High frequency welded pipe production quality on-line monitoring intelligent diagnosis system

Abstract

The present application relates to the technical field of high-frequency welding equipment, and discloses an intelligent diagnosis system for online monitoring of high-frequency welded pipe production quality, comprising: a data acquisition unit configured to synchronously acquire high-frequency voltage and current waveforms generated by a high-frequency welding circuit under the common excitation of a main welding frequency signal and an injected probe signal; a feature extraction unit connected with the data acquisition unit and configured to extract at least one passive electromagnetic fingerprint and one active response fingerprint based on the waveforms and to generate a welding state vector by fusion; and an intelligent diagnosis unit connected with the feature extraction unit and configured to compare the welding state vector with a preset health model to determine the current high-frequency welded pipe welding quality. The present application can more comprehensively represent the welding state through the strategy of active and passive information fusion, and significantly improves the detection sensitivity and diagnosis accuracy for early and weak defects.

Description

Technical Field

[0001] This invention relates to the field of high-frequency welding equipment technology, specifically to an intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality. Background Technology

[0002] High-frequency induction welding is one of the mainstream processes for producing high-quality straight seam welded pipes in the current industrial field. This process, with its significant advantages such as high welding speed, narrow heat-affected zone, and high production efficiency, is widely used in many industries, including petroleum, natural gas, construction, and machinery manufacturing. During high-frequency welding, the edge of the steel strip is rapidly heated to a molten or plastic state by the high-frequency induced current as it passes through the V-shaped guide zone, and then solid-state welding is achieved under the strong pressure of the extrusion rollers, forming a continuous and dense weld.

[0003] The quality of the weld directly determines the overall strength, sealing performance, and service life of the welded pipe, making it a core control point in the entire production process. However, high-frequency welding is a complex dynamic process involving strong coupling of multiple physical fields such as electromagnetic, thermal, and mechanical fields. It is highly susceptible to various uncertainties such as fluctuations in steel strip material, changes in equipment condition, and environmental interference, which can induce various welding defects such as cold welding, overheating, inclusions, and cracks.

[0004] Traditional welded pipe quality control methods largely rely on offline inspections, such as destructive tests like flattening and flaring of finished pipes, or non-destructive testing techniques like ultrasonic and eddy current testing. While these methods can assess the final product's quality, they are essentially "post-production inspections" with significant time lag. By the time defects are detected, a large number of substandard pipes have often already been produced, resulting in severe material waste and economic losses, and they fail to provide any effective information for real-time adjustment of process parameters or prevention of defects.

[0005] To overcome the drawbacks of offline inspection, the industry has developed various online monitoring technologies. For example, infrared thermal imagers are used to monitor the temperature distribution in the weld area, or macroscopic electrical parameters such as the output voltage, current, and power of the welding power source are directly monitored. However, these existing online monitoring methods still have significant limitations. Infrared thermal imaging methods are highly susceptible to interference from environmental factors such as welding spatter, water vapor, and fumes, and changes in the emissivity of the metal surface also affect the accuracy of temperature measurement; they can only capture the thermal manifestations of the process. While monitoring the macroscopic electrical parameters of the welding power source can reflect the stability of the overall energy supply, its sensitivity to local, transient physical processes that occur in the welding core area, i.e., the V-shaped convergence zone, and are crucial for defect formation, is often insufficient. The complexity of the welding process means that single-dimensional information monitoring cannot comprehensively and accurately depict its true state; a slight anomaly in one parameter may be masked by changes in other parameters, leading to missed or misjudged defects.

[0006] Therefore, there is an urgent need in this field for an online monitoring and diagnostic technology that can delve into the physical essence of the welding process, provide real-time, sensitive, and multi-dimensional information, in order to achieve early warning and accurate diagnosis of welding quality anomalies, thereby providing technical support for the intelligent quality control of high-frequency welded pipe production. Summary of the Invention

[0007] The technical problem to be solved by the present invention is that the existing high-frequency welded pipe quality monitoring technology has limitations such as insufficient real-time performance, inability to effectively detect internal defects in the weld, and difficulty in accurately diagnosing the root cause of defects.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] The first aspect of this invention provides an intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality, the system comprising:

[0010] A data acquisition unit is configured to synchronously acquire high-frequency voltage waveforms and high-frequency current waveforms in a high-frequency welding circuit, wherein the high-frequency voltage waveforms and high-frequency current waveforms are generated by the high-frequency welding circuit under the joint excitation of the main welding frequency signal and the injected probe signal;

[0011] A feature extraction unit, connected to the data acquisition unit, is configured to extract at least one passive electromagnetic fingerprint and one active response fingerprint based on the high-frequency voltage waveform and the high-frequency current waveform, and fuse them to generate a welding state vector; the passive electromagnetic fingerprint includes at least one of instantaneous complex impedance and / or harmonic energy ratio, and the active response fingerprint is the response feature of the high-frequency welding circuit to the probe signal;

[0012] An intelligent diagnostic unit, connected to the feature extraction unit, is configured to compare the welding state vector with a preset health model to determine the current welding quality of the high-frequency welded pipe.

[0013] Preferably, the system further includes:

[0014] A perturbation signal injection module, coupled to the high-frequency welding circuit, is configured to inject the probe signal into the high-frequency welding circuit; the frequency of the probe signal is different from the fundamental frequency and integer multiples of the harmonic frequency of the main welding frequency signal.

[0015] In one specific embodiment, the feature extraction unit is configured to extract the passive electromagnetic fingerprint as follows:

[0016] Calculate the ratio between the high-frequency voltage waveform v(t) and the high-frequency current waveform i(t) to obtain the instantaneous complex impedance Z. eq (t), whose mathematical expression is:

[0017] Z eq (t)=R eq (t)+jX eq (t);

[0018] Among them, R eq (t) represents the equivalent resistance characterizing energy dissipation, X eq (t) represents the equivalent reactance characterizing electromagnetic energy storage;

[0019] and / or

[0020] Spectral analysis of the high-frequency current waveform yields the time-frequency spectrum I(ω,t), which is used to obtain the harmonic energy ratio HER, characterizing the nonlinearity of the welding process. n (t), whose mathematical expression is:

[0021]

[0022] Among them, f w is the fundamental frequency of the main welding frequency signal, and n is the harmonic order.

