Performance test system for permanent magnet electric roller based on sensor

Abstract

The invention discloses a permanent magnet electric roller performance test system based on a sensor, which relates to the technical field of electric roller performance test and comprises a data acquisition module, a feature extraction and modeling module, a state judgment module, an atlas matching analysis module and a health assessment and early warning module. Parameters such as current, voltage, temperature, rotating speed and vibration are collected through a multi-source real-time sensor, and a roller operation data matrix is constructed; extracting a multi-dimensional feature vector group and constructing a nonlinear coupling state mapping model; calculating an operation deviation index and matching the operation deviation index with a historical working condition map to generate a full-parameter dynamic working condition fingerprint map; and finally, according to an evaluation result and a deviation degree, outputting a roller health state grade and early warning information. The system can realize accurate identification and trend analysis of the running state of the roller under multiple working conditions, and has high reliability, intelligence and application and popularization value.

Description

Technical Field

[0001] The present invention relates to the technical field of electric roller performance testing, and in particular to a sensor-based permanent magnet electric roller performance testing system. Background Art

[0002] Electric drums are drive devices that enclose an electric motor and transmission mechanism within a drum shell. They are widely used in equipment such as conveyors. Permanent magnet electric drums, in particular, have attracted widespread attention for their high efficiency, high power density, and excellent control performance. However, during actual operation, the working state of permanent magnet electric drums is affected by multiple factors such as temperature, load, current, voltage, and speed. Especially under high-frequency variable loads or complex operating conditions, problems such as reduced efficiency, abnormal temperature rise, and torque fluctuations may occur, affecting their service life and system stability.

[0003] Currently, testing the performance of permanent magnet electric rollers primarily relies on intermittent manual testing or the collection of only a single parameter. This results in poor test accuracy and real-time performance, making it difficult to implement dynamic monitoring and analysis in multiple dimensions and states, and thus unable to fully evaluate their operating performance. Therefore, a performance testing system that can integrate multiple sensor signals, acquire key operating parameters of permanent magnet electric rollers in real time, and implement intelligent, automated evaluation and data visualization is urgently needed to improve roller operation safety and maintenance efficiency. This has become a pressing technical issue to be addressed. Summary of the Invention

[0004] The purpose of the present invention is to provide a sensor-based permanent magnet electric drum performance testing system to address the deficiencies in the background technology.

[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solution: a sensor-based permanent magnet electric drum performance testing system, comprising:

[0006] The data acquisition module collects multi-source real-time operating parameters of the permanent magnet electric drum under set operating conditions, including current, voltage, temperature, speed and vibration signals, and constructs a multi-dimensional operating data matrix of the drum;

[0007] A modeling module extracts a multidimensional feature vector group of the drum operation based on the temporal variation trend of each parameter in the data matrix and the nonlinear coupling relationship between the parameters, and constructs a drum performance evaluation model. The model uses a nonlinear coupling state mapping function to establish a mapping relationship between the parameter space and the state space;

[0008] A state discrimination module compares and analyzes the drum performance evaluation model with the preset performance index boundary interval of the drum to generate a drum operation deviation index for determining whether there is an abnormal deviation in the current operation state of the drum;

[0009] The pattern matching analysis module matches the feature vector group of the current operating state with the working condition feature map of the corresponding roller model in the historical roller feature database to form a full-parameter dynamic working condition fingerprint map and output a performance evaluation result set;

[0010] The health assessment and early warning module calculates the health status level of the drum based on the assessment result set and the drum operation deviation index, and outputs status early warning information.

[0011] Preferably, in the modeling module, extracting a multidimensional feature vector group of the drum operation includes:

[0012] Construct parameter response surface plots based on multidimensional operation data matrices to identify the nonlinear evolution trajectory of each parameter over time;

[0013] The drum working condition conversion tensor is used to construct data slices of different time periods into a three-order tensor array;

[0014] Based on the local weighted reconstruction method of parameter coupling tension coefficient, the highly correlated parameter groups are dimensionally recompressed to highlight the feature contribution value;

[0015] The compressed feature subspace is normalized and phase aligned to generate a multidimensional feature vector group for drum performance evaluation modeling.

