A cloud computing-based energy-saving motor concentricity data acquisition method and system

Abstract

This invention relates to the field of data acquisition and analysis technology, and provides a cloud-based energy-saving motor concentricity data acquisition method and system. By acquiring motor operating information in real time, and determining environmental interference characteristics, motor operating conditions, and extracting key signal features related to motor concentricity based on this information, the system can determine whether the motor operating information contains potential abnormal information, and only upload it to the cloud if potential abnormal information is found. This solves the problem in existing technologies where, in complex industrial environments, environmental noise, changes in operating conditions, and the superposition of multiple fault sources make it difficult for cloud platforms to accurately identify early and weak concentricity anomalies.

Description

Technical Field

[0001] This application relates to the field of data acquisition and analysis technology, and more specifically, to a cloud computing-based energy-saving motor concentricity data acquisition method and system. Background Technology

[0002] In modern industrial production, large electric motors are the core power source driving various key equipment, and their stable operation directly affects production efficiency and energy consumption. The concentricity between the motor rotor and stator is a key indicator for measuring its operational health and energy efficiency. Deviations in concentricity not only lead to premature bearing wear and increased mechanical vibration but also significantly increase electrical energy loss. To achieve preventative maintenance and energy-saving goals, a cloud-based online motor concentricity monitoring system has been introduced. This system uses vibration, temperature, and current sensors to collect motor operating data in real time and uploads it to the cloud for in-depth analysis. This identifies concentricity anomalies and generates early warning information, transforming traditional periodic shutdown inspections into condition-based predictive maintenance.

[0003] However, in actual industrial settings, electric motors are typically deployed in heavy industrial environments filled with various mechanical noises and environmental vibrations, such as continuous, broadband vibrations generated by adjacent heavy equipment, large fans, and fluid pulsations in pipelines. These environmental vibrations can propagate through the foundation, pipelines, or air to the casing of the monitored motor and the sensor mounting point, resulting in a large amount of external interference signals mixed in with the data collected by the sensors. The amplitude of these interference signals is sometimes even comparable to the weak vibration signals caused by internal motor faults, making it difficult for the cloud platform to accurately separate the true motor concentricity anomaly signal from the complex background noise during signal processing.

[0004] Furthermore, the operating conditions of an electric motor are not static. Changes in load and fluid medium adjustments can cause normal fluctuations in vibration, temperature, and current signals. These signal fluctuations caused by changes in operating conditions can, in some cases, be very similar to abnormal signals caused by poor concentricity. This can easily lead the cloud platform to misinterpret normal operating condition fluctuations as concentricity anomalies, resulting in false alarms. Simultaneously, during long-term operation, the wear and aging of internal components can also generate weak vibrations, localized heating, or current harmonics. These signal characteristics may overlap with or influence those caused by poor concentricity, making it difficult for the cloud platform to accurately distinguish multiple fault sources.

[0005] Faced with these challenges, cloud platforms often need to handle significant uncertainties in their concentricity assessment methods. Due to environmental noise interference, the complexity of changing operating conditions, and the superposition of multiple fault sources, the features extracted from raw data are often not pure or clear enough. This ambiguity in data interpretation easily puts traditional rule-based or threshold-based diagnostic methods in a dilemma: overly sensitive thresholds lead to numerous false alarms, reducing system reliability; overly conservative thresholds may miss early warnings of concentricity deviations. Therefore, under the current data processing framework, while ensuring a low false alarm rate, the system struggles to simultaneously achieve real-time and accurate identification of early, subtle concentricity anomalies, significantly diminishing its energy-saving and preventative maintenance value.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] This application discloses a cloud computing-based energy-saving motor concentricity data acquisition method and system, aiming to solve the technical problem that in the prior art, in complex industrial environments, the online monitoring system for motor concentricity is easily affected by environmental noise, changes in operating conditions and the superposition of multiple fault sources, making it difficult for the cloud platform to accurately identify early and weak concentricity anomalies, thereby reducing the reliability of the system and affecting the value of energy saving and preventive maintenance.

[0008] The technical solution of this application is as follows:

[0009] In a first aspect, this application discloses a cloud computing-based energy-saving motor concentricity data acquisition method, applied to a data acquisition unit, the method comprising:

[0010] Real-time acquisition of motor operating information; this motor operating information includes vibration signals, temperature signals, and current signals;

[0011] Based on the motor's operating information, we determine the environmental interference characteristics, the motor's operating conditions, and extract key signal features related to the motor's concentricity, and determine the changing trend of these key signal features.

[0012] Based on the changing trend of this key signal characteristic, determine whether the motor operation information contains potential abnormal information;

[0013] In response to the motor operation information containing potential abnormal information, the motor operation information is uploaded to the cloud.

[0014] Furthermore, this application also discloses a cloud-based energy-saving motor concentricity data acquisition method, wherein, in response to the motor operating information containing potential abnormal information, the motor operating information is uploaded to the cloud, including: in response to the motor operating information containing potential abnormal information, the motor operating information, environmental interference characteristics, and motor operating conditions are uploaded to the cloud; the cloud performs noise separation on the motor operating information based on the environmental interference characteristics and motor operating conditions, and analyzes the separated data to identify motor concentricity abnormalities.

[0015] Based on this, this application further proposes a cloud-based energy-saving motor concentricity data acquisition method. This method further includes: performing frequency domain analysis on the vibration signal to extract vibration energy related to the motor speed; monitoring changes in the vibration energy; identifying whether the vibration energy exhibits intermittent fluctuations; in response to the presence of intermittent fluctuations that are unrelated to changes in motor operating conditions, marking them as potential external interference and sending a summary of the potential external interference to the cloud; when the cloud receives potential external interference of a specific frequency from the same data acquisition unit within a certain period, and none of the potential external interferences cause the cloud to identify abnormal motor concentricity, then sending a confirmation interference mode command to the data acquisition unit.

[0016] Furthermore, this application also discloses a cloud computing-based energy-saving motor concentricity data acquisition method, wherein, based on the vibration signal, frequency domain analysis is performed to extract vibration energy related to motor speed, monitor changes in the vibration energy, and identify whether the vibration energy has intermittent fluctuations, including: performing fast Fourier transform analysis on the vibration signal to extract second harmonic vibration energy; obtaining the moving average sequence of the second harmonic vibration energy within a time window and determining it as the key signal feature; analyzing the moving average sequence to identify whether the vibration energy has intermittent fluctuations, and recording the characteristics of the intermittent fluctuations; wherein, the intermittent fluctuation is characterized by a sudden increase in the moving average of the second harmonic vibration energy exceeding a preset relative threshold within a short period of time, followed by a rapid drop back to near the original level within the next short period of time, with a quiet period between the increase and the drop.

[0017] In some preferred embodiments, this application also discloses a cloud computing-based method for acquiring concentricity data of energy-saving motors, wherein the characteristics of the intermittent fluctuations include: occurrence time, duration, peak frequency, and peak amplitude.

[0018] As a technical improvement, this application also discloses a cloud-based energy-saving motor concentricity data acquisition method. The method involves sending an interference mode confirmation command to the data acquisition unit, followed by: identifying interference features of the intermittent fluctuation and adding them to the environmental interference features to update the environmental interference features; processing the vibration energy based on the updated environmental interference features to suppress interference, and performing trend analysis on the vibration energy after interference suppression; when the trend analysis indicates that the vibration energy continuously rises above a preset threshold, it is marked as a potential concentricity anomaly, and the motor operating information, the updated environmental interference features, the motor operating conditions, and the changing trend of the key signal features are sent to the cloud.