[0023] In one specific embodiment, the feature extraction unit is configured to extract the active response fingerprint as follows:

[0024] A digital phase-locked amplification algorithm is used to lock and extract the response component caused by the probe signal from the high-frequency current waveform and / or high-frequency voltage waveform to obtain the active response fingerprint.

[0025] The active response fingerprint includes the amplitude attenuation A of the response component. p (t) and / or phase shift φ p(t), whose mathematical expression is:

[0026]

[0027] φ p (t)=atan2(C Q (t), C I (t));

[0028] Among them, C I (t) and C Q (t) represents the DC component obtained by mixing the input signal with in-phase and quadrature reference signals and then low-pass filtering.

[0029] Preferably, when the intelligent diagnostic unit compares the welding state vector with a preset health model, it is configured as follows:

[0030] The probability distribution model established based on multiple historical welding state vectors collected during the production of qualified welded pipes is used as the health model.

[0031] The Mahalanobis distance D between the current welding state vector S(t) and the health model is calculated. M The deviation is quantified by (t), and the welding quality is determined based on the deviation. The mathematical expression for the Mahalanobis distance is:

[0032]

[0033] Where, μ H and ∑ H These are the mean vector and covariance matrix of the health model, respectively.

[0034] Furthermore, the intelligent diagnostic unit also includes a defect evolution path tracking module, which is configured to store a time-series welding state vector to form an evolution path and identify defect patterns based on the dynamic properties of the evolution path.

[0035] In one specific embodiment, the defect evolution path tracing module is configured as follows:

[0036] The geometric and dynamic properties of the evolution path in the multidimensional feature space are analyzed, including the velocity, curvature and direction changes of the trajectory, to classify the defect pattern into one of point defects, linear defects or periodic defects.

[0037] Preferably, the system is further configured to:

[0038] When the defect evolution path tracking module identifies a preset defect pattern, it triggers the perturbation signal injection module to adjust or inject a specific probe signal sequence to target and confirm the defect pattern.

[0039] In one specific embodiment, the data acquisition unit is further configured to synchronously acquire at least one external physical signal, the external physical signal including at least one of thermal field information of the weld area or geometric information of the weld bead;

[0040] The feature extraction unit is also configured to integrate the features extracted from the external physical signal into the welding state vector.

[0041] A second aspect of this invention provides an intelligent diagnostic method for online monitoring of the production quality of high-frequency welded pipes, the method comprising the following steps:

[0042] Acquisition steps: Synchronously acquire the high-frequency voltage waveform and high-frequency current waveform generated by the high-frequency welding circuit under the joint excitation of the main welding frequency signal and the injected probe signal;

[0043] Feature extraction steps: Based on the high-frequency voltage waveform and high-frequency current waveform, at least one passive electromagnetic fingerprint and one active response fingerprint are extracted and fused to generate a welding state vector; the passive electromagnetic fingerprint includes at least one of instantaneous complex impedance and / or harmonic energy ratio, and the active response fingerprint is the response feature of the high-frequency welding circuit to the probe signal;

[0044] Diagnostic steps: Compare the welding state vector with a preset health model to determine the current welding quality of the high-frequency welded pipe.

[0045] This invention provides an intelligent online monitoring and diagnostic system for the production quality of high-frequency welded pipes. It has the following beneficial effects:

[0046] 1. This invention injects a probe signal of a specific frequency into the high-frequency welding circuit by setting up a perturbation signal injection module, and the feature extraction unit collaboratively extracts passive electromagnetic fingerprints and active response fingerprints. Since the active response fingerprint is extracted from the response to a known probe signal through algorithms such as digital lock-in amplification, this process can effectively suppress broadband noise and environmental electromagnetic interference from the main welding frequency signal. Therefore, it significantly improves the system's sensitivity to detecting subtle changes in the electrical properties of the weld material, enhances the ability to identify early or minor defects, and improves the signal-to-noise ratio and robustness of the diagnosis.

[0047] 2. This invention, by incorporating a defect evolution path tracking module within the intelligent diagnostic unit, constructs a continuous time series, i.e., an evolution path, from discrete instantaneous welding state vectors. By analyzing the dynamic attributes of this path in the feature space, such as velocity and curvature, different defect patterns can be distinguished into point-like, linear, or periodic types. This ability to analyze the dynamic process of defects surpasses traditional single-point, static threshold judgments, providing richer technical information for tracing the root cause of defects (e.g., whether it is a random material problem or a systemic equipment problem), thereby improving the depth and accuracy of diagnosis.

[0048] 3. This invention achieves collaborative diagnostic refinement by establishing a feedback mechanism between the defect evolution path tracking module and the perturbation signal injection module. When the system identifies an uncertain defect pattern, it can proactively adjust or inject specific probe signal sequences for targeted exploration, and confirm or rule out the preliminary diagnosis based on the response results. This closed-loop workflow of discovery-exploration-confirmation avoids the misjudgments that may result from relying solely on passive monitoring, thereby effectively reducing the false alarm rate of diagnostic results and improving the overall confidence and reliability of the system's output conclusions. Attached Figure Description

[0049] Figure 1 This is a structural block diagram of an intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality according to an embodiment of the present invention;

[0050] Figure 2 This is a typical spectrum diagram of the mixed signal acquired in an embodiment of the present invention;

[0051] Figure 3 This is a flowchart of an embodiment of the intelligent diagnostic method for online monitoring of high-frequency welded pipe production quality according to the present invention;

[0052] Figure 4 This is a time series graph of Mahalanobis distance displayed on the system monitoring interface in an embodiment of the present invention.

[0053] Among them, 10 is the perturbation signal injection module; 20 is the data acquisition unit; 30 is the feature extraction unit; and 40 is the intelligent diagnosis unit. Detailed Implementation

[0054] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] See attached document Figure 1 , Figure 1This is a structural block diagram of an intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality according to an embodiment of the present invention. The system is used for online monitoring and diagnosis of the welding production process of high-frequency welded pipes.

[0056] In one specific embodiment, the system includes: a perturbation signal injection module 10, a data acquisition unit 20, a feature extraction unit 30, and an intelligent diagnostic unit 40.

