Small motor operation monitoring method
By building a motor structure model and analyzing data using convolutional neural networks, the problem that traditional motor fault diagnosis methods are difficult to deal with multiple types of faults is solved, real-time monitoring and fault prediction of motor operating status is realized, and the level of motor intelligence and system are improved. reliability.
Patent Information
- Application Number
- CN202411076949.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-11-22
AI Technical Summary
Traditional motor fault diagnosis methods are difficult to effectively select feature extraction methods, resulting in poor detection of multiple fault types, making it difficult to achieve intelligent maintenance and improve system reliability.
The small motor operation monitoring method is adopted, and the motor structure model is constructed, data is collected using multiple sensors, and data analysis and model training is used to achieve real-time monitoring of motor operation status and fault prediction.
It improves the intelligent level of motor operation, optimizes maintenance strategies, reduces failure rate, and improves the reliability and efficiency of the entire system, providing multi-dimensional motor status analysis and prediction reports.
Smart Images

Figure CN119001437A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of motor monitoring, and in particular to a method for monitoring the operation of a small motor. Background Art
[0002] Motors are an indispensable power source in industrial automation and modern manufacturing systems. The normal operation of motors is crucial to ensuring the continuity of production lines and product quality. By monitoring the operating status of motors, potential faults can be discovered in time, unexpected downtime can be prevented, maintenance costs can be reduced, equipment life can be extended, and overall system reliability can be improved.
[0003] The main steps of traditional motor fault diagnosis methods are data preprocessing, feature extraction, fault diagnosis, etc. Since traditional methods have achieved good results for single fault detection tasks; with the increasing number of motor fault types, it is often difficult to select feature extraction methods when applying traditional methods to motor fault diagnosis.
[0004] Therefore, it is an urgent problem for those skilled in the art to propose a small motor operation monitoring method to solve the difficulties existing in the prior art. Summary of the invention
[0005] In view of this, the present invention provides a small motor operation monitoring method, which can improve the intelligence level of motor operation, optimize maintenance strategies, reduce failure rates, and improve the reliability and efficiency of the entire system.
[0006] In order to achieve the above object, the present invention adopts the following technical solution:
[0007] A method for monitoring the operation of a small motor comprises the following steps:
[0008] S1, data acquisition step: acquiring and constructing a structural model of the target motor to be monitored;
[0009] S2, data collection step: based on the structure model of the target motor to be monitored obtained in S1, arrange the sensing and data transmission modules to collect the physical quantity data related to the motor operation and send them;
[0010] S3, data analysis step: receiving the motor operation related physical quantity data sent in S2, and preprocessing and feature extracting the data in turn, and training a motor operation status monitoring model based on a convolutional neural network;
[0011] S4, operation and maintenance step: Analyze the motor operation status based on the motor operation status monitoring model obtained in S3 to obtain a motor operation status report.
[0012] In the above method, optionally, the specific contents in S1 include:
[0013] The key parameters of the motor are obtained through experimental tests, including rated power, torque, speed, resistance, and inductance. MATLAB is used to assist in building a motor model. By comparing the experimental data with the model prediction results, the accuracy of the model is verified and the model parameters are adjusted.
[0014] In the above method, optionally, the sensing and data transmission module in S2 includes:
[0015] Hall effect sensors, electromagnetic induction sensors, temperature sensors, and wireless transmission units;
[0016] The output ends of the Hall effect sensor, the electromagnetic induction sensor, and the temperature sensor are all connected to the wireless transmission unit for sending the collected physical quantity data related to the motor operation.
[0017] In the above method, optionally, a Hall effect sensor is installed on the three-phase stator winding of the motor, and the current is indirectly measured by detecting the change of the magnetic field, and the motor stator input current and voltage signals are collected to calculate the motor operating parameters, including: current effective value, voltage effective value, input power, speed, output torque, and output power.
[0018] The above method, optionally, an electromagnetic induction sensor:
[0019] According to the monitoring target motor structure model obtained in S1, a hollow coil sensor is set to collect the motor magnetic field changes, the appropriate hollow coil size is selected, the hollow coil is fixed to the outside of the motor and connected to the signal conditioning circuit, and filtering is performed to obtain the motor magnetic field change data.
[0020] In the above method, optionally, the specific contents in S3 include:
[0021] Receive the motor operation related physical quantity data sent by the wireless transmission unit, including: motor operation parameters, motor magnetic field change data and motor temperature data, extract the effective features in the motor magnetic field change data, and train a preliminary motor operation status monitoring model through a convolutional neural network based on the motor operation parameters and motor temperature data, and update the network weights through the back propagation algorithm to minimize the prediction error.
