A motor fault monitoring system for a sintering trolley

Abstract

The present invention belongs to the technical field of sintering trolley, and specifically discloses a motor fault monitoring system for sintering trolley, which includes a data acquisition module, a model construction module, a hidden danger judgment module, an early warning module, a cloud collaborative module and a database. The data acquisition module collects temperature, vibration, current and speed data, and analyzes features such as temperature rise rate, spectral asymmetry, harmonic content and speed fluctuation; the model construction module extracts feature sets of winding overheating, bearing wear, current imbalance and speed anomaly based on historical fault data, and establishes an intelligent recognition model; the hidden danger judgment module analyzes and determines the fault risk according to the matching degree between the real-time operation data and the model, and the early warning module evaluates the fault type and level according to the matching degree. Through multi-source data fusion and intelligent model matching, this system realizes real-time monitoring, accurate diagnosis and active early warning of motor faults, and effectively improves the reliability of equipment operation.

Description

Technical Field

[0001] The invention relates to the technical field of sintering trolleys, and in particular to a sintering trolley motor fault monitoring system. Background Art

[0002] In the modern steel production process, sintering process is a vital link. As the core equipment of the sintering process, the sintering trolley carries the key task of mixing and pelletizing various powdered iron-containing raw materials and sintering them into bulk sintered ore at high temperature.

[0003] Due to the harsh operating environment of the sintering trolley, which is exposed to high temperature, high dust and heavy load for a long time, its motor is very prone to failure. Once the motor fails, it will not only cause the sintering trolley to stop running, but also directly affect the output and quality of sintered ore, and then cause a serious chain reaction in the entire steel production process, increasing production costs. Therefore, it is extremely important to implement effective fault monitoring and management of the sintering trolley motor.

[0004] Existing sintering trolley technologies mostly focus on the optimization and improvement of equipment structure or functional expansion to improve sintering efficiency and product quality. For example, document CN117367124A discloses a sintering trolley that can improve sintering quality. The air leakage is reduced by the matching design of the sealing groove and the driving gear disc. At the same time, a driving motor is used to drive the stirring rod to achieve mineral powder uniformity, and an exhaust fan is used to improve combustion efficiency. Although this solution optimizes the sealing performance and combustion effect of the sintering trolley, its application to the driving motor is still limited to functional execution, and does not involve real-time monitoring and fault analysis of the motor's operating status. Similarly, document CN119289690A discloses a trolley sintering furnace with an automatic discharging function, which realizes automatic discharging of products by driving the trolley body through a driving motor, and reduces impurity pollution with the help of a protective plate. However, this technology also does not monitor the health status of the driving motor, and only improves the discharging flexibility and finished product quality through structural improvements.

[0005] The above-mentioned prior art shows that the current research and development direction for sintering carts is mainly focused on mechanical structure optimization and process improvement, while there is an obvious gap in the intelligent monitoring and management of the motor operating status. Since the sintering cart is in high temperature and high load conditions for a long time, its drive motor is prone to failure due to overheating, abnormal vibration or current fluctuation, which in turn leads to equipment shutdown, reduced sintering efficiency and even production accidents. The existing technology lacks real-time collection, analysis and fault warning mechanisms for motor operating parameters, making it difficult to achieve early diagnosis and preventive maintenance of faults. Therefore, there is an urgent need for an intelligent monitoring system for sintering cart motors to make up for the shortcomings of the existing technology and improve equipment operation reliability and production continuity. Summary of the invention

[0006] In view of the above problems, the present invention proposes a sintering trolley motor fault monitoring system to realize the function of monitoring the sintering trolley.

[0007] The technical solution adopted by the present invention to solve its technical problem is: the present invention provides a sintering trolley motor fault monitoring system, including: a data acquisition module: real-time acquisition of multi-source operation data of the motor, including temperature data, vibration data, current data and speed data, and pre-processing.

[0008] Model building module: Based on the operating data of the motor's historical faults, the feature set of winding overheating, bearing wear, current imbalance, and unstable speed faults is extracted to build identification models for various faults.

[0009] Hidden danger judgment module: The real-time operation data of the motor is input into the fault identification model for matching analysis to determine whether the motor has hidden dangers of fault.