[0057] The perturbation signal injection module 10 is electrically coupled to the high-frequency welding circuit and is used to apply a probe signal to the circuit. The sensing component of the data acquisition unit 20 is coupled to the high-frequency welding circuit and is used to acquire the electrical parameter signals of the circuit. The output terminal of the data acquisition unit 20 is electrically connected to the input terminal of the feature extraction unit 30. The output terminal of the feature extraction unit 30 is electrically connected to the input terminal of the intelligent diagnostic unit 40. In one embodiment, a control output terminal of the intelligent diagnostic unit 40 is connected to a control input terminal of the perturbation signal injection module 10 to form a feedback path.

[0058] Specifically, the function of the perturbation signal injection module 10 is to generate a probe signal of a specific frequency and inject it into the high-frequency welding circuit. The frequency of the probe signal is set to be different from the fundamental frequency and integer multiples of the harmonics of the main welding frequency signal of the high-frequency welding circuit to ensure its distinguishability in the spectrum.

[0059] The function of the data acquisition unit 20 is to synchronously acquire the high-frequency voltage waveform and high-frequency current waveform generated by the high-frequency welding circuit under the joint excitation of the main welding frequency signal and the probe signal. This unit converts the acquired analog waveform signals into digital signal streams and outputs them to subsequent units for processing.

[0060] The feature extraction unit 30 receives a digital signal stream from the data acquisition unit 20. This unit performs algorithmic processing to extract at least one passive electromagnetic fingerprint and one active response fingerprint. The passive electromagnetic fingerprint is based on the analysis of the response to the main welding frequency signal, and the active response fingerprint is based on the analysis of the response to the probe signal. The unit further combines and normalizes the extracted fingerprint features to generate a multi-dimensional welding state vector.

[0061] The intelligent diagnostic unit 40 receives welding state vectors from the feature extraction unit 30. This unit internally stores a preset health model, which is a probability distribution model of state vectors based on historical qualified welded pipe production data. The unit mathematically compares the real-time received welding state vectors with this health model to quantify the degree of deviation from the current welding state, and determines the welding quality of the current welded pipe based on this deviation, ultimately outputting the judgment result.

[0062] See attached document Figure 1 The specific structure and operation of the perturbation signal injection module 10 in the embodiments of the present invention will be described in detail.

[0063] The perturbation signal injection module 10 is configured to generate a probe signal with a stable frequency and amplitude and apply it to the high-frequency welding circuit.

[0064] In one embodiment, the perturbation signal injection module 10 achieves inductive coupling with the induction coil or output busbar of the high-frequency welding circuit through a high-frequency isolation transformer. The secondary winding of the isolation transformer is connected in series or parallel to the high-frequency welding circuit, superimposing the probe signal onto the main welding current. This coupling method provides electrical isolation, preventing the high-power main welding circuit from impacting the low-voltage control circuit within the perturbation signal injection module 10.

[0065] In another embodiment, the perturbation signal injection module 10 can be directly connected to the control system of the high-frequency welding power supply. By modulating the control signal driving the power device of the high-frequency inverter (e.g., pulse width modulation signal), the characteristics of the probe signal are indirectly superimposed on the output voltage or current waveform of the power supply.

[0066] The perturbation signal injection module 10 internally includes a digital frequency synthesizer (DDS), a linear power amplifier, and a control processor.

[0067] The control processor sets the frequency and amplitude of the probe signal according to preset parameters or instructions from the intelligent diagnostic unit 40. A digital frequency synthesizer then generates a high-precision low-voltage sine wave signal. This signal is subsequently amplified by a linear power amplifier, with its output power controlled within a predetermined range (e.g., 1 to 10 watts). This power value is significantly lower than the kilowatt or megawatt-level main output power of the high-frequency welding power supply, ensuring that the injection of the probe signal does not affect the macroscopic welding heat input and welding quality.

[0068] The frequency f of the probe signal p The selection of [the component] follows these principles to ensure the separability of its response signal:

[0069]

[0070] Among them, f w The fundamental frequency of the main welding frequency signal is n, where n is a positive integer.

[0071] This frequency selection principle can be referred to in the appendix. Figure 2 To gain an intuitive understanding. (Attached) Figure 2 This is a typical spectrum diagram of the mixed signal acquired in an embodiment of the present invention. As can be seen from the figure, the signal energy is mainly concentrated at the main welding frequency f. w(e.g., 400kHz as shown in the figure) and its integer multiples of harmonic frequencies (e.g., the second harmonic 2f). w Third harmonic 3f w Several high-amplitude spectral peaks were formed at locations such as (etc.).

[0072] The key to this invention lies in the selected probe signal frequency f. p (For example, 50 kHz as shown in the figure) is intentionally placed in the "spectral gap" between these main harmonic peaks.

[0073] This frequency setting principle avoids the overlap between the spectrum of the probe signal and the spectrum of its response signal, and the spectrum of each harmonic of the main welding frequency signal. This ensures that in the feature extraction unit 30, the active response fingerprint generated solely by the probe signal excitation can be unambiguously locked and extracted through algorithms such as digital phase-locked amplification.

[0074] In normal monitoring mode, the perturbation signal injection module 10 outputs a continuous, single-frequency sine wave as a probe signal. Furthermore, the control processor of the perturbation signal injection module 10 is configured to receive control commands from the intelligent diagnostic unit 40. Upon receiving a specific command, the perturbation signal injection module 10 can switch operating modes and output a specially modulated probe signal sequence, for example, performing a small-range frequency scan (linear frequency sweep signal), or outputting a signal with a specific amplitude envelope for targeted detection of specific defect patterns.

[0075] See attached document Figure 1 The sensor array included in the data acquisition unit 20 in this embodiment of the invention will be described in detail. This sensor array is used to acquire multiphysics measurement data related to the high-frequency welded pipe welding process.

[0076] The data acquisition unit 20 includes a sensor group for measuring electromagnetic signals and a sensor group for measuring external physical signals.

[0077] The electromagnetic signal sensor array is configured to non-invasively measure high-frequency voltage and current in high-frequency welding circuits.

