Petrochemical equipment fault monitoring device and method based on CT multiplexing and multi-modal fusion

Abstract

The invention discloses a petrochemical equipment fault monitoring device and method based on CT multiplexing and multi-modal fusion, and belongs to the field of petrochemical equipment monitoring, and the device comprises a CT power taking and supplying subsystem, a multi-modal sensing subsystem, a switching control unit, a feature extraction and early warning unit, and a wireless transmission unit. Wherein in the charging mode, the current transformer charges the energy storage module, and in the detection mode, the current transformer collects current signals, and the vibration sensor synchronously collects vibration signals. Through time-sharing switching of the same CT coil between the charging mode and the detection mode, the deployment requirements of special power supply equipment and an independent current sensor are eliminated, and the number of components and occupied installation space are reduced.

Description

Technical Field

[0001] This invention belongs to the field of petrochemical equipment monitoring technology, and particularly relates to a petrochemical equipment fault monitoring device and method based on CT multiplexing and multimodal fusion. Background Technology

[0002] Currently, there are two main technical solutions in the field of petrochemical equipment monitoring: one is a remote intelligent environmental monitoring and alarm device for power equipment (such as patent CN103326468A), which achieves passive power supply by switching between solar energy and CT coils, and monitors environmental parameters such as temperature, humidity, and air pressure, as well as basic electrical parameters such as current and voltage; the other is a fault diagnosis system based on current analysis, which relies on drones to collect inspection data and fuse it with sensor data, and builds a fault early warning model through a remote monitoring center. Both solutions adopt a single sensing dimension—the former only collects electrical parameters, and the latter only analyzes current signals.

[0003] The existing technologies have the following shortcomings: (1) Limited power supply: In patent CN103326468A, the CT coil is only used for power supply and is not reused as a current sensor, which leads to the need to deploy additional current detection equipment; while the current analysis scheme relies on external mains power or solar power, and has no self-powering capability, which cannot meet the requirements of large-scale deployment in petrochemical plants. (2) Single monitoring dimension: Neither of the two schemes can collect vibration signals and current signals simultaneously, which makes it difficult to distinguish between mechanical faults (bearing wear, shaft misalignment) and electrical faults (rotor bar breakage, inter-turn short circuit). (3) Defects in real-time performance and cost: The current analysis scheme needs to transmit data to a remote center for processing, which has high latency and high hardware costs; at the same time, the existing technologies are not adapted to petrochemical explosion-proof scenarios (such as explosion-proof enclosures, IP67 protection), which poses safety hazards. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a petrochemical equipment fault monitoring device based on CT multiplexing and multimodal fusion, thereby resolving the issues of "hardware redundancy, lack of explosion-proof design, and limited passive power supply" in the existing technologies.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a petrochemical equipment fault monitoring device based on CT multiplexing and multimodal fusion, comprising:

[0006] The CT power supply and electronic system is used to obtain power frequency current from the motor power line through a current transformer.

[0007] A multimodal sensing subsystem includes a vibration sensor and a current transformer multiplexed as a current sensor;

[0008] A switching control unit is used to control the current transformer to switch to charging mode or detection mode in a time-sharing manner;

[0009] The feature extraction and early warning unit is used to output fault type codes based on the vibration and current signals in the detection mode;

[0010] A wireless transmission unit is used to send the fault type code to the central control system;

[0011] In the charging mode, the current transformer charges the energy storage module; in the detection mode, the current transformer collects current signals and the vibration sensor simultaneously collects vibration signals.

[0012] Preferably, the CT power supply and electronics system includes:

[0013] The rectifier circuit consists of full-bridge rectifier diodes and filter capacitors;

[0014] The voltage regulator circuit uses an LDO regulator to output DC power.

[0015] The energy storage module uses a high-temperature resistant lithium battery;

[0016] The rectifier circuit and voltage regulator circuit are encapsulated in a 316L stainless steel explosion-proof housing.

[0017] Preferably, the switching control unit includes:

[0018] The timing module is used to trigger mode switching after the charging mode has been in operation for a first preset duration.

[0019] The switching execution circuit includes a dual-contact relay, a TVS diode, and a bleeder resistor;

[0020] The main control chip is used to control the switching execution circuit to connect the current transformer to the rectifier circuit or the current detection module.

