An ANN-based system for evaluating hydrodynamic bearing performance

Abstract

A neural network-based system for evaluating the performance of hydrodynamic bearings, consisting of: a material selection module, comprehensive: a fuzzy AHP submodule configured to determine weighting factors for material selection criteria; an ELECTRE sub-module configured to evaluate and select optimal bearing materials based on criteria such as hardness, yield strength, modulus of elasticity, thermal conductivity, density, thermal expansion and cost; an analysis module configured to simulate bearing performance with input parameters rotational speed and eccentricity to generate output parameters such as maximum pressure and indexing angle by using a dedicated processor to apply computational fluid dynamics to perform a simulation based on the input parameters; a bearing test system for conducting experimental studies based on the material selected by the material selection module and the output parameters obtained by the analytical study module, wherein the bearing test system comprises a vertically mounted shaft, a variable speed motor, a load application mechanism and pressure and speed sensors; a statistical analysis module configured to perform an analysis of variance (ANOVA) and a regression analysis of the results obtained from the analysis module and an experiment carried out using a stock testing system; a data acquisition unit configured to acquire data on storage materials and their properties, wherein the data acquisition unit is further configured to retrieve the output data obtained from the material selection module, the analysis module and the experiment using a storage testing system; an artificial neural network (ANN) module comprising a dedicated processor with dedicated memory, wherein the processor is trained on input data, which is the data acquired by the data acquisition unit, to predict the performance of a hydrodynamic bearing, wherein the processor is configured to preprocess and normalize the input data and predict the bearing performance using a feedforward neural network with a backpropagation algorithm; and a data visualization interface connected to a user interface, wherein the data visualization interface performs the visualization of the predicted output of the artificial neural network (ANN) module and wherein the user interface displays the visualized data to the user.

Description

FIELD OF THE INVENTION

[0001] The present disclosure relates to an ANN-based system for evaluating the performance of hydrodynamic bearings. More specifically, the present invention relates to an artificial neural network-based system for predicting performance parameters of hydrodynamic bearings, wherein the ANN model is trained using data obtained from analytical and experimental results, and fuzzy AHP and ELECTRE techniques are used for bearing material selection. BACKGROUND OF THE INVENTION

[0002] In the rapidly evolving field of mechanical engineering, hydrodynamic bearings play a critical role in numerous industrial applications, from automotive systems to precision machinery. Despite their widespread use, determining optimal bearing performance and material selection remains a complex and challenging task. Conventional evaluation methods are often based on limited experimental approaches or simplified computational models that frequently fail to capture the complex interactions between material properties, operating conditions, and performance characteristics.

[0003] The proposed invention addresses these fundamental challenges by introducing a comprehensive, multidimensional approach to hydrodynamic bearing analysis. By integrating advanced techniques such as fuzzy AHP (Analytical Hierarchy Process), the ELECTRE method, computational fluid dynamics, sophisticated experimental testing, and artificial neural network modeling, this system provides an unprecedented level of insight into bearing performance. The approach enables a holistic assessment that simultaneously considers multiple critical factors, including material properties, operating parameters, and performance metrics, something previously impossible with conventional assessment techniques. SUMMARY OF THE INVENTION

[0004] This disclosure relates to a neural network (ANN)-based system for evaluating the performance of hydrodynamic bearings. The proposed system uses the fuzzy AHP technique to determine weighting factors for various criteria and an ELECTRE technique to determine the best material. The selected bearing material is then used for analytical and experimental investigations, with ANSYS Fluent being used for analytical investigations and a journal bearing test rig being used for experimental investigations. A statistical analysis is performed, using ANOVA and regression analyses on the analytical and experimental results. Finally, an artificial neural network model is developed, trained with the same parameters used for analysis and experiment, which also provides some predicted parameters.Finally, the parameters predicted by the ANN are compared with the analysis and experimental results and their deviations are observed.

