Industrial fan permanent magnet motor rotor dynamic balance detection and correction process

Abstract

The invention discloses a rotor dynamic balance online detection and correction process and system of an industrial fan permanent magnet motor. According to the technology, in the rotating working state of the rotor, a non-contact sensor is used for collecting vibration signals in real time, a signal processing system is used for calculating the magnitude and the phase of the unbalance amount, a correction executing mechanism integrated on the rotor is immediately driven to conduct online compensation, and the rotor can be restored to the dynamic balance state without shutdown. The correction executing mechanism is preferably a movable permanent magnet counterweight device, and accurate correction is achieved by changing the counterweight position. Automatic and online dynamic balance maintenance is realized, and the operation reliability, the service life and the energy efficiency of the permanent magnet motor of the industrial fan are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of permanent magnet motor technology, and in particular to the dynamic balance detection and correction process for the rotor of a permanent magnet motor for industrial fans. Background Technology

[0002] Industrial fans, as important ventilation, cooling, and gas transport equipment, are widely used in power, metallurgy, chemical, tunnel, and data center industries. Their core drive component—the permanent magnet synchronous motor—has become the mainstream choice for modern industrial fans due to its advantages such as high efficiency, high power density, and ease of maintenance. As a high-speed rotating component, the dynamic balance accuracy of the motor's rotor directly determines the overall vibration level, noise level, operational reliability, and service life of the machine.

[0003] In actual operation, imbalance in fan rotor systems is difficult to completely avoid. Its causes are complex, including uneven material composition of rotor castings or magnets, assembly errors, wear, dust accumulation, or corrosion of components due to long-term operation, and micro-deformation of the rotor structure caused by harsh operating conditions (such as thermal stress and impact loads). Even small imbalances can generate significant centrifugal forces at high speeds, triggering harmful mechanical vibrations. These vibrations not only produce noise pollution but also lead to accelerated bearing wear, seal failure, and loosening of fasteners. In severe cases, they can cause serious malfunctions such as rotor and stator rubbing, resulting in unplanned downtime and significant economic losses and safety risks.

[0004] Currently, the industry mainly relies on the following two traditional processes for controlling and correcting the dynamic balance of such motor rotors: 1. Offline Pre-Balancing Process: This is the most common initial balancing method. After assembly or during maintenance, the rotor needs to be removed from the motor and installed on a dedicated offline dynamic balancing machine for testing and calibration. Calibration typically involves drilling holes at both ends of the rotor's balancing surface to remove weight or adding fixed counterweights (such as balancing putty or screw counterweights). This method has significant limitations: First, it is a "static" balancing process, unable to simulate the rotor's actual state under operating temperature, electromagnetic force, and load; the balancing effect often deviates after installation and operation. Second, the entire process requires stopping, disassembling, transporting, online balancing, and reinstallation, which is time-consuming and labor-intensive, severely impacting equipment availability. Finally, it is a "one-time" and irreversible calibration, unable to address new imbalances caused by dynamic factors such as wear and scale buildup during operation.

[0005] 2. Periodic Shutdown and On-Site Balancing: For equipment already in operation, when vibration levels exceed limits, maintenance personnel use a portable on-site dynamic balancing instrument for diagnosis and correction. While this method does not require disassembling the rotor, it still necessitates a complete shutdown of the equipment. Technicians must temporarily install trial weights on the exposed shaft ends or fan blades of the rotor while it is stopped, then restart the machine to measure the vibration response. This process involves multiple iterations of "trial weight-start-measurement" before the final counterweight scheme can be calculated and implemented. This process also leads to production interruptions, is highly dependent on the experience of technicians, and results in inconsistent balancing accuracy and efficiency. More importantly, it is a passive, periodic maintenance method that cannot provide immediate response to sudden or gradual imbalances.

