Computer-implemented method for generating a proposal for a spare part in a hydraulic system

Abstract

The invention relates to a computer-implemented method for generating a proposal for a spare part in a hydraulic system, wherein the method comprises the following steps: - Receiving component information of a component to be replaced (S24); - Providing a selection of available spare parts (S26); - Entering the component information into a performance model (S28) to determine a performance spectrum for the component to be replaced; - Identifying at least one replacement part (S30) that reflects the performance range of the component to be replaced; and - automatic generation of a proposal (S32) for at least one identified spare part.

Description

[0001] The invention relates to a computer-implemented method for generating a proposal for a spare part in a hydraulic system. State of the art

[0002] In industry, the process of selecting a suitable replacement part for a damaged or worn component in a hydraulic system is a complex, multi-stage procedure that includes a thorough analysis of both the technical specifications and the operational requirements. The first step is to analyze the original component in detail to determine its physical and functional parameters. Depending on the component, these include its dimensions, nominal pressure, flow rate, temperature and media resistance, as well as specific design features such as seals, material properties, and tolerances. In many cases, this specific data is available in technical documentation such as design drawings, maintenance manuals, or catalogs from the original manufacturer.

[0003] Once the parameters of the defective component have been determined, the search for a suitable replacement part begins. This step requires comparing the determined specifications with the technical data of available replacement parts. In practice, the aim is often to find replacement parts either from the original manufacturer or from an authorized third-party supplier that provides components of equivalent or improved quality. A key criterion here is the compatibility of the mechanical and hydraulic interfaces. It is particularly important to ensure that the replacement part is capable of handling the same forces, pressures, and temperatures as the original part, as any deviation in these aspects can lead to malfunctions or even safety risks.

[0004] In addition to technical parameters, legal and economic aspects also play a role. Standards such as ISO or DIN may need to be adhered to in order to ensure operational safety and meet insurance requirements. Furthermore, the availability of the spare part must be verified. In industrial practice, it can happen that the original component is no longer manufactured, necessitating a search for functionally equivalent components. In such cases, so-called "retrofit solutions" are developed, using modern components that are made compatible with the old systems.

[0005] Another important criterion is the quality and durability of the replacement component. Especially with highly stressed components like valves or pistons, it can be beneficial to switch to higher-quality materials or improved technologies to minimize future wear. This may require examining the system's load collectives and operating profiles to ensure the replacement part can withstand the specific stresses. In some cases, special tests, such as pressure tests or flow simulations, are performed to verify a replacement part's suitability under the specific operating conditions.

[0006] The selection process often involves consulting experts with extensive specialist knowledge. For complex hydraulic systems, detailed consultations with the manufacturers are essential to ensure that the replacement part meets the required functionalities. The final stage of the decision-making process involves a cost-benefit analysis, which considers not only the price of the replacement part but also its impact on the overall system's operating costs.

[0007] One of the main problems when selecting replacement parts is the incomplete or inaccurate recording of the original component's technical specifications. Often, the original technical data sheets are no longer available, or the documentation is incomplete, which can lead to important parameters such as precise tolerances, pressure resistance, or permissible temperature ranges being overlooked. Especially in older systems that may have been modified over many years, discrepancies can exist between the installed component and the one originally specified. This makes it difficult to precisely determine the exact requirements for the replacement part.

[0008] Another common problem is the incompatibility of spare parts that appear similar externally or based on basic specifications, but differ in critical details upon closer inspection. Particularly with hydraulic components such as valves or pistons, even slight differences in dimensions or material composition can lead to sealing problems, insufficient pressure resistance, or undesirable flow characteristics. Even if the mechanical interfaces appear to fit, differences in hydraulic performance characteristics can result in inefficient operation or even premature component failure. For example, valves designed for lower flow rates can cause uncontrolled flow in a high-pressure system, ultimately leading to damage to other system components.

[0009] The availability of spare parts is also a frequent problem, especially with older or specialized hydraulic systems. If the original components are no longer manufactured, it may be necessary to switch to alternative parts, which require not only technical but also design modifications. Such adaptations often necessitate extensive modifications to the system, which can lead to significant costs and downtime. Furthermore, with retrofit solutions, the new components may not be fully compatible with the existing system architecture, requiring the additional integration of control electronics or other systems. This can create unforeseen technical challenges, particularly if the new components interact with other existing components whose behavior changes during operation.

