Electric control ball valve and ball valve fault diagnosis system thereof

Abstract

The invention discloses an electric control ball valve and a ball valve fault diagnosis system thereof, belongs to the technical field of electric control ball valves, and solves the problems that high-frequency fine adjustment causes slow deterioration of a valve, annual increase of torque, visual recession of a self-adaptive algorithm as working condition fluctuation, threshold widening, drifting filtering, software and hardware negative cooperation for covering abrasion, early warning failure, and valve sudden jamming or internal leakage. Comprising a valve body, an electric actuator, a valve rod, a valve ball, a power coupling module and a synchronous acquisition unit, wherein the electric actuator is mounted on the valve body through a connecting seat; the valve rod is in transmission connection with an output shaft of the electric actuator and extends into the valve body; the valve ball is fixedly connected to the lower end of the valve rod and is positioned in the valve body; according to the invention, through weak electric signal excitation, monitoring and building of healthy fingerprints, continuous comparison of stripping working conditions, wear identification, dual-reference algorithm drift prevention, real evaluation and active intervention, misjudgment is completely eradicated.

Description

Technical Field

[0001] This invention relates to the field of electrically controlled ball valve technology, and in particular to an electrically controlled ball valve and its fault diagnosis system. Background Technology

[0002] In continuous production processes such as petrochemicals, electrically controlled ball valves are suitable for controlling the feed of ultra-high pressure polymerization reactions. Their structure features a robust valve body, a high-rigidity valve stem, and reinforced seals to withstand erosion and wear under harsh operating conditions. The drive unit is an integrated electric actuator with a built-in multi-stage gear reduction system, ensuring sufficient output torque for precise small-aperture control.

[0003] The intelligence of this electrically controlled ball valve is reflected in its circuit system. Its control module not only receives standard control signals but also continuously collects current and angle data from high-precision torque and position sensors. These real-time operating parameters, together with the valve's unique mechanical characteristics, constitute a dynamic health model. Through built-in algorithms, the valve learns and compares data trends, ideally identifying early signs of deterioration such as torque creep or positioning deviation caused by slow wear, thus issuing a warning before actual mechanical failure occurs.

[0004] In other words, the valve is in a harsh operating condition of high frequency and minute adjustment for a long time. Its valve seat and valve stem and other components will undergo extremely slow and progressive physical deterioration due to media erosion and fretting friction, which is manifested as a slow increase in driving torque year by year.

[0005] However, adaptive health diagnostic algorithms designed to improve system robustness may misinterpret this linear trend representing equipment degradation as normal operating condition fluctuations, thus silently relaxing alarm thresholds. At the same time, electrical deviations such as minor signal drift from position sensors are also treated as noise by the algorithm's filtering function.

[0006] This dangerous negative synergy between software intelligent adaptation and slow hardware wear and tear causes fundamental mechanical health deterioration to be masked and certified as normal by advanced digital models, resulting in severe distortion of condition perception and complete failure of preventive maintenance and early warning functions. Ultimately, when the accumulated physical wear and tear reaches a critical point, the electrically controlled ball valve will suddenly jam or leak severely without warning, directly causing critical production process interruptions, loss of control, or even safety accidents.

[0007] Therefore, an electrically controlled ball valve and its fault diagnosis system are proposed to solve or alleviate the above problems. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of existing technologies by proposing an electrically controlled ball valve and its fault diagnosis system.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: An electrically controlled ball valve includes a valve body, an electric actuator mounted thereon via a connecting seat, a valve stem that is drivenly connected to the output shaft of the electric actuator and extends into the valve body, a valve ball fixedly connected to the lower end of the valve stem and located within the valve body, a power coupling module connected in series with the motor power supply circuit in the electric actuator, and a synchronous acquisition unit connected in parallel. The synchronous acquisition unit acquires the voltage transient response at the motor drive end of the electric actuator and the current transient response in the power supply circuit in parallel using a precision clock as a reference, and converts these two analog signals into digital signals with high fidelity. The power coupling module receives weak diagnostic signals, linearly amplifies them, and injects them directly into the power supply circuit of the motor in the electric actuator through a broadband isolation transformer.

[0010] Preferably, the power coupling module includes a power operational amplifier and an isolation coupler. The non-inverting input of the power operational amplifier serves as the signal input of the power coupling module. The output of the power operational amplifier is connected to the first end of the primary winding of the isolation coupler. The second end of the primary winding of the isolation coupler is grounded through a sampling resistor. The inverting input of the power operational amplifier is connected to its output through a feedback resistor. The two ends of the secondary winding of the isolation coupler serve as the power output of the power coupling module.

