Capacitive cylinder differential expansion measuring device for steam turbine

Abstract

The embodiment of the invention provides a capacitance type cylinder expansion difference measuring device for a steam turbine. The capacitance type cylinder expansion difference measuring device comprises a capacitance type cylinder expansion sensor and a conditioning circuit. The capacitance type cylinder expansion sensor comprises a first capacitance plate and a second capacitance plate, the first capacitance plate is arranged on the side, facing the differential expansion support, of the rotor side measuring disc, the second capacitance plate is arranged on the side, facing the measuring disc, of the differential expansion support, and the second capacitance plate and the first capacitance plate are oppositely arranged in a spaced mode; the capacitive cylinder expansion sensor is used for detecting the relative displacement change between the measuring disc and the differential expansion bracket and converting the displacement change into a capacitance signal; the conditioning circuit is used for conditioning the capacitance signal, converting the capacitance signal into a direct current signal in direct proportion to differential expansion, and sending the direct current signal to an upper computer; and the upper computer is used for determining the expansion difference value of the cylinder body according to the received direct current signal. And the measurement result is more accurate and reliable.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of steam turbine detection, and in particular to a capacitive cylinder expansion difference measuring device for a steam turbine. Background Art

[0002] Rotating machinery monitoring systems serve as the "stethoscope" that safeguards unit operation. The accuracy of their measured values ​​and the reliability of their protective actions are crucial to the safe and economical operation of rotating machinery. Sensors, acting as the sensing front end of monitoring, measure values ​​that directly reflect the unit's operating status. Some of these values ​​also contribute to the design of protection logic, making the accuracy of sensor measurements crucial to the unit's safe and stable operation.

[0003] Cylinder expansion is also a key measurement in the monitoring system. When a steam turbine heats up during startup, cools down during operation, or experiences load changes, the turbine cylinder and rotor expand and contract to a certain extent. Due to differences in mass and heated surface area, the expansion and contraction rates differ, resulting in a relative difference (referred to as differential expansion). When this differential expansion reaches a certain value, it can cause friction between the moving and stationary components. Existing turbine cylinder expansion sensors generally use a displacement principle. A sensor is mounted on the turbine cylinder to measure the relative displacement between the cylinder and rotor. This principle converts the distance measurement between the cylinder and rotor surfaces into the distance measurement between a point mounted on the cylinder (the displacement sensor) and the rotor. Therefore, this type of displacement sensor places extremely high demands on the sensor's mounting position: the displacement measurement point must be sufficiently flat and representative of the overall cylinder operating status. Summary of the Invention

[0004] The embodiment of the present invention aims to solve at least one of the technical problems existing in the prior art, and provides a capacitive cylinder expansion difference measuring device for a steam turbine.

[0005] An embodiment of the present invention provides a capacitive cylinder expansion differential measurement device for a steam turbine, comprising: a capacitive cylinder expansion sensor and a conditioning circuit; the capacitive cylinder expansion sensor comprises a first capacitor plate and a second capacitor plate, the first capacitor plate being disposed on a side of a rotor-side measuring disk facing a differential expansion bracket, and the second capacitor plate being disposed on a side of the differential expansion bracket facing the measuring disk and spaced relative to the first capacitor plate; The capacitive cylinder expansion sensor is used to detect the relative displacement change between the measuring disk and the expansion differential bracket, and convert the displacement change into a capacitive signal; The conditioning circuit is used to condition the capacitance signal, convert it into a DC signal proportional to the expansion difference, and send the DC signal to the host computer; The host computer is used to determine the expansion difference value of the cylinder body according to the received DC signal.

[0006] In some possible embodiments, the capacitive cylinder expansion sensor further includes a first polyvinyl chloride sheet and a second polyvinyl chloride sheet; The first polyvinyl chloride sheet is fixed to a side of the measuring disk facing the expansion differential bracket, and the first capacitor plate is bonded to a side of the first polyvinyl chloride sheet facing the expansion differential bracket; The second polyvinyl chloride sheet is fixed to a side of the differential expansion bracket facing the measuring disk, and the second capacitor plate is bonded to a side of the second polyvinyl chloride sheet facing the measuring disk.

