A steel structure machining platform structure fault diagnosis method and system

Abstract

The present application relates to the technical field of steel structure machining platform structure fault diagnosis, and particularly relates to a steel structure machining platform structure fault diagnosis method and system, the method comprising the following steps: in the platform non-production period, a preset excitation is applied to the platform structure through a vibration suppression device, and excitation response data of the platform structure to the preset excitation is collected; according to the excitation response data, the performance state of the vibration suppression device is determined, and the change trend of the performance state is recorded; during the platform operation, the operation response data of the platform structure is continuously collected, and the abnormality in the operation response data is identified to obtain abnormal characteristics; the abnormal characteristics are compared with the change trend of the performance state to determine whether the abnormality is caused by the performance change of the vibration suppression device; the influence of the known interference is eliminated from the operation response data, and the damage characteristics of the platform main structure are identified from the operation response data after the known interference is eliminated. The present application improves the accuracy and reliability of fault diagnosis.

Description

Technical Field

[0001] This invention relates to the technical field of structural fault diagnosis for steel structure processing platforms, and specifically to a method and system for structural fault diagnosis of steel structure processing platforms. Background Technology

[0002] In modern heavy equipment manufacturing plants, large steel structure processing platforms are key equipment for precision machining. During long-term service, they must withstand enormous static loads and alternating loads such as impacts and vibrations generated during processing. These continuous mechanical and thermal stresses can lead to potential failures such as micro-cracks within the steel structure, loosening of connectors, or abnormal stress distribution. Traditional detection methods struggle to detect these early, hidden structural anomalies in real time, failing to provide early warnings and posing potential risks to production safety and machining accuracy. Summary of the Invention

[0003] The purpose of this invention is to address the aforementioned shortcomings by proposing a method and system for diagnosing structural faults in steel structure processing platforms.

[0004] A method for diagnosing structural faults in a steel structure processing platform, the method comprising the following steps:

[0005] During non-production periods of the platform, a preset excitation is applied to the platform structure through a vibration suppression device, and the excitation response data of the platform structure to the preset excitation is collected.

[0006] Based on the excitation response data, determine the performance status of the vibration suppression device and record the trend of performance status changes;

[0007] During platform operation, continuous collection of operational response data of the platform structure is performed, and anomalies in the operational response data are identified to derive anomaly characteristics;

[0008] Compare the abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from a change in the performance of the vibration suppression device.

[0009] When the anomaly is not determined to be caused by a change in the performance of the vibration suppression device, the influence of known interference is eliminated from the operational response data, and the damage characteristics of the main structure of the platform are identified from the operational response data after the known interference is eliminated.

[0010] Known disturbances include transient thermal stress, environmental noise, power fluctuations, and periodic vibrations caused by specific processing conditions.

[0011] This technical solution effectively distinguishes between platform main structure failures, vibration suppression device performance degradation, and other known interferences, avoiding misjudgments and omissions caused by multiple factors in traditional diagnostic methods, and significantly improving the accuracy and reliability of fault diagnosis.

[0012] This application also discloses a structural fault diagnosis system for a steel structure processing platform, applied to the aforementioned structural fault diagnosis method for a steel structure processing platform. The system includes:

[0013] The data acquisition module applies a preset excitation to the platform structure through a vibration suppression device during non-production periods and collects the excitation response data of the platform structure to the preset excitation.

[0014] The recording module determines the performance status of the vibration suppression device based on the excitation response data and records the trend of performance status changes.

[0015] The identification module continuously collects operational response data of the platform structure during platform operation, identifies anomalies in the operational response data, and derives anomaly characteristics.

[0016] The judgment module compares the abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from the performance change of the vibration suppression device.

[0017] The processing module, when the anomaly is not determined to be caused by a change in the performance of the vibration suppression device, eliminates the influence of known interference from the operation response data and identifies the damage characteristics of the main structure of the platform from the operation response data after eliminating known interference.

[0018] Known disturbances include transient thermal stress, environmental noise, power fluctuations, and periodic vibrations caused by specific processing conditions.

[0019] This application provides a system capable of implementing the aforementioned fault diagnosis method. Through modular design, it achieves comprehensive monitoring and analysis of platform structure, vibration suppression device performance, and environmental interference, providing hardware and software support for the safe operation of the steel structure processing platform.

[0020] This application can significantly improve the accuracy and reliability of fault diagnosis, avoid false alarms or missed alarms caused by multiple factors in traditional diagnostic systems, and provide a more reliable guarantee for the safe operation and precision machining of large steel structure processing platforms.

[0021] To further understand the features and technical content of the present invention, please refer to the following detailed description and accompanying drawings. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0022] Figure 1 This is a flowchart of a method for diagnosing structural faults in a steel structure processing platform according to the present invention;

[0023] Figure 2This is a structural schematic diagram of a steel structure processing platform fault diagnosis system according to the present invention. Detailed Implementation

[0024] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0025] This embodiment provides a method and system for diagnosing structural faults in a steel structure processing platform, combined with... Figure 1 and Figure 2 As shown.

[0026] refer to Figure 1 A method for diagnosing structural faults in a steel structure processing platform, the method comprising the following steps:

[0027] During non-production periods of the platform, a preset excitation is applied to the platform structure through a vibration suppression device, and the excitation response data of the platform structure to the preset excitation is collected.

[0028] Based on the excitation response data, determine the performance status of the vibration suppression device and record the trend of performance status changes;

[0029] During platform operation, continuous collection of operational response data of the platform structure is performed, and anomalies in the operational response data are identified to derive anomaly characteristics;

[0030] Compare the abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from a change in the performance of the vibration suppression device.

[0031] When the anomaly is not determined to be caused by a change in the performance of the vibration suppression device, the influence of known interference is eliminated from the operational response data, and the damage characteristics of the main structure of the platform are identified from the operational response data after the known interference is eliminated.

[0032] Known disturbances include transient thermal stress, environmental noise, power fluctuations, and periodic vibrations caused by specific processing conditions.

[0033] Among them, "platform structure" usually refers to the main load-bearing structure of the steel structure processing platform, including beams, columns, connectors, etc., which is the basis for processing operations.

[0034] "Vibration suppression devices" refer to equipment installed on the platform structure to actively or passively reduce platform vibration, such as active mass dampers and piezoelectric actuators. Their purpose is to counteract or weaken harmful vibrations caused by the machining process, motor operation, etc., during platform operation by generating a counterforce or damping, thereby ensuring machining accuracy and equipment lifespan.

[0035] "Preset excitation" refers to a known and controllable excitation signal applied to the platform structure by the vibration suppression device during non-production periods. This signal may be a sine wave of a specific frequency and amplitude, a swept frequency signal, or a random excitation. The purpose of this excitation is to stimulate the response of the platform structure under controlled conditions in order to evaluate the performance of the vibration suppression device and the baseline state of the platform structure.

[0036] "Excitation response data" refers to the dynamic response data of the platform structure collected by sensors (such as accelerometers and strain gauges) after the platform structure is subjected to a preset excitation. These data reflect the vibration characteristics, stiffness, damping, and other information of the platform structure under specific excitation.

[0037] "Operational response data" refers to the dynamic response data of the platform structure continuously collected during normal production operation. This data includes the platform's comprehensive response under actual workload, environmental interference, and the impact of potential failures.

[0038] "Abnormal features" refer to characteristics identified from operational response data that deviate significantly from the response pattern under normal platform operation, such as sudden changes in vibration amplitude, abnormal frequency components, and changes in damping characteristics.

[0039] "Known disturbances" refer to other factors that may affect the operational response data during platform operation, besides damage to the main platform structure and changes in the performance of vibration suppression devices. These factors include thermal deformation caused by changes in ambient temperature and periodic vibrations caused by specific working conditions during processing.

[0040] The core of the structural fault diagnosis method for steel structure processing platforms proposed in this application lies in achieving accurate identification of platform structural faults through phased and multi-dimensional data analysis.

[0041] During non-production periods on the platform, a preset excitation is applied to the platform structure using a vibration suppression device, and the excitation response data of the platform structure to the preset excitation is collected. The purpose of this step is to obtain the response of the platform structure under known excitation under controlled conditions, so as to subsequently evaluate the performance of the vibration suppression device. For example, a piezoelectric actuator can be used as the vibration suppression device, and its output of a sinusoidal excitation with a specific frequency and amplitude can be controlled by a function generator. At the same time, accelerometers are placed at key locations on the platform structure to collect the vibration acceleration signal of the platform structure when it is excited. As a preferred embodiment, an electromagnetic exciter is used as the vibration suppression device, which is driven by a power amplifier to generate a swept frequency excitation, and the displacement response data of the platform structure is collected using a laser displacement sensor.