[0005] The present disclosure is intended to provide an ANN-based system for evaluating hydrodynamic bearing performance. The system includes: a material selection module including a fuzzy AHP sub-module configured to determine weighting factors for material selection criteria, and an ELECTRE sub-module configured to rank and select optimal bearing materials based on criteria such as hardness, yield strength, elastic modulus, thermal conductivity, density, thermal expansion, and cost; an analysis module configured to simulate bearing performance with input parameters of rotational speed and eccentricity to generate output parameters such as maximum pressure and indexing angle using a dedicated processor configured to utilize computational fluid dynamics to perform the simulation based on the input parameters;a bearing testing system for conducting experimental studies based on the material selected by the material selection module and the output parameters obtained by the analysis study module, the bearing testing system comprising a vertically mounted shaft, a variable-speed motor, a load application mechanism, and pressure and speed sensors; a statistical analysis module configured to perform an analysis of variance (ANOVA) and to perform a regression analysis on the results obtained by the analysis module and the experiment conducted with the bearing testing system; a data acquisition unit configured to obtain data on bearing materials and their properties, the data acquisition unit further configured to retrieve the output data obtained from the material selection module, the analysis module, and the experiment using a bearing testing system;an artificial neural network (ANN) module comprising a dedicated processor with dedicated memory, the processor being trained on the input data, which is the data acquired by the data acquisition unit, to predict the performance of a hydrodynamic bearing, the processor being configured to preprocess and normalize the input data and predict the bearing performance using a feedforward neural network using a backpropagation algorithm; and a data visualization interface connected to a user interface, the data visualization interface performing the visualization of the predicted output of the artificial neural network (ANN) module, and the user interface displaying the visualized data to the user.

[0006] An objective of the present disclosure is to provide an ANN-based system for evaluating hydrodynamic bearing performance.

[0007] Another objective of the present disclosure is to conduct analytical and experimental investigations of hydrodynamic bearings.

[0008] Another objective of the present disclosure is to develop an artificial neural network model trained on the parameters used for analytical and experimental hydrodynamics.

[0009] Another objective of the present disclosure is to compare the parameters predicted by the ANN with the analysis and experimental results and to observe their deviations.

[0010] To further clarify the advantages and features of the present disclosure, a more detailed description of the invention will be given with reference to specific embodiments thereof, which are illustrated in the accompanying drawings. It should be noted that these drawings represent only typical embodiments of the invention and are therefore not to be considered limiting its scope. The invention will be described and explained in additional detail and in greater detail with reference to the accompanying drawings. SHORT DESCRIPTION OF THE FIGURE

[0011] These and other features, aspects, and advantages of the present disclosure will be better understood when the following detailed description is read with reference to the accompanying drawings, in which like characters represent like parts throughout. Where: Fig.1 shows a block diagram of an ANN-based system for evaluating hydrodynamic bearing performance according to an embodiment of the present disclosure.

[0012] Furthermore, those skilled in the art will appreciate that elements in the drawings are shown for convenience and may not necessarily be drawn to scale. For example, the flowcharts illustrate the method by key steps to enhance understanding of aspects of the present disclosure. Furthermore, with respect to the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawing may show only the specific details relevant to understanding embodiments of the present disclosure in order not to clutter the drawings with details that would be readily apparent to those skilled in the art who would benefit from the description herein. DETAILED DESCRIPTION:

[0013] To facilitate an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and described in specific language. It is to be understood, however, that no limitation on the scope of the invention is intended thereby, since such changes and further modifications of the illustrated system and such further applications of the principles of the invention as illustrated therein are contemplated as would normally occur to one skilled in the art to which the invention pertains.

[0014] It will be understood by one skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be limiting thereof.

[0015] References in this specification to "one aspect," "another aspect," or similar language mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, occurrences of the phrase "in one embodiment," "in another embodiment," and similar language throughout this specification may or may not all refer to the same embodiment.

[0016] The terms "comprises," "comprising," or other variations thereof are intended to cover non-exclusive inclusion, such that a process or method comprising a list of steps not only includes those steps, but may also include other steps not expressly listed or inherent in such process or method. Likewise, one or more devices or subsystems or elements or structures or components preceded by "comprises...a" does not preclude, without further limitation, the existence of other devices or other subsystems or other elements or other structures or other components or additional devices or additional subsystems or additional elements or additional structures or additional components.

[0017] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The system, methods, and examples provided herein are for illustrative purposes only and should not be considered limiting.