[0010] Early warning module: When there is a potential fault in the motor, the fault type is analyzed and the fault level is evaluated, and a corresponding fault early warning signal is generated.

[0011] Cloud collaboration module: encrypts and transmits the motor's multi-source operating data and fault warning signals to the cloud server, and receives fault optimization instructions issued by the cloud.

[0012] Database: Stores the operating data of the motor when it has historically failed.

[0013] Based on the above embodiment, the data acquisition module includes a temperature data acquisition unit, a vibration data acquisition unit, a current data acquisition unit and a speed data acquisition unit, wherein the specific working process of the temperature data acquisition unit is: setting the duration of the monitoring cycle, and setting each sampling time point within the monitoring cycle according to a preset time interval.

[0014] The temperature of the motor casing and three-phase winding at each sampling time point during the monitoring period is collected by an infrared temperature sensor, and a temperature change trend curve of the motor casing and three-phase winding during the monitoring period is drawn.

[0015] According to the temperature change trend curves of the motor housing and the three-phase winding during the monitoring period, the peak value of the temperature rise rate and the peak value of the temperature fluctuation are obtained respectively, and the temperature rise data and temperature fluctuation data of the motor are obtained accordingly.

[0016] The temperature change trend curves of the motor housing and the three-phase winding during the monitoring period are longitudinally compared to obtain the peak temperature difference between the motor housing and the winding and the peak temperature difference between the three-phase windings during the monitoring period to obtain the temperature distribution data of the motor.

[0017] The temperature rise data, temperature fluctuation data and temperature distribution data of the motor are summarized to obtain the temperature data of the motor.

[0018] Based on the above embodiments, the specific working process of the vibration data acquisition unit is as follows: collect the vibration spectrograms in the radial, axial, and tangential directions of the motor bearing during the monitoring period through a triaxial vibration sensor.

[0019] According to the vibration spectrogram in the radial direction of the motor bearing during the monitoring period, obtain the amplitudes of each frequency component and the phase relationship between each frequency component, and substitute them into the relationship comparison table between the amplitude of the frequency component, the phase relationship between the frequency components, and the degree of vibration asymmetry preset, and match to obtain the degree of vibration asymmetry in the radial direction of the motor bearing.

[0020] Similarly, according to the analysis method of the degree of vibration asymmetry in the radial direction of the motor bearing, obtain the degrees of vibration asymmetry in the axial and tangential directions of the motor bearing.

[0021] Construct the vibration data of the motor according to the vibration spectrograms and the degrees of vibration asymmetry in the radial, axial, and tangential directions of the motor bearing.

[0022] Based on the above embodiments, the specific working process of the current data acquisition unit is as follows: collect the three-phase current waveforms of the motor during the monitoring period through a Hall current sensor, and further obtain the harmonic content, the peak value of the instantaneous current fluctuation, and the current waveform distortion rate of the three-phase current of the motor during the monitoring period, and construct the current data of the motor.

[0023] Based on the above embodiments, the specific working process of the speed data acquisition unit is as follows: monitor the speed of the motor through an optical encoder, capture the peak value of the speed fluctuation when the motor runs at a constant speed during the monitoring period, and collect the peak value of the overshoot and the peak value of the response duration when the motor speed switches, and construct the speed data of the motor.

[0024] Compared with the prior art, the sintering trolley motor fault monitoring system of the present invention has the following beneficial effects: 1. Multi-source data real-time monitoring, improving the comprehensiveness of fault detection: The present invention can comprehensively cover the key operating parameters of the motor by collecting and preprocessing the real-time operating data of the motor temperature, vibration, current, and speed, avoiding the limitations of a single data source and ensuring the early discovery of potential faults.

[0025] 2. Accurate fault identification model driven by historical data: The present invention extracts features based on the historical fault data of the motor and constructs an identification model, which can accurately distinguish typical fault types such as winding overheating, bearing wear, current imbalance, and speed instability, improving the pertinence and accuracy of diagnosis.