[0042] Based on the excitation response data, the performance status of the vibration suppression device is determined, and the trend of its performance status is recorded. This step aims to evaluate the operational capability of the vibration suppression device in its current state. For example, the transfer function between the output force of the vibration suppression device at a specific frequency and the platform structure response can be calculated by analyzing the excitation response data. If the transfer function deviates significantly from the design baseline value, it indicates that the performance of the vibration suppression device may be degrading. As a preferred implementation, the suppression effect of the vibration suppression device on the platform structure vibration is analyzed by comparing the excitation response data collected at different time points, for example, by calculating the vibration energy attenuation rate before and after vibration suppression. If the attenuation rate continues to decrease, it indicates that the performance of the vibration suppression device is degrading. These quantitative indicators of performance status, such as transfer function deviation and suppression effect attenuation rate, will be recorded to form a trend of performance status changes.

[0043] During platform operation, operational response data of the platform structure is continuously collected, and anomalies in the operational response data are identified to derive abnormal characteristics. This step aims to monitor the operational status of the platform structure in real time. For example, multiple fiber optic grating sensors can be installed on the platform structure to collect strain data during operation. By monitoring this strain data in real time, anomalies can be identified once the strain amplitude exceeds a preset threshold, the strain frequency components change significantly, or the statistical characteristics of the strain signal (such as mean and variance) are abnormal. As a preferred implementation, a non-contact laser vibrometer is used to continuously monitor the vibration velocity at key points of the platform structure. By performing spectral analysis on the vibration velocity signal, abnormal frequency components or harmonic enhancement phenomena that do not conform to the normal operating mode are identified, thereby deriving abnormal characteristics.

[0044] The abnormal characteristics are compared with the trend of performance status changes to determine whether the abnormality originates from a performance change in the vibration suppression device. This step is one of the key innovations of this application, aiming to distinguish between platform main structure failures and abnormalities caused by the degradation of the vibration suppression device itself. For example, if the identified abnormal characteristic is an increase in vibration at a specific frequency, and the performance status change trend of the vibration suppression device shows a significant decrease in its suppression capability within that frequency range, then it can be preliminarily determined that the abnormality may originate from a performance change in the vibration suppression device. As a preferred implementation, a correlation model is established between the performance degradation of the vibration suppression device and the abnormal response of the platform structure. When the identified abnormal characteristics highly match the abnormal pattern predicted by the model as caused by the degradation of the vibration suppression device, it is determined that the abnormality originates from a performance change in the vibration suppression device.

[0045] When the anomaly is not determined to originate from a performance change in the vibration suppression device, the influence of known interferences is eliminated from the operational response data, and damage characteristics of the platform's main structure are identified from the operational response data after eliminating known interferences. This step aims to further focus on the actual damage to the platform's main structure. For example, if it is known that the platform will undergo periodic thermal deformation under specific processing conditions, the response component caused by this thermal deformation can be subtracted from the operational response data using a pre-established thermal deformation model. As a preferred implementation, an adaptive filtering algorithm is used to identify and eliminate known interference signals caused by environmental noise, power fluctuations, etc., in the operational response data. After eliminating these known interferences, the remaining operational response data is further analyzed, for example, by using modal parameter identification technology to identify changes in modal parameters such as the platform's main structure's natural frequencies, damping ratios, and mode shapes. If these modal parameters change significantly, it indicates that the platform's main structure may be damaged, and damage characteristics, such as decreased stiffness, increased damping, or local mode shape distortion, can be identified based on the changing patterns.

[0046] The structural fault diagnosis method for steel structure processing platforms proposed in this application works by systematically addressing the problem of traditional diagnostic methods' inability to distinguish different sources of anomalies in complex industrial environments through multi-stage and multi-dimensional analysis. First, during non-production periods on the platform, by applying a preset excitation to the vibration suppression device and collecting excitation response data, a performance baseline for the vibration suppression device can be established, and its performance status can be continuously tracked. This step is crucial because it provides key reference information for subsequent differentiation of anomaly sources.

[0047] During normal platform operation, the system continuously collects operational response data of the platform structure and identifies abnormal features in real time. These abnormal features may be caused by various factors, including damage to the main platform structure, performance degradation of the vibration suppression device, and other known disturbances. To accurately locate the fault source, this application compares the identified abnormal features with pre-recorded trends in the performance status of the vibration suppression device. If the abnormal features closely match the performance degradation trend of the vibration suppression device, it can be determined that the anomaly mainly originates from a problem with the vibration suppression device itself, thereby avoiding misjudging the degradation of the vibration suppression device as damage to the main platform structure.

[0048] When an anomaly is not determined to originate from a performance change in the vibration suppression device, the system further eliminates the influence of known disturbances from the operational response data. These known disturbances may include thermal deformation caused by changes in ambient temperature, periodic vibrations generated under specific operating conditions during processing, etc. By eliminating these known disturbances, the remaining operational response data can more purely reflect the true state of the platform's main structure. Finally, damage characteristics of the platform's main structure are identified from the operational response data after eliminating known disturbances. This series of steps forms a logically rigorous and progressive diagnostic process, ensuring accurate and reliable identification of the fault type and source of the steel structure processing platform in complex and ever-changing environments.

[0049] The following is a detailed analysis of known interferences, specifically those factors that do not originate from damage to the platform's main structure or changes in the performance of vibration suppression devices, but significantly affect the platform's structural operational response data, and whose mechanisms or characteristics can be identified, modeled, or predicted in advance. If these interference signals are not eliminated, they may be confused with actual structural damage signals, leading to misdiagnosis or missed diagnosis by the diagnostic system.

[0050] The known interferences mainly include the following aspects:

[0051] First, and most prominent, known disturbances are the effects of transient thermal stress.

[0052] Due to their large size and material properties, steel structure processing platforms are highly sensitive to changes in ambient temperature. Inside a factory, ambient temperature can fluctuate drastically and frequently. For example, daytime production activities or the start-up of nearby high-heat-generating equipment (such as heat treatment furnaces or electric arc furnaces) can cause temperatures to rise, while temperatures can drop sharply at night or during shutdowns. This non-uniform temperature variation leads to non-uniform thermal expansion and contraction in different parts of the platform structure, inducing transient thermal stresses and strains within the structure. The amplitude of these thermal signals can be considerable and exhibit complex time-varying characteristics. They are superimposed on the strain generated by mechanical loads and the stresses remaining from the structural manufacturing process, resulting in a large number of temperature-related components in the operational response data collected by sensors.

[0053] For example, when a region of the platform rapidly heats up due to a nearby heat source, that region expands, generating localized stress. This may manifest as changes in strain or vibration characteristics in sensor data. This change is not due to cracks or loose connections in the structure itself, but rather a normal physical response caused by temperature. If this is not eliminated, the diagnostic system may misinterpret a rapid localized thermal expansion event as abnormal stress concentration or structural damage, thus triggering unnecessary alarms.

[0054] Secondly, environmental noise is also a common known source of interference.

[0055] In industrial production environments, the operation of various equipment (such as fans, pumps, compressors, and cranes) generates continuous background noise. This noise propagates through the air or structures and is collected by sensors on the platform structure. Although this noise is usually random or has specific frequency characteristics, it can mix into the platform structure's operational response data, reducing the signal-to-noise ratio and affecting the identification of the actual structural response.

[0056] For example, the low-frequency vibrations generated by a large ventilation system running continuously in a workshop may be detected by an accelerometer on the platform. This vibration is inherent to the environment and does not indicate damage to the platform structure, but it can mask or interfere with weak damage signals caused by factors such as the propagation of microcracks.

[0057] Furthermore, power fluctuations may also introduce known interference.

[0058] Steel structure fabrication platforms are typically equipped with complex electrical systems, including drive motors and control systems. Fluctuations in power supply voltage or current, especially during the start-up and shutdown of large equipment or changes in grid load, can introduce noise or spurious signals into sensor signal lines or data acquisition systems through electromagnetic coupling or direct conduction.

[0059] For example, when a high-power motor starts, the instantaneous current surge may cause the power supply voltage to drop, which in turn affects the power supply of sensors or signal transmission, generating transient electromagnetic interference, which manifests as spikes or baseline drift in the operating response data.

[0060] Finally, periodic vibrations caused by specific processing conditions also fall under the category of known disturbances.

[0061] When performing cutting, welding, milling, and other operations on a steel structure processing platform, different processing techniques and workpiece characteristics will produce specific, predictable periodic vibrations. These vibrations are inherent products of the processing and not a direct manifestation of damage to the platform structure.

[0062] For example, during high-speed milling operations, the meshing frequency between the tool and the workpiece generates specific high-frequency vibrations. This vibration is part of the normal machining process, but its amplitude can be large and easily misinterpreted as a structural anomaly. By pre-establishing benchmark response models for different machining conditions, the effects of these periodic vibrations can be identified and eliminated.

[0063] In summary, "known disturbances" encompass a variety of predictable or modelable non-destructive factors, including transient thermal stress, environmental noise, power fluctuations, and periodic vibrations caused by specific processing conditions. By eliminating the influence of these known disturbances during fault diagnosis, operational response data can more purely reflect the true health status of the platform's main structure, thereby significantly improving the accuracy and reliability of fault diagnosis.