[0018] Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0019] The functional units described in this specification have been referred to as devices. A device may be implemented in programmable hardware devices such as processors, digital signal processors, central processing units, field-programmable gate arrays, programmable array logic, programmable logic devices, cloud processing systems, or the like. The devices may also be implemented in software for execution by various types of processors. An identified device may contain executable code and may, for example, comprise one or more physical or logical blocks of computer instructions, which may be organized, for example, as an object, procedure, function, or other construct.However, the executable file of an identified device need not be physically located together, but may comprise different instructions stored in different locations which, when logically linked together, constitute the device and fulfill the stated purpose of the device.

[0020] Indeed, executable code of a device or module may be a single instruction or multiple instructions, and may even be distributed across several different code segments, among different applications, and across multiple storage devices. Similarly, operational data may be identified and represented within the device and embodied in any suitable form and organized in any suitable type of data structure. The operational data may be captured as a single set of data or distributed across different locations, including different storage devices, and may exist, at least in part, as electronic signals in a system or network.

[0021] References in this specification to "a selected embodiment," "an embodiment," or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosed subject matter. Therefore, the appearances of the phrases "a selected embodiment," "in an embodiment," or "in an embodiment" in various places in this specification do not necessarily refer to the same embodiment.

[0022] Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the disclosed subject matter. However, one of ordinary skill in the art will recognize that the disclosed subject matter may be practiced without one or more of the specific details, or with different methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the disclosed subject matter.

[0023] According to the example embodiments, the disclosed computer programs or modules may be executed in many example ways, for example, as an application stored in the memory of a device or as a hosted application running on a server and communicating with the device application or browser using a variety of standard protocols such as TCP / IP, HTTP, XML, SOAP, REST, JSON, and other suitable protocols. The disclosed computer programs may be written in example programming languages ​​that execute from memory on the device or from a hosted server, for example, BASIC, COBOL, C, C++, Java, Pascal, or scripting languages ​​such as JavaScript, Python, Ruby, PHP, Perl, or other suitable programming languages.

[0024] Some of the disclosed embodiments comprise or otherwise involve data transmission over a network, such as communicating various inputs or files over the network. The network may comprise, for example, one or more of the following: the Internet, wide area networks (WANs), local area networks (LANs), analog or digital wired and wireless telephone networks (e.g., a PSTN, Integrated Services Digital Network (ISDN), a cellular network, and Digital Subscriber Line (xDSL)), radio, television, cable, satellite, and / or any other delivery or tunneling mechanism for transmitting data. The network may comprise multiple networks or subnetworks, each of which may comprise, for example, a wired or wireless data path. The network may comprise a circuit-switched voice network, a packet-switched data network, or any other network capable of transmitting electronic communications.For example, the network may include networks based on Internet Protocol (IP) or Asynchronous Transfer Mode (ATM), and may support voice using, for example, VoIP, Voice over ATM, or other comparable protocols used for voice data communications. In one implementation, the network includes a cellular network configured for the exchange of text or SMS messages.

[0025] Examples of the network include, but are not limited to, a Personal Area Network (PAN), a Storage Area Network (SAN), a Home Area Network (HAN), a Campus Area Network (CAN), a Local Area Network (LAN), a Wide Area Network (WAN), a Metropolitan Area Network (MAN), a Virtual Private Network (VPN), an Enterprise Private Network (EPN), the Internet, a Global Area Network (GAN), and so on.

[0026] Fig.1 shows a block diagram of an ANN-based system (100) for evaluating hydrodynamic bearing performance according to an embodiment of the present disclosure.