[0026] 3. Dynamic matching degree analysis and hidden danger classification and early warning: The present invention adopts deviation matching degree analysis, combines preset thresholds and comprehensive fault probability mapping rules, realizes the dynamic quantitative evaluation and level division of fault hidden dangers, helps to give priority to handling high-risk faults, and reduces the risk of sudden shutdown. Brief Description of the Drawings

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0028] Figure 1 It is a system module connection diagram of the present invention.

[0029] Figure 2 It is a composition diagram of the multi-source operation data acquisition module of the present invention.

[0030] Figure 3 It is a flowchart for monitoring and analyzing the faults of the sintering trolley motor of the present invention. Detailed Embodiments

[0031] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0032] Please refer to Figure 1 and Figure 3 As shown, the present invention provides a sintering trolley motor fault monitoring system, including a data acquisition module, a model construction module, a hidden danger judgment module, an early warning module, a cloud collaboration module, and a database.

[0033] The model construction module is respectively connected to the data acquisition module and the hidden danger judgment module. The early warning module is respectively connected to the hidden danger judgment module and the cloud collaboration module. The database is connected to the model construction module.

[0034] The data acquisition module collects multi-source operation data of the motor in real time, including temperature data, vibration data, current data, and speed data, and performs preprocessing.

[0035] Further, referring to Figure 2 As shown, the data acquisition module includes a temperature data acquisition unit, a vibration data acquisition unit, a current data acquisition unit, and a speed data acquisition unit. The specific working process of the temperature data acquisition unit is as follows: set the duration of the monitoring period, and set each sampling time point within the monitoring period at a preset time interval.

[0036] The temperature of the motor casing and three-phase winding at each sampling time point during the monitoring period is collected by an infrared temperature sensor, and a temperature change trend curve of the motor casing and three-phase winding during the monitoring period is drawn.

[0037] According to the temperature change trend curves of the motor housing and the three-phase winding during the monitoring period, the peak value of the temperature rise rate and the peak value of the temperature fluctuation are obtained respectively, and the temperature rise data and temperature fluctuation data of the motor are obtained accordingly.

[0038] It should be noted that the peak values ​​of the temperature rise rates of the motor housing and the three-phase winding during the monitoring period are obtained by obtaining the temperature rise rates corresponding to the data points on the temperature change trend curve according to the temperature change trend curve of the motor housing and the three-phase winding, and comparing them with each other to obtain the peak values ​​of the temperature rise rates of the motor housing and the three-phase winding.

[0039] It should be noted that the peak value of the temperature fluctuation of the motor casing and the three-phase winding during the monitoring period is obtained by the following method: according to the temperature change trend curve of the motor casing and the three-phase winding, the temperature fluctuation of each data point on the temperature change trend curve relative to the adjacent previous data point is obtained, and the temperature fluctuations are compared with each other to obtain the peak value of the temperature fluctuation of the motor casing and the three-phase winding.

[0040] The temperature change trend curves of the motor housing and the three-phase winding during the monitoring period are longitudinally compared to obtain the peak temperature difference between the motor housing and the winding and the peak temperature difference between the three-phase windings during the monitoring period to obtain the temperature distribution data of the motor.

[0041] The temperature rise data, temperature fluctuation data and temperature distribution data of the motor are summarized to obtain the temperature data of the motor.

[0042] In a specific embodiment, the infrared temperature sensor is installed on the surface of the motor housing and the three-phase winding.

[0043] As a preferred solution, detection position points can be respectively arranged in the motor housing and the three-phase winding, and the average values ​​of the temperatures corresponding to the detection position points in the motor housing and the three-phase winding at each sampling time point during the monitoring period are calculated and used as the temperatures of the motor housing and the three-phase winding at each sampling time point during the monitoring period.

[0044] As a preferred solution, a temperature change trend curve of the motor housing and three-phase winding during the monitoring period is drawn. The specific method is: a coordinate system is established with the sampling time point as the horizontal axis and the temperature as the vertical axis. According to the temperature of the motor housing and the three-phase winding at each sampling time point during the monitoring period, the corresponding data points are marked in the coordinate system. The mathematical model establishment method is used to draw a trend curve of the temperature change of the motor housing and the three-phase winding over time during the monitoring period, and it is recorded as the temperature change trend curve of the motor housing and the three-phase winding during the monitoring period.