[0064] The structural fault diagnosis method for steel structure processing platforms proposed in this application has significant advantages and innovations compared to existing technologies. Traditional diagnostic systems, when faced with the harsh industrial environment in which steel structure processing platforms operate, often struggle to distinguish between signal anomalies caused by transient thermal effects, degradation of the active vibration suppression system itself, and actual faults in the platform's main structure. This inherent ambiguity leads to high false positive and false negative rates, severely impacting the reliability of the diagnostic system.

[0065] The core innovation of this application lies in the introduction of an independent assessment and trend recording mechanism for the performance status of vibration suppression devices. By applying preset excitations to the vibration suppression devices during non-production periods on the platform and collecting excitation response data to determine their performance status and record their changing trends, this application can establish a "health record" for the vibration suppression devices. During platform operation, when an anomaly is detected, the system first compares the abnormal characteristics with the performance change trend of the vibration suppression devices. This comparative analysis is not available in existing technologies; it can effectively distinguish between anomalies caused by the performance degradation of the vibration suppression devices themselves and anomalies caused by failures in the main structure of the platform. For example, if a decrease in the suppression capability of the vibration suppression devices leads to an increase in vibration at a specific frequency, and this increase is consistent with the recorded performance degradation trend, then the system will not misjudge it as damage to the main structure of the platform.

[0066] Furthermore, this application also considers eliminating the influence of known interferences. When an anomaly is not determined to originate from a performance change in the vibration suppression device, the system eliminates the influence of known interferences (such as thermal deformation) from the operational response data, thereby more purely identifying the damage characteristics of the platform's main structure. This step ensures that, after excluding the degradation of the vibration suppression device and known environmental interferences, the identified damage characteristics can more accurately reflect the true health condition of the platform's main structure.

[0067] In summary, this application constructs a more comprehensive and accurate fault diagnosis framework by introducing innovative steps such as performance evaluation and trend recording of vibration suppression devices, comparison of abnormal characteristics and performance trends, and elimination of known interference. This method can effectively solve the diagnostic ambiguity problem caused by the coupling of multiple abnormal sources in complex industrial environments, which is a problem of traditional diagnostic systems. It significantly improves the accuracy and reliability of fault diagnosis for steel structure processing platforms, providing solid technical support for ensuring production safety and processing accuracy.

[0068] This application further proposes the following steps for determining the performance status of the vibration suppression device based on excitation response data and recording the trend of performance status changes:

[0069] During non-production periods on the platform, a preset excitation is applied to the platform structure through a vibration suppression device, and self-feedback data inside the vibration suppression device is collected simultaneously. The self-feedback data includes the actual excitation output information of the vibration suppression device.

[0070] Based on the actual excitation output information, the original excitation output deviation of the vibration suppression device is quantified;

[0071] The excitation response data is compensated based on the original excitation output deviation to obtain the compensated excitation response data.

[0072] Based on the compensated excitation response data, the performance status of the vibration suppression device is determined, and the trend of performance status change is recorded.

[0073] Specifically, during non-production periods on the platform, when the vibration suppression device applies a preset excitation to the platform structure, it simultaneously collects self-feedback data from within the device. This self-feedback data can be understood as data directly acquired by the sensors or controllers within the vibration suppression device, characterizing its own operating state and output characteristics. Its purpose is to obtain the true output of the vibration suppression device. In practical applications, the self-feedback data specifically refers to the actual excitation output information of the vibration suppression device. For example, it may include real-time measurements of physical quantities such as force, displacement, and acceleration applied by the device, or control parameters such as drive current and voltage. Its purpose is to accurately reflect the actual operating state of the vibration suppression device when the preset excitation is applied.

[0074] Furthermore, based on the actual excitation output information, the original excitation output deviation of the vibration suppression device is quantified. The original excitation output deviation refers to the difference between the actual excitation output information of the vibration suppression device and the preset excitation, and its purpose is to evaluate the accuracy and stability of the vibration suppression device when executing the preset excitation. For example, the original excitation output deviation may include amplitude deviation, phase deviation, frequency deviation, or nonlinear distortion, etc., which reflect the degree of deviation between the ideal output and the actual output of the vibration suppression device.

[0075] Based on this, the excitation response data is compensated for according to the original excitation output deviation, resulting in compensated excitation response data. The compensation process involves correcting the excitation response data collected from the platform structure based on the quantified original excitation output deviation. Its purpose is to eliminate or reduce the impact of the vibration suppression device's own output deviation on the excitation response data. For example, if the actual excitation output amplitude of the vibration suppression device is lower than a preset value, the excitation response data can be amplified accordingly; if phase lag exists, phase correction can be performed on the excitation response data. In this way, the compensated excitation response data can more accurately reflect the response characteristics of the platform structure under ideal preset excitation.

[0076] Finally, based on the compensated excitation response data, the performance status of the vibration suppression device is determined, and the trend of its changes is recorded. The performance status can be understood as the health condition and functional performance of the vibration suppression device under current operating conditions, aiming to provide a reliable benchmark for subsequent fault diagnosis. For example, the performance status may include parameters such as the stiffness, damping, efficiency, or response speed of the vibration suppression device. By recording the trend of its changes, potential degradation or malfunctions of the vibration suppression device can be detected in a timely manner.

[0077] The proposed solution obtains the actual excitation output information of the vibration suppression device by simultaneously collecting self-feedback data from within the device while applying a preset excitation during non-production periods on the platform. The ability to accurately quantify the original excitation output deviation of the vibration suppression device makes precise compensation processing of the excitation response data collected from the platform structure possible. This compensation effectively eliminates the influence of the vibration suppression device's own output deviation on the excitation response data, allowing the compensated excitation response data to more accurately reflect the platform structure's response characteristics under ideal excitation. Therefore, determining the performance status of the vibration suppression device based on the compensated excitation response data avoids performance evaluation errors caused by inaccurate output from the vibration suppression device itself, thereby improving the accuracy and reliability of performance status judgment.

[0078] In some preferred embodiments, it is assumed that the vibration suppression device is designed to apply a pre-set sinusoidal excitation with a frequency of 10Hz and an amplitude of 100N during non-production periods on the platform. In actual operation, the force and displacement sensors inside the vibration suppression device synchronously collect their self-feedback data. After processing, this self-feedback data yields the actual excitation output information. For example, it is found that the actual applied excitation amplitude is only 95N, with a phase lag of 5 degrees. At this point, the quantized original excitation output deviation is the amplitude deviation of 5N and the phase deviation of 5 degrees. Based on this deviation, the excitation response data collected by the accelerometers arranged on the platform structure is compensated. Specifically, the amplitude of the excitation response data can be appropriately amplified to compensate for the 5N deficiency, and the data can be corrected for a 5-degree phase lead. After compensation, the obtained compensated excitation response data will more accurately reflect the response of the platform structure under an ideal 100N, 10Hz sinusoidal excitation. Ultimately, based on these compensated excitation response data, the performance parameters of the vibration suppression device, such as stiffness and damping, can be determined more accurately, and their changing trends over time can be recorded, thus providing a more reliable basis for fault diagnosis of the platform structure.

[0079] Specifically, the steps for quantifying the original excitation output deviation of the vibration suppression device based on the actual excitation output information include:

[0080] Time-domain analysis is performed on the actual excitation output information to extract the instantaneous amplitude, rise time, and decay rate of the actual excitation output information, forming the time-domain analysis results;

[0081] Frequency domain analysis is performed on the actual excitation output information to extract the main frequency, harmonic components, and bandwidth of the actual excitation output information, thus forming the frequency domain analysis results.

[0082] By correlating the time-domain analysis results with the frequency-domain analysis results, a set of nonlinear response characteristics of the vibration suppression device under the current preset excitation is constructed;

[0083] The set of nonlinear response characteristics is compared with the linear response characteristics of the preset excitation to quantify the original excitation output deviation of the vibration suppression device.

[0084] The actual excitation output information can be understood as the detailed data on the actual output excitation fed back by the internal sensors or control system of the vibration suppression device when a preset excitation is applied. This information may include, but is not limited to, real-time measurements of physical quantities such as force, displacement, and acceleration. Time-domain analysis refers to analyzing the changes of the actual excitation output information on the time axis to obtain its dynamic characteristics. Instantaneous amplitude reflects the intensity of the excitation signal at a certain moment; rise time represents the time required for the excitation signal to rise from the starting point to the peak value, reflecting the response speed of the excitation; decay rate describes the rate at which the energy of the excitation signal dissipates after the excitation signal ends. These parameters together constitute a complete description of the excitation signal in the time dimension. Frequency-domain analysis refers to converting the actual excitation output information from the time domain to the frequency domain for analysis using methods such as Fourier transform to reveal its frequency components. The dominant frequency refers to the most important frequency component in the excitation signal, usually corresponding to its main vibration mode; harmonic components refer to frequency components that are integer multiples of the fundamental frequency, and their presence often indicates the nonlinear characteristics of the system; bandwidth represents the frequency range occupied by the excitation signal, reflecting the complexity of the signal.