[0027] With reference to Fig.1, the system (100) comprises a material selection module (102) including a fuzzy AHP sub-module (102a) configured to determine weighting factors for material selection criteria, and an ELECTRE sub-module (102b) configured to rank and select optimal bearing materials based on criteria such as hardness, yield strength, elastic modulus, thermal conductivity, density, thermal expansion, and cost; an analytical module (104) configured to simulate bearing performance with the input parameters of rotational speed and eccentricity to generate output parameters such as maximum pressure and indexing angle by using a dedicated processor configured to apply computational fluid dynamics to perform a simulation based on the input parameters;a bearing testing system (106) for conducting experimental studies based on the material selected by the material selection module and the output parameters obtained from the analytical study module, the bearing testing system (106) comprising a vertically mounted shaft, a variable-speed motor, a load application mechanism, and pressure and speed sensors; a statistical analysis module (108) configured to perform an analysis of variance (ANOVA) and regression analysis on the results obtained by the analysis module and the experiment conducted using a bearing testing system; a data acquisition unit (110) configured to obtain data on bearing materials and their properties, the data acquisition unit (110) further configured to retrieve the output data obtained from the material selection module, the analysis module, and the experiment using a bearing testing system;an artificial neural network (ANN) module (112) comprising a dedicated processor with a dedicated memory, wherein the processor is trained to predict the performance of a hydrodynamic bearing based on the input data, which is the data acquired by the data acquisition unit, the processor being configured to preprocess and normalize the input data and predict the bearing performance using a feedforward neural network using a backpropagation algorithm; and a data visualization interface (114) connected to a user interface (114a), wherein the data visualization interface (114) performs the visualization of the predicted output of the artificial neural network (ANN) module (112), and wherein the user interface (114a) displays the visualized data to the user.

[0028] In one embodiment, the material selection module (102) is further configured to apply the fuzzy AHP technique for pairwise comparison of material properties, create a weighted comparison matrix based on expert advice, and evaluate materials using the ELECTRE technique taking into account desirable and undesirable criteria, wherein the material selection module evaluates stock materials based on the following criteria: hardness, yield strength, elastic modulus, thermal conductivity, density, thermal expansion, and cost per kilogram.

[0029] In one embodiment, the bearing testing system (106) further comprises: a plain bearing assembly; a linear variable differential transformer (LVDT) for measuring eccentricity; a load cell for applying variable loads between 300 N and 600 N; a stepper motor configured to rotate the bearings up to 180 ° in 90 ° increments; an on-site pressure sensor for measuring oil film pressure distribution; a continuous lubrication system; an oil-immersed bearing assembly; and a controller for displaying the angular position of the pressure sensor.

[0030] In one embodiment, the bearing testing system (106) is connected to a data acquisition system (106a) for recording speed, pressure and indexing angle and a data transmission interface (106b) for sending experimental data to the data acquisition system (106a).

[0031] In one embodiment, the analytical module (104) is configured to: utilize ANSYS Workbench for computer simulation, model bearing performance using eccentricity as the primary input parameter, generate predictions for maximum pressure and indexing angle, and compare simulated results with experimental data.

[0032] In one embodiment, the statistical analysis module (108) is configured to perform a comparative analysis between analytical and experimental results, perform an ANOVA to evaluate the statistical significance of variations, and generate regression models to establish correlations between input and output parameters.

[0033] In one embodiment, the data acquisition unit (110) is further configured to acquire data from the following sources: literature research, market surveys, material property databases, and direct experimental measurements.

[0034] In one embodiment, the artificial neural network module (112) is further configured to normalize input data including material properties, rotation speed, and load; implement a feedforward backpropagation neural network; train the network using data from material selection, analytical, and experimental evaluations; denormalize predicted outputs for direct comparison with experimental results; and calculate performance errors between predicted and actual values.

[0035] In one embodiment, the artificial neural network module further comprises a data preprocessing module (112a) configured to standardize input data from different sources, prepare data for training the neural network, and ensure compatibility across different measurement scales.

[0036] In one embodiment, the material selection module (102), the analysis module (104), the bearing inspection system (106), the statistical analysis module (108), the data acquisition unit (110), the artificial neural network (ANN) module (112), and the data visualization interface (114) may be implemented in programmable hardware devices such as processors, digital signal processors, central processing units, field-programmable gate arrays, programmable array logic, programmable logic devices, cloud processing systems, or the like.

[0037] The present invention relates to a system for conducting analytical and experimental studies of hydrodynamic bearings, wherein the system uses an artificial neural network with a fuzzy MCDM approach. In this invention, the fuzzy AHP technique is first used to determine weighting factors of various criteria, after which the ELECTRE method is applied to determine the best material. The selected best material is further used to conduct analytical and experimental studies, wherein the input parameters load and speed are varied to obtain the output parameters maximum pressure and angle of maximum pressure. The analytical study is performed using ANSYS Fluent, and the experimental study is carried out using a plain bearing test bench. The analytical and experimental results are compared.The proposed system also performs statistical analyses, performing ANOVA and regression analyses based on analytical and experimental results. In this invention, an artificial neural network model is developed, trained with the same parameters used for the analysis and experiment, and also provides some predicted parameters. The predicted parameters of the artificial neural network are finally compared with the analytical and experimental results, and their variations are observed.