[0006] In summary, existing technologies suffer from a fundamental contradiction: rotor imbalance is a dynamic problem that arises and changes during operation, while the main correction methods rely on static intervention through shutdown and offline processes. This leads to high maintenance costs, reduced overall equipment efficiency (OEE), and an inability to fundamentally guarantee continuous and optimal operating conditions. Therefore, the industry urgently needs an innovative technology that can detect, diagnose, and automatically correct rotor imbalance in real time without interrupting equipment operation, enabling a shift from "periodic intervention" to "continuous autonomy" in operation and maintenance. This invention is proposed against this backdrop of industry needs and technological challenges. Summary of the Invention

[0007] This invention proposes a process for dynamic balancing detection and correction of the rotor of a permanent magnet motor for industrial fans, which solves the aforementioned problems existing in the use of existing technologies.

[0008] The technical solution of this invention is implemented as follows: An online detection and correction process for the rotor dynamic balance of an industrial fan permanent magnet motor, characterized by the following steps: Online detection phase: While the motor rotor is in a rotating operating state, vibration signals of the rotor in the radial plane are collected in real time by at least two non-contact vibration sensors, and the vibration signals are transmitted to the signal processing and control system. Data processing and unbalance calculation stage: The signal processing and control system filters, amplifies and converts the received vibration signal into digital signal, and calculates the magnitude and phase of the unbalance mass of the rotor on at least one correction plane based on the preset rotor dynamics model and influence coefficient method. Online correction phase: Based on the calculated magnitude and phase of the unbalanced mass, the signal processing and control system generates a correction command to drive a dynamic balancing correction actuator integrated on the rotor to perform an action, so as to realize online compensation of the unbalanced mass while the rotor is rotating continuously, until the vibration signal value drops below the preset threshold.

[0009] Preferably, in the online detection stage, at least two non-contact vibration sensors are arranged along the axial direction of the motor housing and respectively aligned with the bearing seats at both ends of the rotor or the housing position near the bearing seats, so as to collect vibration signals at different axial positions for dual-plane dynamic balance calculation.

[0010] Preferably, the online detection stage further includes acquiring the key phase or zero-position mark signal of the rotor through a phase reference sensor that rotates coaxially with the rotor, so as to accurately determine the phase reference of the unbalance mass.

[0011] Preferably, the dynamic balancing actuator is a built-in non-contact adjustable permanent magnet counterweight device, which includes: Fixed base: Fixedly mounted on the rotor shaft or end plate; Movable counterweight: embedded with a permanent magnet, and can be moved in a controlled manner along the circumferential and / or radial direction of the fixed base; Drive unit: Receives the correction command, drives the movable counterweight to move to the target phase and radial position, and generates a compensating torque by changing the position of the counterweight relative to the rotor's center of gravity.

[0012] Preferably, the drive unit is a micro stepper motor or a piezoelectric ceramic driver, which is connected to the movable counterweight through a worm gear, lead screw, or linear motor mechanism.

[0013] Preferably, in the data processing and unbalance calculation stage, the signal processing and control system adopts an adaptive filtering algorithm to separate the fundamental frequency vibration component that is the same as the rotor rotation frequency from the collected vibration signal, and uses it to calculate the unbalance.

[0014] Preferably, the process further includes a self-learning and optimization stage: after multiple online detection and correction cycles, the signal processing and control system records and analyzes historical imbalance data and correction effects, automatically optimizes the influence coefficient of the rotor dynamics model, and predicts the deterioration trend of the rotor's dynamic balance state.

[0015] The online detection and correction process and system for dynamic balance of permanent magnet motor rotors for industrial fans involved in this invention has the following significant advantages compared with the prior art: 1. Achieve true "zero downtime" maintenance, greatly improving overall equipment efficiency (OEE). The most significant advantage of this invention lies in overcoming the technical bottleneck of traditional dynamic balancing, which requires machine downtime. Through online real-time detection and correction, the entire process is completed while the motor is running continuously, without interrupting production. This directly eliminates unplanned downtime caused by balancing maintenance, offering substantial economic benefits for continuous production industries such as metallurgy and chemical engineering.

[0016] 2. Shifting from a "reactive maintenance" to a "proactive prevention" operation and maintenance model to improve reliability. Traditional methods respond passively after vibration exceeds the limit, while this invention can monitor minute changes in imbalance in real time and automatically perform micro-correction before it develops to a harmful threshold. This enables early warning and immediate self-regulation of rotor imbalance faults, eliminating potential vibration faults in their infancy, greatly improving the operational reliability and long-term stability of the motor and the entire fan system, and effectively preventing the occurrence of cascading mechanical failures.

