Method and system for testing vibration characteristics of a supercharger system under external excitation

Abstract

The application provides a kind of based on booster system vibration characteristic test method and system under the excitation of exogenous, comprising: based on the exogenous excitation test platform built, the multi-source sensing data of booster rotor system under the simulation of different working conditions exogenous excitation is collected;The critical state boundary division of booster rotor system is executed by adaptive state division method;Based on the critical state boundary divided, and according to the state classification model constructed in advance, the critical state of booster rotor system is classified and predicted to output vibration classification prediction result, so as to identify the critical state of booster rotor system.The application can effectively reflect the influence of exogenous excitation on the vibration characteristics of booster structure, realize the quantitative correlation between excitation input and structure response;At the same time, the critical operating state of rotor system can be effectively determined, which provides reliable experimental basis and data support for stability evaluation, fault warning and structure optimization design of booster system.

Description

Technical Field

[0001] This application belongs to the field of mechanical engineering and testing technology, and in particular relates to a method and system for testing the vibration characteristics of a turbocharger system based on external excitation. Background Technology

[0002] With the development trend of high boost, high speed, and miniaturization in internal combustion engines, the vibration problem of the turbocharger rotor system under complex operating conditions is becoming increasingly prominent. Traditional turbocharger vibration testing methods mostly rely on the self-excited response during operation, inferring the system vibration characteristics through engine surface acceleration sensors or acoustic testing methods. However, due to the large noise interference and strong uncertainty of the excitation signal in the test environment, these methods are difficult to achieve quantitative characterization of key components (such as impellers, bearings, and housings) and cannot accurately reflect the dynamic characteristics of the system under external excitation.

[0003] Currently, there is a lack of a scheme to simultaneously test the dynamic vibration response, shaft displacement response, and blade tip clearance changes of a turbocharger system under controllable external excitation during operation, which restricts the research on turbocharger vibration mechanism and structural optimization design. Summary of the Invention

[0004] In view of this, this application aims to propose a method and system for testing the vibration characteristics of a turbocharger system based on external excitation, so as to solve the problem in the prior art that it is difficult to achieve quantitative analysis of the dynamic response of blade clearance and the vibration characteristics of the whole machine through multi-sensor collaborative acquisition under external excitation.

[0005] To achieve the above objectives, the technical solution of this application is implemented as follows:

[0006] Firstly, this application provides a method for testing the vibration characteristics of a turbocharger system based on external excitation, including:

[0007] Based on the external excitation test platform, multi-source sensor data of the turbocharger rotor system under different operating conditions and external excitation simulation were collected.

[0008] The critical state boundary of the turbocharger rotor system is delineated using an adaptive state partitioning method. The adaptive state partitioning method includes constructing a nonlinear mapping relationship between external excitation parameters and the change in blade tip clearance, calculating the response sensitivity based on the fitted curve, and obtaining the preliminary interval boundaries of the stable region, transition region, and strong impact region through an optimal monotonic three-segment partitioning algorithm.

[0009] Based on the defined critical state boundaries, and according to the pre-constructed state classification model, the critical state of the turbocharger rotor system is classified and predicted to output the vibration classification prediction results, thereby identifying the critical state of the turbocharger rotor system; wherein, the state classification model is a hybrid intelligent model based on the combination of the crayfish optimization algorithm and support vector machine.

[0010] Secondly, based on the same inventive concept, this application also provides a test system for the vibration characteristics of a turbocharger system under external excitation. The test system consists of a vibration table, a turbocharger self-circulation test bench, a multi-source sensing system, a data acquisition module, and a signal processing and analysis module. The multi-source sensing system includes a blade clearance sensor and a vibration acceleration sensor.

[0011] The turbocharger is mounted on a turbocharger self-circulation test bench with a rigid support structure. The bottom of the turbocharger self-circulation test bench is bolted to the excitation table to apply external vibration excitation to the turbocharger under test.

[0012] The blade clearance sensor is installed at the compressor inlet to measure the dynamic changes in the clearance between the blade and the casing in real time; the vibration acceleration sensor is installed on the casing to collect structural vibration response signals.

