A method and system for evaluating the stability of a whole vehicle of a maglev train

Abstract

The present disclosure relates to a method and system for evaluating the ride quality of a maglev train, which can solve the problem that local suspension / guiding abnormalities are easily covered by the average index of the whole train due to insufficient coverage of the measurement points and space arrangement in the existing measurement of the ride quality of the maglev train. The present scheme includes: mapping the suspension / guiding action points to the car body measurement nodes according to the spatial arrangement and structural transmission relationship, calculating the single-point ride quality characteristic value of each node according to the effective vibration response time history of the corresponding position; based on the comprehensive weight and single-point ride quality characteristics of each car body measurement node, using a nonlinear multi-point fusion model to obtain the comprehensive index of the whole train ride quality, calculating the contribution degree of each node and identifying the local high-response suspension / guiding action area. The present invention introduces the spatial topological relationship of the suspension / guiding action points, avoids the local suspension / guiding abnormalities being covered by the average index of the whole train, and improves the accuracy of the evaluation of the ride quality of the whole maglev train and the positioning of local abnormalities.

Description

Technical Field

[0001] This disclosure relates to the field of magnetic levitation train dynamics testing and operational status evaluation technology, and more specifically, to a method and system for evaluating the overall stability of a magnetic levitation train. Background Technology

[0002] Vehicle ride comfort is a crucial indicator for evaluating the dynamic performance and ride quality of rail transit vehicles. Existing methods for evaluating rail vehicle ride comfort typically use the vibration response of a few typical measuring points in the car body or passenger compartment as input, and then obtain the overall vehicle ride comfort index through frequency weighting, time-domain statistics, or comfort evaluation models. These methods have good engineering applicability in traditional wheel-rail vehicles and can reflect the overall vibration level in the passenger area of ​​the car body.

[0003] However, maglev trains differ significantly from traditional wheel-rail vehicles in their support and guidance mechanisms. Maglev trains achieve support and guidance through the electromagnetic interactions between multiple levitation and guidance electromagnets and the track. Their vibrations are not only related to the vehicle structure itself but also closely coupled with factors such as the suspension frame, electromagnets, levitation gaps, control current, and track irregularities. For maglev vehicles, vertical, lateral, and longitudinal vibration responses can all be influenced by local excitations at multiple levitation / guidance points.

[0004] Existing methods for assessing and measuring overall vehicle stability have the following shortcomings: First, measurement points are usually concentrated in a few typical locations on the car body or in the passenger compartment, making it difficult to reflect the spatial response differences between multiple suspension points. Second, the response of typical measurement points on the car body is a comprehensive result of multiple local excitations transmitted through the suspension frame, car body structure, and suspension system. Local high responses may be smoothed during the transmission process, making it difficult to identify abnormal suspension areas. Third, existing evaluation methods focus on vibration indices in a single direction or a few directions, making it difficult to fully characterize the multi-directional vibration coupling characteristics during high-speed maglev operation. Fourth, traditional multi-measurement point fusion usually uses fixed weights or linear averaging, making it difficult to adaptively adjust based on operating speed, track conditions, suspension gap fluctuations, and sensor health status. In addition, existing maglev train vibration monitoring methods typically treat sensor measurement points as independent measurement locations, failing to incorporate the spatial topological relationships between the levitation electromagnets, guide electromagnets, and their respective suspension frames into the overall vehicle stability evaluation process. When the vehicle body position corresponding to some floating / guiding action points is limited by equipment, structure or installation space and sensors cannot be directly deployed, existing methods usually lack a virtual measurement point reconstruction mechanism based on adjacent action points, symmetrical action points or vehicle body structure transmission relationship.

[0005] Therefore, it is difficult to accurately identify the response differences in local suspension / guiding areas, abnormal connected areas, and the correlation between suspension gaps, current fluctuations, and vehicle vibration. Summary of the Invention

[0006] The purpose of this invention is to overcome the problems of existing methods for measuring the overall stability of maglev trains, such as relying on a few typical measuring points, not fully utilizing the spatial topological relationships of levitation / guidance points, insufficient spatial response coverage due to limited measuring point layout, and the ease with which local levitation / guidance anomalies are masked by the overall vehicle average index. This invention provides a method and system for evaluating the overall stability of maglev trains. The method or system incorporates the vibration response of multiple levitation points corresponding to the vehicle body measurement locations into the evaluation system, and adaptively fuses the contributions of the measuring points based on vehicle structure, vibration transmission path, and operating status, thereby more accurately reflecting the overall vehicle stability and the vibration state of local levitation systems.

[0007] To achieve the above technical objectives, this disclosure proposes a method for evaluating the overall stability of a maglev train, comprising the following steps: The levitation / guiding action points are mapped to vehicle body measurement nodes according to their spatial arrangement and structural transmission relationship; based on whether sensors can be deployed at the corresponding positions of each vehicle body measurement node, the corresponding positions are divided into actual measurement nodes and virtual measurement nodes; based on the vibration response time history of each vehicle body measurement node, the single-point stability characteristic value of each node in each vibration direction is calculated, wherein the vibration response time history of the virtual measurement node is estimated from the vibration response time history of the adjacent, symmetrical, and / or structurally transmitted force path of the actual measurement node; for each vehicle body measurement node i, based on the absolute value of the difference between the single-point stability characteristic values ​​of the i-th vehicle body measurement node and its adjacent or symmetrical nodes, the fluctuation of the levitation gap corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window, and the levitation electric field corresponding to the i-th vehicle body measurement node... The fluctuation of the current flow or guiding current within the preset time window or preset mileage window is used to calculate the adaptive weight of the operating state of the i-th vehicle body measurement node in the d-th direction. Based on the structural prior weight, operating state adaptive weight, and measurement reliability correction coefficient of each vehicle body measurement node, the comprehensive weight of each vehicle body measurement node is determined. The measurement reliability correction coefficient of the actual measurement node is determined based on the inherent accuracy level and calibration error of the sensor, while the measurement reliability correction coefficient of the virtual measurement node is determined based on the measurement reliability correction coefficient of the actual measurement node participating in the estimation of its vibration response time history, as well as at least one of historical calibration error, coherence coefficient, and transfer function fitting accuracy. Based on the comprehensive weight and single-point stability characteristics corresponding to each vehicle body measurement node, a nonlinear multi-point fusion model including a power-mean fusion term, a spatial dispersion correction term, and a local high-response enhancement term is used to obtain the overall vehicle stability index, the contribution of each node, and anomaly identification.