[0045] As a preferred solution, the peak temperature difference between the motor housing and the winding within the monitoring period is obtained. The specific method is as follows: Calculate the temperature difference between the motor housing temperature and the temperatures of each three-phase winding at the same sampling time points within the monitoring period, and then select the maximum temperature difference and record it as the temperature difference between the corresponding motor housing and the winding at the corresponding acquisition time point.

[0046] Compare the temperature differences between the motor housing and the winding at each sampling time point within the monitoring period, and select the maximum temperature difference as the peak temperature difference between the motor housing and the winding within the monitoring period.

[0047] As a preferred solution, the peak temperature difference between the three-phase windings within the monitoring period is obtained. The specific method is as follows: Compare the temperatures of the three-phase windings at each sampling time point within the monitoring period to obtain the maximum temperature difference between the three-phase windings at each sampling time point within the monitoring period, and then compare them with each other to further obtain the peak temperature difference between the three-phase windings within the monitoring period.

[0048] It should be noted that through the extraction of multi-dimensional temperature data such as the temperature rise rate, temperature fluctuation amount, and peak temperature difference, the present invention can accurately identify local overheating or abnormal heat dissipation, and avoid misjudgment of traditional single-temperature monitoring.

[0049] Furthermore, the specific working process of the vibration data acquisition unit is as follows: Collect the vibration spectrograms of the radial, axial, and tangential directions of the motor bearing within the monitoring period through a three-axis vibration sensor.

[0050] According to the vibration spectrogram of the radial direction of the motor bearing within the monitoring period, obtain the amplitude of each frequency component and the phase relationship between each frequency component, and substitute them into the relationship comparison table preset between the amplitude of the frequency component, the phase relationship between the frequency components, and the degree of vibration asymmetry to match and obtain the degree of vibration asymmetry of the radial vibration of the motor bearing.

[0051] Similarly, according to the analysis method of the degree of vibration asymmetry of the radial vibration of the motor bearing, obtain the degrees of vibration asymmetry of the axial and tangential vibrations of the motor bearing.

[0052] Construct the vibration data of the motor based on the vibration spectrograms and the degrees of vibration asymmetry of the radial, axial, and tangential directions of the motor bearing.

[0053] It should be noted that the three-axis vibration sensor is installed on the motor bearing seat.

[0054] It should be noted that the asymmetry of the radial, axial, and tangential vibrations of the motor bearing refers to the uneven and asymmetric characteristics of the radial, axial, and tangential vibrations of the motor bearing in different directions during the operation of the motor bearing.

[0055] It should be noted that in the case of asymmetric radial vibration of the motor bearing, the vibration intensities in different directions are different, which will cause changes in the amplitudes of some frequency components in the vibration spectrum. For example, in the direction with stronger vibration, the amplitudes of the relevant frequency components will increase significantly, while in the direction with weaker vibration, the amplitudes are relatively small. By comparing the differences in the spectrum amplitudes in different directions, the degree of asymmetry of the radial vibration can be intuitively understood.

[0056] It should be noted that when the radial vibration of the motor bearing is asymmetric, the phases of the vibrations in different directions may change, which will also be reflected in the spectrum. For example, during normal symmetric vibration, the phases of the vibrations in each direction may have a certain regularity, while when asymmetric vibration occurs, the phase relationship becomes complex. By analyzing the changes in the phases, the characteristics of the radial vibration asymmetry can be further understood in depth.

[0057] It should be noted that by using the three-axis vibration spectrum diagram and the analysis of the degree of asymmetry, the present invention can capture the subtle characteristics of bearing wear and improve the early warning ability of bearing faults.

[0058] Furthermore, the specific working process of the current data acquisition unit is as follows: the three-phase current waveforms of the motor during the monitoring period are collected through Hall current sensors, and further the harmonic content, the peak value of the instantaneous current fluctuation, and the current waveform distortion rate of the three-phase current of the motor during the monitoring period are obtained to construct the current data of the motor.

[0059] It should be noted that the Hall current sensor is integrated in the motor power supply circuit.