[0038] The bearing test setup is used in this invention to conduct experimental investigations. The journal bearing test rig (JBTR) equipment features a vertically mounted shaft driven by a variable-speed motor. A brass bearing is connected to the base of the setup via a metal bellows, while a friction torque load cell is located at the top of the bearing. A speed sensor attached to the driven pulley monitors the speed. The bearings are rotated by a stepper motor, which can adjust their position up to 180° in 90° increments. An integrated pressure sensor measures the oil film pressure distribution within the bearing. A lever mechanism applies radial loads using dead weights, and the journal bearing assembly is immersed in oil to ensure consistent lubrication during operation.The angular position of the pressure sensor relative to the load line, as well as the corresponding pressure and friction torque values, are displayed on a controller. These values ​​are transmitted to a PC via a data acquisition cable for further analysis. The JBTR system includes key components such as the journal assembly, bearing assembly, load assembly, lubrication system, sensors, controller, data acquisition software, and associated cables. Both the normal load and speed can be adjusted to suit different test conditions.

[0039] To train the artificial neural network, first, data of seven types of bearing materials along with their properties are collected. Then, using fuzzy AHP, some weighting factors are assigned to the criteria, and the ELECTRE method is applied to select the appropriate bearing material for the mechanical application. A literature review and market analysis were conducted to collect data, including material cost (measured in Rs. / kg of ingot). Seven inexpensive and readily available materials were selected for the material selection process. It was found during analytical and experimental studies that certain materials are susceptible to deformation due to thermal expansion. To prioritize efficiency and cost-effectiveness, thermal expansion and cost were categorized as undesirable criteria.In contrast, properties such as hardness, elastic modulus, yield strength, thermal conductivity, and density were identified as desirable, as higher values ​​of these properties improve the smoother flow of the analysis. Based on literature review and consultation with tribology experts, seven parameters were identified that are crucial for tribological studies. The selected materials were evaluated for their relevance as plain bearing materials, as well as their availability and economic feasibility. The various available bearing materials are ranked according to hardness, yield strength, elastic modulus, thermal conductivity, density, thermal expansion, and cost. To select the most suitable bearing material for the mechanical application, the following criteria must be considered, as mentioned above: 1) Hardness [H(HB)] is desirable -(C1) 2) Yield strength (Mpa) is desirable -(C2) 3) Elastic modulus (Gpa) is desirable -(C3) 4) Thermal conductivity (W / mK) is desirable - (C4) 5) Density (g / cm 3 ) is desirable -(C5) 6) Thermal expansion (µm / mK) is undesirable -(C6) 7) Costs ( / kg) are not desirable -(C7)

[0040] The seven alternatives are: 1) Leaded red brass (C83600) -(A1) 2) High Strength Brass (C86400) -(A2) 3) Lead-containing tin bronze (C92200) -(A3) 4) High Strength Brass (C86800) -(A4) 5) Tin bronze with high lead content (C93200) -(A5) 6) Tin Bronze (C90300) -(A6) 7) Silicon Brass (C87500) -(A7) Table 1: Key data from the literature review on the seven selected materials hardness Yield limit Modulus of elasticity Thermal conductivity You te Thermal expansion Cost units H(HB) MPa Grade point average W / mK g / cm3 µm / mK Rs. / Kg Type of criteria + ve + ve + ve + ve + ve - we - we Leaded red brass (C83600) 57.5 90 95 72 8.9 19 368 High-strength brass (C86400) 90 170 96 28 8.33 20 310 Lead-containing tin bronze (C92200) 65 130 96 70 8.64 18 810 High-strength brass (C86800) 160 315 100 65 8.1 19 310 Tin bronze with high lead content (C93200) 67.5 110 100 59 8.8 18 500 Tin bronze (C90300) 70 145 96 75 8.8 18 450 Silicon brass (C87500) 134 207 120 27.7 8.5 17 400

[0041] In one embodiment, a fuzzy AHP technique is used to determine the weighting factors for material selection. The pairwise comparison matrix was created after consultation with several experts and faculty members in my department, as well as my supervisor, who have previously worked with these types of materials and have the experience to evaluate them accordingly. The final weights obtained are as shown in Table 2 below. Table 2: Final weight factors used for material selection criterion C1 C2 C3 C4 C5 C6 C7 Weight factors 0.1789 0.3598 0.1651 0.1651 0.0756 0.0281 0.0271

[0042] Based on the final weight factors, ELECTRE engineering evaluates and selects optimal bearing materials based on criteria such as hardness, yield strength, elastic modulus, thermal conductivity, density, thermal expansion and cost.

