromatogram images reflecting retention times of components of a gas sample. A matrix of time-series data is created and evaluated using a sliding-window algorithm.
[0006] Patent application CN 103657152 A discloses a simulated moving-bed separation system comprising connected chromatographic columns. The moving-bed separation system is designed to separate three components of a sample with increased purity.
[0007] A large number of applications require customized analysis devices that are designed to be application-specific. Such an application-specific design is time-consuming, complex, and error-prone. In particular, optimizing such a design is difficult and complex. There is a need for a method that allows rapid, reliable, and precise design and manufacture of analysis devices. The invention is based on the object of providing a possibility that offers an improvement in at least one of the aspects outlined. This object is achieved by a method according to the invention for producing an analysis device. The analysis device can be designed, in particular, as a gas analysis device. The analysis device to be produced has at least one separation device, for example a chromatographic separation column, and a plurality of pneumatic modules.The pneumatic modules can be designed, for example, as pressure regulators, throttles or valves. A pneumatic module fluid-dynamically influences a fluid flow in the analysis device to be produced. In a first step of the method, a pneumatically coupled structure is provided which comprises virtual representations of the separation device and the pneumatic modules. The pneumatic coupling of the virtual representations in the provided structure makes it possible to simulate a chained flow through the separation device and the pneumatic modules in the form of their virtual representations. The pneumatically coupled structure can be designed as a data set which specifies how the at least one separation column is to be coupled to the pneumatic modules in the analysis device to be produced.Furthermore, the pneumatically coupled structure can include virtual representations of connecting lines and their fluid dynamic parameters, via which the separating device and the pneumatic modules are to be connected. The pneumatically coupled structure can be provided, for example, by a user who performs the claimed method.
[0008] In addition, the method comprises a second step in which a composition of a fluid flow is specified, which is to be determined using the analysis device to be manufactured. Likewise, in the second step, an operating parameter for the pneumatically coupled structure is specified, for example an ambient temperature, an ambient pressure and / or a delivery pressure with which the fluid flow is to be moved through the analysis device to be manufactured. The operating parameter can be used to characterize an application-specific design case for which the analysis device to be manufactured is to be suitable. The composition of the fluid flow and / or the at least one operating parameter can each be specified by a user, a higher-level software component or artificial intelligence.
[0009] The method further comprises a third step in which at least one flow parameter of the fluid flow is determined which results when it flows through the at least one separation device. For this purpose, in the third step the fluid flow is simulated in the virtual representation of the at least one separation device. The fluid flow can comprise a substance sample and / or a carrier gas. In the third step the fluid flow is likewise provided in the form of a virtual representation. The flow parameter determined in the third step can be a physical or chemical variable of at least one component of the fluid flow which can be influenced by the separation device. The flow parameter can in particular be a variable by which a separation behavior of the fluid flow in the separation device can be described. The determined flow parameter is output to the user and / or to a data interface.The data interface may be designed to make the flow parameter available to a computer program product, for example a graphical user interface, or to make it available to further processing within the scope of the claimed method or outside the claimed method.
[0010] The method further comprises a fourth step in which the flow parameter determined in the third step is compared with a predeterminable setpoint. The setpoint can be specified, for example, by the user or an algorithm. The setpoint can define a requirement which results from a requirement placed on the analysis device to be manufactured. In the fourth step, a check is carried out to determine whether the flow parameter corresponds to the predeterminable setpoint, i.e. whether the requirement defined thereby is met. If the flow parameter corresponds to the predeterminable setpoint, the analysis device is manufactured in accordance with the pneumatically coupled structure which was specified in the first step.
[0011] According to the invention, the flow parameter is determined in the third step based on a section-by-section simulation which smoothly follows the fluid flow, i.e. a virtual representation of the fluid flow. The propagation, i.e. the advance, of the fluid flow in the separation device is recorded accordingly. In an area wetted by the fluid flow, a section of the separation device is recreated by simulation and interactions between the fluid flow and the separation device are simulated therein. Sections upstream of the fluid flow, i.e. sections of the separation device which are not yet wetted by the fluid flow, are consequently ignored. Sections which the fluid flow has already passed through and which lie outside the simulated section are also ignored.The simulated region slides in that it is essentially firmly coupled to at least one component of the fluid flow, i.e. its virtual representation. This implements the principle of a sliding window simulation. The invention is based, among other things, on the surprising discovery that such a simulation technique offers sufficient precision and thus realism for a fluid flow flowing through a separation device. Interactions between the fluid flow and the separation device can thus be exactly predicted on the basis of their virtual representations, and thus an expected chromatogram for an appropriately manufactured analysis device that is fed with a corresponding fluid flow. Complex simulations in which the entire separation device is consistently simulated are therefore unnecessary.The claimed method can be implemented with a high degree of automation and minimized user input. Through the precise determination of the flow parameter, which is performed in the third step by simulation, the analysis device to be manufactured can be designed more precisely for the intended application, thus quickly providing a needs-based analysis device.