[0060] Furthermore, the specific working process of the rotational speed data acquisition unit is as follows: the rotational speed of the motor is monitored through an optical encoder, the peak value of the rotational speed fluctuation during the constant-speed operation of the motor in the monitoring period is captured, and the peak values of the overshoot and the response duration during the speed switching of the motor are collected to construct the rotational speed data of the motor.

[0061] It should be noted that the optical encoder is connected to the motor shaft.

[0062] It should be noted that the constant-speed operation of the motor means that the motor operates stably at a certain constant speed.

[0063] It should be noted that the peak value of the overshoot refers to the percentage of the maximum difference between the actual speed and the target speed to the target speed during the speed switching process of the motor.

[0064] It should be noted that the peak value of the response duration refers to the longest time experienced by the motor from the issuance of the speed switching command until the actual speed reaches the target speed and stabilizes within a certain error range.

[0065] As a preferred solution, the preprocessing of the multi-source operation data of the motor refers to the preprocessing operations of denoising, normalizing, and timestamp alignment for the multi-source operation data.

[0066] It should be noted that the present invention effectively identifies current imbalance and abnormal speed by collecting dynamic parameters such as harmonic content, waveform distortion rate, and speed fluctuation amount, enhancing the adaptability to complex working conditions.

[0067] It should be noted that through the real-time collection and preprocessing of multi-source operation data such as temperature, vibration, current, and speed, the system of the present invention can comprehensively cover the key operation parameters of the motor, avoid the limitations of a single data source, and ensure the early detection of potential faults.

[0068] The model construction module extracts the feature sets of winding overheating, bearing wear, current imbalance, and speed instability faults based on the operation data during the historical faults of the motor, and constructs the identification models for various faults.

[0069] Furthermore, the specific working process of the model construction module is as follows: extract the operation data during the historical faults of the motor stored in the database, and screen the temperature data during the historical winding overheating of the motor, the vibration data during the bearing wear, the current data during the current imbalance, and the speed data during the speed instability.

[0070] According to the temperature data during the historical winding overheating of the motor, obtain the peak values of the temperature rise rate, the peak value of the temperature fluctuation amount, the peak value of the temperature difference between the motor housing and the winding, and the peak value of the temperature difference between the three-phase windings during the historical winding overheating of the motor, obtain the feature set of the winding overheating fault, and construct an identification model for the winding overheating fault with the fault feature as the input and the fault type as the output.

[0071] Similarly, construct the identification models for bearing wear, current imbalance, and speed instability faults.

[0072] It should be noted that the construction process of the motor winding overheating fault identification model is as follows: S1. Data collection and annotation; collect the historical temperature data during the operation of the motor, including the data in the normal state and the winding overheating fault state, where the data includes the timestamp, the motor housing temperature, the three-phase winding temperature, and the corresponding fault type label.

[0073] S2. Feature extraction: Extract the following key features from the temperature data: the peak value of the temperature rise rate, the peak value of the temperature fluctuation amount, the peak value of the temperature difference between the housing and the winding, and the peak value of the temperature difference between the three-phase windings.

[0074] S3. Data preprocessing: Standardize or normalize the features, eliminate the dimension difference, and divide the training set and the test set (such as in an 8:2 ratio).

[0075] S4. Model Selection and Training: Select a classification model (such as random forest, SVM, neural network, etc.). The input is the feature set, and the output is the fault type label. Train the model through supervised learning.

[0076] S5. Model Evaluation: Evaluate the model performance using metrics such as accuracy and confusion matrix, and optimize the hyperparameters through cross-validation.

[0077] It should be noted that for a feasible simulation process, assuming that the peak temperature rise rates of the motor housing and the three-phase windings are both 5°C / min, the peak temperature fluctuation amounts are both 3.0°C, the peak temperature difference between the motor housing and the windings is 8°C, and the peak temperature difference between the three-phase windings is 2°C, the fault type recognition results can be obtained. For specific details, refer to Table 1.

[0078] Table 1. Recognition Results of Winding Overheating Fault Types

[0079]

[0080] It should be noted that the feature set of the bearing wear fault includes the vibration spectrograms in the radial, axial, and tangential directions of the motor bearing and the degree of vibration asymmetry.