[0017] 3. High balancing accuracy, and better matches actual working conditions. Traditional offline balancing methods cannot reproduce the rotor's state under actual operating temperature, electromagnetic field, and load conditions. This invention performs measurement and correction directly during operation, and the collected vibration data and final balance state fully reflect the real operating conditions, resulting in higher correction accuracy and more lasting effects. The introduction of adaptive algorithms also continuously optimizes model parameters, making the system increasingly "intelligent" with use.

[0018] 4. Significantly reduce total lifecycle maintenance costs This technology eliminates the reliance on dedicated balancing machines and frequent on-site manual services. Its automated and intelligent features significantly reduce labor costs, tooling costs, and production losses due to downtime. Simultaneously, by maintaining the rotor in a consistently excellent balanced state, it effectively extends the lifespan of critical mechanical components such as bearings and seals, reduces the frequency of spare parts replacement, and optimizes the overall lifecycle maintenance costs of the equipment.

[0019] 5. Enhance safety and environmental friendliness Continuously suppressing abnormal vibrations fundamentally reduces the risk of safety accidents such as mechanical breakage and loosening caused by excessive vibration. At the same time, lower vibration levels also mean lower operating noise, which helps improve the working environment and meets environmental protection and occupational health requirements.

[0020] 6. High system integration and strong adaptability By integrating detection, analysis, and execution mechanisms into the motor body, a smart closed-loop system is formed. This integrated design facilitates installation and application, and is particularly suitable for applications with high space and reliability requirements. The application of wireless transmission technology further simplifies the connection challenges between rotating and stationary components, improving the system's engineering applicability and reliability.

[0021] In summary, this invention not only solves a specific technical problem, but also promotes a new paradigm of more efficient, intelligent, and economical equipment health management, providing key technical support for the intelligent upgrading of industrial fans and other rotating machinery. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] Example This embodiment discloses an online detection and correction process for the rotor dynamic balance of an industrial fan permanent magnet motor, characterized by the following steps: Online detection phase: While the motor rotor is in a rotating operating state, vibration signals of the rotor in the radial plane are collected in real time by at least two non-contact vibration sensors, and the vibration signals are transmitted to the signal processing and control system. Data processing and unbalance calculation stage: The signal processing and control system filters, amplifies and converts the received vibration signal into digital signal, and calculates the magnitude and phase of the unbalance mass of the rotor on at least one correction plane based on the preset rotor dynamics model and influence coefficient method. Online correction phase: Based on the calculated magnitude and phase of the unbalanced mass, the signal processing and control system generates a correction command to drive a dynamic balancing correction actuator integrated on the rotor to perform an action, so as to realize online compensation of the unbalanced mass while the rotor is rotating continuously, until the vibration signal value drops below the preset threshold.

[0024] Preferably, in the online detection stage, at least two non-contact vibration sensors are arranged along the axial direction of the motor housing and respectively aligned with the bearing seats at both ends of the rotor or the housing position near the bearing seats, so as to collect vibration signals at different axial positions for dual-plane dynamic balance calculation.

[0025] Preferably, the online detection stage further includes acquiring the key phase or zero-position mark signal of the rotor through a phase reference sensor that rotates coaxially with the rotor, so as to accurately determine the phase reference of the unbalance mass.

[0026] Preferably, the dynamic balancing correction actuator is a built-in non-contact adjustable permanent magnet counterweight device, comprising: a fixed base: fixedly mounted on the rotor shaft or end plate; a movable counterweight block: embedded with a permanent magnet, and controllably movable along the circumferential and / or radial direction of the fixed base; and a drive unit: receiving the correction command and driving the movable counterweight block to move to the target phase and radial position, generating a compensating torque by changing the position of the counterweight block relative to the rotor's center of gravity. The drive unit is a micro stepper motor or a piezoelectric ceramic driver, which is connected to the movable counterweight block through a worm gear, lead screw, or linear motor mechanism. In the data processing and unbalance calculation stage, the signal processing and control system employs an adaptive filtering algorithm to separate the fundamental frequency vibration component with the same frequency as the rotor rotation from the collected vibration signal, for calculating the unbalance. The process also includes a self-learning and optimization phase: after multiple online detection and correction cycles, the signal processing and control system records and analyzes historical imbalance data and correction effects, automatically optimizes the influence coefficients of the rotor dynamics model, and predicts the deterioration trend of the rotor's dynamic balance state.

