A multi-source information fusion evaluation method and system for health status of a coal mining machine motor

Abstract

This invention relates to the field of coal mine equipment condition assessment technology, specifically disclosing a multi-source information fusion assessment method and system for the health status of a coal mining machine motor. By real-time acquisition of three-phase current, bearing temperature, and casing vibration signals of the coal mining machine motor, the system completes the working condition classification according to preset rules and standards, determines the basic damage factors of each indicator, identifies mechanical faults through vibration spectrum analysis, calculates additional damage factors to obtain the total vibration damage factor, records the duration of the working condition, calculates the time amplification factor using a nonlinear model, fuses the multi-source synergistic amplification factors to calculate the comprehensive damage rate, obtains the cumulative equivalent operating time through time integration, and calculates the remaining health level based on the motor's design life, thus achieving quantitative assessment and graded early warning of health status. This invention, through precise quantification of time accumulation and multi-source coupled damage, can identify subtle mechanical faults, providing support for predictive maintenance and effectively improving the operational reliability of the coal mining machine motor.

Description

Technical Field

[0001] This invention relates to the field of coal mine equipment condition assessment technology, and in particular to a multi-source information fusion assessment method and system for the health status of a coal mining machine motor. Background Technology

[0002] Coal mining machines are the core production equipment in fully mechanized coal mining faces. Their cutting motors and traction motors, as core power output components, operate continuously underground under harsh conditions of heavy load, variable load, strong mechanical vibration, and high dust and humidity. Problems such as abnormal current fluctuations, excessive temperature, and excessive vibration intensity will continuously cause aging of motor insulation and fatigue damage to mechanical structure. The health status of the motor is directly related to the reliability of coal mining machine operation and the safety and efficiency of coal mine production.

[0003] Currently, the health status monitoring and assessment of coal mining machine motors still relies on traditional technologies, which have many unresolved technical shortcomings. Among these shortcomings, existing assessment methods rely solely on single signal thresholds such as current, temperature, or vibration for alarms, failing to integrate the coupled characteristics of multi-source information and comprehensively reflect the overall health status of the motor. They also do not consider the nonlinear cumulative effect of operating duration on motor damage. Under the same abnormal operating conditions, the degree of damage differs greatly between short-term and long-term continuous operation, and traditional methods cannot accurately quantify this difference. Health status can only achieve a binary judgment of normal or abnormal, lacking continuously quantifiable health indicators. This makes it impossible to provide data support for predictive maintenance of equipment and to predict the remaining service life of the motor based on its health status. Instead, it can only passively respond to faults, easily leading to unplanned shutdowns and seriously affecting continuous coal mine production.

[0004] Therefore, there is an urgent need for a multi-source information fusion assessment method and system for the health status of coal mining machine motors to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-source information fusion assessment method for the health status of a coal mining machine motor, comprising the following steps: Real-time acquisition of current, temperature, and vibration signals from the coal mining machine motor during operation; According to the preset working condition classification rules, the collected current signals are classified into load levels, the temperature signals are classified into temperature rise levels, and the vibration signals are classified into vibration intensity levels according to international standards. Based on the classification of current, temperature and vibration conditions, the corresponding basic damage factors are determined, and the vibration signal is subjected to spectrum analysis. The mechanical fault type is identified according to the extracted characteristic frequency and the corresponding additional damage factor is calculated, thereby determining the total vibration damage factor. Record the duration of the current working condition, and calculate the duration amplification factor of current, temperature and vibration respectively using a preset nonlinear function model based on the duration and working condition level. Based on the basic damage factor, additional damage factor, duration amplification factor, and synergistic amplification factor among multi-source information, the comprehensive damage rate at the current moment is calculated. The cumulative equivalent operating time is obtained by integrating the comprehensive damage rate over time. Finally, the remaining health is calculated based on the cumulative equivalent operating time and the motor's design life to complete the quantitative assessment of the motor's health status.

[0006] Furthermore, this invention also discloses a multi-source information fusion assessment system for the health status of a coal mining machine motor, including: The data acquisition module is used to acquire the current, temperature, and vibration signals of the coal mining machine motor in real time during operation. The grading module is used to classify the load level of the collected current signal, the temperature rise level of the temperature signal, and the vibration intensity level of the vibration signal according to the preset working condition grading rules. The first calculation module is used to determine the corresponding basic damage factor based on the divided current, temperature and vibration condition levels, and to perform spectrum analysis on the vibration signal. Based on the extracted characteristic frequencies, the mechanical fault type is identified and the corresponding additional damage factor is calculated, thereby determining the total vibration damage factor. The second calculation module is used to record the duration of the current working condition and calculate the duration amplification factor of current, temperature and vibration respectively using a preset nonlinear function model based on the duration and working condition level. The quantitative assessment module is used to calculate the comprehensive damage rate at the current moment based on the basic damage factor, the additional damage factor, the time amplification coefficient, and the synergistic amplification coefficient among the multi-source information. The cumulative equivalent running time is obtained by integrating the comprehensive damage rate over time. Finally, the remaining health is calculated based on the cumulative equivalent running time and the motor design life to complete the quantitative assessment of the motor's health status.

[0007] Furthermore, the first computing module includes: The setting unit is used to set the corresponding current-based damage factor, temperature-based damage factor and vibration-based damage factor for each level based on the classification results of current, temperature and vibration working conditions. The first calculation unit is used to perform spectrum transformation processing on the collected vibration signal to obtain a spectrum diagram, calculate the rotor rotation frequency based on the current speed of the motor, and read the bearing geometric parameters from the pre-stored bearing parameter database, and calculate the bearing fault characteristic frequency in combination with the rotor rotation frequency. The adjustment unit is used to search for peak values ​​at each characteristic frequency in the spectrum of the vibration signal. If the peak value exceeds the preset threshold, it is determined that there is a corresponding fault type, and an additional damage factor is assigned according to the fault type. When multiple faults exist at the same time, the additional damage factors are summed. The second calculation unit is used to combine the basic vibration damage factor and the additional damage factor to calculate the total vibration damage factor.

[0008] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described multi-source information fusion evaluation method for the health status of a coal mining machine motor.

[0009] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for multi-source information fusion evaluation of the health status of a coal mining machine motor.

[0010] The beneficial effects of this application are as follows: Firstly, this invention addresses the shortcomings of existing coal mining machine motor health assessment technologies by achieving accurate assessment through multi-source information fusion and time accumulation effect quantification. It simultaneously collects and analyzes multi-source signals such as current, temperature, and vibration, breaking through the limitations of single signal monitoring and comprehensively covering the all-dimensional operating status of the motor in terms of electrical, thermal, and mechanical aspects, thus significantly improving the comprehensiveness and accuracy of health status assessment.

[0011] Secondly, this invention uses logarithmic, exponential and other nonlinear function models to calculate the duration amplification factor, accurately quantifying the nonlinear cumulative effect of duration on motor damage under different working conditions, and solving the problem that traditional methods cannot distinguish between short-term abnormal and long-term continuous abnormal damage.

[0012] Third, this invention obtains the cumulative equivalent operating time by integrating the comprehensive damage rate, and calculates the remaining health status by combining the motor design life, thereby realizing a continuous quantitative assessment of the health status, replacing the traditional binary judgment, and directly supporting the predictive maintenance decision of the coal mining machine motor.

[0013] Fourth, this invention achieves four-level health status classification and alarm based on remaining health, and can predict the remaining service life of the motor, enabling advance maintenance planning, avoiding unplanned downtime, and significantly improving the reliability of coal mining machine motor operation and coal mine production efficiency; the full-process logic of multi-source signal acquisition, working condition classification, and damage quantification is adapted to harsh underground working conditions, the data preprocessing and parameter calculation rules are standardized, and the method has strong universality and field adaptability. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of a method flow proposed in an embodiment of this application.

[0015] Figure 2 This is a schematic diagram of the system structure proposed in an embodiment of the present invention.

[0016] Figure 3This is a flowchart illustrating a specific implementation of an embodiment of the present invention.