[0007] In some possible embodiments, the orthographic projection of the first polyvinyl chloride sheet on the first capacitor plate falls outside the first capacitor plate; The orthographic projection of the second polyvinyl chloride sheet on the second capacitor plate falls outside the second capacitor plate.

[0008] In some possible embodiments, the conditioning circuit includes: an inverse proportional circuit, a high-pass filter circuit, a rectifier circuit, a low-pass filter circuit, and an AD conversion circuit; The reverse proportional circuit is used to convert the capacitance signal into an AC voltage signal; The high-pass filter circuit is used to extract the AC component in the AC voltage signal; The rectifier circuit is used to convert the AC voltage signal into a positive pulsating DC signal; The low-pass filter circuit is used to remove the ripple signal in the pulsating DC signal, smooth the output, and convert it into a DC voltage signal; The AD conversion circuit is used to convert the DC voltage signal into a digital signal and send it to the host computer.

[0009] In some possible embodiments, the high-pass filtering circuit is a high-pass filter, the rectifier circuit is a bridge rectifier circuit, and the low-pass filtering circuit is a low-pass filter.

[0010] In some possible embodiments, the host computer is further configured to: After removing outliers from the received DC signal, the signal is fed into a pre-trained deep learning model for prediction; wherein the deep learning model adopts an LSTM neural network architecture and is trained based on historical differential expansion data to predict the normal differential expansion range under the current working conditions; When the deviation between the predicted value and the actual measured value exceeds the set threshold, it is judged as an abnormal state and a risk warning signal is issued according to the degree of deviation.

[0011] In some possible embodiments, the host computer is further configured to: The Z-Score standardization method is used to detect and eliminate outliers in the DC signal, and interpolation is performed on missing data; Use time-frequency analysis methods and variational autoencoder VAE to extract the time-frequency features and probability distribution features of the signal; Build a time series prediction model based on the LSTM network and predict the future output voltage value through a sliding window method; Establish a spatial vector representation model for four states: normal, slightly abnormal, moderately abnormal, and severely abnormal; Calculate the spatial distance between the current data features and the four types of status in real time, and determine the device status based on the nearest neighbor principle; When an abnormal state is detected, an early warning signal of the corresponding level is triggered according to the abnormality level.

[0012] In some possible embodiments, the conditioning circuit is arranged in a shielding box, the shielding box is made of metal, and is arranged inside the cylinder body by a rigid connection.

[0013] In some possible embodiments, the conditioning circuit is electrically connected to the first capacitor plate and the second capacitor plate via a shielded wire.

[0014] In some possible embodiments, the first capacitor plate and the second capacitor plate have a length ranging from 80 mm to 100 mm and a width ranging from 30 mm to 50 mm.

[0015] The capacitive cylinder expansion differential measurement device for a steam turbine according to an embodiment of the present invention has a capacitive cylinder expansion sensor with a simple structure, small size, and low cost. By applying an excitation signal to the capacitive cylinder expansion sensor, measuring the capacitance between the capacitor plates, and processing the capacitance signal, the process is simple and accurate. This invention achieves cylinder expansion value measurement by outputting a DC signal through a conditioning circuit, reducing costs while converting the distance between a point and a surface into the distance between surfaces. This reduces the flatness requirements for the plane and makes the measurement results more accurate and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] Figure 1 Schematic diagram of the structure of a capacitive cylinder expansion difference measuring device for a steam turbine according to an embodiment of the present invention; Figure 2 for Figure 1 A bottom view of the installation location of a capacitive cylinder expansion sensor in a capacitive cylinder differential expansion measuring device for a steam turbine shown in FIG; Figure 3 Schematic diagram of the differential expansion measurement process according to an embodiment of the present invention; Figure 4 Schematic diagram of the structure of the conditioning circuit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention is further described below in conjunction with the accompanying drawings and specific embodiments. It is apparent that the described embodiments are only a portion of the embodiments of the present invention, rather than all of them. Based on the described embodiments of the present invention, all other embodiments obtained by those skilled in the art without requiring creative effort are within the scope of protection of the present invention.

