Power plant steam turbine cylinder switching operation vibration detection device

Abstract

The invention relates to the technical field of power plant steam turbine vibration detection, and discloses a power plant steam turbine cylinder switching operation vibration detection device which comprises a detection unit which comprises a vibration sensor, a temperature sensor and a noise sensor and is used for synchronously collecting a vibration signal, a temperature signal and a noise signal, the vibration sensor, the temperature sensor and the noise sensor are respectively mounted at preset positions of the steam turbine body; the data acquisition and processing unit is connected with the detection unit and is used for receiving and processing the vibration signal, the temperature signal and the noise signal to obtain fused signal data; and the control unit communicates with the data acquisition and processing unit and is used for remote monitoring, real-time detection result display, alarm triggering and data storage. Different data signals are synchronously collected through the vibration sensor, the temperature sensor and the noise sensor, data fusion processing is carried out, the reliability and stability of the signals are enhanced, the monitoring comprehensiveness and accuracy are improved, and the phenomena of false alarm and missing alarm are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration detection of steam turbines in power plants, and in particular to a vibration detection device for cylinder cutting operation of steam turbines in power plants. Background Art

[0002] As a key device for converting thermal energy into mechanical energy, a power plant's steam turbine's operating efficiency and safety are directly related to the overall plant's production efficiency and stability. The efficient operation of a steam turbine plays an irreplaceable role in the effective utilization of energy and the reliability of power supply.

[0003] During turbine operation, turbine cylinder trimming in power plants can rationally distribute steam flow, reduce energy waste, protect turbine equipment from overload damage, and optimize turbine efficiency. Abnormal vibration during turbine cylinder trimming is often an early sign of equipment failure. To ensure proper operation during turbine cylinder trimming, vibration anomaly detection is necessary. Timely and accurate detection of vibration anomalies is crucial for ensuring safe and stable turbine operation.

[0004] Existing devices for detecting abnormal vibration during turbine cutout operation in power plants primarily rely on real-time monitoring using vibration sensors installed at the appropriate locations on the turbine cutout. However, due to the complex operating environment during cutout, vibration sensors are susceptible to interference during operation, resulting in false alarms and missed alarms. This makes it difficult to quickly and accurately detect abnormal vibration during turbine cutout, posing a potential risk to the safe operation of the equipment. Summary of the Invention

[0005] In view of this, the present invention provides a vibration detection device for power plant steam turbine cylinder cutting operation to solve the problem that the vibration sensor in the prior art is easily interfered with when detecting the vibration signal during power plant steam turbine cylinder cutting operation, resulting in false alarms and missed alarms.

[0006] In a first aspect, the present invention provides a device for detecting vibration during cylinder-cut operation of a steam turbine in a power plant, comprising:

[0007] A detection unit, comprising a vibration sensor, a temperature sensor and a noise sensor, for synchronously collecting vibration signals, temperature signals and noise signals, wherein the vibration sensor, the temperature sensor and the noise sensor are respectively installed at preset positions of the steam turbine body;

[0008] A data acquisition and processing unit is connected to the detection unit and is used to receive and process vibration signals, temperature signals and noise signals to obtain fused signal data;

[0009] The control unit communicates with the data acquisition and processing unit and is used for remote monitoring, real-time display of detection results, triggering alarms and storing data.

[0010] During vibration detection, the vibration, temperature, and noise sensors in a power plant's turbine operating during cylinder trimming simultaneously collect vibration, temperature, and noise signals from the turbine body and transmit these signals in real time to a data acquisition and processing unit. The data acquisition and processing unit processes and fuses the signals to generate fused signal data. This fused signal data is then sent to a control unit, which remotely monitors the results, displays them in real time, triggers alarms when necessary, and stores the data. By synchronously collecting diverse data signals from multiple sensors, including vibration, temperature, and noise sensors, multi-dimensional monitoring of the turbine's operating status during cylinder trimming is achieved, improving both comprehensiveness and accuracy. Data fusion processing of vibration, temperature, and noise signals enhances signal reliability and stability, reducing false alarms and missed alarms.