[0011] Preferably, the synchronous acquisition unit includes an analog-to-digital converter (ADC), a differential amplifier, and a current sensor. The non-inverting and inverting inputs of the differential amplifier serve as the voltage sampling terminals of the synchronous acquisition unit. The output of the differential amplifier is connected to the first channel non-inverting analog input of the ADC. The first channel inverting analog input of the ADC is connected to a reference voltage. The positive output of the current sensor is connected to the second channel non-inverting analog input of the ADC. The negative output of the current sensor is connected to the second channel inverting analog input of the ADC. The serial data output, serial clock input, and chip select terminals of the ADC serve as the data output of the synchronous acquisition unit. The master clock input of the ADC serves as one of the synchronous clock outputs of the synchronous acquisition unit.

[0012] Preferably, the synchronous acquisition unit further includes a clock generator, a clock buffer, and a multi-output isolated power supply module. The output terminal of the clock generator is connected to the input terminal of the clock buffer. The first output terminal of the clock buffer serves as the synchronous clock output terminal of the synchronous acquisition unit and is connected to the main clock input terminal of the analog-to-digital converter. The power input terminal of the multi-output isolated power supply module is powered on. The first regulated output terminal of the multi-output isolated power supply module serves as the isolated power output terminal of the synchronous acquisition unit. The second regulated output terminal of the multi-output isolated power supply module is connected to the analog power supply terminals of the differential amplifier and the analog-to-digital converter. The third regulated output terminal of the multi-output isolated power supply module is connected to the digital power supply terminal of the analog-to-digital converter.

[0013] This invention also provides a ball valve fault diagnosis system, including the electrically controlled ball valve as described above, a signal injection unit, and a core processing unit. The signal injection unit receives instructions from the core processing unit to accurately generate and output a composite test signal composed of a multi-frequency sine wave and a pseudo-random sequence as a diagnostic excitation source. The signal output terminal of the signal injection unit is connected to the signal input terminal of the power coupling module. The power output terminal of the power coupling module is connected in series with the power supply circuit of the motor in the electric actuator through an isolation coupler. The voltage sampling terminal of the synchronous acquisition unit is connected in parallel with the power supply circuit of the motor in the electric actuator. The current sampling terminal of the synchronous acquisition unit is connected to the electric actuator... The power supply circuit of the motor is connected in series. The synchronous clock output terminal of the synchronous acquisition unit is connected to the synchronous clock input terminal of the signal injection unit. The isolation power output terminal of the synchronous acquisition unit is connected to the power input terminal of the signal injection unit. The data output terminal of the synchronous acquisition unit is connected to the data input terminal of the core processing unit. The diagnostic result output terminal of the core processing unit is connected to the host computer. The core processing unit receives the voltage and current data uploaded by the synchronous acquisition unit to perform high-speed parallel processing to extract multi-dimensional physical features. It compares and analyzes these features with the pre-stored gold fingerprint database to calculate the health index and complete the fault diagnosis decision. Finally, it outputs the diagnostic result and / or control command.

[0014] Preferably, the signal injection unit includes a direct digital frequency synthesizer, a first operational amplifier, an active filter, and an analog switch. The reference clock input of the direct digital frequency synthesizer serves as the synchronization clock input of the signal injection unit. The reference clock input of the direct digital frequency synthesizer is connected to the synchronization clock output of the synchronization acquisition unit. The first current output of the direct digital frequency synthesizer is connected to the inverting input of the first operational amplifier through a first resistor. The second current output of the direct digital frequency synthesizer is connected to the inverting input of the first operational amplifier through a second resistor. The non-inverting input of the first operational amplifier is grounded. The output of the first operational amplifier is connected to its inverting input through a third resistor. The output of the first operational amplifier is connected to the signal input of the active filter. The signal output of the active filter is connected to the first channel signal input of the analog switch. The second channel signal input of the analog switch is connected to the digital signal output of the core processing unit. The common output of the analog switch serves as the signal output of the signal injection unit.

[0015] Preferably, the core processing unit includes a field-programmable gate array (FPGA) and an embedded processor. Multiple general-purpose input / output terminals of the FPGA serve as data input terminals of the core processing unit and are respectively connected to the serial data output terminal, serial clock input terminal, and chip select terminal of the analog-to-digital converter (ADC). The FPGA is connected to the embedded processor via an internal bus, and the serial communication peripheral interface of the embedded processor is connected to the diagnostic result output terminal of the core processing unit via a communication interface.

[0016] Preferably, the core processing unit further includes a non-volatile memory, and a digital isolator is provided between the serial communication peripheral interface of the embedded processor and the signal input terminal of the communication interface. The serial clock terminal and the serial data input / output terminal of the non-volatile memory are respectively connected to the corresponding serial peripheral interface of the embedded processor.

