Method and system for identifying external cooling boiling of supercritical water cooled reactor

Abstract

The invention provides a supercritical water-cooled reactor external cooling boiling identification method and system, and relates to the technical field of supercritical water-cooled reactor cooling, and the method comprises the steps: obtaining an external image of a supercritical water-cooled reactor, and obtaining a lower end socket boundary according to image identification; the image comprises the lower sealing head and cooling water outside the lower sealing head; preprocessing the image; performing gas film detection and reconstruction on the preprocessed image to obtain a reconstructed gas film interface; and determining a boiling mode according to the boundary of the lower sealing head and the reconstructed gas film interface. According to the embodiment of the invention, the lower end enclosure boundary and the gas film interface are determined based on the external image of the supercritical water cooled reactor, and the current boiling mode is specifically determined according to the spatial characteristics of the gas film interface and whether the gas film interface is in contact with the lower end enclosure boundary in time, so that the cooling capacity is accurately evaluated.

Description

Technical Field

[0001] The invention relates to the technical field of supercritical water-cooled reactor cooling, and in particular to a supercritical water-cooled reactor external cooling boiling identification method and system. Background Art

[0002] The pressure vessel of a supercritical water reactor is usually a sealed shell with a lower head underneath, which is responsible for containing the radioactive materials inside the reactor. The safety of a supercritical water reactor is of vital importance, and external cooling is required to prevent damage to the lower head and ensure its integrity.

[0003] Currently, water is usually used to cool the lower head. When low-temperature water comes into contact with a high-temperature reactor, it will boil violently. The existing scheme cannot accurately determine the boiling mode, and different boiling modes correspond to different cooling capacities, resulting in inaccurate assessment of the cooling capacity. Summary of the invention

[0004] To solve the above problems, an embodiment of the present invention provides a method for identifying boiling in external cooling of a supercritical water-cooled reactor, comprising: acquiring an image of the outside of a supercritical water-cooled reactor, and obtaining a lower head boundary based on the image recognition; the image includes the lower head and cooling water outside the lower head; preprocessing the image; performing air film detection and reconstruction on the preprocessed image to obtain a reconstructed air film interface; determining a boiling mode based on the lower head boundary and the reconstructed air film interface; the boiling modes include nucleate boiling, film boiling and transition boiling.

[0005] The supercritical water-cooled reactor external cooling boiling identification method provided by the embodiment of the present invention determines the lower head boundary and the air film interface based on the image of the outside of the supercritical water-cooled reactor, and specifically determines the current boiling mode according to the spatial characteristics of the air film interface and whether the air film interface is in contact with the lower head boundary in time, thereby accurately evaluating the cooling capacity.

[0006] Optionally, the air film detection and reconstruction are performed on the preprocessed image to obtain a reconstructed air film interface, including: performing air film detection on the preprocessed image to determine the boundary of the air film, wherein the boundary includes multiple breakpoints; inputting the curvature of the breakpoint into a pretrained machine learning model, predicting the curvature extended from the breakpoint, and determining the position of the next point based on the curvature extended from the breakpoint; inputting the curvature of the next point into the machine learning model, predicting the curvature extended from the next point, and determining the position of the next adjacent point based on the curvature extended from the next point; repeating the step of determining the position of the next adjacent point until the breakpoint is completely repaired to obtain the reconstructed air film interface.

[0007] The embodiment of the present invention provides a method for repairing the breakpoint by predicting the curvature extending from the breakpoint based on a machine learning model, which can reconstruct a complete air-film interface.

[0008] Optionally, determining the boiling mode according to the lower head boundary and the reconstructed air film interface includes: if the air film interface is not closed in space, and the air film interface is not in contact with the lower head boundary within a preset time length, then determining that the boiling mode is film boiling; if the air film interface is not closed in space, and the air film interface is in intermittent contact with the lower head boundary within a preset time length, then determining that the boiling mode is transitional boiling; if the air film interface is closed in space and presents an elliptical curvature, and the air film interface is in intermittent contact with the lower head boundary within a preset time length, then determining that the boiling mode is nucleate boiling.