[0021] Preferably, the switching control unit further includes:

[0022] The vibration signal monitoring module is used to calculate the RMS value of the vibration signal in real time; when the RMS value of the vibration signal exceeds the baseline threshold, it forcibly triggers the switch from charging mode to detection mode.

[0023] Preferably, the switching execution circuit uses a MOSFET instead of a dual-contact relay, and the gate of the MOSFET is connected to the PWM output terminal of the main control chip.

[0024] Preferably, the switching control unit resets the timer after completing wireless data transmission and controls the current transformer to return to charging mode.

[0025] Preferably, the feature extraction and early warning unit performs the following:

[0026] Calculate the RMS value, kurtosis index, and envelope spectrum of the vibration signal;

[0027] The negative sequence current is calculated using the phase difference correction method and the symmetrical component method, and the Park vector magnitude is also calculated.

[0028] A multimodal feature vector is generated based on the RMS value, kurtosis index, envelope spectrum, negative sequence current, and Park vector mode.

[0029] Preferably, the feature extraction and early warning unit further performs the following:

[0030] A weighted decision fusion method is applied to the multimodal feature vectors to obtain a fusion result; wherein the multimodal feature vectors include vibration signals and current signals;

[0031] Based on the fusion result, a preset fault threshold is matched, and the corresponding mechanical or electrical fault code is output.

[0032] Secondly, the present invention also provides a method for fault monitoring of petrochemical equipment based on CT multiplexing and multimodal fusion, the method comprising: (The second aspect is missing from the original text.)

[0033] The power frequency current is obtained from the motor power line through the current transformer of the CT power supply and electronic system.

[0034] The current transformer, which includes a vibration sensor and is multiplexed as a current sensor, is configured through a multimodal sensing subsystem.

[0035] The current transformer is controlled by the switching control unit to switch between charging mode and detection mode in a time-sharing manner.

[0036] The feature extraction and early warning unit outputs a fault type code based on the vibration and current signals in the detection mode;

[0037] The fault type code is transmitted to the central control system via a wireless transmission unit;

[0038] In the charging mode, the current transformer charges the energy storage module; in the detection mode, the current transformer collects current signals and the vibration sensor simultaneously collects vibration signals.

[0039] Thirdly, the present invention also discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0040] Compared with the prior art, the present invention has the following advantages and technical effects:

[0041] This invention provides a petrochemical equipment fault monitoring device based on CT multiplexing and multimodal fusion, comprising: a CT power supply and electronic system for obtaining power frequency current from the motor power line via a current transformer; a multimodal sensing subsystem including a vibration sensor and the current transformer multiplexed as a current sensor; a switching control unit for controlling the current transformer to switch between charging mode and detection mode in a time-sharing manner; a feature extraction and early warning unit for outputting a fault type code based on the vibration signal and current signal in the detection mode; and a wireless transmission unit for sending the fault type code to the central control system; wherein, in charging mode, the current transformer charges the energy storage module, and in detection mode, the current transformer collects the current signal while the vibration sensor simultaneously collects the vibration signal.

[0042] This invention eliminates the need for dedicated power supply equipment and independent current sensors by switching the same CT coil between charging mode (powering the energy storage module) and detection mode (reused as a current sensor) in a time-division manner, thereby reducing the number of components and installation space required.

[0043] This invention simultaneously acquires vibration signals (vibration sensor) and current signals (multiplexed CT) in detection mode, enabling parallel monitoring of mechanical faults (bearing wear, shaft misalignment) and electrical faults (rotor bar breakage, inter-turn short circuit), avoiding misjudgments caused by a single signal dimension.

[0044] This invention uses CT power supply to achieve complete self-powering, eliminating reliance on external power sources; combined with a wireless transmission unit to remotely send fault codes, it avoids the need for low-voltage electrical renovation in high-risk petrochemical areas and meets the deployment requirements for explosion-proof scenarios. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0046] Figure 1 This is a structural diagram of the device according to an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the explosion-proof enclosure according to an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of the workflow of an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram of switching and control according to an embodiment of the present invention;

[0050] Figure 5 This is a schematic diagram of the device installation according to an embodiment of the present invention. Detailed Implementation

[0051] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0052] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0053] The technical terms used in the following embodiments will be explained first.