[0019] Unless otherwise specified, the technical terms or scientific terms used in the embodiments of the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The terms "including" or "comprising" used in the embodiments of the present invention neither limit the shapes, numbers, steps, actions, operations, components, originals and / or their groups mentioned, nor exclude the appearance or addition of one or more other different shapes, numbers, steps, actions, operations, components, originals and / or their groups, or the addition of these. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number and order of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0020] Unless otherwise specifically stated, the relative arrangements of the components and steps, numerical expressions, and numerical values ​​set forth in these embodiments do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn in accordance with actual proportional relationships, and that the techniques, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods, and devices shown should be considered part of the authorized specification. In all examples shown and discussed herein, any specific other examples may have different values. It should be noted that similar symbols and letters represent similar items in the following figures, and therefore, once an item is defined in one figure, it does not need to be further discussed in subsequent figures.

[0021] In the description of the embodiments of the present invention, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In the embodiments of the present invention, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in the embodiments of the present invention and the features of different embodiments or examples, unless they are mutually inconsistent.

[0022] Below, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0023] Figure 1 Schematic diagram of the structure of a capacitive cylinder expansion differential measuring device for a steam turbine according to an embodiment of the present invention. Figure 2 for Figure 1 The bottom view of the installation location of the capacitive cylinder expansion sensor in the capacitive cylinder expansion differential measurement device for steam turbines is shown in FIG. Figure 1 and Figure 2 As shown, an embodiment of the present invention relates to a capacitive cylinder expansion differential measurement device for a steam turbine, comprising a capacitive cylinder expansion sensor and a conditioning circuit. The capacitive cylinder expansion sensor comprises a first capacitor plate 1 and a second capacitor plate 2. The first capacitor plate 1 is disposed on the side of the rotor-side measuring disc 100 facing the expansion differential support 200. The second capacitor plate 2 is disposed on the side of the expansion differential support 200 facing the measuring disc 100, spaced apart from the first capacitor plate 1.

[0024] Specifically, such as Figure 1 As shown, the capacitive cylinder expansion sensor is used to detect the relative displacement change between the measuring disk 100 and the expansion differential bracket 200, and convert the displacement change into a capacitance signal; the conditioning circuit is used to condition the capacitance signal, convert it into a DC signal proportional to the expansion differential, and send the DC signal to the host computer 300; the host computer 300 is used to determine the expansion differential value of the cylinder body based on the received DC signal.

[0025] The capacitive cylinder expansion differential measurement device for a steam turbine according to an embodiment of the present invention has a capacitive cylinder expansion sensor with a simple structure, small size, and low cost. By applying an excitation signal to the capacitive cylinder expansion sensor, measuring the capacitance between the capacitor plates, and processing the capacitance signal, the process is simple and accurate. This invention achieves cylinder expansion value measurement by outputting a DC signal through a conditioning circuit, reducing costs while converting the distance between a point and a surface into the distance between surfaces. This reduces the flatness requirements for the plane and makes the measurement results more accurate and reliable.

[0026] For example, Figure 1 As shown, the capacitive cylinder expansion sensor also includes a first polyvinyl chloride sheet 3 and a second polyvinyl chloride sheet 4; the first polyvinyl chloride sheet 3 is fixed to the side of the measuring disk 100 facing the expansion differential bracket 200, and the first capacitor plate 1 is bonded to the side of the first polyvinyl chloride sheet 3 facing the expansion differential bracket 200; the second polyvinyl chloride sheet 4 is fixed to the side of the expansion differential bracket 200 facing the measuring disk 100, and the second capacitor plate 2 is bonded to the side of the second polyvinyl chloride sheet 4 facing the measuring disk 100.