[0012] In one embodiment of the claimed method, the separate simulation in the third step is carried out separately for at least a first and a second component of the fluid flow. Likewise, in the third step, a flow parameter is determined for each of the first and second components. The section-by-section simulation is carried out in a separate section which smoothly follows the first or second component. Accordingly, the section-by-section simulations can be spatially partially overlapping or spatially disjoint when flow is simulated through the separation device at the same time. In particular, the first component of the fluid flow can be simulated in a first simulation window which follows the first component in the separation device. Analogously, the second component can be simulated in a second simulation window which follows the second component in the separation device.The simulation windows can be designed independently of one another and at least partially overlap during the claimed process. This allows, among other things, a separation of the fluid flow into at least the first and second components to be simulated. Overall, the different physical or chemical properties of different components can thus be precisely simulated.
[0013] In a further embodiment of the claimed method, the flow parameter is determined in the third step using a Crank-Nicolson method. With the Crank-Nicolson method, heat conduction equations and partial differential equations can be solved using a finite difference method. Furthermore, the Crank-Nicolson method is numerically stable without conditions and is therefore particularly suitable for the claimed method. The Crank-Nicolson method is also suitable for carrying out a section-by-section simulation which follows a reference point, i.e. the fluid flow, in a sliding manner. The Crank-Nicolson method is therefore particularly suitable for a sliding window simulation. Plausibility checks within the claimed method, with which unrealistic behavior of the simulation carried out with it can be checked and, if necessary, prevented, are therefore unnecessary.Among other things, the present invention is based on the surprising discovery that physical interactions between a fluid flow and a separation device can be quickly and precisely reproduced using the Crank-Nicolson method. Alternatively or additionally, the third step can also be carried out based on the Euler backward method.
[0014] Furthermore, in the third step of the claimed method, a retention behavior of the fluid flow in the at least one separation device can be simulated. The retention behavior comprises an adsorption behavior and a desorption behavior between a component of the fluid flow and a retention coating of a wall and / or a filling of the separation device. The retention behavior is simulated for at least two components of the fluid flow so that a separation behavior of the separation device between these components can be simulated. The retention behavior of components of the fluid flow can be simulated quickly and at the same time precisely, in particular by means of the Crank-Nicolson method. The at least one flow parameter which is determined in the third step can in particular be an indication which describes a separation between two components of the fluid flow achieved with the separation device.In particular, to simulate the separation process, a separate section-by-section simulation can be carried out for each component of the fluid flow, which smoothly follows the respective component.
[0015] Furthermore, the fluid flow can comprise at least one gas component and / or at least one liquid component, which is delayed by retention by the wall and / or the filling of the separation device. A combination of a gas component with a liquid component can, for example, be in the form of an aerosol. The liquid component can thus be a spray or a vapor. By means of the claimed method, the retention behavior of gases and liquids can be adjusted, thus allowing analysis devices for a wide range of applications to be quickly designed and manufactured as needed.
[0016] In a further embodiment of the claimed method, in the third step at least one flow parameter of the fluid flow is determined as it flows through at least one pneumatic module. For example, pressure losses of the fluid flow at a pneumatic module can in many cases be calculated algebraically. This makes it possible to quickly and accurately determine the behavior of the fluid flow, viewed along its flow direction, upstream and downstream of the separating device, and thus the overall behavior of the pneumatically coupled structure. A plurality of conceivable pneumatically coupled structures can be simulated, for example, by systematic variation in the course of the claimed method, and thus an analysis device particularly suitable for the intended purpose can be provided.