[0078] In one embodiment, the current sensor is a high-bandwidth Rogowski coil wound around a conductor connecting the high-frequency welding power source and the induction coil. The bandwidth of the Rogowski coil is set to cover the frequency from the main welding frequency f... w to probe signal frequency f p Its harmonic response range, for example, has a -3dB bandwidth of not less than 5MHz. The voltage sensor is a high-voltage differential probe with its probe terminals connected in parallel across the induction coil. This probe has a bandwidth matched to that of the current sensor, as well as a high common-mode rejection ratio (CMRR) to suppress the influence of common-mode noise voltage on differential-mode voltage measurements.

[0079] An external physical signal sensor array is configured to acquire thermal and geometric information directly related to weld formation. In one embodiment, thermal field information is acquired by a multi-point infrared thermometer array. Multiple probes in this array are precisely aligned with the apex of the weld's V-shaped zone and the heated areas along the edges of the strip on both sides. Each thermometer is of narrow-band spectral response type to reduce the impact of emissivity fluctuations caused by changes in material surface conditions on the temperature measurement results, and its response time is set to milliseconds or less to capture transient temperature changes during the welding process.

[0080] Geometric information is acquired by a laser profilometer mounted after the welding extrusion rollers and before the weld bead cutting station. This instrument employs the principle of laser triangulation, projecting a laser line across the weld area, with its profile captured by an image sensor. The instrument operates continuously at a preset sampling frequency (e.g., acquiring 2000 profile lines per second), measuring the weld bead's height, width, and cross-sectional shape in real time. Its measurement resolution is set at the micrometer level to detect minute deviations in weld geometry.

[0081] Furthermore, the synchronization and acquisition hardware inside the data acquisition unit 20 in this embodiment of the invention will be described in detail. This hardware is used to ensure strict alignment of multiple heterogeneous signals from the sensor array on a time base and to convert them into digital data with high fidelity.

[0082] In one embodiment, the data acquisition unit 20 employs a network synchronization scheme based on the Precision Time Protocol (PTP, IEEE 1588). This scheme includes an industrial Ethernet switch supporting the PTP protocol and one or more distributed data acquisition modules. The switch or a dedicated device in the network acts as the master clock, broadcasting high-precision time synchronization messages to all data acquisition modules in the network. Each data acquisition module acts as a slave clock, receiving the message and continuously calibrating its local internal clock, ensuring that the clock deviation between all modules and the master clock is controlled within 1 microsecond. When each module acquires sensor signals, it appends a timestamp based on this high-precision synchronization clock to each sampling point or data packet.

[0083] In another embodiment, synchronization is achieved through a central clock distribution system. This system includes a highly stable master oscillator, such as a temperature-compensated crystal oscillator (TCXO), to generate a uniform sampling clock signal and a periodic synchronization trigger signal. These signals are distributed via impedance-matched shielded coaxial cables to the external clock and external trigger inputs of each data acquisition device (e.g., multiple data acquisition cards mounted in the same industrial computer or PXI chassis). This method forces all data acquisition channels to sample on the same clock edge, thereby achieving hardware-level synchronization.

[0084] The data acquisition unit 20 uses a multi-channel, synchronous sampling data acquisition card or module. To ensure complete capture of all information contained in the high-frequency voltage and current waveforms, including the main welding frequency signal, probe signal, and its response, each channel of the acquisition hardware has a sampling rate of no less than 10 MS / s (ten million samples per second) and a vertical resolution of no less than 16 bits. This resolution ensures accurate quantization of smaller signal components against a background of a large-amplitude main signal. The acquisition card transmits the converted digital data with high-precision timestamps to the host computer's memory via a high-speed bus (e.g., PCIe) for subsequent processing by the feature extraction unit 30.

[0085] See attached document Figure 1 The specific process of the feature extraction unit 30 extracting passive electromagnetic fingerprints in this embodiment of the invention will be described in detail. Passive electromagnetic fingerprints are features extracted based on the response of the high-frequency welding circuit to the main welding frequency signal, and this process does not depend on the injected probe signal.

[0086] In one embodiment, the feature extraction unit 30 receives synchronized high-frequency voltage waveform digital signal v(t) and high-frequency current waveform digital signal i(t) from the data acquisition unit 20.

[0087] To obtain the instantaneous complex impedance Z eq (t), the feature extraction unit 30 first performs Hilbert transform on the received v(t) and i(t) respectively to construct their corresponding analytic signals. The analytic signal v of the voltage. a (t) and the analytic signal i of the current a The expression for (t) is:

[0088]

[0089]

[0090] in, Let j denote the Hilbert transform operator, where j is the imaginary unit. The instantaneous complex impedance Z can be obtained by calculating the ratio of two analytic signals. eq(t):

[0091]

[0092] The real part R of the calculation result eq (t) is the equivalent resistance, and its value varies with the Joule heat dissipation in the welded area. The imaginary part X eq (t) is the equivalent reactance, the value of which varies with the geometry of the molten pool and the change in the permeability of the material.

[0093] In order to obtain the harmonic energy ratio HER n (t), the feature extraction unit 30 performs a short-time Fourier transform (STFT) on the high-frequency current waveform digital signal i(t). This process includes:

[0094] The continuous i(t) signal stream is divided into short-time analysis windows with a certain overlap rate;

[0095] Apply a window function (e.g., Hanning window) to the signal within each analysis window to suppress spectral leakage;

[0096] Perform a Fast Fourier Transform (FFT) on the windowed signal segment to obtain the discrete-time spectrum I(ω,t) corresponding to that time period.

[0097] Feature extraction unit 30 extracts the main welding frequency f from the calculated time spectrum I(ω,t). w The energy of the fundamental component and the energies of its harmonic components (e.g., n = 2, 3, 5) are calculated, and their ratios are determined. The energy ratio of the nth harmonic to HRE is given. n The formula for calculating (t) is:

[0098]

[0099] The numerical change of this feature reflects the change in the degree of nonlinearity of the system caused by physical phenomena such as electric arc and plasma discharge during the welding process. The feature extraction unit 30 repeats this calculation for each analysis window over time, thereby generating a series of time-varying passive electromagnetic fingerprint feature values.

