A bearing vibration anomaly monitoring method in motor load testing

Abstract

The application discloses a bearing vibration abnormality monitoring method in motor load testing, and relates to the technical field of motor testing, comprising the following steps: obtaining a plurality of structure sampling positions having a mechanical connection relationship with a bearing installation position, determining a structure sampling position corresponding to the bearing installation position as a target bearing sampling position, and forming an incoming position sequence and a bearing outgoing position sequence; taking a target load switching time as a unified response starting point, dividing time sampling windows for vibration response data of each structure sampling position, setting window numbers and extracting window peak values. The application can position and compare vibration peak value changes at the target bearing sampling position in a mechanical connection path by obtaining a plurality of structure sampling positions having a mechanical connection relationship with the bearing installation position and forming the incoming position sequence and the bearing outgoing position sequence, and can investigate vibration responses transmitted in a mechanical connection direction of a load end and a mechanical connection direction of a mounting base.

Description

Technical Field

[0001] This invention relates to the field of motor testing technology, specifically to a method for monitoring abnormal bearing vibration during motor load testing. Background Technology

[0002] During motor load testing, load switching can cause vibration responses at multiple mechanical connection points inside the motor and its mounting base. There are usually mechanical connections between the bearing mounting location, the load end connection structure, the mounting base connection structure, and the structural locations distributed along the bearing's outward transmission direction. Existing vibration monitoring methods can typically collect vibration response data from the bearing mounting location or adjacent structural locations using vibration sensors, and analyze the motor's operating status based on vibration amplitude, frequency components, peak value changes, or time-domain statistical characteristics to help determine the vibration status of the bearing and related structures.

[0003] In actual load switching tests, the peak rebound at the target bearing sampling location does not necessarily originate directly from abnormal vibration of the bearing itself. It may be related to the vibration response transmitted from the mechanical connection direction of the load end or the mechanical connection direction of the mounting base, or it may be related to the vibration response transmitted step by step from the bearing mounting location to the external structure. If the judgment is based solely on the change in the single-point vibration peak value at the target bearing sampling location, it is difficult to distinguish whether the peak rebound is formed by the transmission from the previous structure sampling location or by the target bearing sampling location as the starting point and continuously received from the subsequent structure sampling locations.

[0004] Furthermore, in motor load testing scenarios, it is necessary to combine multiple structural sampling positions that have mechanical connections with the bearing installation position to establish the input position sequence and the bearing external transmission position sequence. The vibration response data is then divided into time sampling windows, numbered, and peak values ​​are extracted, using the target load switching moment as a unified response starting point. Further, by searching the preceding recovery window, judging the subsequent change direction, judging the continuous succession of the following recovery windows, and confirming the stability of the window numbers under multiple rounds of target load switching, the abnormal vibration results of the bearing itself are determined. Therefore, this invention proposes a method for monitoring abnormal bearing vibration in motor load testing. Summary of the Invention

[0005] The purpose of this invention is to provide a method for monitoring abnormal bearing vibration during motor load testing, so as to solve the problems mentioned in the background art.

[0006] This invention can be achieved through the following technical solution: a method for monitoring abnormal bearing vibration during motor load testing, comprising:

[0007] Step 1: Obtain multiple structural sampling positions that have a mechanical connection with the bearing installation position, determine the structural sampling position corresponding to the bearing installation position as the target bearing sampling position, and form the input position sequence and the external bearing position sequence;

[0008] Step 2: Using the target load switching time as the unified response starting point, divide the vibration response data of each structural sampling location into time sampling windows, set window numbers, and extract the peak values ​​of the windows;

[0009] Step 3: Determine the peak recovery sampling window based on the peak difference between adjacent windows that reach the preset recovery difference threshold in the target bearing sampling position; in the structural sampling positions that are earlier than the peak recovery sampling window in the input position sequence, the number difference is within the preset input window number difference range, the peak value of the previous recovery window is higher than the peak recovery sampling window, and the subsequent change direction meets the preset input change consistency condition. If obtained, an input source confirmation result is formed; if not obtained, an input source exclusion result is formed.

[0010] Step 4: After the source of the incoming transmission is eliminated, search for the subsequent recovery window corresponding to at least two consecutive subsequent structural sampling positions in the bearing external transmission position sequence; when the window number of each subsequent recovery window is later than the window number of the peak recovery sampling window, and the window number meets the preset external transmission number difference condition, the window peak value meets the preset external transmission attenuation condition, and the subsequent change direction meets the preset external transmission change consistency condition, the external transmission continuous acceptance result is formed.

[0011] Step 5: In the multi-round target load switching, the target load switching round that simultaneously generates the incoming source exclusion result and the outgoing continuous acceptance result is determined as the valid confirmation round; when the number of valid confirmation rounds reaches the preset number of confirmation rounds, and the window number difference of the peak recovery sampling window corresponding to the target bearing sampling position in each valid confirmation round does not exceed the preset tolerance range, the bearing initial confirmation result is formed, and the bearing's own vibration abnormality result is output.

[0012] A further technical improvement of the present invention lies in: a method for forming the input position order, comprising:

[0013] Using the target bearing sampling position as the input termination position, candidate structure sampling positions pointing to the target bearing sampling position are obtained along the mechanical connection direction of the load end and the mechanical connection direction of the mounting base;

[0014] Within a preset verification time range after the target load switch, obtain the response start time, response amplitude change, and response duration at each candidate structure sampling location;

[0015] Among the adjacent candidate structure sampling positions in the same mechanical connection direction, the candidate structure sampling position that is far away from the target bearing sampling position is determined as the first candidate structure sampling position, and the candidate structure sampling position that is close to the target bearing sampling position is determined as the second candidate structure sampling position.

[0016] And judge based on the following conditions:

[0017] a1: The response start time of the first candidate structure sampling position is earlier than the response start time of the second candidate structure sampling position;

[0018] a2: The change in response amplitude at both the first candidate structure sampling position and the second candidate structure sampling position is not lower than the preset input detectable amplitude threshold;

[0019] a3: The difference in response duration between the sampling positions of the first and second candidate structures is within the preset range of input duration difference;

[0020] When a1, a2, and a3 are satisfied simultaneously, the sampling positions of the first and second candidate structures are included in the input position order.

[0021] When a1, a2, and a3 are not simultaneously satisfied, the sampling position of the first candidate structure is determined as the input continuity interruption position, and the sampling positions of candidate structures located on the side away from the target bearing sampling position at this position are stopped from being included in the input position sequence. The sampling positions of candidate structures included in the input position sequence are arranged according to the mechanical connection sequence from the sampling position away from the target bearing sampling position to the sampling position close to the target bearing sampling position, forming the input position sequence corresponding to the mechanical connection direction of the load end and the mechanical connection direction of the mounting foundation, respectively.

[0022] A further technical improvement of the present invention lies in: a method for forming the bearing external transmission position sequence, comprising:

[0023] Using the target bearing sampling position as the starting point of the outward transmission, multiple candidate outward transmission positions that have a mechanical connection relationship with the target bearing sampling position and are distributed along directions away from the target bearing sampling position are obtained along different outward transmission mechanical connection directions.

[0024] Within each outward mechanical connection direction, obtain the target bearing sampling position and the response start time, response amplitude change, and mechanical connection distance of each outward candidate position;

[0025] Within the same external mechanical connection direction, the target bearing sampling position is determined as the current reference position, and the response start time of the current reference position is determined as the current reference time;

[0026] Among the candidate positions of external transmission that are not included in the bearing external transmission position sequence within the same external transmission mechanical connection direction, find the candidate position of external transmission whose response start time is later than the current reference time, whose response amplitude change is not lower than the lower limit of the detectable amplitude in the external transmission direction, and whose mechanical connection distance is the smallest relative to the current reference position. The found candidate position of external transmission is determined as the current external transmission connection position.

[0027] Incorporate the current external transmission connection position into the bearing external transmission position sequence, update the current external transmission connection position to the current reference position, and update the response start time of the current external transmission connection position to the current reference time;

[0028] The search and update of the current external transmission connection position is repeated within the same external transmission mechanical connection direction until no current external transmission connection position is found among the external transmission candidate positions not included in the bearing external transmission position sequence. The external transmission candidate positions that have been included in the bearing external transmission position sequence are arranged in the order of inclusion to form the bearing external transmission position sequence.