[0017] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] like Figure 1 As shown, this application provides a multi-source information fusion assessment method for the health status of a coal mining machine motor, including the following steps: S1, real-time acquisition of three-phase current signals, bearing temperature signals and casing vibration signals of the coal mining machine motor during operation; S2, according to the preset working condition classification rules, classifies the collected current signal into load levels, classifies the temperature signal into temperature rise levels, and classifies the vibration signal into vibration intensity levels according to the ISO10816-3 standard. S3. Based on the divided current, temperature and vibration condition levels, the corresponding basic damage factors are determined respectively, and the vibration signal is subjected to spectrum analysis. The mechanical fault type is identified according to the extracted characteristic frequency and the corresponding additional damage factor is calculated, thereby determining the total vibration damage factor. S4 records the duration of the current working condition and calculates the duration amplification factor of current, temperature and vibration respectively using a preset nonlinear function model based on the duration and working condition level. S5. Based on the basic damage factor, additional damage factor, duration amplification factor, and synergistic amplification factor among multi-source information, calculate the comprehensive damage rate at the current moment. Obtain the cumulative equivalent operating time by integrating the comprehensive damage rate over time. Finally, calculate the remaining health status based on the cumulative equivalent operating time and the motor design life to complete the quantitative assessment of the motor's health status.

[0020] As described in steps S1-S5 above, this invention completes the working condition classification, damage factor calculation, duration amplification coefficient solution, multi-source information fusion and comprehensive damage rate integration by real-time acquisition of multi-source physical signals during the operation of the coal mining machine motor, and finally calculates the remaining health status by combining the motor design life, thereby realizing the multi-source information fusion and quantitative assessment of the health status of the coal mining machine motor.

[0021] Coal mining machine motors operate continuously under complex underground conditions. Fluctuations in current load, abnormal temperature increases, and excessive vibration intensity can all cause motor damage, and the degree of damage accumulates over time. A single monitoring indicator cannot fully characterize the motor's health status. Therefore, it is necessary to construct an assessment system that integrates multi-source information and considers the cumulative effect over time. However, existing coal mining machine motor health assessment technologies rely solely on a single signal threshold for alarms, failing to integrate the coupling effect of multi-source information or quantify the cumulative damage effect caused by the duration of the operating conditions. Consequently, they cannot achieve accurate quantification of health status. This invention addresses the various shortcomings of existing technologies through a complete technical process including multi-source signal acquisition, operating condition classification, damage factor calculation, duration amplification coefficient calculation, comprehensive damage rate fusion integration, and remaining health calculation.

[0022] By real-time acquisition of three-phase current signals, bearing temperature signals, and casing vibration signals from the coal mining machine motor during operation, signal acquisition is achieved through Hall current sensors installed in the motor's three-phase power supply lines, PT100 platinum resistance temperature sensors embedded in the bearing housing and stator windings, and piezoelectric accelerometers fixed in the bearing seats at the motor's drive and non-drive ends. This simultaneously acquires three core operational data categories: electrical, thermal, and mechanical, comprehensively covering key causes of motor damage. Furthermore, based on preset operating condition classification rules, the acquired current signals are classified into load levels, the temperature signals into temperature rise levels, and the vibration signals into vibration intensity levels according to the ISO10816-3 standard. This transforms continuous physical signals into standardized operating condition levels, with current classified according to load and current imbalance characteristics, temperature classified according to insulation tolerance thresholds, and vibration classified according to internationally accepted standards. This achieves unified quantification processing of multi-source signals, eliminating evaluation biases caused by different physical dimensions.

[0023] Based on the defined current, temperature, and vibration operating condition levels, corresponding basic damage factors are determined, and the vibration signal is subjected to spectral analysis. Mechanical fault types are identified based on the extracted characteristic frequencies, and corresponding additional damage factors are calculated. This leads to the determination of the total vibration damage factor. Fixed basic damage factors are matched to each operating condition level to characterize the degree of basic damage. Then, the vibration signal is processed using Fast Fourier Transform to extract the rotor rotation frequency and bearing fault characteristic frequencies, identify mechanical faults, and calculate additional damage factors. Finally, the total vibration damage factor is obtained through fusion, accurately characterizing the comprehensive damage caused by vibration. Next, the duration of the current operating condition is recorded, and based on the duration and operating condition level, a preset nonlinear function model is used to calculate the duration amplification coefficients of current, temperature, and vibration. The operating condition state machine monitors and records the duration of each indicator operating condition level in real time. For current, a logarithmic function model is used; for temperature, an exponential function model; and for vibration, a combination of logarithmic and linear models are used to calculate the duration amplification coefficients. This quantifies the cumulative amplification effect of time on damage under different operating conditions, addressing the problem that traditional methods do not consider the influence of operating condition duration.

[0024] Finally, based on the basic damage factor, additional damage factor, time amplification factor, and synergistic amplification factor among multi-source information, the comprehensive damage rate at the current moment is calculated. The cumulative equivalent operating time is obtained by integrating the comprehensive damage rate over time. Finally, the remaining health is calculated based on the cumulative equivalent operating time and the motor design life to complete the quantitative assessment of the motor's health status. By first fusing the single information source damage index and the multi-source synergistic amplification factor to obtain the comprehensive damage rate, and then converting the comprehensive damage rate into the cumulative equivalent operating time under the benchmark working condition by time integration, the remaining health is calculated in combination with the motor design life, realizing the continuous quantitative and graded assessment of the motor's health status, and providing an accurate basis for the predictive maintenance of coal mining machine motors.

[0025] The overall logic of this invention is as follows: First, by real-time acquisition of three core signals—three-phase current, bearing temperature, and casing vibration—during the operation of the coal mining machine motor, raw operating data in the electrical, thermal, and mechanical dimensions of the motor are obtained. Then, based on preset grading rules and the ISO10816-3 international standard, the three types of continuous physical signals are converted into standardized load, temperature rise, and vibration intensity operating condition levels. Next, based on the operating condition levels, the basic damage factors of current and temperature are determined. Then, mechanical faults are identified and additional damage factors are calculated through vibration signal spectrum analysis to obtain the total vibration damage factor, achieving preliminary quantification of damage in each dimension. Subsequently, the duration of each operating condition is recorded, and the corresponding duration amplification coefficient is calculated using a preset nonlinear function model to quantify the nonlinear cumulative effect of operating condition duration on motor damage. Finally, the basic damage factor, additional damage factor, duration amplification coefficient, and multi-source information synergistic amplification coefficient are integrated to calculate the comprehensive damage rate. By integrating the comprehensive damage rate over time, the cumulative equivalent operating time is obtained. Then, combined with the motor's design life, the remaining health is calculated, thereby completing the quantitative assessment of the motor's health status. Its core principle relies on the physical damage mechanism of heat accumulation caused by overload of coal mining machine motor current, accelerated insulation aging due to excessive temperature, and mechanical fatigue caused by abnormal vibration. It integrates the coupling amplification effect of multi-source signal anomalies and the time accumulation damage effect under different working conditions, and uniformly maps the actual running time under complex variable working conditions to the equivalent damage time under the benchmark working condition. The remaining health degree is used as a quantitative indicator to intuitively represent the overall health status of the motor, which fundamentally solves the problems of traditional evaluation methods that rely on only a single indicator, do not consider time accumulation and multi-source coupling damage, and cannot accurately quantify the health status.

[0026] In one embodiment, step S1, which involves real-time acquisition of the three-phase current signal, bearing temperature signal, and casing vibration signal of the coal mining machine motor during operation, specifically includes: S11. The instantaneous values ​​of the U, V, and W phase currents of the motor are synchronously collected at a sampling frequency of not less than 5kHz by three high-precision Hall current sensors installed on the three-phase power supply line of the motor, and the effective values ​​of the three-phase currents are calculated in real time. At the same time, the current imbalance is calculated. The current imbalance is obtained by determining the maximum value, minimum value, and average value of the effective values ​​of the three-phase currents, and then calculating the percentage of the difference between the maximum value and the minimum value to the average value. S12 uses three PT100 platinum resistance temperature sensors embedded in the front and rear bearing chambers and stator windings of the motor to collect temperature signals from two bearing measuring points and one winding measuring point at a sampling frequency of 1Hz, calculates the average temperature of the three measuring points, and performs moving average filtering on the collected temperature signals to eliminate instantaneous interference. S13 uses two piezoelectric accelerometers installed on the motor drive end bearing housing and the non-drive end bearing housing to collect the acceleration signal of the housing vibration at a sampling frequency of not less than 12.8kHz. The acceleration signal is converted into a velocity signal through a hardware integration circuit or digital integration algorithm, and then the effective value of the vibration velocity is calculated. At the same time, the original waveform data of the vibration signal is recorded for subsequent spectrum analysis. In addition, a vibration signal phase synchronization deviation index is introduced. This index is obtained by calculating the normalized value of the difference between the absolute value of the phase difference between the fundamental frequency components of the vibration signal at the two bearing housing measuring points and the standard value, and is used to evaluate the consistency of the vibration transmission path.