[0100] Furthermore, the specific process of the feature extraction unit 30 extracting the active response fingerprint in this embodiment of the invention will be described in detail. The active response fingerprint is a feature extracted based on the response of the high-frequency welding circuit to a known probe signal. This process employs a digital lock-in amplification algorithm to separate the weak probe signal response component from a background containing strong main frequency signals and broadband noise.

[0101] Feature extraction unit 30 receives the synchronous high-frequency current waveform digital signal i(t) from data acquisition unit 20, and obtains the frequency f of the current probe signal from perturbation signal injection module 10 or system preset parameters. p .

[0102] The first step of the algorithm is to generate two mutually orthogonal digital reference signals, namely, the in-phase reference signal s. ref，I (t) and orthogonal reference signal s ref，Q (t). The frequencies of these two signals are the same as the probe signal frequency f. p The same applies; its mathematical expression is:

[0103] s ref，I (t)=sin(2πf p t);

[0104] s ref，Q (t)=cos(2πf p t);

[0105] The second step of the algorithm is frequency mixing. The feature extraction unit 30 multiplies the input current signal i(t) with the two reference signals point by point to obtain the two mixed signals. This operation converts the input signal with frequency f... p The component is down-converted to DC (0Hz) and doubled (2f) frequency. p ) place.

[0106] The third step of the algorithm is low-pass filtering. The feature extraction unit 30 passes the two mixed signals through a digital low-pass filter to remove the second harmonic component and other high-frequency noise components, retaining only the DC component. This process is equivalent to... int The integral operation on the surface ultimately yields the in-phase DC component C. I (t) and the orthogonal DC component C Q (t). Its mathematical expression is:

[0107]

[0108] Wherein, the integration period T int The choice of T is related to the cutoff frequency of the low-pass filter. int The value of needs to be much larger than the period 1 / f of the probe signal. p .

[0109] The fourth step of the algorithm is to calculate the amplitude and phase. Based on the obtained in-phase and quadrature DC components, the feature extraction unit 30 calculates the amplitude attenuation A of the probe signal response. p (t) and phase shift φ p (t). These two together constitute the active response fingerprint. Its mathematical expression is:

[0110]

[0111] φ p (t)=atan2(C Q (t), C I (t));

[0112] Where atan2 is a two-parameter arctangent function.

[0113] Feature extraction unit 30 continues to execute the above steps to generate A that changes over time. p (t) and φ p (t) Eigenvalue sequence. In another embodiment, only the high-frequency voltage waveform v(t) can be used as input, or both v(t) and i(t) can be used for calculation.

[0114] After calculating the passive electromagnetic fingerprint and the active response fingerprint, the feature extraction unit 30 performs a fusion step to construct a unified welding state vector. This vector serves as the mathematical input for subsequent analysis by the intelligent diagnostic unit 40.

[0115] First, the feature extraction unit 30 collects all the calculated feature values ​​at the same time point t.

[0116] In one embodiment, the set includes the equivalent resistance R in a passive electromagnetic fingerprint. eq (t), equivalent reactance X eq (t) and the second harmonic energy ratio HER2(t), and the amplitude attenuation A in the active response fingerprint. p (t) and phase shift φ p (t). In embodiments that include external physical signal sensors, features extracted from thermal field information or geometric information (e.g., weld peak temperature, weld width) are also included in this set.

[0117] Because the features in the set have different physical units and numerical ranges (e.g., ohms, radians, dimensionless ratios), direct combination can lead to some features having unbalanced weights in the subsequent mathematical model. Therefore, the feature extraction unit 30 normalizes each feature sequence. In one specific embodiment, the Z-score normalization method is used. For any feature value x... i Its standardized value x′ i The calculation is as follows:

[0118]

[0119] Where, μ i It is the mean of a large number of samples collected under normal historical production conditions for this feature, σ iThis is its corresponding standard deviation. These two parameters (μ) i ,σ i The statistic is pre-calculated and stored in the system as a baseline statistic for that feature.

[0120] Finally, the feature extraction unit 30 collects all the standardized feature values ​​x′ at time point t. i Arranged in a predetermined order, they form a multi-dimensional column vector. This vector is defined as the welding state vector S(t) at that moment. Its structure is as follows:

[0121]

[0122] This welding state vector S(t) provides a dimensionless, multi-dimensional quantitative description of the current welding process and is transmitted to the intelligent diagnostic unit 40 as the final output of the feature extraction unit 30.

[0123] See attached document Figure 1 The specific functions of the intelligent diagnostic unit 40 in this embodiment of the invention will be described in detail. The core function of this unit is to perform abnormal detection based on a preset health model.

[0124] The intelligent diagnostic unit 40 receives a continuous welding state vector sequence S(t) from the feature extraction unit 30. This unit internally stores a health model, which is a mathematical description of the statistical distribution of welding state vectors during the production of qualified welded pipes. The model is built in an offline training phase. Specifically, firstly, within a production cycle confirming the production of defect-free welded pipes, a large number of welding state vector samples are collected, forming a health state sample set {S}. H}. Then, the mean vector μ of this sample set is calculated. H The sum of the covariance matrix ∑ H These two factors together constitute the health model in this embodiment. The calculation formula is as follows:

[0125]

[0126] Where N is the total number of samples in the healthy state sample set, and S H，k It is the k-th health state vector in the sample set. The mean vector μ H The covariance matrix ∑ represents the central location of the health status. H It describes the fluctuation range of each characteristic component under healthy conditions and their linear correlation with each other.

[0127] During the online diagnostic phase, the intelligent diagnostic unit 40 calculates the Mahalanobis distance D between each real-time received welding state vector S(t) and the established health model. MMahalanobis distance is an efficient multidimensional spatial distance measure that considers the correlation between features and is insensitive to the scale of the features. Its calculation formula is as follows:

[0128]

[0129] Where S(t) is the current welding state vector, μ H and ∑ H These are the mean vector and covariance matrix of the health model, respectively. It is the inverse of the covariance matrix. The calculated D... M (t) is a scalar value that quantifies the statistical distance of the current state from the center of the healthy state.