[0011] In an optional embodiment, the vibration sensor, temperature sensor, and noise sensor are detachably mounted on the turbine body via mounting brackets. When the sensors need maintenance, calibration, or replacement, personnel can simply remove the mounting brackets and remove the sensors from the turbine body. This convenient operation reduces equipment downtime and improves maintenance efficiency.

[0012] In one optional embodiment, the mounting bracket between the temperature sensor and the turbine body is constructed of a thermally conductive material, while the mounting brackets between the temperature sensor and the noise sensor and the turbine body are constructed of an insulating material. The use of a highly thermally conductive material for the temperature sensor mounting bracket ensures that temperature changes in the turbine body are quickly transmitted to the temperature sensor, improving the sensitivity and accuracy of temperature detection. The use of an insulating material for the noise sensor mounting bracket isolates the noise sensor from interference from the turbine body's heat, preventing noise signal distortion caused by temperature fluctuations and ensuring that the temperature and noise sensors can operate stably and accurately in high-temperature environments.

[0013] In an optional embodiment, the data acquisition and processing unit includes:

[0014] Multiple independent acquisition modules are respectively connected to the vibration sensor, temperature sensor and noise sensor to receive vibration signals, temperature signals and noise signals in a classified manner;

[0015] The data cleaning module is connected to the independent acquisition module. The data cleaning module is used to interpolate and fill in missing data. The vibration signal adopts polynomial interpolation, the temperature signal adopts mean or median filling, and the noise signal adopts exponentially weighted moving average filtering;

[0016] The data analysis module is used to identify outliers and generate fused signal data of vibration signals, temperature signals and noise signals through grouping and aggregation operations.

[0017] During operation, the independent acquisition module first classifies and receives the signals from each sensor, and then passes the signals to the data cleaning module. The data cleaning module interpolates and fills missing data based on the characteristics of different types of signals. Vibration signals restore missing parts through polynomial interpolation, temperature signals use the mean or median to fill missing values, and noise signals use exponentially weighted moving average filtering to fill missing parts and smooth noise. The cleaned data enters the data analysis module, which identifies and processes outliers. Then, through grouping and aggregation operations, the vibration, temperature, and noise signal data are fused into fused signal data, providing comprehensive and accurate data support for subsequent equipment status assessments, and realizing accurate monitoring of the turbine cylinder cutting operation status.

[0018] In an optional implementation, the data cleaning module performs exponential smoothing on the vibration signal and the noise signal after the interpolation and filling process to eliminate high-frequency interference, and screens valid data segments through Boolean indexing.

[0019] The interpolated vibration and noise signals first enter exponential smoothing, removing high-frequency noise and interference, making them smoother and more stable. The data cleaning module then uses Boolean indexing to filter valid data segments from the smoothed signals based on preset conditions and rules, eliminating invalid or interfering data. This further improves data quality and usability, ensuring high accuracy and reliability for subsequent analysis and processing.

[0020] In an optional embodiment, the control unit is configured to generate a diagnostic report based on the fusion results of multi-sensor data and predict potential failure trends based on a preset model.

[0021] The control unit receives fused signal data from the data acquisition and processing unit and uses built-in data analysis algorithms and fault diagnosis models to conduct in-depth mining and analysis of the fused data. This data is then used to generate a diagnostic report, accurately assessing the current operating status of the turbine cylinder cutout. Furthermore, based on accumulated historical data and pre-set fault prediction models, the control unit predicts potential fault trends related to the turbine cylinder cutout, identifying potential hidden faults in advance. This provides a scientific basis for the development of preventive maintenance and repair strategies, effectively reducing the risk of sudden equipment failures.

[0022] In an optional embodiment, the control unit includes a storage module, which adopts a multi-body cross storage architecture for classified storage of original data, fused signal data, alarm records and analysis reports of vibration signals, temperature signals and noise signals.

[0023] The storage module receives and stores raw vibration, temperature, and noise signal data from the data acquisition and processing unit, as well as the processed fused signal data. This module also stores alarm records generated by the control unit when an alarm is triggered, as well as analysis reports generated based on data analysis. The multi-body cross-storage architecture enables efficient storage and management of various data types based on category, chronological order, or other rules. This facilitates rapid query, recall, and analysis of historical data, providing comprehensive data support for long-term equipment operation monitoring, fault analysis, and performance optimization, ensuring data security and integrity.