[0013] The data acquisition module is used to realize multi-channel synchronous sampling and signal storage;

[0014] The signal processing and analysis module is used to execute a method for testing the vibration characteristics of a turbocharger system based on external excitation, as described in the first aspect.

[0015] Thirdly, based on the same inventive concept, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements a method for testing the vibration characteristics of a turbocharger system based on external excitation as described in the first aspect.

[0016] Fourthly, based on the same inventive concept, this application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions for causing the computer to execute a method for testing the vibration characteristics of a turbocharger system under external excitation as described in the first aspect.

[0017] Compared with existing technologies, the method and system for testing the vibration characteristics of a turbocharger system based on external excitation described in this application have the following advantages:

[0018] This application enables comprehensive, reliable, and repeatable testing of the structural response of turbochargers under different excitation intensities, providing technical support for the extraction of key dynamic parameters and identification of critical states in rotor systems. This method is applicable to experimental verification and engineering applications in the study of dynamic performance of high-speed rotating machinery. Attached Figure Description

[0019] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a flowchart of a method for testing the vibration characteristics of a turbocharger system based on external excitation, as described in an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of the fitting curve described in the embodiments of this application;

[0022] Figure 3 This is a schematic diagram illustrating the critical state division of the turbocharger under external excitation as described in the embodiments of this application;

[0023] Figure 4 This is a flowchart illustrating the vibration state classification of a turbocharger rotor system based on the COA-SVM classification model as described in an embodiment of this application.

[0024] Figure 5 This is a schematic diagram of the vibration state identification results of the turbocharger rotor system described in the embodiments of this application;

[0025] Figure 6 This is a schematic diagram of a test system for the vibration characteristics of a turbocharger system based on external excitation, as described in an embodiment of this application.

[0026] Figure 7 This is a schematic diagram of the hardware structure of the electronic device described in an embodiment of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0028] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0029] The embodiments of this application are described in detail below with reference to the accompanying drawings.

[0030] Example 1

[0031] Please see Figure 1 As shown in the figure, this embodiment provides a method for testing the vibration characteristics of a turbocharger system based on external excitation, which specifically includes the following steps:

[0032] Step S10: Based on the constructed external excitation test platform, collect multi-source sensor data of the turbocharger rotor system under different operating conditions and external excitation simulation.

[0033] Specifically, in this embodiment, a centrifugal intensifier is selected as the test object. The centrifugal intensifier is installed on a self-circulating test bench with a rigid support structure, and its bottom is connected to an electric vibration table by bolts. Three sets of triaxial vibration acceleration sensors are arranged on the housing to capture response signals at different positions. Two eddy current displacement sensors are installed at the compressor inlet to collect the displacement changes of the shaft system. A blade gap sensor is installed at the drilled hole at the tip of the compressor impeller blade to capture the transient gap change between the impeller blade and the compressor housing under high-speed rotation. All sensors are connected to the signal acquisition module to collect multi-source signals simultaneously.

[0034] During the test, the turbocharger was stabilized at the test speed by controlling the turbocharger self-circulation test bench. Then, the remote excitation control system controlled the excitation table to output excitation signals, including excitation signals under different working conditions such as sweep frequency signals and impact signals. At the same time, vibration response and blade tip clearance signals were collected.

[0035] The specific process is as follows:

[0036] (1) Fix the booster on the worktable of the excitation table to ensure its stable position and reduce the influence of external interference on the experiment; at the same time, connect the pipes of each part of the booster self-circulation experimental system to ensure that the whole system can operate normally and be tested.

[0037] (2) Before the test begins, all measuring devices and sensors are calibrated to ensure their measurement accuracy and linearity, and to ensure the accuracy and consistency of the collected data. At the same time, they are installed and connected according to the layout of each sensor.

