A method and system for monitoring torsional vibration of a shaft system of a constant speed device
By optimizing the signal processing logic and improving the algorithm for monitoring torsional vibration of generator shaft systems, the trigger edge is identified and high-order interpolation reconstruction is performed, solving the problems of high hardware cost and insufficient sampling accuracy in the existing technology, and realizing low-cost, high-reliability torsional vibration monitoring and risk warning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ANHUI DETONG ZHILIAN TECH CO LTD
- Filing Date
- 2026-01-29
- Publication Date
- 2026-06-09
AI Technical Summary
Existing technologies for monitoring torsional vibration in generator shaft systems suffer from high hardware costs, underutilization of existing speed measurement resources, and mismatched data processing models, resulting in insufficient sampling accuracy and analysis errors, making it difficult to achieve efficient and accurate torsional vibration monitoring.
By performing logical judgment and algorithm optimization on data with a fixed sampling rate, the signal trigger edge is identified, the instantaneous angular velocity is calculated, and high-order interpolation reconstruction and filtering are performed to extract the peak characteristic index of the torsional vibration angular displacement signal, thereby achieving low-cost and high-reliability torsional vibration monitoring.
Without altering the existing speed measurement hardware architecture, it improves sampling density and signal accuracy, accurately extracts torsional vibration characteristic information, provides clear equipment health status assessment and risk warning, and reduces hardware modification costs.
Smart Images
Figure CN122171008A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power equipment fault diagnosis technology, specifically to a method and system for monitoring torsional vibration of shaft systems in equipment operating at constant speeds. Background Technology
[0002] The safe and stable operation of power systems is directly related to the national economy and social life. As the core power equipment of the power system, generator sets bear complex mechanical and electromagnetic torques on their shaft systems during long-term operation. Torsional vibration refers to periodic angular velocity fluctuations superimposed on the average rotational speed of the shaft system, resulting from the combined effects of torsional elastic deformation of the shaft system and external excitation. Even small-amplitude torsional vibrations that persist for a long time can lead to fatigue accumulation in the shaft materials, potentially causing serious accidents such as shaft cracks or even fractures, resulting in huge economic losses and risks to power grid operation. Therefore, effective and accurate online monitoring and early warning of torsional vibration in generator set shaft systems is of great significance for ensuring equipment safety and achieving predictive maintenance.
[0003] Currently, the monitoring of shaft torsional vibration in engineering mainly relies on high-precision direct measurement or complex signal analysis methods, but these methods generally have the following limitations: First, the high cost of hardware limits large-scale application. Traditional high-precision torsional vibration monitoring solutions typically rely on specialized sensors such as high-line-count photoelectric encoders or laser velocimeters mounted on the shaft system, and require a dedicated data acquisition system with an extremely high sampling rate. Such specialized equipment is expensive, complex to install and maintain, and difficult to deploy widely in existing generator sets or in generator sets with strict cost controls.
[0004] Secondly, existing conventional speed measurement resources are not being fully utilized. In fact, most generator sets are standardly equipped with speed measuring gears and corresponding magnetoelectric or Hall effect speed sensors to monitor speed and provide overspeed protection, capable of outputting pulse sequences synchronized with shaft rotation. However, the subsequent data acquisition modules typically use a fixed, relatively low sampling rate for signal acquisition. This fixed sampling rate acquisition method suffers from insufficient sampling accuracy and time quantization errors when dealing with minute torsional vibration fluctuations under high-speed operating conditions, making it impossible to directly and accurately extract effective torsional vibration characteristic information from the raw sampled data.
[0005] Finally, the data processing model does not match the actual situation, introducing analytical errors. Since torsional vibration is essentially an angular velocity fluctuation, the time experienced by the shaft system when rotating at the same angular interval is not equal. Therefore, the signal acquired based on angle-triggered mechanisms such as speed-measuring gears is essentially a non-uniform sampling sequence on the time axis. Ignoring this physical nature and directly performing conventional spectral analysis (such as Fast Fourier Transform) on the original acquired time-domain signal based on the assumption of uniform sampling will introduce spectral leakage and frequency identification errors, affecting the accuracy of torsional vibration characteristic frequency and amplitude analysis.