[0027] As described in steps S11-S13 above, by conducting multi-sensor, high-synchronization, and high-precision acquisition of the three-phase current signal of the coal mining machine motor, the bearing and winding temperature signal, and the dual-measuring-point casing vibration signal, and simultaneously completing numerical calculation, filtering, signal conversion, and feature index extraction, the accurate acquisition of multi-source data of motor operation is achieved, providing a stable and reliable data foundation for subsequent working condition classification and health status assessment.

[0028] Coal mining machine motors operate in a complex environment of strong interference, variable load, and continuous operation underground for extended periods. Electrical, thermal, and mechanical operating signals directly reflect the motor's health status. Single-point or low-precision acquisition introduces noise interference and data errors, failing to accurately reproduce the motor's actual operating state. Therefore, standardized, multi-dimensional, and interference-resistant acquisition and preprocessing of these three core signals are necessary. Traditional motor signal acquisition technologies often employ single-point, low-sampling-rate equipment, lacking preprocessing logic designed for harsh underground conditions, resulting in significant signal distortion and limited data dimensions. This invention utilizes dedicated sensors paired with a preset sampling frequency to complete signal acquisition, simultaneously performing numerical calculations, filtering, and feature extraction, specifically addressing the data errors and insufficient dimensionality issues of traditional acquisition methods.

[0029] Specifically, multi-source signal acquisition and preprocessing are accomplished through three types of sensors. First, the acquisition and calculation of three-phase current signals are performed. Three high-precision Hall current sensors, installed on the three-phase power supply lines of the motor, synchronously acquire the instantaneous values ​​of the U, V, and W phase currents of the motor at a sampling frequency of no less than 5kHz. This sampling frequency can completely capture the transient changes in current and avoid the loss of high-frequency electrical fault signals. After the acquisition is completed, the effective values ​​of the three-phase currents are calculated in real time. Then, the current imbalance is obtained by calculating the percentage of the difference between the maximum and minimum values ​​and the average value among the maximum, minimum, and average values ​​of the three-phase current effective values. The current imbalance can directly reflect electrical asymmetry faults such as phase loss and inter-turn short circuits in the motor, making up for the deficiency that the effective values ​​of the current alone cannot identify such faults. Next, temperature signal acquisition and filtering are performed. Three PT100 platinum resistance temperature sensors embedded in the front and rear bearing chambers and stator windings of the motor are used to acquire temperature signals from two bearing measuring points and one winding measuring point at a sampling frequency of 1Hz. This sampling frequency is adapted to the physical characteristics of slow temperature change and avoids generating redundant data. After acquisition, the average temperature of the three measuring points is calculated. At the same time, the temperature signal is subjected to moving average filtering to eliminate instantaneous temperature fluctuations caused by downhole electromagnetic and mechanical interference, ensuring the stability and accuracy of the temperature data.

[0030] Finally, the vibration signal is acquired, converted, and its characteristics are calculated. Two piezoelectric accelerometers are installed on the motor drive end bearing housing and the non-drive end bearing housing to acquire the acceleration signal of the housing vibration at a sampling frequency of not less than 12.8kHz. This sampling frequency meets the Nyquist criterion for vibration spectrum analysis and can completely retain the fault characteristic frequency information. After acquisition, the acceleration signal is converted into a velocity signal through a hardware integration circuit or digital integration algorithm, and then the effective value of vibration velocity is calculated. At the same time, the original waveform data of the vibration signal is recorded for subsequent spectrum analysis. In addition, the normalized value of the difference between the absolute value of the phase difference between the fundamental frequency components of the vibration signal at the two bearing housing measuring points and the standard value is calculated to obtain the phase synchronization deviation of the vibration signal. This index can accurately assess the consistency of the vibration transmission path and identify mechanical anomalies such as bearing housing loosening and base deformation in advance.

[0031] In one embodiment, step S2 involves classifying the acquired current signal into load levels according to preset operating condition classification rules, classifying the temperature signal into temperature rise levels, and classifying the vibration signal into vibration intensity levels according to the ISO10816-3 standard. Specifically, this includes: S21. Classification of current load levels: Based on the ratio of the average value of the three-phase current effective value to the rated current of the motor, combined with the current imbalance degree for comprehensive determination, the current operating conditions are divided into six levels: underload, light load, rated, overload, heavy load, and overloading. The specific classification rules are as follows: When the average current is less than 0.8 times the rated current and the current imbalance degree is less than 5%, it is underload; when the average current is greater than or equal to 0.8 times the rated current and less than 1.0 times the rated current, and at the same time the current imbalance degree is less than 5%, it is light load; when the average current is greater than or equal to 1.0 times the rated current and less than 1.1 times the rated current, and at the same time the current imbalance degree is less than 5%, it is rated; when the average current is greater than or equal to 1.1 times the rated current and less than 1.3 times the rated current, or the current imbalance degree is greater than or equal to 5%, it is overload; when the average current is greater than or equal to 1.3 times the rated current and less than 1.5 times the rated current, and at the same time the current imbalance degree is less than 8%, it is heavy load; when the average current is greater than or equal to 1.5 times the rated current, or the current imbalance degree is greater than or equal to 8%, it is overloading. Among them, when the current imbalance degree exceeds the threshold value, even if the average current does not reach the overload standard, it will be classified into a higher load level to reflect the electrical fault risks such as open phase or inter-turn short circuit; S22. Classification of temperature levels: By comparing the highest temperature among the temperatures of two bearing measurement points and one winding measurement point with the temperature threshold corresponding to the motor insulation class, the temperature operating conditions are divided into four levels: normal, concerned, overheated, and over-temperature. The specific classification rules are as follows: When the highest temperature is lower than 120°C, it is normal; when the highest temperature is greater than or equal to 120°C and lower than 135°C, it is concerned; when the highest temperature is greater than or equal to 135°C and lower than 155°C, it is overheated; when the highest temperature is greater than or equal to 155°C, it is over-temperature. At the same time, the temperature change rate index is introduced. When the temperature change rate exceeds 2°C per minute, even if the highest temperature does not reach the overheat threshold, the early warning mechanism is activated and the operating condition level is adjusted in advance; S23. Classification of vibration severity levels: According to the vibration rating criteria for large motors with a power greater than 300 kW and a rotational speed in the range of 600 to 12,000 revolutions per minute under rigid mounting in ISO10816-3 standard, the vibration operating conditions are divided into five levels: excellent, good, qualified, alarm, and dangerous based on the vibration velocity effective value. The specific classification rules are as follows: When the vibration velocity effective value is less than 1.12 mm / s, it is excellent; when it is greater than or equal to 1.12 mm / s and less than 2.8 mm / s, it is good; When it is greater than or equal to 2.8 mm / s and less than 4.5 mm / s, it is qualified; When it is greater than or equal to 4.5 mm / s and less than 7.1 mm / s, it is alarm; When it is greater than or equal to 7.1 mm / s, it is dangerous.

[0032] As described in steps S21-S23 above, the load level, temperature rise level, and vibration intensity level of the current signal, temperature signal, and vibration signal of the coal mining machine motor are standardized and classified respectively. Combined with the current imbalance and temperature change rate, auxiliary judgment is completed. According to the vibration classification logic of ISO10816-3 international standard, the accurate conversion of multi-source continuous physical signals to a unified working condition level is realized, providing a standardized and unified judgment basis for the subsequent determination of basic damage factors.

