Optimization method and system suitable for non-uniform time series data in torsional vibration measurement

Abstract

The application discloses an optimization method and system suitable for non-uniform timing data in torsional vibration measurement, and relates to the technical field of rotating machinery state monitoring.The application comprises the following steps: receiving original square wave signals output by a rotary encoder, detecting pulse edge types, extracting accurate time stamps of each edge, and generating an original time stamp sequence; based on the original time stamp sequence, a time interval sequence is calculated, and a reference time interval is calculated based on an arithmetic average value; double-criterion abnormality identification is performed on the time interval sequence, and an abnormality type is determined; corresponding interpolation correction is performed according to the abnormality type identification result, and an optimized time stamp sequence is generated; and the optimized time stamp sequence is subjected to uniformity verification and output as input data for torsional vibration analysis.The application reconstructs a high-precision torsional vibration waveform through feature identification and deterministic interpolation compensation of the time stamp sequence, and provides reliable data basis for rotating machinery load health monitoring and fault diagnosis.

Description

Technical Field

[0001] This invention relates to the field of rotating machinery condition monitoring technology, specifically to an optimization method and system for non-uniform time-series data in torsional vibration measurement. Background Technology

[0002] Rotating machinery is core equipment in industries such as energy, power, shipbuilding, and aviation. Its operating status directly affects production safety and efficiency. In torsional vibration analysis of rotating machinery, high-precision sampling of the instantaneous rotational speed fluctuation of the shaft relies on the pulse edge signal output by the shaft-end encoder or speed sensor. By measuring the time interval between adjacent pulse edges, the angular velocity can be calculated and the torsional vibration time-domain waveform can be constructed. However, in complex industrial environments, the idealized uniform sampling conditions described above are difficult to guarantee at all times. Various factors, such as sensor aging, loose installation, strong electromagnetic interference, or mechanical vibration transmitted to the sensor body, can cause deterministic distortion of the output pulse signal, mainly manifested in two types of problems: First, non-equal interval sampling, that is, the time interval between the measured adjacent pulse edges continuously or randomly deviates from its theoretical design value (i.e., the theoretical period at constant speed); Second, pulse loss, that is, due to instantaneous signal drops, acquisition card trigger failure, etc., the pulse edges that should appear are not effectively captured, making the measured time interval approximately an integer multiple of the normal value (e.g., 2 times, 3 times). These problems disrupt the uniformity of the torsional vibration signal time series, making the original timestamp sequence non-uniform time-series data.

[0003] Existing technologies employ non-uniform resampling or dynamic clock synchronization strategies, but these methods are mainly applicable to variable speed or wide speed range conditions. They do not provide a dedicated correction mechanism for discrete sampling loss and interval distortion caused by hardware anomalies under steady speed conditions. Therefore, this invention proposes an optimization method and system suitable for non-uniform time-series data in torsional vibration measurement. Summary of the Invention

[0004] The purpose of this invention is to provide an optimization method and system for non-uniform time-series data in torsional vibration measurement. To address the problem of non-uniform time series caused by abnormal sensor signals, a high-precision torsional vibration waveform is reconstructed through feature identification and deterministic interpolation compensation of timestamp sequences, providing a reliable data foundation for load health monitoring and fault diagnosis of rotating machinery.

[0005] According to a first aspect of the present invention, in order to achieve the above-mentioned objective, the present invention provides the following technical solution: a method for optimizing non-uniform time-series data in torsional vibration measurement, comprising the following steps: Receive the raw square wave signal output by the rotary encoder, detect the pulse edge type and extract the precise timestamp of each edge to generate the raw timestamp sequence, where the edge type includes rising edge or falling edge; Based on the original timestamp sequence, the difference between adjacent timestamps is calculated to obtain the time interval sequence, and the baseline time interval is calculated based on the arithmetic mean. Dual-criteria anomaly identification is performed on the time interval sequence, and the anomaly type is determined, wherein the anomaly type includes pulse missing anomaly and sampling error anomaly; Based on the anomaly type identification results, perform corresponding interpolation corrections to generate an optimized timestamp sequence. ; The optimized timestamp sequence is subjected to uniformity verification. When the verification passes, the output is used as input data for torsional vibration analysis.