[0019] As a further improvement, this application also discloses a cloud computing-based energy-saving motor concentricity data acquisition method. This method determines environmental interference characteristics and motor operating conditions based on the motor's operating information, including: during motor shutdown or stable operation, identifying the frequency range and intensity of environmental noise through time-frequency analysis of the vibration signal, and generating environmental interference characteristics; and classifying the motor's operating state into no-load, light-load, or heavy-load based on a comparison of the current signal with a preset threshold, thus determining the motor's operating condition. This technical solution enables accurate identification of environmental interference characteristics and motor operating conditions, providing crucial contextual information for subsequent signal processing and anomaly detection, and effectively distinguishing the impact of environmental noise and changes in operating conditions on motor operating information.

[0020] As a system extension, this application also discloses a cloud computing-based energy-saving motor concentricity data acquisition method, which extracts key signal features related to motor concentricity, including: performing real-time spectrum analysis on the vibration signal to extract the second harmonic vibration energy related to motor speed and the first vibration energy of the frequency component related to component failure; determining whether the increase of the second harmonic vibration energy is accompanied by a synchronous increase of the first vibration energy or shows a specific correlation; when the second harmonic vibration energy increases and the first vibration energy remains stable or shows an uncorrelated change, the second harmonic vibration energy is marked as a key signal feature related to motor concentricity.

[0021] To improve the solution, this application also discloses a cloud computing-based energy-saving motor concentricity data acquisition method, wherein, based on the changing trend of the key signal feature, it is determined whether the motor operation information contains potential abnormal information, including: calculating the signal moving average of the key signal feature in real time, performing linear trend analysis on the signal moving average; and, based on the linear trend analysis results, determining that the motor operation information contains potential abnormal information when the growth rate of the key signal feature exceeds a preset threshold and continues for a certain period of time.

[0022] Secondly, this application also discloses a cloud-based energy-saving motor concentricity data acquisition system, applied to a data acquisition unit. The system includes: an acquisition module for real-time acquisition of motor operating information, including vibration signals, temperature signals, and current signals; a determination module for determining environmental interference characteristics and motor operating conditions based on the motor operating information, and extracting key signal features related to motor concentricity, and determining the changing trend of the key signal features; a judgment module for determining whether the motor operating information contains potential abnormal information based on the changing trend of the key signal features; and an upload module for uploading the motor operating information to the cloud in response to the presence of potential abnormal information.

[0023] Beneficial effects

[0024] This application discloses a cloud-based energy-saving motor concentricity data acquisition method. By collecting motor operating information in real time and determining environmental interference characteristics, motor operating conditions, and extracting key signal features related to motor concentricity based on this information, the method determines whether the motor operating information contains potential abnormal information and only uploads it to the cloud when such information is present. This method effectively solves the problem in existing technologies where, in complex industrial environments, environmental noise, changes in operating conditions, and the superposition of multiple fault sources make it difficult for cloud platforms to accurately identify early and weak concentricity anomalies. By performing preliminary intelligent analysis and screening locally on the data acquisition unit, this application significantly reduces unnecessary data uploads, lowers communication bandwidth usage and cloud processing load, thereby improving the overall operating efficiency and energy-saving effect of the system. Simultaneously, this local preprocessing mechanism allows the cloud to receive more targeted data, enabling in-depth analysis based on environmental interference characteristics and operating conditions. This greatly improves the accuracy and timeliness of motor concentricity anomaly identification, avoiding false alarms and missed alarms, and providing reliable technical support for achieving preventative maintenance and energy-saving goals for motors. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the steps of the cloud computing-based energy-saving motor concentricity data acquisition method disclosed in an embodiment of the present invention.

[0026] Figure 2 This is a schematic diagram of the structure of the cloud computing-based energy-saving motor concentricity data acquisition system disclosed in an embodiment of the present invention. Detailed Implementation

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which these embodiments belong; the terminology used herein and in the specification of the application is for the purpose of describing particular embodiments only and is not intended to limit these embodiments; the terms "comprising" and "having," and any variations thereof, in the specification of these embodiments and the foregoing drawings, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification of these embodiments and the foregoing drawings are used to distinguish different objects, not to describe a particular order.

[0028] The implementation details of the technical solution in this embodiment are described in detail below:

[0029] In heavy industrial environments, traditional motor concentricity monitoring systems often suffer from insufficient purity and clarity in the features extracted from raw data due to environmental noise interference, complex operating conditions, and the superposition of multiple fault sources. This ambiguity in data interpretation easily leads traditional rule-based or threshold-based diagnostic methods into a dilemma: overly sensitive thresholds result in numerous false alarms, reducing system reliability; overly conservative thresholds may miss early warnings of concentricity deviations. Therefore, within the current data processing framework, it is difficult for the system to simultaneously ensure a low false alarm rate while achieving real-time and accurate identification of early, subtle concentricity anomalies, significantly diminishing its value in energy saving and preventative maintenance.

[0030] To address this, this application proposes a cloud computing-based energy-saving motor concentricity data acquisition method, applied to a data acquisition unit, such as... Figure 1 As shown, the method includes:

[0031] S101, Real-time acquisition of motor operation information; the motor operation information includes vibration signals, temperature signals, and current signals;

[0032] S102, determine the environmental interference characteristics and motor operating conditions based on the motor operating information, and extract key signal features related to the motor concentricity, and determine the changing trend of the key signal features;

[0033] S103, Based on the changing trend of the key signal characteristics, determine whether the motor operation information contains potential abnormal information;

[0034] S104, in response to the motor operation information containing potential abnormal information, the motor operation information is uploaded to the cloud.

[0035] This application effectively filters out a large amount of normal operating data and known interference by performing preliminary intelligent analysis and judgment at the data acquisition unit side. Relevant information is only uploaded to the cloud when a potential anomaly is detected, thus significantly reducing data transmission volume and cloud processing load, achieving energy-saving goals. Simultaneously, through comprehensive analysis of environmental interference characteristics, motor operating conditions, and the changing trends of key signal characteristics, the accuracy of anomaly detection is improved, avoiding false alarms and missed alarms, and providing more reliable data support for preventative motor maintenance.

[0036] To better understand the technical solution proposed in this application, it is necessary to explain some key terms involved. A data acquisition unit refers to a hardware device deployed at the motor site, responsible for real-time monitoring of the motor's operating status. It integrates sensors, a data processor, and a communication module. Motor operating information refers to the raw data acquired by the data acquisition unit through sensors, including vibration signals, temperature signals, and current signals. These signals are the basis for assessing the motor's health status. Environmental interference characteristics refer to external factors that affect the accuracy of motor operating information, such as environmental noise and vibration from adjacent equipment. Identifying these factors helps in subsequent data purification. Motor operating conditions refer to the motor's operating status within a specific time period, such as load size and speed. These conditions affect the normal fluctuation range of motor signals. Key signal characteristics are signal indicators directly related to abnormal motor concentricity; their changing trends are the core basis for judging whether the motor has a concentricity problem. Potential anomaly information refers to data patterns that, based on preliminary analysis, may indicate abnormal motor concentricity. The cloud refers to a remote data processing and storage center responsible for receiving anomaly information uploaded by the data acquisition unit and performing deeper diagnostic analysis.