[0009] In the embodiment of the present invention, the current boiling mode can be accurately identified based on the spatial characteristics of the air film interface and whether the air film interface is in contact with the lower head boundary in time.

[0010] Optionally, the preprocessing of the image includes: performing wavelet denoising on the image, and / or performing binarization processing on the image.

[0011] The embodiment of the present invention provides a specific method of preprocessing, thereby enhancing the air film characteristics in the image.

[0012] Optionally, the method further includes: acquiring a tangent line of the breakpoint, and calculating a curvature of the breakpoint according to a slope of the tangent line.

[0013] In the embodiment of the present invention, the curvature can be calculated based on the slope of the breakpoint, so as to repair the breakpoint.

[0014] Optionally, performing air film detection on the preprocessed image to determine the boundary of the air film includes: performing air film detection on the preprocessed image using a Laplace Gaussian operator to determine the boundary of the air film.

[0015] In the embodiment of the present invention, the Laplace Gaussian operator is specifically used to detect the boundary of the air film, and features such as edges can be accurately found in the image.

[0016] Optionally, the method further includes: obtaining curvatures of bubbles of various sizes under different boiling modes as training sets and test sets; and training a machine learning model for predicting the curvatures of bubbles of different sizes based on the training sets and the test sets.

[0017] In the embodiment of the present invention, a machine learning model is trained based on the curvature of bubbles of various sizes under different boiling modes to achieve curvature prediction.

[0018] Optionally, the machine learning model is a machine learning model based on an extreme gradient boosting algorithm.

[0019] In the embodiment of the present invention, a machine learning model of an extreme gradient boosting algorithm is specifically adopted to achieve curvature prediction.

[0020] Optionally, a dynamic threshold method is used for binarization.

[0021] In the embodiment of the present invention, a dynamic threshold method is specifically used for binarization, and the threshold is calculated based on local statistical information of the image, so that each region can obtain a suitable binarization effect.

[0022] The embodiment of the present invention provides a supercritical water-cooled reactor external cooling boiling recognition system, comprising: an image acquisition module, used to acquire an image of the outside of a supercritical water-cooled reactor, and obtain a lower head boundary according to the image recognition; the image includes the lower head and cooling water outside the lower head; a preprocessing module, used to preprocess the image; an air film detection module, used to perform air film detection and reconstruction on the preprocessed image to obtain a reconstructed air film interface; a boiling mode recognition module, used to determine a boiling mode according to the lower head boundary and the reconstructed air film interface; the boiling modes include nucleate boiling, film boiling and transition boiling.

[0023] The supercritical water-cooled reactor external cooling boiling identification system of the embodiment of the present invention can achieve the same technical effect as the supercritical water-cooled reactor external cooling boiling identification method described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0025] Figure 1 A schematic flow chart of a method for identifying boiling in a supercritical water-cooled reactor externally provided in an embodiment of the present invention;

[0026] Figure 2 A schematic diagram of breakpoint repair in an embodiment of the present invention;

[0027] Figure 3 is a schematic flow chart of an air film breakpoint reconstruction algorithm in an embodiment of the present invention;

[0028] Figure 4 A schematic structural diagram of a supercritical water-cooled reactor external cooling boiling identification system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0030] Since low-temperature water will boil violently when it contacts a high-temperature reactor, and different boiling modes have different cooling capacities, it is impossible to effectively evaluate the cooling capacity without accurately identifying the boiling mode. Based on this, an embodiment of the present invention provides a method for identifying boiling in external cooling of a supercritical water-cooled reactor, which obtains the lower head boundary and the air film interface through image recognition, and classifies based on the spatiotemporal judgment criteria to accurately determine the boiling mode.

[0031] Figure 1 A schematic flow chart of a method for identifying boiling in a supercritical water-cooled reactor externally provided in an embodiment of the present invention is shown. The method comprises:

[0032] S102, acquiring an image of the exterior of the supercritical water-cooled reactor, and obtaining a lower head boundary based on the image recognition.