[0006] Furthermore, the raw square wave signal output from the rotary encoder is received, the pulse edge type is detected, and the precise timestamp of each edge is extracted to generate the raw timestamp sequence, as follows: (21) The square wave signal is captured in real time by a high-speed digital acquisition card with a sampling rate of not less than 10 times the highest edge frequency of the encoder; (22) Rising edge or falling edge shall be used as the effective edge triggering condition. (23) Extract the precise timestamps of each valid edge. Constructing the original timestamp sequence , where the sequence length .

[0007] Furthermore, based on the original timestamp sequence, the difference between adjacent timestamps is calculated to obtain a time interval sequence, and the baseline time interval is calculated based on the arithmetic mean, as follows: Calculate the difference between adjacent timestamps: The time interval sequence is as follows: To eliminate the impact of the start-stop transition process, the first and last five time intervals of the time interval sequence were removed. For the remaining... The arithmetic mean of the time intervals is calculated as the baseline time interval: In the formula, As the reference time interval, The difference between adjacent timestamps. It is a time interval sequence.

[0008] Furthermore, a dual-criteria anomaly identification is performed on the time interval sequence, and the anomaly type is determined. The anomaly types include pulse missing anomalies and sampling error anomalies, as detailed below. For each time interval to be detected , to The dual criteria shall be applied in the following order: (41) First criterion: identification of missing integer multiples Calculate the ratio Take the integer part; like And it satisfies: Among them, tolerance threshold If the interval is determined to be a pulse missing anomaly, the number of missing pulses is [number missing]. ; (42) Second criterion: Significant offset identification If the first criterion is not met, and the following conditions are satisfied: Among them, tolerance threshold If so, the interval is determined to be an abnormal sampling error; Otherwise, it is considered a normal interval.

[0009] Furthermore, uniform interpolation compensation is performed for pulse missing anomalies, as follows: When a missing value is identified At each pulse, at the original timestamp and Insert evenly between A new timestamp: .

[0010] Furthermore, interpolation correction based on local neighborhood verification is performed to address sampling error anomalies, as follows: Define the normal interval set: Select 3 positions before and after the interval to be corrected, and filter those that meet the requirements. The time intervals constitute the set of normal intervals in the local neighborhood, where ; Calculate the local average interval: Where L is the number of normal intervals; Updated timestamps: Perform a synchronous shift on all timestamps p > i+1: In the formula, represent Corrected timestamp represent Subsequent update timestamps.

[0011] Furthermore, the optimized timestamp sequence Perform uniformity verification. When the verification passes, the output will be used as input data for torsional vibration analysis, as follows: Integrate the original timestamps with the interpolated new timestamps, reorder them chronologically, and generate an optimized timestamp sequence. ; Verify the optimization effect: Calculate the standard deviation of adjacent time intervals after optimization. The following requirements must be met: in The standard deviation of the original sequence is required; at the same time, the maximum relative deviation is required to be no more than 10%, that is: When the verification passes, output As input data for torsional vibration analysis.

[0012] According to a second aspect of the present invention, the present invention provides an optimization system for non-uniform time-series data in torsional vibration measurements, for implementing the optimization method for non-uniform time-series data in torsional vibration measurements described in the first aspect, comprising: The edge detection and timestamp extraction module is used to receive the raw square wave signal output by the rotary encoder, detect the pulse edge type and extract the precise timestamp of each edge to generate the raw timestamp sequence, wherein the edge type includes rising edge or falling edge; The time interval calculation module is used to calculate the difference between adjacent timestamps based on the original timestamp sequence to obtain the time interval sequence, and to calculate the baseline time interval based on the arithmetic mean. An abnormal interval identification module is used to perform dual-criteria abnormal identification on the time interval sequence and determine the abnormal type, wherein the abnormal type includes pulse missing abnormality and sampling error abnormality; The interpolation correction and output module is used to perform corresponding interpolation corrections based on the anomaly type identification results, generating an optimized timestamp sequence. ; The verification module is used to verify the uniformity of the optimized timestamp sequence. When the verification is successful, the output is used as input data for torsional vibration analysis.

[0013] According to a third aspect of the present invention, a terminal device is provided, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores the computer program capable of running on the processor, and when the processor loads and executes the computer program, it employs the optimization method for non-uniform time-series data in torsional vibration measurement described in the first aspect.