[0017] In the claimed method, the flow parameter which is determined in the third step can be a peak pattern of the fluid flow separated into its components by the separation device or its virtual representation in the pneumatically coupled structure. Such a peak pattern can be a graph of a chromatogram, for example. The peak pattern comprises a peak for each component of the fluid flow, i.e. a diagram tip which, when viewed by a detector, indicates an intensity indication of a measurement signal. The area essentially enclosed by the peak represents a measure of the existing quantity of the associated component of the fluid flow. The peak pattern thus comprises information about the distance from one another between individual components of the fluid flow emerging from the separation device.Consequently, the peak pattern indicates an expected measurement result from a suitable detector connected essentially immediately downstream of the separation device.
[0018] Furthermore, in the claimed method, a broadening mechanism for at least one peak in the peak pattern can also be simulated in the third step. The broadening mechanism deforms a peak compared to its idealized shape. For example, a peak is broadened by diffusion and a non-uniform radial velocity profile in the fluid flow. Likewise, broadening and / or a peak shape due to a non-ideal fluid flow, for example a sample injection, can be simulated relatively easily. Likewise, non-ideal behavior of the detector can be simulated, for example diffusion or convection in a thermal conductivity detector. The claimed method is therefore suitable for realistically simulating even non-ideal operating states of the separation device with relatively little computational effort. As a result, the technical advantages described above are achieved to a greater extent.
[0019] Furthermore, in the claimed method, the shape of at least one peak in the peak pattern can be determined using a Gaussian function. The shape of a Gaussian function, i.e. its graph, can be adapted using relatively few coefficients. Furthermore, there are known relationships for different mechanisms, in particular broadening mechanisms, which allow coefficients for a Gaussian function to be adapted accordingly. Using the Gaussian function, and thus its coefficients, a peak in the peak pattern can be precisely described using a reduced data set. The claimed method is overall data-economical and can therefore be carried out quickly in the third step. Likewise, the shape of the peak can be determined using a superposition of several Gaussian functions. Likewise, the section-by-section simulation in the third step can be carried out on a width, i.e. a window width, that is fixedly specified by the user and / or an algorithm.The width here refers to the dimension of the simulated section along the direction of the fluid flow, i.e., the window width of at least one simulation window of the sliding-window simulation. Alternatively, the width, i.e., the window width on which the section-by-section simulation is performed, can also be dynamically specified during the third step, for example, using artificial intelligence.
[0020] In a further embodiment of the claimed method, a flow parameter for the fluid flow, i.e. its virtual representation, which is present in a connecting line, is also determined. For this purpose, a virtual representation of the connecting line is present in the pneumatically coupled structure. The separating device can be connected to a pneumatic module or two pneumatic modules to one another via a connecting line. This also applies analogously to their respective virtual representations. In particular, line-length-dependent pressure losses and / or peak broadenings, for example due to diffusion, in the fluid flow or its virtual representation can be determined. This further increases the realism in the third step of the claimed method.
[0021] Furthermore, in the claimed method the first, second and third steps can be carried out repeatedly by varying the pneumatically coupled structure. The variation of the pneumatically coupled structure can be a change in length, a diameter, an internal coating, a filling of the separation device or its virtual representation. Analogously, at least one pneumatic module and / or a connecting line or their respective virtual representation can be varied. The variation can be carried out systematically, by means of an optimization algorithm and / or for example by means of artificial intelligence. The variation takes place as long as the flow parameter determined in the third step falls short of the associated predefinable target value, i.e. does not meet the requirement thereby established for the analysis device to be produced.Since the claimed method can be carried out with relatively low computational effort, frequent repetition is also possible within an acceptable time as a result of the above-described variations. Accordingly, the claimed method is also suitable for small-step variations of the pneumatically coupled structure, whereby a structural optimum for the analytical device to be manufactured can be quickly achieved. Overall, the claimed method can be used to produce analytical devices whose technical potential is increasingly exploited.
[0022] The underlying problem is also solved by a computer program product according to the invention. The computer program product is designed to determine at least one flow parameter of a fluid flow that flows through a separating device of an analysis device. The computer program product is suitable for storing a pneumatically coupled structure with at least one separating device and a plurality of pneumatic modules, i.e. their respective virtual representations, and for simulating a fluid flow or its virtual representation that flows through the pneumatically coupled structure. The computer program product is likewise suitable for setting a composition of the fluid flow or its virtual representation.According to the invention, the computer program product is designed to at least partially implement a method according to one of the embodiments outlined above, in particular at least its first, second, third and / or fourth step. For this purpose, the computer program product can be stored, at least temporarily, in a memory, for example a hard disk, a RAM, an optical data carrier and / or an FPGA. The computer program product allows for rapid, at least partially automatic design of analysis devices manufactured using a corresponding method.