[0027] Example 1: A typical online dynamic balancing correction system and process The process steps in this embodiment are as follows: S1: System initialization and self-test. Before the motor starts, the control system performs a power-on self-test, driving the two counterweights to move to the initial zero point position (e.g., 0-degree phase angle).

[0028] S2: Online Detection and Data Acquisition. The motor starts and reaches a stable operating speed (e.g., 980 RPM). Two vibration sensors continuously acquire time-domain vibration signals, while a key phase sensor synchronously acquires pulse signals per revolution. Data is acquired in real-time and transmitted to the control system at a sampling rate of 10 kHz.

[0029] S3: Signal processing and imbalance calculation. The software within the control system executes the following algorithm flow: Signal preprocessing: Bandpass filtering (center frequency is the rotation frequency) is performed on the original vibration signal to remove high-frequency noise and low-frequency interference.

[0030] Phase locking and spectrum analysis: Using the key phase signal, the vibration signal is synchronously sampled and averaged over an integer period to extract the fundamental frequency vibration component that is in sync with the rotational frequency. The amplitudes A1 and A2 of the fundamental frequency vibration at two measurement points and the phase angles φ1 and φ2 relative to the key phase pulse are obtained through Fast Fourier Transform (FFT).

[0031] Unbalance calculation: The control system has a pre-stored "influence coefficient matrix" for this rotor model.

[0032] S4: Online calibration execution. The control system converts the calculated target phase angle into the number of pulses required by the stepper motors and drives the two stepper motors to move the two counterweights along the guide rails to positions θ1 and θ2, respectively. This process continues while the motors are running and takes approximately 5-10 seconds.

[0033] S5: Effect Verification and Iteration. After the counterweight is moved into place, the system waits approximately 30 seconds for the vibration to stabilize, then collects the vibration signal again and calculates the residual vibration value. If the residual vibration amplitude is lower than the preset safety threshold (e.g., 2.0 mm / s RMS), the correction is complete, and the system enters monitoring mode. If the target is not met, the current state is used as the new initial imbalance, and steps S3-S4 are repeated for fine-tuning iterations. Usually, 1-2 iterations are sufficient to meet the requirements.

[0034] S6: Continuous Monitoring and Adaptation. During long-term motor operation, the system automatically performs a rapid detection and micro-correction once per hour. All historical imbalance data and correction records are stored for trend analysis. The adaptive algorithm in the system software periodically (e.g., weekly) uses historical data to make minor adjustments to the pre-stored influence coefficient matrix, making the model increasingly closer to the dynamic characteristics of the specific rotor in actual operation.

[0035] Example 2: Another embodiment of actuator driving method The main difference from Embodiment 1 lies in the driving method of the dynamic balancing actuator. In this embodiment, the driving unit employs a piezoelectric ceramic micro-actuator. The counterweight is connected to the fixed base via a flexible hinge structure. The piezoelectric ceramic actuator generates precise micrometer-level extension and contraction based on the control voltage, pushing the counterweight to move within a very small arc range. By controlling the magnitude and timing of the voltage applied to the piezoelectric ceramic, the circumferential position of the counterweight can be precisely controlled. This method is suitable for micro-balancing scenarios requiring extremely high precision and fast response (millisecond level), but its correction range is relatively small, making it more suitable for fine maintenance of rotors that have already undergone coarse balancing.

[0036] Example 3: Optimization Example of Control Algorithm In the data processing phase, in addition to the classic influence coefficient method, this embodiment also introduces an artificial neural network (ANN) algorithm into the control system as a supplement. This ANN model takes real-time acquired multi-dimensional data (including vibration amplitude, phase, motor current, and winding temperature) as input and the optimal correction position of the counterweight as output. The model is trained using a large amount of historical correction data. During online operation, the ANN model and the physical model (influence coefficient method) are calculated in parallel, and the final decision is made by combining the outputs of both. This method can better handle nonlinear and time-varying system characteristics, such as imbalance prediction and compensation under severe thermal deformation conditions, further improving the accuracy and robustness of the correction.