[0033] When a coal mining machine motor operates continuously underground, fluctuations in current load can cause electrical damage, abnormal temperature increases can accelerate the aging of insulation materials, and excessive vibration intensity can lead to mechanical fatigue. The continuous values ​​of these three types of signals cannot directly correlate with the degree of equipment damage, and the rate of damage accumulation varies significantly under different operating conditions. Therefore, it is necessary to transform multi-dimensional physical signals into clearly categorized operating conditions to achieve accurate quantification of the degree of damage. Traditional motor operating condition grading techniques only use a single signal threshold for judgment, without combining auxiliary characteristic indicators for comprehensive assessment. Vibration grading also does not follow internationally accepted standards, resulting in low matching between grading results and the actual damage state of the equipment, failing to provide reliable support for health assessment. This invention addresses the problems of poor standardization and insufficient accuracy of traditional grading methods through a comprehensive solution that integrates multi-indicator judgment, provides early warning based on auxiliary characteristics, and categorizes according to international standards.

[0034] Specifically, the current load level classification is implemented, with the ratio of the average value of the three-phase current to the rated current of the motor as the core basis. At the same time, the current imbalance is combined to carry out a comprehensive judgment, and the current condition is divided into six levels. During the classification process, the current imbalance is used as a key auxiliary indicator. When the current imbalance exceeds the set threshold, even if the average current does not reach the overload standard, the condition will be classified into a higher level. This accurately reflects the potential risks brought about by electrical faults such as motor phase loss or inter-turn short circuit, and ensures that the current condition classification can cover all scenarios of electrical abnormalities.

[0035] Next, temperature level classification is performed. The highest temperature among two bearing measuring points and one winding measuring point is selected as the judgment basis and compared with the temperature threshold corresponding to the motor insulation level. The temperature conditions are divided into four levels. At the same time, the temperature change rate index is introduced. When the temperature change rate exceeds 2°C per minute, the early warning mechanism can be activated and the operating condition level can be adjusted without waiting for the highest temperature to reach the over-temperature threshold. This can identify the risk of thermal aging caused by rapid temperature rise in advance and adapt to the rapid accumulation characteristics of thermal damage to motor insulation materials.

[0036] Finally, the vibration intensity level classification was carried out, strictly in accordance with the vibration rating criteria for rigidly installed large motors with power greater than 300kW and speed in the range of 600 to 12000 rpm in the ISO10816-3 standard. The effective value of vibration velocity was used as the judgment index, and the vibration condition was divided into five levels. Adopting internationally accepted standards can improve the universality and authority of vibration classification, ensure that the correspondence between vibration condition and mechanical damage degree conforms to industry-standard specifications, and provide a reliable benchmark for subsequent vibration damage factor calculation.

[0037] In one embodiment, step S3, based on the divided current, temperature, and vibration condition levels, determines the corresponding basic damage factors, performs spectral analysis on the vibration signal, identifies the mechanical fault type based on the extracted characteristic frequencies, calculates the corresponding additional damage factors, and then determines the total vibration damage factor, specifically including: S31, based on the current load level classification results, the basic current damage factors corresponding to the six levels of underload, light load, rated, overload, heavy load and overload are set to 0.7, 0.9, 1.0, 3.5, 8.0 and 18.0 respectively. Based on the temperature level classification results, the basic temperature damage factors corresponding to the four levels of normal, attention, over-temperature and over-temperature are set to 1.0, 2.0, 5.0 and 15.0 respectively. Based on the vibration level classification results, the basic vibration damage factors corresponding to the five levels of excellent, good, qualified, alarm and danger are set to 0.8, 1.0, 1.5, 4.0 and 10.0 respectively. S32, perform Fast Fourier Transform (FFT) processing on the acquired vibration acceleration signal to obtain the vibration signal spectrum. Calculate the rotor rotational frequency based on the current motor speed. The rotor rotational frequency is obtained by dividing the motor speed value by 60. Read geometric parameters such as the number of rolling elements, rolling element diameter, bearing pitch diameter, and contact angle from a pre-stored motor bearing parameter database. Based on these parameters and the rotor rotational frequency, calculate the bearing fault characteristic frequencies, specifically including the inner ring fault frequency, outer ring fault frequency, rolling element fault frequency, and cage fault frequency. The formula for calculating the inner ring fault frequency is: ; Among them, the This indicates the failure frequency of the inner race, where n represents the number of rolling elements. Indicates the rotor rotation frequency. Indicates the diameter of the rolling element. This indicates the bearing pitch diameter (i.e., the diameter of the raceway center circle). This indicates the contact angle (usually 0°).

[0038] Among them, the outer ring failure frequency The calculation formula is: ; Among them, rolling element failure frequency The calculation formula is: ; Among them, cage failure frequency The calculation formula is: ; S33. Search the vibration signal spectrum for the rotor rotation frequency, twice the rotor rotation frequency, half the rotor rotation frequency, and the peak amplitude of each bearing fault characteristic frequency and its harmonics. If the peak amplitude at a certain characteristic frequency exceeds three times the average amplitude of its neighborhood baseline, it is determined that there is a corresponding fault type. The corresponding additional damage factor is assigned according to the fault type. The rotor imbalance fault is assigned 0.2, the misalignment fault is assigned 0.3, the looseness fault is assigned 0.4, the inner ring fault is assigned 0.6, the outer ring fault is assigned 0.5, the rolling element fault is assigned 0.7, and the cage fault is assigned 0.3. When multiple faults exist at the same time, the additional damage factors corresponding to each fault are summed, but the upper limit of the summation is set to 1.5. S34, Determine the total vibration damage factor: Combine the basic vibration damage factor and the additional damage factor to calculate the total vibration damage factor.

[0039] As described in steps S31-S34 above, by setting corresponding basic damage factors for each operating condition level of current, temperature, and vibration, the vibration signal is subjected to spectrum analysis and the fault characteristic frequency is calculated. Based on the peak value of the spectrum, the mechanical fault type is identified and the additional damage factor is calculated. Finally, the total vibration damage factor is obtained by fusion, realizing the quantitative characterization of the three types of damage of motor electrical, thermal, and mechanical, and providing accurate damage parameters for subsequent comprehensive damage rate calculation.

[0040] When a coal mining machine motor is running underground, fluctuations in electrical load, abnormal temperature increases, and excessive mechanical vibration can all cause continuous damage. Relying solely on operating condition levels cannot directly reflect the specific degree of damage, and vibration intensity grading cannot identify subtle mechanical damage such as rotor imbalance and bearing failure. This damage accumulates and eventually leads to equipment failure. Therefore, it is necessary to convert operating condition levels into quantified damage factors and supplement this with additional damage from faults through spectral analysis to fully reflect the true damage state of the motor. Traditional motor damage assessment only uses a single signal threshold to determine anomalies, without setting quantified basic damage factors for different operating conditions, nor identifying subtle mechanical faults and calculating additional damage through vibration spectral analysis. The damage assessment results are one-sided and lack accuracy. This invention provides a complete solution by assigning basic damage factors in a graded manner, extracting fault features through spectral analysis, determining faults based on peak values ​​and calculating additional damage, and fusing these to obtain a total damage factor. This solution specifically addresses the problems of incomplete damage quantification and the inability to identify subtle faults in traditional methods.

[0041] Specifically, the process involves determining the basic damage factors. Based on the grading results of current, temperature, and vibration, a fixed value of basic damage factor is assigned to each grading level. The six load levels of current correspond to basic damage factors of 0.7, 0.9, 1.0, 3.5, 8.0, and 18.0, respectively; the four temperature rise levels correspond to basic damage factors of 1.0, 2.0, 5.0, and 15.0, respectively; and the five intensity levels of vibration correspond to basic damage factors of 0.8, 1.0, 1.5, 4.0, and 10.0, respectively. By assigning grading values, the abstract grading levels are transformed into calculable quantitative parameters that directly characterize the basic damage degree of each indicator.