[0022] The above method optionally uses a back propagation algorithm to update the network weights, including:
[0023] Forward propagation: input feature data and calculate predicted output;
[0024] Calculate loss: Based on the predicted output and the actual motor state, the mean square error is used to calculate the loss function value;
[0025] Back propagation: Calculate the gradient of the loss function with respect to the network parameters through the chain rule;
[0026] Weight update: Update the network weights according to the gradient descent method, and iterate the optimization until the model converges.
[0027] In the above method, optionally, the specific contents of S4 include:
[0028] Use the motor operation status monitoring model to predict the real-time feature data and output the classification of the motor operation status; based on the prediction results, analyze the health of the motor, identify potential failure modes, and convert the model prediction results into a readable report format, including: real-time charts of motor status, historical trend analysis, and failure mode identification; based on the prediction results of the motor operation status monitoring model, set abnormal thresholds when abnormal conditions are detected.
[0029] It can be seen from the above technical solution that, compared with the prior art, the small motor operation monitoring method of the present invention has the following beneficial effects:
[0030] The motor operation monitoring method proposed in the present invention acquires multi-dimensional data of motor operation by constructing a motor structure model and using a variety of sensors, and uses a convolutional neural network to perform data analysis and model training to achieve real-time monitoring of the motor operation status and fault prediction; it can improve the intelligent level of motor operation, optimize maintenance strategies, reduce failure rates, and improve the reliability and efficiency of the entire system;
[0031] The key parameters of the motor are obtained through experimental tests, and a motor model is established using software tools to verify the accuracy of the model and adjust the model parameters to improve the prediction performance. The data acquisition module uses a multi-point sensor network for real-time data acquisition, and the data processing module uses distributed computing for data cleaning, normalization and preliminary clustering analysis. The state monitoring module groups and recognizes the operating status of the motor and generates a status report. The visual analysis module combines convolutional neural networks and isolation forest algorithms to identify abnormal image patterns, the abnormal pattern recognition module analyzes abnormal patterns of the data, the network analysis module analyzes key components and fault propagation paths, and the pattern prediction module performs pattern discovery and trend prediction. These technical steps together constitute a comprehensive motor operation monitoring system that can provide multi-dimensional motor state analysis and prediction reports. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0033] Figure 1A flow chart of a small motor operation monitoring method provided by the present invention. DETAILED DESCRIPTION
[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0035] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, the elements defined by the sentence "comprise one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0036] The present invention can be used in many general or special computing device environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multi-processor devices, distributed computing environments including any of the above devices or devices, etc.
[0037] Reference Figure 1 As shown, the present invention discloses a small motor operation monitoring method, comprising the following steps:
[0038] S1, data acquisition step: acquiring and constructing a structural model of the target motor to be monitored;
[0039] S2, data collection step: based on the structure model of the target motor to be monitored obtained in S1, arrange the sensing and data transmission modules to collect the physical quantity data related to the motor operation and send them;
[0040] S3, data analysis step: receiving the motor operation related physical quantity data sent in S2, and preprocessing and feature extracting the data in turn, and training a motor operation status monitoring model based on a convolutional neural network;
[0041] S4, operation and maintenance step: Analyze the motor operation status based on the motor operation status monitoring model obtained in S3 to obtain a motor operation status report.
[0042] Furthermore, the specific contents of S1 include:
[0043] The key parameters of the motor are obtained through experimental tests, including rated power, torque, speed, resistance, and inductance. MATLAB is used to assist in building a motor model. By comparing the experimental data with the model prediction results, the accuracy of the model is verified and the model parameters are adjusted to improve the prediction performance of the model.
[0044] Furthermore, the sensing and data transmission module in S2 includes:
[0045] Hall effect sensors, electromagnetic induction sensors, temperature sensors, and wireless transmission units;
[0046] The output ends of the Hall effect sensor, the electromagnetic induction sensor, and the temperature sensor are all connected to the wireless transmission unit for sending the collected physical quantity data related to the motor operation.
[0047] Furthermore, the Hall effect sensor is installed on the three-phase stator winding of the motor, and indirectly measures the current by detecting the change of the magnetic field, collects the motor stator input current and voltage signals, and calculates the motor operating parameters, including: current effective value, voltage effective value, input power, speed, output torque, and output power.
[0048] Specifically, the Hall effect sensor usually includes a Hall element placed near the stator winding of the motor to sense the magnetic field passing through the wire. When current flows through the stator winding, the Hall element generates a voltage signal proportional to the current, which is then amplified and converted into a usable output signal for monitoring and control.