[0006] Therefore, this invention proposes a method and system for monitoring torsional vibration of shaft systems in equipment operating at constant speeds. Summary of the Invention
[0007] The purpose of this invention is to provide a method and system for monitoring torsional vibration of shaft systems in equipment operating at constant speeds. Without changing the existing speed measurement hardware architecture, the accuracy of data at a fixed sampling rate is improved through algorithm optimization, thereby achieving a low-cost and highly reliable torsional vibration monitoring solution.
[0008] 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 monitoring torsional vibration of shaft systems in equipment operating at constant speeds, comprising the following steps: Receive the original square wave time signal sequence; Logical judgment is performed on the original square wave time signal sequence to identify the trigger edge of the signal and calculate the time interval Δt between adjacent trigger edges. The trigger edge is the time when the trigger block passes through the sensor, specifically including the rising edge and the falling edge. Based on the time interval Δt and the physical angular interval Δθ of the trigger block, the instantaneous angular velocity of the shaft system is calculated to obtain the original angular velocity sequence; The instantaneous angular velocity sequence is subjected to high-order interpolation reconstruction and filtering to obtain a smooth angular velocity fluctuation signal; The angular velocity fluctuation signal is integrated to obtain the torsional angular displacement signal, and its peak characteristic index is extracted. The peak-to-peak value characteristic index is compared with a preset safety threshold, and a torsional vibration risk alarm is triggered when the threshold is exceeded.
[0009] Furthermore, the original square wave time signal sequence is obtained by continuously acquiring the time sequence signal output by the speed sensor using a general-purpose fixed-frequency data acquisition card at a fixed sampling rate under the condition of stable shaft speed.
[0010] Furthermore, the fixed sampling rate is set based on the rated speed of the shaft system and the range of torsional vibration frequency, and is not less than twice the highest frequency of the output signal of the speed sensor.
[0011] Furthermore, logical judgments are performed on the original square wave time signal sequence to identify the trigger edges of the signal, and the time interval Δt between adjacent trigger edges is calculated, as follows: (41) For the original square wave signal Discrete sequences obtained through sampling sampling frequency , To eliminate high-frequency noise, a digital low-pass filter is first applied to the signal's fundamental frequency. in This represents an FIR filter with linear phase characteristics, and its cutoff frequency. ; (42) Establish a dual-threshold hysteresis window: in This represents the high threshold of square wave amplitude. This represents the low threshold of square wave amplitude. , The normalized threshold coefficient satisfies To ensure noise tolerance; (43) Define state variables ,initialization State transition condition: Falling edge event detection flag: in This is the state transition condition for state number n. This represents a falling edge event detection flag; when At that time, record discrete indexes satisfy: in This is the filtered square wave. For discrete indexes, Low threshold for square wave amplitude (44) At the sampling point and Perform first-order linear interpolation between them to calculate the threshold time. : in The amplitude of the filtered square wave. For square wave amplitude low threshold, The sampling period is ; T Calculate the time interval between adjacent falling edges : In the formula, For falling edge intervals, Indicates the first Each sampling point timestamp.
[0012] Furthermore, the instantaneous angular velocity ω of the shaft system is calculated as follows: ω=Δθ / Δt In the formula, Δθ is the physical angular interval of the trigger block, and Δt is the time interval between adjacent trigger edges.