[0042] Next, vibration signal spectrum analysis and fault characteristic frequency calculation are performed. The collected vibration acceleration signal is processed by fast Fourier transform to obtain the spectrum. The rotor rotation frequency is calculated by dividing the current motor speed by 60. The number of rolling elements, rolling element diameter, bearing pitch diameter and contact angle are read from the pre-established motor bearing parameter database. Then, combined with the rotor rotation frequency, the inner ring fault frequency, outer ring fault frequency, rolling element fault frequency and cage fault frequency are calculated in sequence to provide standard frequency basis for accurate identification of mechanical faults.

[0043] Then, fault type identification and additional damage factor calculation are performed. This step retrieves the peak amplitude at the rotor rotation frequency, twice the rotor rotation frequency, half the rotor rotation frequency, and the characteristic frequency of each bearing fault in the vibration spectrum. When the peak amplitude at a certain characteristic frequency exceeds three times the average amplitude of the neighborhood baseline, a corresponding mechanical fault is determined to exist. Simultaneously, a fixed additional damage factor is assigned according to the fault type. When multiple faults exist simultaneously, the additional damage factors are summed, with an upper limit of 1.5 set to supplement minor mechanical damage not covered by the vibration intensity classification, thus improving the completeness of vibration damage assessment. Finally, the total vibration damage factor is determined. The formula for calculating the total vibration damage factor is: ; Among them, the Indicates the total vibration damage factor. Indicates the vibration-based damage factor. This represents the additional damage factor, which is used to fully characterize all mechanical damage caused by vibration to the motor by calculating the total vibration damage factor.

[0044] In one embodiment, step S4 records the duration of the current operating condition and calculates the duration amplification factors of current, temperature, and vibration using a preset nonlinear function model based on the duration and operating condition level. Specifically, this includes: S41, establish a working condition state machine to monitor the changes in the working condition level of each indicator such as current, temperature and vibration in real time. When the working condition level of any indicator changes, record the time when the change occurs and calculate the duration of the indicator under the current working condition. The duration is obtained by subtracting the time of the last change in the working condition level of the indicator from the current time. For the three different indicators of current, temperature and vibration, the timing of their duration is independent of each other. S42. Based on the current load level, the current duration amplification factor is calculated using a logarithmic function model to reflect the nonlinear change of heat accumulation effect over time. The specific calculation rules are as follows: For underload levels, the current duration amplification factor is obtained by subtracting the natural logarithm of the sum of 0.03 multiplied by 1 and the duration divided by 1800. For light load and rated levels, the duration amplification factor is constant at 1.0. For overload levels, the duration amplification factor is obtained by adding 0.15 multiplied by 1 and the natural logarithm of the sum of the duration divided by 600. For heavy load levels, the duration amplification factor is obtained by adding 0.4 multiplied by 1 and the natural logarithm of the sum of the duration divided by 120. For overload levels, the duration amplification factor is obtained by adding 0.8 multiplied by 1 and the natural logarithm of the sum of the duration divided by 20. The unit of duration is seconds. S43. Based on the temperature level, the temperature duration amplification factor is calculated using an exponential function model to simulate the accelerated thermal aging effect of insulation materials based on the Arrhenius equation. The specific calculation rule is as follows: for the normal level, the temperature duration amplification factor is constant at 1.0. For the level of concern, the temperature duration amplification factor is obtained by adding 0.05 multiplied by 1 and the natural logarithm of the sum of duration divided by 600; For overtemperature levels, the temperature duration amplification factor is calculated using an exponential function with the natural constant e as the base and 0.0003 multiplied by the duration as the exponent. For overheating levels, the temperature duration amplification factor is calculated using an exponential function with the natural constant e as the base and 0.0008 multiplied by the duration as the exponent, where the duration is in seconds. S44. Based on the vibration level and combined with the mechanical fatigue accumulation characteristics, the vibration duration amplification factor is calculated using a model combining logarithmic and linear functions. The specific calculation rule is as follows: For the excellent level, the vibration duration amplification factor is obtained by adding 0.98 to the natural logarithm of the sum of 0.02 multiplied by one and the duration divided by 43200. For the good grade, the vibration duration amplification factor is constant at 1.0; for the acceptable grade, the vibration duration amplification factor is obtained by adding 0.08 multiplied by 1 and the natural logarithm of the sum of duration and 1800. For alarm levels, the vibration duration amplification factor is obtained by adding 0.2 multiplied by the natural logarithm of the sum of the duration and 300, plus 0.0001 multiplied by the duration. For the hazard level, the vibration duration amplification factor is obtained by adding 0.5 multiplied by the natural logarithm of the sum of 1 and the duration divided by 30, and then adding 0.0005 multiplied by the duration, where the duration is in seconds.

[0045] As described in steps S41-S44 above, by establishing a working condition state machine to independently record the working condition duration of each index such as current, temperature, and vibration, and for the physical damage characteristics of different working condition levels of the three types of indices, a matched nonlinear function model is used to calculate the duration amplification coefficient respectively, so as to realize the accurate quantification of the cumulative effect of working condition duration on motor damage, and provide time dimension correction parameters for subsequent single information source damage index calculation.

[0046] During operation, coal mining machine motors suffer from various damages, including heat accumulation due to current overload, insulation aging caused by excessive temperature, and mechanical fatigue due to abnormal vibration. The degree of damage varies non-linearly with the duration of the operating condition. Under the same operating condition level, the duration directly determines the actual damage to the motor. Therefore, the cumulative effect of time must be incorporated into the damage assessment system, and the degree of damage under different durations must be corrected using a duration amplification factor. Traditional motor health assessment techniques only use fixed parameters to characterize damage, failing to consider the non-linear correlation between operating condition duration and damage degree. They cannot distinguish between short-term and long-term sustained anomalies, resulting in significant deviations between assessment results and the actual aging state of the motor. This invention addresses the core problem of traditional techniques failing to quantify the cumulative damage over time through an overall scheme that independently times the operating condition duration and matches non-linear function models by type.

[0047] First, the operation duration recording and judgment operation is performed. A working condition state machine is established to monitor the changes in the working condition level of three indicators: current, temperature, and vibration in real time. When the working condition level of any indicator changes, the time of occurrence of the change is recorded immediately. By subtracting the time of the last change in the working condition level of the indicator from the current time, the duration of the indicator under the current working condition is calculated. The duration timing of the three indicators of current, temperature, and vibration is independent of each other, which can adapt to the actual scenario where the changes of various indicators are not synchronized under complex working conditions downhole, and ensure the authenticity and accuracy of the duration data.

[0048] Next, the current duration amplification factor is calculated. This step is performed using a logarithmic function model based on the current load level. This model can accurately reflect the marginal diminishing characteristics of the heat accumulation effect of current overload. The current duration amplification factor for the underload level is obtained by subtracting the natural logarithm of the sum of 0.03 multiplied by 1 and the duration divided by 1800. The current duration amplification factor for the light load level and the rated level remains constant at 1.0. The current duration amplification factor for the overload, heavy load, and overload levels is calculated according to the corresponding logarithmic formula. The unit of duration is uniformly set to seconds to ensure accurate quantification of the time accumulation effect of current damage.

[0049] Then, the temperature duration amplification factor calculation is performed. The calculation is completed using an exponential function model based on the temperature level. This model strictly follows the accelerated thermal aging law of insulation materials based on the Arrhenius equation. The temperature duration amplification factor for the normal level is constant at 1.0. The temperature duration amplification factor for the concern level is calculated using a logarithmic formula. The temperature duration amplification factor for the over-temperature level is calculated with the natural constant e as the base and 0.0003 multiplied by the duration as the exponent. The temperature duration amplification factor for the over-temperature level is calculated with the natural constant e as the base and 0.0008 multiplied by the duration as the exponent. The duration is in seconds, which realizes the accurate quantification of accelerated damage caused by abnormal temperature.