[0008] To achieve the above technical objectives, this disclosure proposes a magnetic levitation train overall vehicle stability testing and evaluation system, comprising: a single-point stability characteristic calculation module, configured to map the suspension / guidance action points to vehicle body measurement nodes according to spatial arrangement and structural transmission relationship; dividing the corresponding positions of each vehicle body measurement node into actual measurement nodes and virtual measurement nodes based on whether sensors can be deployed at the corresponding positions; calculating the single-point stability characteristic value of each node in each vibration direction based on the vibration response time history of each vehicle body measurement node, wherein the vibration response time history of the virtual measurement node is estimated from the vibration response time history of the adjacent, symmetrical, and / or structural force transmission path of the actual measurement node; and an operating state adaptive weight calculation module, configured to, for each vehicle body measurement node i, calculate the single-point stability characteristic value based on the absolute value of the difference between the single-point stability characteristic value of the i-th vehicle body measurement node and its adjacent or symmetrical nodes, the fluctuation amount of the suspension gap corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window, and the suspension gap corresponding to the i-th vehicle body measurement node. The fluctuation of the current or guiding current within the preset time window or preset mileage window is used to calculate the adaptive weight of the operating state of the i-th vehicle body measurement node in the d-th direction. The comprehensive weight calculation module is configured to determine the comprehensive weight of each vehicle body measurement node based on the structural prior weight, the adaptive weight of the operating state, and the measurement reliability correction coefficient. The measurement reliability correction coefficient of the actual measurement node is determined based on the inherent accuracy level and calibration error of the sensor, and the measurement reliability correction coefficient of the virtual measurement node is determined based on the measurement reliability correction coefficient of the actual measurement node whose vibration response time history is estimated, as well as at least one of historical calibration error, coherence coefficient, and transfer function fitting accuracy. The stability test evaluation and anomaly identification module is configured to obtain the overall vehicle stability index, the contribution of each node, and anomaly identification based on the comprehensive weight and single-point stability characteristics corresponding to each vehicle body measurement node, using a constructed nonlinear multi-point fusion model including a power mean fusion term, a spatial dispersion correction term, and a local high response enhancement term.

[0009] Compared with the prior art, the present invention has the following advantages.

[0010] (1) The present invention maps the suspension / guidance action points to vehicle body measurement nodes according to the spatial arrangement and structural transmission relationship, and constructs a topological model of vehicle body measurement nodes, so that the overall vehicle stability evaluation no longer depends on isolated measurement points, but can reflect the adjacent relationship, left-right symmetry relationship and front-back adjacent relationship within the same suspension frame.

[0011] (2) By combining actual measurement location with virtual measurement location, this invention can still estimate the response of the corresponding vehicle body measurement node by using the transmission relationship between adjacent nodes, symmetrical nodes and vehicle body structure when some action points cannot be directly deployed with sensors, thereby improving the spatial coverage capability under the condition of limited measurement point layout.

[0012] (3) The present invention uses the suspension gap fluctuation, suspension current fluctuation and / or guide current fluctuation corresponding to the vehicle body measurement node as associated operating status data, which is used to dynamically adjust the adaptive weight of the operating status of the vehicle body measurement node, and combined with the node contribution, topological correlation and main vibration direction, to provide a basis for distinguishing and locating sensor abnormalities, vehicle body local structure transmission abnormalities, suspension / guide control status abnormalities and line local excitation abnormalities.

[0013] (4) The present invention utilizes spatial dispersion correction term and local high response enhancement term to improve the contribution of local anomalies to the overall vehicle performance and avoids the local suspension / guidance area anomalies being masked by the multi-point averaging process.

[0014] (5) Based on the contribution of vehicle body measurement nodes and topological connectivity, the present invention identifies abnormal areas and can output the suspension frame number, node number, main vibration direction and associated operating status, which is convenient for debugging, acceptance and operation and maintenance positioning. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of a comprehensive vehicle stability evaluation process according to an embodiment of the present invention.

[0017] Figure 2 This is a structural block diagram of a vehicle stability testing and evaluation system according to an embodiment of the present invention.

[0018] Figure 3 This is a schematic diagram of the arrangement of the suspension action points corresponding to the vehicle body measurement positions according to an embodiment of the present invention. Detailed Implementation

[0019] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the components, steps, data processing methods, weighting parameters, and numerical expressions set forth in these embodiments do not limit the scope of protection of the present invention.

[0020] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0021] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0022] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0023] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0024] (I) Topology model of vehicle body measurement nodes In this disclosure, the stability measurement object is the vibration response location on the vehicle body.

[0025] During stability measurements, each suspension / guiding action point is mapped to a measurement position on the vehicle floor, vehicle load-bearing structure, or vehicle lateral structure according to its spatial arrangement and structural transmission relationship. These measurement positions are then abstracted as vehicle measurement nodes. Adjacent relationships, left-right symmetry relationships, and front-rear adjacent relationships within the same suspension frame are abstracted as edge relationships between vehicle measurement nodes. A topological model of the vehicle measurement nodes corresponding to the suspension / guiding action points is established and denoted as... V is the set of nodes, and E is the set of edges.

[0026] In the above, the levitation / guiding action point is used to characterize the position where the levitation electromagnet, the guiding electromagnet, or the levitation / guiding combined action part generates a supporting or guiding effect on the track.

[0027] The suspension / guidance action points are mapped to vehicle body measurement nodes according to their spatial arrangement and structural transmission relationship, and a topological model of the vehicle body measurement nodes is constructed. This enables the evaluation of vehicle stability to no longer rely solely on isolated measurement points, but to reflect the adjacent relationships, left-right symmetry relationships, and front-rear adjacent relationships within the same suspension frame.