[0013] Furthermore, the instantaneous angular velocity sequence is subjected to high-order interpolation reconstruction and filtering to obtain a smooth angular velocity fluctuation signal, as follows: (61) The original angular velocity sequence Treated as a non-uniformly sampled signal, based on the cumulative time axis Establish an irregular time grid, in which ; (62) Interpolation reconstruction improves sampling density The non-uniformly sampled sequence is resampled into a uniform high-density time series using piecewise cubic Hermite interpolation, with interpolation points... The angular velocity at that point is calculated by the following formula: in , For the derivative estimation at the nodes, the basis functions are: Set resampling frequency and ,in For frequency conversion, This is the highest analysis frequency for torsional vibration, used to increase sampling density to obtain high time resolution sequences. ; (63) Digital filtering improves waveform stability High-density sequences after resampling Bandpass digital filtering is performed using an IIR Butterworth filter, whose transfer function is: Where z is the complex frequency variable in the Z-transform. Delay k sampling points, filter coefficients The lower cutoff frequency is determined by the bilinear transform method from the analog prototype filter. Used to suppress low-frequency trends, upper cutoff frequency. To combat aliasing, the inverse Z-transform yields a difference equation, which, after filtering, produces a smoothed angular velocity fluctuation signal. : In the formula, Δω[m] is the filtered angular velocity fluctuation signal, N is the filter order, m is the discrete time index, ω_dense[m−k] is the angular velocity value of the high-density sequence at time m−k, and Δω[m−k] is the filtered angular velocity fluctuation signal at time m−k.
[0014] Furthermore, the angular velocity fluctuation signal is integrated to obtain the torsional angular displacement signal, and its peak-to-peak value characteristic index is extracted, as follows: (71) For smooth angular velocity fluctuation signals Numerical integration is performed, and the angular displacement is obtained using the cumulative trapezoidal integration method. : in The m-th amplitude of the angular velocity fluctuation signal. To achieve uniform resampling time steps, initial conditions ; (72) Perform polynomial baseline correction and fit the k-th order polynomial trend term using the least squares method. : Where the coefficient By minimizing The corrected torsional angular displacement is obtained. : Extracting peak characteristic indices of torsional vibration angular displacement : In the formula, This is a peak-to-peak value characteristic index for torsional vibration angular displacement signals. This represents the maximum value in the corrected torsional angular displacement sequence. This represents the minimum value in the corrected torsional angular displacement sequence.
[0015] According to a second aspect of the present invention, the present invention provides a torsional vibration monitoring system for shaft systems of equipment operating at constant speeds, for implementing the torsional vibration monitoring method for shaft systems of equipment operating at constant speeds described in the first aspect, comprising: The signal acquisition module is used to continuously acquire the time-series signal output by the speed sensor at a fixed sampling rate to obtain the original square wave time signal sequence; The data processing module is used to perform logical judgments on the original square wave time signal sequence, identify the trigger edge of the signal, and calculate the time interval Δt between adjacent trigger edges. The trigger edge is the time when the trigger block passes through the sensor, specifically including the rising edge and the falling edge. The angular velocity calculation module is used to calculate the instantaneous angular velocity of the shaft system based on the time interval Δt and the physical angular interval Δθ of the trigger block, and obtain the original angular velocity sequence. The signal enhancement module is used to perform high-order interpolation reconstruction and filtering on the instantaneous angular velocity sequence to obtain a smooth angular velocity fluctuation signal. The torsional vibration calculation module is used to perform integral calculation on the angular velocity fluctuation signal to obtain the torsional vibration angular displacement signal and extract its peak-to-peak value characteristic index. The status warning module is used to compare the peak-to-peak value characteristic index with a preset safety threshold, and trigger a torsional vibration risk alarm when the threshold is exceeded.
[0016] Furthermore, the signal acquisition module includes a speed measuring trigger device and a non-contact speed sensor installed on the generator set shaft system. The speed measuring trigger device is a speed measuring gear or a trigger block fixed on the shaft, and the non-contact speed sensor is a magnetoelectric or Hall effect sensor.