[0050] Finally, the vibration duration amplification factor is calculated. This step uses a model combining logarithmic and linear functions based on the vibration level. This model matches the nonlinear characteristics of mechanical fatigue accumulation. The vibration duration amplification factors for the excellent, good, and qualified levels are calculated using corresponding logarithmic formulas. For alarm and danger levels, a linear correction term is added to the logarithmic model. The duration is measured in seconds, fully reflecting the cumulative mechanical fatigue effect of vibration anomalies over time. This provides reliable time correction parameters for subsequent vibration damage index calculations. It should be noted that the vibration duration amplification factor α for the current load... I The formula for calculating (τ) is: Underload: α I (τ)=1-0.05·ln(1+τ / 3600), long-term underload has a slight improvement, but the improvement rate decreases with time; Light load: α I (τ)=1.0, baseline condition, linear accumulation of time; Overload: α I (τ)=1+0.2·ln(1+τ / 300), heat accumulation effect, the initial damage is large, and the damage increases and slows down in the later stage; Overload: α I (τ) = 1 + 0.5·ln(1 + τ / 60), which is due to the combined effect of strong thermal stress and mechanical stress; Overload: α I(τ)=1+1.0·ln(1+τ / 10), rapid thermal aging of insulation; For over-temperature and over-temperature conditions, an exponential function model is used to describe the accelerated aging process of insulation materials. The model coefficients are determined based on the Arrhenius equation and the temperature-life relationship corresponding to the insulation class, enabling precise quantification of the cumulative time of temperature damage. The temperature duration amplification factor α... T The formula for calculating (τ) is: Normal: α T (τ)=1.0, base case; Overtemperature: α T (τ) = exp(0.0002·τ), indicating exponential growth; Overheating: α T (τ) = exp(0.0005·τ), indicating exponential growth; Vibration duration amplification factor (based on ISO standard classification) α V The formula for calculating (τ) is: V1 - Excellent: α V (τ)=0.95+0.05·ln(1+τ / 86400), indicating a slight improvement in long-term excellent performance; V2-Good: α V (τ)=1.0, base case; V3-Qualified: α V (τ) = 1 + 0.1·ln(1 + τ / 3600), representing slow fatigue accumulation; V4-Alarm: α V (τ)=1+0.3·ln(1+τ / 600), indicating relatively rapid damage development; V5 - Danger: α V (τ)=1+0.8·ln(1+τ / 60), rapid accumulation of damage may lead to sudden failure; Where τ is the duration of the current operating condition (in seconds).

[0051] In one embodiment, step S5 calculates the comprehensive damage rate at the current moment based on the basic damage factor, additional damage factor, time amplification factor, and synergistic amplification factor among multi-source information. The cumulative equivalent operating time is obtained by integrating the comprehensive damage rate over time. Finally, the remaining health is calculated based on the cumulative equivalent operating time and the motor's design life to complete the quantitative assessment of the motor's health status. Specifically, this includes: S51, calculate the current damage index, temperature damage index and vibration damage index respectively. The current damage index is obtained by multiplying the current basic damage factor by the current duration amplification factor. The temperature damage index is obtained by multiplying the temperature basic damage factor by the temperature duration amplification factor. The vibration damage index is obtained by multiplying the total vibration damage factor by the vibration duration amplification factor. S52, based on the coupling effect of abnormal current, temperature and vibration states, introduces a synergistic amplification factor. The specific calculation process is as follows: First, multiple synergistic triggering conditions are set, including current and temperature synergistic triggering conditions, current and vibration synergistic triggering conditions, temperature and vibration synergistic triggering conditions, and triple synergistic triggering conditions of current, temperature and vibration; then, it is determined whether each synergistic triggering condition is valid. When the current load level is overload or above and the temperature level is overtemperature or above, the current and temperature synergistic triggering condition is valid and takes a value of 1; otherwise, it is 0. When the current load level is rated or above and the vibration level is alarm or above, the current and vibration coordinated triggering condition is met and the value is 1; otherwise, it is 0. When the temperature level is over-temperature or above and the vibration level is alarm or above, the temperature and vibration coordinated triggering condition is met and the value is 1; otherwise, it is 0. When the current, temperature, and vibration simultaneously reach the threshold of each of the above-mentioned synergistic triggering conditions, the triple synergistic triggering condition is met and takes a value of 1; otherwise, it is 0. Finally, the values ​​of each coordinated triggering condition are multiplied by their corresponding coordinated coefficients and summed to obtain the coordinated amplification coefficients. The coordinated coefficients for current and temperature are 0.1, for current and vibration are 0.15, for temperature and vibration are 0.1, and for the triple coordinated coefficient is 0.2. S53 uses a weighted average method to fuse the damage indices of each information source and multiplies them by a synergistic amplification effect. The specific calculation process is as follows: First, set the weight coefficients for the damage indices of current, temperature and vibration, where the weight of current is 0.35, the weight of temperature is 0.3, the weight of vibration is 0.35, and the sum of the three weight coefficients is 1. Then, the current damage index is multiplied by the current weight, the temperature damage index is multiplied by the temperature weight, and the vibration damage index is multiplied by the vibration weight, and then summed to obtain the weighted basic damage rate. Finally, the weighted basic damage rate is multiplied by one and the sum of the synergistic amplification factor to obtain the comprehensive damage rate at the current moment. S54 maps the actual running time to the equivalent running time under the benchmark condition through the comprehensive damage rate. Specifically, it is obtained by integrating the comprehensive damage rate over time. In the actual discretization processing system, the cumulative sum approximation is used, that is, the comprehensive damage rate in each sampling or calculation cycle is multiplied by the duration of that cycle and then summed to obtain the cumulative equivalent running time. S55 is calculated based on the motor's design life under rated operating conditions. Specifically, it is obtained by subtracting the ratio of the cumulative equivalent operating time to the motor's design life from 1 and then multiplying by 100%. Health status is assessed by classifying the remaining health percentage. When the remaining health percentage is greater than or equal to 80%, it is considered a healthy state; when the remaining health percentage is greater than or equal to 60% but less than 80%, it is considered a sub-healthy state; when the remaining health percentage is greater than or equal to 40% but less than 60%, it is considered a warning state; and when the remaining health percentage is less than 40%, it is considered a dangerous state. Corresponding maintenance suggestions and alarm signals are triggered according to different health levels.

[0052] As described in steps S51-S55 above, by sequentially completing the calculation of single information source damage index, multi-source information collaborative amplification coefficient, comprehensive damage rate, cumulative equivalent operating time, remaining health status and status assessment, the previously obtained damage factors, duration amplification coefficient and other parameters are integrated and quantified in multiple dimensions, and finally the accurate quantitative assessment and graded early warning of the health status of the coal mining machine motor are realized, providing a reliable basis for predictive maintenance of equipment.

[0053] During underground operation, coal mining machine motors experience a coupled amplification effect from three abnormal states: current, temperature, and vibration. The damage level of a single indicator cannot represent the overall damage level of the motor. Actual operating time and motor damage are non-linearly correlated, and directly using actual operating time cannot accurately reflect the degree of motor aging. The remaining health status needs to be presented through a unified quantitative indicator to guide maintenance decisions. Therefore, it is necessary to fuse multi-source damage parameters and convert them into a readily identifiable health index. Traditional motor health assessment techniques simply superimpose the results of single indicators, failing to consider the synergistic amplification damage caused by multi-source abnormal coupling, and failing to convert actual operating time into equivalent damage time. This results in the inability to achieve continuous quantification and accurate grading of health status, and the assessment results cannot support predictive maintenance. This invention addresses the various shortcomings of traditional assessment techniques through a complete scheme that weights and fuses damage indices, calculates synergistic amplification coefficients, integrates to obtain equivalent operating time, calculates remaining health status, and grades it.

[0054] First, the damage index calculation for a single information source is performed. The previously determined basic damage factor and total vibration damage factor are multiplied by their corresponding time amplification coefficients to obtain the current damage index, temperature damage index, and vibration damage index. This integrates three parameters—operating condition level, fault-related damage, and time cumulative effect—into a single-information-source quantitative damage index, completing the initial integration of damage across various dimensions. Next, the multi-source information collaborative amplification coefficient calculation is performed. Based on the physical coupling mechanism of current, temperature, and vibration anomalies, four collaborative triggering conditions are set. The triggering conditions are determined according to the operating condition level of each indicator, and corresponding values ​​are assigned. The triggering values ​​are then multiplied by preset collaborative coefficients and summed to obtain the collaborative amplification coefficients. The current-temperature collaborative coefficient is 0.1, the current-vibration collaborative coefficient is 0.15, the temperature-vibration collaborative coefficient is 0.1, and the triple collaborative coefficient is 0.2. This accurately quantifies the nonlinear damage amplification effect when multiple anomalies occur concurrently. The current damage index... D I ( t The formula for calculating ) is: D I ( t )= l I × α I ( t ) Temperature damage index D T ( t The formula for calculating ) is: D T ( t )= l T × α T ( t ) Vibration damage index (including fault diagnosis) D V ( t The formula for calculating ) is: D V ( t )= × α V ( t ) in, l I Indicates the base current damage rate. l T This indicates the base temperature damage rate.