[0043] Based on the selected material, analytical and experimental investigations are carried out, which are described as follows:

[0044] Table 3 below shows the details of the stock material used for analytical analysis. Table 3: Information on the bearing material Material from the magazine Cr-Ni steel Material bearing Brass material Pin diameter (D j ) 40 mm Length of the bearing 40.2 mm Scope (C) 0.189 C / R j .009 Speed ​​range 1050-1200 rpm Load (W) 300-600 N lubricant SAE 20W40 µ (kinematic viscosity) 0.0981 Pa·s

[0045] The bearing material selected in the material selection process was used in both experimental and analytical studies to analyze the pressure distribution at different loads and speeds. The input parameters for the experimental study included the bearing's rotational speed, measured in rpm, and the applied load, expressed in newtons. The experimental outputs consisted of the maximum pressure, recorded in kilopascals (kPa), and the index angle at which this maximum pressure occurred, measured in degrees. In contrast, the analytical analysis used rotational speed and eccentricity, measured in millimeters, as input parameters. The outputs of the analytical analysis mirrored those of the experimental study: maximum pressure and the corresponding index angle.Since ANSYS Fluent could not directly integrate the load as an input parameter, the bearing was designed with different eccentricities for each condition, which were calculated using data from a linear variable differential transformer (LVDT) connected to the journal bearing test bench.

[0046] The analytical study was conducted using ANSYS Workbench 16.0 following a systematic approach. First, the journal bearing was designed in the Design Modeller tool according to the specific requirements. Key sections of the model were identified and labeled, including the inlet, outlet, inner surface, and outer surface. A fine mesh was then generated to ensure accurate simulations. Boundary conditions were applied, and the material properties of the journal, bearing, and lubricant were introduced, including the properties of the SAE 20W40 lubricant used in the analysis. After the setup was complete, the solution process began with initialization, allowing the software to iteratively solve the equations governing the system. After numerous iterations, the solution converged, indicating that the results were stable and reliable.The final step was to generate a simulation based on the converged solution, the results of which were analyzed and discussed in the “Results and Discussion” section.

[0047] A journal bearing test setup is used to conduct experimental analyses. The journal is housed in a spindle supported by a bearing, with the spindle housing attached to a baseplate. An AC motor drives the journal rotation via a pulley arrangement. The brass bearing is designed with specific dimensions to maintain a uniform length-to-diameter ratio and minimize the likelihood of lateral leakage. A narrow opening in the bearing allows oil to flow into the area where a pressure sensor is mounted to measure the distributed oil film pressure.

[0048] The setup includes a load cell mounted on an index roller driven by a stepper motor operating within an angular range of 180° and moving in 90° increments. A lever mechanism is installed to apply a radial load, with a load cup attached to one end, holding weights from 300 N to 600 N. A linear variable differential transformer (LVDT) is connected to the lever to record eccentricity values ​​before the experiment begins. These eccentricity measurements will later be used to design the bearing under various load and speed conditions for analytical studies using ANSYS software.

[0049] During the test, the plain bearing assembly is immersed vertically in a beaker of lubricant to ensure continuous lubrication. Test parameters include varying loads and speeds. For each set of conditions, the system records the maximum pressure and the corresponding indexing angle at which the pressure is observed. These measurements, including speed, indexing angle, and pressure, are displayed on digital readouts connected to the test fixture.