[0017] The present invention has the following beneficial effects: This invention actively injects a specific weak electrical signal into the motor of the operating electric actuator and simultaneously monitors its electrical response. It learns and memorizes the "fingerprint" of the ball valve in a healthy state as a benchmark, and then continuously repeats this process during operation to extract deep electrical characteristics that reflect the mechanical state. These characteristics are compared with the original benchmark to remove operating condition interference and calculate health residuals. Then, the trend and internal correlation of the residuals are analyzed to identify wear patterns. At the same time, a dual-benchmark protection strategy is used to resist the drift of the algorithm itself. Finally, the true health state of the ball valve is comprehensively evaluated, and the host computer system can be actively intervened when a problem is detected to prevent misjudgment. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the structure of the electrically controlled ball valve in this invention; Figure 2 This is a structural block diagram of the ball valve fault diagnosis system in this invention.

[0020] 1. Valve body; 2. Connecting seat; 3. Electric actuator; 4. Signal injection unit; 5. Power coupling module; 6. Synchronous acquisition unit; 7. Core processing unit. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0022] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0023] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0024] In the description of this invention, it should be understood that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use, or the orientation or positional relationship commonly understood by those skilled in the art. They are only used to facilitate the description of this invention and to simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0025] Furthermore, the terms "first," "second," and "third" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0026] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0027] An electrically controlled ball valve, such as Figure 1 and Figure 2 As shown, the device includes a valve body 1, an electric actuator 3 mounted on it via a connecting seat 2, a valve stem that is drivenly connected to the output shaft of the electric actuator 3 and extends into the valve body 1, a valve ball that is fixedly connected to the lower end of the valve stem and located in the valve body 1, a power coupling module 5 connected in series with the motor power supply circuit in the electric actuator 3, and a synchronous acquisition unit 6 connected in parallel. The synchronous acquisition unit 6 uses a precision clock as a reference to acquire the voltage transient response on the motor drive end of the electric actuator 3 and the current transient response in the power supply circuit in parallel, and converts these two analog signals into digital signals with high fidelity. The power coupling module 5 receives the weak diagnostic signal, linearly amplifies it, and injects it directly into the power supply circuit of the motor in the electric actuator 3 through a broadband isolation transformer.

[0028] The power coupling module 5 includes a power operational amplifier LME49810 and an isolation coupler. The non-inverting input of the power operational amplifier serves as the signal input of the power coupling module 5. The output of the power operational amplifier is connected to the first end of the primary winding of the isolation coupler. The second end of the primary winding of the isolation coupler is grounded through a sampling resistor. The inverting input of the power operational amplifier is connected to its output through a feedback resistor. The two ends of the secondary winding of the isolation coupler serve as the power output of the power coupling module 5.

[0029] Synchronous acquisition unit 6 includes an AD7768 analog-to-digital converter, an ADA4899-1 differential amplifier, a LEMIT 60-S current sensor, a SiT8208 clock generator, an ADCLK914 clock buffer, and a TITPS7A4700 / 3300 series multi-output isolated power supply module. The non-inverting and inverting inputs of the differential amplifier serve as the voltage sampling terminals of synchronous acquisition unit 6. The output of the differential amplifier is connected to the first channel non-inverting analog input of the analog-to-digital converter, and the first channel inverting analog input is connected to a reference voltage. The positive output of the current sensor is connected to the second channel non-inverting analog input of the analog-to-digital converter, and the negative output is connected to the second channel inverting analog input. The analog-to-digital converter also includes a serial data output, a serial clock input, and a chip select terminal. The clock input terminal of the analog-to-digital converter (ADC) serves as one of the synchronous clock output terminals of the synchronous acquisition unit 6. The output terminal of the clock generator is connected to the input terminal of the clock buffer. The first output terminal of the clock buffer serves as the synchronous clock output terminal of the synchronous acquisition unit 6 and is connected to the main clock input terminal of the ADC. The power input terminal of the multi-output isolated power supply module is powered on. The first regulated output terminal of the multi-output isolated power supply module serves as the isolated power output terminal of the synchronous acquisition unit 6. The second regulated output terminal of the multi-output isolated power supply module is connected to the analog power supply terminals of the differential amplifier and the ADC. The third regulated output terminal of the multi-output isolated power supply module is connected to the digital power supply terminal of the ADC.

[0030] This invention also provides a ball valve fault diagnosis system, such as... Figure 2As shown, the device includes the electrically controlled ball valve, signal injection unit 4, and core processing unit 7. Signal injection unit 4 receives instructions from core processing unit 7 to accurately generate and output a composite test signal composed of a multi-frequency sine wave and a pseudo-random sequence as a diagnostic excitation source. The signal output terminal of signal injection unit 4 is connected to the signal input terminal of power coupling module 5. The power output terminal of power coupling module 5 is connected in series to the power supply circuit of the motor in electric actuator 3 via an isolation coupler. The voltage sampling terminal of synchronous acquisition unit 6 is connected in parallel to the power supply circuit of the motor in electric actuator 3, and the current sampling terminal of synchronous acquisition unit 6 is connected in parallel to the power supply circuit of the motor in electric actuator 3. The circuit is connected in series. The synchronous clock output terminal of the synchronous acquisition unit 6 is connected to the synchronous clock input terminal of the signal injection unit 4. The isolation power output terminal of the synchronous acquisition unit 6 is connected to the power input terminal of the signal injection unit 4. The data output terminal of the synchronous acquisition unit 6 is connected to the data input terminal of the core processing unit 7. The diagnostic result output terminal of the core processing unit 7 is connected to the host computer. The core processing unit 7 receives the voltage and current data uploaded by the synchronous acquisition unit 6 to perform high-speed parallel processing to extract multi-dimensional physical features. It compares and analyzes these features with the pre-stored gold fingerprint database to calculate the health index and complete the fault diagnosis decision. Finally, it outputs the diagnostic result and / or control command.