[0016] Preferably, constructing the drum performance evaluation model includes:

[0017] Based on the drum operation feature vector group, a multivariable state evolution trajectory in the parameter space is constructed;

[0018] Set the boundary clusters of typical drum working conditions and map the physical parameter range corresponding to each feature dimension to the continuous interval of the state space;

[0019] A nonlinear coupled state mapping function is established through a differentiable state transformation network to describe the joint driving effect of multi-parameter inputs on the drum state.

[0020] Preferably, the method for generating the drum operation deviation index is: extracting a normalized feature vector from the operation data: ; Where d is the feature dimension and T is the vector transpose;

[0021] Use the drum's historical normal operation data Fitting Gaussian mixture model, N is the total number of running data, assuming the model contains K mixture components, then the joint probability density function for: Where, represents the weight of the k-th Gaussian component, represents the mean vector of the kth component, represents the covariance matrix of the kth component, is a multivariate Gaussian density function; calculate the probability density of the current eigenvector , the expression is: ; Represents the currently collected drum operation feature vector; calculate the drum operation deviation index, the expression is: Where, It is the drum running deviation index.

[0022] Preferably, the pattern matching analysis module includes:

[0023] Construct a hierarchical index system for drum models and perform multi-dimensional label encoding on the historical feature database based on drum structural parameters, power levels, and typical operating conditions.

[0024] The phase correlation comparison method is used to perform time alignment processing on the current feature vector group and the candidate operating condition map;

[0025] Through the dynamic cosine similarity matrix, the local similarity curves between the current working conditions and the historical working conditions under multiple feature dimensions are calculated to form a similarity heat map;

[0026] Based on the similarity distribution morphology, a full-parameter dynamic working condition fingerprint is generated, and the performance evaluation result set is output in combination with the set threshold.

[0027] Preferably, generating a full-parameter dynamic working condition fingerprint and outputting a performance evaluation result set includes:

[0028] Based on the similarity heat map between the current feature vector group and the historical operating condition map, a multi-dimensional similarity response surface is constructed, and the principal axis curve of the response morphology is extracted as the fingerprint structure baseline;

[0029] The concave areas of the similarity response surface are locally amplified to highlight potential atypical deviation behaviors;

[0030] The extracted response morphology is mapped to the standard working condition map template corresponding to the drum model, and the final full-parameter dynamic working condition fingerprint map is formed through spatial warping matching;

[0031] Combined with the set grading discrimination threshold system, the overall deviation degree of the fingerprint spectrum is scored and a performance evaluation result set is output, including the current working condition level, evolution trend and prediction suggestions.

[0032] Preferably, the calculation of the drum health level includes:

[0033] The performance score of the current working condition of the drum ∈[0,1] is combined with the operation deviation index DRDI to construct a two-dimensional evaluation point, and it is projected to the set health level distribution interval, and the health level is set to level five;

[0034] Collect the performance scores of the first three time windows of the current cycle, and use the first-order exponential smoothing method to calculate the smoothed average score , the expression is: ;in are the rating values ​​of the first three cycles respectively;

[0035] Construct the residual function between DRDI and the scoring curve, and calculate the residual value between the current state and it : ; where f(DRDI) is the regression curve of the score and deviation index in the historical normal sample;

[0036] according to The ternary parameters composed of DRDI and residual value R are scored using the set health level decision rules, and the health level and corresponding color code are output to generate drum maintenance recommendations.

[0037] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0038] 1. This invention combines a drum performance evaluation model with a multi-dimensional health status identification mechanism, enabling real-time quantitative analysis of the operating status of permanent magnet motorized drums under complex operating conditions. Compared to traditional methods that rely on single thresholds or static rules, this invention dynamically identifies potential weak anomalies, operating condition drift, and latent degradation behaviors, improving diagnostic sensitivity and early warning accuracy, effectively reducing the risk of sudden drum failure.