[0033] The above image includes the lower head and the cooling water outside the lower head. Optionally, in this embodiment, the image data is obtained based on a high-speed camera.

[0034] The lower head of a supercritical water reactor is usually approximated as a sphere. Based on the original image data, the lower head with a known position can be identified.

[0035] S104, preprocessing the above image.

[0036] The purpose of preprocessing in this embodiment is to enhance the image and achieve the purpose of improving the characteristics of the air film. Optionally, the image preprocessing adopts a wavelet denoising method. Exemplarily, the thresholding method of wavelet transform is used for image denoising steps including: 1. wavelet decomposition of the image signal; 2. threshold quantization of the high-frequency coefficients after decomposition; 3. wavelet reconstruction of the image signal.

[0037] Optionally, a dynamic threshold method is used for binarization. Exemplarily, the average grayscale value of any block is calculated, and the grayscale value of each pixel in the block is compared with the average grayscale value. If it is greater than the average grayscale value, the grayscale value of the pixel is set to 255, otherwise the grayscale value of the pixel is set to 0.

[0038] S106, performing air film detection and reconstruction on the preprocessed image to obtain a reconstructed air film interface.

[0039] Specifically, the Laplace Gaussian operator can be used to detect the air film on the preprocessed image to determine the boundary of the air film. In this embodiment, the Laplace Gaussian operator is used to detect the air film, and its basic equation is:

[0040]

[0041] Among them, 2 is the Laplace operator σ is the standard deviation of the Gaussian kernel, G(x,y) is a two-dimensional Gaussian function with standard deviation σ, and xy is the coordinate axis.

[0042] Due to image signal interference, noise, reflection and other factors, it is difficult to directly obtain a clear air film interface, and what is obtained after air film detection is a discontinuous curve (bubble boundary). Based on the above-mentioned discontinuous characteristics of the air film, the breakpoints of the curve are reconstructed in this embodiment.

[0043] Specifically, the above air film detection and reconstruction can be performed in the following manner:

[0044] First, air film detection is performed on the preprocessed image to determine the boundary of the air film. There are usually multiple breakpoints on the air film.

[0045] Secondly, the curvature of the above breakpoint is input into a pre-trained machine learning model to predict the curvature of the breakpoint, and the position of the next point is determined according to the curvature of the breakpoint. Exemplarily, the machine learning model is a machine learning model based on an extreme gradient boosting algorithm.

[0046] At the breakpoint of the air film, the tangent line of the point can be determined, so as to obtain the slope of the point, and the curvature of the point can be calculated from the slope. After obtaining the slope of the breakpoint, it is input into the machine learning model to predict and complete the curvature, so as to determine the position of the next point.

[0047] Then, the curvature of the next point is input into the machine learning model to predict the curvature of the next point, and the position of the next adjacent point is determined based on the curvature of the next point. The above steps of determining the position of the next adjacent point are repeated until the breakpoint is completely repaired to obtain the reconstructed air-film interface.

[0048] S108, determining a boiling mode according to the lower head boundary and the reconstructed air film interface. The boiling modes can be divided into nucleate boiling, film boiling and transition boiling.

[0049] In this embodiment, there are three boiling modes, namely nucleate boiling, film boiling and transition boiling, which can be classified based on time and space judgment criteria.

[0050] If the air film interface is not closed in space, and the air film interface does not contact the lower head boundary within a preset time, the boiling mode is determined to be film boiling;

[0051] If the air film interface is not closed in space, and the air film interface is in intermittent contact with the lower head boundary within a preset time, the boiling mode is determined to be transition boiling;

[0052] If the air film interface is closed in space and presents an elliptical curvature, and the air film interface is in intermittent contact with the lower head boundary within a preset time, the boiling mode is determined to be nucleate boiling.