[0049] The installation location of the Hall current sensor needs to be carefully selected to ensure that the Hall element can effectively sense the magnetic field. The output signal of the sensor usually needs to be filtered and amplified to improve the signal quality and measurement accuracy. In addition, in order to protect the Hall element from overvoltage damage, appropriate protection measures are usually added to the circuit.
[0050] In one sampling period, the stator current I(t) and stator voltage U(t) of the motor are sampled to obtain the sampling value: I m , U m , (m=1...N) then the effective value of the current is I rms , voltage effective value U rms 、Input power P E , speed n, output torque T, output power P m The calculation is as follows:
[0051]
[0052] Furthermore, the electromagnetic induction sensor: according to the monitoring target motor structure model obtained in S1, a hollow coil sensor is set to collect the motor magnetic field change, the appropriate hollow coil size is selected, the hollow coil is fixed to the outside of the motor and connected to the signal conditioning circuit, and filtering is performed to obtain the motor magnetic field change data.
[0053] Specifically, hollow coils play a core role in electromagnetic induction sensors. They can detect changes in the surrounding magnetic field, thereby measuring the strength and direction of the magnetic field. This sensor uses Faraday's law of electromagnetic induction. The magnitude of the induced electromotive force is proportional to the rate of change of the magnetic field, that is, the changing magnetic field can generate an induced electromotive force, and then generate current. The hollow coil sensor monitors the changes in the induced current or voltage. By measuring this electromotive force, the sensor can calculate the strength and direction of the magnetic field and indirectly measure the changes in the magnetic field.
[0054] The hollow coil sensor does not need to be in direct contact with the object being measured, thus avoiding wear or damage. High sensitivity: it can detect weak changes in the magnetic field and is suitable for precision measurement. Fast response speed: it can respond quickly to changes in the magnetic field and is suitable for dynamic measurement. Simple structure: the hollow coil has a simple structure and is easy to integrate and maintain.
[0055] Setting steps for air-core coil sensor to measure the change of motor magnetic field:
[0056] Choose the right air-core coil size: The diameter of the air-core coil should match the size of the motor to ensure that the coil can be placed tightly around the stator or rotor of the motor to maximize the capture of magnetic field changes.
[0057] Mounting the air-core coil: Fix the air-core coil to the outside of the motor, ensuring there is no air gap between the coil and the part being measured to reduce magnetic field leakage. Brackets or magnetic materials can be used to fix the coil while allowing the coil to move freely to accommodate thermal expansion or mechanical vibration of the motor.
[0058] Connecting to the signal conditioning circuit: The air-core coil sensor usually needs to be connected to the signal conditioning circuit to amplify the weak voltage signal induced in the coil and convert it into a signal that can be used for recording and analysis. Choose the appropriate amplifier and filter to ensure the quality and accuracy of the signal.
[0059] Calibrating the sensor: Before you begin measuring, calibrate the air-core coil sensor to ensure the accuracy of your measurements. This may involve using a standard magnet of known magnetic field strength or calibration equipment to adjust the sensor's gain.
[0060] Take measurements: Start the motor and gradually increase the load, and observe the changes in the output signal of the air-core coil sensor. Record the magnetic field data under different working conditions for subsequent analysis and comparison.
[0061] Data analysis: Use a data acquisition system to collect the output signals of the sensors and perform necessary processing, such as filtering, integration, differentiation, etc., to extract useful information about the changes in the motor's magnetic field. Analyzing this data can help evaluate the motor's operating status and diagnose potential problems.
[0062] Experimental verification and adjustment: Based on the preliminary measurement results, it may be necessary to fine-tune the sensor position or signal conditioning circuitry to obtain the best measurement results.
[0063] Furthermore, the specific contents of S3 include:
[0064] Receive the motor operation related physical quantity data sent by the wireless transmission unit, including: motor operation parameters, motor magnetic field change data and motor temperature data, extract the effective features in the motor magnetic field change data, and train a preliminary motor operation status monitoring model through a convolutional neural network based on the motor operation parameters and motor temperature data, and update the network weights through the back propagation algorithm to minimize the prediction error.
[0065] Specifically, after training is completed, an independent test set is used to evaluate the model performance, including indicators such as prediction accuracy, recall rate, F1 score, etc. Based on the evaluation results, it may be necessary to adjust the network structure (such as the number of layers and nodes), optimize algorithm parameters (such as learning rate and momentum), or increase data enhancement to further improve the generalization ability and robustness of the model.
[0066] Furthermore, the back-propagation algorithm is used to update the network weights, including:
[0067] Forward propagation: input feature data and calculate predicted output;
[0068] Calculate loss: Based on the predicted output and the actual motor state, the mean square error is used to calculate the loss function value;
[0069] Back propagation: Calculate the gradient of the loss function with respect to the network parameters through the chain rule;
[0070] Weight update: Update the network weights according to the gradient descent method, and iterate the optimization until the model converges.