[0038] (3) Start the excitation device on the vibration table and operate it according to the pre-set excitation conditions. Set the characteristics of different excitations, including frequency, amplitude and direction of action. According to the installation method of the turbocharger, select the positive x-axis direction to apply the excitation. According to the frequency and amplitude under different working conditions, apply the engine excitation and impact excitation in sequence to simulate the external excitation conditions under real working conditions.

[0039] (4) Start the data acquisition system and record the blade tip clearance signal, vibration acceleration signal and shaft displacement signal under different working conditions in real time;

[0040] (5) Repeat the above-mentioned working condition setting and data acquisition process to obtain test data under various external excitation conditions, so as to provide data support for subsequent research.

[0041] Step S20: Perform critical state boundary division of the turbocharger rotor system using an adaptive state partitioning method.

[0042] Specifically, in this embodiment, the step includes the following process:

[0043] (1) Based on the external excitation simulation test platform, the blade tip clearance data and shaft displacement data under different excitation intensities were collected, and the data were denoised, aligned and normalized to ensure the reliability of subsequent modeling.

[0044] (2) Using the collected data on external impact intensity and blade tip clearance, a quantitative mapping relationship between impact-induced vibration and blade tip clearance variation is constructed by fitting a model using the nonlinear least squares method. The fitting curve obtained in this step can characterize the nonlinear change law of the system approaching saturation from a linear response.

[0045] (3) Perform sensitivity calculation on the fitted model curve to obtain the instantaneous sensitivity change curve of the system under different excitation intensities. The sensitivity curve can reflect the dynamic characteristics of the system evolving from a stable response to a violent nonlinear interval, and is the core basis for state division.

[0046] (4) The sensitivity curve is treated as a one-dimensional ordered sequence. The optimal monotonic three-segment partitioning algorithm is used to automatically identify the "stable region - transition region - strong impact region" under the premise of satisfying the constraints of intra-segment stationarity, inter-segment monotonicity and minimum segment length. This process does not require manual thresholding and is based entirely on the curve shape and physical laws to achieve adaptive partitioning.

[0047] (5) Perform continuity checks, physical consistency verification and variance constraint tests on the division results to ensure that the three-segment division is reasonable in both statistical and engineering sense. Finally, extract the boundary points between the stable region and the transition region and the boundary points between the transition region and the strong impact region as the adaptive critical state thresholds of the system.

[0048] Specifically, for the turbocharger rotor system under external impact excitation, the critical state boundaries of its operation are delineated and identified based on multi-source sensor data. The dynamic behavior of the rotor system is characterized by the tip clearance response, and a threshold is set based on its numerical changes to delineate the boundaries of different operating states of the rotor system. This process establishes a clear correspondence between tip clearance and system state, providing a benchmark for subsequent state discrimination. A quantitative model between the external impact excitation intensity and tip clearance is constructed using the least squares curve fitting method. To accurately characterize the nonlinear trend of rapid response to impact and displacement tending towards saturation, this embodiment selects a saturated nonlinear function model for fitting, the mathematical expression of which is:

[0049] ;

[0050] In the formula, Indicates platform value; Indicates the initial deviation, which can make Timely satisfaction ; Indicates the attenuation rate; This represents the power parameter. The model parameters are obtained using the least squares criterion, and its optimization objective can be expressed as:

[0051] ;

[0052] In the formula, These are the impact excitation-gap change response data points collected in the experiment; N represents the data pairs. Quantity; For the model in x i The fitted value at that point.

[0053] Further error analysis was performed on the fitted curve. The specific steps were as follows: (1) The experimental data points were... Substitute the values ​​into the fitting equation to obtain the corresponding parameter variables, ensuring that the experimental points have a unique mapping on the fitting curve; (2) Obtain the spatial coordinates corresponding to the fitting curve based on the calculation; (3) Calculate the error between each node and the corresponding fitting point. Its expression is as follows:

[0054] ;

[0055] In the formula, For parameter point numbers; , These are the coordinates of the parameter points and the corresponding fitted points, respectively.