[0054] A current transformer (CT) coil, also known as a current transformer, is an electrical device that converts a large current into a small current. It is mainly used for measuring and protecting circuits. It operates based on Faraday's law of electromagnetic induction. The primary winding is connected in series in the current circuit to be measured, while the secondary winding is connected in series in the measuring instrument and the protection circuit.

[0055] Multimodal sensing synchronously acquires motor current signals (CT power supply) and mechanical vibration signals (MEMS accelerometer) for comprehensive fault diagnosis.

[0056] Example 1

[0057] like Figure 1 As shown, this embodiment provides a petrochemical equipment fault monitoring device based on CT multiplexing and multimodal fusion, including: a CT power supply and electronic power generation system, a multimodal sensing subsystem, a feature extraction and early warning unit, an intelligent switching and control unit, and a wireless transmission unit; specifically including:

[0058] The CT power supply and electronic system is used to obtain power frequency current from the motor power line through a current transformer.

[0059] Furthermore, the CT power supply and electronics system includes:

[0060] The rectifier circuit consists of full-bridge rectifier diodes and filter capacitors;

[0061] The voltage regulator circuit uses an LDO regulator to output DC power.

[0062] The energy storage module uses a high-temperature resistant lithium battery;

[0063] The rectifier circuit and voltage regulator circuit are encapsulated in a 316L stainless steel explosion-proof housing.

[0064] Specifically, the CT power supply and electronic system uses a switch to time-divide the CT coil for device power supply and current detection. CT coil: 3 current transformers connected to the motor power line, outputting mains frequency current (e.g., transformation ratio 1000:1, range 0-100A). Rectifier circuit: Full-bridge rectifier (diode model 1N4007) + filter capacitor (1000μF / 25V). Voltage regulator circuit: LDO regulator (TPS7A4700, output 3.3V±1%). Explosion-proof design: Except for vibration and current transformers, the entire hardware circuit and battery are encapsulated in an explosion-proof housing (material 316L stainless steel). Figure 2 As shown, it meets the IP67 protection rating. Energy storage module: High-temperature resistant lithium battery (3.7V / 1000mAh, operating temperature -40℃~85℃).

[0065] A multimodal sensing subsystem includes a vibration sensor and a current transformer multiplexed as a current sensor;

[0066] Specifically, the multimodal sensing subsystem includes a vibration sensor and a current detection module. The vibration sensor is a triaxial accelerometer, mounted on the motor / pump bearing housing; the current detection module uses a CT coil multiplexed as a current sensor to acquire the motor current waveform (12-bit resolution).

[0067] A switching control unit is used to control the current transformer to switch to charging mode or detection mode in a time-sharing manner;

[0068] Furthermore, the switching control unit includes:

[0069] The timing module is used to trigger mode switching after the charging mode has been in operation for a first preset duration.

[0070] The switching execution circuit includes a dual-contact relay, a TVS diode, and a bleeder resistor; the switching execution circuit uses a MOSFET instead of the dual-contact relay, and the gate of the MOSFET is connected to the PWM output terminal of the main control chip.

[0071] The main control chip is used to control the switching execution circuit to connect the current transformer to the rectifier circuit or the current detection module.

[0072] The vibration signal monitoring module is used to calculate the RMS value of the vibration signal in real time; when the RMS value of the vibration signal exceeds the baseline threshold, it forcibly triggers the switch from charging mode to detection mode.

[0073] In this embodiment, the switching control unit resets the timer after completing wireless data transmission and controls the current transformer to return to charging mode.

[0074] Specifically, the main function of the switching and control unit is to perform charging and detection in a time-sharing manner to avoid signal interference. The switching components include: a dual-contact relay (G6K-2F-Y) + a TVS diode (SMBJ15CA) + a bleeder resistor (10Ω / 50W). The main control chip is an STM32L452 (low-power mode current ≤1μA). The mode switching logic is as follows: the CT coil charges the battery (default state) for 30 minutes; then it switches to data acquisition (detection mode), simultaneously starting vibration and current signal sampling (10-second window).

[0075] The feature extraction and early warning unit is used to output fault type codes based on the vibration and current signals in the detection mode;

[0076] Furthermore, the feature extraction and early warning unit performs the following:

[0077] Calculate the RMS value, kurtosis index, and envelope spectrum of the vibration signal;

[0078] The negative sequence current is calculated using the phase difference correction method and the symmetrical component method, and the Park vector magnitude is also calculated.