[0028] (ii) Multi-point measurement Based on the topology model of the vehicle body measurement nodes corresponding to the suspension / guidance action points, the vehicle body coordinate system, the suspension frame installation reference, and the installable area of ​​the vehicle body, the sensor positions corresponding to each node are determined. When a sensor can be deployed at the position corresponding to a vehicle body measurement node, this position is taken as the actual measurement position of the vehicle body, and the vehicle body measurement node is called the actual measurement node. When a sensor cannot be deployed at the position corresponding to a vehicle body measurement node due to limitations of vehicle body structure, equipment layout, or installation space, this position is taken as the virtual measurement position of the vehicle body, and the vehicle body measurement node is called the virtual measurement node. The vibration response time history is estimated based on the adjacent, symmetrical, and / or structural force transmission paths of the vehicle body measurement nodes.

[0029] By combining actual and virtual measurement locations, when sensors cannot be directly deployed at some measurement points, the response of the corresponding vehicle body measurement node can still be estimated using adjacent nodes, symmetrical nodes, and nodes with structural force transmission paths, thereby improving the spatial coverage capability under the condition of limited measurement point layout.

[0030] The sensors used include multiple unidirectional or tridirectional accelerometers, and gyroscopes, temperature sensors, and mounting attitude detection units can be configured as needed. The levitation / guidance points correspond to the node numbers in the vehicle body measurement point topology model, which are then used to set the sensor mounting hole positions. This ensures that the sensor mounting positions for the same node remain consistent across different vehicles or test batches. An orientation-correction structure is used to restrict the sensor mounting orientation, and orientation calibration markers are used to determine the transformation relationship between the sensor coordinate system and the vehicle coordinate system. The mounting attitude detection unit acquires the sensor mounting attitude and generates coordinate transformations or mounting attitude deviations.

[0031] When using a triaxial accelerometer, the acceleration response in the sensor coordinate system can be transformed to the vehicle coordinate system based on the installation attitude detection results. In the vehicle coordinate system, the vehicle length direction is defined as longitudinal, the vehicle width direction is defined as transverse, and the direction perpendicular to the vehicle floor is defined as vertical.

[0032] All sensors synchronously acquire multi-point, multi-directional vibration response data and operational status data. Within a preset time window or preset mileage window, multi-directional vibration response data at each actual measurement location is collected synchronously, along with train operational status data corresponding to each node. Multi-directional vibration response data includes at least one of vertical, lateral, and longitudinal directions, preferably including vertical, lateral, and longitudinal acceleration. Train operational status data includes one or more of the following: train speed, mileage position, suspension gap, suspension current, guide current, electromagnet voltage, car body load status, and suspension control commands.

[0033] Synchronous acquisition methods can be achieved through hardware triggering, unified clock, train control system timestamp, network time synchronization, or mileage synchronization, so that vibration response data corresponds with operating status data such as suspension gap and current within the same time window or the same mileage window.

[0034] (III) Data Processing (3.1) Vibration response time history Vibration response preprocessing: The acquired multi-directional vibration response data undergoes one or more of the following processing methods: synchronous correction, resampling, detrending, zero-bias removal, outlier removal, bandpass filtering, frequency-weighted filtering, coordinate transformation, installation attitude correction, temperature compensation, and effective test section extraction, to obtain the vibration response time history of each actual measurement node in at least one vibration direction.

[0035] In one implementation, the vibration response time history of the virtual measurement node in the d-th vibration direction is estimated according to the following formula:

[0036] in, Let be the estimated vibration response time history value of the i-th virtual measurement node. This is the set of actual measurement nodes that are adjacent to, symmetrical to, and / or have a structural force transmission path with the i-th virtual measurement node. Let j be the vibration response time history of the j-th actual measurement node. Let be the weight corresponding to the j-th actual measurement node of the i-th virtual measurement node.

[0037] Weight It is determined based on at least one of the following: distance between nodes, relationship with the suspension frame, left-right symmetry, front-rear adjacent relationship, vehicle body structure transfer function, or historical calibration data.

[0038] For example, based on the established criteria, a non-negative response relationship index between the j-th actual measurement node and the i-th virtual measurement position in the d-th vibration direction can be obtained. .

[0039] Normalize according to the following formula: .in, It is at least one of the following: the reciprocal of the distance between nodes, the amplitude of the vehicle body structure transfer function within a preset frequency range, or the ratio of the response amplitude obtained from historical calibration data.

[0040] The weights satisfy:

[0041] During normal train operation, the vibration response time history of the virtual measurement location is generated based on the vibration response time histories of relevant actual measurement nodes that are adjacent to, symmetrical to, and / or have structural force transmission paths with the virtual measurement location, combined with the weights of each relevant actual measurement node.

[0042] (3.2) Single-point stationarity eigenvalues Calculate the single-point stability characteristic value of each vehicle body measurement node. For the vibration response time history of the i-th vehicle body measurement node in the d-th vibration direction, segment it according to a preset time window or a preset mileage window. When using a preset mileage window, determine the time interval corresponding to the mileage window based on the synchronously acquired train speed. Perform spectral analysis on the vibration response time history within each window to obtain the vibration acceleration amplitude and corresponding frequency of each frequency component within the preset evaluation frequency range.

[0043] In one implementation, a stability index calculation method based on frequency weighting and tenth-order nonlinear synthesis is employed. For the k-th frequency component of the i-th vehicle body measurement node in the d-th vibration direction, its stability component is calculated according to the following formula:

[0044] in, The stationarity component corresponding to the k-th frequency component in the d-th vibration direction of the i-th vehicle body measurement node; This represents the amplitude of the vibration acceleration at that frequency component. The corresponding frequency is expressed in Hertz (Hz). For the d-th vibration direction at frequency The corresponding frequency correction factor.

[0045] By performing a tenth-order nonlinear synthesis on each frequency component within a preset evaluation frequency range within the same window, the stationarity baseline value of the i-th vehicle body measurement node in the d-th vibration direction is obtained:

[0046] in, Let K be the stability baseline value of the i-th vehicle body measurement node in the d-th vibration direction, and K be the total number of frequency components in the preset evaluation frequency range within the corresponding window.

[0047] Considering that the stationarity baseline value mainly reflects the comprehensive vibration level of each frequency component within the corresponding window, in order to enhance the sensitivity of the single-point stationarity characteristic value to the short-time local impact response, the local impact factor of the vibration response time history within the corresponding window is further determined. :

[0048] The local impact correction is applied to the stationarity baseline value according to the following formula to obtain the single-point stationarity eigenvalue. :

[0049] in, Let T be the vibration response time history of the i-th vehicle body measurement node in the d-th vibration direction, and let T be the time length of the corresponding window. It is a positive number. This is the impact correction threshold corresponding to the d-th vibration direction. This is a non-negative impact correction factor. This corresponds to the start time of the preset time window.