[0014] According to a fourth aspect of the present invention, the present invention provides a storage medium comprising computer-executable instructions, which, when executed by a computer processor, are used to perform an optimization method for non-uniform time-series data in torsional vibration measurements as described in the first aspect.

[0015] This invention has at least the following beneficial effects: 1. This invention addresses the non-uniform sampling problem caused by sensor hardware malfunctions under steady-speed conditions. By using feature recognition of timestamp sequences and a deterministic interpolation compensation mechanism, it significantly improves the temporal uniformity of torsional vibration signals. It employs a sequential dual-criteria mechanism to prioritize the identification of missing integer multiple pulses and then identify non-integer multiple sampling errors. This deterministic discrimination process effectively avoids the inherent defect of the traditional single threshold method in misjudging missing pulses as speed fluctuations.

[0016] 2. This invention adopts a local interpolation strategy based on neighborhood verification. When correcting sampling errors, it strictly calculates the local average interval according to the adjacent normal interval, ensuring that the interpolation result conforms to the physical change law of the signal. It avoids the secondary distortion that may be introduced by simple global averaging or arbitrary interpolation, and significantly improves the authenticity and smoothness of the reconstructed waveform.

[0017] 3. This invention sets differentiated tolerance thresholds for pulse loss and sampling offset. The first tolerance threshold is set at 5% of the reference interval, which is designed specifically for identifying pulse loss of integer multiples. It can effectively capture real loss events while avoiding false alarms triggered by small speed fluctuations or quantization noise. The second tolerance threshold is set at 25% of the reference interval to distinguish between normal fluctuations and sampling error anomalies. In the anomaly identification process, it can effectively capture real anomalies and has good anti-noise and anti-minor fluctuation capabilities, which improves the practicality and robustness of the method in complex industrial environments.

[0018] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the method described in this invention. Detailed Implementation

[0020] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0021] Example 1: This embodiment provides an optimization method for non-uniform time-series data in torsional vibration measurement. It is applicable to the problem of unequal sampling intervals and pulse loss caused by sensor signal anomalies during steady-speed operation. This method generates a time interval sequence with optimized time uniformity by performing feature extraction, anomaly detection, and deterministic interpolation compensation on the encoder square wave pulse edge timestamp sequence, providing accurate input data for torsional vibration analysis. The method is applicable when the speed fluctuation range does not exceed [a certain value]. Under this constraint, the reference interval It has a clear physical meaning, ensuring the effectiveness of the method.