[0031] The signal injection unit 4 includes a direct digital frequency synthesizer AD9910, a first operational amplifier OPA2188, an active filter, and an analog switch ADG1412. The reference clock input of the direct digital frequency synthesizer serves as the synchronization clock input of the signal injection unit 4. The reference clock input of the direct digital frequency synthesizer is connected to the synchronization clock output of the synchronization acquisition unit 6. The first current output of the direct digital frequency synthesizer is connected to the inverting input of the first operational amplifier through a first resistor. The second current output of the direct digital frequency synthesizer is connected to the inverting input of the first operational amplifier through a second resistor. The non-inverting input of the first operational amplifier is grounded. The output of the first operational amplifier is connected to its inverting input through a third resistor. The output of the first operational amplifier is connected to the signal input of the active filter. The signal output of the active filter is connected to the first channel signal input of the analog switch. The second channel signal input of the analog switch is connected to the digital signal output of the core processing unit 7. The common output of the analog switch serves as the signal output of the signal injection unit 4.

[0032] The core processing unit 7 includes a field-programmable gate array (FPGA) Xilinx Kintex-7 XC7K160T, an embedded processor ARM Cortex-A9, and non-volatile memory DDR3 SDRAM. Multiple general-purpose input / output terminals of the FPGA serve as data input terminals of the core processing unit 7, connected to the serial data output terminal, serial clock input terminal, and chip select terminal of the analog-to-digital converter (ADC), respectively. The FPGA is connected to the embedded processor via an internal bus. The serial communication peripheral interface of the embedded processor is connected to the diagnostic result output terminal of the core processing unit 7 via a communication interface. A digital isolator is provided between the serial communication peripheral interface of the embedded processor and the signal input terminal of the communication interface. The serial clock terminal and serial data input / output terminal of the non-volatile memory are respectively connected to the corresponding serial peripheral interfaces of the embedded processor.