[0130] The intelligent diagnostic unit 40 will calculate the Mahalanobis distance D M (t) and a preset decision threshold D threshold The comparison is performed. This threshold is determined based on the Mahalanobis distance distribution of the healthy sample set (theoretically following a chi-square distribution) and according to a preset confidence level (e.g., 99.7%). If D M (t)>D threshold If D is abnormal, the current welding state is determined to be abnormal, and the system generates an abnormal state signal; M (t)≤D threshold If the condition is met, the current welding status is determined to be normal. This determination result is output to the subsequent defect evolution path tracking module for further analysis, or it can be displayed directly on the human-computer interaction interface.

[0131] Furthermore, the function of the defect evolution path tracking module within the intelligent diagnostic unit 40 is described in detail. This module is used to perform in-depth dynamic analysis of the welding state sequence that passes anomaly detection in order to identify the fundamental pattern of defects.

[0132] This module receives and stores a series of welding state vectors S(t) from previous processing steps. In one embodiment, the module maintains a first-in-first-out (FIFO) circular buffer to store the most recent M welding state vectors. This time-ordered set of vectors {S(t-M+1), ..., S(t)} forms a discrete trajectory in the multidimensional feature space, which is the defect evolution path.

[0133] This module performs quantitative calculations of the geometric and dynamic properties of the evolution path. In one embodiment, this module calculates the path's velocity and curvature. The instantaneous velocity vector v(t) of the path at time t is approximated by performing a first-order difference on the state vector:

[0134]

[0135] Here, Δt is the time interval between two consecutive state vectors. The magnitude of this velocity vector, i.e., the rate ||v(t)||, characterizes the drastic degree of change in the welding state.

[0136] The curvature of a path characterizes the degree of its bending, reflecting the rate of change of the direction of state change. In one embodiment, the module first calculates the path's acceleration vector a(t), which is the first-order difference of the velocity vector:

[0137]

[0138] Subsequently, the curvature of the path is calculated based on the velocity and acceleration vectors. In another embodiment, the curvature is approximated by calculating the rate of change of the angle between the continuous velocity vectors v(t) and v(t-1).

[0139] This module classifies defect patterns based on the time series of the aforementioned dynamic attributes:

[0140] If an isolated, high-amplitude pulse is detected in both the velocity (v(t)) and curvature, followed by a rapid return to the baseline level, the module classifies this event as a point defect. This corresponds to an evolutionary path briefly darting out of a healthy region and then quickly returning.

[0141] If a path is detected migrating from a healthy region to a quasi-stable abnormal region at a high rate, and then continuing to operate at a low rate in that region, the module classifies this event as a linear defect. This corresponds to a persistent shift in the weld state.

[0142] If a path is detected to exhibit a periodic trajectory in the feature space, such as a closed loop or oscillation, and its dynamic properties, such as velocity and curvature, also exhibit corresponding periodic changes, then this module classifies this event as a periodic defect. This module can determine the characteristic frequencies of this periodic change by performing spectral analysis on the time series of one or more components of the state vector.

[0143] Furthermore, the collaborative diagnostic refinement mechanism within the intelligent diagnostic unit 40 in this embodiment of the invention is described in detail. This mechanism utilizes the control feedback path from the intelligent diagnostic unit 40 to the perturbation signal injection module 10.

[0144] The activation condition for this mechanism is when the defect evolution path tracking module is uncertain about the classification result of the current welding state. This uncertain state is defined as the degree of matching between the dynamic attributes of the evolution path (e.g., velocity, curvature) and any preset defect pattern (point-like, linear, periodic) is lower than a preset confidence threshold.

[0145] Under these conditions, the intelligent diagnostic unit 40 automatically generates and sends a probe command to the perturbation signal injection module 10. This command contains parameters for a specific probe signal sequence. Upon receiving the command, the perturbation signal injection module 10 pauses the output of the conventional single-frequency sine wave probe signal and instead generates and injects the probe signal sequence specified in the command.

[0146] In one embodiment, the probe signal sequence is a linear sweep frequency signal. The frequency f of the signal... p (t) within a predetermined time period T sweep Within, from an initial frequency f start Linear variation to a termination frequency f end :

[0147]

[0148] In another embodiment, the probe signal sequence is a polyphonic signal, which is a linear superposition of multiple sine waves of different frequencies.

[0149] During the target exploration, the data acquisition unit 20 and the feature extraction unit 30 operate continuously. The feature extraction unit 30 uses a digital lock-in amplification algorithm to calculate the system's response to the frequency sweep signal, thus obtaining the response amplitude A as a function of frequency. p (f) and response phase φ p (f).

[0150] The intelligent diagnostic unit 40 receives the frequency response feature and compares it with a preset targeted defect feature library. The feature library stores known frequency response curve features (e.g., resonance peaks or absorption valleys appearing at specific frequency points) corresponding to specific physical defects (e.g., internal cracks or inclusions of a specific size).

[0151] If the frequency response curve acquired in real time matches an entry in the feature library, the intelligent diagnostic unit 40 confirms the initial uncertain diagnostic result as the specific defect type corresponding to that entry. If the response curve does not match any defect entry, the intelligent diagnostic unit 40 excludes the initial anomaly determination. After this process is completed, the intelligent diagnostic unit 40 sends a command to the perturbation signal injection module 10 to restore the output of the normal single-frequency probe signal.

[0152] In summary, this invention provides an intelligent diagnostic system and method for online monitoring of high-frequency welded pipe production quality. The system actively injects a low-energy, frequency-specific probe signal into the high-frequency welding circuit and simultaneously acquires high-frequency voltage and current waveforms, as well as external physical field signals, generated under the combined excitation of the main welding signal and the probe signal, thus achieving multi-dimensional information acquisition of the welding process. The feature extraction unit within the system uses a specialized algorithm to extract passive electromagnetic fingerprints and active response fingerprints from the responses to the main signal and probe signal, respectively, and fuses them into a unified, standardized welding state vector. The intelligent diagnostic unit utilizes a probabilistic model based on historical health data to quickly detect abnormal states by calculating the Mahalanobis distance between the real-time state vector and the model. Furthermore, the system analyzes the evolution path of the state vector in the feature space to classify the dynamic patterns of defects and triggers a closed-loop targeted detection mechanism. By changing the form of the probe signal, it accurately identifies specific defects, thereby improving the accuracy and reliability of the diagnosis.

