Motor fault diagnosis system and method based on big data analysis

Abstract

The invention discloses a motor fault diagnosis system and method based on big data analysis, relates to the technical field of motor fault diagnosis, and solves the technical problem that fault diagnosis cannot be reasonably carried out due to the fact that different signals cannot be collected and recognized at different time periods in the prior art, in particular to a performance state detection unit. Performing operation performance state evaluation detection on the motor, performing vibration signal acquisition on the motor through a vibration sensor, and dividing the operation state of the motor according to vibration signal analysis to obtain a floating state and a stable state; the time-phased state influence evaluation unit is used for performing time-phased state influence evaluation on the motor, and marking the corresponding floating time periods as floating time periods and stable time periods according to the floating state and the stable state respectively; carrying out time-phased evaluation; and the fault analyzing and processing unit is used for acquiring and analyzing according to the operating parameters of the motor when the time-phased state influence evaluation of the motor is normal.

Description

Technical Field

[0001] The present invention relates to the technical field of motor fault diagnosis, and in particular to a motor fault diagnosis system and method based on big data analysis. Background Art

[0002] Motor fault diagnosis is a technical process that monitors the motor's operating status, collects and analyzes data to identify the fault type, locate the fault, and determine the cause of the fault. It is a key link in ensuring the safe and stable operation of the motor and reducing downtime losses, and is widely used in industrial manufacturing, energy, transportation and other fields.

[0003] However, in the existing technology, it is impossible to infer the real-time operating status based on the internal vibration signal of the motor, and it is impossible to divide the time period according to the operating status, resulting in the inability to collect, identify and process different signals in different time periods, so that fault diagnosis cannot be performed reasonably, reducing the efficiency of fault diagnosis. In addition, multiple feature collection, monitoring and calculation processing are not performed when the motor is evaluated to be normal, and fault prediction cannot be performed in a timely manner.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the Invention

[0005] The purpose of the present invention is to solve the above-mentioned problems and to propose a motor fault diagnosis system and method based on big data analysis.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The motor fault diagnosis system based on big data analysis includes a motor fault diagnosis platform, where the motor fault diagnosis platform has communication connections with:

[0008] The performance status detection unit evaluates and detects the motor's operating performance status, collects vibration signals from the motor through a vibration sensor, and divides the motor's operating status into floating state and stable state based on vibration signal analysis;

[0009] A time-based state impact assessment unit is configured to perform time-based state impact assessment on the motor, wherein the corresponding floating periods are marked as floating periods and stable periods according to the floating state and stable state, respectively; and the assessment is performed in time periods;

[0010] The fault analysis and processing unit collects and analyzes the motor's operating parameters when the motor's time-phased status impact assessment is normal.

[0011] As a preferred embodiment of the present invention, the process of the performance status detection unit is as follows:

[0012] The vibration sensor is used to collect vibration signals from the motor, and the time domain signals are converted into frequency domain signals through fast Fourier transform (FFT). The rolling bearing inside the motor is used as the analysis component for vibration generation, and the frequency of each part of the rolling bearing is analyzed to obtain the frequency of the rolling element passing through the inner ring. Specifically, , where n r is the rotor speed in r / min; d is the rolling element diameter, D is the bearing pitch diameter, α is the contact angle, Z is the number of rolling elements, and fi is the frequency of the rolling elements passing through the inner ring;

[0013] Collect the frequency fo of the rolling element passing through the outer ring, specifically: ;

[0014] Collect the cage frequency fc, specifically: ;

[0015] Collect the rolling body spin frequency fb, specifically: .

[0016] As a preferred embodiment of the present invention, the vibration frequencies generated by the rolling bearings in the motor are summarized, and the frequencies are monitored for floating according to the increase in the motor running time. If the increase in the vibration amplitude corresponding to any type of vibration frequency exceeds the corresponding set single amplitude threshold, or the cumulative increase in the vibration amplitude of all types of vibration frequencies exceeds the set multiple amplitude thresholds, it is inferred that the running state of the motor is a floating state; if the increase in the vibration amplitude corresponding to any type of vibration frequency does not exceed the corresponding set single amplitude threshold, and the cumulative increase in the vibration amplitude of all types of vibration frequencies does not exceed the set multiple amplitude thresholds, it is inferred that the running state of the motor is a stable state.