[0022] Please see Figure 1 This invention provides a technical solution: an optimization method for non-uniform time-series data in torsional vibration measurement, comprising the following steps: S1. Receive the raw square wave signal output by the rotary encoder, detect the pulse edge type and extract the precise timestamp of each edge, and generate the raw timestamp sequence, wherein the edge type includes rising edge or falling edge; (S11) The square wave signal is captured in real time by a high-speed digital acquisition card with a sampling rate of not less than 10 times the highest edge frequency of the encoder; (S12) Rising edge or falling edge shall be used as the valid edge triggering condition. (S13) Extract the precise timestamps of each valid edge. Constructing the original timestamp sequence , where the sequence length To ensure the validity of statistics; It should be noted that this embodiment is limited to using only one type of trigger condition, either rising edge or falling edge, as the effective trigger condition. This choice of technique stems from the stringent requirements of torsional vibration calculation for the consistency of the angle reference. Specifically, when rising edge detection is selected, the detection circuit needs to identify the zero-crossing point of the square wave signal transitioning from a logic low level to a logic high level, and the physical time t of this zero-crossing point... i The transition edge of the output state of the hardware comparator is used for calibration; similarly, the falling edge detection corresponds to the transition moment from high to low level. This invention abandons the scheme of simultaneous detection of both edges because: encoder manufacturing tolerances and installation eccentricity may cause the angular positions corresponding to the rising edge and the falling edge to not be strictly separated by 180 mechanical angles. If the two types of edges are mixed, a systematic phase error will be introduced. This error will be amplified into a relative deviation on the order of Δω / ω ≈ (Δθ / θ)·(Δt / t) in the calculation of torsional vibration angular velocity, where Δθ is the angular asymmetry and θ is the nominal angular interval; S2. Based on the original timestamp sequence, calculate the difference between adjacent timestamps to obtain the time interval sequence, and calculate the baseline time interval based on the arithmetic mean; (S21) Calculate the difference between adjacent timestamps: The time interval sequence is as follows: (S22) Remove the first and last 5 time intervals from the time interval sequence to eliminate the impact of the start-stop transition process, and then remove the remaining 5 time intervals from the sequence. The arithmetic mean of the time intervals is calculated as the baseline time interval: In the formula, As the reference time interval, The difference between adjacent timestamps. It is a time interval sequence; It should be further explained that, in order to ensure the physical accuracy of the benchmark value, this embodiment does not directly calculate the average value of the entire sequence when calculating the benchmark time interval. Instead, it first executes the preset boundary clearing logic, that is, clears the first and last 5 time intervals of the time interval sequence. Its physical meaning is to eliminate the non-smooth interruption of the rotating machinery at the moment of data interruption or during the start-stop transition. S3. Perform dual-criteria anomaly identification on the time interval sequence and determine the anomaly type, wherein the anomaly type includes pulse missing anomaly and sampling error anomaly; For each time interval to be detected , to The dual criteria shall be applied in the following order: (S31) First criterion: Identification of missing integer multiples Calculate the ratio Take the integer part; like And it satisfies: Among them, tolerance threshold If the interval is determined to be a pulse missing anomaly, the number of missing pulses is [number missing]. ; (S32) Second criterion: Significant offset identification If the first criterion is not met, and the following conditions are satisfied: Among them, tolerance threshold If so, the interval is determined to be an abnormal sampling error; Otherwise, it is considered a normal interval; The first criterion prioritizes identifying pulse loss that is an integer multiple, while the second criterion identifies sampling errors that are not integer multiples. The execution of the criteria has a deterministic order. Through the dual anomaly criterion mechanism executed in sequence, it can accurately distinguish between pulse loss and general sampling errors. The criterion logic is clear and the execution is highly deterministic, effectively avoiding the ambiguity and misjudgment of existing general methods in anomaly type identification, and providing a reliable basis for subsequent targeted correction. S4. Perform corresponding interpolation correction based on the anomaly type identification result to generate an optimized timestamp sequence. ; (S41) Perform uniform interpolation compensation for pulse missing anomalies, as follows: When a missing value is identified At each pulse, at the original timestamp and Insert evenly between A new timestamp: .

[0023] (S42) Perform interpolation correction based on local neighborhood verification for sampling error anomalies, as follows: Define the normal interval set: Select 3 positions before and after the interval to be corrected, and filter those that meet the requirements. The time intervals constitute the set of normal intervals in the local neighborhood, where ; Calculate the local average interval: Where L is the number of normal intervals; Updated timestamps: Perform a synchronous shift on all timestamps p > i+1: ; In the formula, represent Corrected timestamp represent Subsequent update timestamps; Specifically, for sampling error anomalies, the screening criteria for normal intervals within the local neighborhood are strictly limited. To ensure the physical authenticity of the interpolated data, a local interpolation strategy based on neighborhood verification is adopted. When correcting sampling errors, the local average interval is calculated strictly according to the adjacent normal interval, ensuring that the interpolation results conform to the physical change law of the signal and avoiding the secondary distortion that may be introduced by simple global averaging or arbitrary interpolation, which significantly improves the authenticity and smoothness of the reconstructed waveform. S5. Perform uniformity verification on the optimized timestamp sequence. When the verification passes, output the data as input for torsional vibration analysis. Integrate the original timestamps with the interpolated new timestamps, reorder them chronologically, and generate an optimized timestamp sequence. ; Verify the optimization effect: Calculate the standard deviation of adjacent time intervals after optimization. The following requirements must be met: in The standard deviation of the original sequence is required; at the same time, the maximum relative deviation is required to be no more than 10%, that is: When the verification passes, output As input data for torsional vibration analysis.