[0039] 2. This invention incorporates a thermal matching algorithm based on the similarity between the full-parameter dynamic operating condition fingerprint and the historical operating condition fingerprint, enabling multi-condition, full-dimensional, and multi-cycle visual identification and trend tracking of the drum's operating status. By combining score smoothing, deviation residual analysis, and a multi-level health level output mechanism, this system supports intelligent drum maintenance scheduling and lifecycle management. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0041] Figure 1 It is a flow chart of the system modules of the present invention. DETAILED DESCRIPTION

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0043] For examples, see Figure 1 As shown, the sensor-based permanent magnet electric drum performance testing system described in this embodiment includes:

[0044] The data acquisition module collects multi-source real-time operating parameters of the permanent magnet electric drum under set operating conditions, including current, voltage, temperature, speed and vibration signals, and constructs a multi-dimensional operating data matrix of the drum;

[0045] A modeling module extracts a multidimensional feature vector group of the drum operation based on the temporal variation trend of each parameter in the data matrix and the nonlinear coupling relationship between the parameters, and constructs a drum performance evaluation model. The model uses a nonlinear coupling state mapping function to establish a mapping relationship between the parameter space and the state space;

[0046] A state discrimination module compares and analyzes the drum performance evaluation model with the preset performance index boundary interval of the drum to generate a drum operation deviation index for determining whether there is an abnormal deviation in the current operation state of the drum;

[0047] The pattern matching analysis module matches the feature vector group of the current operating state with the working condition feature map of the corresponding roller model in the historical roller feature database to form a full-parameter dynamic working condition fingerprint map and output a performance evaluation result set;

[0048] The health assessment and early warning module calculates the health status level of the drum based on the assessment result set and the drum operation deviation index, and outputs status early warning information.

[0049] In this embodiment, a multi-source data acquisition solution for permanent magnet motor performance testing is provided. The specific process is as follows:

[0050] First, before the permanent magnet electric drum is put into operation, basic sampling parameters are set according to the drum's designed speed range and load level. The system introduces a dynamic sampling frequency adjustment mechanism. This involves configuring a microcontroller unit (MCU) at the sensor access front end to adjust the sampling frequency in real time based on speed acceleration and load fluctuations. For example, when the drum enters a heavy-load startup state, the system increases the sampling frequency of current, voltage, and vibration signals to 5kHz to capture transient fluctuations. During steady-state operation, the frequency is automatically reduced to 1kHz to reduce redundant data.

[0051] Secondly, the current and voltage parameters are collected through tension-isolated sensors installed on the power supply line. These sensors use magnetic coupling to sense electromagnetic changes in the main power supply line, thereby avoiding electrical interference with the original control system and ensuring that the test system operates independently without changing the roller control circuit.

[0052] Temperature data acquisition adopts a distributed thermocouple network structure. Several thermocouples (such as K-type) are pasted or embedded on the surface of the drum shell and the bearing parts to form an axial and radial multi-point temperature monitoring array. The drum thermal field distribution is reconstructed through the difference interpolation method.

[0053] Furthermore, because the collected signals originate from asynchronous sampling channels (temperature, vibration, current, and other signals have different sampling rates), they require time-domain alignment processing in the data acquisition module. This module employs a unified timestamp management and interpolation mechanism to ensure that different signals have corresponding time tags within the same sampling period. Ultimately, a multidimensional operating data matrix with high temporal consistency is constructed, which can be directly input into the state assessment model, laying the foundation for subsequent performance analysis.

[0054] The state assessment model of this invention is constructed based on a multidimensional drum feature vector set. A "differentiable state transformation network" is used to establish a nonlinear coupled mapping between the "parameter space" and the "state space." The modeling process includes extracting multidimensional parameter responses from time series data and constructing a third-order tensor array of "time x parameter dimension x operating condition number." State mapping is learned using a continuously differentiable deep neural network (such as a feedforward network with a Reluctant Unit (ReLU) activation function). Using historical normal operating condition data as training samples, the nonlinear relationship between the combined multi-parameter input and the state response is fitted, resulting in the output of the drum's operating deviation index.

[0055] In this embodiment, based on the multi-dimensional operation data matrix of the drum, operation feature vector extraction and performance evaluation model construction are carried out, which specifically includes the following steps:

[0056] First, a parameter response surface plot was constructed based on the operating data collected from the drum under different loads and during the startup phase. Key parameters such as time, current, voltage, temperature, and vibration were converted into a three-dimensional response surface to analyze the nonlinear characteristics of each parameter over time. Observing the surface changes revealed a significant coupling delay effect between the vibration parameters and the temperature rise rate under different operating conditions.

