Data slimming method based on online analysis of reactor core refined calculation data
By setting up a data sampling area in the reactor core for weighted integral calculation and low-resolution reconstruction, the storage range is narrowed and data is converted into basic shared data, the problem of large amount of core refinement calculation is solved, and the core performance is achieved Quick analysis and efficient storage of state.
Patent Information
- Application Number
- CN202210278526.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-21
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-03-21
AI Technical Summary
Refined calculation of reactor cores leads to large amount of data, large amount of storage resources, and difficult and slow post-processing analysis, which leads to difficulty in analyzing the core performance status and affects safety and working performance.
Using the data slimming method based on reactor core, the weighted integral calculation is performed by setting the data sampling area, reconstructing the data at low resolution, narrowing the storage range, converting the data into basic common data and quantitative conversion relationship data, excluding overlapping data, and making correlation Analysis to enable efficient storage and rapid analysis of data.
It realizes rapid and timely analysis of core performance status, reduces storage resources and time consumption, reduces invalid data storage, and improves data processing efficiency.
Smart Images

Figure CN114741376B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of reactor core fine calculation analysis, and in particular to a data slimming method based on online analysis of reactor core fine calculation data. Background Art
[0002] Carrying out refined computational fluid dynamics (CFD) calculations on the reactor core can obtain refined flow and heat transfer data of the core, guide the design and optimization of the core characteristic structure and operation, and is of great significance to the economy and safety of the reactor. However, the reactor core fine calculation has many problems, such as a large number of fine grids, a large amount of calculation data, a large amount of calculation storage resources, and difficult and slow post-processing and analysis of calculation data. The main reasons are: the characteristic structures such as the core positioning grid and fuel elements are geometrically complex, including a variety of curvature design, including many millimeter-level and micron-level small size designs, and the number is tens of thousands or even hundreds of thousands. The fine calculation of the whole core requires tens of billions or even hundreds of billions of fine grids, and the huge amount of grids causes a huge amount of calculation data storage. At the same time, a large amount of computing memory is required for post-processing of large-scale calculation data of the reactor core, and the transmission time of large data is long, and the post-processing analysis is prone to jamming. The fine data is difficult to be analyzed quickly, and therefore, due to the lack of analysis conclusions, it is difficult to quickly determine whether the performance status of each area inside the core will lead to reduced safety and working performance. In addition, if all parameters of the entire core are stored at the same level of high resolution (such as sub-millimeter level), a large amount of useless and over-high-resolution data will be stored, which will increase the burden of data transmission and storage.
[0003] Therefore, it is an urgent problem for those skilled in the art to propose a data slimming method based on online analysis of reactor core refined calculation data to solve the problems existing in the prior art. Summary of the invention
[0004] In view of this, the present invention provides a data slimming method based on online analysis of reactor core refined calculation data, which reduces the amount of data, reduces the storage resources and storage time consumed by core refined calculation, and realizes rapid and timely analysis and post-processing display of core performance status.
[0005] In order to achieve the above object, the present invention adopts the following technical solution:
[0006] The data slimming method based on online analysis of reactor core refined calculation data includes the following steps:
[0007] S101: setting one or more of point, line, surface, and volume data sampling and analysis areas at different spatial positions in the reactor core to obtain fine data DATA1;
[0008] S102: during analysis or after analysis is paused or stopped, and before the fine data DATA1 is stored, weighted integral calculation of important parameters of different core partitions is performed to complete low-resolution reconstruction of the fine data DATA1, and obtain data DATA2;
[0009] S103: analyzing whether the performance status of each spatial position of the core is abnormal and important, determining whether data is stored in each area and what time and space resolution is used for storage, and determining that the stored data becomes DATA3;
[0010] S104: reducing the data to be stored in the core to a specific time and space range, reducing the data DATA3 to data DATA4 within the specific time and space range;
[0011] S105: converting part of the data in the data DATA4 into common basic data and quantitative conversion relationship data, and arranging the unconverted data in the planned storage data DATA4 and the quantitative conversion relationship data formed after the conversion in DATA4 into the data to be stored DATA5;
[0012] S106: transforming the data DATA5 to be stored by a quantized mathematical formula, and replacing the corresponding transformed data with the transformed mathematical formula data to form the data DATA6 to be stored;
[0013] S107: Perform correlation analysis between the data DATA6 to be stored and the existing stored data, exclude the data in DATA6 that overlap with the existing data, and determine the new data DATA7 to be stored;
[0014] S108: Continue to analyze the correlation between the data to be stored DATA7 and the existing stored data, exclude the data in DATA7 that can be quantized and reconstructed by the existing data, and add the quantized and reconstructed mathematical relationship as storage data to DATA7 to form new data to be stored DATA8.