[0081] It should be noted that the feature set of the current imbalance fault includes the harmonic content of the three-phase current of the motor, the peak value of the instantaneous current fluctuation, and the current waveform distortion rate.

[0082] It should be noted that the feature set of the unstable speed fault includes the peak value of the speed fluctuation during the constant-speed operation of the motor, the peak value of the overshoot during speed switching, and the peak value of the response duration.

[0083] It should be noted that based on the historical fault data of the motor, the present invention extracts features and constructs an identification model, which can accurately distinguish typical fault types such as winding overheating, bearing wear, current imbalance, and unstable speed, improving the pertinence and accuracy of diagnosis.

[0084] The hidden danger judgment module inputs the real-time operation data of the motor into the fault identification model for matching degree analysis to judge whether there are potential fault hazards in the motor.

[0085] Further, the specific working process of the hidden danger judgment module includes: inputting the temperature data of the motor into the identification model of the winding overheating fault, obtaining the difference between each sub-item in the motor temperature data and the corresponding sub-item in the winding overheating fault identification model, and recording it as the deviation of each sub-item in the motor temperature data relative to the model. Then input this deviation into the preset relationship model between the deviation of the temperature data relative to the model and the matching degree of the winding overheating fault identification model to output the matching degree between the motor and the winding overheating fault identification model. The relationship model contains the quantitative mapping relationship between the deviation of the temperature data relative to the model and the matching degree of the winding overheating fault identification model.

[0086] Similarly, obtain the matching degrees of the motor and bearing wear fault recognition model, the current imbalance fault recognition model, and the rotational speed instability fault recognition model.

[0087] It should be noted that for a feasible simulation process, based on the deviation of each sub-item in the motor temperature data relative to the model, analyze the matching degree of the motor and the winding overheating fault recognition model. The specific implementation process is as follows: Steps1. Prepare data: Obtain a set of real-time motor temperature data, and calculate the deviation of each sub-item relative to the corresponding sub-item in the model. For specific reference, see Table 2.

[0088] Table 2. Deviation of each sub-item of the real-time temperature data relative to the corresponding sub-item in the model

[0089]

[0090] Steps2. Determine the matching degree relationship model: Assume that the preset relationship model between the deviation of the temperature data relative to the model and the winding overheating fault recognition model is a linear relationship: Matching degree = 1 - Σ(absolute value of deviation / model value).

[0091] Steps3. Calculate the matching degree: Calculate according to the above formula: Σ(absolute value of deviation / model value) = (0.5 / 5) + (0.2 / 6) + (0.2 / 3) + (0.4 / 4) + (0.5 / 8) + (0.2 / 2) ≈ 0.4625; Matching degree = 1 - 0.4625 = 0.5375, expressed as a percentage is 53.75%.

[0092] Steps4. Final table presentation, see Table 3.

[0093] Table 3. Calculation results

[0094]

[0095] It should be noted that to obtain the deviation of the vibration spectrograms of the radial, axial, and tangential directions of the motor bearing in the motor vibration data relative to the model, the specific method is as follows: Obtain the similarity between the radial vibration spectrogram of the motor bearing in the motor vibration data and the radial vibration spectrogram of the bearing wear fault recognition model, and substitute it into the control relationship between the similarity of the vibration spectrogram and the deviation of the vibration spectrogram relative to the model to obtain the deviation of the radial vibration spectrogram of the motor bearing relative to the model.

[0096] Similarly, obtain the deviations of the axial and tangential vibration spectrograms of the motor bearing relative to the model.

[0097] Further, the specific working process of the potential hazard judgment module further includes: comparing the matching degrees of the motor with various fault recognition models with a preset matching degree threshold. If the matching degrees of the motor with all the fault recognition models are less than the preset matching degree threshold, the motor has no potential faults; otherwise, the motor has potential faults.

[0098] When the motor has potential faults, the warning module analyzes the fault types and evaluates the fault levels, and generates corresponding fault warning signals.