[0053] The supercritical water-cooled reactor external cooling boiling identification method provided by the embodiment of the present invention determines the lower head boundary and the air film interface based on the image of the outside of the supercritical water-cooled reactor, and specifically determines the current boiling mode according to the spatial characteristics of the air film interface and whether the air film interface is in contact with the lower head boundary in time, thereby accurately evaluating the cooling capacity.

[0054] In this embodiment, the model is trained using image data of the exterior of a supercritical water-cooled reactor in a public database, and the image may include bubbles of various sizes. Specifically, the curvatures of bubbles of various sizes under different boiling modes are first obtained as training sets and test sets; a machine learning model for predicting the curvatures of bubbles of different sizes is obtained based on the training sets and test sets.

[0055] The geometric curvature of bubbles of different sizes is obtained from the public database as a training set for training, and the bubble curvature based on a database different from the above public database is used as a test set to test the effect of machine learning. After learning, at the current air film breakpoint, the tangent line of the point is obtained as the slope. The curvature can be directly calculated based on the slope. After obtaining the curvature of the breakpoint, machine learning prediction is performed, and then the curvature is completed.

[0056] Figure 2 FIG. 2 shows a schematic diagram of breakpoint repair in an embodiment of the present invention. Figure 2 Point A is a breakpoint on the air-film interface (the point on its right is missing), line 1 is the tangent through point A, points a, b, and c are the points to be repaired later, and lines 2, 3, and 4 are the tangents through the above points a, b, and c, respectively. Based on the above machine learning model and the curvature of point A, the curvature of point a can be predicted to determine the position of point a. In the same way, the positions of b and c can be obtained in sequence until they are connected to the breakpoint on the right.

[0057] Figure 3 A schematic flow chart of the air film breakpoint reconstruction algorithm in an embodiment of the present invention is shown, and the extreme gradient boosting algorithm autonomously learns the air film interface curvature as an example for explanation, which specifically includes the following steps:

[0058] S301, calculating the breakpoint curvature.

[0059] S302, the curvature slope is matched with the extreme gradient boosting algorithm to obtain the optimal solution curvature.

[0060] S303, obtain the position of the next point by limit approximation.

[0061] S304, repair of air film breakpoint.

[0062] S305, determine whether the interface is closed. If yes, the air film reconstruction is completed; if no, return to S301.

[0063] For example, in this embodiment, based on a public database, an extreme gradient boosting algorithm is used to autonomously learn the curvature of the air film interface under different boiling modes. At both ends of the breakpoint, based on the optimal solution slope of the extreme gradient boosting algorithm, the slope can determine the next position of the breakpoint repair extension. The above process can be calculated simultaneously at both ends of the breakpoint until the breakpoint is completely repaired and the interface is closed.

[0064] In this embodiment, the tangent of the breakpoint can be obtained, and then the curvature of the breakpoint can be calculated based on the slope of the tangent. The curvature can be calculated based on the slope as follows:

[0065] Assume that the rectangular coordinate equation of the boundary curve of the air film is y = f(x), and y = f(x) has a second-order derivative, and the slope of the tangent line of the curve at point M is y′ = tanα, so

[0066]

[0067] again Therefore, the curvature of the curve at point M is:

[0068]

[0069] Figure 4 The schematic diagram of the structure of the supercritical water-cooled reactor external cooling boiling identification system provided by an embodiment of the present invention is shown, and the system includes:

[0070] An image acquisition module 401 is used to acquire an image of the exterior of the supercritical water-cooled reactor and obtain a lower head boundary according to the image recognition; the image includes the lower head and cooling water outside the lower head;

[0071] A preprocessing module 402, used for preprocessing the image;

[0072] An air film detection module 403 is used to perform air film detection and reconstruction on the preprocessed image to obtain a reconstructed air film interface;

[0073] The boiling mode recognition module 404 is used to determine the boiling mode according to the lower head boundary and the reconstructed air film interface; the boiling mode includes nucleate boiling, film boiling and transition boiling.