[0027] In some embodiments, as Figure 1 As shown, the orthographic projection of the first polyvinyl chloride sheet 3 on the first capacitor plate 1 falls outside the first capacitor plate 1; the orthographic projection of the second polyvinyl chloride sheet 4 on the second capacitor plate 2 falls outside the second capacitor plate 2. In other words, Figure 1 As shown, the PVC sheet is slightly larger than the capacitor plate and is rigidly connected to the measuring disc / cylinder expansion bracket on all sides. The majority of the PVC sheet bonded to the capacitor plate is used to rigidly connect to the cylinder body. This ensures insulation between the electrodes and the measuring disc / cylinder body. In some embodiments, the first capacitor plate 1 and the second capacitor plate 2 are both elongated strips, each bonded to the corresponding PVC sheet using a J-2012 epoxy resin adhesive.

[0028] In some embodiments, the capacitor plates mentioned above can be made of conductive metals such as copper or stainless steel. Their size can be determined based on the size and spacing of the measuring disk, measurement accuracy, and other requirements. Unless otherwise specified, the length can be between 80 and 100 mm, and the width can be between 30 and 50 mm.

[0029] When assembling the above-mentioned capacitive cylinder expansion differential measuring device for steam turbines, first, the first capacitor plate 1 and the second capacitor plate 2 are fixed to the center of the first polyvinyl chloride sheet 3 and the second polyvinyl chloride sheet 4 using J-2012 epoxy resin adhesive. The first polyvinyl chloride sheet 3 and the second polyvinyl chloride sheet 4 are connected to the measuring disk 100 and the expansion differential bracket 200 respectively using a rigid structure of screws 5 around them. Ensure that the capacitor plates are insulated from the measuring disk 100 and the expansion differential bracket 200. In some embodiments, the first capacitor plate 1 and the second capacitor plate 2 are connected to the signal conditioning circuit via a shielded wire 6 to reduce interference caused by external stray signals on the electrode output signals.

[0030] For example, Figure 1 and Figure 3 As shown, the conditioning circuit includes: an inverse proportional circuit A, a high-pass filter circuit B, a rectifier circuit C, a low-pass filter circuit D, and an AD conversion circuit E. The inverse proportional circuit A is used to convert the capacitor signal into an AC voltage signal; the high-pass filter circuit B is used to extract the AC component from the AC voltage signal; the rectifier circuit C is used to convert the AC voltage signal into a positive pulsating DC signal; the low-pass filter circuit D is used to remove the ripple signal from the pulsating DC signal, smooth the output, and convert it into a DC voltage signal; the AD conversion circuit E is used to convert the DC voltage signal into a digital signal and send it to the host computer 300. In some embodiments, the high-pass filter circuit is a high-pass filter, the rectifier circuit is a bridge rectifier circuit, and the low-pass filter circuit is a low-pass filter.

[0031] Specifically, in this embodiment, due to the relative displacement of the cylinder body and the rotor during thermal expansion / contraction, the spacing between the capacitor plates mounted on the surfaces of the two will change. At this time, the output capacitance value is inversely proportional to the expansion difference value. The capacitance signal is processed by an inverting proportional circuit A. At this time, the output voltage is inversely proportional to the capacitance value, that is, it is proportional to the expansion difference value. The capacitance signal is processed by the conditioning circuit. When relative displacement occurs between the cylinder body and the rotor, the conditioning circuit will output a voltage, and the output voltage value will be rectified and filtered. The DC voltage after processing is proportional to the relative displacement, thereby achieving the purpose of measuring the cylinder expansion difference. Due to the use of a new type of capacitance sensor, the sensor can be used to measure the spacing between non-flat surfaces, improve sensitivity, reduce the influence of interference, and make the measurement results more accurate. At the same time, the sensor has a low cost, a simple structure, strong resistance to harsh environments, and can be widely used in various industrial processes. At the same time, based on the prediction of future data by the model, abnormal conditions of the cylinder expansion value can be discovered, and risk warning can be achieved.