[0099] Further, the specific working process of the warning module includes: if the matching degree of the motor with a certain fault recognition model is greater than or equal to the preset matching degree threshold, classify this type of fault into the fault types of the motor, and statistically obtain the set of fault types of the motor.

[0100] Further, the specific working process of the warning module further includes: setting a relation comparison table of the matching degree of the motor with various fault recognition models - the comprehensive fault probability of the motor according to a preset rule, and screening out the comprehensive fault probability of the motor according to the matching degrees of the motor with various fault recognition models in the set of fault types.

[0101] Substitute the comprehensive fault probability of the motor into the mapping relation between the preset comprehensive fault probability and the fault level to obtain the fault level of the motor and generate corresponding fault warning signals.

[0102] The cloud collaboration module encrypts and transmits the multi-source operation data and fault warning signals of the motor to the cloud server, and receives the optimized fault instructions sent from the cloud.

[0103] It should be noted that the optimized fault instructions can specifically be parameter dynamic adjustment instructions, operation mode switching instructions, maintenance and repair strategy instructions, fault suppression and fault tolerance control instructions, system-level collaboration control instructions, etc.

[0104] It should be noted that by adopting deviation matching degree analysis and combining with the preset threshold and comprehensive fault probability mapping rules, the present invention realizes the dynamic quantitative evaluation and level division of potential faults, which helps to prioritize the treatment of high-risk faults and reduce the risk of sudden shutdown.

[0105] The database stores the operation data of the motor during historical faults.

[0106] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should fall within the protection scope of the present invention.

Claims

1. A sintering trolley motor fault monitoring system, characterized in that: include: Data acquisition module: collects multi-source operation data of the motor in real time, including temperature data, vibration data, current data and speed data, and performs pre-processing; Model building module: Based on the operating data of the motor's historical faults, the temperature rise, temperature fluctuation and temperature distribution characteristics when the winding is overheated are extracted; the spectrum and asymmetric characteristics of vibration in all directions when the bearing is worn; the harmonic, instantaneous current and waveform distortion rate characteristics when the current is unbalanced; Speed ​​fluctuation and speed switching overshoot and response time characteristics when the speed is unstable; build identification models for various faults; Hidden danger judgment module: input the real-time operation data of the motor into the fault identification model for matching analysis to determine whether the motor has hidden dangers of faults; Early warning module: analyzes the fault type and evaluates the fault level when there is a potential fault in the motor, and generates a corresponding fault early warning signal; Cloud collaboration module: encrypts and transmits the motor's multi-source operating data and fault warning signals to the cloud server, and receives fault optimization instructions issued by the cloud; Database: stores the running data of the motor during historical faults; The specific working process of the model building module is as follows: Extract the running data of the motor when it has a historical fault stored in the database, and filter the temperature data when the motor has a historical winding overheating, the vibration data when the bearing is worn, the current data when the current is unbalanced, and the speed data when the speed is unstable; According to the temperature data of the motor winding overheating in the past, the peak values ​​of the temperature rise rate of the motor housing and the three-phase winding, the peak value of the temperature fluctuation, the peak value of the temperature difference between the motor housing and the winding, and the peak value of the temperature difference between the three-phase windings are obtained when the motor winding is overheating in the past, and the feature set of the winding overheating fault is obtained, and an identification model of the winding overheating fault is constructed with the input as the fault feature and the output as the fault type; Similarly, an identification model for bearing wear, current imbalance, and speed instability faults is constructed.

2. A sintering trolley motor fault monitoring system according to claim 1, characterized in that: The data acquisition module includes a temperature data acquisition unit, a vibration data acquisition unit, a current data acquisition unit and a speed data acquisition unit, wherein the specific working process of the temperature data acquisition unit is as follows: Set the duration of the monitoring cycle and set each sampling time point within the monitoring cycle according to the preset time interval; The temperature of the motor housing and the three-phase winding at each sampling time point during the monitoring period is collected by an infrared temperature sensor, and a temperature change trend curve of the motor housing and the three-phase winding during the monitoring period is drawn; According to the temperature change trend curves of the motor housing and the three-phase winding during the monitoring period, the peak value of the temperature rise rate and the peak value of the temperature fluctuation are obtained respectively, and the temperature rise data and temperature fluctuation data of the motor are obtained accordingly; Compare the temperature change trend curves of the motor housing and the three-phase windings in the monitoring period longitudinally, obtain the peak temperature difference between the motor housing and the windings, and the peak temperature difference between the three-phase windings in the monitoring period, and obtain the temperature distribution data of the motor; The temperature rise data, temperature fluctuation data and temperature distribution data of the motor are summarized to obtain the temperature data of the motor.