[0029] A further technical improvement of the present invention is that the method for setting the window number in step two includes:

[0030] Using the target load switching time as a unified response starting point, the vibration response data of each structural sampling location are divided into multiple time sampling windows of the same length, and the starting time difference of each time sampling window relative to the target load switching time is recorded.

[0031] Based on the start time difference of each time sampling window, determine the window arrangement position of the corresponding time sampling window, and set the time sampling windows with the same window arrangement position to the same window number;

[0032] For any given window number, count the number of sampling points and the sampling time coverage of each structural sampling location within the time sampling window corresponding to that window number; when the number of sampling points at any structural sampling location is lower than the preset valid sampling number, the window number is determined to be an invalid window number; when the sampling time coverage of any structural sampling location does not reach the preset window coverage, the window number is determined to be an invalid window number.

[0033] Exclude time sampling windows corresponding to invalid window numbers, determine the time sampling windows corresponding to the remaining window numbers as the time sampling windows participating in window peak extraction, and retain the original numbering order of the remaining window numbers relative to the target load switching time.

[0034] A further technical improvement of the present invention is that the method for determining the peak recovery sampling window in step three includes:

[0035] In each time sampling window corresponding to the target bearing sampling position, the difference between the adjacent window peak values ​​between the window peak value of the next time sampling window and the window peak value of the previous time sampling window is calculated in ascending order of window number.

[0036] The sampling window after which the peak difference between adjacent windows reaches the preset recovery difference threshold is determined as the recovery window to be verified, and the peak difference between the adjacent windows is determined as the benchmark recovery difference.

[0037] Starting with the recovery window to be verified, multiple subsequent time sampling windows are continuously extracted along the direction of increasing window number, and the peak difference between adjacent subsequent time sampling windows is calculated respectively.

[0038] Calculate the ratio between the peak difference of each subsequent window and the baseline recovery difference to obtain multiple difference ratios, and arrange the multiple difference ratios in ascending order according to the window number of the corresponding subsequent time sampling window;

[0039] Compare the difference ratios at the top and bottom of the list, and check for adjacent ratio pairs where the latter difference ratio is greater than the former difference ratio among multiple difference ratios.

[0040] When the difference ratio at the end of the list is less than the difference ratio at the beginning of the list, there are no adjacent ratio pairs among multiple difference ratios where the next difference ratio is greater than the previous difference ratio, and the peak difference of each subsequent window does not exceed the baseline recovery difference, the recovery window to be verified is determined as the peak recovery sampling window.

[0041] When the difference ratio at the end of the list is not less than the difference ratio at the beginning of the list, the recovery window to be verified is not determined as the peak recovery sampling window; when there are two consecutive difference ratios that increase sequentially among multiple difference ratios, the recovery window to be verified is not determined as the peak recovery sampling window; when there is a subsequent window peak difference that exceeds the baseline recovery difference, the recovery window to be verified is not determined as the peak recovery sampling window.

[0042] A further technical improvement of the present invention is that the method for determining in step three that the subsequent changing directions of the preceding recovery window and the peak recovery sampling window satisfy a preset condition for consistent input changes includes:

[0043] In the sequence of each input position where the previous position rise window has been obtained, the corresponding previous position rise window is extracted respectively, and the first time sampling window after the previous position rise window is taken as the starting point, and multiple subsequent input time sampling windows are extracted in the direction of increasing window number.

[0044] In the target bearing sampling location, starting with the first time sampling window after the peak recovery sampling window, extract the same number of target subsequent time sampling windows as the number of input subsequent time sampling windows in the direction of increasing window number.

[0045] Calculate the peak difference of the input window between adjacent subsequent input time sampling windows, and calculate the peak difference of the target window between adjacent subsequent target time sampling windows;

[0046] According to the window number correspondence, calculate the absolute value of the difference between the peak difference of the input window corresponding to each input position sequence and the peak difference of the target window, and determine the absolute value of the difference as the difference deviation.

[0047] The deviations of each difference in the same input position sequence are accumulated to obtain the cumulative value of the progressive deviation of the corresponding input position sequence;

[0048] The input position sequence with the smallest cumulative deviation value is determined as the target input position sequence. When the cumulative deviation value corresponding to the target input position sequence does not exceed the preset cumulative deviation threshold, the preceding recovery window and the peak recovery sampling window in the target input position sequence are determined as the subsequent change direction to meet the preset input change consistency condition.

[0049] A further technical improvement of the present invention is that the method for forming the outward transmission continuous reception result in step four includes:

[0050] After the incoming source exclusion result is formed, at least two consecutive rear structure sampling positions are obtained along the bearing external transmission position, and the rear recovery window corresponding to each rear structure sampling position is determined respectively.

[0051] Based on the hierarchical succession relationship formed by each subsequent recovery window along the bearing's external transmission position, an external transmission number succession relationship is formed;

[0052] Based on the changes in the peak value of each subsequent recovery window along the bearing's outward transmission position, an outward transmission peak value succession relationship is formed;

[0053] Based on the peak value change process after each subsequent recovery window, an external transmission and transfer relationship is formed;

[0054] When external transmission numbering succession relationship, external transmission peak succession relationship, and external transmission variation succession relationship are formed, an external transmission continuous succession result is formed.

[0055] A further technical improvement of the present invention lies in: a method for forming the sequence of external transmission numbers, comprising:

[0056] Among a series of consecutive rear structure sampling positions, the rear structure sampling position adjacent to the target bearing sampling position is determined as the first rear structure sampling position.

[0057] In the time sampling window corresponding to the first subsequent structure sampling position, the difference between the peak value of the next time sampling window and the peak value of the previous time sampling window is calculated in ascending order of window number. The difference is determined as the peak value difference between adjacent windows corresponding to the first subsequent structure sampling position. In the time sampling windows where the window number is later than the peak recovery sampling window, the next time sampling window where the peak value difference between adjacent windows reaches the preset recovery difference threshold is found, and the next time sampling window with the smallest window number is determined as the first subsequent recovery window.

[0058] The next subsequent structural sampling position is selected sequentially along the bearing external transmission position, and the previous subsequent recovery window is used as the current search reference window.

[0059] Within the time sampling window corresponding to the current subsequent structure sampling position, calculate the difference between the peak value of the next time sampling window and the peak value of the previous time sampling window in ascending order of window number. Determine the difference as the peak value difference between adjacent windows corresponding to the current subsequent structure sampling position. Within the time sampling windows whose window number is later than the current reference window, find the next time sampling window whose peak value difference between adjacent windows reaches the preset recovery difference threshold, and determine the next time sampling window with the smallest window number as the subsequent recovery window corresponding to the current subsequent structure sampling position.

[0060] Calculate the difference between the window number of the next subsequent rise window and the window number of the previous subsequent rise window in two adjacent subsequent rise windows to obtain the difference between adjacent outgoing numbers; when the differences between adjacent outgoing numbers are all positive and the differences between adjacent outgoing numbers meet the preset outgoing number difference conditions, an outgoing number succession relationship is formed.

[0061] A further technical improvement of the present invention is that the method for forming the bearing initial confirmation result in step five includes:

[0062] In the multi-round target load switching, the peak recovery sampling window, the incoming source exclusion result and the outgoing continuous acceptance result corresponding to each round of target load switching are extracted respectively.

[0063] The target load switching rounds that have formed the results of excluding incoming sources and the results of continuous outgoing transmission are determined as valid confirmation rounds;

[0064] Extract the window number of the peak recovery sampling window corresponding to the target bearing sampling position in each valid confirmation round, and arrange them according to the order of target load switching rounds to form the recovery window number arrangement result;

[0065] Subtract the minimum window number from the maximum window number in the result of sorting the rising window numbers to obtain the maximum number difference;

[0066] When the number of valid confirmation rounds reaches the preset number of confirmation rounds and the maximum number difference does not exceed the preset tolerance range, the bearing initial confirmation result is formed.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] This invention acquires multiple structural sampling positions that are mechanically connected to the bearing installation location, and forms an incoming position sequence and a bearing outgoing position sequence. This allows the vibration peak change at the target bearing sampling position to be located and compared within the mechanical connection path. By finding the preceding peak rise sampling window earlier than the incoming position sequence, and combining the number difference, window peak value, and preset incoming change consistency conditions for judgment, the vibration response transmitted in the load end mechanical connection direction and the installation foundation mechanical connection direction can be investigated, thereby forming an incoming source confirmation result or an incoming source exclusion result.