[0024] In summary, this invention addresses the non-uniform sampling problem caused by sensor hardware malfunctions under steady-speed conditions. By employing feature recognition of timestamp sequences and a deterministic interpolation compensation mechanism, it significantly improves the temporal uniformity of torsional vibration signals. The invention utilizes a sequential dual-criteria mechanism to prioritize the identification of missing integer multiple pulses, followed by the identification of non-integer multiple sampling errors. This deterministic discrimination process effectively avoids the inherent defect of the traditional single threshold method in misjudging missing pulses as speed fluctuations. A local interpolation strategy based on neighborhood verification is adopted. When correcting sampling errors, the local average interval is calculated strictly according to the adjacent normal interval. This ensures that the interpolation result conforms to the physical change law of the signal and avoids the secondary distortion that may be introduced by simple global averaging or arbitrary interpolation. This significantly improves the authenticity and smoothness of the reconstructed waveform. Different tolerance thresholds are set for pulse loss and sampling offset. The first tolerance threshold is set at 5% of the reference interval, which is designed to identify pulse loss of integer multiples. It can effectively capture real loss events while avoiding false alarms triggered by small speed fluctuations or quantization noise. The second tolerance threshold is set at 25% of the reference interval to distinguish between normal fluctuations and sampling error anomalies. In the anomaly identification process, it can effectively capture real anomalies and has good noise resistance and resistance to slight fluctuations, which improves the practicality and robustness of the method in complex industrial environments.

[0025] Example 2: This embodiment provides an optimization system suitable for non-uniform time-series data in torsional vibration measurement, used to implement the optimization method for non-uniform time-series data in torsional vibration measurement described in Embodiment 1, including: The edge detection and timestamp extraction module is used to receive the raw square wave signal output by the rotary encoder, detect the pulse edge type and extract the precise timestamp of each edge to generate the raw timestamp sequence, wherein the edge type includes rising edge or falling edge; The time interval calculation module is used to calculate the difference between adjacent timestamps based on the original timestamp sequence to obtain the time interval sequence, and to calculate the baseline time interval based on the arithmetic mean. An abnormal interval identification module is used to perform dual-criteria abnormal identification on the time interval sequence and determine the abnormal type, wherein the abnormal type includes pulse missing abnormality and sampling error abnormality; The interpolation correction and output module is used to perform corresponding interpolation corrections based on the anomaly type identification results, generating an optimized timestamp sequence. ; The verification module is used to verify the uniformity of the optimized timestamp sequence. When the verification is successful, the output is used as input data for torsional vibration analysis.

[0026] Example 3: The present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor loads and executes the computer program, it employs the optimization method for non-uniform time-series data in torsional vibration measurement described in Embodiment 1.

[0027] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer, or a cloud server, and the terminal device includes, but is not limited to, a processor and a memory. For example, the terminal device may also include input / output devices, network access devices, and buses.

[0028] Furthermore, the processor can be a central processing unit (CPU). Of course, depending on the actual use, other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. can also be used. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it in this regard.

[0029] Example 4: The present invention provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the optimization method described in Embodiment 1 for non-uniform time-series data in torsional vibration measurement.

[0030] The computer program can be stored in a computer-readable medium. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or certain middleware. The computer-readable medium includes any entity or device capable of carrying computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the computer-readable medium includes, but is not limited to, the above-mentioned components.

[0031] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0032] For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on," "mounted on," "fixed to," or "set on" another element, it may be directly on the other element or there may be an intermediate element present. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be an intermediate element present. The terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible embodiments.

[0033] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0034] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

Claims

1. An optimization method for non-uniform time-series data in torsional vibration measurement, characterized in that, Includes the following steps: Receive the raw square wave signal output by the rotary encoder, detect the pulse edge type and extract the precise timestamp of each edge to generate the raw timestamp sequence, where the edge type includes rising edge or falling edge; Based on the original timestamp sequence, the difference between adjacent timestamps is calculated to obtain the time interval sequence, and the baseline time interval is calculated based on the arithmetic mean. Dual-criteria anomaly identification is performed on the time interval sequence, and the anomaly type is determined, wherein the anomaly type includes pulse missing anomaly and sampling error anomaly; Based on the anomaly type identification results, perform corresponding interpolation corrections to generate an optimized timestamp sequence. ; The optimized timestamp sequence is subjected to uniformity verification. When the verification passes, the output is used as input data for torsional vibration analysis.

2. The optimization method for non-uniform time-series data in torsional vibration measurement according to claim 1, characterized in that: The system receives the raw square wave signal output from the rotary encoder, detects the pulse edge type, extracts the precise timestamp of each edge, and generates the raw timestamp sequence, as follows: (21) The square wave signal is captured in real time by a high-speed digital acquisition card with a sampling rate of not less than 10 times the highest edge frequency of the encoder; (22) Rising edge or falling edge shall be used as the effective edge triggering condition. (23) Extract the precise timestamps of each valid edge. Constructing the original timestamp sequence , where the sequence length .