[0056] In this embodiment, based on the tip clearance displacement values ​​obtained from experiments under different external impact conditions, the displacement change is calculated with reference to normal operating conditions, and then the displacement change is fitted. Through continuous optimization, the results of each parameter are obtained. , , , The sum of squared errors is 0.000222, and the formula for the fitted curve is: ,like Figure 2 As shown.

[0057] To accurately delineate different stages, an adaptive state partitioning method based on curve sensitivity is proposed. Under external impact excitation, the steady-state response of the turbocharger rotor system can be expressed as the impact excitation intensity. The function of the impeller tip clearance response is denoted as When the system is in the weakly nonlinear region, its inherent dynamic characteristics remain essentially unchanged, and the equivalent stiffness, damping, and mass parameters of the structure are all within a stable range. Therefore, the external excitation and the system response exhibit an approximately linear mapping relationship, which can be expressed as:

[0058] ;

[0059] In the formula, This represents the equivalent sensitivity of the system in the steady-state region. This is the static response bias of the system under zero excitation. The intensity of the impact excitation.

[0060] In the stable phase, the system's dynamic characteristics remain unchanged, and the slope of the response curve changes gently and steadily. As the excitation increases, the system enters a strongly nonlinear region, significantly increasing the sensitivity of the system response to the excitation intensity. This is manifested in the curve's slope changing sharply and becoming no longer monotonic. Based on this mechanism, starting from the fitted response curve, three types of sensitivity quantification indices are introduced: the first derivative, the second derivative, and the discrete second-order difference, to characterize the system's evolution from the stable phase to the nonlinear or critical phase.

[0061] The first-order derivative sensitivity measures the local rate of change, reflecting the system's linear response to excitation; the second-order derivative sensitivity describes the curve curvature change, characterizing abrupt changes in structural dynamic characteristics. When experimental data points are limited or noisy, direct differentiation may lead to instability, amplified errors, or false abrupt changes. To improve robustness to sparse data and noisy conditions, a second-order difference is introduced based on the fitted curve to characterize the nonlinear enhancement trend of the system response with excitation. The discrete second-order difference directly reflects the "curvature" of the curve, with its peak value corresponding to the occurrence of critical states or impact behaviors. After normalization and weighted combination, a comprehensive sensitivity index reflecting the system's nonlinear enhancement can be obtained.

[0062] The formula for first-order sensitivity is:

[0063] ;

[0064] The formula for second-order sensitivity is:

[0065] ;

[0066] To characterize the response of the displacement curve to external excitation, a second-order difference is introduced to reflect the local sensitivity of the curve to change. The second-order difference formula is:

[0067] ;

[0068] To improve the detection capability of "abrupt changes in curve curvature", a discrete second-order difference term is added to the original sensitivity index.

[0069] ;

[0070] In the formula, , , Indicates weight, For the first curve The coordinates of the discrete sampling points.

[0071] This embodiment proposes a critical state partitioning method based on optimal monotonic three-segment partitioning to automate the interval division of the system response sensitivity curve under external impact excitation intensity. This method treats the sensitivity sequence as one-dimensional ordered data that increases with excitation intensity. By establishing the assumption of intra-segment stationarity, inter-segment monotonicity constraints, and a globally optimal objective function, it achieves structured analysis of the sensitivity curve. Within this framework, the system state is divided into three categories: stable segment, transition segment, and strong impact segment. The boundaries of these segments are adaptively determined by the sensitivity variation trend, without relying on human experience or preset thresholds.

[0072] The sensitivity sequence obtained from the design calculation is as follows:

[0073] ;

[0074] In the formula, S is the sensitivity sequence; S i For different orders of sensitivity.

[0075] To achieve the optimal monotonically tri-segment partition, two split points need to be found in the sequence. and Divide it into three consecutive intervals. For any interval To measure its intra-segment stationarity, a constant model is used to make an optimal approximation of the interval. In the least squares sense, the optimal constant for the interval is the sample mean of that interval. :

[0076] ;

[0077] After obtaining the optimal constant approximation for each interval, the fitting error of the three segments can be further quantified. The squared errors of the three intervals are summed to construct the cost function:

[0078] ;

[0079] The cost function essentially characterizes a given split point. , The stationarity of the sequence within the three segments.