[0024] In an optional embodiment, the control unit includes an alarm module, which includes an audible and visual alarm component and a notification module, and is configured to trigger an audible and visual alarm according to a preset threshold value and send an alarm signal to the outside.

[0025] The alarm module monitors the fused signal data transmitted by the data acquisition and processing unit in real time. When vibration, temperature, or noise indicators in the data exceed preset normal thresholds, the audio and visual alarm component immediately activates, alerting on-site personnel to equipment anomalies through flashing lights and piercing beeps. Simultaneously, the notification module converts the alarm signal into a text message, email, or other communication protocol message, and promptly sends it to remote operations and maintenance personnel. This ensures that relevant personnel are immediately aware of equipment failures and can take timely measures to address them, effectively reducing fault handling response time and improving the safety and reliability of equipment operation.

[0026] In an optional embodiment, a power supply unit is also included to provide power to the detection unit, data acquisition and processing unit, and control unit. This ensures that all components of the entire monitoring system can operate normally, preventing data acquisition interruptions, data processing errors, or control function failures caused by voltage fluctuations or power failures, thereby ensuring the reliability of the turbine cylinder cutting operation vibration detection device.

[0027] In an optional embodiment, the power supply unit includes an external power supply module and at least one set of backup power supply modules arranged in parallel. The external power supply module is responsible for providing power to the turbine cylinder cutting operation vibration detection device to ensure the normal operation of the detection device. Once the external power supply module fails or becomes abnormal, such as low voltage, power outage, etc., the backup power supply module will be put into use immediately to connect the power supply task, provide continuous power support for the turbine cylinder cutting operation vibration detection device, enhance the redundancy and reliability of the power supply system, ensure that the turbine cylinder cutting operation vibration detection device can operate uninterruptedly under any circumstances, and will not cause data loss or monitoring interruption due to power problems, providing solid power guarantee for the safe operation of the turbine cylinder cutting. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0029] Figure 1 A schematic diagram of a power plant steam turbine cylinder cutting operation vibration detection device provided by an embodiment of the present invention.

[0030] Explanation of the accompanying drawings: 100, detection unit, 110, vibration sensor, 120, temperature sensor, 130, noise sensor, 200, data acquisition and processing unit, 300, control unit, 310, communication module, 320, display module, 330, alarm module, 340, storage module, 400, power supply unit. DETAILED DESCRIPTION

[0031] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0032] The following combination Figure 1 , describing embodiments of the present invention.

[0033] According to an embodiment of the present invention, on one hand, a vibration detection device for a power plant steam turbine during cylinder cutting operation is provided, comprising a detection unit 100 , a data acquisition and processing unit 200 , and a control unit 300 .

[0034] The detection unit 100 includes a vibration sensor 110, a temperature sensor 120, and a noise sensor 130, which are used to synchronously collect vibration, temperature, and noise signals. The vibration sensor 110, temperature sensor 120, and noise sensor 130 are respectively installed at predetermined locations on the turbine body. The data acquisition and processing unit 200 is electrically connected to the detection unit 100 and is used to receive and process the vibration, temperature, and noise signals to generate fused signal data. The control unit 300 communicates with the data acquisition and processing unit 200 for remote monitoring, real-time display of detection results, triggering of alarms, and data storage.

[0035] When the vibration detection device for a power plant turbine operating during cylinder trimming performs vibration detection, the vibration sensor 110, temperature sensor 120, and noise sensor 130 first synchronously collect vibration, temperature, and noise signals from the turbine body. These collected signals are transmitted in real time to the data acquisition and processing unit 200. After receiving these signals, the data acquisition and processing unit 200 processes and fuses them to generate fused signal data. This fused signal data is then sent to the control unit 300. The control unit 300's primary responsibilities include remote monitoring, real-time display of detection results, triggering alarms when necessary, and storing data. By using multiple sensor types—the vibration sensor 110, temperature sensor 120, and noise sensor 130—to synchronously collect different types of signals, multi-dimensional monitoring of the turbine's cylinder trimming operation status can be achieved, thereby improving the comprehensiveness and accuracy of the monitoring. By fusing the vibration, temperature, and noise signals, the reliability and stability of the signals can be effectively enhanced, thereby reducing false alarms and missed alarms.