[0017] As a preferred embodiment of the present invention, the process of the time-period status impact assessment unit is as follows:

[0018] During the floating period, the electrical signal or vibration signal of the motor is monitored in real time, such as the current and the amperage. The signal collected by the real-time monitoring is set as a real-valued signal, and the symbol x(t) is set, and t represents time. The real-valued signal is converted into an analytical signal z(t) through the Hilbert transform as an operation method.

[0019] According to the real-valued signal x(t) collected in real time, the signal y(t) after Hilbert transformation is obtained by calculation; the calculation formula is: ,in, Expressed as an integral variable;

[0020] After obtaining y(t), the analytical signal z(t) is composed of the original real-valued signals x(t) and y(t), that is, z(t)=x(t)+jy(t), where j represents the imaginary unit;

[0021] According to the above signal conversion, the parameters required for fault feature extraction are calculated.

[0022] As a preferred embodiment of the present invention, the amplitude envelope A(t) is calculated using the following formula: ;

[0023] Instantaneous phase , the specific calculation formula is: ;

[0024] The instantaneous frequency f(t) is obtained by taking the derivative of the instantaneous phase and then performing a certain transformation. The specific formula is: ;

[0025] During the floating period, the amplitude envelope, instantaneous phase, and instantaneous frequency are counted, and the fluctuation of the corresponding values ​​at each moment is recorded. If the corresponding floating span of adjacent moments increases significantly and the significant increase in span persists, it indicates that the current motor state is abnormal. A floating high impact signal is generated and sent to the motor fault diagnosis platform. A significant increase is indicated by the floating span exceeding 1.2 times the set threshold.

[0026] If the corresponding floating span at adjacent moments does not increase significantly, or the significant increase in span occurs only once, it indicates that the current state of the motor is normal, and a floating low impact signal is generated and sent to the motor fault diagnosis platform.

[0027] As a preferred embodiment of the present invention, during a stable period, the electrical signal or vibration signal of the motor is monitored in real time, and the corresponding performance parameters of the signal are recorded according to the real-time signal. The performance parameters at the time when the signal is generated are statistically analyzed at each moment according to the time when the signal is generated and the motor operation intensity in the time period corresponding to the time. When there is no span and no obvious increase, the performance parameters at all moments are statistically analyzed and a parameter range is constructed.

[0028] Within the parameter range, the operating scenarios are classified based on the signal generation time and the fluctuation of the motor operating intensity. For example, if the generated signal is a current rising or falling signal, the motor operating intensity increases or decreases; the corresponding operating scenarios are divided into signal aggravation and signal deceleration, and intensity increase and intensity decrease.

[0029] The operating scenario signal trend is set as the embodiment end of the intensity trend, and the intensity trend is the generation end of the signal trend; when the floating trends of each stage of the embodiment end and the generation end are synchronized, it indicates that there is no deviation in the motor state impact assessment; on the contrary, when the floating trends of each stage of the embodiment end and the generation end are not synchronized, it indicates that the motor state impact assessment is deviated, and the hardware equipment required for motor state detection needs to be re-debugged and the motor operation period needs to be reselected.

[0030] As a preferred embodiment of the present invention, when there is no deviation, the trend of parameter alternation at each moment within the parameter range is recorded, and while recording the overall trend, the fluctuations at each adjacent moment are simultaneously recorded, and then the proportional relationship between the intensity parameter and the signal type and the corresponding recorded parameter fluctuation trend is determined;

[0031] If the proportional relationship between the current parameter floating trend and the intensity floating trend in the recorded parameter floating trend is not synchronized, the motor operating state is abnormal, and a job adaptive control abnormality signal is generated and sent to the motor fault diagnosis platform. The motor fault diagnosis platform then debugs the motor's operating performance and changes the operating scenario.