[0057] Subsequently, to capture the coupled temporal behavior between parameters, a third-order tensor representation is used to model the operational data. The specific method is as follows:

[0058] Slice the operating data of different periods into a "parameter × time" matrix;

[0059] Stack in time order to form a third-order tensor;

[0060] By constructing a "coupling tension coefficient matrix" based on the local covariance change rate between each parameter dimension, the time delay coupling relationship between the parameters is reflected;

[0061] Based on this tensor structure, coupling delays can be identified, such as a covariance jump in vibration intensity 0.5s after a temperature rise.

[0062] Next, we move on to model construction. First, based on the multidimensional feature vectors extracted above, we use historical drum operating condition samples to define a set of typical operating condition boundary clusters, dividing each dimension into "normal," "critical," and "abnormal" state intervals. Subsequently, we construct a differentiable state transformation function network to simulate the nonlinear combined effects of multiple physical parameters under multiple operating conditions, establishing a mapping relationship from "parameter space" to "state space."

[0063] Finally, to ensure model accuracy, experimental calibration data is used to perform residual verification on the model output. That is, the difference between the model predicted state and the measured state is analyzed point by point, and the model boundary response is adjusted using the deviation weight mechanism, thus forming a complete performance evaluation model suitable for multiple working conditions of the drum.

[0064] The technical means shown in this embodiment ensure that the evaluation model has stability, interpretability and generalization capabilities under multi-parameter input conditions, providing an accurate basis for subsequent state judgment and fault prediction.

[0065] In the state discrimination module, the normalized feature vector is extracted from the running data: ; where d is the characteristic dimension, such as temperature gradient, vibration root mean square value, current fluctuation rate, etc., and T is the vector transpose.

[0066] Use the drum's historical normal operation data Fit the GMM model, N is the total number of running data, and assume that the model contains K mixed components, then the joint probability density function is: Where, represents the weight of the k-th Gaussian component, represents the mean vector of the kth component, represents the covariance matrix of the kth component, is a multivariate Gaussian density function; calculate the probability density of the current eigenvector , the expression is: ;

[0067] in Represents the currently collected drum operation feature vector, which has the same dimension as the input dimension in the training model and contains multiple normalized parameters such as temperature, current, voltage, vibration, etc. It is used to evaluate the probability density of the current state in the training model to determine whether it is a normal operating state. A higher value indicates that the current state is close to the normal area in the training model; if If the value is low, it means the current state may be abnormal. Calculate the roller operation deviation index using the expression: Where, It is the drum running deviation index.

[0068] If DRDI≈0, it means the status is close to normal; if , indicating that the state deviates seriously from the model density center, that is, a potential anomaly.

[0069] In this embodiment, for a certain model of permanent magnet electric drum (rated power 11kW, belt diameter 315mm), historical operating data is collected under multiple typical operating conditions (no-load start, full-load operation, and high-temperature environment). The data includes five dimensions: current RMS value, voltage fluctuation amplitude, shell temperature rise rate, spindle speed stability, and axial vibration RMS value.

[0070] When constructing the historical operating condition database, the present invention adopts a multi-level classification label of "model + power level + typical operating condition"; the encoding structure is: [model_ID]_[power level]_[operating condition type], such as MGD315_11kW_FL represents a full-load operating condition; the matching algorithm uses phase correlation comparison, that is, through Fourier transform, the maximum phase consistency point is found in the frequency domain, corresponding to the optimal alignment position of the time series data on the time axis.

[0071] When the drum is running in real time, the system obtains the five-dimensional eigenvalues ​​within the current time window to form a feature vector group. For example, the temperature rise rate in the current window is 1.6°C / min, and the vibration RMS is 2.1mm / s. The system calls the corresponding historical operating condition map and uses a phase-correlation matching algorithm to compare the matching degree of the feature sequence on the time axis. Specifically, after comparing the current operating condition with the historical map, a multi-dimensional cosine similarity heat map is calculated; the heat map is projected to form a response principal axis curve and divided into 67 equal-width segments; the similarity score is calculated for each segment, and the weighted average is calculated after setting the weight of each segment to output the final performance score.