[0055] Then, the comprehensive damage rate calculation operation is performed. The current weight is set to 0.35, the temperature weight to 0.3, and the vibration weight to 0.35. The sum of the three weight coefficients is 1. A weighted average method is used to fuse the three types of damage indices to obtain the weighted basic damage rate. Then, the weighted basic damage rate is multiplied by the sum of the synergistic amplification coefficients to obtain the comprehensive damage rate at the current moment. D total ( t,t The calculation formula is: D total ( t,t )=( w I · D I + w T · D T + w V · D V )×(1+ ); in, w I Indicates current weighting. w T Indicates temperature weighting. w V This represents the vibration weight.

[0056] The This represents the total cooperative amplification factor. The calculation formula is: ; The synergistic amplification factor was set based on the physical mechanism of multi-fault interaction and experimental verification data: there is a direct thermal effect relationship between current overload and temperature rise; there is an electromagnetic-mechanical coupling relationship between abnormal current and excessive vibration; and there is a material performance degradation relationship between temperature rise and increased vibration. Each factor was determined through regression analysis of a large amount of historical operating data and accelerated life tests, reflecting the nonlinear damage amplification effect under concurrent multi-fault conditions.

[0057] It should be noted that, in practice, the coordination coefficient can be adaptively adjusted according to the motor type, operating history, and actual working conditions to improve the accuracy of health assessment.

[0058] The specific calculation methods for each collaborative item are as follows: Current-temperature synergy : Current-vibration coordination : Temperature-vibration coordination : Triple Synergy : Where R represents the rated current, Represents current. Indicates the effective value of vibration velocity. Indicates temperature.

[0059] The synergy coefficients are determined by the following method: first, a preliminary range is determined based on the physical mechanism of each fault coupling; then, experimental data is obtained through orthogonal accelerated life tests; the optimal estimated value of the coefficients is determined by multivariate nonlinear regression analysis; and finally, engineering corrections are made in conjunction with field operation statistics.

[0060] Specifically, the current-temperature co-coefficient =0.15, current-vibration coordination coefficient =0.12, temperature-vibration compatibility factor =0.10, triple synergy coefficient k ITV =0.25, used to fully integrate multi-source damage and synergistic effects, to obtain the core parameter characterizing the overall damage rate of the motor.

[0061] Next, the cumulative equivalent running time calculation operation is performed. The comprehensive damage rate is integrated over time to map the actual running time to the equivalent running time under the baseline operating condition. In the discretization processing system, an accumulation and approximate calculation is used. The comprehensive damage rate of each cycle is multiplied by the cycle duration and then accumulated to obtain the cumulative equivalent running time. This unifies the time damage scale under different operating conditions and eliminates the evaluation bias caused by differences in operating conditions. The calculation of the cumulative equivalent running time is as follows: The actual running time is converted into the equivalent running time under the baseline operating conditions (light load, normal temperature, and good vibration). The calculation formula is: Divide the different operating conditions within a certain period into k segments, and calculate the equivalent running time of each segment. The cumulative equivalent running time during this period is... for: Finally, the remaining health calculation and status assessment are performed. The remaining health is obtained by subtracting the ratio of the cumulative equivalent operating time to the motor's design life from 1, and then multiplying by 100%. In this remaining health calculation, the design life of the motor under baseline operating conditions is assumed to be L. design (Unit: hours), then the remaining health RH is: in, T eq,total ( d The daily cumulative equivalent operating time is used. The motor status is divided into four levels—healthy, sub-healthy, warning, and dangerous (shutdown for maintenance)—based on the remaining health value range. Different levels correspond to different maintenance recommendations and alarm signals. This design enables a clear presentation of the health status and accurate early warning, providing quantitative support for the safe operation and maintenance management of coal mining machine motors.

[0062] When RH≥80% (healthy), the maintenance suggestion of "normal operation, periodic inspection" will be output; When 60%≤RH<80% (sub-healthy), the maintenance suggestion is "strengthen monitoring and analyze the cause". When 40%≤RH<60% (early warning), a maintenance suggestion of "planned maintenance preparation" will be output. When RH < 40% (shutdown for maintenance), a maintenance suggestion of "schedule a shutdown for maintenance in the near future" will be output.

[0063] It should be noted that this embodiment also discloses a method for predicting remaining useful life, specifically: based on the maximum value of the historical daily average damage rate over the past 30 days. D m Predicted Remaining Useful Life (RUL): RUL =( RH / 100%)× L design / D m Among them, the RUL Indicates the remaining useful life.

[0064] The daily average damage rate is calculated using a time-slice weighted average method, which specifically includes: Time points were collected and recorded, recording all moments of change in current, temperature, and vibration condition levels. These data were then merged and sorted to obtain the time series t0, t1, ..., t N ; Time segmentation: Divide a day into N time segments [t] k−1 ,t k Within each time period, the operating conditions of each indicator remain constant; For each time period k, the overall damage rate D for that time period is calculated. tk ; Weighted average, calculate the daily average damage rate: ; The maximum value among the calculated historical daily average damage rates over the most recent 30 days is taken as the [value]. D m。

[0065] like Figure 2 As shown, the present invention also discloses a multi-source information fusion assessment system for the health status of a coal mining machine motor, comprising: The data acquisition module is used to acquire the current, temperature, and vibration signals of the coal mining machine motor in real time during operation. The grading module is used to classify the load level of the collected current signal, the temperature rise level of the temperature signal, and the vibration intensity level of the vibration signal according to the preset working condition grading rules. The first calculation module is used to determine the corresponding basic damage factor based on the divided current, temperature and vibration condition levels, and to perform spectrum analysis on the vibration signal. Based on the extracted characteristic frequencies, the mechanical fault type is identified and the corresponding additional damage factor is calculated, thereby determining the total vibration damage factor. The second calculation module is used to record the duration of the current working condition and calculate the duration amplification factor of current, temperature and vibration respectively using a preset nonlinear function model based on the duration and working condition level. The quantitative assessment module is used to calculate the comprehensive damage rate at the current moment based on the basic damage factor, the additional damage factor, the time amplification coefficient, and the synergistic amplification coefficient among the multi-source information. The cumulative equivalent running time is obtained by integrating the comprehensive damage rate over time. Finally, the remaining health is calculated based on the cumulative equivalent running time and the motor design life to complete the quantitative assessment of the motor's health status.

[0066] like Figure 3 As shown, the present invention also discloses an implementation flowchart for multi-source information fusion assessment of the health status of a coal mining machine motor.

[0067] In one embodiment, the first computing module includes: The setting unit is used to set the corresponding current-based damage factor, temperature-based damage factor and vibration-based damage factor for each level based on the classification results of current, temperature and vibration working conditions. The first calculation unit is used to perform spectrum transformation processing on the collected vibration signal to obtain a spectrum diagram, calculate the rotor rotation frequency based on the current speed of the motor, and read the bearing geometric parameters from the pre-stored bearing parameter database, and calculate the bearing fault characteristic frequency in combination with the rotor rotation frequency. The adjustment unit is used to search for peak values ​​at each characteristic frequency in the spectrum of the vibration signal. If the peak value exceeds the preset threshold, it is determined that there is a corresponding fault type, and an additional damage factor is assigned according to the fault type. When multiple faults exist at the same time, the additional damage factors are summed. The second calculation unit is used to combine the basic vibration damage factor and the additional damage factor to calculate the total vibration damage factor.

[0068] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described multi-source information fusion evaluation method for the health status of a coal mining machine motor.

[0069] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for multi-source information fusion evaluation of the health status of a coal mining machine motor.

[0070] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0071] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0072] The above description is merely a preferred embodiment of the present invention and does not limit the scope of this application. Any equivalent results or equivalent process transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.