[0010] Another common problem arises when using third-party alternative parts. Although these parts may appear technically equivalent or even superior, their quality can vary significantly, especially if production technologies or quality controls are not on par with the original manufacturer. In practice, this sometimes results in seemingly compatible replacement parts having a shorter lifespan or experiencing unpredictable failures, as they have not been tested under the same conditions as the original parts. These quality issues often only become apparent after installation and during operation, potentially leading to additional downtime and repair costs.

[0011] The interfaces between mechanical and electronic components can also cause problems. Modern hydraulic systems are often equipped with control units and sensors that are precisely matched to the original components. Replacing a component with a non-equivalent replacement part can lead to problems with signal processing or feedback, affecting the efficiency and accuracy of the control system. In some cases, this requires recalibration or even adjustments to the control algorithms, which means additional effort and requires specialized knowledge.

[0012] Last but not least, the economic aspects of the spare parts sourcing process can also be problematic. The effort involved in searching for a suitable spare part, checking its compatibility, and potentially adapting the system must be economically proportionate to the costs of a new system or a more comprehensive modernization. Companies are often under pressure to find cost-effective solutions in the short term, which can lead to the use of inferior or insufficiently tested spare parts, thus posing long-term risks to operational safety and the reliability of the overall system.

[0013] To overcome these disadvantages, expertise from specialists is often relied upon. However, this expertise is not always available and is not uniformly applicable.

[0014] The invention is therefore based on the objective of proposing a method for identifying spare parts for components in hydraulic systems. Furthermore, parameterization or calibration is generated for digitally configurable components to ensure better adaptation.

[0015] The problem is solved by the subject matter of the independent claims. Disclosure of the invention

[0016] According to a first aspect of the invention, this problem is solved by a computer-implemented method for generating a proposal for a spare part in a hydraulic system, the method comprising the following steps: - Receiving component information for a component to be replaced; - Providing a selection of available spare parts; - Entering the component information into a performance model to determine a performance spectrum for the component to be replaced; - Identifying at least one replacement part that covers the performance range of the component to be replaced; and - Automatic generation of a suggestion for at least one identified spare part.

[0017] This procedure is implemented to simplify the process of finding and selecting a replacement part for a hydraulic system. Therefore, the procedure can be carried out, in particular, by an automated system into which the component information of the part to be replaced is entered.

[0018] The selection of available spare parts includes information on which spare parts are generally available for replacement. This information includes the performance specifications of the spare parts.

[0019] The invention is based on the fact that the component information is entered into a performance model, with which the performance spectrum of the component is created.

[0020] The performance spectrum indicates how the component behaves in operation and what properties it possesses. From this information, further parameters of the component to be replaced can be derived, such as the materials it may be made of, its electronic, fluid-dynamic, and / or mechanical connections, its temperature range, and / or the standards it meets. If specific details regarding these properties are already included in the component information, they can be used for the further course of the proposed procedure.

[0021] The performance spectrum determined using the performance model is then compared with the performance spectra of the available spare parts. From this comparison, at least one spare part is selected that meets the requirements of the component to be replaced. The selection can be made, for example, by choosing the spare part whose performance spectrum most closely matches that of the component being replaced. Individual aspects of the performance spectrum can be weighted differently. For instance, it may be more important that the spare part can withstand a certain pressure than a specific behavior during operation at a defined pressure.

[0022] Once at least one potential replacement part has been identified, it can be suggested. For example, a user can then install the replacement part in the hydraulic system, order the replacement part directly from a webshop, or take other suitable actions.

[0023] The proposed method significantly simplifies the selection process for a replacement part for a hydraulic system. In this way, the invention solves its problem.

[0024] In one embodiment, the component to be replaced is a valve.

[0025] Valves are often the most wear-prone components in hydraulic systems. They typically have moving parts that are electronically controlled and are subjected to heavy loads. Therefore, valve replacement is frequently necessary. However, the market offers a wide range of valves, and especially for older hydraulic systems, compatible valves may not be available.