[0032] If the proportional relationship between the current floating trend of the parameter and the intensity floating trend is synchronized in the recorded parameter floating trend, the motor operation status is normal, and a normal operation adaptive control signal is generated and sent to the motor fault diagnosis platform;

[0033] If, during a parameter floating trend, the floating trends of adjacent parameters fluctuate but the overall trend remains constant, and the current signal type indicates a favorable trend for the corresponding parameter, a motor status affected signal is still generated and sent to the motor fault diagnosis platform; the motor fault diagnosis platform adjusts the corresponding operating parameters of the motor and records the signal change process;

[0034] If the current signal type indicates an unfavorable trend in the corresponding parameter, a motor status repair signal is still generated and sent to the motor fault diagnosis platform; the motor fault diagnosis platform suspends the operation of the motor and maintains the motor hardware involved in the unfavorable trend parameter;

[0035] If the floating trend of adjacent parameters is reciprocating with the overall trend, the favorable or unfavorable trend of the parameter floating will be reflected by the current signal type, and a corresponding normal signal or abnormal signal will be generated and sent to the motor fault diagnosis platform.

[0036] As a preferred embodiment of the present invention, the process of the fault analysis processing unit is as follows:

[0037] According to the historical maintenance process, the abnormal operating parameters are set as fault characteristics; the fault factor set is constructed based on the fault characteristics, and the fault factor set U{u1, u2…, u n}, and construct the comment set V{v1, v2…, v m}, specifically a description set of fault severity or fault possibility; specifically, v1 represents no fault and v2 represents a mild fault;

[0038] According to the importance of each fault symptom to fault diagnosis, the weight vector A{a1, a2…, a n},and ;

[0039] After completing the weight vector setting, the fault features corresponding to the fault factor set are judged one by one to obtain their membership to each element in the comment set V and construct the fuzzy evaluation matrix, which is specifically: ; where r ij Representation factor u i Comments v j The membership degree of ; i and j are natural numbers from 1 to n and 1 to m respectively;

[0040] Get the comprehensive evaluation result vector B, and , It is a fuzzy composite operation;

[0041] And the comprehensive evaluation result vector B={b1, b2…, b m}, where b j Indicates that the motor belongs to the comment v j The motor fault is determined based on the size of each element in the comprehensive evaluation result vector B. If b2 is the largest, the possibility of a mild motor fault is the greatest. A fault prediction signal is generated and sent to the motor fault diagnosis platform according to the type of fault prediction degree.

[0042] Motor fault diagnosis method based on big data analysis, the fault diagnosis method is as follows:

[0043] Performance status detection: evaluate the motor's operating performance status, collect vibration signals from the motor through a vibration sensor, and classify the motor's operating status into floating state and stable state based on vibration signal analysis;

[0044] Time-based state impact assessment: The motor is assessed for time-based state impact, wherein the corresponding floating periods are marked as floating periods and stable periods according to the floating state and stable state respectively; and the assessment is performed in time periods;

[0045] Fault analysis and processing: When the motor's time-phased status impact assessment is normal, data is collected and analyzed based on the motor's operating parameters.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. In the present invention, the motor's operating performance status is evaluated and tested, and the motor's operating status is divided according to the performance status detection, so that targeted fault analysis and processing can be performed according to different operating statuses, thereby improving the accuracy of motor fault diagnosis. At the same time, the real-time performance status evaluation can also reflect whether there is a fault in the current operating status, so as to facilitate timely maintenance and reduce the wear and tear on the operating hardware caused by the motor fault.

[0048] 2. In the present invention, the motor state impact assessment is performed in different time periods, wherein the corresponding floating time periods are marked as floating time periods and stable time periods according to the floating state and stable state, respectively. Different algorithms are used to perform state impact assessment for different types of time periods to infer whether there are operating risks for the motor in the current operating period, thereby performing targeted impact assessment to improve the comprehensiveness of fault diagnosis and ensure diagnostic efficiency.

[0049] 3. In the present invention, when the motor's time-segment status impact assessment is normal, data is collected and analyzed based on the motor's operating parameters to achieve the effect of fault analysis and processing, thereby diagnosing the motor's fault, improving the motor's operating efficiency and stability, and ensuring that the motor fault is diagnosed in a timely manner through parameter analysis when it fails, which is conducive to timely maintenance of the motor fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0051] Figure 1 It is a principle block diagram of the present invention;

[0052] Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0053] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] See also Figure 1 As shown, the motor fault diagnosis system based on big data analysis includes a motor fault diagnosis platform, wherein the motor fault diagnosis platform is communicatively connected to a performance status detection unit, a time period status impact assessment unit, and a fault analysis and processing unit;

[0056] The motor fault diagnosis platform generates a performance status detection signal and sends it to the performance status detection unit;

[0057] After receiving the performance status detection signal, the performance status detection unit evaluates and detects the motor's operating performance status. The motor's operating status is divided according to the performance status detection, so that targeted fault analysis and processing can be carried out according to different operating states, thereby improving the accuracy of motor fault diagnosis. At the same time, the real-time performance status evaluation can also reflect whether there is a fault in the current operating state, so as to facilitate timely maintenance and reduce the wear and tear caused by motor faults on the operating hardware.