[0036] In one embodiment, the vibration sensor 110 , the temperature sensor 120 , and the noise sensor 130 are fixed to a preset detection position of the steam turbine body via a detachable mounting bracket, and the mounting bracket includes a locking mechanism and a quick release interface.

[0037] When the vibration sensor 110 , the temperature sensor 120 or the noise sensor 130 needs to be maintained, calibrated or replaced, the operator can remove the target sensor together with the mounting bracket from the turbine body by releasing the locking mechanism and separating the quick release interface.

[0038] Furthermore, the mounting bracket can utilize one or more of the following fixing methods: magnetic attraction, snap-on, or threaded connection to accommodate various installation scenarios. This removable mounting method simplifies sensor assembly and disassembly, eliminating the complex operations required by traditional welding or bolting. This minimizes equipment downtime, improves maintenance efficiency, and enhances system maintainability.

[0039] In one embodiment, the mounting bracket between temperature sensor 120 and the turbine body is constructed from a metal material with high thermal conductivity, such as a copper alloy or aluminum-based composite material, to ensure that temperature changes in the turbine body are quickly transferred to the sensitive elements of temperature sensor 120 through heat conduction. The mounting brackets for noise sensor 130 and vibration sensor 110 are constructed from a material with low thermal conductivity, such as ceramic fiber or polyimide composite material, to prevent the impact of thermal radiation from the turbine body on the internal circuitry of noise sensor 130.

[0040] Furthermore, a thermal insulation layer or heat dissipation structure can be added between the mounting brackets for the noise sensor 130 and vibration sensor 110 and the turbine body. For example, a heat-reflective coating or integrated micro-heat sink can be applied to the bracket surface to further optimize temperature isolation, thereby ensuring the reliability of multi-source data fusion analysis. Sound isolators can be installed on the mounting brackets for the vibration sensor 110 and temperature sensor 120 to reduce the impact of noise on the vibration sensor 110 and temperature sensor 120, thereby ensuring signal transmission.

[0041] In one embodiment, the data acquisition and processing unit 200 includes multiple independent acquisition modules, a data cleaning module, and a data analysis module. The independent acquisition modules are connected to the vibration sensor 110, temperature sensor 120, and noise sensor 130 via shielded cables or wireless transmission protocols, respectively. They utilize time-division multiplexing to receive vibration, temperature, and noise signals at a preset sampling rate and convert them into digital signals via analog-to-digital converters. The data cleaning module is connected to the multiple independent acquisition modules and is used to restore missing vibration signals using cubic spline polynomial interpolation, dynamically select a mean or median fill-in strategy for missing temperature signal values based on historical data, and smooth missing noise signal segments using an adaptive exponentially weighted moving average filter algorithm. The data analysis module identifies outliers using boxplots or a standard score (Z-score) algorithm, performs aggregation calculations on vibration signals grouped by frequency domain features, temperature signals grouped by time windows, and noise signals grouped by energy spectral density, ultimately generating fused signal data containing multi-dimensional indicators. In some other embodiments, the data cleaning module may integrate wavelet transforms or Kalman filtering algorithms to further enhance data repair capabilities under complex operating conditions.

[0042] When the independent acquisition module is running, it first classifies and receives the signals from each sensor, and then passes the signals to the data cleaning module. The data cleaning module interpolates and fills missing data based on the characteristics of different signal types. Vibration signals restore missing parts through polynomial interpolation, temperature signals use the mean or median to fill missing values, and noise signals use exponentially weighted moving average filtering to fill missing parts and smooth noise. The cleaned data enters the data analysis module, which identifies and processes outliers. Then, through grouping and aggregation operations, it fuses the vibration, temperature, and noise signal data into fused signal data, providing comprehensive and accurate data support for subsequent equipment status assessments and achieving precise monitoring of the turbine cylinder cutting operating status.