[0050] (3.3) Overall weight Combination Figure 1As shown, in this embodiment, the comprehensive weight of the vehicle body measurement node is decomposed into structural prior weight, operating state adaptive weight, and measurement reliability correction coefficient.

[0051] The structural prior information includes at least one of the following: the location of the suspension frame to which the vehicle body measurement node belongs, the type of the corresponding suspension / guidance action point, the relative position of the node to the vehicle's center of gravity, the node's symmetry relationship, the vehicle body mode shape, and the vehicle body structural vibration transmission path. Based on this structural prior information, the structural prior weight of the i-th vehicle body measurement node in the d-th vibration direction is determined. Structural prior weights are used to characterize the inherent importance of measurement points in the vehicle structure. Structural prior weights can typically be obtained through various methods, such as: structural design experience; modal shapes or modal participation; finite element transfer analysis of the vehicle body; historical track test calibration; normal operation data of the same vehicle model; and expert pre-set scores.

[0052] The adaptive weight β of the operating state is used to characterize the dynamic impact of node response on the stability of the whole vehicle under different operating states. It can be determined based on the operating state information and the single-point stability characteristic value. The operating state information includes suspension gap fluctuation, suspension current fluctuation, guide current fluctuation and the difference in response of adjacent nodes.

[0053] The measurement reliability correction coefficient of the i-th vehicle body measurement node in the d-th vibration direction is determined based on the node data quality of the vehicle body measurement nodes. For actual measurement locations, node data quality is related to sensor accuracy. The measurement reliability correction coefficient is determined based on at least one of sensor bias, noise, saturation, drift, packet loss, and installation attitude deviation. For virtual measurement locations, node data quality is related to the accuracy of the sensors involved in the estimation, as well as the correlation in physical space between nodes and the accuracy of function fitting. The measurement reliability correction coefficient is determined based on the measurement reliability correction coefficient of the actual measurement nodes involved in estimating their vibration response time histories, and at least one of historical calibration error, coherence coefficient, and transfer function fitting accuracy.

[0054] In one implementation, the comprehensive weight of the i-th vehicle body measurement node in the d-th direction Determined according to the following formula:

[0055] in, Let be the structural prior weight of the i-th vehicle body measurement node in the d-th vibration direction. The adaptive weights are used to define the running state of the i-th vehicle body measurement node in the d-th direction. Let be the measurement reliability correction coefficient for the i-th vehicle body measurement node in the d-th vibration direction, with the denominator being the summation of the weight factors corresponding to all measurement points and directions. Through normalization, the weights of each measurement point in each direction can be compared within the same evaluation period.

[0056] In one implementation, the adaptive weights of the running state of the i-th vehicle body measurement node in the d-th direction are... Determined according to the following formula:

[0057] in, The fluctuation of the suspension gap corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window. The fluctuation of the levitation current or guiding current corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window. It is the absolute value of the difference between the single-point stability eigenvalues ​​of the i-th vehicle body measurement node and its adjacent or symmetrical nodes. to The coefficients are non-negative. , , This is the normalization function; in, It can be the standard deviation, peak-to-peak value, or root mean square fluctuation of the floating gap within a preset window; It can be the standard deviation, peak-to-peak value, or root mean square fluctuation of the floating current or guiding current. This can be the absolute value or normalized difference of the single-point stability characteristic value difference between the i-th vehicle body measurement node and its adjacent or symmetrical nodes. When the fluctuations in suspension gap, suspension current, guide current, or the response differences between adjacent nodes increase, the adaptive weight of the corresponding node's operating state is increased. The peak-to-peak value is the difference between the maximum and minimum signal values ​​within a preset window.

[0058] (3.4) Overall vehicle stability index A nonlinear multi-point fusion model is used to generate a comprehensive vehicle stability index. Unlike simple averaging or fixed-weight linear fusion of single-point stability feature values ​​from multiple vehicle measurement nodes, this embodiment uses a nonlinear multi-point fusion model to fuse the comprehensive weights and single-point stability feature values.

[0059] The nonlinear multi-point fusion model includes a power-means fusion term, a spatial dispersion correction term, and a local high-response enhancement term to improve the impact of local abnormal regions on the overall vehicle evaluation results and avoid the local suspension / guidance area anomalies being masked by the multi-point averaging process.

[0060] In one implementation, the overall vehicle stability index I is expressed as:

[0061] in, For direction weights, Let q be the order of the power-mean fusion of the weighted power-mean values ​​for each vehicle body measurement node in the d-th direction, and let q be the order of the power-mean fusion of the weighted power-mean values ​​for different vibration directions. , , This is a spatial dispersion correction term. For local high response enhancement terms, This is the spatial dispersion correction coefficient. This is a correction factor for local high response. It is adjusted by... , q, and It can enhance the sensitivity to local high-response suspension areas while maintaining the ability to characterize the average vibration level of the whole vehicle.

[0062] when When, the single-point stability characteristic values ​​of each vehicle body measurement node within the d-th vibration direction are fused using a weighted arithmetic average; when At this time, a larger single-point stationarity eigenvalue increases the impact on the node fusion result within the corresponding direction. Similarly, when When different vibration directions are involved, the fusion results of nodes within the same direction are fused using a weighted arithmetic mean; when At that time, the impact of the fusion results of nodes in larger directions on the overall stability index of the vehicle increases.

[0063] Spatial dispersion correction term It can be determined according to the following formula:

[0064] in

[0065] Local high response enhancement term Determined according to the following formula:

[0066] in, This represents the weighted average single-point stationarity eigenvalue in the d-th vibration direction. To prevent positive numbers with a denominator of zero, The power-mean fusion term is used to improve the impact of local large response points on the overall vehicle performance; the spatial dispersion correction term is used to characterize the unevenness of response distribution across multiple nodes; and the local high response enhancement term is used to enhance the maximum local positive deviation.

[0067] The overall vehicle stability index I is used to characterize the overall vibration stability level of the maglev train within a preset time window or preset mileage window. The larger the overall vehicle stability index I is, the stronger the vibration response of the whole vehicle and the worse the overall vehicle stability.