[0023] The object set out above is likewise achieved by a digital twin of an analysis device according to the invention. The analysis device modeled, i.e. simulated, in the digital twin comprises at least one separating device and a plurality of pneumatic modules. For this purpose, the digital twin comprises virtual representations of the at least one separating device and the pneumatic modules. According to the invention, the digital twin is produced by a method according to one of the embodiments presented above. The digital twin can in particular be designed as an intermediate product of the method which is present after the fourth step. The digital twin can be designed as a digital twin within the meaning of the application US 2017 / 0286572 A1. The disclosure content of US 2017 / 0286572 A1 is incorporated into the present application by reference.
[0024] The problem outlined at the outset is also solved by an analysis device according to the invention. The analysis device comprises a separating device and a plurality of pneumatic modules. The separating device of the analysis device is pneumatically connected to the pneumatic modules. According to the invention, the analysis device is produced by means of a method which is designed according to one of the embodiments described above. The method can be used in particular to produce a digital twin which in turn serves as a template and / or construction specification for the claimed analysis device. In particular, a control data set for a machine tool on which the pneumatic module is manufactured can be determined from the digital twin manually or automatically, for example by means of an optimization algorithm or artificial intelligence.The claimed analysis device can be manufactured with a higher degree of automation based on the underlying method, which accelerates its production and makes it more cost-effective. The advantages of the underlying method and the resulting digital twin are specifically realized by the analysis device according to the invention and represent an economically relevant embodiment of the claimed method.
[0025] Furthermore, the underlying object is achieved by an inventive use of a so-called sliding window solver. The sliding window solver is used therein to determine a flow parameter of a fluid flow. According to the invention, the fluid flow, which is simulated or calculated with the sliding window solver in the form of its virtual representation, flows through a separating device of an analysis device, i.e. a virtual representation thereof. In particular, the sliding window solver can be used to implement one of the methods according to one of the outlined embodiments, for example in its third step. The invention is based, among other things, on the surprising finding that a sliding window solver is suitable for quickly and precisely simulating a pneumatically coupled structure of an analysis device, i.e. its virtual representation.In particular, such simulation using the sliding-window solver is so powerful that it allows for the rapid and precise design of an analysis device. The technical advantages of the underlying method are achieved to a greater extent by the sliding-window solver. These features of the corresponding method are thus transferable to the claimed use of the sliding-window solver.
[0026] The invention is explained in more detail below with reference to individual embodiments in figures. The figures are to be read as complementary to one another in that identical reference numerals in different figures have the same technical meaning. Furthermore, the individual features of the embodiments shown in the figures can also be combined with one another and with the features outlined above. They show in detail: FIG. 1 shows a schematic sequence of an embodiment of the claimed method;
[0027] FIG 2 shows a schematic sequence of a stage of a first embodiment of the claimed method.
[0028] FIG. 1 schematically shows a sequence of one embodiment of the claimed method 100. The method 100 is directed to producing an analysis device 10 which can be designed as a gas analysis device and whose virtual representation 35 is shown in FIG. 1. The analysis device 10 to be produced comprises, as separation devices 12, a first and a second separation device 12. 1, 12. 2, which is each pneumatically connected to a plurality of pneumatic modules 14. The pneumatic modules 14 can each be designed as a valve 16, as a throttle 17 or as a connecting line 18. The pneumatic modules 14 are each designed to act on at least one flow parameter 32 of a fluid flow 20 which flows through the respective pneumatic module 14.The pneumatic modules 14 and at least one of the separating devices 12 belong to a pneumatically coupled structure 30 which is provided in a first step 110 in the claimed method 100.
[0029] In the pneumatically coupled structure 30, a virtual representation 35 is stored for each of the at least one separating device 12 and the associated pneumatic modules 14. Consequently, the pneumatically coupled structure 30 comprises a plurality of functionally combined virtual representations 35. Furthermore, the pneumatically coupled structure 30 comprises gas inlets 11, through each of which a fluid flow 20 can be supplied, i.e., their virtual representations 35. Similarly, the pneumatically coupled structure 30 comprises gas outlets 13, through each of which a fluid flow 20 can be discharged. The pneumatically coupled structure 30 is designed so that a fluid flow 20, i.e., its virtual representation 35, enters the pneumatically coupled structure 30 via a gas inlet 11 and exits it again via a gas outlet 13. The fluid flow 25 can comprise a substance sample 15 and / or a carrier gas 25.The pneumatic modules 14 represent a flow resistance 19 for the fluid flow 20 passing through them. The flow resistance 19 influences one of the flow parameters 32 of the fluid flow 20. Furthermore, the flow resistance 19 is determined by design parameters of the associated pneumatic module 14 (not shown in detail).