[0085] Correlating time-domain and frequency-domain analysis results aims to comprehensively capture the nonlinear response characteristics of vibration suppression devices. For example, by analyzing the correspondence between the amplitude of specific frequency components and the instantaneous amplitude in the time domain, nonlinear distortion can be identified more accurately. Thus, the constructed set of nonlinear response characteristics can comprehensively reflect the actual operating state of the vibration suppression device under preset excitation. The linear response characteristics of the preset excitation refer to the theoretical response that the vibration suppression device should produce when receiving the preset excitation under ideal conditions. By comparing the actually measured set of nonlinear response characteristics with these linear response characteristics, the original excitation output deviation of the vibration suppression device can be accurately quantified. This deviation can manifest as differences in amplitude, phase, or frequency components, thereby providing a quantitative basis for evaluating the performance degradation of the vibration suppression device.

[0086] This application's solution employs multi-dimensional and refined analysis of the actual excitation output information of the vibration suppression device, enabling more accurate quantification of its original excitation output deviation. Specifically, time-domain analysis captures the transient characteristics of the excitation signal, such as instantaneous amplitude, rise time, and decay rate. These parameters are crucial for identifying the dynamic response and energy transfer efficiency of the excitation. Simultaneously, frequency-domain analysis reveals the spectral composition of the excitation signal, including the dominant frequency, harmonic components, and bandwidth. This information offers unique advantages in identifying frequency distortion and nonlinear behavior of the excitation. By correlating the results of time-domain and frequency-domain analyses, a more comprehensive set of nonlinear response characteristics can be constructed, thus avoiding information omissions or misjudgments that may result from single-dimensional analysis. Finally, by comparing this set of nonlinear response characteristics with the ideal linear response characteristics of the preset excitation, the deviations of the vibration suppression device in actual operation can be accurately identified and quantified, such as nonlinear distortion, amplitude decay, or phase lag, providing a reliable data foundation for subsequent performance status evaluation and fault diagnosis.

[0087] The above technical solution enables precise quantification of the original excitation output deviation of vibration suppression devices. Compared to relying solely on a single dimension or coarse performance indicators, this solution, by combining comprehensive analysis in the time and frequency domains and constructing a set of nonlinear response characteristics, can more comprehensively and accurately capture subtle performance changes and potential nonlinear behaviors of vibration suppression devices in actual operation. This significantly improves the accuracy and sensitivity of performance status assessment for vibration suppression devices, helps in the early detection of device performance degradation, and thus provides more reliable input data for fault diagnosis of steel structure processing platform structures, thereby enhancing the accuracy and effectiveness of the entire diagnostic method.

[0088] The original excitation output deviations mentioned above in this application include nonlinear distortion, amplitude deviation, and phase deviation.

[0089] Nonlinear distortion refers to the nonlinear components in the output response of a vibration suppression device when a preset excitation is applied, which do not conform to the linear response of the preset excitation. These nonlinear components may originate from factors such as internal friction, gaps, material nonlinearity, or control system saturation. Specifically, nonlinear distortion can be quantified by comparing the spectral differences or harmonic content between the actual excitation output information and the ideal linear response. Its purpose is to accurately characterize the nonlinear effects generated by the vibration suppression device during operation.

[0090] Amplitude deviation refers to the difference between the instantaneous amplitude or peak value of the actual excitation output information of a vibration suppression device and the ideal amplitude of the preset excitation. This deviation may be caused by insufficient driving capability, power supply fluctuations, or sensor calibration errors. In practical applications, amplitude deviation can be determined by calculating the percentage error or absolute difference between the amplitude of the actual excitation output information and the amplitude of the preset excitation. Its purpose is to evaluate whether the output strength of the vibration suppression device meets expectations.

[0091] Phase deviation refers to the lag or lead of the actual excitation output information of a vibration suppression device relative to the preset excitation on the time axis. Phase deviation may be caused by factors such as system inertia, control delay, or signal transmission delay. For example, phase deviation can be quantified by comparing the phase angle of the zero-crossing point or a specific frequency component of the actual excitation output information with the preset excitation, with the aim of reflecting the timeliness and synchronicity of the vibration suppression device's response.

[0092] The proposed solution decomposes the original excitation output deviation into nonlinear distortion, amplitude deviation, and phase deviation, enabling a more precise and comprehensive quantification of the vibration suppression device's performance status. When performing time-domain and frequency-domain analysis on the actual excitation output information and constructing a set of nonlinear response characteristics, these specific deviation components can serve as key indicators for comparing the nonlinear response characteristic set with the linear response characteristics of the preset excitation. By quantifying these deviations separately, the performance degradation mode of the vibration suppression device can be identified more accurately, such as whether it is due to enhanced nonlinear effects, decreased output capability, or increased response delay.

[0093] The above technical solution enables a more detailed and comprehensive characterization of the original excitation output deviation of the vibration suppression device. This refinement helps to gain a deeper understanding of the performance degradation mechanism of the vibration suppression device, such as distinguishing between the increase in nonlinearity caused by mechanical wear, the amplitude attenuation caused by aging of the drive circuit, and the phase drift caused by control system delay. Therefore, it can provide more accurate input data for subsequent performance status assessment and fault diagnosis, thereby improving the accuracy and reliability of platform structure fault diagnosis.

[0094] This application further proposes the following steps before compensating the stimulus response data:

[0095] Multiple temperature sensors are deployed on the platform structure to acquire temperature data at different locations on the platform structure in real time.

[0096] Based on the temperature data, calculate the temperature gradient in each region of the platform structure;

[0097] The thermal deformation response of the platform structure is evaluated based on the temperature gradient.

[0098] Based on the thermal deformation response, identify the temperature-related thermal deformation components in the excitation response data;

[0099] The thermal deformation component is separated from the excitation response data to obtain excitation response data with the influence of thermal deformation removed;

[0100] Based on the original excitation output deviation, the excitation response data after removing the influence of thermal deformation is compensated to obtain the compensated excitation response data.

[0101] Specifically, deploying multiple temperature sensors on the platform structure refers to installing high-precision temperature sensors in key areas of the steel structure fabrication platform, such as load-bearing beams, connection nodes, and support columns. These sensors are configured to continuously collect temperature values ​​of the platform structure at different spatial locations in real time. The purpose of acquiring real-time temperature data at different locations on the platform structure is to comprehensively understand the temperature distribution of the platform structure during the application of a preset excitation, providing fundamental data for subsequent thermal deformation analysis.

[0102] The calculation of temperature gradients in different regions of the platform structure based on temperature data can be understood as constructing a temperature field model of the overall platform structure by interpolating, fitting, or performing finite element analysis on the collected discrete temperature data, and calculating the rate of temperature change between different regions or within the same region. For example, the ratio of temperature difference to distance between adjacent sensors can be calculated, or the partial derivatives of the temperature field can be solved using numerical methods. The purpose is to quantify the potential areas of thermal stress generated within the platform structure due to uneven temperature distribution.

[0103] In practical applications, evaluating the thermal deformation response of a platform structure based on a temperature gradient involves using thermoelasticity principles or finite element analysis software, taking the calculated temperature gradient as input, to simulate and analyze the platform structure's thermal expansion and contraction under the current temperature distribution, as well as the resulting stress and deformation. For example, the linear expansion, bending deformation of each component, and the displacement field of the overall structure can be calculated. The aim is to accurately predict the impact of temperature changes on the platform structure's geometry and mechanical properties.

[0104] Furthermore, identifying the temperature-related thermal deformation components in the excitation response data based on the thermal deformation response involves comparing and analyzing the evaluated thermal deformation response with the acquired excitation response data. Since thermal deformation alters the natural frequencies and mode shapes of the structure, signal processing techniques, such as modal analysis, wavelet transform, or empirical mode decomposition, can be used to separate the low-frequency or quasi-static response components caused by thermal deformation from the excitation response data. The aim is to isolate the temperature effect from the pure vibration response.

[0105] Therefore, separating the thermal deformation component from the excitation response data to obtain excitation response data free from the influence of thermal deformation refers to removing the identified thermal deformation component from the original excitation response data through methods such as digital filtering, signal reconstruction, or residual analysis. For example, an adaptive filtering algorithm can be used to filter out slowly changing signals caused by thermal deformation from rapidly changing vibration signals. The aim is to obtain purer excitation response data that better reflects the true dynamic characteristics of the structure.

[0106] Finally, based on the original excitation output deviation, the excitation response data after removing the influence of thermal deformation is compensated to obtain compensated excitation response data. This means that, after eliminating the influence of temperature, the data is corrected using the original excitation output deviation of the vibration suppression device, following the steps described above for determining the performance status of the vibration suppression device and recording the trend of performance status changes. The purpose is to ensure the accuracy of the compensation process and avoid interference from temperature factors in determining the performance status of the vibration suppression device.