[0072] The system then constructs a similarity matrix, calculating the cosine similarity between the real-time and standard profiles for each feature dimension, generating a two-dimensional heat map. For example, the average similarity between the current operating condition and the "full load" profile is 0.72, but the vibration feature is only 0.41, indicating a possible structural anomaly.

[0073] The system performs weighted projection on the similarity heat map along the characteristic dimension, extracts the main axis response curve, and aligns it with the standard working condition map. It forms a five-dimensional dynamic working condition fingerprint map through the spatial warp matching method, and records the "working condition signature" of the drum in the form of an image.

[0074] In the process of constructing the full-parameter dynamic working condition fingerprint map, the present invention adopts the following map registration and scoring process:

[0075] Calculate the five-dimensional eigenvector group of the current working condition;

[0076] Compare historical standard maps and construct a cosine similarity heat map of "current working conditions vs. standard maps";

[0077] Perform phase correlation matching on the time axis and feature axis, and use interpolation + sliding window to find the maximum similarity position of the current feature curve in the standard spectrum;

[0078] The atlas registration is completed by aligning the response surface structures of the two through spatial warping matching methods (such as dynamic time warping DTW or weighted deformation function).

[0079] The registered heat map is divided into 67 equal-width segments on the time axis;

[0080] The average similarity score is calculated for each segment and recorded as (i=1..67);

[0081] Setting segment weights (Equal by default or weighted according to the importance of the working conditions), the final total score is: ; This score is the current working condition fingerprint score (range 0~1), which is used for health level judgment and trend prediction.

[0082] During this process, if there are continuous low-similarity bands in the similarity matrix (such as the current + temperature rise is continuously low), the system automatically triggers local anomaly enhancement processing to improve the image resolution and mark suspicious areas.

[0083] The system performs an overall score on the fingerprint map and sets the status classification thresholds as follows:

[0084] Normal: score ≥0.85;

[0085] Metastable: 0.70≤score<0.85;

[0086] critical: 0.55≤score<0.70;

[0087] Abnormal: score <0.55;

[0088] The current operating condition score is 0.61, falling into the critical range. Because it's close to the scoring boundary, the system further introduces a fuzzy state factor (calculating the degree of fluctuation overlap between features) to perform interval corrections. The final output is: State Level: Critical; Major Deviation Parameters: Vibration and Temperature Rise; Trend Recommendation: Cleaning the drum heat pipes and inspecting the bearings are recommended; Graph Recording: Saving the current operating condition graph to a database for subsequent trend analysis.

[0089] The present invention tracks the changing trend of the score value sequence through a three-period sliding window, specifically in the following way:

[0090] Record the current moment score , and the scores of the first two cycles and ;

[0091] Apply first-order exponential smoothing (EWMA) to generate the trend mean: ; α∈(0,1) is the smoothing factor, and the recommended value is 0.3~0.5;

[0092] if , it indicates that the trend is worsening;

[0093] If the scores fluctuate widely but the mean is stable, the trend is considered “metastable”.

[0094] This trend can be used to predict the direction of health grade change and serve as input for drum maintenance strategy.

[0095] In this embodiment, a technical solution for grading the operating health status of a permanent magnet motor drum is provided. A multi-factor health determination system is constructed by combining the performance evaluation score, deviation index (DRDI) and score residual. The system is suitable for outputting drum status levels and supporting maintenance decisions under complex working conditions.

[0096] During the operation of the drum, the system calculates the comprehensive performance score of its current working condition in real time , ranging from 0 to 1. Scoring indicators include: temperature rise rate, current fluctuation, voltage imbalance, vibration intensity and other parameters.

[0097] At the same time, the system uses the trained Gaussian mixture model to calculate the operating deviation index DRDI of the current working condition, which reflects the degree of deviation of the drum state from the historical normal state distribution.

[0098] The system is based on ( Construct operational status evaluation points for two-dimensional input for subsequent grade mapping.

[0099] In order to suppress the influence of occasional disturbances on state judgment, the first-order exponentially weighted average (EWMA) algorithm is introduced to perform weighted processing on the last three scoring cycles and calculate the smoothed average score. , the expression is: ;in These are the rating values ​​for the previous three periods. This operation can effectively reduce the interference of short-term fluctuations on the evaluation level.