[0037] The core of the cloud-based energy-saving motor concentricity data acquisition method proposed in this application lies in achieving efficient and accurate motor concentricity monitoring through a combination of local intelligent processing and cloud-based collaborative analysis. Specifically, this method first acquires motor operating information in real time. For example, the data acquisition unit can be configured with a triaxial accelerometer to acquire vibration signals, a thermocouple or PT100 sensor to acquire temperature signals, and a Hall effect sensor or current transformer to acquire current signals. These sensors are installed in key parts of the motor, such as bearing housings and the motor casing, to ensure the acquisition of representative operating data. The acquisition frequency and accuracy can be adjusted according to the type of motor and monitoring requirements; for example, the sampling rate of the vibration signal can be set to 10kHz, and the sampling rates of the temperature and current signals can be set to 100Hz.

[0038] Next, based on the collected motor operating information, environmental interference characteristics and motor operating conditions are determined, and key signal features related to motor concentricity are extracted, while the changing trends of these key signal features are also determined. For example, when determining environmental interference characteristics, the data acquisition unit can perform time-frequency analysis on vibration signals during motor shutdown or stable operation to identify the frequency range and intensity of environmental noise and store it as environmental interference characteristics. When determining motor operating conditions, the average or effective value of the current signal can be monitored and compared with a preset threshold to classify the motor operating state as no-load, light-load, or heavy-load. When extracting key signal features, real-time spectrum analysis of the vibration signal can be performed to identify the second harmonic vibration energy related to motor speed, as second harmonic vibration energy is usually closely related to poor motor concentricity. Simultaneously, frequency components related to component failures such as bearing failures and gear failures can be extracted as the first vibration energy. By comparing whether the rise in second harmonic vibration energy is accompanied by a synchronous rise in the first vibration energy or shows a specific correlation, the second harmonic vibration energy can be more accurately marked as a key signal feature related to motor concentricity. For example, when the second harmonic vibration energy increases while the first harmonic vibration energy remains stable or shows no correlation, the change in the second harmonic vibration energy can be considered more likely to be related to concentricity anomalies. The changing trend of key signal characteristics can be determined by calculating the moving average of the key signal characteristics in real time and performing linear trend analysis on this moving average to assess its growth rate and duration.

[0039] Subsequently, based on the changing trends of key signal characteristics, it is determined whether the motor operation information contains potential anomalies. For example, if the linear trend analysis results show that the growth rate of key signal characteristics exceeds a preset threshold and persists for a certain period of time, it can be determined that the motor operation information contains potential anomalies. This preset threshold and duration can be set based on historical data and expert experience to balance the false alarm rate and the false negative rate. Finally, in response to the presence of potential anomalies in the motor operation information, the data acquisition unit uploads the motor operation information to the cloud. For example, when potential anomalies are identified, the data acquisition unit can package and upload the original motor operation information and preliminary analysis results (such as environmental interference characteristics, motor operating conditions, and changing trends of key signal characteristics) to the cloud via communication methods such as cellular networks, Wi-Fi, or Ethernet.

[0040] The method proposed in this application effectively filters out a large amount of normal operating data and known interference by performing preliminary intelligent analysis and judgment at the data acquisition unit side. It only uploads relevant information to the cloud when a potential anomaly is detected, thereby significantly reducing data transmission volume and cloud processing load, achieving energy-saving goals. Traditional methods often upload all collected raw data to the cloud indiscriminately, resulting in massive data volumes, high transmission bandwidth and storage costs, and the cloud needing to process a large amount of redundant information. In contrast, this application, through localized intelligent judgment, uploads only information containing potential anomalies, greatly optimizing the data flow. For example, during normal motor operation, the data acquisition unit can continuously monitor and perform local analysis without frequent communication with the cloud. The data upload mechanism is only triggered when the growth rate of a key signal characteristic (such as second harmonic vibration energy) exceeds a preset threshold and persists for a period of time. This on-demand upload strategy not only reduces network traffic and cloud computing resource consumption but also allows the cloud to focus more on in-depth diagnosis of truly abnormal data, improving the overall system efficiency and response speed. Furthermore, through comprehensive analysis of environmental interference characteristics, motor operating conditions, and the changing trends of key signal characteristics, this application improves the accuracy of anomaly detection, avoids false alarms and missed alarms, and provides more reliable data support for preventive maintenance of motors. This local intelligence and cloud-based collaborative model enables the system to ensure a low false alarm rate while also achieving real-time and accurate identification of early, subtle concentricity anomalies, thereby maximizing its energy-saving and preventive maintenance value.

[0041] In some embodiments described above, when the motor operating information contains potential abnormal information, the data acquisition unit uploads the motor operating information to the cloud. However, in practical applications, simply uploading the raw motor operating information may pose challenges for the cloud in identifying concentricity anomalies. Specifically, due to the lack of key contextual information such as environmental interference characteristics and motor operating conditions, the cloud may find it difficult to effectively distinguish between genuine abnormal signals and normal fluctuations caused by environmental noise or changes in operating conditions, thus affecting the accuracy and efficiency of anomaly detection. Therefore, this application further proposes a scheme to refine the data uploaded to the cloud and clarify the cloud processing procedure.

[0042] In response to this, this application further proposes the following steps for uploading motor operating information to the cloud in response to the motor operating information containing potential abnormal information: uploading the motor operating information, environmental interference characteristics, and motor operating conditions to the cloud; the cloud performs noise separation on the motor operating information based on the environmental interference characteristics and motor operating conditions, and analyzes the separated data to identify abnormal motor concentricity.

[0043] Specifically, when the data acquisition unit determines that the motor operating information contains potential anomalies, it not only uploads the original motor operating information but also simultaneously uploads pre-determined environmental interference characteristics and the current motor operating condition. The motor operating information may include vibration signals, temperature signals, and current signals. Environmental interference characteristics refer to external factors that may affect the acquisition and analysis of motor operating information, such as the frequency range and intensity of ambient noise. The motor operating condition describes the current operating state of the motor, such as no-load, light-load, or heavy-load.

[0044] Furthermore, after receiving this information, the cloud platform utilizes the uploaded environmental interference characteristics and motor operating conditions to perform refined noise separation processing on the motor operating information. For example, based on the noise frequency and intensity recorded in the environmental interference characteristics, adaptive filtering, wavelet denoising, and other techniques can be used to effectively remove environmental noise components from the vibration signal. Simultaneously, combined with the motor operating conditions, signal characteristics under different operating conditions can be normalized or baseline adjusted to eliminate the impact of operating condition changes on the judgment of concentricity anomalies. After completing noise separation, the cloud platform will conduct in-depth analysis of the processed data to identify motor concentricity anomalies.

[0045] The solution presented in this application improves the accuracy of motor concentricity anomaly identification because it provides environmental interference characteristics and motor operating conditions simultaneously when uploading motor operating information. This provides the cloud with the contextual information needed for more precise data processing and analysis. It is precisely because the cloud can effectively separate noise and compensate for operating conditions based on this additional information that it becomes possible to accurately extract true anomaly signals related to motor concentricity from complex signals. For example, the introduction of environmental interference characteristics allows the cloud to selectively filter out external noise, avoiding misjudging noise as concentricity anomalies; while providing motor operating conditions helps the cloud understand the normal behavior patterns of the motor under different loads, thereby more accurately identifying concentricity anomalies that deviate from the normal pattern.