[0050] To compare the analytical and experimental analysis results, ANOVA and regression analyses are performed for the analytical and experimental results. The results obtained from the ANOVA analysis for the experimental angle of maximum pressure show that load is the most important factor influencing this parameter. This conclusion is supported by the observation that the p-value for load is the smallest among the analyzed variables. Furthermore, the interaction of the load with itself (load-load) shows a significant influence on the experimental angle of maximum pressure, further highlighting its importance. For the experimental maximum pressure, the analysis shows that both load and speed are the most influential factors. The minimal p-values ​​associated with these terms underline their crucial role in determining the experimental maximum pressure.Furthermore, the interaction between load and load shows a notable influence on this parameter, highlighting the complex interplay of these variables. The ANOVA results for the angle of maximum pressure generated by ANSYS simulations also point to load as the predominant influencing factor. Similar to the experimental results, the p-value for load is the lowest, indicating its strong influence on the simulated angle of maximum pressure. The interaction between loads also shows a significant effect in this context, which is consistent with the experimental observations. For the maximum pressure generated by ANSYS, the analysis confirms that load remains the most influential term. The minimal p-value for load highlights its crucial role in determining the simulated maximum pressure.The interaction between loads also has a significant effect, further confirming the importance of this factor for the system's behavior.

[0051] In one embodiment, before training the ANN model using the collected data, the data is preprocessed, converting the data into an interchangeable scale for the input to train the artificial neural network. The data is preprocessed by performing normalization using equation (1) below, as it is observed that the collected data exhibits significant deviations in numerical values. y=(x−xminxmax−xmin) Where, x actual value, X max Maximum value of x, X min Minimum value of x, y normalized value corresponding to x.

[0052] The normalized input and output values ​​are then used to train the ANN model. The present invention uses a feedforward-back neural network with a backpropagation learning algorithm. The performance of the training model was verified by determining the error between the practical value and the predicted value. The following parameters were then set in the ANN toolbox: Table 4: Topology settings parameter Value Number of input variables 02 Number of output variables 02 Number of hidden levels 01 Hidden layer nodes 10 Table 5: Training settings parameter Value Training function Levenberg Marquardt (TRAINLM) Performance function Mean squared error (MSE) Transfer function Logarithmic sigmoid curves Epochs 1000 iterations Minimum gradient 1E - 11 Maximum error 600

[0053] The obtained normalized predicted outputs from the neural network model are denormalized to convert the obtained outputs to their actual size. The predicted output data are then compared with the experimental data. Normalization is performed using the following equation (2): X=(xmax−xmin)+xmin

[0054] Where y is the normalized value corresponding to x, x is the data in actual scale, x max is the maximum value of x and x min is the minimum value of x.

[0055] The observed discrepancies between analytical and experimental data can be attributed to several factors. Analytical studies conducted using software are often performed under assumed conditions that may not fully reflect real-world scenarios. In contrast, experimental studies are conducted in dynamic environments where external factors such as temperature and humidity constantly fluctuate, leading to discrepancies in the results. Furthermore, the condition of the equipment used for the experiments plays a significant role. Machinery that has been used for a long time, is inadequately maintained, or is subject to wear and tear often leads to measurement errors.

[0056] Comparison of the predicted results of the artificial neural network (ANN) with those from analytical and experimental methods showed that the ANN predictions closely match the observed results. The ANN model developed in this invention has proven accurate in its predictions.

[0057] Furthermore, the artificial neural network demonstrates the ability to effectively predict output parameters when fed with random input parameters, as long as they are within the range used to train the network. This underscores the robustness and reliability of the ANN model in replicating and extending the understanding of system behavior.

[0058] The drawings and the foregoing description provide examples of embodiments. Those skilled in the art will recognize that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be separated into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of the processes described herein may be changed and are not limited to the manner described herein. Furthermore, the actions of any flowchart need not be implemented in the order shown; nor do all actions necessarily need to be performed. Also, those actions that are not dependent on other actions may be performed in parallel with the other actions. The scope of the embodiments is in no way limited by these specific examples.Numerous variations, whether explicitly stated in the specification or not, such as differences in structure, dimensions, and use of materials, are possible. The scope of the embodiments is at least as broad as indicated in the following claims.