[0107] This application's solution effectively solves the problem of temperature variation interfering with the accuracy of excitation response data by introducing a step to identify and eliminate the temperature effect on the platform structure before compensating the excitation response data. Specifically, by deploying temperature sensors and acquiring temperature data, a comprehensive understanding of the temperature distribution of the platform structure can be obtained; by calculating the temperature gradient and evaluating the thermal deformation response, the impact of temperature changes on the structural mechanical behavior can be quantified; and by identifying and separating the thermal deformation component in the excitation response data, spurious responses caused by temperature can be removed from the true structural dynamic response. These steps ensure that subsequent compensation processing based on the original excitation output deviation can be applied to purer excitation response data, thereby avoiding interference from temperature factors in determining the performance status of the vibration suppression device and ensuring the reliability of the diagnostic results.

[0108] Through the above technical solution, this application can effectively eliminate the influence of ambient temperature changes on the dynamic response data of the steel structure processing platform, improving the purity and accuracy of the excitation response data. Compared with compensation methods that only consider the performance deviation of the vibration suppression device itself, this application further eliminates the thermal deformation component caused by temperature, making the judgment of the performance status of the vibration suppression device more accurate and avoiding misjudgments caused by temperature factors. In addition, by providing more accurate excitation response data after compensation, it also provides a more reliable basis for subsequent damage identification of the main structure of the platform, thereby improving the robustness and reliability of the entire fault diagnosis method.

[0109] In some preferred embodiments, this application is implemented as follows:

[0110] Assuming that fault diagnosis of the steel structure processing platform is required during non-production periods, 20 high-precision thermocouple temperature sensors are first deployed at key locations on the platform structure, such as the middle of the main beam, connection nodes, and the bottom of the support columns. These sensors collect temperature data at each point in real time at a frequency of once per second.

[0111] While collecting excitation response data, the system uses data from these 20 temperature sensors to construct the real-time temperature field of the platform structure using a three-dimensional interpolation algorithm, and calculates the temperature gradient of each region. For example, if a 3°C temperature difference is found between the upper and lower surfaces of the main beam, the corresponding temperature gradient will be calculated.

[0112] Subsequently, based on the calculated temperature gradient, combined with the material properties and geometric model of the platform structure, the thermal deformation response of the platform structure under the current temperature distribution is evaluated using finite element analysis software (such as ANSYS or ABAQUS), including the linear expansion of each component and the minute bending or torsion of the overall structure.

[0113] Next, the evaluated thermal deformation response is compared with the excitation response data acquired by the accelerometer. For example, if the thermal deformation evaluation shows that the platform has a slow, quasi-static downward bending with its frequency components mainly concentrated below 0.1 Hz, the system will use signal processing techniques such as low-pass filters or empirical mode decomposition to identify and separate the low-frequency components corresponding to this thermal deformation from the excitation response data.

[0114] After successfully separating the thermal deformation component, the temperature-related portion of the original excitation response data is removed, resulting in excitation response data free from the influence of thermal deformation. Finally, based on the original excitation output deviation of the vibration suppression device (e.g., amplitude and phase deviations obtained through self-feedback data quantization), these excitation response data free from the influence of thermal deformation are compensated to obtain more accurate compensated excitation response data, which is used to determine the performance status of the vibration suppression device.

[0115] This application further proposes a step for comparing abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from a change in the performance of the vibration suppression device, including:

[0116] Before applying a preset excitation to the vibration suppression device, assess the high-frequency interference around the force sensor inside the vibration suppression device.

[0117] When high-frequency interference is detected, a high-frequency energy transient absorption and isolation mechanism is activated, which includes hardware filtering, software noise reduction, active cancellation, and physical isolation operations.

[0118] After high-frequency interference is absorbed or isolated, clean data from the force sensor inside the vibration suppression device is collected.

[0119] Based on the clean data, the actual excitation output deviation of the vibration suppression device is quantified, and the actual performance degradation record of the vibration suppression device is updated based on the actual excitation output deviation of the vibration suppression device.

[0120] During platform operation, when an anomaly is detected, the anomaly characteristics are compared with the actual performance degradation records of the vibration suppression device to obtain the comparison results;

[0121] If the comparison results match, the anomaly is determined to be due to a true performance degradation of the vibration suppression device.

[0122] If the comparison results do not match, the transient thermal stress effect is compensated from the operational response data, and the damage characteristics of the platform's main structure are identified from the compensated operational response data.

[0123] Specifically, before applying the preset excitation to the vibration suppression device, the high-frequency interference around the force sensor inside the device is evaluated. This aims to ensure that the subsequently acquired data accurately reflects the performance status of the vibration suppression device and avoids the intrusion of external noise. High-frequency interference can be understood as noise components with frequencies higher than the target signal, caused by factors such as power fluctuations, electromagnetic radiation, and mechanical vibration. When such high-frequency interference is detected, a high-frequency energy transient absorption and isolation mechanism is activated. This mechanism aims to effectively suppress or eliminate these transient high-frequency energies through hardware filtering, software noise reduction, or active cancellation, thereby providing a cleaner measurement environment for the force sensor.

[0124] The activation of the high-frequency energy transient absorption and isolation mechanism specifically refers to a series of operations taken in the structural fault diagnosis method for steel structure processing platforms to eliminate or suppress high-frequency interference detected by the system around the force sensor inside the vibration suppression device before the device applies a preset excitation. Its core purpose is to ensure that the force sensor can collect clean data, thereby accurately quantifying the true excitation output deviation of the vibration suppression device and avoiding the impact of external noise on performance evaluation.

[0125] The "high-frequency energy transient absorption and isolation mechanism" specifically refers to a comprehensive technical approach for effectively suppressing or eliminating transient high-frequency energy. Here, "high-frequency interference" refers to noise components with frequencies higher than the target signal. This noise may originate from various industrial environmental factors, such as power fluctuations, electromagnetic radiation, and vibrations from nearby machinery. This mechanism, through the synergistic effect of multiple technical approaches, provides a cleaner measurement environment for the force sensor, ensuring the accuracy of data acquisition.

[0126] Specifically, the mechanism may include the following operating methods:

[0127] The first step is hardware filtering. This typically involves integrating physical electronic filters, such as low-pass filters, at the front end of the force sensor signal path. These filters selectively block or attenuate high-frequency noise signals, allowing only signals within the target frequency range to pass through, thus pre-cleaning the signal before it enters digital processing.

[0128] Secondly, there is software noise reduction. After the sensor data is digitized, it can be further denoised using digital signal processing algorithms. For example, digital filters (such as FIR or IIR filters), adaptive filtering algorithms, or wavelet denoising techniques can be used to identify and remove high-frequency noise components from the original data, thereby extracting a cleaner signal.

[0129] Another method is active cancellation. This is a more advanced noise reduction method that monitors high-frequency interference signals in real time and generates an anti-noise signal with the same amplitude but opposite phase as the interference signal. This anti-noise signal is then superimposed on the interfered signal, thereby achieving active cancellation of the interference. This may require additional actuators or complex control systems to implement.

[0130] In addition, physical isolation methods, such as setting up an electromagnetic shielding layer around the force sensor or using vibration damping materials for mechanical isolation, can be used to prevent external high-frequency electromagnetic waves or mechanical vibrations from directly coupling to the sensor.

[0131] For example, during fault diagnosis of a steel structure processing platform, if electromagnetic interference from nearby high-frequency welding equipment is detected, this interference can propagate through the air or structure, affecting the measurement accuracy of the force sensor inside the vibration suppression device. In this case, the system immediately activates a high-frequency energy transient absorption and isolation mechanism. This may include: activating the digital filter at the front end of the force sensor to filter out high-frequency electromagnetic noise generated by the welding equipment; and, if an electromagnetic shielding layer is installed around the sensor, ensuring it is in an effective working state to physically isolate external electromagnetic radiation. Through these operations, the data collected by the force sensor will not be affected by high-frequency electromagnetic interference, thus ensuring the accuracy of the vibration suppression device performance evaluation.

[0132] After high-frequency interference is absorbed or isolated, the data collected from the force sensors inside the vibration suppression device is called purified data. Based on this purified data, the actual excitation output deviation of the vibration suppression device can be quantified more accurately. Actual excitation output deviation refers to the difference between the output excitation of the vibration suppression device under actual operating conditions and the theoretically preset excitation. This deviation can more accurately reflect the actual performance of the vibration suppression device. Based on the actual excitation output deviation, the actual performance degradation record of the vibration suppression device can be updated, forming a dynamic and accurate performance baseline.

[0133] During platform operation, when the system detects an anomaly, it compares this anomaly with the latest updated performance degradation records of the vibration suppression device. This comparison aims to determine whether the current anomaly matches the known performance degradation pattern of the vibration suppression device. If the comparison results match, it can be definitively determined that the anomaly originates from the actual performance degradation of the vibration suppression device. Conversely, if the comparison results do not match, it indicates that the anomaly is not caused by the performance degradation of the vibration suppression device. In this case, to more accurately identify the damage characteristics of the platform's main structure, it is necessary to compensate for the transient thermal stress effects from the operational response data. Transient thermal stress effects refer to the instantaneous deformation and stress changes in the platform structure caused by factors such as changes in ambient temperature, local heating, or cooling. These changes may introduce interference into the operational response data, affecting the accuracy of damage identification. Through compensation processing, a cleaner operational response data, free from the influence of thermal stress, can be obtained, thereby enabling the identification of damage characteristics of the platform's main structure.