[0043] In one embodiment, the data cleaning module performs double exponential smoothing on the interpolated vibration and noise signals, suppressing high-frequency interference components by adjusting the smoothing coefficient; then, Boolean indexing combined with sliding window technology is used to screen out valid data segments whose vibration amplitude is within a preset safety threshold and whose main frequency band of the noise spectrum is within the range of 50-2000Hz. The data cleaning module performs exponential smoothing on the interpolated vibration and noise signals to eliminate high-frequency interference, and screens out valid data segments through Boolean indexing. The interpolated vibration and noise signals first enter the exponential smoothing process to remove high-frequency noise and interference components in the signal, making the signal smoother and more stable. Then, the data cleaning module uses the Boolean indexing function to screen out valid data segments from the smoothed signal according to preset conditions and rules, and excludes invalid or interfering data, further improving data quality and availability, and ensuring that the data for subsequent analysis and processing is highly accurate and reliable.

[0044] In some other embodiments, abnormal data segments can be automatically identified and marked using machine learning algorithms such as the isolation forest algorithm, thereby avoiding the subjectivity of manually setting thresholds and improving the intelligence level of data screening.

[0045] In one embodiment, the control unit 300 calls a preset fault diagnosis knowledge base containing typical fault characteristic spectra such as bearing wear, blade cracks, and shaft misalignment based on the fused signal data, generates a diagnostic report, and details the abnormality type, severity level, and recommended treatment measures; at the same time, a prediction model constructed using a long short-term memory neural network or a random forest algorithm is used to analyze the vibration trend, temperature rise rate, and noise spectrum offset in the historical data, and output the probability distribution of potential faults in the next 72 hours.

[0046] In some other embodiments, digital twin technology can also be introduced to compare measured data with theoretical operating parameters in real time through a three-dimensional simulation model of the turbine to improve the accuracy of fault prediction.

[0047] The control unit 300 receives the fused signal data from the data acquisition and processing unit 200 and uses built-in data analysis algorithms and fault diagnosis models to conduct in-depth mining and analysis of the fused data. This data is then used to generate a diagnostic report, accurately assessing the current operating status of the turbine cylinder cutout. Furthermore, based on accumulated historical data and pre-set fault prediction models, the control unit 300 predicts potential fault trends associated with cylinder cutouts, identifying potential failures in advance. This provides a scientific basis for the development of preventive maintenance and repair strategies, effectively reducing the risk of sudden equipment failures.

[0048] In one embodiment, the storage module utilizes a multi-bank cross-storage architecture, storing the original time-domain waveform data of the vibration signal, the temperature-time series data of the temperature signal, and the frequency-domain spectrum data of the noise signal in independent solid-state storage units. Redundancy check mechanisms ensure data integrity. Alarm records and analysis reports utilize a hybrid storage model of relational and unstructured databases. Blockchain technology can be deployed for encrypted storage and distributed backup of critical data, preventing data tampering and enhancing anti-attack capabilities.

[0049] The storage module receives and stores raw vibration, temperature, and noise signal data from the data acquisition and processing unit 200, as well as the processed fused signal data. This module also stores alarm records generated by the control unit 300 when an alarm is triggered, as well as analysis reports generated based on data analysis. The multi-body cross-storage architecture enables efficient storage and management of various data types based on category, chronological order, or other rules. This facilitates rapid query, access, and analysis of historical data, providing comprehensive data support for long-term equipment operation monitoring, fault analysis, and performance optimization, ensuring data security and integrity.

[0050] In one embodiment, the control unit 300 includes an alarm module 330 , which includes an audible and visual alarm component and a notification module, and is configured to trigger an audible and visual alarm according to a preset threshold and send an alarm signal to the outside.

[0051] The alarm module 330 monitors the fused signal data transmitted by the data acquisition and processing unit 200 in real time. When vibration, temperature, or noise indicators in the data exceed preset normal thresholds, the audio and visual alarm component immediately activates, alerting on-site personnel to the equipment anomaly through flashing lights and piercing beeps. Simultaneously, the notification module converts the alarm signal into a text message, email, or other communication protocol message, and promptly sends it to remote operations and maintenance personnel. This ensures that relevant personnel are immediately aware of the equipment failure and can take timely measures to address it, effectively reducing fault handling response time and improving the safety and reliability of equipment operation.