[0080] To ensure the rationality of interval partitioning and avoid excessive oscillations due to noise, two types of constraints are introduced in the optimal three-segment partitioning process: minimum segment length constraint and monotonicity constraint. The minimum segment length constraint requires that the number of samples in each of the three intervals is no less than a preset value. This ensures that the interval mean estimation has sufficient statistical stability, avoiding the loss of physical meaning or numerical oscillations caused by excessively short intervals. The monotonicity constraint requires that the means of the three intervals satisfy... This reflects the physical law that the system's response sensitivity gradually increases with the excitation intensity from "stable-transition-strong impact".

[0081] Under the above constraints, by traversing By combining and calculating the corresponding cost function, the globally optimal partition can be obtained. The cost function is as follows:

[0082] .

[0083] After obtaining the preliminary interval boundaries of the stable region, transition region and strong impact region based on the optimal monotonic three-segment division, the consistency of the segmentation results is further verified, and the final state recognition strategy is constructed by combining the residual threshold.

[0084] First, based on the optimal split point of the three segments. , Calculate the average sensitivity value of the three intervals. Based on the physical characteristic that sensitivity gradually increases with excitation intensity, the three average values ​​correspond to the system's stable region (State=1), transition region (State=2), and strong impact region (State=3), respectively. Therefore, each excitation point can be directly assigned its corresponding state based on the interval position, realizing sequence-level state mapping.

[0085] Secondly, to ensure the reliability of the three-segment division, a residual threshold is established using the fitting residuals of the sensitivity curve, and possible local outliers are identified. The residual sequence is defined as:

[0086] ;

[0087] Where, r i For the first The residuals at each stimulus point Indicates the first The actual sensitivity value of each excitation point. These are the predicted values ​​from the global curve fitting. Calculate the statistical characteristics of the absolute values ​​of the residuals:

[0088] ;

[0089] ;

[0090] And construct the residual threshold: In the formula, σ is the average of the absolute values ​​of all residuals. r The standard deviation of the absolute value of the residuals; This is an empirical coefficient.

[0091] When certain points satisfy When this occurs, it indicates that there may be transient jumps or local excitation anomalies. However, without disrupting the overall monotonic trend, these points are only recorded as auxiliary references and do not affect the main segment structure.

[0092] Finally, based on the segmentation and residual verification results, a final state determination strategy is constructed. This strategy is applied when the three consecutive segments satisfy both the minimum segment length constraint and the monotonicity constraint of the mean. If all three intervals contain sufficient samples, then the optimal three-segment partition is directly adopted as the final state result:

[0093] ;

[0094] If a three-segment structure that satisfies the constraints cannot be found due to curve fluctuations or data noise, a backup method based on residual thresholds and clustering is used to generate state labels to ensure the robustness of the algorithm under different data conditions.

[0095] In this embodiment, the critical state division of the turbocharger under external excitation is as follows: Figure 3 As shown, 0~7.5g is the stable stage, 7.5~15g is the transition stage, and greater than 15g is the strong impact stage.

[0096] Step S30: Based on the defined critical state boundaries, and according to the pre-built state classification model, the critical state of the turbocharger rotor system is classified and predicted to output the vibration classification prediction results, thereby identifying the critical state of the turbocharger rotor system.

[0097] Specifically, in this embodiment, after completing the division of the critical operating states of the rotor system, shaft displacement signals and vibration acceleration signals are further introduced, and multi-source information fusion technology is used to jointly model and analyze the rotor system, thereby improving the accuracy and robustness of state identification.