[0074] The supercritical water-cooled reactor external cooling boiling identification system provided by the embodiment of the present invention determines the lower head boundary and the air film interface based on the image of the outside of the supercritical water-cooled reactor, and specifically determines the current boiling mode according to the spatial characteristics of the air film interface and whether the air film interface is in contact with the lower head boundary in time, thereby accurately evaluating the cooling capacity.

[0075] The embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is read and executed by a processor, the method provided in the above embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0076] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the control device through a computer, and the program can be stored in a computer-readable storage medium. When the program is executed, it may include the processes of the above-mentioned method embodiments, wherein the storage medium may be a memory, a disk, an optical disk, etc.

[0077] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the claims.

[0078] In this article, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0079] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0080] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying boiling in external cooling of a supercritical water-cooled reactor, characterized in that: include: Acquire an image of the exterior of a supercritical water-cooled reactor, and obtain a lower head boundary based on the image recognition; the image includes the lower head and cooling water outside the lower head; Preprocessing the image; Perform air film detection and reconstruction on the preprocessed image to obtain the reconstructed air film interface; The boiling mode is determined according to the lower head boundary and the reconstructed air film interface; the boiling mode includes nucleate boiling, film boiling and transition boiling.

2. The method according to claim 1, characterized in that The air film detection and reconstruction of the preprocessed image to obtain a reconstructed air film interface includes: Performing air film detection on the preprocessed image to determine the boundary of the air film, wherein the boundary includes a plurality of breakpoints; Inputting the curvature of the breakpoint into a pre-trained machine learning model, predicting the curvature of the breakpoint, and determining the position of the next point according to the curvature of the breakpoint; Inputting the curvature of the next point into the machine learning model, predicting the curvature of the next point, and determining the position of the next adjacent point according to the curvature of the next point; Repeat the step of determining the position of the next adjacent point until the breakpoint is completely repaired to obtain a reconstructed air-film interface.

3. The method according to claim 1 or 2, characterized in that: The determining of the boiling mode according to the lower head boundary and the reconstructed air film interface includes: If the air film interface is not closed in space, and the air film interface does not contact the lower head boundary within a preset time, then the boiling mode is determined to be film boiling; If the air film interface is not closed in space, and the air film interface is in intermittent contact with the lower head boundary within a preset time, then the boiling mode is determined to be transition boiling; If the air film interface is closed in space and presents an elliptical curvature, and the air film interface is in intermittent contact with the lower head boundary within a preset time length, then the boiling mode is determined to be nucleate boiling.

4. The method according to claim 1 or 2, characterized in that: The preprocessing of the image includes: performing wavelet denoising on the image, and / or performing binarization processing on the image.

5. The method according to claim 2, characterized in that: The method further comprises: A tangent line of the breakpoint is obtained, and the curvature of the breakpoint is calculated according to the slope of the tangent line.

6. The method according to claim 2, characterized in that The step of performing air film detection on the preprocessed image to determine the boundary of the air film includes: The Laplace Gaussian operator is used to detect the air film on the preprocessed image to determine the boundary of the air film.

7. The method according to claim 2, characterized in that: The method further comprises: The curvatures of bubbles of various sizes under different boiling modes are obtained as training sets and test sets; A machine learning model for predicting the curvature of bubbles of different sizes is obtained based on the training set and the test set.

8. The method according to claim 7, characterized in that The machine learning model is a machine learning model based on the extreme gradient boosting algorithm.

9. The method according to claim 4, characterized in that The dynamic threshold method is used for binarization.

10. A supercritical water-cooled reactor external cooling boiling identification system, characterized in that: include: An image acquisition module is used to acquire an image of the outside of the supercritical water-cooled reactor and obtain a lower head boundary according to the image recognition; the image includes the lower head and cooling water outside the lower head; A preprocessing module, used for preprocessing the image; An air film detection module is used to detect and reconstruct the air film on the preprocessed image to obtain the reconstructed air film interface; A boiling mode recognition module is used to determine the boiling mode according to the lower head boundary and the reconstructed air film interface; the boiling mode includes nucleate boiling, film boiling and transition boiling.

Images