[0017] Furthermore, the signal enhancement module integrates a high-order interpolation algorithm to reduce time-domain quantization errors caused by a fixed sampling rate, and the signal enhancement module also has a built-in bandpass filter.
[0018] This invention has at least the following beneficial effects: 1. This invention, through deep loading of signal processing logic, replaces the reliance on high-line-count, high-cost sensors with software algorithm optimization. While maintaining the existing hardware architecture, it achieves the measurement results of standard sensors through dual-edge logic judgment and re-timing processing, providing practical and feasible hardware monitoring technical support for realizing high-performance, low-power system solutions.
[0019] 2. This invention achieves a doubling of sampling density without changing the physical scaling accuracy of the sensor hardware by synchronously identifying the rising and falling edges of the original square wave signal. This adjacent edge triggering mechanism can capture more instantaneous feature points, making the depiction of the motion state of the system trigger block more sensitive, effectively avoiding the information omissions caused by the traditional single-edge detection mode, thereby improving the overall resolution of the system at the algorithm level.
[0020] 3. This invention effectively suppresses the inherent quantization error of a fixed sampling rate by using precise trigger edge identification and sub-sampling period-level timestamp interpolation algorithms. Combined with high-order polynomial interpolation reconstruction and zero-phase bandpass filtering techniques, it can accurately separate weak torsional vibration components from the original pulse sequence, achieving precise restoration of angular velocity fluctuations.
[0021] 4. This invention directly obtains the peak angular displacement, reflecting the physical quantity of torsional deformation of the shaft system, through integral calculation. This indicator is intuitive and stable, and is directly related to the mechanical stress and fatigue accumulation of the shaft system. It provides a clear and reliable direct criterion for equipment health status assessment and risk warning, greatly improving the engineering practicality and accuracy of condition judgment. 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
[0022] Figure 1 This is a flowchart illustrating the method described in this invention. Detailed Implementation
[0023] 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.
[0024] Example 1: Please see Figure 1 This invention provides a technical solution: a method for monitoring torsional vibration of shaft systems in equipment operating at constant speeds, comprising the following steps: S1. Receive the original square wave time signal sequence; The original square wave time signal sequence was obtained by continuously acquiring the time-series signal output by the speed sensor using a general-purpose fixed-frequency data acquisition card at a fixed sampling rate under stable shaft speed conditions. The fixed sampling rate was determined based on the rated speed and torsional vibration frequency range of the shaft system, and was not less than twice the highest frequency of the speed sensor output signal. This method eliminates the need for expensive photoelectric encoders or dedicated high-speed acquisition cards, directly reusing the speed measuring gears and speed sensors already configured on-site in the generator set. Signal acquisition can be achieved through conventional hardware, significantly reducing hardware modification costs. Furthermore, the fixed sampling rate mode is fully compatible with general-purpose data acquisition cards in industrial settings, avoiding the need for customized acquisition equipment and facilitating large-scale deployment in existing generator sets. It should be further explained that the original square wave time signal sequence refers to the digital pulse sequence generated by the alternating passage of the tooth tip and tooth valley of the speed measuring gear in response to the stable rotational speed of the shaft system. This sequence is characterized by the periodic transition of the logic level between "0" and "1". Ideally, each transition corresponds to the geometric moment when the mechanical cross section of the trigger block passes through the sensor's sensitive axis. S2. Perform logical judgment on the original square wave time signal sequence, identify the trigger edge of the signal, and calculate the time interval Δt between adjacent trigger edges. The trigger edge is the moment when the trigger block passes through the sensor, specifically including the rising edge and the falling edge. The specific steps are as follows: Taking the falling edge as an example For the original square wave signal Discrete sequences obtained through sampling sampling frequency ( (Assuming the signal's fundamental frequency), first implement digital low-pass filtering to eliminate high-frequency noise: in This represents an FIR filter with linear phase