[0079] A multimodal feature vector is generated based on the RMS value, kurtosis index, envelope spectrum, negative sequence current, and Park vector mode.

[0080] A weighted decision fusion method is applied to the multimodal feature vectors to obtain a fusion result; wherein the multimodal feature vectors include vibration signals and current signals;

[0081] Based on the fusion result, a preset fault threshold is matched, and the corresponding mechanical or electrical fault code is output.

[0082] Specifically, the feature extraction and early warning unit includes: (1) Vibration signal: calculating RMS value, kurtosis index, and envelope spectrum (used to detect bearing pitting, gear tooth breakage, and shaft misalignment faults). (2) Current signal: phase difference correction method + symmetrical component method (calculating negative sequence current and unbalance, diagnosing inter-turn short circuits), Park vector mode (detecting rotor bar breakage). (3) Multimodal fusion algorithm: using weighted decision fusion (vibration weight 60%, current weight 40%), with adjustable threshold. Example: if vibration kurtosis > 4 and current THD > 8%, it is determined to be a combined fault of bearing wear and rotor bar breakage.

[0083] A wireless transmission unit is used to send the fault type code to the central control system;

[0084] Specifically, the wireless transmission unit is used to send signal characteristic data (such as vibration kurtosis, vibration amplitude, impact energy, current amplitude, current frequency, current THD, negative sequence current, sideband components, etc.) to the central control system; and to send fault codes (such as F001: bearing fault, F002: broken bar fault) to the central control system.

[0085] The workflow of this embodiment is as follows: Figure 3 As shown: (1) Power on and initialize, enter charging mode; (2) After the timer reaches 30 minutes, switch to detection mode; (3) Start ADC sampling and store data; (4) Extract features for fault warning; (5) Send data through wireless transmission unit, and reset timer after completion.

[0086] Switching control diagram, such as Figure 4 As shown, in charging mode, the current transformer charges the energy storage module; in detection mode, the current transformer collects current signals and the vibration sensor simultaneously collects vibration signals.

[0087] Installation diagram of the device, as shown Figure 5 As shown, the vibration sensor is installed on the surface of the monitored motor or pump unit by means of adhesive or bolts; three current transformers are respectively fitted on the three-phase power lines of the motor, and the signals from the vibration sensor and the current transformers are connected to the monitoring device.

[0088] In this embodiment, the trigger condition is replaced; when the vibration RMS value exceeds the baseline by 20%, the detection mode is immediately activated. Detection is triggered based on battery voltage (e.g., detection is activated when the voltage is ≥3.6V).

[0089] In this embodiment, hardware substitution is used, with MOSFETs (such as IRF540N) replacing relays to reduce switching power consumption.

[0090] In this embodiment, the CT coil has dual functions: a dynamic switching method for the same CT in charging / detection modes, that is, the same CT is used for power extraction (charging mode) and current detection (detection mode) in a time-sharing manner.

[0091] The beneficial effects of this invention are:

[0092] This invention achieves time-division multiplexing of the CT coil through a timed switching mode, allowing it to be used alternately for device power supply and current detection. This design significantly reduces system costs, decreasing the need for dedicated power supply equipment and additional current detection devices. Simultaneously, it saves installation space, making it particularly suitable for petrochemical plant sites with limited space. Multimodal data fusion provides more comprehensive and accurate device status information, helping to promptly detect potential faults and improve monitoring accuracy. Furthermore, the equipped wireless transmission unit supports wireless data transmission and remote monitoring, allowing maintenance personnel to monitor device status anytime, anywhere, and respond promptly to anomalies, reducing maintenance costs and improving safety. Since the device achieves autonomous power supply through the CT coil and data transmission through the wireless transmission unit, extensive on-site modifications to the power grid and other low-voltage systems are unnecessary, simplifying deployment and significantly reducing deployment costs. These technical features collectively enhance the innovation and practicality of this device.

[0093] Example 2

[0094] Based on the same inventive concept, this embodiment also provides a petrochemical equipment fault monitoring method based on CT multiplexing and multimodal fusion, used in the device described in Embodiment 1, the method comprising:

[0095] The power frequency current is obtained from the motor power line through the current transformer of the CT power supply and electronic system.

[0096] The current transformer, which includes a vibration sensor and is multiplexed as a current sensor, is configured through a multimodal sensing subsystem.

[0097] The current transformer is controlled by the switching control unit to switch between charging mode and detection mode in a time-sharing manner.