[0052] Furthermore, a multi-level response mechanism can be integrated into the alarm module 330: when the vibration amplitude exceeds the first-level threshold, the sound and light alarm component triggers a yellow warning light and intermittent buzzing; when the temperature or noise index exceeds the second-level threshold, it switches to a red warning light and a continuous buzzing; the notification module sends an alarm message containing a fault code, real-time data and positioning information to the preset terminal.

[0053] In one embodiment, a power supply unit 400 is also included to provide power to the detection unit 100, the data acquisition and processing unit 200, and the control unit 300. This ensures that all components of the entire monitoring system can operate normally, and prevents data acquisition interruptions, data processing errors, or control function failures caused by voltage fluctuations or power failures, thereby ensuring the reliability of the steam turbine cylinder cutting operation vibration detection device.

[0054] In one embodiment, the power supply unit 400 includes an external power supply module and at least one backup power supply module connected in parallel. The external power supply module is responsible for providing power to the turbine cylinder cut-off vibration detection device, ensuring its normal operation. If the external power supply module experiences a fault or anomaly, such as low voltage or a power outage, the backup power supply module immediately takes over, providing continuous power support to the turbine cylinder cut-off vibration detection device. This enhances the redundancy and reliability of the power supply system, ensuring uninterrupted operation of the turbine cylinder cut-off vibration detection device under all circumstances, preventing data loss or monitoring interruptions due to power failures, and providing a solid power guarantee for the safe operation of the turbine cylinder cut-off. The backup power supply can be a lithium-ion battery pack, a fuel cell, or a flywheel energy storage device. When the voltage of the external power supply module drops below 70% of the rated value or the frequency deviation exceeds ±2 Hz, the switching circuit switches power to the backup power supply module. In some other embodiments, a hybrid power supply system can be configured with solar photovoltaic panels and wind turbines, suitable for deployment in the field or off-grid power plants.

[0055] In this embodiment, the control unit 300 includes a communication module 310, a display module 320, an alarm module 330, and a storage module 340, which are used to communicate, alarm, and store data from the data acquisition and processing unit 200. The communication module 310 enables wireless data transmission, thereby enabling personnel to remotely monitor abnormal vibration during the power plant steam turbine cylinder cut operation. The display module 320 and the alarm module 330 provide timely reminders and warnings of abnormal vibration during the power plant steam turbine cylinder cut operation, thereby ensuring that personnel can immediately handle the abnormal vibration during the power plant steam turbine cylinder cut operation. The storage module 340 stores the alarm information to facilitate subsequent improvements to the power plant steam turbine cylinder cut operation. The communication module 310 includes multiple wireless communication methods, and the communication module 310 enables wireless data transmission, thereby enabling personnel to remotely monitor abnormal vibration during the power plant steam turbine cylinder cut operation. The display module 320 includes several display screens for classifying and displaying different data of the detection unit 100. The display module 320 and the alarm module 330 can provide timely reminders and warnings of abnormal vibrations during the cylinder cutting operation of the power plant turbine, thereby ensuring that the staff can deal with the abnormal vibrations during the cylinder cutting operation of the power plant turbine in the first time.

[0056] The power plant steam turbine cylinder cutting operation vibration detection device provided in this embodiment collects different variables through multiple vibration sensors 110, temperature sensors 120 and noise sensors 130. The collected data enters the detection unit 100, and the detection unit 100 loads the data collected by the three sensors into the Python Pandas library. The vibration sensor 110, the temperature sensor 120 and the noise sensor 130 are cleaned through the data stored in the Python Pandas library. First, missing value processing is performed, and the missing values of the vibration sensor 110 are processed by the polynomial interpolation method. The missing values of the temperature sensor 120 are filled by using the mean, median or mode. The missing values of the noise sensor are filled by exponential weighted moving average filtering. Then, outliers are identified by the histogram of the visualization tool. After identification, if the number of outliers is small, they are deleted or corrected. If the number of outliers is large, the vibration sensor 110, the temperature sensor 120 and the noise sensor 130 need to be checked for status. The formats and units of the vibration sensor 110, the temperature sensor 120, and the noise sensor 130 are unified. High-frequency noise or interference signals are removed from the data of the vibration sensor 110 and the noise sensor 130 through exponential smoothing. Then, the data is selected and filtered according to indexes, conditions, labels, etc., and the data is filtered using Boolean indexes or conditional expressions. The cleaned data is grouped according to the values of one or more columns, and then various aggregation operations are performed on each group to obtain the sum, average, maximum, and minimum values of the data. The data of the vibration sensor 110, the temperature sensor 120, and the noise sensor 130 are merged and connected to create a pivot table. The data of the vibration sensor 110, the temperature sensor 120, and the noise sensor 130 are then summarized and counted. The detection unit 100 then transmits the summarized data to the data acquisition and processing unit 200, and performs corresponding operations based on the data. This can effectively reduce the situation where abnormal vibration of the power plant steam turbine cylinder cutting cannot be discovered in time due to missed reports or false reports by a single sensor.