[0153] See attached document Figure 3 , Figure 3 This is a flowchart of an intelligent diagnostic method for online monitoring of high-frequency welded pipe production quality according to an embodiment of the present invention. In one specific embodiment, the method may include the following steps:

[0154] S100 injects a probe signal of a preset frequency into the high-frequency welding circuit and simultaneously acquires the high-frequency voltage waveform and high-frequency current waveform generated by the high-frequency welding circuit under the joint excitation of the main welding frequency signal and the probe signal.

[0155] This step is performed by the perturbation signal injection module 10 and the data acquisition unit 20. Current and voltage signals are acquired through Rogowski coils and high-voltage differential probes, respectively, and the multiple analog signals are converted into digital signal streams with high-precision timestamps by acquisition hardware based on Precision Time Protocol (PTP) or a central clock distribution system at a sampling rate of not less than 10 MS / s.

[0156] S200, based on the high-frequency voltage waveform and high-frequency current waveform, extract at least one passive electromagnetic fingerprint and one active response fingerprint, and fuse them to generate a welding state vector.

[0157] This step is performed by the feature extraction unit 30. First, the instantaneous complex impedance is calculated by performing a Hilbert transform on the voltage and current waveforms, and the harmonic energy ratio is calculated by performing a short-time Fourier transform on the current waveform, thereby obtaining the passive electromagnetic fingerprint. Simultaneously, a digital lock-in amplification algorithm is used to lock and extract the amplitude attenuation and phase shift of the probe signal response component from the voltage and / or current waveforms, thereby obtaining the active response fingerprint. Subsequently, all extracted feature values ​​are Z-score normalized and arranged in a predetermined order to form a multidimensional, dimensionless welding state vector.

[0158] S300, compare the welding state vector with the preset health model to determine the current high-frequency welded pipe welding quality.

[0159] This step is performed by the intelligent diagnostic unit 40. The degree to which the current welding state deviates from normal operating conditions is quantitatively assessed by calculating the Mahalanobis distance between the current welding state vector and a mean vector and covariance matrix (i.e., a health model) established based on historical health data. This Mahalanobis distance is compared with a preset judgment threshold; if the distance exceeds the threshold, the current welding quality is determined to be abnormal.

[0160] In a further embodiment, after step S300, the method further includes:

[0161] For state vector sequences identified as anomalous, their evolutionary paths in a multidimensional feature space are traced. By calculating dynamic properties such as instantaneous velocity and curvature of this path, the defect patterns are classified into one of the following: point defects, linear defects, or periodic defects, in order to make a preliminary judgment on the nature of the defects.

[0162] In a further embodiment, if the classification result of the aforementioned defect pattern is uncertain, a collaborative diagnostic refinement step is triggered:

[0163] The intelligent diagnostic unit 40 sends a control command to the perturbation signal injection module 10, causing it to inject a specific probe signal sequence, such as a linear frequency sweep signal. By analyzing the frequency response characteristics of this specific sequence and comparing it with a preset defect feature library, the defect type can be accurately identified.

[0164] To more clearly illustrate the technical solution of the present invention, a specific application embodiment is given below in conjunction with the accompanying drawings.

[0165] Example:

[0166] On a production line manufacturing high-frequency straight seam welded pipe (ERW) conforming to API 5L standard, with an outer diameter of 114 mm and a wall thickness of 4.0 mm, the main welding frequency f of the high-frequency welding power source is... wThe frequency was 400 kHz, and the steel pipe's traveling speed was 60 meters per minute. The online monitoring and intelligent diagnostic system of this invention was deployed on this production line.

[0167] Step 1: System Initialization and Normal Monitoring

[0168] After the system starts up, the perturbation signal injection module 10 begins to work, injecting a frequency f into the induction coil of the high-frequency welding circuit. p A sinusoidal probe signal of 50kHz and 5W power. This frequency avoids the harmonics of the main welding frequency.

[0169] The Rogowski coil and high-voltage differential probe in the data acquisition unit 20 continuously acquire high-frequency current and voltage signals, and digitize them at a sampling rate of 20 MS / s via a synchronous acquisition card. The feature extraction unit 30 calculates the welding state vector S(t) in real time. In this embodiment, this vector is a five-dimensional vector, including the standardized equivalent resistance R. eq Equivalent reactance X eq Second harmonic energy ratio HER2, probe response amplitude attenuation A p and phase shift φ p .

[0170] The intelligent diagnostic unit 40 receives the vector sequence and calculates its Mahalanobis distance D with the preset health model (established based on 10,000 previously generated qualified samples). M (t).

[0171] See attached document Figure 4 , Figure 4 This is a time series graph of the Mahalanobis distance displayed on the system monitoring interface in this embodiment. Under normal production conditions, the welding state is stable, and the calculated Mahalanobis distance D... M (t) (shown by the solid line in the figure) fluctuates within a small range at a relatively low level, always remaining within the preset anomaly detection threshold D. threshold (As shown by the dashed line in the figure) below.

[0172] Step 2, Defect Occurrence and Anomaly Detection:

[0173] At a certain point in production, an extrusion roller used for weld seam pressing begins to exhibit slight, periodic radial runout due to bearing wear. This mechanical vibration causes the convergence angle of the V-zone and the extrusion pressure to also change periodically at the same frequency. This is a typical source of a "periodic defect" that is not easily detected by traditional methods.

[0174] This periodic physical change is immediately reflected in the electromagnetic fingerprint:

[0175] Passive fingerprint: Periodic changes in the geometry of the V-shaped region lead to equivalent reactance X eqPeriodic fluctuations occur; changes in extrusion pressure affect contact resistance and molten pool condition, leading to an increase in equivalent resistance R. eq The harmonic energy ratio HER2 also oscillates accordingly.

[0176] Active fingerprinting: The periodic change in the overall impedance of the welding circuit also modulates the transmission path of the 50kHz probe signal, causing its response amplitude A to... p and phase φ p It also exhibits the same periodic fluctuations.