[0069] Furthermore, after the source of the incoming transmission is eliminated, the system further searches for the subsequent recovery window corresponding to at least two consecutive subsequent structural sampling positions in the bearing transmission position sequence. The system then uses preset transmission number difference conditions, preset transmission attenuation conditions, and preset transmission variation consistency conditions to form a continuous transmission result. As a result, the peak recovery sampling window of the target bearing sampling position not only corresponds to the time continuity of the subsequent structural sampling positions, but also corresponds to the change of the window peak value along the bearing transmission position sequence and the subsequent change direction, thus improving the reliability of the bearing installation position as the vibration initiation position.

[0070] On the other hand, by jointly confirming the peak rebound sampling window, the incoming source exclusion result and the outgoing continuous acceptance result in the multi-round target load switching, and further comparing the window number difference of the peak rebound sampling window corresponding to the sampling position of the target bearing in each round, the bearing initial confirmation result is formed and the bearing vibration abnormality result is output when the window number difference does not exceed the preset tolerance range. This can be combined with the response stability under multi-round target load switching to confirm, thereby improving the consistency of bearing vibration abnormality monitoring results in motor load testing. Attached Figure Description

[0071] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0072] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0073] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0074] Please see Figure 1 As shown, the present invention provides a method for monitoring abnormal bearing vibration during motor load testing, comprising:

[0075] Step 1: Obtain multiple structural sampling positions that have a mechanical connection with the bearing installation position, determine the structural sampling position corresponding to the bearing installation position as the target bearing sampling position, and form the input position sequence and the external bearing position sequence;

[0076] Specifically, first, obtain multiple structural sampling locations that have a mechanical connection with the bearing installation position, and then determine the target bearing sampling location, including:

[0077] In this embodiment, based on the mechanical assembly relationship of the motor load test object, multiple mechanical connection positions with the bearing mounting position are first determined, which may be rigid, elastic, or contact connections. Vibration sampling points are then set at each mechanical connection position. Each vibration sampling point capable of collecting vibration response data is designated as a structural sampling position. For each structural sampling position, its corresponding mechanical connection position, its mechanical connection direction, the connection path between it and the bearing mounting position, and its arrangement order within that connection path are recorded. Subsequently, the structural sampling position corresponding to the bearing mounting position is designated as the target bearing sampling position. For the remaining structural sampling positions with mechanical connection relationships to the target bearing sampling position, structural sampling positions pointing towards the target bearing sampling position are extracted along the mechanical connection direction at the load end and the mechanical connection direction of the mounting base, and these extracted structural sampling positions are designated as candidate structural sampling positions. Furthermore, structural sampling positions distributed from the target bearing sampling position away from it are extracted along different outward mechanical connection directions, and these extracted structural sampling positions are designated as outward candidate positions.

[0078] The baseline data used to calculate the response start time, the change in response amplitude, and the response duration includes:

[0079] Before switching the target load, a preset reference sampling time period is set, and the vibration response amplitude of each structural sampling location is collected within the preset reference sampling time period. For any structural sampling location, the average vibration response amplitude of that sampling location within the preset reference sampling time period is calculated, and this average vibration response amplitude is used as the reference vibration response amplitude of that sampling location. Subsequently, the absolute difference between the vibration response amplitude of each sampling point within the preset reference sampling time period and the reference vibration response amplitude is calculated, and the maximum value of the absolute difference is determined as the reference fluctuation upper limit value of that structural sampling location. Within a preset verification time range after the target load switch, the absolute difference between the vibration response amplitude of each sampling point and the reference vibration response amplitude is compared point by point along the sampling time sequence. The sampling time corresponding to the sampling point that first exceeds the upper limit of the reference fluctuation is determined as the response start time. The vibration response amplitude of each sampling point within the preset verification time range is extracted, and the absolute difference between the vibration response amplitude of each sampling point and the reference vibration response amplitude is calculated. The maximum value of the absolute difference is determined as the change in response amplitude. Starting from the sampling point corresponding to the response start time, the number of sampling points where the absolute difference between the vibration response amplitude and the reference vibration response amplitude is continuously greater than the upper limit of the reference fluctuation is counted. The number of sampling points is multiplied by the sampling time interval between adjacent sampling points to obtain the response duration.

[0080] To determine the order of incoming positions, including:

[0081] When forming the input position sequence, the target bearing sampling position is used as the input termination position. Candidate structure sampling positions pointing to the target bearing sampling position are processed along both the load-end mechanical connection direction and the mounting base mechanical connection direction. In the same mechanical connection direction, adjacent candidate structure sampling positions are selected sequentially according to the mechanical connection order from furthest from the target bearing sampling position to closest to it. The candidate structure sampling position furthest from the target bearing sampling position is determined as the first candidate structure sampling position, and the candidate structure sampling position closest to the target bearing sampling position is determined as the second candidate structure sampling position. Subsequently, the response start time, response amplitude change, and response duration of the first and second candidate structure sampling positions are compared: when the response start time of the first candidate structure sampling position is earlier than that of the second candidate structure sampling position, and the response amplitude change of both the first and second candidate structure sampling positions is not lower than a preset input detectable amplitude threshold, and the difference in response duration between the first and second candidate structure sampling positions is within a preset input duration difference range, the first and second candidate structure sampling positions are included in the input position sequence. If not all three judgment results are met, the sampling position of the first candidate structure is determined as the input continuity interruption position, and the inclusion of candidate structure sampling positions located on the side of the input continuity interruption position away from the target bearing sampling position in the input position sequence is stopped. After completing the judgment of each adjacent candidate structure sampling position within the same mechanical connection direction, the candidate structure sampling positions included in the input position sequence are arranged according to the mechanical connection sequence from the sampling position away from the target bearing sampling position to the sampling position closer to the target bearing sampling position, forming the input position sequence for the corresponding load end mechanical connection direction and the mounting foundation mechanical connection direction, respectively.

[0082] The sequence of bearing external position transmission includes:

[0083] When determining the bearing external transmission position sequence, the target bearing sampling position is used as the external transmission starting point. Candidate external transmission positions distributed along directions away from the target bearing sampling position are processed along different external transmission mechanical connection directions. Within each external transmission mechanical connection direction, the length of the connection segment between two adjacent mechanical connection positions is recorded sequentially according to the arrangement order of the mechanical connection positions corresponding to the structural sampling positions. For the current reference position and any external transmission candidate position, the lengths of all connection segments traversed between the current reference position and the candidate position are added together to obtain the mechanical connection distance of the candidate position relative to the current reference position. Subsequently, within the same external transmission mechanical connection direction, the target bearing sampling position is determined as the current reference position, and the response start time of the target bearing sampling position is determined as the current reference time. Within the same external transmission mechanical connection direction, among the candidate external transmission positions not yet included in the bearing external transmission position sequence, first, screen for candidate external transmission positions whose response start time is later than the current reference time and whose response amplitude change is not lower than the lower limit of the detectable amplitude in the external transmission direction. Among the screened candidate external transmission positions, compare the mechanical connection distances of each candidate external transmission position relative to the current reference position, and determine the candidate external transmission position with the smallest mechanical connection distance as the current external transmission connection position. When multiple candidate external transmission positions have the same mechanical connection distance, determine the candidate external transmission position with the earliest response start time as the current external transmission connection position. After including the current external transmission connection position in the bearing external transmission position sequence, update the current reference position to the current external transmission connection position, and update the current reference time to the response start time of the current external transmission connection position. Continue performing the above search, inclusion, and update actions within the same external transmission mechanical connection direction until no candidate external transmission position meeting the above screening conditions is found among those not included in the bearing external transmission position sequence. Finally, arrange the candidate external transmission positions already included in the bearing external transmission position sequence according to the order of inclusion to form the bearing external transmission position sequence corresponding to the external transmission mechanical connection direction.