[0057] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A power plant steam turbine cylinder cutting operation vibration detection device, characterized in that: include: A detection unit (100) includes a vibration sensor (110), a temperature sensor (120), and a noise sensor (130), and is used to synchronously collect vibration signals, temperature signals, and noise signals, wherein the vibration sensor (110), the temperature sensor (120), and the noise sensor (130) are respectively installed at preset positions of the steam turbine body; A data acquisition and processing unit (200) is connected to the detection unit (100) and is used to receive and process the vibration signal, the temperature signal and the noise signal to obtain fused signal data; The control unit (300) communicates with the data acquisition and processing unit (200) and is used for remote monitoring, real-time display of detection results, triggering alarms and storing data.

2. The power plant steam turbine cylinder cutting operation vibration detection device according to claim 1, characterized in that: The vibration sensor (110), the temperature sensor (120), and the noise sensor (130) are detachably mounted on the steam turbine body via a mounting bracket.

3. The power plant steam turbine cylinder cutting operation vibration detection device according to claim 2, characterized in that: The mounting bracket between the temperature sensor (120) and the steam turbine body is made of a heat-conducting material, and the mounting brackets between the temperature sensor (120) and the noise sensor (130) and the steam turbine body are made of a heat-insulating material.

4. The power plant steam turbine cylinder cutting operation vibration detection device according to any one of claims 1 to 3, characterized in that: The data acquisition and processing unit (200) comprises: A plurality of independent acquisition modules are respectively connected to the vibration sensor (110), the temperature sensor (120) and the noise sensor (130), and are used to receive vibration signals, temperature signals and noise signals in a classified manner; A data cleaning module is connected to the independent acquisition module, and is used to interpolate and fill missing data, wherein the vibration signal adopts polynomial interpolation, the temperature signal adopts mean or median filling, and the noise signal adopts exponentially weighted moving average filtering; The data analysis module is used to identify outliers and generate fused signal data of vibration signals, temperature signals and noise signals through grouping and aggregation operations.

5. The power plant steam turbine cylinder cutting operation vibration detection device according to claim 4, characterized in that: The data cleaning module performs exponential smoothing on the vibration signal and noise signal after interpolation and filling processing to eliminate high-frequency interference, and screens valid data segments through Boolean indexing.

6. The power plant steam turbine cylinder cutting operation vibration detection device according to any one of claims 1 to 3, characterized in that: The control unit (300) is configured to generate a diagnostic report based on the fusion results of multi-sensor data and predict potential fault trends based on a preset model.

7. The power plant steam turbine cylinder cutting operation vibration detection device according to any one of claims 1 to 3, characterized in that: The control unit (300) includes a storage module, which adopts a multi-body cross storage architecture and is used for classified storage of original data, fused signal data, alarm records and analysis reports of vibration signals, temperature signals and noise signals.

8. The power plant steam turbine cylinder cutting operation vibration detection device according to any one of claims 1 to 3, characterized in that: The control unit (300) includes an alarm module (330), which includes an audible and visual alarm component and a notification module, and is used to trigger an audible and visual alarm according to a preset threshold value and send an alarm signal to the outside.

9. The power plant steam turbine cylinder cutting operation vibration detection device according to any one of claims 1 to 3, characterized in that: It also includes a power supply unit (400) for providing power support for the detection unit (100), the data acquisition and processing unit (200) and the control unit (300).

10. The power plant steam turbine cylinder cutting operation vibration detection device according to claim 9, characterized in that: The power supply unit (400) comprises an external power supply module and at least one set of backup power supply modules arranged in parallel.

Images