[0046] Through the above technical solution, this application can significantly improve the accuracy and reliability of cloud-based identification of motor concentricity anomalies. Because the cloud fully considers the impact of environmental interference and operating conditions when analyzing motor operating information, it can effectively reduce false alarms and missed alarms, avoiding erroneous judgments caused by environmental noise or changes in normal operating conditions. Furthermore, this refined data upload and cloud processing mechanism enables the cloud to perform data analysis more efficiently, thereby providing a more accurate basis for predictive maintenance and fault diagnosis of motors, further improving the energy efficiency and intelligence of the entire data acquisition method.

[0047] In some preferred embodiments, suppose a data acquisition unit detects a sudden increase in the second harmonic vibration energy of a motor's vibration signal during nighttime operation, initially identifying it as a potential anomaly. According to the solution of this application, the data acquisition unit uploads not only the vibration signal but also current environmental interference characteristics (e.g., specific frequency noise generated by the startup of a large piece of equipment in a factory at night) and the motor's operating condition (e.g., the motor is operating under light load). Upon receiving this information, the cloud first uses the environmental interference characteristics to separate noise from the vibration signal. It finds that the elevated portion of the vibration signal highly matches the frequency of the environmental noise, and after removing the noise, the second harmonic vibration energy does not show a continuous upward trend. Simultaneously, considering the light-load operating condition of the motor, the cloud determines that the increase in vibration energy is not caused by abnormal motor concentricity but by external environmental interference. Therefore, the cloud avoids misjudging environmental interference as abnormal motor concentricity, thereby improving the accuracy of the diagnosis.

[0048] In some embodiments described above, the data acquisition unit primarily relies on the changing trends of key signal characteristics when determining whether motor operating information contains potential abnormal information. However, in actual industrial environments, motors may be affected by various external disturbances during operation, such as the start-up and shutdown of nearby equipment, structural resonance, or instantaneous impacts. These disturbances may cause vibration signals to exhibit fluctuations similar to motor concentricity anomalies, leading to misjudgments as potential abnormal information. If this problem is not addressed, the system may frequently upload non-concentricity anomaly data to the cloud, increasing data transmission burden and cloud processing costs, while simultaneously reducing the system's efficiency and accuracy in identifying true concentricity anomalies. Therefore, this application further proposes an optimization scheme that improves the accuracy of motor concentricity anomaly judgment by performing more refined analysis of vibration signals to identify and confirm external interference patterns.

[0049] The above method further includes: performing frequency domain analysis based on the vibration signal to extract vibration energy related to the motor speed, monitoring changes in the vibration energy, and identifying whether the vibration energy exhibits intermittent fluctuations; in response to the presence of intermittent fluctuations, and if the intermittent fluctuations are unrelated to changes in the motor's operating conditions, marking them as potential external interference and sending a summary of the potential external interference to the cloud; when the cloud receives potential external interference of a specific frequency from the same data acquisition unit within a certain period of time, and none of the potential external interferences cause the cloud to identify abnormal motor concentricity, then sending a confirmation interference mode command to the data acquisition unit.

[0050] Specifically, frequency domain analysis based on vibration signals involves processing the real-time acquired vibration signals using methods such as Fourier transform to convert them from the time domain to the frequency domain, allowing observation of the energy distribution of different frequency components. Vibration energy related to motor speed typically refers to the vibration energy at the motor's fundamental frequency and its harmonics, such as second harmonic vibration energy. Its changes are closely related to abnormal motor concentricity. By continuously monitoring the vibration energy at these specific frequencies, its trends over time can be captured. Identifying intermittent fluctuations in vibration energy refers to detecting sudden increases or decreases in vibration energy within a short period, followed by a return to normal levels. Such fluctuations may indicate transient external disturbances.

[0051] Furthermore, when such intermittent fluctuations are detected, it is necessary to determine whether they are related to changes in the motor's operating conditions. For example, if operating parameters such as motor speed and load change significantly, fluctuations in vibration energy may be a normal response. However, if the fluctuations are unrelated to changes in operating conditions, there is reason to suspect that they are potential external disturbances. In this case, the fluctuation will be marked as a potential external disturbance, and its summary information will be sent to the cloud. The summary information may include key characteristics such as the fluctuation's occurrence time, duration, peak frequency, and peak amplitude, so that the cloud can perform further analysis.

[0052] Upon receiving a summary of potential external interference from the data acquisition unit, the cloud does not immediately interpret it as a concentricity anomaly. Instead, it continuously collects such information from the same data acquisition unit. If, over a predetermined period (e.g., several hours or days), the cloud repeatedly receives summaries of potential external interference at a specific frequency, and during this period, after performing noise separation and analysis on the uploaded motor operating information, the cloud fails to identify a genuine motor concentricity anomaly, it determines that these recurring fluctuation patterns that do not cause concentricity anomalies are likely a stable external interference pattern. At this point, the cloud sends a confirmation instruction to the data acquisition unit, informing it that the specific pattern has been identified as environmental interference.

[0053] This application's solution, by introducing frequency domain analysis of vibration signals and an intermittent fluctuation identification mechanism, can proactively identify and distinguish vibration fluctuations caused by external interference from vibration changes caused by abnormal motor concentricity. When the data acquisition unit detects intermittent vibration fluctuations unrelated to the motor's operating conditions, it does not immediately upload all motor operating information as a potential anomaly. Instead, it first marks it as a potential external interference and sends a summary message to the cloud. Through long-term observation and pattern matching, the cloud can confirm that these recurring fluctuation patterns that do not lead to actual concentricity anomalies are stable external interferences. Once the interference pattern is confirmed, the data acquisition unit can use this information to preprocess subsequent vibration signals locally, thereby effectively suppressing the influence of known interferences, making subsequent concentricity anomaly judgments more accurate, and avoiding false alarms caused by external interference.

[0054] Through the above technical solution, this application can significantly improve the accuracy and reliability of motor concentricity anomaly detection. By identifying and confirming external interference patterns, false alarms caused by environmental noise or instantaneous external events can be effectively reduced, avoiding unnecessary uploading of motor operation information and cloud processing, thereby reducing data transmission bandwidth and cloud computing resource consumption. Furthermore, this solution allows the data acquisition unit to focus more on actual motor concentricity anomalies, improving the system's response speed and diagnostic efficiency for critical faults, and providing more accurate data support for predictive maintenance of equipment.

[0055] In some preferred embodiments, it is assumed that a data acquisition unit monitors a motor located in a production workshop. During normal operation, the second harmonic vibration energy of the motor's vibration signal typically remains at a stable level. However, at 10:00 AM and 3:00 PM each day, when a large stamping machine in the workshop starts up, a momentary impact is generated, causing a brief, intermittent fluctuation in the vibration signal of the monitored motor at a specific frequency. This fluctuation lasts for about 10 seconds before quickly returning to normal.

[0056] According to the scheme of this application, the data acquisition unit performs frequency domain analysis based on the vibration signal and identifies intermittent fluctuations in second-harmonic vibration energy related to motor speed. Since these fluctuations are unrelated to changes in motor operating conditions (such as speed and load), the data acquisition unit marks them as potential external interference and sends a summary containing information such as occurrence time, duration, peak frequency, and peak amplitude to the cloud. Over a period of time, such as a week, the cloud continuously receives summary information of potential external interference from the data acquisition unit regarding the same specific frequency (e.g., the operating frequency of the press). During this period, the cloud performs noise separation and analysis on the uploaded motor operating information and does not identify any abnormal motor concentricity. Based on this, the cloud determines that this recurring fluctuation pattern that does not cause concentricity abnormalities is a stable external interference pattern. Subsequently, the cloud sends a confirmation interference pattern command to the data acquisition unit, informing it that this specific fluctuation pattern has been confirmed as environmental interference. In this way, the data acquisition unit can identify and effectively process the confirmed interference pattern during subsequent monitoring, avoiding misjudging it as an abnormality in motor concentricity, thereby improving the accuracy of anomaly detection and reducing unnecessary cloud data uploads and processing.