[0134] This application's solution introduces an evaluation and processing mechanism for high-frequency interference before applying a preset excitation to the vibration suppression device, ensuring the purity of the data collected by the force sensors inside the device. This allows for the quantification of the actual excitation output deviation of the vibration suppression device based on more reliable and pure data, and the updating of its actual performance degradation record accordingly. This process effectively avoids the problem of inaccurate performance evaluation caused by environmental noise or sensor interference in traditional methods, making the judgment of the vibration suppression device's performance status more accurate. Furthermore, during platform operation, when an anomaly is detected, comparing the anomaly characteristics with the purified and calibrated actual performance degradation record significantly improves the accuracy of anomaly source identification. When the anomaly is not determined to originate from the vibration suppression device, this solution compensates for the transient thermal stress effect in the operational response data, eliminating structural response interference caused by temperature changes, thus providing a purer and more reliable data foundation for subsequent identification of damage characteristics of the platform's main structure. Due to the strict control of the data source and the refined processing of interference factors, this application's solution can more accurately distinguish between vibration suppression device malfunctions and damage to the platform's main structure, avoiding misjudgments and omissions.

[0135] Through the above technical solutions, this application can significantly improve the accuracy and reliability of structural fault diagnosis for steel structure processing platforms. Specifically, by activating absorption and isolation mechanisms in the presence of high-frequency interference, the purity of the vibration suppression device performance evaluation data is ensured, thereby making the quantified true excitation output deviation and updated true performance degradation records more accurate. This makes the correlation judgment between abnormal features and vibration suppression device performance degradation more accurate during platform operation, effectively avoiding the situation where abnormalities caused by other interferences are misjudged as vibration suppression device failures. In addition, when the abnormality does not originate from the vibration suppression device, by compensating for the transient thermal stress effect in the operating response data, the interference of temperature changes on the structural response is eliminated, providing a purer data basis for identifying damage characteristics of the platform's main structure, thereby improving the accuracy of damage identification of the platform's main structure and reducing the misdiagnosis rate.

[0136] In some preferred embodiments, it is assumed that a steel structure processing platform undergoes routine diagnostics during non-production hours at night. Before applying a preset excitation to the vibration suppression device, the system first assesses whether there is high-frequency interference around its internal force sensors. For example, if electromagnetic interference generated by nearby high-frequency welding equipment is detected, the system immediately activates a high-frequency energy transient absorption and isolation mechanism, such as by activating a digital filter and electromagnetic shielding layer at the sensor front end, to ensure that the data collected by the force sensors is not affected by external high-frequency noise. After the high-frequency interference is effectively absorbed or isolated, the force sensors collect clean data, which is used to quantify the actual excitation output deviation of the vibration suppression device and update its actual performance degradation record accordingly. For example, the record shows that the damping coefficient of the vibration suppression device has a slight downward trend over the past month.

[0137] Subsequently, during platform operation, the monitoring system detected an abnormal vibration response in a certain area of ​​the platform, manifested as a sudden increase in energy at a specific frequency. At this point, the system compares this abnormal characteristic with previously updated records of actual performance degradation of the vibration suppression device. If the comparison shows that the abnormal characteristic matches the vibration mode caused by a decrease in the damping coefficient of the vibration suppression device, the system determines that the abnormality originates from actual performance degradation of the vibration suppression device and recommends maintenance of the device.

[0138] However, if the comparison results do not match, meaning the abnormal characteristic does not match the actual performance degradation record of the vibration suppression device, the system will further compensate for the transient thermal stress effect from the operational response data. For example, by using a temperature sensor network integrated into the platform structure, temperature data for each area is acquired in real time, and the transient thermal deformation and thermal stress response caused by temperature changes are calculated using a finite element model. These thermal stress effects are then accurately compensated for from the operational response data. In the compensated operational response data, the system identifies a significant decrease in the local stiffness of a certain connection part of the platform's main structure, thereby accurately identifying the damage characteristics of the platform's main structure, such as loose bolts or weld cracks, and triggering corresponding maintenance alarms. Through this series of refined processing, the solution in this application can effectively avoid misjudgments and ensure the accuracy of fault diagnosis.

[0139] The steps described above for identifying damage characteristics of the platform's main structure from operational response data after eliminating known interferences include:

[0140] Frequency domain analysis was performed on the operational response data after eliminating known interference to extract the energy distribution characteristics of the operational response data in multiple preset frequency ranges.

[0141] Based on energy distribution characteristics, identify local stiffness changes, damping changes, or mode distortion characteristics of the platform structure at preset monitoring locations;

[0142] By comparing local stiffness changes, damping changes, or mode distortion characteristics with the historical health status characteristics of the platform structure, it can be determined whether there are any abnormal changes that deviate from the normal evolution trend.

[0143] When abnormal changes persist and the corresponding abnormal changes are concentrated in some monitoring locations, the corresponding monitoring locations are identified as potential damage areas of the platform's main structure, and the damage characteristics of the platform's main structure are identified.

[0144] Frequency domain analysis of the operational response data after eliminating known interference involves using signal processing techniques such as Fourier transform to convert the time-domain operational response data to the frequency domain, allowing observation of its energy distribution at different frequencies. Extracting the energy distribution characteristics of the operational response data after eliminating known interference within multiple preset frequency ranges aims to quantify the response intensity of the platform structure under specific vibration modes. These frequency ranges are typically associated with the critical modal frequencies or potential damage-sensitive frequencies of the platform structure.

[0145] Furthermore, based on energy distribution characteristics, identifying local stiffness changes, damping changes, or mode shape distortion characteristics of the platform structure at preset monitoring locations refers to inferring changes in the platform structure's physical parameters by analyzing changes in frequency domain energy distribution, such as shifts in specific frequency peak values, increases or decreases in amplitude, or broadening of bandwidth. Preset monitoring locations refer to sensor positions pre-arranged on the platform structure. These locations are typically chosen in areas of structural stress concentration, connection nodes, or vulnerable parts to more effectively capture local damage information. Local stiffness changes may manifest as a decrease in natural frequency, damping changes may manifest as a change in vibration decay rate, and mode shape distortion may be reflected by a decrease in the modal correlation coefficient.

[0146] Furthermore, comparing local stiffness variations, damping variations, or modal distortion characteristics with the historical health status characteristics of the platform structure aims to establish a benchmark to determine whether the current state of the platform structure deviates from its normal operating trajectory. Historical health status characteristics can be extracted from benchmark data collected when the platform structure is undamaged or in a known healthy state. By comparing these characteristics, it can be determined whether there are any abnormal changes deviating from the normal evolutionary trend, i.e., whether the current structural parameters have exceeded the normal fluctuation range or exhibit a continuously deteriorating trend.

[0147] Specifically, when abnormal changes persist and are concentrated in certain monitoring locations, these locations are identified as potential damage areas for the platform's main structure, and damage characteristics are determined. Persistence implies that the anomaly is not instantaneous or sporadic but rather persistent, increasing its reliability as a genuine damage indicator. Concentration in certain monitoring locations suggests localized damage, facilitating precise location of the damaged area and identification of specific damage characteristics, such as cracks, loose connections, or material degradation.

[0148] The proposed solution employs refined frequency domain analysis of operational response data after eliminating known disturbances. This effectively decomposes complex time-domain signals into different frequency components, revealing the energy distribution of the platform structure under various vibration modes. This analytical method allows for the sensitive capture of changes in key structural dynamic parameters such as local stiffness, damping, and mode shapes. By rigorously comparing these identified changes with the historical health characteristics of the platform structure, it is possible to distinguish between normal operational fluctuations and genuine structural anomalies. In particular, when anomalous changes exhibit persistence and localized concentration, the influence of incidental disturbances or global environmental factors can be eliminated, thereby accurately pinpointing potential damage areas in the main platform structure and further identifying specific damage types. Therefore, this solution provides a systematic and reliable damage identification mechanism, offering a solid technical basis for subsequent maintenance and repair decisions.

[0149] Through the above technical solution, this application enables early and accurate identification of damage to the main structure of the steel structure processing platform. By performing frequency domain analysis on the operational response data and extracting energy distribution characteristics, it can sensitively capture minute changes in local stiffness, damping, or mode shapes, thus identifying potential problems in the early stages of damage. Compared with methods that only perform simple anomaly detection, this solution effectively avoids false alarms and improves diagnostic accuracy by comparing with historical health status characteristics. Furthermore, by judging the persistence and local concentration of abnormal changes, the damaged area can be accurately located, providing clear guidance for targeted maintenance, significantly improving the safety and reliability of platform operation, and reducing maintenance costs.

[0150] Specifically, the steps for identifying damage characteristics of the platform's main structure can be further refined as follows:

[0151] When the abnormal change is manifested as a local high-frequency energy enhancement and a change in damping characteristics, the potential damage area is determined to have structural crack-like damage characteristics.

[0152] When abnormal changes manifest as changes in the distribution of low- and medium-frequency energy and alterations in stiffness characteristics, the potential damage area is determined to have structural connection loosening or stiffness degradation damage characteristics.