[0098] like Figure 4 As shown, the acquired vibration acceleration and shaft displacement signals are first preprocessed. Empirical Mode Decomposition (EMD) is used to filter the signals. After filtering and frame-by-frame windowing, multi-domain feature extraction is performed. The extracted time-domain features include dimensional and dimensionless indices. Ten dimensional indices are extracted: maximum, minimum, average, root mean square (RMS), variance, standard deviation, peak value, peak-to-peak value, short-time zero-crossing rate, and short-time energy. Six dimensionless indices are extracted: skewness, kurtosis, impulse factor, peak factor, margin factor, and waveform factor. The total time-domain features are 16-dimensional. The extracted frequency-domain features include spectral kurtosis, centroid frequency, mean square frequency, RMS frequency, frequency variance, and frequency standard deviation, totaling six dimensions.

[0099] To further reduce redundant information and enhance model computational efficiency, the UMAP (Uniform Manifold Approximation and Projection) method is used to reduce the dimensionality of the extracted multidimensional features. Let the high-dimensional feature dataset be... Each data sample has a dimension of D, based on a specified number of nearest neighbors. A k-nearest neighbor (KNN) graph is constructed for each sample. Then, for each pair of adjacent samples in the graph ( Calculate the conditional probability, where, from point... Time conditional probability and from point Time conditional probability They are respectively:

[0100] ;

[0101] ;

[0102] In the formula, For point and points The distance between them; For point and points The distance between them; Indicates from data points The distance to its first nearest neighbor; Indicates from data points The distance to its first nearest neighbor; and is the scale parameter to ensure that the nearest neighbor distribution of each point has the same entropy.

[0103] UMAP replaces the perplexity parameter in t-SNE with the nearest neighbor number k. The nearest neighbor number is defined as the k nearest data points around a given data point, which directly determines the number of neighbors connected to each data point in the KNN graph.

[0104] ;

[0105] In the formula, N represents all sample points. Let be the joint probability between data point i and point j.

[0106] When constructing a high-dimensional adjacency graph, due to the local scaling parameter of each node... and The differences may lead to differences from point Time conditional probability With from point Time conditional probability Inconsistent. To ensure graph symmetry, UMAP introduces joint probabilities for weight merging:

[0107] ;

[0108] Let each high-dimensional sample be mapped to a point in the low-dimensional space. To ensure that the similarity in low-dimensional embeddings reproduces the structure of the high-dimensional adjacency graph as closely as possible, for any pair of embedding points... , Similarity is measured using the following kernel function:

[0109] ;

[0110] Among them, relative to and It is symmetrical. and This is a hyperparameter, and in UMAP the default value is... , .set up and The goal is to fit a piecewise function:

[0111] ;

[0112] Among them, when Larger and When the distance is small, the growth of the denominator is suppressed, resulting in all distances being less than 1. The points are almost completely connected, when A smaller kernel size results in a more rapid decay of distance in the kernel function, thus emphasizing the separation of distant points. The main adjustment is to scale the kernel function to smoothly transition between "local compactness" and "global distribution".

[0113] After constructing the high-dimensional and low-dimensional similarity maps, UMAP optimizes a binary cross-entropy (BCE) loss function to ensure that the low-dimensional embeddings reproduce the high-dimensional adjacency relationships as closely as possible. Specifically, let the high-dimensional similarity be... Low-dimensional similarity is Then the loss function Defined as:

[0114] , ;

[0115] Based on the above method, a 3D sensitive feature set for identifying the critical state of the turbocharger rotor system is obtained, which serves as the input for the subsequent Support Vector Machine (SVM). After data preprocessing, feature extraction, and fusion, SVM is selected to classify the data and a classification model is established. The SVM model is then used to classify and predict the critical vibration state of the turbocharger rotor system, thereby determining whether the turbocharger is safe under external impact excitation.

[0116] This embodiment employs the Crayfish Optimization Algorithm (COA) to optimize model parameters, thereby constructing a state classification model based on COA-SVM (Crayfish Optimization Algorithm and Support Vector Machine) for identifying the critical state of the turbocharger rotor system under external impact excitation. The specific process is as follows:

[0117] (1) Model parameter initialization: Set the penalty coefficient of SVM and kernel function influence coefficient The search space is defined, and these two parameters are used as the position vectors of individual crayfish in COA. Algorithm parameters are set according to the biological behavioral characteristics of COA, including population size, maximum number of iterations, and temperature threshold. Foraging probability etc., among which, temperature parameter The introduction of this method can effectively simulate the seasonal migration behavior of crayfish.