characteristics, and its cutoff frequency. ; Establish a dual-threshold hysteresis window: in , The normalized threshold coefficient satisfies To ensure noise tolerance; Define state variables ,initialization (High level state), state transition condition: Falling edge event detection flag: in This is the state transition condition for state number n. This represents a falling edge event detection flag; when At the time of recording the discrete index Satisfies: At the sampling point and Perform first-order linear interpolation between them to calculate the threshold time. : in The amplitude of the filtered square wave. For square wave amplitude low threshold, The sampling period is ; T Calculate the time interval between adjacent falling edges (Period Sequence): In the formula, For falling edge intervals, Indicates the first Timestamp of each sampling point; Triggering edges specifically include two types of shapes: rising edge and falling edge. The former represents the transition process from tooth root to tooth tip, while the latter corresponds to the transition process from tooth tip to tooth root. Both physically uniquely map the angular position increment Δθ of the axis system. S3. Based on the time interval Δt and the physical angular interval Δθ of the trigger block, calculate the instantaneous angular velocity of the shaft system to obtain the original angular velocity sequence, as follows: ω=Δθ / Δt In the formula, Δθ is the physical angular interval of the trigger block, and Δt is the time interval between adjacent trigger edges; S4. Perform high-order interpolation reconstruction and filtering on the instantaneous angular velocity sequence to obtain a smooth angular velocity fluctuation signal, as detailed below: Original angular velocity sequence Treated as a non-uniformly sampled signal, based on the cumulative time axis Establish an irregular time grid, in which ; Interpolation reconstruction improves sampling density The non-uniformly sampled sequence is resampled into a uniform high-density time series using piecewise cubic Hermite interpolation, with interpolation points... The angular velocity at that point is calculated by the following formula: in , For the derivative estimation at the nodes, the basis functions are: Set resampling frequency and ,in For frequency conversion, To achieve the highest analysis frequency for torsional vibration, the sampling density was increased by at least 5 times to obtain high temporal resolution sequences. ; Digital filtering improves waveform stability High-density sequences after resampling Bandpass digital filtering is performed using an IIR Butterworth filter, whose transfer function is: Where z is the complex frequency variable in the Z-transform. This represents the filter coefficients after a delay of k sampling points. The lower cutoff frequency is determined by the bilinear transform method from the analog prototype filter. Used to suppress low-frequency trends, upper cutoff frequency. For anti-aliasing purposes, the difference equation is obtained through inverse Z-transform as follows. The filtered, smoothed angular velocity fluctuation signal is then obtained by calculating the difference equation. : In the formula, Δω[m] is the filtered angular velocity fluctuation signal, N is the filter order, m is the discrete time index, ω_dense[m−k] is the angular velocity value of the high-density sequence at time m−k, and Δω[m−k] is the filtered angular velocity fluctuation signal at time m−k. This treatment effectively eliminates noise interference and enhances the stability of torsional vibration characteristic waveforms; S5. Integrate the angular velocity fluctuation signal to obtain the torsional vibration angular displacement signal, and extract its peak-to-peak value characteristic index, as follows: For smooth angular velocity fluctuation signals Numerical integration is performed, and the angular displacement is obtained using the cumulative trapezoidal integration method. : in The m-th amplitude of the angular velocity fluctuation signal. To achieve uniform resampling time steps, initial conditions ; Polynomial baseline correction is performed, and the k-th order polynomial trend term is fitted using the least squares method. : Where the coefficient By minimizing The corrected torsional angular displacement is obtained. : Extracting peak characteristic indices of torsional vibration angular displacement : In the formula, This is a peak-to-peak value characteristic index for torsional vibration angular displacement signals. This represents the maximum value in the corrected torsional angular displacement sequence. This represents the minimum value in the corrected torsional angular displacement sequence. S6. Compare the peak-to-peak characteristic index with the preset safety threshold. When the threshold is exceeded, a torsional vibration risk alarm is triggered. This allows for the direct calculation of the peak angular displacement from the fluctuating angular velocity, which can intuitively reflect the stress and deformation of the shaft system, providing a direct basis for health assessment.