[0030] The method 100 comprises a first step 110 in which the pneumatically coupled structure 30, as shown by way of example in FIG. 1, is provided, for example by a user who carries out the method 100 or by a template. The pneumatically coupled structure 30 can be displayed via a graphical user interface and adapted by the user. The pneumatically coupled structure 30 is designed to be pneumatically coupled such that the fluid flow 20, that is to say its virtual representation 35, can only enter via at least one gas inlet 11 and can only exit via at least one gas outlet 13.
[0031] Furthermore, a second step 120 belongs to the claimed method 100, in which a composition 22 of the fluid flow 20 is specified, i.e. its virtual representation 35. Likewise, at least one operating parameter is specified which acts on the fluid flow 20 to be simulated, for example an ambient temperature. The fluid flow 20 comprises a plurality of components 21 and their flow through the pneumatically coupled structure 30 is to be simulated in the claimed method 100. The components 21 are separated by the separation device 12 in such a way that they arrive one after the other on reaching a detector (not shown in detail) and cause a signal peak 24 for each component 12 at the detector. The fluid flow 20 is designed as a material sample 15 which is to be passed through one of the separation devices 12, namely the first separation device 12.1.whose behavior is simulated via their respective virtual representations 35. The composition 22 of the fluid flow 20 is specified in the second step 120, for example, by the user. The claimed method 100 also includes a third step 130, in which at least one flow parameter 32 of the fluid flow 20 is determined, which is established when flowing through the separation device 12, i.e. the first separation device 12.1. For this purpose, a flow behavior and retention behavior of individual components 21 of the fluid flow 20, not shown in detail in FIG. 1, is simulated at least in the separation device 12 with their respective virtual representations 35. Embodiments of the third step 130 are shown in more detail in FIGS. 2, 3. Furthermore, in the third step 130, the determined flow parameter 32 is output to the user and / or a data interface (not shown in more detail).By outputting it to the data interface, the determined flow parameter 35 can be stored at least temporarily.
[0032] Furthermore, the claimed method 100 comprises a fourth step 140 in which the flow parameter 32 determined in the third step 130 is compared with a predefinable target value 33. The target value 33 can be specified by the user or determined from a user input. If the determined flow parameter 32 falls short of the target value 33, i.e. does not meet the requirement represented thereby for the pneumatically coupled structure 30, the first, second and third steps 110, 120, 130 are carried out repeatedly. The repeated implementation takes place here by varying 36 the pneumatically coupled structure 30. In this case, design parameters of at least one of the separating devices 12, 12.1, 12.2 and / or at least one pneumatic module 14 are varied by means of an artificial intelligence 50. The artificial intelligence 50 is designed as a so-called neural network 52, which can be trained with training data sets.Accordingly, by varying 36, a plurality of pneumatically coupled structures 30 is generated, in which at least one flow resistance 19 changes. Each of the pneumatically coupled structures 30 thus generated thus has a plurality of virtual representations 35 of different pneumatic modules 14 and / or separating devices 12, 12.1, 12.2. The first, second, and third steps 120, 120, 130 are carried out while varying 36 the pneumatically coupled structure 30 until the at least one flow parameter 32 corresponds to the associated target value 35.
[0033] If the determined flow parameter 32 corresponds to the target value 33, i.e. fulfills the requirement represented thereby, the pneumatically coupled structure 30 is output and the analysis device 10 is manufactured based thereon. The pneumatic structure 30, in which the at least one flow parameter 32 corresponds to the target value 33, represents an intermediate product in the claimed method 100, which is at least temporarily stored as a so-called digital twin 55 of the analysis device 10 to be manufactured. The digital twin 55 of the analysis device 10 to be manufactured can be output to the user. The claimed method 100 can be executed by a computer program product 60, which can be monolithic or can comprise a plurality of subprograms that interact in a random operation and thus realize the functionality of the method 100.An operating behavior of the analysis device 10, which is produced by means of the claimed method 100, can be reproduced, i.e. simulated, by the digital twin 55.