[0068] (iv) Vehicle stability assessment This scheme ultimately yields comprehensive vehicle stability indicators, contribution rates, and identification results of local anomaly areas. Simultaneously, it calculates the in-direction contribution of each vehicle body measurement node in each vibration direction. The contribution of the i-th measurement point in the d-th direction is also calculated. Determine according to the following formula.

[0069]

[0070] In the formula: The contribution of the i-th vehicle body measurement node to the in-direction node in the power-mean fusion term of the d-th vibration direction is given by [the term name]. Let be the comprehensive weight of the i-th vehicle body measurement node in the d-th vibration direction. The order of the node power-mean fusion is used to perform weighted power-mean fusion of all vehicle body measurement nodes within the d-th vibration direction, where N is the total number of vehicle body measurement nodes. For any vibration direction d, the sum of the node contributions within the direction of each node is 1.

[0071] The contribution of nodes within the direction is used to characterize the relative influence of each vehicle body measurement node on the power-mean fusion term of the corresponding vibration direction, and is used to screen high-contribution candidate nodes, construct candidate high-response regions, and determine the main vibration directions.

[0072] Anomaly identification based on node contribution and topological connectivity includes: determining the intra-directional node contribution in the d-th vibration direction. Vehicle body measurement nodes exceeding a first threshold within M consecutive preset time windows or preset mileage windows are identified as high-contribution candidate nodes in the d-th vibration direction; connected regions formed by at least two adjacent high-contribution candidate nodes in the same vibration direction in the topology model are identified as candidate high-response regions. For a single high-contribution candidate node, the vibration direction with the largest node contribution within its direction is identified as the main vibration direction; for a candidate high-response region R, its regional contribution in the d-th vibration direction is determined according to the following formula: Regional contribution The direction of the greatest vibration is determined as the main vibration direction of the candidate high-response region. Based on whether the high-contribution candidate nodes are isolated, the topological connectivity of the candidate high-response region, the main vibration direction, the corresponding suspension gap fluctuation, suspension current fluctuation and / or guide current fluctuation, and the mileage recurrence under different operating batches, the abnormal mode is determined, and at least one of the following is output: suspected sensor abnormality warning, vehicle body local high response warning, suspension state associated abnormality warning, guide state associated abnormality warning, and line local excitation abnormality warning.

[0073] In one implementation, when a single high-contribution candidate node corresponding to the actual measurement location is isolated and the associated operating status parameters do not exceed their respective state thresholds, a suspected sensor anomaly warning is generated; when multiple adjacent nodes within a candidate high-response area simultaneously exhibit high contributions and the associated operating status parameters do not exceed their respective state thresholds, a local high-response warning for the vehicle body is generated; when the main contributing vibration direction of the candidate high-response area is vertical and the suspension gap fluctuation and / or suspension current fluctuation exceed their respective state thresholds, a suspension state associated anomaly warning is generated; when the main contributing vibration direction is lateral and the guide current fluctuation exceeds the corresponding state threshold, a guide state associated anomaly warning is generated; when a high-contribution candidate node or candidate high-response area appears repeatedly within the same preset mileage range in different operating batches, a local excitation anomaly warning for the line is generated.

[0074] The anomaly alerts are used to assist in anomaly screening and maintenance location. Anomaly identification results include the suspension frame number, vehicle body measurement node number, main vibration direction, anomaly alert type, and corresponding train speed, suspension gap fluctuation, suspension current fluctuation, and guide current fluctuation. The cause of the anomaly can be further confirmed by combining retest results, maintenance records, and other operational status data. The first threshold, the status thresholds corresponding to each operational status parameter, and the number of consecutive windows M are determined based on historical normal operation data, line test calibration data, or vehicle acceptance requirements.

[0075] A magnetic levitation train overall stability evaluation system implemented according to the above description has the following structural block diagram: Figure 2 As shown, the system includes a topology modeling module, a measurement point mapping module, a positioning and installation component, a multi-measurement point sensor component, a synchronous acquisition component, an operational status access component, a signal processing module, a single-point feature calculation module, a weight calculation module, a vehicle index fusion module, an anomaly identification module, and an evaluation output module. Each module is described below.

[0076] The topology modeling module is configured to create a topology model of the vehicle body measurement points corresponding to the levitation / guidance action points.

[0077] The measurement point mapping module is configured to determine the actual measurement position and the virtual measurement position based on the topology model, the vehicle coordinate system, the suspension frame installation datum, and the vehicle body installation area.

[0078] The positioning and installation components include a detachable positioning template, sensor mounting holes, a direction error-proofing structure, measurement point numbering markers, and direction calibration markers. The detachable positioning template determines the position and orientation of the sensor mounting holes based on the vehicle body reference position and the vehicle body measurement node number, and is removed after sensor installation. The sensor mounting holes are set to correspond to the node numbers in the topology model of the vehicle body measurement points corresponding to the levitation / guiding action points, ensuring that the sensor installation position corresponding to the same node remains consistent across different vehicles or different test batches. The direction error-proofing structure restricts the sensor installation orientation, and the direction calibration markers determine the transformation relationship between the sensor coordinate system and the vehicle coordinate system.

[0079] The multi-point sensor assembly, including multiple unidirectional or tridirectional accelerometers, and optionally configured with gyroscopes, temperature sensors, and a mounting attitude detection unit, is designed to acquire multi-directional vibration response data at each actual measurement location. The mounting attitude detection unit is used to obtain the sensor mounting attitude and generate coordinate transformations or mounting attitude deviations.

[0080] The synchronous acquisition component is configured to perform multi-channel synchronous sampling to ensure that the multi-point sensor component and the operating status access component are on the same time reference or mileage reference.

[0081] The operational status access component is configured to acquire train speed, suspension gap, suspension current, and / or guide current.

[0082] The signal processing module is configured to generate the vibration response time history of each actual measurement node and to estimate the vibration response time history of the node corresponding to the virtual measurement location.

[0083] The single-point feature calculation module is configured to calculate the single-point stability feature value of each node in each vibration direction.

[0084] The weight calculation module is configured to calculate the overall weight based on the prior structural weights, the adaptive weights of the running state, and the measurement confidence correction coefficient.