[0059] Advantages, other advantages, and solutions to problems have been described above with reference to specific embodiments. However, the advantages, benefits, solutions to problems, and any components that may cause an advantage or solution to occur or become more apparent are not to be construed as a critical, required, or essential feature or component of any or all of the claims. REFERENCES 100 An ANN-based system for evaluating hydrodynamic bearing performance. 102 Material selection module 102a Fuzzy AHP submodule 102b ELECTRE submodule 104 Analytical Module 106 Bearing testing system 106a Data acquisition system 106b Data transmission interface 108 Statistical Analysis Module 110 Data acquisition unit 112 Module for artificial neural networks (ANN) 112a Data preprocessing module 114 Data visualization interface 114a User interface

Claims

[1] A neural network-based system for evaluating the performance of hydrodynamic bearings, consisting of: a material selection module comprising: a fuzzy AHP sub-module configured to determine weighting factors for material selection criteria; an ELECTRE sub-module configured to evaluate and select optimal bearing materials based on criteria such as hardness, yield strength, elastic modulus, thermal conductivity, density, thermal expansion, and cost; an analysis module configured to simulate bearing performance with input parameters of speed and eccentricity to generate output parameters such as maximum pressure and indexing angle by using a dedicated processor to apply computational fluid dynamics to perform a simulation based on the input parameters; a bearing testing system for conducting experimental studies based on the material selected by the material selection module and the output parameters obtained by the analytical study module, the bearing testing system comprising a vertically mounted shaft, a variable speed motor, a load application mechanism, and pressure and speed sensors; a statistical analysis module configured to perform an analysis of variance (ANOVA) and a regression analysis of the results obtained from the analysis module and an experiment conducted using a bearing testing system; a data acquisition unit configured to acquire data about bearing materials and their properties, the data acquisition unit further configured to retrieve the output data obtained from the material selection module, the analysis module, and the experiment using a bearing testing system; an artificial neural network (ANN) module comprising a dedicated processor with a dedicated memory, the processor being trained to predict the performance of a hydrodynamic bearing based on the input data, which is the data acquired by the data acquisition unit, the processor being configured to preprocess and normalize the input data and predict the bearing performance using a feedforward neural network with backpropagation algorithm; and a data visualization interface connected to a user interface, wherein the data visualization interface performs the visualization of the predicted output of the artificial neural network (ANN) module, and wherein the user interface displays the visualized data to the user. [2] The system of claim 1, wherein the material selection module is further configured to apply the fuzzy AHP technique for pairwise comparison of material properties, create a weighted comparison matrix based on expert advice, and evaluate materials using the ELECTRE technique taking into account desirable and undesirable criteria, and wherein the material selection module evaluates bearing materials based on the following criteria: hardness, yield strength, elastic modulus, thermal conductivity, density, thermal expansion, and cost per kilogram. [3] The system of claim 1, wherein the bearing testing system further comprises: a plain bearing arrangement; a linear variable differential transformer (LVDT) for measuring eccentricity; a load cell for applying variable loads between 300 N and 600 N; a stepper motor that can rotate bearings up to 180° in 90° increments; an in-situ pressure sensor for measuring the oil film pressure distribution; a continuous lubrication system; an oil-filled bearing arrangement; and a controller for displaying the angular position of the pressure sensor; [4] The system (100) of claim 1, wherein the bearing testing system is connected to a data acquisition system for recording speed, pressure, and indexing angle, and includes a data transmission interface for sending experimental data to the data acquisition system. [5] The system of claim 1, wherein the analytical module is configured to use ANSYS Workbench for computer simulation, to model bearing performance using eccentricity as the primary input parameter, to generate predictions for maximum pressure and indexing angle, and to compare simulated results with experimental data. [6] The system of claim 1, wherein the statistical analysis module is configured to perform a comparative analysis between analytical and experimental results, to perform an ANOVA to evaluate the statistical significance of variations, and to generate regression models to establish correlations between input and output parameters. [7] The system of claim 1, wherein the data acquisition unit is further configured to obtain data from the following sources: literature searches, market surveys, material property databases, and direct experimental measurements. [8] The system of claim 1, wherein the artificial neural network module is further configured to normalize input data including material properties, rotation speed, and load; implement a feedforward backpropagation neural network; train the network using data from material selection, analytical, and experimental evaluations; denormalize predicted outputs for direct comparison with experimental results; and calculate performance errors between predicted and actual values. [9] The system of claim 1, wherein the artificial neural network module further comprises a data preprocessing module configured to standardize input data from different sources, prepare data for training the neural network, and ensure compatibility across different measurement scales.

Images