[0026] The proposed method is therefore particularly suitable for valves, as previously it was often necessary to rely on specialist personnel with the appropriate expertise.

[0027] In one embodiment, the performance spectrum includes at least one characteristic curve of two parameters of the component.

[0028] A characteristic curve is the graphical representation of the relationship between two physical quantities that is characteristic of a component, assembly, or device. This relationship is represented as a line in a planar coordinate system. The characteristic curve serves to illustrate the relationship, but also to represent it quantitatively when an algebraic function of the relationship is unknown. Characteristic curves can be derived directly from measured values. However, this is not always possible, so an approximation of the characteristic curve can be determined using a model. For this purpose, mathematical functions, particularly fluid dynamic functions, can be used that describe the behavior of the component, at least approximately.

[0029] Characteristic curves have the advantage over other theoretically usable parameters of being easily interpretable. Therefore, when replacing a component, specialists can better predict the behavior of the replacement part within the hydraulic system.

[0030] In one embodiment, the performance model includes a simulation of the hydraulic system.

[0031] The simulation of the hydraulic system can be, in particular, a physical, preferably a fluid dynamic simulation of the hydraulic system. This involves simulating the behavior of the component under specific conditions. These conditions can include, for example, different operating points or external influences on the component, especially forces.

[0032] Input parameters for the simulation can be derived from the component data. These include, for example, the component's dimensions or other structural properties such as flow cross-sections and load limits. The simulation output is the performance spectrum of the component to be replaced, containing all the information required for comparison with the selection of replacement parts. This can include, for example, individual values ​​from a characteristic curve, system states at specific operating points, or other features.

[0033] In one embodiment, the determination of the at least one spare part is carried out using a simulation of the at least one spare part in the hydraulic system, wherein the at least one spare part is parameterized for the simulation, and the result of the simulation includes the performance spectrum of the at least one spare part for the selected parameterization.

[0034] The simulation of the spare parts can preferably be based on the same model as the simulation of the component to be replaced in other embodiments. Therefore, the simulation of the spare parts can also be based on a physical model, in particular a fluid dynamic model.

[0035] Parameterization is used to adapt the performance and / or other properties of the potential replacement part to the performance range of the component being replaced.

[0036] For example, if several valves are available as replacement parts, all of which exceed the performance of the component being replaced in one or more respects, these valves can be parameterized so that their performance range matches that of the component being replaced as closely as possible. For instance, a flow cross-section can be reduced, or the actuation times can be adjusted to match the behavior of the replacement part, such as the response time, and to simulate the behavior of the component being replaced.

[0037] During parameterization, specific values ​​are assigned to the curves that describe the behavior of the available spare parts, with which the behavior of the respective spare part can be adjusted.

[0038] In one embodiment, the simulation of the at least one spare part is performed in several runs, with the parameterization of the at least one spare part being adjusted in each run.

[0039] Performing multiple runs advantageously results in more accurate simulation results for the provided spare parts, as each spare part is better adapted to the performance range to be met, and the effect of this adaptation can lead to an optimized selection of the proposed spare parts.

[0040] In each iteration, the individual control parameters of the spare parts can be cycled through or permuted in order to test different modes for use.

[0041] In one embodiment, the parameterization is adjusted using a machine learning algorithm.

[0042] The machine learning algorithm is trained to adjust the parameters in each iteration so that the resulting performance spectrum matches the required performance spectrum of the component being replaced as closely as possible. To determine the quality of this match, the machine learning algorithm can calculate a loss function that is minimized in each iteration until either a maximum number of iterations is reached or the result of the loss function no longer changes significantly.

[0043] To optimize the parameterization of spare parts, various machine learning models are suitable, each approaching the task in a different way. A common choice is reinforcement learning (RL), particularly models like Q-learning or Deep Q-Networks (DQN). These models learn through interaction with the respective spare part and adjust their decisions using rewards or punishments. RL models are especially suitable when the spare part requires dynamic adjustments and the effects of actions in a time-varying environment are not directly predictable. RL learns through trial and error, which makes it robust in changing environments. However, it requires a large number of interactions to find optimal settings.