[0085] The vehicle index fusion module is configured to generate a comprehensive vehicle stability index according to a nonlinear multi-point fusion model.

[0086] The anomaly identification module is configured to identify local high-response levitation / guiding regions based on node contribution, topological connectivity, and the correlation between levitation gap fluctuations and current fluctuations.

[0087] The evaluation output module is configured to output the overall vehicle stability index, the contribution of each node, and the anomaly identification results.

[0088] This invention also provides a magnetic levitation train stability testing and evaluation device for implementing the above-described evaluation method. The testing and evaluation device includes a multi-point sensor assembly, a positioning and installation assembly, a synchronous acquisition assembly, an operating status access assembly, an edge computing assembly, and a display and storage assembly. The edge computing assembly is used to perform real-time signal processing, single-point feature calculation, weight calculation, vehicle index fusion, and anomaly identification; the display and storage assembly is used to display and store test data and evaluation results.

[0089] In one application embodiment, the overall vehicle stability of a single-section high-speed maglev vehicle is measured. According to... Figure 3 As shown, taking the suspension action points as an example, a single-section magnetic levitation vehicle is equipped with four suspension frames along its length: suspension frame 1, suspension frame 2, suspension frame 3, and suspension frame 4. Each suspension frame corresponds to four suspension action points, and these four suspension action points correspond to four vehicle body measurement points. The four suspension frames together form vehicle body measurement nodes 1 to 16. Each measurement point is located within the vehicle body measurement position corresponding to the suspension / guidance action point, used to characterize the vehicle body vibration response near the corresponding suspension / guidance action point. A topology model containing 16 nodes is established. Each node records its suspension frame number, left / right position, front / back position, action point type, and corresponding sensor number. Adjacent nodes within the same suspension frame establish adjacent edges, corresponding left / right nodes establish left / right symmetrical edges, and nodes at the same position in the front and rear suspension frames establish front / back adjacent edges.

[0090] When a sensor cannot be installed at a node in the suspension frame 2 due to equipment obstruction, the node is set as a virtual measurement node; otherwise, the node is set as an actual measurement node.

[0091] For virtual measurement nodes, the vibration response time history of the node is generated based on the relationship between the distance between nodes and the suspension frame, the left-right symmetry relationship, the front-rear adjacent relationship, the vehicle body structure transfer function, or the weights of adjacent nodes, left-right symmetric nodes and historical calibration data within the same suspension frame.

[0092] During train operation, triaxial acceleration, velocity, suspension gap, suspension current, and guide current are simultaneously collected within a 1-second time window or a preset mileage window. After synchronous correction, filtering, frequency weighting, and coordinate transformation, the vibration response time history of each actual and virtual measurement node in the vehicle coordinate system is obtained.

[0093] Based on the vibration response time history of each vehicle body measurement node at the corresponding location, calculate the single-point stability characteristic value of each node.

[0094] Next, structural prior weights are generated based on the location of the node on the suspension frame, the type of corresponding suspension or guidance action point, the relative position of the node to the vehicle's center of gravity, the symmetry relationship of the node, and the transmission path of the vehicle structure. Adaptive weights for the operating state are generated based on suspension gap fluctuations, suspension current fluctuations, guidance current fluctuations, and differences in response between adjacent nodes. Measurement reliability correction coefficients are generated based on sensor signal quality. A comprehensive weight is then calculated based on the structural prior weights, adaptive weights for the operating state, and the measurement reliability correction coefficients.

[0095] Finally, the comprehensive weights and single-point stability eigenvalues ​​are input into the nonlinear multi-point fusion model to obtain the overall vehicle stability index I, and the node contribution is output. Nodes are sorted according to their contribution, and nodes whose contribution exceeds a first threshold within a consecutive preset time window or preset mileage window are identified as high-contribution candidate nodes. Based on the topological connectivity between high-contribution candidate nodes, at least two adjacent high-contribution candidate nodes in the same vibration direction are grouped into candidate high-response regions.

[0096] When a single high-contribution candidate node corresponding to the actual measurement location is isolated and the associated operating status parameters do not exceed their respective state thresholds, a suspected sensor anomaly warning is generated; when multiple adjacent nodes within a candidate high-response area simultaneously exhibit high contributions and the associated operating status parameters do not exceed their respective state thresholds, a local high-response warning for the vehicle body is generated; when the main contributing vibration direction of the candidate high-response area is vertical and the suspension gap fluctuation and / or suspension current fluctuation exceed their respective state thresholds, a suspension state-related anomaly warning is generated; when the main contributing vibration direction is lateral and the guide current fluctuation exceeds the corresponding state threshold, a guide state-related anomaly warning is generated; when a high-contribution candidate node or candidate high-response area appears repeatedly within the same preset mileage range in different operating batches, a local excitation anomaly warning for the line is generated.

[0097] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Without departing from the concept of the present invention, those skilled in the art can adjust or combine the number of measurement points, the arrangement of measurement points, the type of single-point index, the weight calculation method, the parameters of the fusion model, and the format of the output results, and such adjustments or combinations should all fall within the scope of protection of the present invention.

[0098] This invention can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the invention.

[0099] For example, a method for evaluating the overall stability of a maglev train includes the following steps: The levitation / guiding action points are mapped to vehicle body measurement nodes according to their spatial arrangement and structural transmission relationship; based on whether sensors can be deployed at the corresponding positions of each vehicle body measurement node, the corresponding positions are divided into actual measurement nodes and virtual measurement nodes; based on the vibration response time history of each vehicle body measurement node, the single-point stability characteristic value of each node in each vibration direction is calculated, wherein the vibration response time history of the virtual measurement node is estimated from the vibration response time history of the adjacent, symmetrical, and / or structural force transmission paths of the actual measurement node; for each vehicle body measurement node i, the method is based on the absolute value of the difference between the single-point stability characteristic values ​​of the i-th vehicle body measurement node and its adjacent or symmetrical nodes, the fluctuation of the levitation gap corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window, and the levitation current or guiding current corresponding to the i-th vehicle body measurement node. Within the preset time window or preset mileage window, the adaptive weight of the operating state of the i-th vehicle body measurement node in the d-th direction is calculated based on the fluctuation amount. The comprehensive weight of each vehicle body measurement node is determined based on its structural prior weight, adaptive operating state weight, and measurement reliability correction coefficient. The measurement reliability correction coefficient of the actual measurement node is determined based on the inherent accuracy level and calibration error of the sensor, while the measurement reliability correction coefficient of the virtual measurement node is determined based on the measurement reliability correction coefficient of the actual measurement node participating in the estimation of its vibration response time history, as well as at least one of historical calibration error, coherence coefficient, and transfer function fitting accuracy. Based on the comprehensive weight and single-point stability characteristics corresponding to each vehicle body measurement node, a nonlinear multi-point fusion model including a power-mean fusion term, a spatial dispersion correction term, and a local high-response enhancement term is used to obtain the overall vehicle stability index, the contribution of each node, and anomaly identification.