[0057] Specifically, the steps described above—performing frequency domain analysis based on vibration signals to extract vibration energy related to motor speed, monitoring changes in vibration energy, and identifying whether intermittent fluctuations in vibration energy—can be further refined into the following operations: performing frequency domain analysis based on the vibration signals to extract vibration energy related to motor speed, monitoring changes in the vibration energy, and identifying whether intermittent fluctuations in the vibration energy exist, including:

[0058] Fast Fourier Transform analysis was performed on the vibration signal to extract the second harmonic vibration energy;

[0059] Obtain the moving average sequence of the second harmonic vibration energy within a time window, and determine it as the key signal feature;

[0060] Analyze the moving average sequence to identify whether the vibration energy has intermittent fluctuations, and record the characteristics of the intermittent fluctuations; wherein, the intermittent fluctuation is that within a short period of time, the moving average of the second harmonic vibration energy suddenly rises above a preset relative threshold, and then quickly falls back to near the original level in the next short period of time, and there is a quiet period between the rise and fall.

[0061] Specifically, Fast Fourier Transform (FFT) analysis is performed on the vibration signal to convert the time-domain vibration signal into a frequency-domain signal, thereby revealing its various frequency components and their corresponding energy or amplitude. Based on this, the second harmonic vibration energy is extracted, because abnormal motor concentricity typically produces a significant vibration response at the second harmonic of the motor's rotational frequency.

[0062] Furthermore, to smooth the data and better capture trends, a moving average sequence of second harmonic vibration energy within a time window is obtained. This moving average sequence is identified as a key signal feature related to motor concentricity because it reflects the dynamic changes in second harmonic vibration energy. The length of the time window can be adjusted according to the actual application scenario and the required response speed, for example, it can be set to several seconds to tens of seconds.

[0063] Based on this, the moving average sequence is analyzed to identify whether intermittent fluctuations in vibration energy exist. These intermittent fluctuations are precisely defined as follows: within a short time period, the moving average of the second harmonic vibration energy suddenly rises and exceeds a preset relative threshold, then rapidly falls back to near its original level within the next short time period, with a quiet period between the rise and fall. The preset relative threshold can be set based on historical data or expert experience to distinguish between normal fluctuations and abnormal rises. The quiet period refers to the time between the rise and fall during which the vibration energy remains at a high level or changes slowly, which helps distinguish between instantaneous impacts and persistent disturbances. Once such intermittent fluctuations are identified, their characteristics, such as the time of occurrence, duration, peak frequency, and peak amplitude, are recorded.

[0064] This application's solution utilizes Fast Fourier Transform (FFT) analysis on vibration signals, focusing on extracting second-harmonic vibration energy. This effectively separates features highly correlated with motor concentricity anomalies from complex vibration signals. Further obtaining the moving average sequence of second-harmonic vibration energy effectively filters out transient noise, making the trend analysis of vibration energy changes more stable and reliable. Crucially, this solution precisely defines "intermittent fluctuations," requiring them to exhibit specific patterns of "sudden rise," "rapid fall," and "quiet periods." This precise pattern recognition mechanism allows the data acquisition unit to more accurately distinguish between transient vibration increases caused by external instantaneous impacts or environmental disturbances and persistent or trend-based vibration changes caused by motor concentricity anomalies. This approach avoids misjudging intermittent interference caused by non-concentricity anomalies as potential concentricity problems, thereby improving the efficiency and accuracy of subsequent cloud-based analysis.

[0065] Through the above technical solution, this application can significantly improve the accuracy of identifying potential external interference in motor vibration signals. By performing moving average processing on the second harmonic vibration energy and accurately defining the characteristics of intermittent fluctuations, random noise and atypical fluctuations can be effectively filtered out, thereby more accurately capturing the unique signal patterns of external interference. This refined identification mechanism allows the data acquisition unit to perform preliminary intelligent judgment on the vibration signal locally, reducing the need to upload a large amount of irrelevant data to the cloud due to false alarms, thus reducing data transmission volume and cloud processing load, and further improving the energy efficiency and effectiveness of the entire data acquisition method. At the same time, it also provides a cleaner and more reliable data foundation for subsequent cloud-based identification of motor concentricity anomalies.

[0066] Specifically, the characteristics of the aforementioned intermittent fluctuations can be further clarified. These intermittent fluctuations include: occurrence time, duration, peak frequency, and peak amplitude.

[0067] Specifically, the occurrence time refers to the point in time when the intermittent fluctuation begins, usually recorded as a timestamp, with the aim of accurately tracking the moment the abnormal event occurs. Duration refers to the length of time from the start to the end of the intermittent fluctuation, used to quantify the sustained impact of the fluctuation. Peak frequency refers to the frequency component corresponding to the peak vibrational energy during the intermittent fluctuation, which helps identify the nature of the interference source. Peak amplitude refers to the maximum amplitude of the vibrational energy during the intermittent fluctuation, used to measure the intensity of the interference.

[0068] The solution presented in this application, by clearly defining the specific characteristics of intermittent fluctuations, enables more refined identification and analysis of potential external interference. By recording the occurrence time, duration, peak frequency, and peak amplitude, the data acquisition unit can provide more comprehensive interference information. These detailed characteristics help the system more accurately understand the nature of the fluctuations, such as distinguishing between transient impacts and persistent interference, or identifying resonance phenomena at specific frequencies. This provides more accurate and discriminative data support for subsequent noise separation and interference pattern confirmation in the cloud.

[0069] The aforementioned technical solutions enable standardized and quantifiable descriptions of intermittent fluctuations, significantly improving the accuracy and traceability of identifying potential external interference. This detailed feature recording helps the cloud more effectively identify and differentiate genuine motor concentricity anomalies from external interference, avoiding false alarms caused by external interference and thus enhancing the reliability and energy efficiency of the entire data acquisition method. Furthermore, long-term monitoring and analysis of these features can establish an interference pattern library, further optimizing interference suppression strategies.

[0070] In some of the embodiments described above in this application, although potential external interference can be identified and confirmed, if the confirmed interference patterns are not effectively processed, false alarms may occur in the subsequent judgment of motor concentricity anomalies, affecting the accuracy of the diagnosis. Therefore, this application further proposes that after receiving the instruction to confirm the interference pattern, the method further includes:

[0071] The intermittent fluctuations are identified as disturbance features and added to the environmental disturbance features to update the environmental disturbance features;

[0072] The vibration energy is processed based on the updated environmental disturbance characteristics to suppress the disturbance, and a trend analysis is performed on the vibration energy after the disturbance is suppressed.

[0073] When the trend analysis indicates that the vibration energy continues to rise beyond a preset threshold, it is marked as a potential concentricity anomaly, and the motor operation information, the updated environmental interference characteristics, the motor operating conditions, and the changing trends of the key signal characteristics are sent to the cloud.