[0037] Experimental results: The system from Example 1 was tested on a 75kW industrial centrifugal fan permanent magnet motor (2950 RPM). A 5g unbalanced mass was simulated on one side of the rotor. With the motor running at full speed, the system automatically completed detection, calculation, and correction within 90 seconds, reducing the effective value of the casing vibration velocity from 7.5 mm / s to below 1.8 mm / s. In a subsequent 500-hour simulated dust adhesion experiment, the system, through periodic micro-corrections, consistently maintained the vibration value below 2.5 mm / s, verifying the effectiveness of its online continuous maintenance.

[0038] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0039] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A process for online detection and correction of rotor dynamic balance of an industrial fan permanent magnet motor, characterized by, The method comprises the following steps: An online detection stage: in the rotating state of the rotor, the vibration signals of the rotor in the radial plane are collected by at least two non-contact vibration sensors in real time, and the vibration signals are transmitted to a signal processing and control system; A data processing and unbalance amount calculation stage: the signal processing and control system performs filtering, amplification and analog-to-digital conversion processing on the received vibration signals, and calculates the size and phase of the unbalance mass of the rotor in at least one correction plane based on a preset rotor dynamics model and an influence coefficient method; An online correction stage: according to the size and phase of the unbalance mass calculated, the signal processing and control system generates a correction instruction to drive an integrated dynamic balance correction execution mechanism on the rotor to act, so as to realize online compensation of the unbalance mass in the state of uninterrupted rotation of the rotor until the vibration signal value is reduced to below a preset threshold.

2. A process for online detection and correction of rotor dynamic imbalance of an industrial fan permanent magnet motor as claimed in claim 1, wherein In the online detection stage, the at least two non-contact vibration sensors are arranged axially along the motor shell and are respectively aligned with the two end bearing seats of the rotor or the shell positions close to the bearing seats to collect vibration signals at different axial positions for double-plane dynamic balance calculation.

3. A process for on-line detection and correction of dynamic imbalance of a rotor of an industrial fan permanent magnet motor as claimed in claim 1 or 2, wherein The online detection stage further comprises collecting the key phase or zero position mark signal of the rotor by a phase reference sensor rotating coaxially with the rotor to accurately determine the phase reference datum of the unbalance mass.

4. A process for on-line detection and correction of dynamic imbalance of a rotor of an industrial fan permanent magnet motor as claimed in claim 1, wherein The dynamic balance correction execution mechanism is an embedded non-contact adjustable permanent magnet counterweight device, which comprises: A fixed base fixedly installed on the shaft or end plate of the rotor; A movable counterweight block embedded with a permanent magnet and controllably movable along the circumference and / or radius of the fixed base; A driving unit receiving the correction instruction to drive the movable counterweight block to move to the target phase and radial position to generate a compensation torque by changing the position of the counterweight block relative to the gravity center of the rotor.

5. A process for on-line detection and correction of dynamic imbalance of a rotor of an industrial fan permanent magnet motor as claimed in claim 4 wherein, The driving unit is a micro stepping motor or a piezoelectric ceramic driver connected to the movable counterweight block through a worm gear, a lead screw or a linear motor mechanism.

6. A process for on-line detection and correction of dynamic imbalance of a rotor of an industrial fan permanent magnet motor as claimed in claim 1, wherein In the data processing and unbalance amount calculation stage, the signal processing and control system uses an adaptive filtering algorithm to separate the fundamental frequency vibration component with the same frequency as the rotor rotation frequency from the collected vibration signals for calculating the unbalance amount.

7. A process for on-line detection and correction of dynamic imbalance of a rotor of an industrial fan permanent magnet motor as claimed in claim 1, wherein The process further comprises a self-learning and optimization stage: after multiple online detection and correction cycles, the signal processing and control system records and analyzes historical unbalance amount data and correction effects, automatically optimizes the influence coefficient of the rotor dynamics model, and predicts the dynamic balance state degradation trend of the rotor.