Claims

1. A multi-source information fusion assessment method for the health status of a coal mining machine motor, characterized in that, Includes the following steps: Real-time acquisition of current, temperature, and vibration signals from the coal mining machine motor during operation; According to the preset working condition classification rules, the collected current signals are classified into load levels, the temperature signals are classified into temperature rise levels, and the vibration signals are classified into vibration intensity levels according to international standards. Based on the classification of current, temperature and vibration conditions, the corresponding basic damage factors are determined, and the vibration signal is subjected to spectrum analysis. The mechanical fault type is identified according to the extracted characteristic frequency and the corresponding additional damage factor is calculated, thereby determining the total vibration damage factor. Record the duration of the current working condition, and calculate the duration amplification factor of current, temperature and vibration respectively using a preset nonlinear function model based on the duration and working condition level. Based on the basic damage factor, additional damage factor, duration amplification factor, and synergistic amplification factor among multi-source information, the comprehensive damage rate at the current moment is calculated. The cumulative equivalent operating time is obtained by integrating the comprehensive damage rate over time. Finally, the remaining health is calculated based on the cumulative equivalent operating time and the motor's design life to complete the quantitative assessment of the motor's health status.

2. The multi-source information fusion assessment method for the health status of a coal mining machine motor according to claim 1, characterized in that, The steps for real-time acquisition of current, temperature, and vibration signals from the coal mining machine motor during operation specifically include: The instantaneous values ​​of the three-phase current are collected synchronously by current sensors installed on the three-phase power supply line of the motor, and the effective value of the three-phase current and the current imbalance are calculated in real time. Temperature signals from bearing and winding measuring points are collected and filtered by temperature sensors embedded in the motor bearing housing and stator windings. The acceleration signal of the housing vibration is collected by a vibration sensor installed on the motor bearing housing, and converted into a velocity signal. The effective value of the vibration velocity is calculated, and the original waveform data of the vibration signal is recorded for subsequent spectrum analysis.

3. The multi-source information fusion assessment method for the health status of a coal mining machine motor according to claim 1, characterized in that, The steps of classifying the acquired current signal into load levels according to preset operating condition classification rules, classifying the temperature signal into temperature rise levels, and classifying the vibration signal into vibration intensity levels according to international standards specifically include: Based on the ratio of the average value of the three-phase current to the rated current of the motor, and combined with the current imbalance, a comprehensive judgment is made to classify the current operating conditions into multiple load levels. Based on the comparison between the highest temperature among the bearing measuring point and the winding measuring point and the temperature threshold corresponding to the motor insulation class, the temperature conditions are divided into multiple temperature rise levels. Vibration conditions are classified into multiple vibration intensity levels based on the effective value of vibration velocity.

4. The multi-source information fusion assessment method for the health status of a coal mining machine motor according to claim 1, characterized in that, The steps of determining the basic damage factor based on the divided current, temperature, and vibration condition levels, performing spectral analysis on the vibration signal, identifying the mechanical fault type based on the extracted characteristic frequencies, calculating the corresponding additional damage factor, and then determining the total vibration damage factor specifically include: Based on the classification results of various working conditions such as current, temperature and vibration, corresponding basic damage factors for current, temperature and vibration are set for each level. The collected vibration signal is processed by spectrum transformation to obtain a spectrum diagram. The rotor rotation frequency is calculated based on the current motor speed. The bearing geometric parameters are read from the pre-stored bearing parameter database and the bearing fault characteristic frequency is calculated in combination with the rotor rotation frequency. The peak values ​​at each characteristic frequency are searched in the spectrum of the vibration signal. If the peak value exceeds the preset threshold, the corresponding fault type is determined, and an additional damage factor is assigned according to the fault type. When multiple faults exist at the same time, the additional damage factors are summed. The total vibration damage factor is calculated by combining the basic vibration damage factor and the additional damage factor.

5. The multi-source information fusion assessment method for the health status of a coal mining machine motor according to claim 1, characterized in that, The steps of recording the duration of the current operating condition and calculating the duration amplification factors of current, temperature, and vibration using a preset nonlinear function model based on the duration and operating condition level specifically include: Real-time monitoring of changes in the operating condition levels of various indicators such as current, temperature, and vibration, and independent recording of the duration of each indicator under the current operating condition; The current duration amplification factor is calculated using a logarithmic function model based on the current load level. The temperature duration amplification factor is calculated using an exponential function model based on the temperature level. Based on the vibration level, the vibration duration amplification factor is calculated using a model combining logarithmic and linear functions.

6. The multi-source information fusion assessment method for the health status of a coal mining machine motor according to claim 1, characterized in that, The steps of calculating the comprehensive damage rate at the current moment based on the basic damage factor, additional damage factor, time amplification factor, and synergistic amplification factor among multi-source information, obtaining the cumulative equivalent operating time by integrating the comprehensive damage rate over time, and finally calculating the remaining health based on the cumulative equivalent operating time and the motor design life, specifically include: The current damage index, temperature damage index, and vibration damage index are calculated separately. The current damage index is obtained by using the current basic damage factor and the current duration amplification factor. The temperature damage index is obtained by using the temperature basic damage factor and the temperature duration amplification factor. The vibration damage index is obtained by using the total vibration damage factor and the vibration duration amplification factor. Based on the coupling effect of abnormal current, temperature and vibration, multiple coordinated triggering conditions are set, and the validity of the coordinated triggering conditions is determined according to each working condition level. The values ​​of each triggering condition are multiplied by the corresponding coordination coefficient and then summed to obtain the coordinated amplification coefficient. The damage indices of each information source are fused using a weighted average method to obtain the overall damage rate at the current moment; The actual running time is mapped to the equivalent running time under the baseline working condition through the comprehensive damage rate, and the cumulative equivalent running time is obtained by integrating the comprehensive damage rate over time. Based on the motor's design life, calculate the current remaining health level, and then conduct a health status classification assessment based on the range of remaining health level values.

7. A multi-source information fusion assessment system for the health status of a coal mining machine motor, characterized in that, include: The data acquisition module is used to acquire the current, temperature, and vibration signals of the coal mining machine motor in real time during operation. The grading module is used to classify the load level of the collected current signal, the temperature rise level of the temperature signal, and the vibration intensity level of the vibration signal according to the preset working condition grading rules. The first calculation module is used to determine the corresponding basic damage factor based on the divided current, temperature and vibration condition levels, and to perform spectrum analysis on the vibration signal. Based on the extracted characteristic frequencies, the mechanical fault type is identified and the corresponding additional damage factor is calculated, thereby determining the total vibration damage factor. The second calculation module is used to record the duration of the current working condition and calculate the duration amplification factor of current, temperature and vibration respectively using a preset nonlinear function model based on the duration and working condition level. The quantitative assessment module is used to calculate the comprehensive damage rate at the current moment based on the basic damage factor, the additional damage factor, the time amplification coefficient, and the synergistic amplification coefficient among the multi-source information. The cumulative equivalent running time is obtained by integrating the comprehensive damage rate over time. Finally, the remaining health is calculated based on the cumulative equivalent running time and the motor design life to complete the quantitative assessment of the motor's health status.

8. The multi-source information fusion assessment system for the health status of a coal mining machine motor according to claim 7, characterized in that, The first computing module includes: The setting unit is used to set the corresponding current-based damage factor, temperature-based damage factor and vibration-based damage factor for each level based on the classification results of current, temperature and vibration working conditions. The first calculation unit is used to perform spectrum transformation processing on the collected vibration signal to obtain a spectrum diagram, calculate the rotor rotation frequency based on the current speed of the motor, and read the bearing geometric parameters from the pre-stored bearing parameter database, and calculate the bearing fault characteristic frequency in combination with the rotor rotation frequency. The adjustment unit is used to search for peak values ​​at each characteristic frequency in the spectrum of the vibration signal. If the peak value exceeds the preset threshold, it is determined that there is a corresponding fault type, and an additional damage factor is assigned according to the fault type. When multiple faults exist at the same time, the additional damage factors are summed. The second calculation unit is used to combine the basic vibration damage factor and the additional damage factor to calculate the total vibration damage factor.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images