[0025] It should be noted that the preset safety threshold θ is calculated based on the shaft system specifications. θ=M L / (G I)(mrad) Where M is the upper limit of material torque safety, and G is the material shear modulus (N / m). 2 I is the moment of inertia of the axial section (m) 4 ), L-axis segment length.
[0026] Example 2: This embodiment provides a shaft torsional vibration monitoring system for constant-speed equipment, used to implement the shaft torsional vibration monitoring method for constant-speed equipment described in Embodiment 1, including: The signal acquisition module is used to continuously acquire the time-series signal output by the speed sensor at a fixed sampling rate to obtain the original square wave time signal sequence; The data processing module is used to perform logical judgments on the original square wave time signal sequence, identify the trigger edge of the signal, and calculate the time interval Δt between adjacent trigger edges. The trigger edge is the time when the trigger block passes through the sensor, specifically including the rising edge and the falling edge. The angular velocity calculation module is used to calculate the instantaneous angular velocity of the shaft system based on the time interval Δt and the physical angular interval Δθ of the trigger block, and obtain the original angular velocity sequence. The signal enhancement module is used to perform high-order interpolation reconstruction and filtering on the instantaneous angular velocity sequence to obtain a smooth angular velocity fluctuation signal. The torsional vibration calculation module is used to perform integral calculation on the angular velocity fluctuation signal to obtain the torsional vibration angular displacement signal and extract its peak-to-peak value characteristic index. The status warning module is used to compare the peak-to-peak value characteristic index with a preset safety threshold, and trigger a torsional vibration risk alarm when the threshold is exceeded.
[0027] Regarding the technical solution of this embodiment, the signal acquisition module includes a speed-measuring trigger device and a non-contact speed sensor installed on the generator set shaft. The speed-measuring trigger device is a speed-measuring gear or a trigger block fixed on the shaft. The non-contact speed sensor is a magnetoelectric or Hall effect sensor. This signal acquisition module configuration, through the combination of the speed-measuring gear or shaft-mounted trigger block and the non-contact speed sensor, achieves full reuse of the existing hardware of the generator set. The speed-measuring gear, as a standard component of rotating machinery, provides an angular position reference without the need for an additional precision encoder, significantly reducing modification costs and installation complexity. The magnetoelectric or Hall effect sensor adopts a non-contact measurement principle, avoiding mechanical wear and additional dynamic loads on the shaft. It exhibits long-term stability in harsh industrial environments such as oil and dust, and its standard industrial interface has good electrical compatibility with general-purpose data acquisition cards, facilitating large-scale deployment in existing units, thus achieving a balance between reliability, economy, and ease of engineering implementation.
[0028] Regarding the technical solution of this embodiment, the signal enhancement module integrates a high-order interpolation algorithm to reduce time-domain quantization errors caused by a fixed sampling rate. The signal enhancement module also incorporates a bandpass filter. The high-order interpolation algorithm addresses the time-domain quantization errors present when acquiring square wave signals at a fixed sampling rate, reconstructing and smoothing the original non-uniform angular velocity sequence, significantly improving the signal's temporal resolution and continuity, thus laying a data foundation for subsequent accurate analysis. The built-in bandpass filter is designed based on the frequency characteristics of shaft torsional vibration, filtering out DC components and high-frequency noise from the signal, thereby accurately extracting the target frequency band angular velocity fluctuation component reflecting the torsional vibration state. The two are combined sequentially: first, interpolation improves signal quality and unifies the time reference; then, filtering achieves effective feature separation, jointly overcoming the limitations of low-cost hardware in terms of sampling accuracy and anti-interference capabilities.
[0029] 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.
[0030] 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.
[0031] 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.