[0177] The synchronous periodic changes of these features cause the welding state vector S(t) to begin deviating from the healthy central region in the feature space. Therefore, the calculated Mahalanobis distance D... M (t) also began to rise and fall periodically. (See attached diagram) Figure 4 As shown in the "Defect Occurrence" area, D M The peak value of (t) repeatedly exceeded the anomaly detection threshold D. threshold The system immediately determined that the current welding status was abnormal.

[0178] Step 3: Defect Pattern Recognition and Diagnosis. Upon detecting an anomaly, the intelligent diagnostic unit 40 immediately activates the defect evolution path tracing module. This module analyzes the trajectory formed by the state vector S(t) sequence over a recent period (e.g., the last 2 seconds) in the five-dimensional feature space.

[0179] The defect evolution path tracing module, by calculating the velocity and curvature of the trajectory, discovered that it exhibits significant periodic characteristics. The module further analyzes any component of the state vector (e.g., X...) eq The frequency of the periodic fluctuation was calculated to be approximately 1.2 Hz by performing a Fast Fourier Transform on the time series (t).

[0180] Based on this, the system classifies the defect pattern as "periodic defect" and pushes a precise diagnostic message to the operator's monitoring interface: "Warning: Periodic welding defect detected. Defect frequency: approximately 1.2Hz. Possible cause: mechanical vibration. Please check the condition of the extrusion rollers and forming rollers."

[0181] Following the clear instructions, the operator quickly located the compression roller experiencing radial runout and replaced its worn bearing. After troubleshooting, see attached... Figure 4 As shown in the "Defect Elimination" region, the Mahalanobis distance D M (t) quickly falls below the threshold, and the system returns to normal monitoring status.

[0182] This application example demonstrates how the present invention, through active-passive fingerprint fusion, can not only promptly detect welding quality anomalies, but also classify and trace the root causes of defects by deeply analyzing the defect evolution path, thereby greatly improving the efficiency and accuracy of fault diagnosis and ensuring the production quality of high-frequency welded pipes.

[0183] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-frequency welded pipe production quality online monitoring and intelligent diagnostic system, characterized in that, include: The data acquisition unit is configured to synchronously acquire high-frequency voltage waveforms and high-frequency current waveforms in the high-frequency welding circuit. The high-frequency voltage waveforms and high-frequency current waveforms are generated by the high-frequency welding circuit under the joint excitation of the main welding frequency signal and the injected probe signal. The feature extraction unit, connected to the data acquisition unit, is configured to extract at least one passive electromagnetic fingerprint and one active response fingerprint based on the high-frequency voltage waveform and high-frequency current waveform, and fuse them to generate a welding state vector; the passive electromagnetic fingerprint includes at least one of instantaneous complex impedance and / or harmonic energy ratio, and the active response fingerprint is the response feature of the high-frequency welding circuit to the probe signal; The intelligent diagnostic unit, connected to the feature extraction unit, is configured to compare the welding state vector with a preset health model to determine the current high-frequency welded pipe welding quality. A perturbation signal injection module, coupled to the high-frequency welding circuit, is configured to inject the probe signal into the high-frequency welding circuit; the frequency of the probe signal is different from the fundamental frequency and integer multiples of the harmonic frequency of the main welding frequency signal. The intelligent diagnostic unit includes: The defect evolution path tracking module is configured to store a time-series welding state vector to form an evolution path and to identify defect patterns based on the dynamic properties of the evolution path. The defect evolution path tracking module identifies defect patterns based on the dynamic attributes of the evolution path, and is specifically configured as follows: The geometric and dynamic properties of the evolution path in the multidimensional feature space are analyzed, including the velocity, curvature and direction changes of the trajectory, in order to classify the defect pattern into one of point defects, linear defects or periodic defects. The system is also configured to: When the defect evolution path tracking module identifies a preset defect pattern, it triggers the perturbation signal injection module to adjust or inject a specific probe signal sequence to target and confirm the defect pattern.

2. The intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality according to claim 1, characterized in that, The feature extraction unit extracts the passive electromagnetic fingerprint, specifically configured as follows: The ratio between the high-frequency voltage waveform and the high-frequency current waveform is calculated to obtain the instantaneous complex impedance, which includes the equivalent resistance characterizing energy dissipation and the equivalent reactance characterizing electromagnetic energy storage. and / or The high-frequency current waveform is subjected to spectral analysis to obtain the harmonic energy ratio, which characterizes the degree of nonlinearity in the welding process.

3. The intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality according to claim 1, characterized in that, The feature extraction unit extracts the active response fingerprint, specifically configured as follows: A digital lock-in amplification algorithm is used to lock and extract the response component caused by the probe signal from the high-frequency current waveform and / or high-frequency voltage waveform to obtain the active response fingerprint; The active response fingerprint includes amplitude attenuation and / or phase shift of the response component.

4. The intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality according to claim 1, characterized in that, The intelligent diagnostic unit compares the welding state vector with a preset health model, specifically configured as follows: The health model is a probability distribution model established based on multiple historical welding state vectors collected during the production of qualified welded pipes. The intelligent diagnostic unit quantifies the degree of deviation by calculating the Mahalanobis distance between the current welding state vector and the health model, and determines the welding quality based on the degree of deviation.

5. The intelligent diagnostic system for online monitoring of high-frequency welded pipe production quality according to claim 1, characterized in that, The data acquisition unit is also configured to synchronously acquire at least one external physical signal, which includes at least one of the thermal field information of the weld area or the geometric information of the weld bead. The feature extraction unit is also configured to integrate the features extracted from the external physical signal into the welding state vector.

6. A method for online monitoring and intelligent diagnosis of high-frequency welded pipe production quality, based on the system described in any one of claims 1-5, characterized in that, Includes the following steps: The high-frequency voltage and current waveforms generated by the high-frequency welding circuit under the joint excitation of the main welding frequency signal and the injected probe signal are acquired synchronously. Based on the high-frequency voltage waveform and high-frequency current waveform, at least one passive electromagnetic fingerprint and one active response fingerprint are extracted and fused to generate a welding state vector; the passive electromagnetic fingerprint includes at least one of instantaneous complex impedance and / or harmonic energy ratio, and the active response fingerprint is the response characteristics of the high-frequency welding circuit to the probe signal; The welding state vector is compared with a preset health model to determine the current welding quality of the high-frequency welded pipe.

Images