[0074] Specifically, when the data acquisition unit receives a confirmation instruction for the interference pattern from the cloud, it indicates that the cloud has confirmed that the previously reported intermittent fluctuations were not caused by abnormal motor concentricity, but by a specific external interference source. At this time, the characteristics of the intermittent fluctuations, such as their occurrence time, duration, peak frequency, and peak amplitude, will be identified as an interference feature. This interference feature is then added to the environmental interference features, thereby updating the environmental interference features. The environmental interference features can be understood as a set of noise and interference patterns describing factors outside the motor's operating environment. By incorporating the confirmed intermittent fluctuation patterns, the data acquisition unit can learn and adapt to specific external interferences.

[0075] Furthermore, based on the updated environmental interference characteristics, the real-time acquired vibration energy is processed to suppress interference. This processing aims to effectively remove or reduce components matching the identified interference patterns from the original vibration signal, thereby obtaining purer vibration energy data that better reflects the motor's operating state. For example, adaptive filtering, notch filtering, or pattern-matching-based signal separation techniques can be used to perform interference suppression. After interference suppression is completed, trend analysis is performed on the vibration energy after interference suppression. The trend analysis can employ methods such as moving average, linear regression, or exponential smoothing to monitor the long-term trend of vibration energy changes, rather than short-term fluctuations caused by interference.

[0076] When the trend analysis indicates that the vibration energy continues to rise and exceeds a preset threshold, it suggests that the motor vibration level is still abnormally increasing even after excluding known interferences. In this case, the situation will be marked as a potential concentricity anomaly. In response, the data acquisition unit will send detailed data to the cloud, including the motor operating information, updated environmental interference characteristics, motor operating conditions, and the changing trends of key signal characteristics. This is intended to report the abnormal information, which has been pre-processed and initially assessed locally, to the cloud for further diagnosis and analysis.

[0077] This application's solution incorporates intermittent fluctuation patterns confirmed in the cloud into local environmental interference characteristics, enabling the data acquisition unit to adaptively learn and identify external interference. Due to this learning mechanism, the data acquisition unit can more accurately suppress known interferences during subsequent vibration energy processing using updated environmental interference characteristics, effectively avoiding misjudging external interference as motor concentricity anomalies. By performing trend analysis on the vibration energy after interference suppression, this application ensures that only when the motor's own vibration level shows a continuous and significant abnormal increase will it be marked as a potential concentricity anomaly and reported to the cloud. This mechanism allows the local data acquisition unit to make more intelligent preliminary judgments, reducing the processing burden on the cloud and improving the overall accuracy and efficiency of diagnosis.

[0078] Through the above technical solution, this application can significantly improve the accuracy of motor concentricity anomaly detection and effectively reduce false alarms caused by external intermittent interference. By integrating the interference patterns confirmed in the cloud into the local environmental interference characteristics, the data acquisition unit gains adaptive learning and anti-interference capabilities, enabling it to operate stably and reliably in complex and ever-changing environments. Furthermore, this solution optimizes the data upload strategy, uploading detailed data only when a potential concentricity anomaly is confirmed, thereby reducing data transmission volume and cloud processing load, and improving the energy efficiency and operational efficiency of the entire system.

[0079] In some preferred embodiments, a specific example is given below. Suppose that during the operation of a motor in a factory, its data acquisition unit continuously monitors intermittent fluctuations in the vibration signal. After a period of observation and cloud analysis, the cloud confirms that these intermittent fluctuations are caused by a nearby heavy piece of equipment that starts periodically, and are not due to abnormal motor concentricity. At this point, the cloud sends a confirmation interference mode command to the data acquisition unit. Upon receiving the command, the data acquisition unit immediately identifies the characteristics of the intermittent fluctuations caused by the heavy equipment (e.g., a specific frequency range, duration, amplitude, etc.) as interference features and adds them to the locally stored environmental interference features, thereby updating the environmental interference features. Subsequently, when the heavy equipment starts again and generates similar interference, the data acquisition unit adaptively processes the real-time acquired vibration energy based on the updated environmental interference features, effectively suppressing the interference component caused by the heavy equipment. Then, the data acquisition unit performs trend analysis on the vibration energy after interference suppression. For example, by calculating its long-term moving average and monitoring its rate of change. If, even after excluding interference from heavy equipment, the trend analysis of the vibration energy still indicates a continuous increase and exceeds a preset threshold, then the data acquisition unit will mark it as a potential concentricity anomaly. At this point, the data acquisition unit will upload detailed information, including the original motor operating information, updated environmental interference characteristics, current motor operating conditions, and the changing trends of key signal characteristics, to the cloud for final diagnosis. In this way, the system avoids false alarms caused by known external interference, ensuring that only genuine motor concentricity anomalies are accurately identified and reported.

[0080] Specifically, in the aforementioned cloud-based energy-saving motor concentricity data acquisition method, the steps for determining environmental interference characteristics and motor operating conditions based on motor operating information can be further refined. Determining environmental interference characteristics and motor operating conditions based on the aforementioned motor operating information includes:

[0081] During the motor shutdown or stable operation phase, the vibration signal is analyzed by time and frequency to identify the frequency range and intensity of environmental noise and generate environmental interference characteristics;

[0082] Based on the comparison between the current signal and the preset threshold, the motor operating status is classified into no-load, light-load, or heavy-load to determine the motor operating condition.

[0083] Specifically, the determination of environmental interference characteristics is carried out during the motor shutdown or stable operation phase. During this phase, the motor's inherent vibration is relatively small or stable, and the vibration signals collected at this time mainly reflect the influence of environmental noise. By performing time-frequency analysis on the vibration signals, such as using short-time Fourier transform (STFT) or wavelet transform, the energy distribution and intensity of environmental noise in different frequency ranges can be identified, thereby generating detailed environmental interference characteristics. These environmental interference characteristics can include information such as the frequency components, amplitude, and duration of the noise. Their purpose is to provide a benchmark for subsequent motor operation information analysis, so as to effectively distinguish between the motor's own abnormal signals and external environmental interference in actual operation.

[0084] The determination of the motor's operating condition is achieved through the analysis of current signals. Current signals are a direct indicator of the motor's load state. By comparing the real-time acquired current signals with preset thresholds, the motor's operating state can be classified into different conditions, such as no-load, light-load, or heavy-load. For example, when the current signal is below the first preset threshold, the motor is considered to be in an no-load state; when the current signal is between the first and second preset thresholds, it is considered to be in a light-load state; and when the current signal is above the second preset threshold, it is considered to be in a heavy-load state. The purpose is that the motor's vibration characteristics and concentricity anomalies may differ under different operating conditions, and accurate identification of the operating condition helps to more precisely analyze the key signal characteristics related to motor concentricity.

[0085] This application's solution, through time-frequency analysis during motor shutdown or stable operation, effectively captures and quantifies the characteristics of environmental noise. This allows for effective differentiation between the motor's own vibration signals and environmental noise in subsequent motor operation information analysis, avoiding misjudgments. Simultaneously, by classifying motor operating conditions based on current signals, the judgment of motor concentricity anomalies can be combined with specific load conditions, improving the accuracy and reliability of diagnosis. This phased, multi-dimensional data analysis method provides a cleaner and more clearly defined data foundation for subsequent extraction of key signal features related to motor concentricity.

[0086] The above technical solutions enable accurate identification of environmental interference characteristics and motor operating conditions. Specifically, by performing time-frequency analysis on vibration signals at specific stages, an environmental noise model can be accurately established, effectively filtering out or compensating for the influence of environmental noise on vibration signals, thereby improving the sensitivity and accuracy of motor concentricity anomaly detection. Furthermore, by analyzing current signals to determine motor operating conditions, the judgment of motor concentricity anomalies is no longer a single-dimensional process, but rather a comprehensive assessment combining the actual load conditions of the motor. This avoids false alarms caused by changes in operating conditions and significantly improves the reliability and practicality of data acquisition and analysis.