[0033] When the above-mentioned electrically controlled ball valve and ball valve fault diagnosis system are working, the following steps are included: S1. Benchmark establishment steps: The control signal injection unit 4 and the synchronous acquisition unit 6 cooperate to acquire the electromechanical response data of the valve in a healthy state, and establish a gold fingerprint library and environmental compensation model in the core processing unit 7. S101, control the valves to be positioned sequentially to multiple preset calibration opening points; S102. At each calibration opening point and multiple preset ambient temperature points, the control signal injection unit 4 injects standard multi-frequency test signals. S103, control the synchronous acquisition unit 6 to acquire the corresponding voltage and current response signals with high precision; S104. In the core processing unit 7, the reference complex impedance spectrum, reference nonlinear coefficient and reference cross-correlation entropy under each combination of opening degree and temperature are calculated and stored to form a gold fingerprint library. S105. In the core processing unit 7, the relationship between the reference complex impedance spectrum and temperature is analyzed, and an environmental compensation model is established by fitting. S2. Online excitation and synchronous acquisition steps: During valve operation, the control signal injection unit 4 injects a composite test signal into the valve motor circuit and controls the synchronous acquisition unit 6 to synchronously acquire voltage response and current response signals, while recording operating parameters. S201. At the preset periodic diagnostic time or when a trigger command is received, record the current valve command opening degree, process pressure and medium temperature; S202, the control signal injection unit 4 generates and injects a composite test signal containing at least two sinusoidal components of different frequencies into the motor power supply circuit; S203, control the synchronous acquisition unit 6, and synchronously acquire the voltage response waveform at both ends of the motor drive terminal and the current response waveform flowing through the circuit with a clock that is strictly synchronized with the signal injection. S204. Perform synchronous averaging on the acquired multi-cycle waveforms to suppress random noise; S3. Multidimensional physical feature extraction steps: In the core processing unit 7, the acquired signal is processed and multidimensional physical feature vectors, including frequency domain impedance features, nonlinear distortion features, and time domain statistical features, are extracted in parallel. S301. Frequency Domain Impedance Feature Extraction: Perform Fast Fourier Transform on the voltage and current response signals to calculate the complex impedance spectrum; identify the main resonance peaks in the complex impedance amplitude spectrum, extract the center frequency corresponding to each resonance peak as the resonance frequency, and calculate the corresponding modal damping ratio based on the negative 3 dB bandwidth of each resonance peak. S302. Nonlinear Distortion Feature Extraction: Analyze the spectrum of the current response signal, measure the amplitude of the third-order intermodulation distortion component generated by the two fundamental frequency components in the composite test signal; calculate the ratio of the sum of the amplitudes of the intermodulation distortion components to the sum of the amplitudes of the fundamental components to obtain the intermodulation distortion ratio; calculate the spectral kurtosis of the current spectrum in the narrow band near the fundamental frequency. S303, Time-domain statistical feature extraction: Calculate the normalized cross-correlation coefficient sequence of the injected composite test signal and the current response signal under different time delays; calculate the probability distribution information entropy of the cross-correlation coefficient sequence as the cross-correlation entropy; S4. Health Residual Vector Calculation Steps: In the core processing unit 7, the benchmark value in the gold fingerprint library is called according to the current working condition, and the deviation between the multi-dimensional physical feature vector and the corresponding benchmark value is calculated to form the health residual vector. S401. Based on the current operating parameters recorded in step S2, especially the valve opening and ambient temperature, query the corresponding reference complex impedance spectrum, reference nonlinear coefficient and reference cross-correlation entropy from the gold fingerprint database. S402. Using an environmental compensation model, perform temperature compensation correction on the queried baseline value; S403. Subtract the current resonant frequency, modal damping ratio, intermodulation distortion ratio, spectral kurtosis and cross-correlation entropy eigenvalues ​​extracted in step S3 from the corresponding corrected reference values ​​respectively. S404. Combine all the obtained differences to form a healthy residual vector; S5. Wear mode correlation analysis steps: In the core processing unit 7, time trend analysis and internal correlation analysis are performed on the health residual vector, and it is matched with the pre-stored fault modes to preliminarily determine the wear type. S501. Trend significance test: Select the key components in the health residual vector, calculate their Spearman rank correlation coefficients in multiple consecutive diagnostic periods, and determine whether there is a statistically significant trend of monotonically increasing or decreasing. S502, Cross-dimensional Correlation Analysis: Calculate the Pearson correlation coefficient between different feature components within the health residual vector to analyze the linkage relationship between feature changes; S503, Fault Mode Matching: The observed trend combinations and correlation patterns are compared with the fault-feature mapping matrix pre-stored in the core processing unit 7; the matrix defines the expected feature change patterns corresponding to different wear types. S504. Based on the matching results, output one or more possible preliminary wear types and their corresponding confidence scores; S6. Anti-drift baseline protection steps: In the core processing unit 7, an adaptive health baseline is maintained and managed. Based on the properties of the health residual vector and the preset rigid protection rules, it is determined whether to slowly update, freeze, or force reset the baseline. S601. Maintain an adaptive health baseline feature vector representing recent health status; S602. Baseline update condition judgment: If the absolute values ​​of all major components of the healthy residual vector are less than a small change threshold and no statistically significant trend is detected after analysis in step S5, it is judged as normal fluctuation and proceeds to step S603; otherwise, proceed to step S604 or S605. S603, Slow Update: Using a first-order low-pass filtering algorithm, the currently extracted multi-dimensional physical feature vector is incorporated into the adaptive healthy baseline with a small weight, so as to achieve slow and smooth tracking of the baseline. S604, Trend Freeze: When step S5 determines that there is a clear wear trend, immediately stop the adaptive health baseline update process and freeze it in the current state; S605, Over-limit Reset: When any component in the health residual vector exceeds its preset maximum allowable drift limit, the adaptive health baseline is forcibly reset to the corresponding benchmark value in the gold fingerprint database. S7. Fusion Diagnosis and Decision-Making Steps: In the core processing unit 7, based on the health residual vector and the adaptive health baseline, the comprehensive health index is calculated, and combined with the preliminary judgment of the wear type, the final health status diagnosis result and maintenance instructions are generated. S701, Dual-path health index calculation: Calculate the weighted Euclidean distance between the current multidimensional physical feature vector and the benchmark value of the gold fingerprint database, and map it to the absolute health index; calculate the weighted Euclidean distance between the current multidimensional physical feature vector and the adaptive health baseline, and map it to the relative health index; S702, Multi-criteria Fusion Decision: Combining the values ​​and trends of the absolute health index and relative health index, as well as the preliminary wear type and confidence level output from step S5, the final health status level is determined according to a predefined decision rule table; the health status level includes at least four levels: "normal", "concern", "deterioration" and "severe". S703, Command Output: When the diagnostic result is "Attention" or above, in addition to generating a diagnostic report containing specific wear descriptions and maintenance recommendations, the core processing unit 7 also generates a digital command, which is sent to the valve's main control system via the communication interface to lock or reset the parameters of its internal adaptive learning algorithm to prevent the algorithm model from continuing to drift.