[0100] For example, a magnetic levitation train overall vehicle stability testing and evaluation system includes: a single-point stability characteristic calculation module, configured to map the suspension / guiding action points to vehicle body measurement nodes according to spatial arrangement and structural transmission relationship; classifying the corresponding positions of each vehicle body measurement node into actual measurement nodes and virtual measurement nodes based on whether sensors can be deployed at the corresponding positions; calculating the single-point stability characteristic value of each node in each vibration direction based on the vibration response time history of each vehicle body measurement node, wherein the vibration response time history of the virtual measurement node is estimated from the vibration response time history of the adjacent, symmetrical and / or actual measurement nodes with structural force transmission paths; and an operating state adaptive weight calculation module, configured to, for each vehicle body measurement node i, calculate the single-point stability characteristic value based on the absolute value of the difference between the single-point stability characteristic values ​​of the i-th vehicle body measurement node and its adjacent or symmetrical nodes, the fluctuation of the suspension gap corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window, and the suspension current or guiding current corresponding to the i-th vehicle body measurement node. The fluctuation of the current within the preset time window or preset mileage window is used to calculate the adaptive weight of the operating state of the i-th vehicle body measurement node in the d-th direction. The comprehensive weight calculation module is configured to determine the comprehensive weight of each vehicle body measurement node based on the structural prior weight, the adaptive weight of the operating state, and the measurement reliability correction coefficient. The measurement reliability correction coefficient of the actual measurement node is determined based on the inherent accuracy level and calibration error of the sensor, and the measurement reliability correction coefficient of the virtual measurement node is determined based on the measurement reliability correction coefficient of the actual measurement node whose vibration response time history is estimated, as well as at least one of historical calibration error, coherence coefficient, and transfer function fitting accuracy. The stability test evaluation and anomaly identification module is configured to obtain the overall vehicle stability index, the contribution of each node, and anomaly identification based on the comprehensive weight and single-point stability characteristics corresponding to each vehicle body measurement node, using a constructed nonlinear multi-point fusion model including a power mean fusion term, a spatial dispersion correction term, and a local high response enhancement term.

[0101] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0102] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0103] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, Python, etc., and conventional procedural programming languages ​​such as "C" or similar languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0104] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0105] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0106] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0107] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It will be known to those skilled in the art that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are equivalent.

[0108] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of the invention is defined by the appended claims.

Claims

1. A method for evaluating the overall stability of a maglev train, characterized in that, Includes the following steps: The suspension / guiding action points are mapped to vehicle body measurement nodes according to the spatial arrangement and structural transmission relationship. Based on whether sensors can be deployed at the corresponding positions of each vehicle body measurement node, the corresponding positions of the vehicle body measurement nodes are divided into actual measurement nodes and virtual measurement nodes. Based on the vibration response time history of each vehicle body measurement node, the single-point stability characteristic value of each node in each vibration direction is calculated. The vibration response time history of the virtual measurement node is estimated from the vibration response time history of the actual measurement node that is adjacent to it, symmetrical to it and / or has a structural transmission path. For each vehicle body measurement node i, the adaptive weight of the running state of the i-th vehicle body measurement node in the d-th direction is calculated based on the absolute value of the difference between the single-point stability characteristic value between the i-th vehicle body measurement node and its adjacent or symmetrical nodes, the fluctuation of the suspension gap corresponding to the i-th vehicle body measurement node in the preset time window or preset mileage window, and the fluctuation of the suspension current or guide current corresponding to the i-th vehicle body measurement node in the preset time window or preset mileage window. The comprehensive weight of each vehicle body measurement node is determined based on the structural prior weight, operating state adaptive weight, and measurement reliability correction coefficient. The measurement reliability correction coefficient of the actual measurement node is determined based on the inherent accuracy level and calibration error of the sensor, while the measurement reliability correction coefficient of the virtual measurement node is determined based on the measurement reliability correction coefficient of the actual measurement node participating in the estimation of its vibration response time history, as well as at least one of historical calibration error, coherence coefficient, and transfer function fitting accuracy. Based on the comprehensive weights and single-point stability characteristics of each vehicle body measurement node, a nonlinear multi-point fusion model including a power mean fusion term, a spatial dispersion correction term, and a local high response enhancement term is used to obtain the comprehensive stability index of the whole vehicle, the contribution of each node, and anomaly identification.

2. The method according to claim 1, characterized in that, The overall vehicle stability index is determined as follows: in, Let be the single-point stability characteristic value of the i-th node in the d-th vibration direction; For comprehensive weighting; For directional weights; Let q be the order of the power-mean fusion of the weighted power-mean values ​​for each vehicle body measurement node within the d-th vibration direction, and let q be the order of the power-mean fusion of the weighted power-mean values ​​for different vibration directions. , ; This is a spatial dispersion correction term; This is a local high-response enhancement term; and These are the spatial dispersion correction coefficient and the local high response correction coefficient, respectively. Spatial dispersion correction term Determined according to the following formula: in in, This represents the weighted average single-point stationarity eigenvalue in the d-th vibration direction. To prevent positive numbers with a denominator of zero; Local high response enhancement term Determined according to the following formula: 。 3. The method according to claim 1, characterized in that, Node contribution The calculation is as follows: In the formula: Let i be the contribution of the i-th node in the d-th vibration direction. Let be the comprehensive weight of the i-th vehicle body measurement node in the d-th direction, and N be the total number of vehicle body measurement nodes.