[0153] The phrase "enhanced local high-frequency energy and altered damping characteristics" refers to the finding, during frequency domain analysis of operational response data after the elimination of known interferences, that vibration energy significantly increases in a specific local region at higher frequencies, while the energy dissipation capacity (i.e., damping) in that region also exhibits abnormal changes. This phenomenon is typically associated with the formation of microcracks or defects within the structure, as the presence of cracks alters the continuity and stiffness of the local material, thereby affecting its high-frequency vibration response and generating additional energy dissipation at the crack interface.

[0154] "Structural crack-related damage characteristics" refer to macroscopic or microscopic fractures that occur inside or on the surface of a material due to factors such as fatigue, stress concentration, and corrosion. This type of damage may not be easily detected in its early stages, but as the cracks propagate, it poses a serious threat to the load-bearing capacity and safety of the structure.

[0155] "Changes in low-to-mid-frequency energy distribution and stiffness characteristics" refers to the discovery of anomalies in energy distribution in the low-to-mid-frequency range during frequency domain analysis of operational response data after the elimination of known interferences. This could manifest as a shift in certain natural frequencies or a significant change in response amplitude at specific frequencies, coupled with a decrease in local or overall structural stiffness as assessed through modal analysis or other methods. Such changes are typically related to factors such as loosening of structural connectors, stiffness degradation due to material aging, or local structural deformation.

[0156] "Structural connection loosening or stiffness degradation damage characteristics" refers to the reduction in connection stiffness between structural components or the decrease in the elastic modulus of the structural material itself due to reasons such as loose bolt connections, weld cracking, rivet failure, material fatigue or aging, which changes the overall or local load-bearing capacity and deformation characteristics of the structure.

[0157] This application's approach, through detailed analysis of energy distribution characteristics and changes in structural dynamic parameters (such as damping and stiffness) across different frequency ranges, can further concretize the general concept of "abnormal changes" into damage types with clear physical significance. Increased high-frequency energy and changes in damping characteristics are considered typical fingerprints of structural crack damage, because crack initiation causes minor changes in local stiffness and alterations in energy dissipation mechanisms, which are reflected in the high-frequency response. Conversely, changes in mid- and low-frequency energy distribution and significant alterations in stiffness characteristics are more likely to indicate loosening of structural connections or degradation of material stiffness, as these damage modes typically affect the overall or broader dynamic characteristics of the structure, thus being more pronounced in the mid- and low-frequency vibration response. This differentiation of damage types based on frequency and dynamic parameter differences makes the fault diagnosis process more targeted.

[0158] The aforementioned technical solution enables refined identification of damage characteristics to the platform's main structure. Compared to simply determining the presence of anomalies, this application can further clarify the nature and type of damage, such as structural cracks, loose connections, or stiffness degradation. This precise damage classification provides a more reliable and specific basis for subsequent repair decisions and maintenance strategy formulation. For example, crack-type damage may require welding repair or component replacement; while loose connections may only require tightening. This significantly improves repair efficiency and accuracy, reduces maintenance costs, and effectively enhances the operational safety and reliability of the steel structure processing platform.

[0159] In this regard, this application further proposes steps for obtaining historical health status characteristics of the platform structure, including:

[0160] During non-production periods or when the platform is in a stable operating state, and under the condition that the vibration suppression device is in a known performance state, the baseline response data of the monitoring position corresponding to the platform structure is collected.

[0161] Feature extraction is performed on the collected baseline response data to form reference features characterizing the platform structure under a state of no structural damage;

[0162] Store the reference features as historical health status features of the platform structure.

[0163] Specifically, data collection is conducted during non-production periods or periods of stable operation of the platform. This aims to minimize the impact of external environmental factors (such as shocks caused by production operations, load changes, etc.) on the platform's structural response data, thereby obtaining purer and more stable benchmark data. Non-production periods can be understood as times when the platform does not perform any processing or transportation operations, such as nighttime or holidays. Periods of stable operation refer to the periods when the platform's vibration response is stable under specific, constant loads or operating modes.

[0164] Furthermore, the condition that the vibration suppression device is in a known performance state means that, when collecting reference response data, the performance parameters of the vibration suppression device (such as its excitation output capability, damping characteristics, etc.) are clearly known and stable, or its performance degradation trend has been accurately recorded. This is to ensure that the collected reference response data can truly reflect the inherent characteristics of the platform structure in a damage-free state, and will not be confused by the uncertainty of the vibration suppression device's own performance.

[0165] In one preferred embodiment, the baseline response data of the platform structure at the corresponding monitoring location can be collected by arranging acceleration sensors, strain sensors, or displacement sensors at key parts of the platform structure (such as beams, columns, and connection nodes). These sensors are configured to acquire vibration response, strain response, or displacement response data of the platform structure under preset excitation or environmental excitation in real time or periodically.

[0166] Feature extraction from the acquired baseline response data can be understood as extracting physical quantities or mathematical characteristics that characterize the health status of the platform structure from the raw response data using signal processing techniques (such as Fourier transform, wavelet analysis, modal analysis, etc.). These characteristics may include, but are not limited to, modal frequencies, modal damping ratios, mode shapes, frequency response functions, energy distribution characteristics, and time-domain statistical characteristics. These extracted features are used to form reference features characterizing the platform structure in a damage-free state; that is, these features represent the inherent dynamic characteristics of the platform structure in a healthy and intact state.

[0167] Therefore, storing the aforementioned reference features as historical health status characteristics of the platform structure aims to establish a reliable benchmark database. This database contains various dynamic response characteristics of the platform structure in a healthy state, which can be used for comparative analysis of operational response data collected during platform operation, thereby effectively identifying damage characteristics of the platform's main structure.

[0168] The proposed solution establishes a clean and reliable database of historical health status features for the platform structure by collecting benchmark response data and extracting features under controlled conditions (during non-production periods or when the platform is in stable operation and the performance of the vibration suppression device is known). Because this database is established without the influence of external interference and the inherent uncertainties of the vibration suppression device, it allows for accurate comparison of any identified abnormal features with these historical health status features during platform operation. This comparison mechanism effectively distinguishes between genuine anomalies caused by damage to the platform's main structure and false anomalies caused by other factors (such as environmental changes and fluctuations in the performance of the vibration suppression device), thereby improving the accuracy and reliability of damage identification.

[0169] Through the above technical solution, this application can establish a highly reliable and representative historical health status profile of the platform structure. This enables comparisons based on an accurate "health" benchmark during subsequent fault diagnosis, significantly improving the sensitivity and accuracy of identifying damage characteristics of the platform's main structure and effectively reducing false alarm and false negative rates. Furthermore, by continuously updating and maintaining this historical health status profile, long-term monitoring and trend analysis of the platform's structural health status can be achieved, providing more robust data support for preventative maintenance and safety management of the platform.

[0170] refer to Figure 2 This application proposes a structural fault diagnosis system for a steel structure processing platform, applied to the aforementioned structural fault diagnosis method for a steel structure processing platform. The system includes:

[0171] The data acquisition module applies a preset excitation to the platform structure through a vibration suppression device during non-production periods and collects the excitation response data of the platform structure to the preset excitation.

[0172] The recording module determines the performance status of the vibration suppression device based on the excitation response data and records the trend of performance status changes.

[0173] The identification module continuously collects operational response data of the platform structure during platform operation, identifies anomalies in the operational response data, and derives anomaly characteristics.

[0174] The judgment module compares the abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from the performance change of the vibration suppression device.

[0175] The processing module, when the anomaly is not determined to be caused by a change in the performance of the vibration suppression device, eliminates the influence of known interference from the operation response data and identifies the damage characteristics of the main structure of the platform from the operation response data after eliminating known interference.

[0176] Known disturbances include transient thermal stress, environmental noise, power fluctuations, and periodic vibrations caused by specific processing conditions.

[0177] Specifically, the data acquisition module can be understood as the hardware and software components responsible for data acquisition. Its function is to apply a preset excitation to the platform structure through a vibration suppression device during non-production periods, and simultaneously acquire the platform structure's excitation response data to the preset excitation. This module can include various sensors (such as accelerometers, strain sensors, displacement sensors, etc.), as well as a data acquisition card, an analog-to-digital converter, and corresponding driver software. These sensors are strategically placed at key locations on the platform structure to ensure comprehensive and accurate capture of the platform structure's dynamic response under excitation.

[0178] The recording module can be understood as a component responsible for data processing, analysis, and storage. Its function is to determine the performance state of the vibration suppression device based on the excitation response data acquired by the acquisition module, using a built-in algorithm, and continuously record the trend of this performance state. This module may include a data processor, memory, and software programs for performing data analysis and state assessment. The determination of the performance state can be based on feature extraction and pattern recognition of the excitation response data, such as analyzing the transfer function between the vibration suppression device's output and the platform response, and energy dissipation characteristics.