[0058] The vibration sensor is used to collect the vibration signal of the motor, and the time domain signal is converted into a frequency domain signal through fast Fourier transform (FFT). It should be explained that when a rolling bearing fails, a specific characteristic frequency will appear;

[0059] The rolling bearing inside the motor is used as the analysis component for vibration generation, and the frequency of each part of the rolling bearing is analyzed to obtain the frequency of the rolling element passing through the inner ring, specifically: , where n r is the rotor speed in r / min; d is the rolling element diameter, D is the bearing pitch diameter, α is the contact angle, Z is the number of rolling elements, and fi is the frequency of the rolling elements passing through the inner ring;

[0060] Collect the frequency fo of the rolling element passing through the outer ring, specifically: ;

[0061] Collect the cage frequency fc, specifically: ;

[0062] Collect the rolling body spin frequency fb, specifically: ;

[0063] The vibration frequencies generated by the rolling bearings in the motor are aggregated and monitored for fluctuations as the motor's operating time increases. If the increase in vibration amplitude corresponding to any type of vibration frequency exceeds a corresponding single amplitude threshold, or if the cumulative increase in vibration amplitude for all types of vibration frequencies exceeds multiple amplitude thresholds, the motor's operating state is inferred to be floating.

[0064] If the increase in vibration amplitude corresponding to any type of vibration frequency does not exceed the corresponding set single amplitude threshold, and the cumulative increase in vibration amplitude of all types of vibration frequencies does not exceed the set multiple amplitude thresholds, it is inferred that the operating state of the motor is stable;

[0065] After obtaining the vibration frequencies of each type, a red line threshold range is set, and the motor state abnormality assessment is performed based on whether any vibration frequency is continuously within the corresponding red line threshold range. If yes, a state abnormality signal is generated and sent to the motor fault diagnosis platform, which debugs the distributed components in the motor; if no, a state normal signal is generated and sent to the motor fault diagnosis platform;

[0066] After receiving the normal state signal, the motor fault diagnosis platform generates a time-segment state impact assessment signal and sends it to the time-segment state impact assessment unit;

[0067] After receiving the data, the time-based state impact assessment unit performs a time-based state impact assessment on the motor. The floating periods are marked as floating periods and stable periods according to the floating state and stable state, respectively. Different algorithms are used to perform state impact assessment for different types of periods to infer whether there are any operating risks for the motor during the current operating period. This allows for targeted impact assessments to improve the comprehensiveness of fault diagnosis and ensure diagnostic efficiency.

[0068] During the floating period, the electrical signal or vibration signal of the motor is monitored in real time, such as the current and the amperage. The signal collected by the real-time monitoring is set as a real-valued signal, and the symbol x(t) is set, and t represents time. The real-valued signal is converted into an analytical signal z(t) through the Hilbert transform as an operation method.

[0069] According to the real-valued signal x(t) collected in real time, the signal y(t) after Hilbert transformation is obtained by calculation; the calculation formula is: ,in, Expressed as an integral variable, it should be explained that the integration of the operation formula is performed on the entire time axis (from negative infinity to positive infinity);

[0070] After obtaining y(t), the analytical signal z(t) is composed of the original real-valued signals x(t) and y(t), that is, z(t) = x(t) + jy(t), where j represents the imaginary unit. It should be explained that this operation expands the real-valued signal to the complex domain.

[0071] According to the above signal conversion, the parameters required for fault feature extraction are calculated:

[0072] Calculate the amplitude envelope A(t). The specific calculation formula is: It should be explained that the amplitude envelope reflects the amplitude changes of the signal at different times. In motor fault detection, for example, when a broken bar fault occurs in the motor rotor, the amplitude envelope of the stator current signal obtained through the Hilbert transform will show characteristics different from the normal state. By analyzing these characteristic changes, it is possible to determine whether a fault exists.