[0084] The system establishes a preset threshold for detectable input amplitude, a preset range for input duration difference, and a lower limit for detectable amplitude in the outward transmission direction. It also provides the input position sequence and bearing outward transmission position sequence for subsequent steps, including:

[0085] In both the load-side mechanical connection direction and the installation foundation mechanical connection direction, the upper limit value of the reference fluctuation corresponding to each candidate structure sampling position is extracted. These upper limit values ​​are then compared, and the maximum value is taken as the preset input detectable amplitude threshold. During multi-round reference load switching tests, the response duration difference between adjacent candidate structure sampling positions in the same mechanical connection direction is calculated according to the response duration calculation method in this embodiment. The minimum value among these response duration differences is determined as the lower limit of the preset input duration difference range, and the maximum value is determined as the upper limit of the preset input duration difference range. The lower and upper limits form the preset input duration difference range. Within each outward transmission mechanical connection direction, the upper limit value of the reference fluctuation corresponding to each outward transmission candidate position is extracted. These upper limit values ​​are then compared, and the maximum value is taken as the lower limit of the detectable amplitude in the outward transmission direction. A preset input detectable amplitude threshold is formed within the mechanical connection direction at the load end and the mechanical connection direction at the mounting base, and is used to form the input position sequence. An external transmission direction detectable amplitude lower limit is formed within each external transmission mechanical connection direction and is used to form the bearing external transmission position sequence. After completing the above processing, subsequent steps call the structural sampling position located before the target bearing sampling position in the input position sequence to find the preceding recovery window; and call the subsequent structural sampling position located after the target bearing sampling position in the bearing external transmission position sequence to find the subsequent recovery window.

[0086] Step 2: Using the target load switching time as the unified response starting point, divide the vibration response data of each structural sampling location into time sampling windows, set window numbers, and extract the peak values ​​of the windows;

[0087] Specifically, the target load switching time is used as the unified response starting point, and a time sampling window is divided, including:

[0088] In this embodiment, after determining the sampling locations of each structure, the vibration response data collected at each sampling location is time-aligned. The sampling time at which the target load switching begins is determined as the target load switching time, and this target load switching time is used as the unified response starting point for each structural sampling location. For any structural sampling location, starting from the unified response starting point, the vibration response data at that location are sequentially divided according to the same time length to obtain multiple time sampling windows. Each time sampling window records its window start sampling time, window end sampling time, number of sampling points within the window, and vibration response amplitude corresponding to each sampling point within the window. Subsequently, the time difference between the window start sampling time and the target load switching time for each time sampling window is calculated, and this time difference is determined as the starting time difference of the corresponding time sampling window relative to the target load switching time. Thus, the vibration response data at each structural sampling location are divided into windows with the same target load switching time as the time zero point, avoiding mismatched window numbers due to different sampling starting points for different structural sampling locations.

[0089] Window numbers are set based on the start time difference, including:

[0090] After obtaining the starting time difference for each time sampling window, the starting time difference is divided by the time length of the time sampling window to obtain the quotient value, which is then rounded down to determine the window arrangement position. Time sampling windows with the same window arrangement position across different structural sampling locations are assigned the same window number; for time sampling windows with different window arrangement positions, different window numbers are assigned according to the ascending order of their window arrangement positions. Each window number corresponds to a defined time interval after the target load switching time, and time sampling windows with the same window number across different structural sampling locations correspond to the same time interval.

[0091] Identify invalid window numbers and preserve the original numbering order, including:

[0092] After setting the window numbers, for each window number, the number of sampling points at each structural sampling location within the corresponding time sampling window is counted, and the sampling time coverage of that time sampling window is calculated. The number of sampling points is obtained by directly counting the actual sampling points collected within the time interval corresponding to that window number; the sampling time coverage is obtained by subtracting the sampling time of the first sampling point from the sampling time of the last sampling point within the time sampling window corresponding to that window number. The initial number of sampling points corresponding to that time sampling window is obtained by multiplying the time length of the time sampling window by the sampling frequency of the vibration response data. When the initial number of sampling points is not an integer, it is rounded down, and the rounded value is determined as the preset effective sampling number; when the initial number of sampling points is an integer, it is determined as the preset effective sampling number. The time interval between adjacent sampling points in the vibration response data is determined as the sampling time interval, and the time range obtained by subtracting the sampling time interval from the time length of the time sampling window is determined as the preset window coverage. When the number of sampling points at any structural sampling location within the time sampling window corresponding to that window number is less than the preset valid sampling number, that window number is determined as an invalid window number. Similarly, when the sampling time coverage of any structural sampling location within the time sampling window corresponding to that window number does not reach the preset window coverage range, that window number is also determined as an invalid window number. After determining the invalid window numbers, the time sampling windows corresponding to the invalid window numbers are excluded, and the time sampling windows corresponding to the remaining window numbers are determined as the time sampling windows participating in window peak extraction. The original numbering order of the remaining window numbers relative to the target load switching time is retained, and the retained window numbers are not renumbered. Therefore, even if there are excluded invalid window numbers in the middle, the difference in subsequent window numbers still reflects the true time interval between each time sampling window and the target load switching time.

[0093] The peak value of the window is extracted based on the time sampling window involved in the peak value extraction, including:

[0094] After determining the time sampling window for peak extraction, for any structural sampling location and any reserved window number, the vibration response amplitude of each sampling point within the time sampling window corresponding to that window number is read, and the maximum value is selected from the vibration response amplitudes. This maximum value is determined as the peak value of that structural sampling location under that window number. For each structural sampling location, the peak values ​​of each window are recorded in ascending order of the reserved window numbers. This allows subsequent steps to calculate the peak value difference between adjacent windows based on the peak values ​​corresponding to adjacent window numbers, and to determine the incoming source confirmation result and the outgoing continuous reception result based on the window number difference between different structural sampling locations.

[0095] Step 3: Determine the peak recovery sampling window based on the peak difference between adjacent windows that reach the preset recovery difference threshold in the target bearing sampling position; in the structural sampling positions that are earlier than the peak recovery sampling window in the input position sequence, the number difference is within the preset input window number difference range, the peak value of the previous recovery window is higher than the peak recovery sampling window, and the subsequent change direction meets the preset input change consistency condition. If obtained, an input source confirmation result is formed; if not obtained, an input source exclusion result is formed.

[0096] Specifically, in this embodiment, based on the time sampling windows, window numbers, and window peak values ​​already established in the previous embodiments, within each time sampling window corresponding to the target bearing sampling position, the preceding and following time sampling windows are selected sequentially in ascending order of window number. The difference between the peak value of the following time sampling window and the peak value of the preceding time sampling window is calculated, and this difference is determined as the peak value difference between adjacent windows. The peak value differences between each adjacent window are sequentially compared with a preset recovery difference threshold. When the peak value difference of a certain adjacent window reaches the preset recovery difference threshold, the following time sampling window that generated the peak value difference is determined as the recovery window to be verified, and the peak value difference between the adjacent windows is determined as the baseline recovery difference. Since the preset recovery difference threshold is a value greater than zero, the baseline recovery difference is a value greater than zero.

[0097] Before determining the recovery window to be verified, the number of subsequent time sampling windows is formed based on multiple rounds of baseline load switching tests. After determining the recovery window to be verified, starting from the recovery window to be verified, multiple subsequent time sampling windows corresponding to the number of subsequent time sampling windows are continuously extracted in the direction of increasing window number. The peak difference between adjacent subsequent time sampling windows is calculated in ascending order of window number. The ratio between the peak difference of each subsequent window and the baseline recovery difference is calculated to obtain multiple difference ratios, which are then arranged in ascending order of the window number of the corresponding subsequent time sampling window. The difference ratio at the beginning and the difference ratio at the end are compared, and the preceding and following difference ratios of two adjacent difference ratios are compared one by one to determine whether there are adjacent ratio pairs where the following difference ratio is greater than the preceding difference ratio. When the difference ratio at the end of the list is less than the difference ratio at the beginning of the list, there are no adjacent ratio pairs among multiple difference ratios where the next difference ratio is greater than the previous difference ratio, and the peak difference of each subsequent window does not exceed the baseline recovery difference, the recovery window to be verified is determined as the peak recovery sampling window.

[0098] After determining the peak recovery sampling window, the input position sequence already formed in the aforementioned embodiments is invoked, and the structural sampling position located before the target bearing sampling position is selected from the input position sequence as the sampling position to be searched. For any sampling position to be searched, its corresponding time sampling windows, window numbers, and window peak values ​​are read, and the difference obtained by subtracting the window peak value of the previous time sampling window from the window peak value of the next time sampling window in ascending order of window number is calculated. This difference is determined as the peak difference between adjacent windows corresponding to the sampling position to be searched. In the time sampling window corresponding to the sampling position to be searched, the search window number is earlier than the time sampling window of the peak recovery sampling window; when the peak difference between adjacent windows reaches a preset recovery difference threshold, the time sampling window that generated the peak difference between adjacent windows is determined as the candidate preceding recovery window. Calculate the difference between the window number of the peak recovery sampling window and the window number of the candidate preceding recovery window, and determine this difference as the number difference. When the candidate preceding recovery window is earlier than the peak recovery sampling window, the number difference is within the range of the preset input window number difference, and the window peak value of the candidate preceding recovery window is higher than the window peak value of the peak recovery sampling window, the candidate preceding recovery window is determined as the preceding recovery window.