[0044] In contrast, supervised learning models, such as linear regression or neural networks, can be used to optimize a predefined objective function. These models learn from historical data and look for patterns that describe the relationship between the parameters and the output. These models rely on a larger amount of historical data and only work well if the behavior of the components is largely stable. The strength of supervised learning models lies in their suitability for predictable, repeatable scenarios where the relationships are linear or nonlinear, but fixed.

[0045] Bayesian optimization offers another way to optimize the parameterization of a spare part. This method uses probabilistic models, often based on Gaussian processes, to find a balance between exploration and exploitation. This means it investigates which parameterizations are still unexplored to potentially achieve better results, while simultaneously refining already known, effective settings. Bayesian optimization is particularly useful when it comes to minimizing the number of iterations or simulations.

[0046] Evolutionary algorithms operate on a principle similar to natural selection. They generate a population of solutions and improve them over several generations through crossing and mutation. This method is particularly suitable when the optimization landscape is very complex and local optima should be avoided. Evolutionary algorithms typically require many evaluations, but they are robust against noisy data and can also be used for non-differentiable or discrete parameterizations.

[0047] In one embodiment, the parameterization is adjusted so that the performance spectrum obtained through the simulation matches the performance spectrum of the component to be replaced.

[0048] In one embodiment, the simulation is performed for several spare parts, wherein the proposal for the at least one identified spare part includes the spare parts whose simulated performance spectrum has the highest degree of similarity to the performance spectrum of the component to be replaced.

[0049] The degree of agreement can be determined in stages and / or with weighting. Depending on the application of the hydraulic system, different parameters, properties, or operating points can be weighted differently.

[0050] In another aspect, the invention relates to a computer program with program code for carrying out a method as described above when the computer program is executed on a computer.

[0051] In another aspect, the invention relates to a computer-readable data carrier containing the program code of a computer program for carrying out a method as described above when the computer program is executed on a computer.

[0052] In another aspect, the invention relates to a system for generating a proposal for a spare part in a hydraulic system, wherein the system is configured to carry out a method as described above.

[0053] The system can be a computing machine, in particular a personal computer or a server, on which a web-based application is run, which can be controlled via a web interface, in particular a web browser.

[0054] In summary, the present invention provides a method for generating a proposal for a spare part in a hydraulic system, a computer program, a computer-readable data carrier, and a system for generating a proposal for a spare part in a hydraulic system.

[0055] The described configurations and training programs can be combined in any way desired.

[0056] Further possible embodiments, developments and implementations of the invention also include combinations of features of the invention described previously or subsequently with regard to the exemplary embodiments that are not explicitly mentioned. Brief description of the drawings

[0057] The accompanying drawings are intended to provide a further understanding of the embodiments of the invention. They illustrate embodiments and, in conjunction with the description, serve to explain the principles and concepts of the invention.

[0058] Other embodiments and many of the aforementioned advantages become apparent with reference to the drawings. The elements depicted in the drawings are not necessarily shown to scale.

[0059] They show: Fig. 1 schematically illustrates the process of the inventive method according to one embodiment; and Fig. 2 schematically illustrates the process of the inventive method according to a further embodiment.

[0060] In the figures of the drawings, identical reference symbols denote identical or functionally equivalent elements, parts or components, unless otherwise stated.

[0061] Fig. Figure 1 schematically illustrates the process of the inventive method according to one embodiment. In this embodiment, the method is executed in two domains, 10 and 12. Domain 10 is a user domain in which a user of the method resides.

[0062] The user can use a computer to generate a request for replacing a component in a hydraulic system in step S10. A user interface can be used for this purpose. With their request, the user enters component information, which is then incorporated into a performance model 14 in step S12. The performance model 14 determines a performance spectrum for the component to be replaced, which should be represented as accurately as possible by the replacement parts. In the illustrated embodiment, the performance model 14 is executed in the user domain. In alternative embodiments, the performance model 14, and thus also step 12, can be executed in a different domain.

[0063] In step S14, a selection of possible products is determined based on the performance spectrum. In the illustrated embodiment, the steps from step S14 onwards are executed in the second domain 12. The second domain 12 can, for example, be a dealer or manufacturer domain in which a database 16 is provided containing technical data and the availability of possible spare parts.