4. The method according to claim 1, characterized in that, The anomaly identification steps include: A topology model is constructed based on the vehicle body measurement nodes and the relationships between them. The contribution of the node in the d-th vibration direction Within M consecutive preset time windows or preset mileage windows, nodes that exceed the first threshold are identified as high contribution candidate nodes in the d-th vibration direction, and the connected region formed by at least two adjacent high contribution candidate nodes in the same vibration direction in the topology model is identified as a candidate high response region. For a single high-contribution candidate node, the vibration direction with the largest node contribution is determined as the main vibration direction; For the candidate high response region, the sum of the node contribution of each vehicle body measurement node in each vibration direction is calculated, and the vibration direction with the largest sum of node contribution is determined as the main vibration direction of the candidate high response region. The abnormal mode is determined based on whether the high-contribution candidate nodes are isolated, the topological connectivity of the candidate high-response region, the main vibration direction, the corresponding suspension gap fluctuation, suspension current fluctuation and / or guide current fluctuation, and the repeated occurrence of mileage in different operating batches. Based on the aforementioned abnormality pattern, at least one of the following is output: suspected sensor abnormality alert, vehicle body local high response alert, suspension state associated abnormality alert, guidance state associated abnormality alert, and line local excitation abnormality alert. The corresponding suspension frame number, vehicle body measurement node number, main vibration direction, and associated operating status are also output.

5. The method according to claim 1, characterized in that, The vibration response time history at the virtual measurement location is estimated using the following formula: in, Let be the estimated vibration response time history value of the i-th virtual measurement node. This is the set of actual measurement nodes that are adjacent to, symmetrical to, and / or have a structural transmission path with the i-th virtual measurement node. Let j be the vibration response time history of the j-th actual measurement node. Let be the weight corresponding to the j-th actual measurement node of the i-th virtual measurement node.

6. The method according to claim 1, characterized in that, The steps for determining the comprehensive weights corresponding to the vehicle body measurement nodes include: Based on at least one of the following: the location of the node on the suspension frame, the type of the corresponding suspension or guiding action point, the relative position of the node to the vehicle's center of gravity, the node's symmetry relationship, and the vehicle structure transmission path, determine the structural prior weight of the i-th vehicle body measurement node in the d-th vibration direction. ; Calculate the adaptive weights of the running state of the i-th vehicle body measurement node in the d-th direction. : in, The fluctuation of the suspension gap corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window. The fluctuation of the levitation current or guiding current corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window. It represents the absolute value of the difference between the single-point stability eigenvalues ​​of the i-th vehicle body measurement node and its adjacent or symmetrical nodes. to The coefficients are non-negative. , , This is the normalization function; Calculate the combined weight of the i-th node in the d-th vibration direction. : in, This is the measurement reliability correction coefficient for the i-th vehicle body measurement node in the d-th vibration direction.

7. The method according to claim 1, characterized in that, The steps for calculating the single-point stationarity eigenvalue include: The vibration response time history of the i-th vehicle body measurement node in the d-th vibration direction is segmented according to a preset time window or a preset mileage window, and the vibration response time history in each window is subjected to spectrum analysis to obtain the vibration acceleration amplitude and corresponding frequency of each frequency component within the preset evaluation frequency range. Based on the vibration acceleration amplitude of each frequency component, the corresponding frequency, and the frequency weighting correction function corresponding to the d-th vibration direction, the stationarity component of each frequency component is calculated, and the stationarity component is nonlinearly synthesized to the power of ten to obtain the stationarity base value of the i-th vehicle body measurement node in the d-th vibration direction. ; The local impact factor within the corresponding window of the vibration response time history is determined by the following formula: : The local impact correction is applied to the stationarity baseline value according to the following formula to obtain the single-point stationarity eigenvalue. : in, Let T be the vibration response time history of the i-th vehicle body measurement node in the d-th vibration direction, and let T be the time length of the corresponding window. It is a positive number. This is the impact correction threshold corresponding to the d-th vibration direction. This is the non-negative impact correction factor. This represents the start time of the corresponding window.

8. The method according to claim 4, characterized in that, The relationships between the measurement nodes of the vehicle body are adjacent relationships within the same suspension frame, left-right symmetrical relationships, and front-back adjacent relationships.

9. A system for testing and evaluating the overall stability of a maglev train, implemented according to any one of claims 1-8, characterized in that, include: The single-point stability characteristic calculation module is configured to map the suspension / guiding action point to vehicle body measurement nodes according to the spatial arrangement and structural transmission relationship. Based on whether sensors can be deployed at the corresponding positions of each vehicle body measurement node, the corresponding positions of the vehicle body measurement nodes are divided into actual measurement nodes and virtual measurement nodes. Based on the vibration response time history of each vehicle body measurement node, the single-point stability characteristic value of each node in each vibration direction is calculated. The vibration response time history of the virtual measurement node is estimated from the vibration response time history of the actual measurement node that is adjacent to it, symmetrical to it and / or has a structural force transmission path. The adaptive weight calculation module for the running state is configured to calculate the adaptive weight of the running state of the i-th vehicle body measurement node in the d-th direction for each vehicle body measurement node i, based on the absolute value of the difference between the single-point stability characteristic value between the i-th vehicle body measurement node and its adjacent or symmetrical nodes, the fluctuation of the suspension gap corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window, and the fluctuation of the suspension current or guide current corresponding to the i-th vehicle body measurement node within the preset time window or preset mileage window. The comprehensive weight calculation module is configured to determine the comprehensive weight of each vehicle body measurement node based on the structural prior weight, the adaptive weight of the operating state, and the measurement reliability correction coefficient. The measurement reliability correction coefficient of the actual measurement node is determined based on the inherent accuracy level and calibration error of the sensor, while the measurement reliability correction coefficient of the virtual measurement node is determined based on the measurement reliability correction coefficient of the actual measurement node participating in the estimation of its vibration response time history, as well as at least one of historical calibration error, coherence coefficient, and transfer function fitting accuracy. The stability test evaluation and anomaly identification module is configured to obtain the overall vehicle stability index, the contribution of each node, and anomaly identification based on the comprehensive weights and single-point stability characteristics corresponding to each vehicle body measurement node and using a constructed nonlinear multi-point fusion model including a power mean fusion term, a spatial dispersion correction term, and a local high response enhancement term.

Images