[0073] Instantaneous phase , the specific calculation formula is: ,It needs to be explained that the instantaneous phase can reflect the phase of the signal at each moment;

[0074] The instantaneous frequency f(t) is obtained by taking the derivative of the instantaneous phase and then performing a certain transformation. The specific formula is: When a motor bearing fails, the instantaneous frequency of the vibration signal will show abnormal changes near the bearing characteristic frequency. This characteristic can be used to diagnose the bearing failure.

[0075] During the floating period, the amplitude envelope, instantaneous phase, and instantaneous frequency are counted, and the fluctuation of the corresponding values ​​at each moment is recorded. If the corresponding floating span of adjacent moments increases significantly and the significant increase in span persists, it indicates that the current motor state is abnormal. A floating high impact signal is generated and sent to the motor fault diagnosis platform. A significant increase is indicated by the floating span exceeding 1.2 times the set threshold.

[0076] If the corresponding floating span at adjacent moments does not increase significantly, or the significant increase in span occurs only once, it indicates that the current state of the motor is normal, and a floating low impact signal is generated and sent to the motor fault diagnosis platform;

[0077] During the stable period, the electrical signal or vibration signal of the motor is monitored in real time, and the corresponding performance parameters of the signal are recorded according to the real-time signal. The performance parameters at the time of signal generation are counted at each moment according to the time when the signal is generated and the motor operation intensity of the time period corresponding to the time. When there is no span and no obvious increase, the performance parameters at all moments are counted and the parameter range is constructed;

[0078] Within the parameter range, the operating scenarios are classified based on the signal generation time and the fluctuation of the motor operating intensity. For example, if the generated signal is a current rising or falling signal, the motor operating intensity increases or decreases; the corresponding operating scenarios are divided into signal aggravation and signal deceleration, and intensity increase and intensity decrease.

[0079] And the operating scenario signal trend is set as the embodiment end of the intensity trend, and the intensity trend is the generation end of the signal trend;

[0080] If the floating trends of each stage of the display end and the generating end are synchronized, it indicates that there is no deviation in the motor state impact assessment. On the contrary, if the floating trends of each stage of the display end and the generating end are not synchronized, it indicates that there is a deviation in the motor state impact assessment. The hardware equipment required for motor state detection, such as sensors, needs to be re-debugged and the motor operation period needs to be re-selected.

[0081] When there is no deviation, the trend of parameter alternation at each moment within the parameter range is recorded. While recording the overall trend, the fluctuation of each adjacent moment is also recorded simultaneously. Then the proportional relationship between the intensity parameter and the signal type and the corresponding recorded parameter fluctuation trend is determined. That is, when the current is used as a parameter, the current increases when the intensity increases, and the current generates an increasing signal. The proportional relationship is sequentially used, and the proportional relationship is constant according to the parameter characteristics;

[0082] If the proportional relationship between the current floating trend of the parameter and the intensity floating trend in the recorded parameter floating trend is not synchronized, the motor operation state is abnormal, and an operation adaptive control abnormality signal is generated and sent to the motor fault diagnosis platform. The motor fault diagnosis platform debugs the motor operation performance and changes the operation scenario. If the proportional relationship between the current floating trend of the parameter and the intensity floating trend in the recorded parameter floating trend is synchronized, the motor operation state is normal, and an operation adaptive control normal signal is generated and sent to the motor fault diagnosis platform. The non-synchronization means that according to the original proportional relationship, when the intensity increases, the parameter is increasing, but the parameter does not show an increasing trend.

[0083] If, during a parameter floating trend, the floating trends of adjacent parameters fluctuate but the overall trend remains constant, and the current signal type indicates a favorable trend for the corresponding parameter, a motor status affected signal is still generated and sent to the motor fault diagnosis platform; the motor fault diagnosis platform adjusts the corresponding operating parameters of the motor and records the signal change process;

[0084] If the current signal type indicates an unfavorable trend in the corresponding parameter, a motor status repair signal is still generated and sent to the motor fault diagnosis platform; the motor fault diagnosis platform suspends the operation of the motor and maintains the motor hardware involved in the unfavorable trend parameter;

[0085] It should be explained that an unfavorable trend is represented by the fluctuation of the corresponding parameter of the signal type, which reflects the decrease in motor efficiency. In this case, the current signal type is considered to be an unfavorable trend. The same is true for a favorable trend. The fluctuation of the corresponding parameter of the signal type reflects the increase in motor efficiency.