[0179] In practical applications, the identification module is specifically responsible for continuous monitoring and anomaly detection during platform operation. Its function is to continuously collect operational response data of the platform structure and identify anomalies in the operational response data in real time, deriving anomaly characteristics. This module can also utilize sensor networks to acquire operational data and, through signal processing techniques (such as time-domain analysis, frequency-domain analysis, wavelet analysis, etc.) and machine learning algorithms, automatically detect abnormal signals deviating from normal patterns from massive amounts of operational data, and extract features characterizing these anomalies, such as the amplitude, frequency components, and duration of the anomalies.

[0180] Furthermore, the judgment module is specifically responsible for determining the source of the fault. Its function is to compare the abnormal features identified by the identification module with the performance status change trends of the vibration suppression device recorded by the recording module to determine whether the detected abnormality originates from a performance change in the vibration suppression device. This module can use pattern matching, threshold comparison, or statistical methods for comparison. For example, if the abnormal features highly match the known performance degradation pattern of the vibration suppression device, then the abnormality is determined to be related to the vibration suppression device.

[0181] Furthermore, the processing module is specifically responsible for further diagnosis and damage identification. Its function is to eliminate the influence of known interferences from the operational response data when the judgment module determines that the anomaly is not due to performance changes in the vibration suppression device, and to identify damage characteristics of the platform's main structure from the operational response data after eliminating known interferences. This module can integrate various signal denoising and feature extraction algorithms. For example, it can eliminate known interferences such as environmental noise and operational loads through filters, and then further analyze the purified data to identify potential damage to the platform's main structure (e.g., beams, columns, connectors), such as cracks, loosening, and stiffness degradation.

[0182] The solution presented in this application modularizes the key steps in the structural fault diagnosis method for steel structure processing platforms and implements them using specialized hardware and software components, thereby constructing an efficient and automated diagnostic system. The content disclosed above is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the description and drawings of this invention are included within the scope of protection of this invention. Furthermore, the elements herein can be updated as technology develops.

Claims

1. A method for diagnosing structural faults in a steel structure processing platform, characterized in that, The method includes the following steps: During non-production periods of the platform, a preset excitation is applied to the platform structure through a vibration suppression device, and the excitation response data of the platform structure to the preset excitation is collected. Based on the excitation response data, the performance status of the vibration suppression device is determined, and the trend of its performance status change is recorded. The steps for determining the performance status of the vibration suppression device and recording its trend include: during non-production periods on the platform, applying a preset excitation to the platform structure through the vibration suppression device and simultaneously collecting self-feedback data from within the vibration suppression device, including the actual excitation output information of the vibration suppression device; quantifying the original excitation output deviation of the vibration suppression device based on the actual excitation output information; compensating the excitation response data based on the original excitation output deviation to obtain compensated excitation response data; and determining the performance of the vibration suppression device based on the compensated excitation response data. The steps for quantifying the original excitation output deviation of the vibration suppression device based on the actual excitation output information include: performing time-domain analysis on the actual excitation output information to extract the instantaneous amplitude, rise time, and decay rate of the actual excitation output information, forming a time-domain analysis result; performing frequency-domain analysis on the actual excitation output information to extract the dominant frequency, harmonic components, and bandwidth of the actual excitation output information, forming a frequency-domain analysis result; correlating the time-domain analysis result with the frequency-domain analysis result to construct a set of nonlinear response characteristics of the vibration suppression device under the current preset excitation; and comparing the set of nonlinear response characteristics with the linear response characteristics of the preset excitation to quantify the original excitation output deviation of the vibration suppression device. During platform operation, continuous collection of operational response data of the platform structure is performed, and anomalies in the operational response data are identified to derive anomaly characteristics; Compare the abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from a change in the performance of the vibration suppression device. When the anomaly is not determined to be caused by a change in the performance of the vibration suppression device, the influence of known interference is eliminated from the operational response data, and the damage characteristics of the main structure of the platform are identified from the operational response data after the known interference is eliminated. Known disturbances include transient thermal stress, environmental noise, power fluctuations, and periodic vibrations caused by specific processing conditions.

2. The method for diagnosing structural faults in a steel structure processing platform as described in claim 1, characterized in that, The original excitation output deviation includes nonlinear distortion, amplitude deviation, and phase deviation.

3. The method for diagnosing structural faults in a steel structure processing platform as described in claim 1, characterized in that, The steps for compensating the excitation response data based on the original excitation output deviation to obtain the compensated excitation response data include: Multiple temperature sensors are deployed on the platform structure to acquire temperature data at different locations on the platform structure in real time. Based on the temperature data, calculate the temperature gradient in each region of the platform structure; The thermal deformation response of the platform structure is evaluated based on the temperature gradient. Based on the thermal deformation response, identify the temperature-related thermal deformation components in the excitation response data; The thermal deformation component is separated from the excitation response data to obtain excitation response data with the influence of thermal deformation removed; Based on the original excitation output deviation, the excitation response data after removing the influence of thermal deformation is compensated to obtain the compensated excitation response data.

4. The method for diagnosing structural faults in a steel structure processing platform as described in claim 1, characterized in that, The steps of comparing abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from a performance change in the vibration suppression device, and when the abnormality is not determined to originate from a performance change in the vibration suppression device, to eliminate the influence of known disturbances from the operational response data and to identify the damage characteristics of the platform's main structure from the operational response data after eliminating known disturbances, include: Before applying a preset excitation to the vibration suppression device, assess the high-frequency interference around the force sensor inside the vibration suppression device. When high-frequency interference is detected, a high-frequency energy transient absorption and isolation mechanism is activated, which includes hardware filtering, software noise reduction, active cancellation, and physical isolation operations. After high-frequency interference is absorbed or isolated, clean data from the force sensor inside the vibration suppression device is collected. Based on the clean data, the actual excitation output deviation of the vibration suppression device is quantified, and the actual performance degradation record of the vibration suppression device is updated based on the actual excitation output deviation of the vibration suppression device. During platform operation, when an anomaly is detected, the anomaly characteristics are compared with the actual performance degradation records of the vibration suppression device to obtain the comparison results; If the comparison results match, the anomaly is determined to be due to a true performance degradation of the vibration suppression device. If the comparison results do not match, the transient thermal stress effect is compensated from the operational response data, and the damage characteristics of the platform's main structure are identified from the compensated operational response data.

5. The method for diagnosing structural faults in a steel structure processing platform as described in claim 1, characterized in that, The steps for identifying damage characteristics of the platform's main structure from operational response data after eliminating known disturbances include: Frequency domain analysis was performed on the operational response data after eliminating known interference to extract the energy distribution characteristics of the operational response data in multiple preset frequency ranges. Based on energy distribution characteristics, identify local stiffness changes, damping changes, or mode distortion characteristics of the platform structure at preset monitoring locations; By comparing local stiffness changes, damping changes, or mode distortion characteristics with the historical health status characteristics of the platform structure, it can be determined whether there are any abnormal changes that deviate from the normal evolution trend. When abnormal changes persist and the corresponding abnormal changes are concentrated in some monitoring locations, the corresponding monitoring locations are identified as potential damage areas of the platform's main structure, and the damage characteristics of the platform's main structure are identified.

6. The method for diagnosing structural faults in a steel structure processing platform as described in claim 5, characterized in that, The steps for identifying damage characteristics of the platform's main structure include: When the abnormal change is manifested as a local high-frequency energy enhancement and a change in damping characteristics, the potential damage area is determined to have structural crack-like damage characteristics. When abnormal changes manifest as changes in the distribution of low- and medium-frequency energy and alterations in stiffness characteristics, the potential damage area is determined to have structural connection loosening or stiffness degradation damage characteristics.

7. The method for diagnosing structural faults in a steel structure processing platform as described in claim 5, characterized in that, The steps to obtain historical health status characteristics of the platform structure include: During non-production periods or when the platform is in a stable operating state, and under the condition that the vibration suppression device is in a known performance state, the baseline response data of the monitoring position corresponding to the platform structure is collected. Feature extraction is performed on the collected baseline response data to form reference features characterizing the platform structure under a state of no structural damage; Store the reference features as historical health status features of the platform structure.

8. A structural fault diagnosis system for a steel structure processing platform, applied to the structural fault diagnosis method for a steel structure processing platform as described in claim 1, characterized in that, The system includes: The data acquisition module applies a preset excitation to the platform structure through a vibration suppression device during non-production periods and collects the excitation response data of the platform structure to the preset excitation. The recording module determines the performance status of the vibration suppression device based on the excitation response data and records the trend of performance status changes. The identification module continuously collects operational response data of the platform structure during platform operation, identifies anomalies in the operational response data, and derives anomaly characteristics. The judgment module compares the abnormal characteristics with the trend of performance status changes to determine whether the abnormality originates from the performance change of the vibration suppression device. The processing module, when the anomaly is not determined to be caused by a change in the performance of the vibration suppression device, eliminates the influence of known interference from the operation response data and identifies the damage characteristics of the main structure of the platform from the operation response data after eliminating known interference. Known disturbances include transient thermal stress, environmental noise, power fluctuations, and periodic vibrations caused by specific processing conditions.

Images