3. A sintering trolley motor fault monitoring system according to claim 2, characterized in that: The specific working process of the vibration data acquisition unit is as follows: The radial, axial and tangential vibration spectra of the motor bearings during the monitoring period are collected by a three-axis vibration sensor; According to the radial vibration spectrum of the motor bearing during the monitoring period, the amplitude of each frequency component and the phase relationship between each frequency component are obtained, and the amplitude and the phase relationship between the frequency components are substituted into the preset relationship comparison table between the amplitude of the frequency component, the phase relationship between the frequency components and the degree of vibration asymmetry, and the degree of asymmetry of the radial vibration of the motor bearing is obtained by matching; Similarly, according to the analysis method of the asymmetric degree of radial vibration of the motor bearing, the asymmetric degree of axial and tangential vibration of the motor bearing is obtained; The vibration data of the motor is constructed based on the radial, axial and tangential vibration spectra and the degree of vibration asymmetry of the motor bearing.

4. A sintering trolley motor fault monitoring system according to claim 2, characterized in that: The specific working process of the current data acquisition unit is as follows: The three-phase current waveform of the motor during the monitoring period is collected by the Hall current sensor, and the harmonic content, instantaneous current fluctuation peak value, and current waveform distortion rate of the three-phase current of the motor during the monitoring period are further obtained to construct the current data of the motor.

5. The sintering trolley motor fault monitoring system according to claim 2 is characterized in that: The specific working process of the speed data acquisition unit is as follows: The motor speed is monitored by a photoelectric encoder to capture the peak value of the speed fluctuation when the motor is running at a constant speed within the monitoring period, and the overshoot peak value and response time peak value when the motor speed is switched are collected to construct the motor speed data.

6. A sintering trolley motor fault monitoring system according to claim 1, characterized in that: The specific working process of the hidden danger judgment module includes: Input the temperature data of the motor into the identification model of the winding overheat fault, obtain the difference between each sub-item in the motor temperature data and the corresponding sub-item in the winding overheat fault identification model, record it as the deviation of each sub-item in the motor temperature data relative to the model, and input the deviation into a preset temperature data relative model deviation-winding overheat fault identification model matching degree relationship model, output the matching degree of the motor and the winding overheat fault identification model, the relationship model includes a quantitative mapping relationship between the temperature data relative model deviation and the winding overheat fault identification model matching degree; Similarly, the matching degree of the motor and bearing wear fault identification model, the current imbalance fault identification model, and the speed instability fault identification model is obtained.

7. A sintering trolley motor fault monitoring system according to claim 6, characterized in that: The specific working process of the hidden danger judgment module also includes: The matching degree between the motor and various fault identification models is compared with a preset matching degree threshold. If the matching degree between the motor and various fault identification models is less than the preset matching degree threshold, the motor has no hidden fault risk. Otherwise, the motor has hidden fault risk.

8. The sintering trolley motor fault monitoring system according to claim 6, characterized in that: The specific working process of the early warning module includes: If the matching degree between the motor and a certain fault identification model is greater than or equal to a preset matching degree threshold, the fault is classified as a fault type of the motor, and a set of motor fault types is obtained by statistics.

9. A sintering trolley motor fault monitoring system according to claim 8, characterized in that: The specific working process of the early warning module also includes: According to preset rules, a comparison table of the relationship between the motor's matching degree with various fault identification models and the motor's comprehensive fault probability is set, and the motor's comprehensive fault probability is screened out according to the matching degree between the motor and various fault identification models in the fault type set; The comprehensive fault probability of the motor is substituted into the preset mapping relationship between the comprehensive fault probability and the fault level to obtain the fault level of the motor and generate a corresponding fault warning signal.

Images