[0032] For reference Figure 4, the first capacitor plate 1 and the second capacitor plate 2 used for detection form a capacitance sensor Cx, which serves as the feedback capacitor of the inverse proportional circuit. The output of the inverse proportional circuit can be expressed as follows:

[0033] Where: U o —output voltage of the inverse proportional circuit; f—frequency of the input excitation voltage; U s —Input excitation voltage; C x —Capacitance under test; C o —Reference capacitor.

[0034] From the parallel plate capacitance formula we know that:

[0035] Where: Cx—capacitance under test; —Dielectric constant of the medium; S—plate area of ​​parallel plates; d—plate distance between parallel plates.

[0036] That is, the capacitance is inversely proportional to the distance between the parallel plate capacitors.

[0037] Where: U o —Output voltage of the inverse proportional circuit; C o —reference capacitance; —dielectric constant of the medium; S—plate area of ​​parallel plates; U s —Input excitation voltage; d—The distance between the parallel plates.

[0038] From the above formula, we can see that the output voltage is positively correlated with the distance between the parallel plates of the capacitor.

[0039] After high-pass filtering to retain the AC component, the output voltage enters the AC-DC conversion filter module. When the bridge rectifier circuit receives AC voltage input, the diode, due to its unidirectional conduction principle, allows current to flow in only one direction through the load. This converts both the positive and negative half-cycles of the AC voltage into voltages in the same direction, namely, DC voltage. This voltage then passes through the filter circuit, removing the AC component while retaining the DC component, reducing the ripple factor. The resulting stable voltage is then converted to analog-to-digital (A / D) before being uploaded.

[0040] For example, Figure 1 and Figure 3As shown, the host computer 300 is specifically used to: remove outliers from the received DC signal and then send it to a pre-trained deep learning model for prediction; wherein, the deep learning model adopts an LSTM neural network architecture and is trained based on historical expansion difference data to predict the normal expansion difference range under the current working conditions; when the deviation between the predicted value and the actual measured value exceeds a set threshold, it is determined to be an abnormal state, and a risk warning signal is issued according to the degree of deviation.

[0041] Specifically, such as Figure 1 and Figure 3 As shown, the host computer 300 is specifically used to: use the Z-Score standardization method to detect and eliminate outliers of the DC signal, and interpolate missing data; use the time-frequency analysis method and variational autoencoder VAE to extract the time-frequency characteristics and probability distribution characteristics of the signal; build a time series prediction model based on the LSTM network, and predict the future output voltage value through a sliding window method; establish a spatial vector representation model of four types of states: normal, slightly abnormal, moderately abnormal, and severely abnormal; calculate the spatial distance between the current data features and the four types of states in real time, and determine the device state according to the nearest neighbor principle; when an abnormal state is detected, trigger a warning signal of the corresponding level according to the abnormality level.

[0042] Specifically, the host computer 300 first uses the Z_Score method to eliminate outliers that clearly exceed the statistical threshold to improve data quality; it also collects historical values ​​from the past 3 to 6 months as samples, establishes a nonlinear mapping through machine learning to extract feature quantities, and further uses a deep learning model based on time series data to predict data for a period of time in the future.

[0043] The specific historical values ​​collected are output voltage values ​​under stable / normal conditions and output voltage values ​​under marked abnormal conditions (minor, moderate, and severe faults). The stable / normal output voltage values ​​are used for data prediction during model training, while the normal and / or abnormal values ​​are used for data early warning. A variational autoencoder (VAE) is used, combining a probabilistic model and a neural network, to automatically extract data distribution features by maximizing the data likelihood function. This extracts features that reflect the underlying patterns from the original time series. These features can be used for subsequent prediction and anomaly detection. These features include time-domain and frequency-domain features, as well as high-level features automatically extracted using an LSTM deep learning model. Data prediction can be directly performed using the trained deep learning model. Simultaneously with data prediction, the output voltage trends under abnormal conditions are trained and summarized, converted into spatial vectors, and classified into four categories: normal, minor, moderate, and severe abnormal conditions. After the actual input voltage time series is input, the characteristics of the time series are extracted and the spatial distance between it and the four conditions is determined. The closest distance is the actual equipment status, which is then used to predict the three abnormality levels of minor, medium, and severe, and issue risk warnings of different levels. That is, the host computer processes the signal to directly realize the calculation, storage, transmission, and risk warning of the cylinder expansion value. The specific prediction steps are as follows: (1) Data processing Remove outliers and use interpolation to supplement them.