[0100] The system uses historical health samples to establish an empirical regression curve f(DRDI) between the score and DRDI, and calculates the residual value between the current state and it. , the expression is: If the residual is too large, it means that there is a structural inconsistency between the score and the deviation index, and there may be a skewed state of "appearing healthy but actually having hidden dangers".

[0101] Construction of the f(DRDI) regression function: During the system modeling phase, use the historical health condition sample set to extract the corresponding DRDI and score pairs. Use the least squares method to fit a regression curve, such as a polynomial fit or a cubic spline function: Where s, b, and c are empirical regression coefficients, representing the nonlinear change adjustment factor (s) for the deviation index, the linear decline rate coefficient (b), and the baseline score under normal operating conditions (c), respectively. This empirical regression function is obtained by performing a least squares fit on the DRDI and score pairs in historical samples. The function f is used to express the "expected change in score with the deviation index under normal distribution." If the residual R > a threshold (e.g., 0.1), it indicates that the score is inconsistent with the degree of deviation, and there may be a "normal appearance but structural hazards" operating condition.

[0102] In this embodiment, the drum health status is divided into five levels, from healthy to abnormal, namely H0, H1, H2, H3 and H4. Each level is determined by three core indicators, namely: the smoothed performance score , running deviation index DRDI and score residual R (i.e. the difference between the current score and the deviation from the expected value).

[0103] H0 (perfectly healthy): When the average performance score of the drum is higher than 0.85, the deviation index DRDI is less than 1.2, and the score residual is less than 0.05, the system determines that the drum is in a completely healthy state with extremely high operating stability and no intervention is required.

[0104] H1 (stable): When the score is between 0.75 and 0.85, the deviation index does not exceed 2.0, and the residual is less than 0.1, the system considers the drum to be basically stable and can continue to operate, but regular inspections are recommended.

[0105] Level H2 (sub-health): When the score drops to the range of 0.65 to 0.75, the deviation index is less than 3.0, and the residual value is below 0.15, the system determines that the drum may have a mild abnormality, such as the initial stage of bearing wear, rapid temperature rise, etc., and it is recommended to arrange an inspection in the short term.

[0106] H3 (critical abnormality): If the score is between 0.50 and 0.65, the deviation index is below 4.0, and the residual value is less than 0.2, it means that the drum is in a critical operating state and the abnormal characteristics are obvious. Pre-maintenance should be arranged as soon as possible.

[0107] Level H4 (Severe Abnormality): When the score is lower than 0.50, the deviation index is greater than or equal to 4.0, and the residual between the score and the model expectation exceeds 0.2, the system determines that there is a serious abnormality risk for the drum, such as continuous vibration exceeding the limit, rapid temperature rise, and uncontrolled operating condition deviation. It is recommended to immediately shut down and repair the drum.

[0108] The system automatically outputs the corresponding health level code (such as "H2") and color code (such as yellow) according to the above rules, and sends diagnostic suggestions and risk levels to the maintenance system through the early warning interface.

[0109] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. Sensor-based permanent magnet electric drum performance test system, characterized by: include: The data acquisition module collects multi-source real-time operating parameters of the permanent magnet electric drum under set operating conditions, including current, voltage, temperature, speed and vibration signals, and constructs a multi-dimensional operating data matrix of the drum; A modeling module extracts a multidimensional feature vector group of the drum operation based on the temporal variation trend of each parameter in the data matrix and the nonlinear coupling relationship between the parameters, and constructs a drum performance evaluation model. The model uses a nonlinear coupling state mapping function to establish a mapping relationship between the parameter space and the state space; A state discrimination module compares and analyzes the drum performance evaluation model with the preset performance index boundary interval of the drum to generate a drum operation deviation index for determining whether there is an abnormal deviation in the current operation state of the drum; The pattern matching analysis module matches the feature vector group of the current operating state with the working condition feature map of the corresponding roller model in the historical roller feature database to form a full-parameter dynamic working condition fingerprint map and output a performance evaluation result set; The health assessment and early warning module calculates the health status level of the drum based on the assessment result set and the drum operation deviation index, and outputs status early warning information.