[0087] In some embodiments described above in this application, the extraction of key signal features related to motor concentricity is proposed. Specifically, the extraction process may include the following steps: Extracting key signal features related to motor concentricity includes: performing real-time spectrum analysis on the vibration signal to extract the second harmonic vibration energy related to motor speed and the first vibration energy of frequency components related to component failure; determining whether the increase in the second harmonic vibration energy is accompanied by a synchronous increase in the first vibration energy or shows a specific correlation; when the second harmonic vibration energy increases, and the first vibration energy remains stable or shows an uncorrelated change, marking the second harmonic vibration energy as a key signal feature related to motor concentricity.

[0088] Specifically, real-time spectrum analysis of vibration signals refers to converting the acquired time-domain vibration signal into a frequency-domain signal using mathematical methods such as Fourier transform, thereby revealing the vibration intensity of different frequency components. Among these, the second harmonic vibration energy refers to the vibration energy occurring at twice the motor's rotational speed frequency, which is generally considered a typical indicator of abnormal motor concentricity (such as shaft misalignment). The first vibration energy can be understood as the vibration energy of specific frequency components related to faults in other internal components of the motor (such as bearings, gears, etc.), and its purpose is to distinguish vibrations caused by different fault sources.

[0089] Furthermore, after extracting the second harmonic vibration energy and the first vibration energy, it is necessary to determine the correlation between the changing trends of these two energies. Specifically, it is necessary to determine whether the increase in the second harmonic vibration energy is accompanied by a synchronous increase in the first vibration energy or shows a specific correlation, in order to rule out complex situations caused by failures in other components that may indirectly affect the second harmonic vibration energy. For example, if a bearing failure leads to an increase in the overall vibration level, it may be reflected in multiple frequencies, including the second harmonic. Therefore, through this correlation determination, the increase in second harmonic vibration energy caused purely by concentricity abnormalities can be identified more accurately.

[0090] Therefore, when the second harmonic vibration energy shows an upward trend, while the first vibration energy remains stable or exhibits unrelated changes, this second harmonic vibration energy can be clearly identified as a key signal feature related to motor concentricity. This identification process ensures that the extracted key signal features are highly relevant to motor concentricity anomalies, rather than other types of faults.

[0091] This application's solution effectively addresses the challenge of accurately identifying abnormal motor concentricity signals under complex operating conditions by performing refined spectral analysis of vibration signals and introducing multi-dimensional correlation judgment of vibration energy. By distinguishing between the second harmonic vibration energy related to motor concentricity and the first vibration energy of other frequency components related to component faults, and performing logical judgments based on their changing relationships, it avoids misjudging vibrations caused by other faults as concentricity anomalies. This method makes the extracted key signal features more specific and accurate, thus providing a reliable data foundation for subsequent anomaly judgment.

[0092] The above technical solution significantly improves the accuracy of motor concentricity anomaly detection and reduces false alarm rates. By precisely identifying and distinguishing vibration characteristics from different sources, the solution allows the data acquisition unit to more effectively focus on the true concentricity issue, avoiding confusion caused by other component failures or environmental interference. This not only improves data analysis efficiency but also provides strong support for timely and accurate maintenance of motor equipment, thereby extending equipment lifespan and reducing operating costs.

[0093] Specifically, the above-mentioned method of determining whether motor operation information contains potential abnormal information based on the changing trend of key signal features includes the following steps: calculating the moving average value of the key signal features in real time, and performing linear trend analysis on the moving average value; based on the linear trend analysis results, when the growth rate of the key signal features exceeds a preset threshold and continues for a certain period of time, determining that the motor operation information contains potential abnormal information.

[0094] Specifically, the real-time calculation of the moving average of the key signal features refers to the continuous moving average processing of key signal features (such as second harmonic vibration energy) extracted from the vibration signal and related to the motor concentricity. The moving average can be obtained by averaging multiple consecutive sampled values ​​of the key signal features within a preset time window. This time window can be configured according to the actual application scenario and motor operating characteristics, with the aim of smoothing the data, eliminating instantaneous noise and short-term fluctuations, thereby better revealing the potential trends of the key signal features.

[0095] Furthermore, performing linear trend analysis on the signal moving average refers to using linear regression or other trend analysis algorithms to evaluate the long-term trend of the smoothed signal moving average sequence. For example, a straight line can be fitted using the least squares method, and the slope of this line represents the growth rate of the key signal feature. In this way, the rate and direction of change of the key signal feature over time can be quantified, providing a quantitative basis for subsequent anomaly detection.

[0096] Specifically, based on the linear trend analysis results, if the growth rate of the key signal characteristic exceeds a preset threshold and persists for a certain period of time, it is determined that the motor operation information contains potential abnormal information. Specifically, the growth rate obtained from the linear trend analysis is compared with a preset threshold. If the growth rate exceeds the threshold, and this over-threshold state persists for a preset period of time, it is considered that there may be a potential abnormality in the motor concentricity. The preset threshold and duration can be calibrated based on historical data, motor type, operating environment, and requirements for anomaly sensitivity to balance the false alarm rate and the missed alarm rate.

[0097] This application's solution, by introducing a moving average and linear trend analysis, can more accurately and robustly determine whether motor operating information contains potential anomalies. The use of the moving average effectively filters out instantaneous noise and occasional fluctuations in key signal features, allowing subsequent trend analysis to focus on the true and continuous trend of motor concentricity changes. Linear trend analysis provides a means to quantify the rate of change of key signal features, ensuring that the judgment of concentricity anomalies no longer relies solely on instantaneous values ​​but is based on long-term or medium-term evolution trends. Furthermore, requiring the growth rate to persist for a certain period before triggering anomaly judgment further enhances the reliability of the judgment and avoids false alarms caused by short-term fluctuations.

[0098] Through the above technical solutions, this application can more accurately and robustly determine whether motor operating information contains potential abnormal information. Specifically, using the moving average of the signal can effectively suppress instantaneous noise and occasional interference, making the trend analysis of key signal features more stable and reliable. The introduction of linear trend analysis transforms the judgment of motor concentricity anomalies from a single instantaneous value comparison to a comprehensive evaluation of the rate of change and its persistence, thereby improving the sensitivity and accuracy of anomaly detection. In addition, requiring the growth rate to persist for a certain period of time before triggering anomaly judgment effectively avoids false alarms caused by short-term fluctuations, improves the reliability and practicality of the system, and helps to achieve early and accurate warnings of motor concentricity anomalies.

[0099] Furthermore, in a second aspect, this application also discloses a cloud-based energy-saving motor concentricity data acquisition system, applied to a data acquisition unit, such as... Figure 2 As shown, the system includes:

[0100] The acquisition module 201 is used to acquire motor operating information in real time; the motor operating information includes vibration signals, temperature signals and current signals;

[0101] The determination module 202 is used to determine environmental interference characteristics and motor operating conditions based on the motor operating information, and to extract key signal features related to motor concentricity, and determine the changing trend of the key signal features;

[0102] The judgment module 203 is used to determine whether the motor operation information contains potential abnormal information based on the changing trend of the key signal characteristics.