[0044] (2) Data standardization Standardize the data (such as Z-Score standardization) so that the data distribution is within a reasonable range (such as [0,1]).

[0045] (3) Construct features and create sliding windows Use the original data and the corresponding time as features. Set the window size to n (which can be based on the number of points collected per minute) and predict one time point in the future. Use the LSTM (Long Short-Term Memory) model for time series data prediction.

[0046] (4) Actual model prediction After the training is completed, by extracting the feature quantities under the normal state in the early stage and training the model, it is possible to output the voltage value at a certain time point in the future by inputting the time series data of the historical cylinder expansion output voltage.

[0047] (5) Actual status warning The model simultaneously inputs data corresponding to four states: normal, mild, intermediate, and severe. The output voltage trends under abnormal operating conditions are trained and summarized, converted into spatial vectors, and classified into four categories: normal, mild, moderate, and severe abnormal conditions. After inputting the actual voltage time series, the model extracts its features and determines their spatial distance to the four conditions. The closest distance represents the actual device state, allowing predictions of the three abnormality levels: mild, intermediate, and severe.

[0048] In some embodiments, as Figure 1 and Figure 3 As shown, the conditioning circuit is arranged in a shielding box 7 made of metal and rigidly connected to the inner side of the cylinder. The conditioning circuit is electrically connected to the first capacitor plate 1 and the second capacitor plate 2 via a shielding wire 6.

[0049] Specifically, the polyvinyl chloride sheet is in the shape of an elongated strip, and is generally 0.5 mm thick. The length and width depend on the size of the installation position of the measuring disk and the cylinder body, and are generally 80 to 100 mm in length and 40 to 60 mm in width. The shielding box 7 adopts a rectangular structure, with a length, width, and height of 100 mm, 100 mm, and 50 mm respectively. It is made of metal, is grounded as a whole, and is fixed on the cylinder body to form an electrostatic shield. There is a small circular hole on the side of the shielding box, with a diameter of about 5 mm, for the passage of the shielding wire. The conditioning circuit is connected to the capacitive cylinder expansion sensor and the host computer 300 respectively through the shielding wire 6 through the shielding box 7. The circuit board uses chip devices to save space, and its conditioning circuit is located inside the shielding box 7 to avoid interference caused by stray signals.

[0050] It will be understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present invention, and the present invention is not limited thereto. Those skilled in the art will appreciate that various modifications and improvements can be made without departing from the spirit and substance of the present invention, and such modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A capacitive cylinder expansion differential measuring device for a steam turbine, characterized in that: include: A capacitive cylinder expansion sensor and conditioning circuit; the capacitive cylinder expansion sensor includes a first capacitor plate and a second capacitor plate. The first capacitor plate is disposed on the side of the rotor-side measuring disc facing the expansion differential bracket. The second capacitor plate is disposed on the side of the expansion differential bracket facing the measuring disc and is spaced apart from the first capacitor plate. The capacitive cylinder expansion sensor is used to detect the relative displacement change between the measuring disk and the expansion differential bracket, and convert the displacement change into a capacitive signal; The conditioning circuit is used to condition the capacitance signal, convert it into a DC signal proportional to the expansion difference, and send the DC signal to the host computer; The host computer is used to determine the expansion difference value of the cylinder body according to the received DC signal.