[0064] The information required to identify suitable spare parts is loaded from database 16. This can include, in particular, key figures or characteristic curves of the spare parts, but also information on the material properties and / or load limit of the respective spare part.

[0065] A machine learning algorithm can be used in step S14 to narrow down the selection of potential spare parts. In the next step, S16, the pre-selected spare parts are simulated, which can also be done using a machine learning algorithm. In particular, a different machine learning algorithm than the one used in step S14 can be used here.

[0066] In step 14, the machine learning algorithm searches for the most suitable selection of spare parts, which essentially corresponds to a classification problem. In step S16, however, various parameter settings for the spare part are tested and simulated so that the performance range of the spare parts is as close as possible to the performance range of the component being replaced. This corresponds to an optimization task.

[0067] In step S18, the results are finally compared. A value for the goodness of fit of the performance spectra can be determined for this purpose. This goodness of fit can preferably also be determined by a machine learning algorithm.

[0068] In step S20, the results are summarized into a proposal for a replacement part for the component to be replaced. Depending on the configuration and / or the user's wishes, this proposal can include one or more replacement part options, which the user can then select and process.

[0069] In step S22, a replacement part suggested by the system is finally installed in the hydraulic system.

[0070] Fig. 2 shows the procedure at a more abstract level than the procedure from Fig. 1.

[0071] In step S24, component information is provided, which can be entered by a user, for example. In addition, step S26 provides a selection of available spare parts. When providing this selection, information is given specifically to evaluate the suitability of the spare parts for replacing the component being replaced.

[0072] In step S28, the provided component information is entered into a performance model to generate a performance spectrum for the component to be replaced. This determines which specifications the proposed replacement part must have.

[0073] In step S30, the determined performance spectrum is compared with the available information on the available spare parts. This step may include simulating the spare parts themselves, so that a performance spectrum is also available for the spare parts, or at least for a narrower selection of them. In the latter case, the performance spectra of the spare parts can be compared with each other and with the performance spectrum of the component to be replaced. In step S32, the spare part whose performance spectrum best matches that of the component to be replaced is presented to the user with an automatically generated suggestion.

Claims

[1] Computer-implemented method for generating a proposal for a spare part in a hydraulic system, the method comprising the following steps: - Receiving component information of a component to be replaced (S24); - Providing a selection of available spare parts (S26); - Entering the component information into a performance model (S28) to determine a performance spectrum for the component to be replaced; - Identifying at least one replacement part (S30) that reflects the performance range of the component to be replaced; and - automatic generation of a proposal (S32) for at least one identified spare part. [2] Computer-implemented method according to claim 1, wherein the component to be replaced is a valve. [3] Computer-implemented method according to one of the preceding claims, wherein the performance spectrum includes at least one characteristic curve of two parameters of the component. [4] Computer-implemented method according to any of the preceding claims, wherein the performance model includes a simulation of the hydraulic system. [5] Computer-implemented method according to one of the preceding claims, wherein the determination of the at least one spare part (S30) is carried out using a simulation of the at least one spare part in the hydraulic system, wherein the at least one spare part is parameterized for the simulation, wherein the result of the simulation includes the performance spectrum of the at least one spare part for the selected parameterization. [6] Computer-implemented method according to claim 5, wherein the simulation of the at least one spare part is carried out in several runs, wherein the parameterization of the at least one spare part is adjusted in each run. [7] Computer-implemented method according to claim 6, wherein the parameterization is adapted using a machine learning algorithm. [8] Computer-implemented method according to claim 7, wherein the parameterization is adapted such that the performance spectrum obtained by the simulation matches the performance spectrum of the component to be replaced. [9] Computer-implemented method according to any one of claims 5 to 8, wherein the simulation is carried out for several spare parts and wherein the proposal for the at least one identified spare part includes the spare parts whose simulated performance spectrum has the highest degree of similarity to the performance spectrum of the component to be replaced. [10] Computer program with program code to execute a method according to any of the preceding claims when the computer program is executed on a computer. [11] Computer-readable data carrier containing program code of a computer program for executing a method according to any one of claims 1 to 9 when the computer program is executed on a computer. [12] System for generating a proposal for a spare part in a hydraulic system, wherein the system is configured to perform a method according to any one of claims 1 to 9.

Images