[0034] A stage of a first embodiment of the claimed method 100 is shown schematically in FIG 2. In detail, FIG 2 shows a third step 130 of the claimed method 100, which can be carried out, for example, in FIG 1. The embodiment according to FIG 2 assumes that the first and second steps 110, 120 of the claimed method 100, as shown by way of example in FIG 1, have already been successfully completed. A separating device 12, in particular a first separating device 12.1, is specified in terms of its structure. The structure, i.e. the associated virtual representation 35 of the separating device 12, is defined by design parameters, for example an inner diameter 38 of the separating device 12, its length 34, and / or a layer thickness 39 of a retention coating 28 on a wall 26 of the separating device 12. These are specified in the first step 110 .Furthermore, a composition 21 of a substance sample 15, i.e. its virtual representation 35, is specified, which flows through the separation device 12 as a fluid flow 20.
[0035] In the third step 130, the separating device 12 and the fluid flow 20 located therein are adjusted section by section, the associated sections 40 following a flow direction of the fluid flow 20, which is shown by arrows in FIG. 2. A first section 41 contains a part of the fluid flow 20 which has, as components 21, a first and a second component 44, 46 which are mixed with one another. The first and second components 44, 46 interact with the retention coating 28 in the separating device 20. The retention coating 28 retains the first and second components 44, 46 for different lengths of time, so that a retention effect 29 occurs. As a result of the retention effect 29 with the retention coating 28, the second component 46 is retained more strongly than the first component 44.In a second section 42 of the separation device 12, the first component 44 is present separately, that is to say separate from the second component 46. Correspondingly, the second component 46 is present separately in a third section 43. The retention effect 29 on the first and second components 44, 46 of the fluid flow 20, that is to say their corresponding virtual representations 25, is initially carried out in the third step 130 only for the first section 41. The first component 56 is simulated in a first simulation window 56, which follows the first component 44. Correspondingly, the second component 46 is simulated in a second simulation window 48. The first and second simulation windows 56, 58 follow the first and second components 44, 46, respectively. In the area of the first section 41, the first and second components 44, 46 are mixed, so that the first and second simulation windows 56, 58 essentially overlap.The retention effect 29 is determined step by step for additional sections 40 located between the first and second sections 41, 42 and the first and third sections 41, 43. A window width 49 of the first, second, and third sections 41, 42, 43, i.e., the simulation windows 56, 58, remains constant.
[0036] The degree of separation between the first and second components 44, 46 represents a flow parameter 32, which is determined in the third step 130. The flow parameter 32 is determined in a plurality of sections 30 that are adjacent to one another and partially overlap. The sections 30 are each simulated separately for the first and second components 44, 46. The sections 30 in which the flow parameter 32 is determined, and thus the simulation windows 56, 58, each follow that of the first and second components 44, 46 of the fluid flow 20 with their respective flow velocities. As a result, the same particles of these components 44, 46 and their associated retention effect 29 are always simulated when transferred to the virtual representations 35 of the first and second components 44, 46 of the fluid flow 20.The adjacent sections 40, each for the first and second components 44, 46, which are simulated in this way, define boundary conditions 47 for each other. Depending on the flow velocity of the first or second component 44, 46 of the fluid flow 20, time steps 45 result which separate the values determined for the respective sections 40 for the flow parameter 32 of the first or second component 44, 46 from each other. The time steps 45 also specify a follow-up 48 with the first or second component 44, 46 for the first and second simulation windows 56, 58. Due to the retention effects 29 on the first and second components 44, 46, the first and second simulation windows 56, 58 diverge and are disjoint, as in FIG. 2. Such a degree of separation 23 is also symbolized in FIG. 2. This results in a total of a follow-up 48 of the sections 40 from the first to the second section 42 .By thus reproducing the retention effect 29 on the first and second components 44, 46 of the fluid flow 20, the principle of a sliding window simulation is realized. This allows the third step 130 to be carried out quickly. The longer the separation device 12, the greater the time savings in the third step 130 compared to a simulation in which the separation device 12 and the fluid flow 20, i.e., their virtual representations 35, are recreated simultaneously with their retention effect 29. Accordingly, the degree of separation 23 can be determined quickly as a flow parameter 32. This makes it possible, with acceptable computational effort, to recreate a large number of pneumatically coupled structures 30 in several runs of the first, second and third steps 110, 120, 130 and thus to determine, by variation 36, a pneumatically coupled structure 30 which corresponds to the target value 33.As a result, an analytical device 10 to be manufactured can be designed at least partially automatically. The invention is based, among other things, on the finding that by using a suitably suitable sliding-window solver 65, the third step 130 can be performed at an increased speed, so that a variation 36 of the pneumatically coupled structure 30 is again practical. This simplifies and accelerates the overall manufacture of a corresponding analytical device 10.