[0086] If the floating trend of adjacent parameters is reciprocating with the overall trend, the current signal type will be used to reflect the favorable or unfavorable trend of the parameter floating, and a corresponding normal signal or abnormal signal will be generated and sent to the motor fault diagnosis platform;

[0087] After the motor fault diagnosis platform receives a normal signal or a normal signal of adaptive operation control, it generates a fault analysis and processing signal and sends it to the fault analysis and processing unit;

[0088] After receiving the data, the fault analysis and processing unit collects and analyzes the motor's operating parameters when the motor's time-phased status impact assessment is normal to achieve the effect of fault analysis and processing, thereby diagnosing the motor fault, improving the motor's operating efficiency and stability, and ensuring timely fault diagnosis through parameter analysis when the motor fails, which is conducive to timely maintenance of the motor fault;

[0089] According to the historical maintenance process, the abnormal operating parameters are set as fault characteristics, such as the vibration frequency of the vibration characteristic, the floating span of the temperature characteristic, etc.; the fault factor set is constructed based on the fault characteristics, and the fault factor set U{u1, u2…, u n}, and construct the comment set V{v1, v2…, v m}, specifically a set of descriptions of fault severity or fault probability; for example, v1 represents no fault and v2 represents a mild fault;

[0090] According to the importance of each fault symptom to fault diagnosis, the weight vector A{a1, a2…, a n},and ;

[0091] After completing the weight vector setting, the fault features corresponding to the fault factor set are judged one by one to obtain their membership to each element in the comment set V and construct the fuzzy evaluation matrix, which is specifically: ; where r ij Representation factor u i Comments v j i and j are natural numbers from 1 to n and 1 to m respectively; it should be explained that when the vibration is at a certain level, the membership degree r of “no fault” is determined by experiment or experience. 11 =0.3, the membership degree r for “mild fault” 12 =0.5;

[0092] Get the comprehensive evaluation result vector B, and , Fuzzy synthesis operation, such as the maximum-minimum synthesis method, the maximum-product synthesis method, etc. in the prior art;

[0093] And the comprehensive evaluation result vector B={b1, b2…, b m}, where b j Indicates that the motor belongs to the comment v j For example, using the maximum-minimum synthesis method, ;

[0094] The motor fault is determined based on the size of each element in the comprehensive evaluation result vector B. If b2 is the largest, the possibility of a mild motor fault is the highest. A fault prediction signal is generated and sent to the motor fault diagnosis platform according to the type of fault prediction degree.

[0095] See also Figure 2 As shown in the figure, the motor fault diagnosis method based on big data analysis is as follows:

[0096] Performance status detection: evaluate the motor's operating performance status, collect vibration signals from the motor through a vibration sensor, and classify the motor's operating status into floating state and stable state based on vibration signal analysis;

[0097] Time-based state impact assessment: The motor is assessed for time-based state impact, wherein the corresponding floating periods are marked as floating periods and stable periods according to the floating state and stable state respectively; and the assessment is performed in time periods;

[0098] Fault analysis and processing: When the motor's time-phased status impact assessment is normal, data is collected and analyzed based on the motor's operating parameters.

[0099] Thresholds, preset values, and preset ranges are set for comparative analysis of results to determine whether they are good or bad. The values ​​are set based on a combination of large-scale model analysis of sample data and manual experience, and can also be adjusted appropriately based on seasonal or common-sense factors.

[0100] The settings of weight ratio coefficients, influencing factors, etc. are assigned specific values ​​according to the influence of each parameter on the result, which ultimately reflects the impact on the result. They are also set and entered into storage through a combination of large-scale model analysis of sample data and manual experience. Appropriate adjustments can also be made based on seasonal or common-sense influencing conditions.