[0118] (2) Fitness evaluation: Five-fold cross-validation was used, with the classification accuracy of the SVM on the training set as the fitness function. The parameter combinations for each individual crayfish were determined. Calculate its average classification accuracy on the validation set to evaluate the quality of the parameter combination;

[0119] (3) Location update mechanism: A dynamic search strategy is established based on the ecological behavior of crayfish: when the ambient temperature... At this time, the crayfish enters its burrow to avoid predators and performs a detailed local search. At the same time, a competitive foraging strategy is adopted to balance global exploration and local development;

[0120] (4) Optimal Individual Selection: After each iteration, select the individual with the highest fitness (i.e., the parameter). and ), and retain the current optimal parameter combination;

[0121] (5) Dynamic parameter adjustment: If the number of iterations or minimum fitness is not met, return to step (2) and proceed to the next iteration;

[0122] (6) Iteration termination condition: If the preset maximum number of iterations is reached, or if the optimal solution does not improve significantly in a number of consecutive iterations, then the iteration stops;

[0123] (7) Model training and validation: using the found optimal parameters and Train the SVM model;

[0124] (8) Performance evaluation and visualization: Use the trained SVM model to make predictions on the test set to evaluate the classification performance of the model.

[0125] The vibration transition state identification results of the rotor system in this embodiment are as follows: Figure 5 As shown, the critical state identification of the turbocharger rotor system under external impact excitation is finally realized, which improves the system's operational safety and reliability, and provides technical support for the intelligent operation and maintenance and fault early warning of turbochargers.

[0126] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0127] Example 2

[0128] Based on the same inventive concept, and corresponding to any of the above embodiments, the embodiments of this application also provide a test system for the vibration characteristics of a turbocharger system under external excitation.

[0129] like Figure 6 As shown, the turbocharger system vibration characteristic testing system based on external excitation consists of an excitation table (controlled by a remote excitation control system, which outputs sinusoidal, frequency sweep, or random excitation signals, with the excitation direction being radial or axial to simulate external vibration loading conditions under different operating conditions); a turbocharger self-circulation test bench; a multi-source sensing system; a data acquisition module; and a signal processing and analysis module. The multi-source sensing system includes a blade clearance sensor and a vibration acceleration sensor.

[0130] The turbocharger is mounted on a turbocharger self-circulation test bench with a rigid support structure. The bottom of the turbocharger self-circulation test bench is bolted to the excitation table to apply external vibration excitation to the turbocharger under test.

[0131] The blade clearance sensor is installed at the compressor inlet to measure the dynamic changes in the clearance between the blades and the casing in real time; the vibration acceleration sensor is installed on the casing to collect structural vibration response signals.

[0132] The data acquisition module is used to achieve multi-channel synchronous sampling and signal storage;

[0133] The signal processing and analysis module is used to execute a method for testing the vibration characteristics of a turbocharger system based on external excitation, as described in Example 1.

[0134] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware.

[0135] The apparatus of the above embodiments is used to implement the corresponding method in any of the foregoing embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0136] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the methods described in any of the above embodiments.

[0137] Figure 7 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0138] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0139] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0140] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0141] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0142] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0143] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0144] The electronic devices described above are used to implement the corresponding methods in any of the foregoing embodiments and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0145] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to perform the methods described in any of the above embodiments.