[0071] Furthermore, the specific contents of S4 include:
[0072] Use the motor operation status monitoring model to predict the real-time feature data and output the classification of the motor operation status; based on the prediction results, analyze the health of the motor, identify potential failure modes, and convert the model prediction results into a readable report format, including: real-time charts of motor status, historical trend analysis, and failure mode identification; based on the prediction results of the motor operation status monitoring model, set abnormal thresholds when abnormal conditions are detected.
[0073] Specifically, based on the prediction results of the motor operation status monitoring model, an abnormal threshold is set. When an abnormal status is detected, an alarm is automatically triggered to notify the maintenance team to take action. At the same time, based on the analysis of the motor status, maintenance suggestions are provided, such as adjusting the workload, performing preventive maintenance or replacing parts to extend the motor life and reduce unplanned downtime.
[0074] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.
[0075] Those skilled in the art may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, computer software, or a combination of both.
[0076] In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.
[0077] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for monitoring the operation of a small motor, characterized in that: The following steps are involved: S1, data acquisition step: acquiring and constructing a structural model of the target motor to be monitored; S2, data collection step: based on the structure model of the target motor to be monitored obtained in S1, arrange the sensing and data transmission modules to collect the physical quantity data related to the motor operation and send them; S3, data analysis step: receiving the motor operation related physical quantity data sent in S2, and preprocessing and feature extracting the data in turn, and training a motor operation status monitoring model based on a convolutional neural network; S4, operation and maintenance step: Analyze the motor operation status based on the motor operation status monitoring model obtained in S3 to obtain a motor operation status report.
2. A small motor operation monitoring method according to claim 1, characterized in that: The specific contents of S1 include: The key parameters of the motor are obtained through experimental tests, including rated power, torque, speed, resistance, and inductance. MATLAB is used to assist in building a motor model. By comparing the experimental data with the model prediction results, the accuracy of the model is verified and the model parameters are adjusted.
3. A small motor operation monitoring method according to claim 1, characterized in that: The sensing and data transmission modules in S2 include: Hall effect sensors, electromagnetic induction sensors, temperature sensors, and wireless transmission units; The output ends of the Hall effect sensor, the electromagnetic induction sensor, and the temperature sensor are all connected to the wireless transmission unit for sending the collected physical quantity data related to the motor operation.
4. A small motor operation monitoring method according to claim 3, characterized in that: The Hall effect sensor is installed on the three-phase stator winding of the motor. It indirectly measures the current by detecting the change of the magnetic field, collects the input current and voltage signals of the motor stator, and calculates the motor operating parameters, including: current effective value, voltage effective value, input power, speed, output torque, and output power.
5. A small motor operation monitoring method according to claim 3, characterized in that: Electromagnetic induction sensor: According to the monitoring target motor structure model obtained in S1, a hollow coil sensor is set to collect the motor magnetic field changes, the appropriate hollow coil size is selected, the hollow coil is fixed to the outside of the motor and connected to the signal conditioning circuit, and filtering is performed to obtain the motor magnetic field change data.
6. A small motor operation monitoring method according to claim 5, characterized in that: The specific contents of S3 include: Receive the motor operation related physical quantity data sent by the wireless transmission unit, including: motor operation parameters, motor magnetic field change data and motor temperature data, extract the effective features in the motor magnetic field change data, and train the motor operation status monitoring model through a convolutional neural network based on the motor operation parameters and motor temperature data, and update the network weights through the back propagation algorithm to minimize the prediction error.
7. A small motor operation monitoring method according to claim 6, characterized in that: The back-propagation algorithm is used to update the network weights, including: Forward propagation: input feature data into the motor operation status monitoring model and calculate the predicted output; Calculate loss: Based on the predicted output and the actual motor state, the mean square error is used to calculate the loss function value; Back propagation: Calculate the gradient of the loss function with respect to the network parameters through the chain rule; Weight update: Update the network weights according to the gradient descent method, and iterate the optimization until the motor operation status monitoring model converges.
8. A small motor operation monitoring method according to claim 1, characterized in that: The specific contents of S4 include: Use the motor operation status monitoring model to predict the real-time feature data and output the classification of the motor operation status; based on the prediction results, analyze the health of the motor, identify potential failure modes, and convert the model prediction results into a readable report format, including: real-time charts of motor status, historical trend analysis, and failure mode identification; based on the prediction results of the motor operation status monitoring model, set abnormal thresholds when abnormal conditions are detected.