[0034] To address the issue of algorithmic adaptive drift and mechanical progressive wear masking each other in predictive maintenance of electrically controlled ball valves, the system's operation begins with a baseline initialization action performed during the critical health period after the ball valve's commissioning or overhaul. At this time, the direct digital frequency synthesizer in signal injection unit 4 generates millivolt-level high-purity multi-frequency sinusoidal or pseudo-random binary sequence test signals under the unified synchronous clock of the clock buffer. These signals are shaped by an anti-aliasing filter composed of a high-precision operational amplifier, selected by an analog switch, and then linearly amplified by a power operational amplifier. They are then safely and distortion-free coupled in series into the power supply bus of the motor in electric actuator 3 via a wide-bandwidth isolation coupler. Simultaneously, the differential amplifier in synchronous acquisition unit 6 measures minute voltage changes between motor terminals with a high common-mode rejection ratio, while the current sensor accurately measures the loop current response. Both are strictly and synchronously sampled by a high-resolution analog-to-digital converter locked by the same clock source and converted into a digital signal stream.

[0035] The field-programmable gate array in the core processing unit 7 receives this data at high speed and executes algorithms such as fast Fourier transform and cross-power spectrum calculation in parallel to extract the electromechanical fingerprints of the valve in a new state, at different opening degrees and temperatures, namely the broad spectrum complex impedance curve, nonlinear intermodulation distortion components and excitation and response cross-correlation entropy. After being corrected by the environmental temperature compensation model, this data is stored in non-volatile memory as an immutable gold fingerprint library.

[0036] This initialization process establishes an absolute, pure physical baseline for the entire lifecycle diagnostics process. It is independent of any subsequent running algorithms, ensuring that the ruler measuring wear itself is not stretched or distorted.

[0037] After entering the online monitoring phase, the system periodically and automatically executes a health monitoring cycle to combat slow collaborative failures. In each cycle, the signal injection unit 4 accurately reproduces the test signal during initialization and injects it into the motor circuit of the electric actuator 3 through the isolation coupler. This active injection action creates a controlled diagnostic window that is separable from the spectrum of strong production process interference, so that weak electromechanical responses directly related to the mechanical state can be effectively extracted from the strong power frequency current and power noise for the purpose of drive control, thus avoiding the state signals being overwhelmed.

[0038] The synchronous acquisition unit 6 further synchronizes the accuracy, capturing the instantaneous response of the injection point voltage and loop current. Its ultra-low noise ensures the effective perception of picoampere-level leakage current or microvolt-level voltage changes, realizing high-fidelity, synchronized data acquisition. This provides reliable raw material for subsequent analysis and avoids feature extraction errors caused by asynchronous or distorted signal acquisition. Such errors are easily misjudged as normal fluctuations by adaptive algorithms.

[0039] The acquired raw data is sent to the core processing unit 7. Here, the embedded processor, in collaboration with the field-programmable gate array, executes the algorithm steps in the perspective method. First, multi-dimensional physical feature extraction is performed. By performing fast Fourier transform on the voltage and current signals and dividing by complex numbers, the real-time broad spectrum complex impedance is obtained, and its resonant frequency and modal damping ratio are identified.

[0040] By performing a detailed analysis of the current spectrum, the third-order intermodulation distortion ratio and spectral kurtosis were calculated.

[0041] By calculating the cross-correlation function of the excitation and response signals and obtaining its information entropy, the cross-correlation entropy value is obtained. The key effect of this step is to upgrade the single-dimensional current or torque signal to a multi-dimensional feature space composed of frequency domain, nonlinear domain, and time domain statistical domain. Mechanical wear, such as valve seat erosion leading to nonlinear changes in stiffness and electrical drift, such as a decrease in sensor gain, will leave completely different marks in the above different feature spaces. The parameter drift of a single algorithm is difficult to synchronously and coordinately create a healthy illusion in all dimensions.

[0042] Next, the algorithm enters the health residual vector calculation step. Based on the actual valve opening, medium temperature and other operating condition parameters recorded at the current monitoring time, the processor calls the corresponding temperature-compensated benchmark feature value from the gold fingerprint database, and subtracts the real-time extracted feature vector from it element by element to generate the health residual vector. This achieves operating condition separation, which effectively separates the legitimate changes in valve dynamic characteristics caused by changes in production processes, such as pressure and flow regulation, from the illegitimate changes in characteristics caused by mechanical wear. The residual vector directly and purely represents the deviation of the current mechanical state from the new healthy state. This makes the slow wear trend emerge from the mixed operating condition noise and no longer easily absorbed by the adaptive algorithm as an adaptable new operating condition.