2. The sensor-based permanent magnet electric drum performance testing system according to claim 1, characterized in that: In the modeling module, the multi-dimensional feature vector group of the drum operation is extracted, including: Construct parameter response surface plots based on multidimensional operation data matrices to identify the nonlinear evolution trajectory of each parameter over time; The drum working condition conversion tensor is used to construct data slices of different time periods into a third-order tensor array; Based on the local weighted reconstruction method of parameter coupling tension coefficient, the highly correlated parameter groups are dimensionally recompressed to highlight the feature contribution value; The compressed feature subspace is normalized and phase aligned to generate a multidimensional feature vector group for drum performance evaluation modeling.

3. The sensor-based permanent magnet electric drum performance testing system according to claim 2, characterized in that: The construction of the drum performance evaluation model includes: Based on the drum operation feature vector group, a multivariable state evolution trajectory in the parameter space is constructed; Set the boundary clusters of typical drum working conditions and map the physical parameter range corresponding to each feature dimension to the continuous interval of the state space; A nonlinear coupled state mapping function is established through a differentiable state transformation network to describe the joint driving effect of multi-parameter inputs on the drum state.

4. The sensor-based permanent magnet electric drum performance testing system according to claim 1, characterized in that: The method for generating the drum operation deviation index is to extract the normalized feature vector from the operation data: ; Where d is the feature dimension and T is the vector transpose; Use the drum's historical normal operation data Fitting Gaussian mixture model, N is the total number of running data, assuming the model contains K mixture components, then the joint probability density function for: Where, represents the weight of the k-th Gaussian component, represents the mean vector of the kth component, represents the covariance matrix of the kth component, is a multivariate Gaussian density function; calculate the probability density of the current eigenvector , the expression is: ; Represents the currently collected drum operation feature vector; calculate the drum operation deviation index, the expression is: Where, It is the drum running deviation index.

5. The sensor-based permanent magnet electric drum performance testing system according to claim 1, characterized in that: The graph matching analysis module includes: Construct a hierarchical index system for drum models and perform multi-dimensional label encoding on the historical feature database based on drum structural parameters, power levels, and typical operating conditions. The phase correlation comparison method is used to perform time alignment processing on the current feature vector group and the candidate operating condition map; Through the dynamic cosine similarity matrix, the local similarity curves between the current working conditions and the historical working conditions under multiple feature dimensions are calculated to form a similarity heat map; Based on the similarity distribution morphology, a full-parameter dynamic working condition fingerprint is generated, and the performance evaluation result set is output in combination with the set threshold.

6. The sensor-based permanent magnet motor performance testing system according to claim 5, characterized in that: Which generates The full-parameter dynamic working condition fingerprint and output performance evaluation result set include: Based on the similarity heat map between the current feature vector group and the historical operating condition map, a multi-dimensional similarity response surface is constructed, and the principal axis curve of the response morphology is extracted as the fingerprint structure baseline; The concave areas of the similarity response surface are locally amplified to highlight potential atypical deviation behaviors; The extracted response morphology is mapped to the standard working condition map template corresponding to the drum model, and the final full-parameter dynamic working condition fingerprint map is formed through spatial warping matching; Combined with the set grading discrimination threshold system, the overall deviation degree of the fingerprint spectrum is scored and a performance evaluation result set is output, including the current working condition level, evolution trend and prediction suggestions.

7. The sensor-based permanent magnet electric drum performance testing system according to claim 1, characterized in that: The calculation of the drum health level includes: The performance score of the current working condition of the drum ∈[0,1] is combined with the operation deviation index DRDI to construct a two-dimensional evaluation point, and it is projected to the set health level distribution interval, and the health level is set to level five; Collect the performance scores of the first three time windows of the current cycle, and use the first-order exponential smoothing method to calculate the smoothed average score , the expression is: ;in are the rating values ​​of the first three cycles respectively; Construct the residual function between DRDI and the scoring curve, and calculate the residual value between the current state and it : ; where f(DRDI) is the regression curve of the score and deviation index in the historical normal sample; according to The ternary parameters composed of DRDI and residual value R are scored using the set health level decision rules, and the health level and corresponding color code are output to generate drum maintenance recommendations.

Images