Claims
Patent claims 1. Method (100) for producing an analysis device (10) with at least one separating device (12, 12.1, 12.2) and a plurality of pneumatic modules (14), comprising the steps of: a) providing a pneumatically coupled structure (30) comprising virtual representations (35) of the at least one separating device (12, 12.1, 12.2) and the plurality of pneumatic modules (14); b) specifying a composition (22) of a fluid flow (20) and at least one operating parameter for the pneumatically coupled structure (30); c) determining at least one flow parameter (32) of the fluid flow (20) as it flows through the at least one separating device (12, 12.1, 12, .) and outputting the at least one flow parameter (32) to a user and / or a data interface; d) manufacturing the analysis device (10) according to the pneumatically coupled structure (30) specified in step a) if the determined flow parameter (32) corresponds to a predefinable target value (33); characterized in that step c) is carried out based on a section-by-section simulation that smoothly follows the fluid flow (20).
2. Method (100) according to claim 1, characterized in that the section-wise simulation in step c) is carried out separately for a first and a second component (44, 46) of the fluid flow (20), for each of which a flow parameter (32) is determined.
3. Method (100) according to claim 1 or 2, characterized in that the determination of the flow parameter (32) in step c) is carried out using a Crank-Nicolson method.
4. Method (100) according to one of claims 1 to 3, characterized in that in step c) a retention behavior (29) of the fluid flow () in the at least one separation device (12, 12.1, 12.2) is simulated.
5. Method (100) according to claim 4, characterized in that the fluid flow (20) comprises at least one gas component, at least one liquid component and / or at least one solid component, which is delayed by retention by a retention coating (28) of a wall (26) and / or a filling of the separation device (12, 12.1, 12.2).
6. Method (100) according to one of claims 1 to 5, characterized in that in step c) at least one flow parameter (32) of the fluid flow (20) is also determined when flowing through at least one pneumatic module (14).
7. Method (100) according to one of claims 1 to 6, characterized in that the flow parameter (32) is a peak pattern of the fluid flow (20) separated into its components (44, 46) by the separation device (12, 12.1, 12.2).
8. Method (100) according to claim 7, characterized in that in step c) a broadening mechanism for at least one peak (24) is adjusted.
9. Method (100) according to claim 7 or 8, characterized in that a shape of at least one peak (24) in the peak pattern is determined by means of a Gaussian function.
10. Method (100) according to one of claims 1 to 9, characterized in that the section-wise simulation takes place on a window width (49) which is fixedly predetermined by a user and / or an algorithm.
11. Method (100) according to one of claims 1 to 10, characterized in that it also includes a flow parameter (32) of the fluid flow (20) in a connecting line (18) is determined.
12. Method (100) according to one of claims 1 to 11, characterized in that steps a), b) and c) are carried out repeatedly while varying the pneumatically coupled structure (30).
13. The method (100) according to claim 12, characterized in that the variation of the pneumatically coupled structure (30) is carried out by means of an optimization algorithm and / or artificial intelligence.
14. Method (100) according to one of claims 1 to 13, characterized in that the composition (22) of the fluid flow (20) and / or the at least one operating parameter in step b) are specified by the user, a higher-level software component or an artificial intelligence.
15. Computer program product (60) for determining a flow parameter (32) of a fluid flow (20) through a separation device (12, 12.1, 12.2) of an analysis device (10), characterized in that the computer program product (60) is designed to carry out at least steps a), b) and c) of a method (100) according to one of claims 1 to 14.
16. Use of a sliding window solver (65) for determining a flow parameter (32) of a fluid flow (20), wherein the fluid flow (20) flows through a separation device (12, 12.2, 12.2) of an analysis device (10), characterized in that the flow parameter (32) is determined with the sliding window solver (65) in a method (100) according to one of claims 1 to 14.