[0043] Subsequently, the wear mode correlation analysis step deeply mines the residual vector. On the one hand, it analyzes whether there is a statistically significant monotonic trend in the time series of each residual component, such as the damping ratio residual slowly increasing year by year. On the other hand, it analyzes the correlation between different residual components, such as whether the intermodulation distortion ratio residual and the cross-correlation entropy residual grow synchronously. These patterns are matched with pre-stored fault and feature mapping matrices, such as valve seat erosion: nonlinear features are strongly correlated with statistical entropy features and weakly correlated with resonant frequency features. This achieves cross-validation and initial fault source judgment across feature dimensions. Even if the adaptive algorithm attempts to adjust the threshold for a certain feature, such as torque-related features, the coordinated change trend between multiple independent physical principle features that conforms to a specific wear mode cannot be simply optimized away by the algorithm. This is equivalent to providing multiple independent evidence for the diagnostic conclusion, significantly improving the robustness of the criterion.

[0044] The system's anti-drift baseline protection step directly addresses the algorithm's own adaptability issues by building a firewall. This step maintains a slowly updatable adaptive health baseline for daily alarms in the core processing unit 7. However, its update logic is strictly constrained by protection rules. The baseline is only updated smoothly at an extremely slow rate when the health residual vector is extremely small and has no trend. Once the relevant analysis steps detect a clear wear trend, the baseline update is immediately frozen.

[0045] If any residual component exceeds the maximum permissible drift limit set according to the physical life of the device, the baseline will be forcibly reset to the original benchmark of the gold fingerprint library. The core effect of this step is to achieve surgical control over the learning ability of the software algorithm. It allows the system to learn moderately to maintain sensitivity to small, non-trend normal fluctuations, but resolutely prevents it from adapting or normalizing to long-term, monotonous trends that characterize real physical degradation in any form, thereby cutting off the path of negative feedback loop between algorithm drift and mechanical wear.

[0046] Finally, in the fusion diagnostic decision-making step, based on the processor's comprehensive absolute health index (based on the gold benchmark), the relative health index (based on the protected adaptive baseline), and the initial wear pattern judgment results, a graded diagnostic conclusion from normal to severe is given. When the warning level is reached, a digitally signed instruction is sent to the host computer control system of the electric ball valve through the digital isolator and communication interface circuit, forcibly locking or rolling back the learning function of its built-in predictive maintenance algorithm. This transforms the diagnostic conclusion at the hardware layer into direct intervention at the software layer, realizing a closed loop of perception, diagnosis, and execution. It not only reveals the masked faults but also proactively prevents the host computer control system from continuing to execute erroneous adaptive behaviors.

[0047] This means that slow mechanical wear will inevitably leave traceable marks on multidimensional physical characteristics such as broad-spectrum impedance. Due to their multidimensionality, correlation, and comparability with absolute benchmarks, these marks cannot be silently concealed by single-dimensional adaptive algorithms. Thus, the gradual degradation that was originally tacitly confirmed by the algorithm as normal is exposed again as a clear and predictable fault development trajectory. Ultimately, this provides a valuable maintenance window and decision-making basis before sudden failures and unplanned downtime occur.

[0048] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An electrically controlled ball valve, characterized in that, The device includes a valve body (1), an electric actuator (3) mounted on it via a connecting seat (2), a valve stem that is drivenly connected to the output shaft of the electric actuator (3) and extends into the valve body (1), a valve ball that is fixedly connected to the lower end of the valve stem and located in the valve body (1), a power coupling module (5) connected in series with the motor power supply circuit in the electric actuator (3), and a synchronous acquisition unit (6) connected in parallel. The synchronous acquisition unit (6) acquires the voltage transient response on the motor drive end of the electric actuator (3) and the current transient response in the power supply circuit in parallel with a precision clock as a reference, and converts these two analog signals into digital signals with high fidelity. The power coupling module (5) receives weak diagnostic signals, amplifies them linearly, and injects them directly into the power supply circuit of the motor in the electric actuator (3) through a broadband isolation transformer.

2. The electrically controlled ball valve according to claim 1, characterized in that, The power coupling module (5) includes a power operational amplifier and an isolation coupler. The non-inverting input terminal of the power operational amplifier serves as the signal input terminal of the power coupling module (5). The output terminal of the power operational amplifier is connected to the first end of the primary winding of the isolation coupler. The second end of the primary winding of the isolation coupler is grounded through a sampling resistor. The inverting input terminal of the power operational amplifier is connected to its output terminal through a feedback resistor. The two ends of the secondary winding of the isolation coupler serve as the power output terminals of the power coupling module (5).

3. The electrically controlled ball valve according to claim 1, characterized in that, The synchronous acquisition unit (6) includes an analog-to-digital converter, a differential amplifier, and a current sensor. The non-inverting and inverting input terminals of the differential amplifier serve as the voltage sampling terminals of the synchronous acquisition unit (6). The output terminal of the differential amplifier is connected to the first channel non-inverting analog input terminal of the analog-to-digital converter. The first channel inverting analog input terminal of the analog-to-digital converter is connected to a reference voltage. The positive output terminal of the current sensor is connected to the second channel non-inverting analog input terminal of the analog-to-digital converter. The negative output terminal of the current sensor is connected to the second channel inverting analog input terminal of the analog-to-digital converter. The serial data output terminal, serial clock input terminal, and chip selection terminal of the analog-to-digital converter serve as the data output terminals of the synchronous acquisition unit (6). The master clock input terminal of the analog-to-digital converter serves as one of the synchronous clock output terminals of the synchronous acquisition unit (6).