[0099] In each input position sequence where the preceding recovery window has been obtained, the corresponding preceding recovery window is extracted. Starting with the first time sampling window after the preceding recovery window, multiple input subsequent time sampling windows corresponding to the number of subsequent time sampling windows are extracted along the direction of increasing window number. In the target bearing sampling position, starting with the first time sampling window after the peak recovery sampling window, the same number of target subsequent time sampling windows are extracted along the direction of increasing window number. The peak difference between adjacent input subsequent time sampling windows is calculated according to the ascending window number, and the peak difference between adjacent target subsequent time sampling windows is also calculated. For each relative arrangement position, the peak difference between the input window and the peak difference between the target window are checked for directional consistency: when the peak difference between the input window and the target window is greater than zero, the peak difference between the target window and the target window is greater than zero; when the peak difference between the input window and the target window is less than zero, the peak difference between the input window and the target window is equal to zero; when all directional consistency checks are valid, the difference deviation and cumulative deviation are calculated for the input position sequence; when the directional consistency check is invalid for any relative arrangement position, the input position sequence is determined as an input position sequence that does not meet the preset input progressive consistency condition, and the difference deviation and cumulative deviation are stopped for the input position sequence.

[0100] When all directional consistency judgments are valid, the peak difference of the input window under the same relative arrangement position is paired with the peak difference of the target window according to the window number correspondence. The absolute value of the difference in each pairing result is calculated, and this absolute value is determined as the difference deviation. The difference deviations in the same input position sequence are accumulated to obtain the cumulative value of the progressive deviation for the corresponding input position sequence. In the input position sequences where all directional consistency judgments are valid and progressive deviation cumulative values ​​are formed, if multiple input position sequences form progressive deviation cumulative values, the input position sequence with the smallest progressive deviation cumulative value is determined as the target input position sequence; if only one input position sequence forms a progressive deviation cumulative value, that input position sequence is determined as the target input position sequence; if none of the input position sequences form a progressive deviation cumulative value, it is determined that no previous rising window that meets the preset input progressive consistency condition has been obtained. When the progressive deviation cumulative value corresponding to the target input position sequence does not exceed the preset input progressive deviation cumulative threshold, the previous rising window and the peak rising sampling window in the target input position sequence are determined as subsequent progressive directions that meet the preset input progressive consistency condition. Among them, the preset input variation consistency condition is used to characterize that the variation direction of the window peak after the previous rise window corresponds to the variation direction of the window peak after the peak rise sampling window, and the variation amplitude deviation between the corresponding window peak differences is limited.

[0101] After completing the lookup of the preceding recovery window and the judgment of the preset input change consistency condition, the results are summarized for the input position sequence corresponding to the load-side mechanical connection direction and the installation foundation mechanical connection direction. When a preceding recovery window is obtained in any input position sequence, and this preceding recovery window is earlier than the peak recovery sampling window, the number difference is within the preset input window number difference range, the window peak value of the preceding recovery window is higher than the window peak value of the peak recovery sampling window, and the subsequent change direction of this preceding recovery window and the peak recovery sampling window meets the preset input change consistency condition, an input source confirmation result is formed; when no preceding recovery window that simultaneously meets the above conditions is obtained in the input position sequence corresponding to the load-side mechanical connection direction and the installation foundation mechanical connection direction, an input source exclusion result is formed.

[0102] In this embodiment, the preset recovery difference threshold is formed through multiple rounds of benchmark load switching tests: the time sampling window and window peak value of the target bearing sampling position in each round of benchmark load switching test are extracted, and the difference obtained by subtracting the window peak value of the previous time sampling window from the window peak value of the next time sampling window in ascending order of window number is calculated to obtain multiple benchmark adjacent window peak value differences; the benchmark adjacent window peak value differences with a value greater than zero are retained and arranged in ascending order of value, and the benchmark adjacent window peak value difference ranked second is selected as the preset recovery difference threshold. This process is used to exclude the small recovery corresponding to the single round minimum positive fluctuation at the top of the list; when the number of retained benchmark adjacent window peak value differences is one, the benchmark adjacent window peak value difference is used as the preset recovery difference threshold. The preset input window number difference range is formed through multiple rounds of benchmark load switching tests: In the input position sequence, the structural sampling position located before the target bearing sampling position and whose response start time is earlier than the target bearing sampling position is selected as the benchmark preceding structural sampling position. The preceding recovery window corresponding to the benchmark preceding structural sampling position and the peak recovery sampling window corresponding to the target bearing sampling position are determined respectively. The difference obtained by subtracting the window number of the preceding recovery window from the window number of the peak recovery sampling window is calculated to obtain multiple benchmark input window number differences. The minimum value among the multiple benchmark input window number differences is determined as the lower limit of the preset input window number difference range, and the maximum value is determined as the upper limit.

[0103] The number of subsequent time sampling windows is determined through multiple rounds of benchmark load switching tests: starting from the determined peak recovery sampling window, the peak difference between adjacent time sampling windows is calculated one by one in the direction of increasing window number; the upper limit of benchmark fluctuation is the maximum absolute difference between the vibration response amplitude of each sampling point at the corresponding structural sampling position and the benchmark vibration response amplitude within the preset benchmark sampling time period; when the absolute value of the peak difference of three consecutive windows does not exceed the upper limit of benchmark fluctuation, the first time sampling window in the three windows is determined as the stable fluctuation start window, and the difference obtained by subtracting the window number of the peak recovery sampling window from the window number of the stable fluctuation start window is calculated to obtain the number of subsequent windows in a single round; the number of subsequent windows in a single round obtained from multiple rounds of benchmark load switching tests is arranged in ascending order of value, and the number of subsequent windows in the first round is taken as the number of subsequent time sampling windows, so that the subsequent time sampling windows, the input subsequent time sampling windows, and the target subsequent time sampling windows are all within the effective attenuation range of the corresponding round. The preset input variation deviation cumulative threshold is formed through multiple rounds of reference load switching tests: According to the judgment method of the preset input variation consistency condition in this embodiment, the cumulative variation deviation between the input position sequence and the target bearing sampling position in each round of reference load switching test is calculated to obtain multiple reference variation deviation cumulative values; the multiple reference variation deviation cumulative values ​​are arranged in ascending order of value, and the last reference variation deviation cumulative value in the arrangement is selected as the preset input variation deviation cumulative threshold.

[0104] Step 4: After the source of the incoming transmission is eliminated, search for the subsequent recovery window corresponding to at least two consecutive subsequent structural sampling positions in the bearing external transmission position sequence; when the window number of each subsequent recovery window is later than the window number of the peak recovery sampling window, and the window number meets the preset external transmission number difference condition, the window peak value meets the preset external transmission attenuation condition, and the subsequent change direction meets the preset external transmission change consistency condition, the external transmission continuous acceptance result is formed.

[0105] In this embodiment, after the source exclusion result is formed, the bearing outgoing position sequence, time sampling window, window number, and window peak value already formed in the previous embodiment are invoked. To avoid structural sampling positions from different outgoing mechanical connection directions being mixed in the same acceptance judgment process, subsequent subsequent structural sampling positions are extracted, subsequent recovery windows are searched, and outgoing continuous acceptance results are judged within the outgoing mechanical connection direction corresponding to the same bearing outgoing position sequence. Along the bearing outgoing position sequence, starting from the first structural sampling position after the target bearing sampling position, subsequent structural sampling positions are extracted sequentially in the arrangement direction away from the target bearing sampling position; when the number of extracted subsequent structural sampling positions reaches at least two, these at least two subsequent structural sampling positions are used as outgoing continuous acceptance judgment objects. If the number of subsequent structural sampling positions after the target bearing sampling position in the bearing outgoing position sequence is less than two, no outgoing continuous acceptance result is formed.

[0106] Among at least two consecutive subsequent structure sampling positions, the subsequent structure sampling position adjacent to the target bearing sampling position is determined as the first subsequent structure sampling position. Within the time sampling window corresponding to the first subsequent structure sampling position, the difference between the peak value of the next time sampling window and the peak value of the previous time sampling window is calculated in ascending order of window number. This difference is determined as the adjacent window peak value difference corresponding to the first subsequent structure sampling position. In time sampling windows where the window number is later than the peak recovery sampling window, a subsequent time sampling window whose adjacent window peak value difference reaches a preset recovery difference threshold is searched. When multiple subsequent time sampling windows meeting the conditions are found, the subsequent time sampling window with the smallest window number is determined as the first subsequent recovery window.