2. The capacitive cylinder expansion differential measuring device for a steam turbine according to claim 1, characterized in that: The capacitive cylinder expansion sensor further includes a first polyvinyl chloride sheet and a second polyvinyl chloride sheet; The first polyvinyl chloride sheet is fixed to a side of the measuring disk facing the expansion differential bracket, and the first capacitor plate is bonded to a side of the first polyvinyl chloride sheet facing the expansion differential bracket; The second polyvinyl chloride sheet is fixed to a side of the differential expansion bracket facing the measuring disk, and the second capacitor plate is bonded to a side of the second polyvinyl chloride sheet facing the measuring disk.

3. The capacitive cylinder expansion differential measuring device for a steam turbine according to claim 2, characterized in that: The orthographic projection of the first polyvinyl chloride sheet on the first capacitor plate falls outside the first capacitor plate; The orthographic projection of the second polyvinyl chloride sheet on the second capacitor plate falls outside the second capacitor plate.

4. The capacitive cylinder expansion differential measuring device for a steam turbine according to any one of claims 1 to 3, characterized in that: The conditioning circuit includes: an inverse proportional circuit, a high-pass filter circuit, a rectifier circuit, a low-pass filter circuit and an AD conversion circuit; The reverse proportional circuit is used to convert the capacitance signal into an AC voltage signal; The high-pass filter circuit is used to extract the AC component in the AC voltage signal; The rectifier circuit is used to convert the AC voltage signal into a positive pulsating DC signal; The low-pass filter circuit is used to remove the ripple signal in the pulsating DC signal, smooth the output, and convert it into a DC voltage signal; The AD conversion circuit is used to convert the DC voltage signal into a digital signal and send it to the host computer.

5. The capacitive cylinder expansion differential measuring device for a steam turbine according to claim 4, characterized in that: The high-pass filter circuit is a high-pass filter, the rectifier circuit is a bridge rectifier circuit, and the low-pass filter circuit is a low-pass filter.

6. The capacitive cylinder expansion differential measuring device for a steam turbine according to any one of claims 1 to 3, characterized in that: The host computer is further used for: After removing outliers from the received DC signal, the signal is fed into a pre-trained deep learning model for prediction; wherein the deep learning model adopts an LSTM neural network architecture and is trained based on historical differential expansion data to predict the normal differential expansion range under the current working conditions; When the deviation between the predicted value and the actual measured value exceeds the set threshold, it is judged as an abnormal state and a risk warning signal is issued according to the degree of deviation.

7. The capacitive cylinder expansion differential measuring device for a steam turbine according to claim 6, characterized in that: The host computer is further used for: The Z-Score standardization method is used to detect and eliminate outliers in the DC signal, and interpolation is performed on missing data; Use time-frequency analysis methods and variational autoencoder VAE to extract the time-frequency features and probability distribution features of the signal; Build a time series prediction model based on the LSTM network and predict the future output voltage value through a sliding window method; Establish a spatial vector representation model for four states: normal, slightly abnormal, moderately abnormal, and severely abnormal; Calculate the spatial distance between the current data features and the four types of status in real time, and determine the device status based on the nearest neighbor principle; When an abnormal state is detected, an early warning signal of the corresponding level is triggered according to the abnormality level.

8. The capacitive cylinder expansion differential measuring device for a steam turbine according to any one of claims 1 to 3, characterized in that: The conditioning circuit is arranged in a shielding box, which is made of metal and is arranged inside the cylinder body by a rigid connection.

9. The capacitive cylinder expansion differential measuring device for a steam turbine according to any one of claims 1 to 3, characterized in that: The conditioning circuit is electrically connected to the first capacitor plate and the second capacitor plate through a shielded wire.

10. The capacitive cylinder expansion differential measuring device for a steam turbine according to any one of claims 1 to 3, characterized in that: The length of the first capacitor plate and the second capacitor plate ranges from 80 mm to 100 mm, and the width ranges from 30 mm to 50 mm.

Images