4. The electrically controlled ball valve according to claim 3, characterized in that, The synchronous acquisition unit (6) further includes a clock generator, a clock buffer, and a multi-output isolated power supply module. The output terminal of the clock generator is connected to the input terminal of the clock buffer. The first output terminal of the clock buffer serves as the synchronous clock output terminal of the synchronous acquisition unit (6) and is connected to the main clock input terminal of the analog-to-digital converter. The power input terminal of the multi-output isolated power supply module is powered on. The first regulated output terminal of the multi-output isolated power supply module serves as the isolated power output terminal of the synchronous acquisition unit (6). The second regulated output terminal of the multi-output isolated power supply module is connected to the analog power supply terminals of the differential amplifier and the analog-to-digital converter. The third regulated output terminal of the multi-output isolated power supply module is connected to the digital power supply terminal of the analog-to-digital converter.

5. A ball valve fault diagnosis system, characterized in that, The system includes an electrically controlled ball valve as described in any one of claims 1-4, a signal injection unit (4), and a core processing unit (7). The signal injection unit (4) receives instructions from the core processing unit (7) to accurately generate and output a composite test signal composed of a multi-frequency sine wave and a pseudo-random sequence as a diagnostic excitation source. The signal output terminal of the signal injection unit (4) is connected to the signal input terminal of the power coupling module (5). The power output terminal of the power coupling module (5) is connected in series with the power supply circuit of the motor in the electric actuator (3) through an isolation coupler. The voltage sampling terminal of the synchronous acquisition unit (6) is connected in parallel with the power supply circuit of the motor in the electric actuator (3). The current sampling terminal of the synchronous acquisition unit (6) is connected to the electric actuator (3) The power supply circuit of the motor is connected in series. The synchronous clock output terminal of the synchronous acquisition unit (6) is connected to the synchronous clock input terminal of the signal injection unit (4). The isolation power output terminal of the synchronous acquisition unit (6) is connected to the power input terminal of the signal injection unit (4). The data output terminal of the synchronous acquisition unit (6) is connected to the data input terminal of the core processing unit (7). The diagnostic result output terminal of the core processing unit (7) is connected to the host computer. The core processing unit (7) receives the voltage and current data uploaded by the synchronous acquisition unit (6) to perform high-speed parallel processing to extract multi-dimensional physical features. It compares and analyzes these features with the pre-stored gold fingerprint database to calculate the health index and complete the fault diagnosis decision. Finally, it outputs the diagnostic result and / or control command.

6. A ball valve fault diagnosis system according to claim 5, characterized in that, The signal injection unit (4) includes a direct digital frequency synthesizer, a first operational amplifier, an active filter, and an analog switch. The reference clock input terminal of the direct digital frequency synthesizer serves as the synchronous clock input terminal of the signal injection unit (4). The reference clock input terminal of the direct digital frequency synthesizer is connected to the synchronous clock output terminal of the synchronous acquisition unit (6). The first current output terminal of the direct digital frequency synthesizer is connected to the inverting input terminal of the first operational amplifier through a first resistor. The second current output terminal of the direct digital frequency synthesizer is connected to the inverting input terminal of the first operational amplifier through a second resistor. The non-inverting input terminal of the first operational amplifier is grounded. The output terminal of the first operational amplifier is connected to its inverting input terminal through a third resistor. The output terminal of the first operational amplifier is connected to the signal input terminal of the active filter. The signal output terminal of the active filter is connected to the first channel signal input terminal of the analog switch. The second channel signal input terminal of the analog switch is connected to the digital signal output terminal of the core processing unit (7). The common output terminal of the analog switch serves as the signal output terminal of the signal injection unit (4).

7. A ball valve fault diagnosis system according to claim 5, characterized in that, The core processing unit (7) includes a field-programmable gate array (FPGA) and an embedded processor. The multiple general-purpose input / output terminals of the FPGA serve as the data input terminals of the core processing unit (7) and are respectively connected to the serial data output terminal, serial clock input terminal, and chip selection terminal of the analog-to-digital converter. The FPGA is connected to the embedded processor through an internal bus. The serial communication peripheral interface of the embedded processor is connected to the diagnostic result output terminal of the core processing unit (7) through a communication interface.

8. A ball valve fault diagnosis system according to claim 7, characterized in that, The core processing unit (7) also includes a non-volatile memory. A digital isolator is provided between the serial communication peripheral interface of the embedded processor and the signal input terminal of the communication interface. The serial clock terminal and the serial data input / output terminal of the non-volatile memory are respectively connected to the corresponding serial peripheral interface of the embedded processor.

Images