[0107] After determining the first subsequent recovery window, the next subsequent structural sampling position is selected sequentially along the bearing's outward transmission position, with the subsequent recovery window corresponding to the previous subsequent structural sampling position serving as the current search reference window. Within the time sampling window corresponding to the current subsequent structural sampling position, the difference between the peak value of the next time sampling window and the peak value of the previous time sampling window is calculated in ascending order of window number. This difference is determined as the peak value difference between adjacent windows corresponding to the current subsequent structural sampling position. In time sampling windows with window numbers later than the current search reference window, the next time sampling window whose peak value difference between adjacent windows reaches a preset recovery difference threshold is searched. When multiple subsequent time sampling windows meeting the conditions are found, the next time sampling window with the smallest window number is determined as the subsequent recovery window corresponding to the current subsequent structural sampling position. This process is repeated sequentially along the bearing's outward transmission position until at least two consecutive subsequent structural sampling positions have obtained corresponding subsequent recovery windows, or until a certain subsequent structural sampling position has not obtained a corresponding subsequent recovery window.

[0108] In this embodiment, the preset external transmission number difference condition is formed through multiple rounds of reference load switching tests. Specifically, within each external transmission mechanical connection direction, multiple rounds of reference load switching tests are repeatedly performed. In each round of reference load switching tests, the rear-position rise window corresponding to at least two consecutive rear-position structural sampling positions is obtained according to the rear-position rise window search action of this embodiment. For two adjacent rear-position rise windows, the difference obtained by subtracting the window number of the previous rear-position rise window from the window number of the next rear-position rise window is calculated to obtain multiple reference adjacent external transmission number differences. The multiple reference adjacent external transmission number differences are arranged in ascending order of value. The reference adjacent external transmission number difference at the top of the arrangement is determined as the lower limit of the preset external transmission number difference condition, and the reference adjacent external transmission number difference at the bottom of the arrangement is determined as the upper limit of the preset external transmission number difference condition. When the adjacent external transmission number difference is not less than the lower limit and not greater than the upper limit, it is determined that the adjacent external transmission number difference meets the preset external transmission number difference condition.

[0109] After obtaining the corresponding rear-position recovery window at each subsequent structural sampling location, the rear-position recovery windows are arranged according to the bearing outward transmission position order, and the window number of each rear-position recovery window is read sequentially. The window numbers of two adjacent rear-position recovery windows are compared one by one, and the difference between the window number of the subsequent rear-position recovery window and the window number of the preceding rear-position recovery window is calculated. This difference is determined as the adjacent outward transmission number difference. When the adjacent outward transmission number differences are all positive and meet the preset outward transmission number difference conditions, an outward transmission number succession relationship is formed. This outward transmission number succession relationship is formed by the progressively updated current lookup reference window, not by comparing each subsequent structural sampling location independently with the peak recovery sampling window.

[0110] After establishing the transmission number succession relationship, the peak values ​​of each subsequent recovery window are extracted and arranged according to the bearing transmission position order. The peak values ​​of two adjacent subsequent recovery windows are compared one by one, and the difference between the peak value of the preceding and following subsequent recovery windows is calculated. This difference is determined as the adjacent transmission peak value difference. When all adjacent transmission peak value differences are non-negative, and the peak value of each subsequent recovery window is not higher than the peak value of the peak recovery sampling window, an transmission peak succession relationship is established. In this embodiment, the preset transmission attenuation condition adopts the attenuation criterion that the peak value of the window does not increase along the bearing transmission position sequence, to characterize that the peak value of the subsequent recovery window does not increase along the direction away from the target bearing sampling position.

[0111] After establishing the outward peak transmission relationship, starting with the first time sampling window after each subsequent rise window, the same number of subsequent time sampling windows are extracted along the direction of increasing window number. The number of subsequent time sampling windows adopts the number of subsequent time sampling windows formed in the aforementioned embodiment, ensuring that the subsequent time sampling windows, the incoming subsequent time sampling windows, and the target subsequent time sampling windows use the same number of parameters. For each subsequent structural sampling position, the peak difference between adjacent subsequent time sampling windows is calculated in ascending order of window number.

[0112] After obtaining the peak difference of the subsequent window corresponding to each subsequent structural sampling position, the peak differences of the subsequent windows corresponding to two adjacent subsequent structural sampling positions are paired according to the bearing external transmission position order. For the peak difference of the subsequent window corresponding to the previous subsequent structural sampling position and the peak difference of the subsequent window corresponding to the next subsequent structural sampling position under each relative arrangement position, a direction consistency judgment is performed respectively: if the peak difference of the previous subsequent window is greater than zero, the peak difference of the next subsequent window is greater than zero; if the peak difference of the previous subsequent window is less than zero, the peak difference of the next subsequent window is less than zero; if the peak difference of the previous subsequent window is equal to zero, the peak difference of the next subsequent window is equal to zero. When all direction consistency judgments are valid, the calculation of external transmission variable deviation and the calculation of the cumulative value of external transmission variable deviation are performed; when the direction consistency judgment under any relative arrangement position is invalid, the corresponding adjacent subsequent structural sampling position is determined to not meet the preset external transmission variable consistency condition, and the calculation of external transmission variable deviation and the calculation of the cumulative value of external transmission variable deviation for that adjacent subsequent structural sampling position are stopped.

[0113] When all directional consistency judgments are valid, for each pairing result, the absolute value of the difference between the peak difference of the subsequent window corresponding to the previous subsequent structural sampling position and the peak difference of the subsequent window corresponding to the next subsequent structural sampling position is calculated, and this absolute value of the difference is determined as the external transmission deviation. The external transmission deviations between the same adjacent subsequent structural sampling positions are accumulated to obtain the cumulative value of the external transmission deviation. When all adjacent subsequent structural sampling positions arranged sequentially along the bearing's external transmission position form cumulative values ​​of external transmission deviation, and each cumulative value of external transmission deviation does not exceed the preset cumulative threshold of external transmission deviation, an external transmission continuity relationship is formed. In this embodiment, the external transmission continuity relationship characterizes the consistency of the gradual transmission after the peak recovery sampling window is gradually transmitted outward through the correspondence of the subsequent gradual process between adjacent subsequent structural sampling positions. The judgment criteria that the directional consistency judgments of the peak differences of the subsequent windows corresponding to each adjacent subsequent structural sampling position are valid under the same relative arrangement position, and the corresponding cumulative value of external transmission deviation does not exceed the preset cumulative threshold of external transmission deviation, are determined as the preset external transmission consistency condition.

[0114] After establishing the outgoing transmission numbering succession relationship, outgoing transmission peak succession relationship, and outgoing transmission variation succession relationship, the three types of relationships are summarized. An outgoing transmission continuous succession result is formed when at least two consecutively arranged subsequent structure sampling positions obtain corresponding subsequent recovery windows and an outgoing transmission numbering succession relationship, an outgoing transmission peak succession relationship, and an outgoing transmission variation succession relationship are formed. An outgoing transmission continuous succession result is not formed when at least two consecutively arranged subsequent structure sampling positions do not obtain corresponding subsequent recovery windows, or when any of the three relationships (outgoing transmission numbering succession relationship, outgoing transmission peak succession relationship, and outgoing transmission variation succession relationship) is not formed.

[0115] In this embodiment, the preset external transmission deviation cumulative threshold is formed through multiple rounds of reference load switching tests. Specifically, in the multiple rounds of reference load switching tests, according to the calculation method of the external transmission deviation cumulative value in this embodiment, the external transmission deviation cumulative value between adjacent subsequent structure sampling positions is calculated respectively, resulting in multiple reference external transmission deviation cumulative values; the multiple reference external transmission deviation cumulative values ​​are arranged in ascending order of value, and the reference external transmission deviation cumulative value at the end of the list is selected as the preset external transmission deviation cumulative threshold, which is used to determine whether the window peak change process of adjacent subsequent structure sampling positions after the subsequent rise window meets the preset external transmission consistency condition.

[0116] Step 5: In the multi-round target load switching, the target load switching round that simultaneously generates the incoming source exclusion result and the outgoing continuous acceptance result is determined as the valid confirmation round; when the number of valid confirmation rounds reaches the preset number of confirmation rounds, and the window number difference of the peak recovery sampling window corresponding to the target bearing sampling position in each valid confirmation round does not exceed the preset tolerance range, the bearing initial confirmation result is formed, and the bearing's own vibration abnormality result is output.