[0146] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0147] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to perform the methods described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0148] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0149] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0150] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A method for testing the vibration characteristics of a turbocharger system based on external excitation, characterized in that, include: Based on the external excitation test platform, multi-source sensor data of the turbocharger rotor system under different operating conditions and external excitation simulation were collected. The critical state boundary delineation of the turbocharger rotor system is performed using an adaptive state delineation method. This method includes constructing a nonlinear mapping relationship between external excitation parameters and the change in blade tip clearance, calculating the response sensitivity based on the fitted curve, and obtaining preliminary boundary intervals for the stable region, transition region, and strong impact region using an optimal monotonic three-segment delineation algorithm. A nonlinear mapping relationship between external impact intensity and blade tip clearance variation was constructed by fitting a nonlinear least squares model. By performing sensitivity calculations on the fitted curves, the instantaneous sensitivity variation curves of the system under different excitation intensities can be obtained; The sensitivity change curve is treated as a one-dimensional ordered sequence. The optimal monotonic three-segment partitioning algorithm is used to identify the "stable region - transition region - strong impact region" under the premise of satisfying the constraints of intra-segment stationarity, inter-segment monotonicity and minimum segment length. Based on the defined critical state boundaries, and according to the pre-built state classification model, the critical state of the turbocharger rotor system is classified and predicted to output the vibration classification prediction results, thereby identifying the critical state of the turbocharger rotor system; wherein, the state classification model is a hybrid intelligent model based on the combination of the crayfish optimization algorithm and support vector machine. Among them, based on the fitted curve, and by introducing three types of sensitivity quantification indicators—first derivative, second derivative, and discrete second difference—the sensitivity sequence reflecting the nonlinear enhancement of the system is obtained through normalization and weighted combination. Two cutoff points are determined based on the obtained sensitivity sequence. The intervals are divided into three continuous intervals based on the location of the cutoff points. The optimal constant approximation of each interval is performed using a constant model. The squared errors of the three intervals are accumulated to construct the cost function. In the least squares sense, the optimal constant of the interval is the sample mean of the corresponding interval. Based on the minimum segment length constraint and monotonicity constraint, the globally optimal partition is obtained by traversing the two split points and calculating the corresponding cost function.

2. The method for testing the vibration characteristics of a turbocharger system based on external excitation according to claim 1, characterized in that: in, The multi-source sensing data includes at least tip clearance signal, vibration acceleration signal, and shaft displacement signal data.

3. The method for testing the vibration characteristics of a turbocharger system based on external excitation according to claim 1, characterized in that, Also includes: Based on the segmentation results, the effectiveness of the segmentation is verified and the critical state threshold is determined. The validity verification includes continuity checks, physical consistency verification, and variance constraint tests. The determination of the critical state threshold includes extracting the boundary points between the stable region and the transition region, as well as the boundary points between the transition region and the strong impact region, as adaptive critical state thresholds of the system.

4. The method for testing the vibration characteristics of a turbocharger system based on external excitation according to claim 1, characterized in that, The method for constructing the state classification model includes: The acquired vibration acceleration signal and shaft displacement signal are preprocessed, and the signals are filtered based on the empirical mode decomposition algorithm. By performing multi-domain feature extraction and fusion on the filtered, framed, and windowed signal, the high-dimensional features are reduced to a three-dimensional sensitive feature set using the UMAP dimensionality reduction algorithm. Support vector machines are used as classifiers, and crayfish optimization algorithms are introduced to optimize model parameters in order to construct a state classification model.

5. A vibration characteristic testing system for a turbocharger system based on external excitation, characterized in that: The testing system consists of a vibration table, a turbocharger self-circulation test bench, a multi-source sensing system, a data acquisition module, and a signal processing and analysis module; wherein, the multi-source sensing system includes a blade clearance sensor and a vibration acceleration sensor. The turbocharger is mounted on a turbocharger self-circulation test bench with a rigid support structure. The bottom of the turbocharger self-circulation test bench is bolted to the excitation table to apply external vibration excitation to the turbocharger under test. The blade clearance sensor is installed at the compressor inlet to measure the dynamic changes in the clearance between the blade and the casing in real time; the vibration acceleration sensor is installed on the casing to collect structural vibration response signals. The data acquisition module is used to realize multi-channel synchronous sampling and signal storage; The signal processing and analysis module is used to execute a method for testing the vibration characteristics of a turbocharger system based on external excitation as described in any one of claims 1-4.

Images