[0032] 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. A method for monitoring torsional vibration of shaft systems in equipment operating at constant speeds, characterized in that, Includes the following steps: Receive the original square wave time signal sequence; Logical judgment is performed on the original square wave time signal sequence to identify the trigger edge of the signal and calculate the time interval Δt between adjacent trigger edges. The trigger edge is the time when the trigger block passes through the sensor, specifically including the rising edge and the falling edge. Based on the time interval Δt and the physical angular interval Δθ of the trigger block, the instantaneous angular velocity of the shaft system is calculated to obtain the original angular velocity sequence; The instantaneous angular velocity sequence is subjected to high-order interpolation reconstruction and filtering to obtain a smooth angular velocity fluctuation signal; The angular velocity fluctuation signal is integrated to obtain the torsional angular displacement signal, and its peak characteristic index is extracted. The peak-to-peak value characteristic index is compared with a preset safety threshold, and a torsional vibration risk alarm is triggered when the threshold is exceeded.
2. The method for monitoring torsional vibration of shaft systems in equipment at constant speed according to claim 1, characterized in that: The original square wave time signal sequence was obtained by continuously acquiring the time sequence signal output by the speed sensor using a general-purpose fixed-frequency data acquisition card at a fixed sampling rate under the stable speed condition of the shaft system.
3. The method for monitoring torsional vibration of shaft systems in equipment at constant speed according to claim 2, characterized in that: The fixed sampling rate is set based on the rated speed of the shaft system and the range of torsional vibration frequency, and is not less than twice the highest frequency of the output signal of the speed sensor.
4. The method for monitoring torsional vibration of shaft systems in equipment at constant speed according to claim 1, characterized in that: The original square wave time signal sequence is logically judged to identify the trigger edge of the signal, and the time interval Δt between adjacent trigger edges is calculated, as follows: (41) For the original square wave signal Discrete sequences obtained through sampling sampling frequency , To eliminate high-frequency noise, a digital low-pass filter is first applied to the signal's fundamental frequency. in This represents an FIR filter with linear phase characteristics, and its cutoff frequency. ; (42) Establish a dual-threshold hysteresis window: in This represents the high threshold of square wave amplitude. This represents the low threshold of square wave amplitude. , The normalized threshold coefficient satisfies To ensure noise tolerance; (43) Define state variables ,initialization State transition condition: Falling edge event detection flag: in This is the state transition condition for state number n. This represents a falling edge event detection flag; when At that time, record discrete indexes satisfy: in This is the filtered square wave. For discrete indexes, for Low threshold; (44) At the sampling point and Perform first-order linear interpolation between them to calculate the threshold time. : in The amplitude of the filtered square wave. For square wave amplitude low threshold, The sampling period is ; T Calculate the time interval between adjacent falling edges : In the formula, For falling edge intervals, Indicates the first Each sampling point timestamp.
5. The method for monitoring torsional vibration of shaft systems in equipment at constant speed according to claim 4, characterized in that: The instantaneous angular velocity of the shaft system is calculated as follows: ω=Δθ / Δt In the formula, Δθ is the physical angular interval of the trigger block, and Δt is the time interval between adjacent trigger edges.