[0117] Specifically, in this embodiment, the multi-round target load switching used to form the bearing initial confirmation result in step five refers to the valid confirmation rounds that simultaneously form the incoming source exclusion result and the outgoing continuous acceptance result in the same set of condition rounds and are included in the recovery window number arrangement result; target load switching rounds that are not the same set of condition rounds and do not simultaneously form the incoming source exclusion result and the outgoing continuous acceptance result are not included in the statistical objects of the window number difference comparison in step five.

[0118] To execute step five, multiple rounds of target load switching are repeated. Before each round of target load switching begins, the load switching amplitude, load switching direction, motor speed value corresponding to the load switching moment, time sampling window duration, and window numbering rules for that round are recorded. The load switching amplitude, load switching direction, motor speed value, time sampling window duration, and window numbering rules of the first round of target load switching are determined as the reference. For each of the remaining rounds of target load switching, comparison results are generated for load switching amplitude, load switching direction, motor speed value, time sampling window duration, and window numbering rules. When there are inconsistent comparison results, the round of target load switching is marked as a non-same-condition round and excluded from subsequent valid confirmation round statistics and window number stability confirmation. Target load switching rounds not marked as non-same-condition rounds are determined as the same-condition round set.

[0119] After forming the same set of conditional rounds, for each round of target load switching in the same set of conditional rounds, the peak rise sampling window of the target bearing sampling position corresponding to that round, the incoming source exclusion result corresponding to that round, and the outgoing continuous reception result corresponding to that round are recorded. The peak rise sampling window is determined by the corresponding action in step three, the incoming source exclusion result is formed by the preceding rise window search and evolution consistency judgment action in step three, and the outgoing continuous reception result is formed by the subsequent rise window step-by-step search and outgoing reception relationship judgment action in step four, thereby ensuring that the three types of input results used in step five all come from the same set of conditional rounds.

[0120] After completing the round-by-round recording of the same set of conditional rounds, each round of target load switching is checked to see if it simultaneously has both incoming source exclusion results and outgoing continuous acceptance results. Target load switching rounds that simultaneously have both incoming source exclusion results and outgoing continuous acceptance results are determined as valid confirmation rounds. Further, the window numbers of the peak recovery sampling windows corresponding to the target bearing sampling positions in each valid confirmation round are extracted and arranged according to the chronological order of the target load switching rounds, forming a recovery window numbering arrangement result.

[0121] After generating the recovery window numbering arrangement results, the maximum window number and the minimum window number are found in the recovery window numbering arrangement results, and the difference between the maximum window number and the minimum window number is calculated. This difference is determined as the maximum number difference, which is used to characterize the fluctuation range of the window number of the peak recovery sampling window in multiple rounds of target load switching.

[0122] After obtaining the maximum number difference, quantity verification and tolerance verification are performed respectively. Quantity verification includes: counting the number of valid confirmation rounds and comparing the count with the preset number of confirmation rounds. Tolerance verification includes: comparing the maximum number difference with the preset tolerance range. When the number of valid confirmation rounds reaches the preset number of confirmation rounds and the maximum number difference does not exceed the preset tolerance range, a bearing initial confirmation result is generated, and this result is used to output the bearing's own vibration anomaly result. When the number of valid confirmation rounds does not reach the preset number of confirmation rounds, no bearing initial confirmation result is generated. When the maximum number difference exceeds the preset tolerance range, no bearing initial confirmation result is generated.

[0123] In this embodiment, the preset number of confirmation rounds and the preset tolerance range are formed through a benchmark load switching test. Specifically, in the benchmark load switching test used to calibrate the bearing initial confirmation result, multiple rounds of target load switching are repeatedly executed. In each round of benchmark load switching test, the peak recovery sampling window, the incoming source exclusion result, and the outgoing continuous reception result are formed according to steps three and four. For each round of benchmark load switching, the load switching amplitude, load switching direction, motor speed value, time sampling window duration, and window number setting rules are also recorded. The same set of rounds under the same conditions is formed according to the aforementioned consistency verification action. For each benchmark load switching test, the number of valid confirmation rounds is counted according to the aforementioned determination action of valid confirmation rounds, resulting in multiple benchmark valid rounds. The multiple benchmark valid rounds are arranged in ascending order of value, and the benchmark valid rounds at the end of the list are selected as the preset number of confirmation rounds. Furthermore, for each reference load switching test, the corresponding maximum number difference is calculated according to the aforementioned maximum number difference calculation action to obtain multiple reference maximum number differences; the multiple reference maximum number differences are arranged in ascending order of value, and the reference maximum number difference at the end of the arrangement is selected as the preset tolerance range to ensure that the stability requirements of multiple window numbers are met in the reference load switching test used to calibrate the bearing start confirmation results.

[0124] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for monitoring abnormal bearing vibration during motor load testing, characterized in that, include: Step 1: Obtain multiple structural sampling positions that have a mechanical connection with the bearing installation position, determine the structural sampling position corresponding to the bearing installation position as the target bearing sampling position, and form the input position sequence and the external bearing position sequence; Step 2: Using the target load switching time as the unified response starting point, divide the vibration response data of each structural sampling location into time sampling windows, set window numbers, and extract the peak values ​​of the windows; Step 3: Determine the peak recovery sampling window based on the peak difference between adjacent windows that reach the preset recovery difference threshold in the target bearing sampling position; in the structural sampling positions that are earlier than the peak recovery sampling window in the input position sequence, the number difference is within the preset input window number difference range, the peak value of the previous recovery window is higher than the peak recovery sampling window, and the subsequent change direction meets the preset input change consistency condition. If obtained, an input source confirmation result is formed; if not obtained, an input source exclusion result is formed. Step 4: After the source of the incoming transmission is eliminated, search for the subsequent recovery window corresponding to at least two consecutive subsequent structural sampling positions in the bearing external transmission position sequence; when the window number of each subsequent recovery window is later than the window number of the peak recovery sampling window, and the window number meets the preset external transmission number difference condition, the window peak value meets the preset external transmission attenuation condition, and the subsequent change direction meets the preset external transmission change consistency condition, the external transmission continuous acceptance result is formed. Step 5: In the multi-round target load switching, the target load switching round that simultaneously generates the incoming source exclusion result and the outgoing continuous acceptance result is determined as the valid confirmation round; when the number of valid confirmation rounds reaches the preset number of confirmation rounds, and the window number difference of the peak recovery sampling window corresponding to the target bearing sampling position in each valid confirmation round does not exceed the preset tolerance range, the bearing initial confirmation result is formed, and the bearing's own vibration abnormality result is output.

2. The method for monitoring abnormal bearing vibration during motor load testing according to claim 1, characterized in that, The methods for determining the input position order include: Using the target bearing sampling position as the input termination position, candidate structure sampling positions pointing to the target bearing sampling position are obtained along the mechanical connection direction of the load end and the mechanical connection direction of the mounting base; Within a preset verification time range after the target load switch, obtain the response start time, response amplitude change, and response duration at each candidate structure sampling location; Among the adjacent candidate structure sampling positions in the same mechanical connection direction, the candidate structure sampling position that is far away from the target bearing sampling position is determined as the first candidate structure sampling position, and the candidate structure sampling position that is close to the target bearing sampling position is determined as the second candidate structure sampling position. And judge based on the following conditions: a1: The response start time of the first candidate structure sampling position is earlier than the response start time of the second candidate structure sampling position; a2: The change in response amplitude at both the first candidate structure sampling position and the second candidate structure sampling position is not lower than the preset input detectable amplitude threshold; a3: The difference in response duration between the sampling positions of the first and second candidate structures is within the preset range of input duration difference; When a1, a2, and a3 are satisfied simultaneously, the sampling positions of the first and second candidate structures are included in the input position order. When a1, a2, and a3 are not simultaneously satisfied, the sampling position of the first candidate structure is determined as the input continuity interruption position, and the sampling positions of candidate structures located on the side away from the target bearing sampling position at this position are stopped from being included in the input position sequence. The sampling positions of candidate structures included in the input position sequence are arranged according to the mechanical connection sequence from the sampling position away from the target bearing sampling position to the sampling position close to the target bearing sampling position, forming the input position sequence corresponding to the mechanical connection direction of the load end and the mechanical connection direction of the mounting foundation, respectively.