[0101] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The motor fault diagnosis system based on big data analysis is characterized by: It includes a motor fault diagnosis platform, where the communication connections of the motor fault diagnosis platform are: The performance status detection unit evaluates and detects the motor's operating performance status, collects vibration signals from the motor through a vibration sensor, and divides the motor's operating status into floating state and stable state based on vibration signal analysis; A time-division state impact assessment unit is used to perform time-division state impact assessment on the motor, wherein corresponding floating periods are marked as floating periods and stable periods according to the floating state and the stable state respectively; and conduct time-phased assessments; The fault analysis and processing unit collects and analyzes the motor's operating parameters when the motor's time-phased status impact assessment is normal.

2. The motor fault diagnosis system based on big data analysis according to claim 1 is characterized in that: The process of the performance status detection unit is as follows: The vibration sensor is used to collect vibration signals from the motor, and the time domain signals are converted into frequency domain signals through fast Fourier transform (FFT). The rolling bearing inside the motor is used as the analysis component for vibration generation, and the frequency of each part of the rolling bearing is analyzed to obtain the frequency of the rolling element passing through the inner ring. Specifically, , where n r is the rotor speed in r / min; d is the rolling element diameter, D is the bearing pitch diameter, α is the contact angle, Z is the number of rolling elements, and fi is the frequency of the rolling elements passing through the inner ring; Collect the frequency fo of the rolling element passing through the outer ring, specifically: ; Collect the cage frequency fc, specifically: ; Collect the rolling body spin frequency fb, specifically: .

3. The motor fault diagnosis system based on big data analysis according to claim 2 is characterized in that: The vibration frequencies generated by the rolling bearings in the motor are summarized, and the frequency fluctuation is monitored as the motor running time increases. If the increase in the vibration amplitude corresponding to any type of vibration frequency exceeds the corresponding set single amplitude threshold, or the cumulative increase in the vibration amplitude of all types of vibration frequencies exceeds the set multiple amplitude thresholds, it is inferred that the running state of the motor is a floating state; if the increase in the vibration amplitude corresponding to any type of vibration frequency does not exceed the corresponding set single amplitude threshold, and the cumulative increase in the vibration amplitude of all types of vibration frequencies does not exceed the set multiple amplitude thresholds, it is inferred that the running state of the motor is a stable state.

4. The motor fault diagnosis system based on big data analysis according to claim 3 is characterized in that: The process of the time-phased status impact assessment unit is as follows: During the floating period, the electrical signal or vibration signal of the motor is monitored in real time, such as the current and the amperage. The signal collected by the real-time monitoring is set as a real-valued signal, and the symbol x(t) is set, and t represents time. The real-valued signal is converted into an analytical signal z(t) through the Hilbert transform as an operation method. According to the real-valued signal x(t) collected in real time, the signal y(t) after Hilbert transformation is obtained by calculation; the calculation formula is: ,in, Expressed as an integral variable; After obtaining y(t), the analytical signal z(t) is composed of the original real-valued signals x(t) and y(t), that is, z(t)=x(t)+jy(t), where j represents the imaginary unit; According to the above signal conversion, the parameters required for fault feature extraction are calculated.

5. The motor fault diagnosis system based on big data analysis according to claim 4 is characterized in that: Calculate the amplitude envelope A(t). The specific calculation formula is: ; Instantaneous phase , the specific calculation formula is: ; The instantaneous frequency f(t) is obtained by taking the derivative of the instantaneous phase and then performing a certain transformation. The specific formula is: ; During the floating period, the amplitude envelope, instantaneous phase, and instantaneous frequency are counted, and the fluctuation of the corresponding values ​​at each moment is recorded. If the corresponding floating span of adjacent moments increases significantly and the significant increase in span persists, it indicates that the current motor state is abnormal. A floating high impact signal is generated and sent to the motor fault diagnosis platform. A significant increase is indicated by the floating span exceeding 1.2 times the set threshold. If the corresponding floating span at adjacent moments does not increase significantly, or the significant increase in span occurs only once, it indicates that the current state of the motor is normal, and a floating low impact signal is generated and sent to the motor fault diagnosis platform.