3. The optimization method for non-uniform time-series data in torsional vibration measurement according to claim 1, characterized in that: Based on the original timestamp sequence, the difference between adjacent timestamps is calculated to obtain the time interval sequence, and the baseline time interval is calculated based on the arithmetic mean, as follows: Calculate the difference between adjacent timestamps: The time interval sequence is as follows: To eliminate the impact of the start-stop transition process, the first and last five time intervals of the time interval sequence were removed. For the remaining... The arithmetic mean of the time intervals is calculated as the baseline time interval: In the formula, As the reference time interval, The difference between adjacent timestamps. It is a time interval sequence.

4. The optimization method for non-uniform time-series data in torsional vibration measurement according to claim 3, characterized in that: Dual-criteria anomaly identification is performed on the time interval sequence, and the anomaly type is determined. The anomaly types include pulse missing anomalies and sampling error anomalies, as detailed below. For each time interval to be detected , to The dual criteria shall be applied in the following order: (41) First criterion: identification of missing integer multiples Calculate the ratio Take the integer part; like And it satisfies: Among them, tolerance threshold If the interval is determined to be a pulse missing anomaly, the number of missing pulses is [number missing]. ; (42) Second criterion: Significant offset identification If the first criterion is not met, and the following conditions are satisfied: Among them, tolerance threshold If so, the interval is determined to be an abnormal sampling error; Otherwise, it is considered a normal interval.

5. The optimization method for non-uniform time-series data in torsional vibration measurement according to claim 4, characterized in that: Uniform interpolation compensation is performed to address pulse missing anomalies, as follows: When a missing value is identified At each pulse, at the original timestamp and Insert evenly between A new timestamp: 。 6. The optimization method for non-uniform time-series data in torsional vibration measurement according to claim 4, characterized in that: For sampling error anomalies, interpolation correction based on local neighborhood verification is performed as follows: Define the normal interval set: Select 3 positions before and after the interval to be corrected, and filter those that meet the requirements. The time intervals constitute the set of normal intervals in the local neighborhood, where ; Calculate the local average interval: Where L is the number of normal intervals; Updated timestamps: Perform a synchronous shift on all timestamps p > i+1: In the formula, represent Corrected timestamp represent Subsequent update timestamps.

7. The optimization method for non-uniform time-series data in torsional vibration measurement according to claim 6, characterized in that: For the optimized timestamp sequence Perform uniformity verification. When the verification passes, the output will be used as input data for torsional vibration analysis, as follows: Integrate the original timestamps with the interpolated new timestamps, reorder them chronologically, and generate an optimized timestamp sequence. ; Verify the optimization effect: Calculate the standard deviation of adjacent time intervals after optimization. The following requirements must be met: in The standard deviation of the original sequence is required; at the same time, the maximum relative deviation is required to be no more than 10%, that is: When the verification passes, output As input data for torsional vibration analysis.

8. An optimization system for non-uniform time-series data in torsional vibration measurement, used to implement the optimization method for non-uniform time-series data in torsional vibration measurement as described in any one of claims 1 to 7, characterized in that, include: The edge detection and timestamp extraction module is used to receive the raw square wave signal output by the rotary encoder, detect the pulse edge type and extract the precise timestamp of each edge to generate the raw timestamp sequence, wherein the edge type includes rising edge or falling edge; The time interval calculation module is used to calculate the difference between adjacent timestamps based on the original timestamp sequence to obtain the time interval sequence, and to calculate the baseline time interval based on the arithmetic mean. An abnormal interval identification module is used to perform dual-criteria abnormal identification on the time interval sequence and determine the abnormal type, wherein the abnormal type includes pulse missing abnormality and sampling error abnormality; The interpolation correction and output module is used to perform corresponding interpolation corrections based on the anomaly type identification results, generating an optimized timestamp sequence. ; The verification module is used to verify the uniformity of the optimized timestamp sequence. When the verification is successful, the output is used as input data for torsional vibration analysis.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, The memory stores a computer program that can run on a processor. When the processor loads and executes the computer program, it employs the optimization method for non-uniform time-series data in torsional vibration measurement as described in any one of claims 1 to 7.

10. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform an optimization method for non-uniform time-series data in torsional vibration measurements as described in any one of claims 1 to 7.

Images