[0098] The feature extraction and early warning unit outputs a fault type code based on the vibration and current signals in the detection mode;

[0099] The fault type code is transmitted to the central control system via a wireless transmission unit;

[0100] In the charging mode, the current transformer charges the energy storage module; in the detection mode, the current transformer collects current signals and the vibration sensor simultaneously collects vibration signals.

[0101] The petrochemical equipment fault monitoring method based on CT multiplexing and multimodal fusion provided in this embodiment has all the advantages of the petrochemical equipment fault monitoring device based on CT multiplexing and multimodal fusion provided in Embodiment 1.

[0102] Example 3

[0103] This embodiment also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0104] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A petrochemical equipment fault monitoring device based on CT multiplexing and multimodal fusion, characterized in that, include: The CT power supply and electronic system is used to obtain power frequency current from the motor power line through a current transformer. A multimodal sensing subsystem includes a vibration sensor and a current transformer multiplexed as a current sensor; A switching control unit is used to control the current transformer to switch to charging mode or detection mode in a time-sharing manner; The feature extraction and early warning unit is used to output fault type codes based on the vibration and current signals in the detection mode; A wireless transmission unit is used to send the fault type code to the central control system; In the charging mode, the current transformer charges the energy storage module; in the detection mode, the current transformer collects current signals and the vibration sensor simultaneously collects vibration signals.

2. The apparatus according to claim 1, characterized in that, The CT power supply and electronic power generation system includes: The rectifier circuit consists of full-bridge rectifier diodes and filter capacitors; The voltage regulator circuit uses an LDO regulator to output DC power. The energy storage module uses a high-temperature resistant lithium battery; The rectifier circuit and voltage regulator circuit are encapsulated in a 316L stainless steel explosion-proof housing.

3. The apparatus according to claim 1, characterized in that, The switching control unit includes: The timing module is used to trigger mode switching after the charging mode has been in operation for a first preset duration. The switching execution circuit includes a dual-contact relay, a TVS diode, and a bleeder resistor; The main control chip is used to control the switching execution circuit to connect the current transformer to the rectifier circuit or the current detection module.

4. The apparatus according to claim 1, characterized in that, The switching control unit further includes: The vibration signal monitoring module is used to calculate the RMS value of the vibration signal in real time; when the RMS value of the vibration signal exceeds the baseline threshold, it forcibly triggers the switch from charging mode to detection mode.

5. The apparatus according to claim 3, characterized in that, The switching execution circuit uses a MOSFET instead of a dual-contact relay, and the gate of the MOSFET is connected to the PWM output terminal of the main control chip.

6. The apparatus according to claim 1, characterized in that, After completing wireless data transmission, the switching control unit resets the timer and controls the current transformer to return to charging mode.

7. The apparatus according to claim 1, characterized in that, The feature extraction and early warning unit performs the following: Calculate the RMS value, kurtosis index, and envelope spectrum of the vibration signal; The negative sequence current is calculated using the phase difference correction method and the symmetrical component method, and the Park vector magnitude is also calculated. A multimodal feature vector is generated based on the RMS value, kurtosis index, envelope spectrum, negative sequence current, and Park vector mode.

8. The apparatus according to claim 1, characterized in that, The feature extraction and early warning unit further performs the following: A weighted decision fusion method is applied to the multimodal feature vectors to obtain a fusion result; wherein the multimodal feature vectors include vibration signals and current signals; Based on the fusion result, a preset fault threshold is matched, and the corresponding mechanical or electrical fault code is output.

9. A method for fault monitoring of petrochemical equipment based on CT multiplexing and multimodal fusion, characterized in that, The method, based on the apparatus according to any one of claims 1-8, comprises: The power frequency current is obtained from the motor power line through the current transformer of the CT power supply and electronic system. The current transformer, which includes a vibration sensor and is multiplexed as a current sensor, is configured through a multimodal sensing subsystem. The current transformer is controlled by the switching control unit to switch between charging mode and detection mode in a time-sharing manner. The feature extraction and early warning unit outputs a fault type code based on the vibration and current signals in the detection mode; The fault type code is transmitted to the central control system via a wireless transmission unit; In the charging mode, the current transformer charges the energy storage module; in the detection mode, the current transformer collects current signals and the vibration sensor simultaneously collects vibration signals.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method of claim 9.

Images