6. The motor fault diagnosis system based on big data analysis according to claim 5 is characterized in that: During the stable period, the electrical signal or vibration signal of the motor is monitored in real time, and the corresponding performance parameters of the signal are recorded according to the real-time signal. The performance parameters at the time of signal generation are counted at each moment according to the time when the signal is generated and the motor operation intensity of the time period corresponding to the time. When there is no span and no obvious increase, the performance parameters at all moments are counted and the parameter range is constructed; Within the parameter range, the operating scenarios are classified based on the signal generation time and the fluctuation of the motor operating intensity. For example, if the generated signal is a current rising or falling signal, the motor operating intensity increases or decreases; the corresponding operating scenarios are divided into signal aggravation and signal deceleration, and intensity increase and intensity decrease. The operating scenario signal trend is set as the embodiment end of the intensity trend, and the intensity trend is the generation end of the signal trend; when the floating trends of each stage of the embodiment end and the generation end are synchronized, it indicates that there is no deviation in the motor state impact assessment; on the contrary, when the floating trends of each stage of the embodiment end and the generation end are not synchronized, it indicates that the motor state impact assessment is deviated, and the hardware equipment required for motor state detection needs to be re-debugged and the motor operation period needs to be reselected.

7. The motor fault diagnosis system based on big data analysis according to claim 6, characterized in that: When there is no deviation, the trend of parameter alternation at each moment within the parameter range is recorded. While recording the overall trend, the fluctuation of each adjacent moment is also recorded simultaneously. Then the proportional relationship between the intensity parameter and signal type and the corresponding recorded parameter fluctuation trend is determined; If the proportional relationship between the current parameter floating trend and the intensity floating trend in the recorded parameter floating trend is not synchronized, the motor operating state is abnormal, and a job adaptive control abnormality signal is generated and sent to the motor fault diagnosis platform. The motor fault diagnosis platform then debugs the motor's operating performance and changes the operating scenario. If the proportional relationship between the current floating trend of the parameter and the intensity floating trend is synchronized in the recorded parameter floating trend, the motor operation status is normal, and a normal operation adaptive control signal is generated and sent to the motor fault diagnosis platform; If, during a parameter floating trend, the floating trends of adjacent parameters fluctuate but the overall trend is constant, and the current signal type is a favorable trend for the corresponding parameter, a motor status affected signal is still generated and sent to the motor fault diagnosis platform; The motor fault diagnosis platform adjusts the corresponding operating parameters of the motor and records the signal change process; If the current signal type indicates an unfavorable trend in the corresponding parameter, a motor status repair signal is still generated and sent to the motor fault diagnosis platform; The motor fault diagnosis platform suspends the operation of the motor and performs maintenance on the motor hardware involved in the unfavorable trend parameters; If the floating trend of adjacent parameters is reciprocating with the overall trend, the favorable or unfavorable trend of the parameter floating will be reflected by the current signal type, and a corresponding normal signal or abnormal signal will be generated and sent to the motor fault diagnosis platform.

8. The motor fault diagnosis system based on big data analysis according to claim 7 is characterized in that: The process of the fault analysis and processing unit is as follows: According to the historical maintenance process, abnormal operating parameters are set as fault characteristics; The fault factor set is constructed by fault characteristics, and the fault factor set U{u1, u2…, u n }, and construct the comment set V{v1, v2…, v m }, specifically a description set of fault severity or fault possibility; specifically, v1 represents no fault and v2 represents a mild fault; According to the importance of each fault symptom to fault diagnosis, the weight vector A{a1, a2…, a n },and ; After completing the weight vector setting, the fault features corresponding to the fault factor set are judged one by one to obtain their membership to each element in the comment set V and construct the fuzzy evaluation matrix, which is specifically: ; where r ij Representation factor u i Comments v j The membership degree of ; i and j are natural numbers from 1 to n and 1 to m respectively; Get the comprehensive evaluation result vector B, and , It is a fuzzy composite operation; And the comprehensive evaluation result vector B={b1, b2…, b m }, where b j Indicates that the motor belongs to the comment v j The motor fault is determined based on the size of each element in the comprehensive evaluation result vector B. If b2 is the largest, the possibility of a mild motor fault is the greatest. A fault prediction signal is generated and sent to the motor fault diagnosis platform according to the type of fault prediction degree.

9. A motor fault diagnosis method based on big data analysis, characterized in that: A motor fault diagnosis system based on big data analysis as described in any one of claims 1 to 8 above.

Images