6. The method for monitoring torsional vibration of shaft systems in equipment at constant speed according to claim 5, characterized in that: The instantaneous angular velocity sequence is subjected to high-order interpolation reconstruction and filtering to obtain a smooth angular velocity fluctuation signal, as follows: (61) The original angular velocity sequence Treated as a non-uniformly sampled signal, based on the cumulative time axis Establish an irregular time grid, in which ; (62) Interpolation reconstruction improves sampling density The non-uniformly sampled sequence is resampled into a uniform high-density time series using piecewise cubic Hermite interpolation, with interpolation points... The angular velocity at that point is calculated by the following formula: in , For the derivative estimation at the nodes, the basis functions are: Set resampling frequency and ,in For frequency conversion, This is the highest analysis frequency for torsional vibration, used to increase sampling density to obtain high time resolution sequences. ; (63) Digital filtering improves waveform stability High-density sequences after resampling Bandpass digital filtering is performed using an IIR Butterworth filter, whose transfer function is: : Where z is the complex frequency variable in the Z-transform. Delay k sampling points, filter coefficients The lower cutoff frequency is determined by the bilinear transform method from the analog prototype filter. Used to suppress low-frequency trends, upper cutoff frequency. For anti-aliasing purposes, the difference equation is obtained through inverse Z-transform as follows. The filtered, smoothed angular velocity fluctuation signal is then obtained by calculating the difference equation. : In the formula, Δω[m] is the filtered angular velocity fluctuation signal, N is the filter order, m is the discrete time index, ω_dense[m−k] is the angular velocity value of the high-density sequence at time m−k, and Δω[m−k] is the filtered angular velocity fluctuation signal at time m−k.
7. The method for monitoring torsional vibration of shaft systems in equipment at constant speed according to claim 6, characterized in that: The angular velocity fluctuation signal is integrated to obtain the torsional angular displacement signal, and its peak-to-peak value characteristic index is extracted, as follows: (71) For smooth angular velocity fluctuation signals Numerical integration is performed, and the angular displacement is obtained using the cumulative trapezoidal integration method. : in The m-th amplitude of the angular velocity fluctuation signal. To achieve uniform resampling time steps, initial conditions ; (72) Perform polynomial baseline correction and fit the k-th order polynomial trend term using the least squares method. : Where the coefficient By minimizing Let m be the index, k be the order, and j be the summation index, then the corrected torsional angular displacement is obtained. : Extracting peak characteristic indices of torsional vibration angular displacement : In the formula, This is a peak-to-peak value characteristic index for torsional vibration angular displacement signals. This represents the maximum value in the corrected torsional angular displacement sequence. This represents the minimum value in the corrected torsional angular displacement sequence.
8. A torsional vibration monitoring system for shaft systems of equipment operating at constant speed, used to implement the torsional vibration monitoring method for shaft systems of equipment operating at constant speed as described in any one of claims 1 to 7, characterized in that, include: The signal acquisition module is used to continuously acquire the time-series signal output by the speed sensor at a fixed sampling rate to obtain the original square wave time signal sequence; The data processing module is used to perform logical judgments on the original square wave time signal sequence, identify the trigger edge of the signal, and calculate the time interval Δt between adjacent trigger edges. The trigger edge is the time when the trigger block passes through the sensor, specifically including the rising edge and the falling edge. The angular velocity calculation module is used to calculate the instantaneous angular velocity of the shaft system based on the time interval Δt and the physical angular interval Δθ of the trigger block, and obtain the original angular velocity sequence. The signal enhancement module is used to perform high-order interpolation reconstruction and filtering on the instantaneous angular velocity sequence to obtain a smooth angular velocity fluctuation signal. The torsional vibration calculation module is used to perform integral calculation on the angular velocity fluctuation signal to obtain the torsional vibration angular displacement signal and extract its peak-to-peak value characteristic index. The status warning module is used to compare the peak-to-peak value characteristic index with a preset safety threshold, and trigger a torsional vibration risk alarm when the threshold is exceeded.
9. A shaft torsional vibration monitoring system for a constant-speed equipment according to claim 8, characterized in that: The signal acquisition module includes a speed measuring trigger device and a non-contact speed sensor installed on the generator set shaft. The speed measuring trigger device is a speed measuring gear or a trigger block fixed on the shaft, and the non-contact speed sensor is a magnetoelectric or Hall sensor.
10. A method for monitoring torsional vibration of shaft systems in equipment at constant speed, as described in claim 9, characterized in that: The signal enhancement module integrates a high-order interpolation algorithm to reduce time-domain quantization errors caused by a fixed sampling rate, and also has a built-in bandpass filter.