3. The method for monitoring abnormal bearing vibration during motor load testing according to claim 1, characterized in that, Methods for forming the bearing external transmission position sequence include: Using the target bearing sampling position as the starting point of the outward transmission, multiple candidate outward transmission positions that have a mechanical connection relationship with the target bearing sampling position and are distributed along directions away from the target bearing sampling position are obtained along different outward transmission mechanical connection directions. Within each outward mechanical connection direction, obtain the target bearing sampling position and the response start time, response amplitude change, and mechanical connection distance of each outward candidate position; Within the same external mechanical connection direction, the target bearing sampling position is determined as the current reference position, and the response start time of the current reference position is determined as the current reference time; Among the candidate positions of external transmission that are not included in the bearing external transmission position sequence within the same external transmission mechanical connection direction, find the candidate position of external transmission whose response start time is later than the current reference time, whose response amplitude change is not lower than the lower limit of the detectable amplitude in the external transmission direction, and whose mechanical connection distance is the smallest relative to the current reference position. The found candidate position of external transmission is determined as the current external transmission connection position. Incorporate the current external transmission connection position into the bearing external transmission position sequence, update the current reference position to the current external transmission connection position, and update the current reference time to the response start time of the current external transmission connection position; The search and update of the current external transmission connection position is repeated within the same external transmission mechanical connection direction until no current external transmission connection position is found among the external transmission candidate positions not included in the bearing external transmission position sequence. The external transmission candidate positions that have been included in the bearing external transmission position sequence are arranged in the order of inclusion to form the bearing external transmission position sequence.

4. The method for monitoring abnormal bearing vibration during motor load testing according to claim 1, characterized in that, The method for setting window numbers in step two includes: Using the target load switching time as a unified response starting point, the vibration response data of each structural sampling location are divided into multiple time sampling windows of the same length, and the starting time difference of each time sampling window relative to the target load switching time is recorded. Based on the start time difference of each time sampling window, determine the window arrangement position of the corresponding time sampling window, and set the time sampling windows with the same window arrangement position to the same window number; For any given window number, count the number of sampling points and the sampling time coverage of each structural sampling location within the time sampling window corresponding to that window number; when the number of sampling points at any structural sampling location is lower than the preset valid sampling number, the window number is determined to be an invalid window number; when the sampling time coverage of any structural sampling location does not reach the preset window coverage, the window number is determined to be an invalid window number. Exclude time sampling windows corresponding to invalid window numbers, determine the time sampling windows corresponding to the remaining window numbers as the time sampling windows participating in window peak extraction, and retain the original numbering order of the remaining window numbers relative to the target load switching time.

5. The method for monitoring abnormal bearing vibration during motor load testing according to claim 1, characterized in that, The method for determining the peak recovery sampling window in step three includes: In each time sampling window corresponding to the target bearing sampling position, the difference between the adjacent window peak values ​​between the window peak value of the next time sampling window and the window peak value of the previous time sampling window is calculated in ascending order of window number. The sampling window after which the peak difference between adjacent windows reaches the preset recovery difference threshold is determined as the recovery window to be verified, and the peak difference between the adjacent windows is determined as the benchmark recovery difference. Starting with the recovery window to be verified, multiple subsequent time sampling windows are continuously extracted along the direction of increasing window number, and the peak difference between adjacent subsequent time sampling windows is calculated respectively. Calculate the ratio between the peak difference of each subsequent window and the baseline recovery difference to obtain multiple difference ratios, and arrange the multiple difference ratios in ascending order according to the window number of the corresponding subsequent time sampling window; Compare the difference ratios at the top and bottom of the list, and check for adjacent ratio pairs where the latter difference ratio is greater than the former difference ratio among multiple difference ratios. When the difference ratio at the end of the list is less than the difference ratio at the beginning of the list, there are no adjacent ratio pairs among multiple difference ratios where the next difference ratio is greater than the previous difference ratio, and the peak difference of each subsequent window does not exceed the baseline recovery difference, the recovery window to be verified is determined as the peak recovery sampling window.

6. The method for monitoring abnormal bearing vibration during motor load testing according to claim 1, characterized in that, The method for determining in step three whether the subsequent change direction of the preceding recovery window and the peak recovery sampling window meets the preset input change consistency condition includes: In the sequence of each input position where the previous position recovery window has been obtained, the first time sampling window after the previous position recovery window is used as the starting point to extract multiple input subsequent time sampling windows, and in the target bearing sampling position, the first time sampling window after the peak recovery sampling window is used as the starting point to extract the same number of target subsequent time sampling windows. Calculate the peak difference of the input window between adjacent input subsequent time sampling windows and the peak difference of the target window between adjacent target subsequent time sampling windows, respectively; When the signs of the peak difference between the input window and the peak difference between the target window are consistent in all relative positions in the same input position sequence, calculate the absolute value of the difference between the two in each relative position and determine it as the difference deviation. The deviations of each difference in the same input position sequence are accumulated to obtain the cumulative value of the progressive deviation; In the sequence of input positions that form the cumulative value of the incremental deviation, the input position sequence with the smallest cumulative value of incremental deviation is determined as the target input position sequence; when the cumulative value of incremental deviation corresponding to the target input position sequence does not exceed the preset cumulative threshold of incremental deviation, the preceding recovery window and the peak recovery sampling window in the target input position sequence are determined as the subsequent incremental direction satisfies the preset condition of consistent incremental change.

7. The method for monitoring abnormal bearing vibration during motor load testing according to claim 1, characterized in that, The method for generating continuous outgoing transmission results in step four includes: After the incoming source exclusion result is formed, at least two consecutive rear structure sampling positions are obtained along the bearing external transmission position, and the rear recovery window corresponding to each rear structure sampling position is determined respectively. Based on the hierarchical succession relationship formed by each subsequent recovery window along the bearing's external transmission position, an external transmission number succession relationship is formed; Based on the changes in the peak value of each subsequent recovery window along the bearing's outward transmission position, an outward transmission peak value succession relationship is formed; Based on the peak value change process after each subsequent recovery window, an external transmission and transfer relationship is formed; When external transmission numbering succession relationship, external transmission peak succession relationship, and external transmission variation succession relationship are formed, an external transmission continuous succession result is formed.

8. The method for monitoring abnormal bearing vibration during motor load testing according to claim 7, characterized in that, The methods for forming the sequence of serial numbers in external publications include: Among the multiple consecutive sampling positions of the subsequent structure, the sampling position of the subsequent structure adjacent to the sampling position of the target bearing is determined as the first sampling position of the subsequent structure, and the peak recovery sampling window is used as the current search reference window for the first sampling position of the subsequent structure. For the current sampling position of the subsequent structure to be found, calculate the difference between the peak value of the next sampling window and the peak value of the previous sampling window in ascending order of window number, and determine the difference as the peak value difference between adjacent windows. In the time sampling window whose window number is later than the current lookup reference window, the next time sampling window whose peak difference between adjacent windows reaches the preset recovery difference threshold and whose window number is the smallest is determined as the corresponding subsequent recovery window. The next subsequent structure sampling position is selected sequentially along the bearing external transmission position, and the previous subsequent recovery window is used as the current search reference window, until the corresponding subsequent recovery window is obtained for multiple consecutively arranged subsequent structure sampling positions. Calculate the difference between the window number of the next subsequent rise window and the window number of the previous subsequent rise window in two adjacent subsequent rise windows to obtain the difference between adjacent outgoing numbers; when the differences between adjacent outgoing numbers are all positive and the differences between adjacent outgoing numbers meet the preset outgoing number difference conditions, an outgoing number succession relationship is formed.

9. The method for monitoring abnormal bearing vibration during motor load testing according to claim 1, characterized in that, The method for generating the bearing initial verification result in step five includes: In the multi-round target load switching, the peak recovery sampling window, the incoming source exclusion result and the outgoing continuous acceptance result corresponding to each round of target load switching are extracted respectively. The target load switching rounds that have formed the results of excluding incoming sources and the results of continuous outgoing transmission are determined as valid confirmation rounds; Extract the window number of the peak recovery sampling window corresponding to the target bearing sampling position in each valid confirmation round, and arrange them according to the order of target load switching rounds to form the recovery window number arrangement result; Subtract the minimum window number from the maximum window number in the result of sorting the rising window numbers to obtain the maximum number difference; When the number of valid confirmation rounds reaches the preset number of confirmation rounds and the maximum number difference does not exceed the preset tolerance range, the bearing initial confirmation result is formed.

Images