[0103] The upload module 204 is used to upload the motor operation information to the cloud in response to the motor operation information containing potential abnormal information.

[0104] The cloud-based energy-saving motor concentricity data acquisition system proposed in this application integrates an acquisition module, a determination module, a judgment module, and an upload module on the data acquisition unit side, achieving a combination of local intelligent processing of motor operation information and cloud-based collaborative analysis. The acquisition module is responsible for acquiring comprehensive motor operation data in real time; the determination module performs preliminary analysis on this data, identifying environmental interference, operating conditions, and extracting key signal features; the judgment module intelligently identifies potential anomalies based on the changing trends of key signal features; and the upload module selectively uploads relevant information to the cloud only when potential anomalies are detected. This system architecture effectively filters out a large amount of normal operating data and known interference, significantly reducing data transmission volume and cloud processing load, thereby achieving energy-saving goals. Simultaneously, through the collaborative work of each module, the accuracy of anomaly judgment is improved, avoiding false alarms and missed alarms, and providing more reliable data support for preventative motor maintenance.

[0105] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A cloud computing-based energy-saving motor concentricity data acquisition method, applied to a data acquisition unit, characterized in that, The method includes: Real-time acquisition of motor operating information; the motor operating information includes vibration signals, temperature signals, and current signals; Based on the motor operating information, determine the environmental interference characteristics, motor operating conditions, and extract key signal features related to motor concentricity, and determine the changing trend of the key signal features; Based on the changing trends of the key signal characteristics, determine whether the motor operation information contains potential abnormal information; In response to the motor operating information containing potential abnormal information, the motor operating information is uploaded to the cloud; The response that the motor operating information includes potential anomaly information includes uploading the motor operating information to the cloud, including: In response to the motor operating information containing potential abnormal information, the motor operating information, environmental interference characteristics, and motor operating conditions are uploaded to the cloud. Based on the environmental interference characteristics and motor operating conditions, the cloud performs noise separation on the motor operating information and analyzes the separated data to identify abnormal motor concentricity. The method further includes: Frequency domain analysis is performed on the vibration signal to extract the vibration energy related to the motor speed, monitor the changes in the vibration energy, and identify whether there are intermittent fluctuations in the vibration energy. In response to the presence of intermittent fluctuations that are not related to changes in motor operating conditions, these fluctuations are marked as potential external interferences, and a summary of the potential external interferences is sent to the cloud. If the cloud receives potential external interference of a specific frequency from the same data acquisition unit within a certain period of time, and the potential external interference does not cause the cloud to identify abnormal motor concentricity, then it sends an interference mode confirmation command to the data acquisition unit.

2. The cloud computing-based energy-saving motor concentricity data acquisition method according to claim 1, characterized in that, The step of performing frequency domain analysis based on the vibration signal to extract vibration energy related to the motor speed, monitoring changes in the vibration energy, and identifying whether there are intermittent fluctuations in the vibration energy includes: Fast Fourier Transform analysis was performed on the vibration signal to extract the second harmonic vibration energy; Obtain the moving average sequence of the second harmonic vibration energy within a time window, and determine it as the key signal feature; Analyze the moving average sequence to identify whether the vibration energy has intermittent fluctuations, and record the characteristics of the intermittent fluctuations; wherein, the intermittent fluctuation is that within a short period of time, the moving average of the second harmonic vibration energy suddenly rises above a preset relative threshold, and then quickly falls back to near the original level in the next short period of time, and there is a quiet period between the rise and fall.

3. The cloud computing-based energy-saving motor concentricity data acquisition method according to claim 2, characterized in that, The characteristics of the intermittent fluctuations include: occurrence time, duration, peak frequency, and peak amplitude.

4. The cloud computing-based energy-saving motor concentricity data acquisition method according to claim 2, characterized in that, After sending an interference mode confirmation command to the data acquisition unit, the method further includes: The intermittent fluctuations are identified as disturbance features and added to the environmental disturbance features to update the environmental disturbance features; The vibration energy is processed based on the updated environmental disturbance characteristics to suppress the disturbance, and a trend analysis is performed on the vibration energy after the disturbance is suppressed. When the trend analysis indicates that the vibration energy continues to rise beyond a preset threshold, it is marked as a potential concentricity anomaly, and the motor operation information, the updated environmental interference characteristics, the motor operating conditions, and the changing trends of the key signal characteristics are sent to the cloud.

5. The cloud computing-based energy-saving motor concentricity data acquisition method according to claim 1, characterized in that, Based on the motor operating information, the environmental interference characteristics and motor operating conditions are determined, including: During the motor shutdown or stable operation phase, the vibration signal is analyzed by time and frequency to identify the frequency range and intensity of environmental noise and generate environmental interference characteristics; Based on the comparison between the current signal and the preset threshold, the motor operating status is classified into no-load, light-load, or heavy-load to determine the motor operating condition.

6. The cloud computing-based energy-saving motor concentricity data acquisition method according to claim 1, characterized in that, The extraction of key signal features related to motor concentricity includes: Real-time spectrum analysis was performed on the vibration signal to extract the second harmonic vibration energy related to the motor speed and the first vibration energy of the frequency components related to component failure. Determine whether the increase in the second harmonic vibration energy is accompanied by a synchronous increase in the first vibration energy or shows a specific correlation; when the second harmonic vibration energy increases and the first vibration energy remains stable or shows an uncorrelated change, mark the second harmonic vibration energy as a key signal feature related to the motor concentricity.

7. The cloud computing-based energy-saving motor concentricity data acquisition method according to claim 6, characterized in that, Based on the changing trends of the key signal characteristics, determine whether the motor operation information contains potential abnormal information, including: Calculate the moving average of the key signal features in real time, and perform linear trend analysis on the moving average of the signal. Based on the linear trend analysis results, when the growth rate of the key signal feature exceeds a preset threshold and continues for a certain period of time, it is determined that the motor operation information contains potential abnormal information.

8. A cloud computing-based energy-saving motor concentricity data acquisition system, applied to a data acquisition unit, characterized in that, The system includes: The acquisition module is used to acquire motor operating information in real time; the motor operating information includes vibration signals, temperature signals, and current signals. The determination module is used to determine environmental interference characteristics and motor operating conditions based on the motor operating information, as well as to extract key signal features related to motor concentricity and determine the changing trend of the key signal features. The judgment module is used to determine whether the motor operation information contains potential abnormal information based on the changing trend of the key signal characteristics. The upload module is used to upload the motor operation information to the cloud in response to the motor operation information containing potential abnormal information; The response that the motor operating information includes potential anomaly information includes uploading the motor operating information to the cloud, including: In response to the motor operating information containing potential abnormal information, the motor operating information, environmental interference characteristics, and motor operating conditions are uploaded to the cloud. Based on the environmental interference characteristics and motor operating conditions, the cloud performs noise separation on the motor operating information and analyzes the separated data to identify abnormal motor concentricity. Also includes: Frequency domain analysis is performed on the vibration signal to extract the vibration energy related to the motor speed, monitor the changes in the vibration energy, and identify whether there are intermittent fluctuations in the vibration energy. In response to the presence of intermittent fluctuations that are not related to changes in motor operating conditions, these fluctuations are marked as potential external interferences, and a summary of the potential external interferences is sent to the cloud. If the cloud receives potential external interference of a specific frequency from the same data acquisition unit within a certain period of time, and the potential external interference does not cause the cloud to identify abnormal motor concentricity, then it sends an interference mode confirmation command to the data acquisition unit.

Images