[0015] Optionally, the important parameters in S102 include temperature, pressure and flow rate, and the weighted integral calculation includes line weighted integral calculation, area weighted integral calculation and volume weighted integral calculation.
[0016] Optionally, the specific content of S103 is: compare the values and distribution of important parameters in the fine data DATA1 or data DATA2 and the spatiotemporal gradients of important parameters with the preset parameter extreme values and gradient extreme values to determine whether there is an abnormality (exceeding the preset extreme values). If there is no abnormality, the data will not be stored or stored at a low resolution. If it is an abnormality, the data will be stored and stored at a high resolution to obtain the storage data of data DATA3.
[0017] Optionally, the specific content of S104 is: according to the storage requirements and important parameters, confirm the spatial scope that affects the core safety and operating performance level, narrow the scope of the data storage area, and at the same time confirm the time period that affects the core performance level, narrow the time range of data storage, and obtain data DATA4 with reduced data storage volume.
[0018] Optionally, in S105 , all or part of the data in the data DATA4 is converted into basic common data and quantization conversion relationship data.
[0019] Optionally, in S106, quantitative configuration characteristic analysis and conversion are performed on part or all of the data in the data DATA5.
[0020] Optionally, the specific content of S107 is: perform a correlation analysis between the data DATA6 and the data stored in the existing database, determine whether all or part of the data of DATA6 is already in the existing database, and if so, establish an associated calling relationship and save the calling relationship data, while excluding the data in DATA6 that can be quantified and reconstructed by the existing data, and add the quantified and reconstructed mathematical relationship to DATA6 as storage data to form new data DATA7 that needs to be stored.
[0021] Optionally, the process further includes S109: dividing the space where the data DATA8 is stored on demand into multiple partitions, and implementing parallel storage for data to be stored in different partitions.
[0022] It can be seen from the above technical solution that, compared with the prior art, the present invention provides a data slimming method based on online analysis of reactor core refined calculation data: reducing the amount of data, reducing the storage resources and storage time consumed by core refined calculations, and realizing rapid and timely analysis and post-processing display of core performance status. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] 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.
[0024] Figure 1 A flow chart of the data slimming method based on online analysis of reactor core refined calculation data provided by the present invention. DETAILED DESCRIPTION
[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0026] Reference Figure 1 As shown, the present invention discloses a data slimming method based on online analysis of reactor core refined calculation data, comprising the following steps:
[0027] S101: setting one or more of point, line, surface, and volume data sampling and analysis areas at different spatial positions in the reactor core to obtain fine data DATA1;
[0028] S102: during analysis or after analysis is paused or stopped, and before the fine data DATA1 is stored, weighted integral calculation of important parameters of different core partitions is performed to complete low-resolution reconstruction of the fine data DATA1, and obtain data DATA2;
[0029] S103: analyzing whether the performance status of each spatial position of the core is abnormal and important, determining whether data is stored in each area and what time and space resolution is used for storage, and determining that the stored data becomes DATA3;
[0030] S104: reducing the data to be stored in the core to a specific time and space range, reducing the data DATA3 to data DATA4 within the specific time and space range;
[0031] S105: converting part of the data in the data DATA4 into common basic data and quantitative conversion relationship data, and arranging the unconverted data in the planned storage data DATA4 and the quantitative conversion relationship data formed after the conversion in DATA4 into the data to be stored DATA5;
[0032] S106: transforming the data DATA5 to be stored by a quantized mathematical formula, and replacing the corresponding transformed data with the transformed mathematical formula data to form the data DATA6 to be stored;
[0033] S107: Perform correlation analysis between the data DATA6 to be stored and the existing stored data, exclude the data in DATA6 that overlap with the existing data, and determine the new data DATA7 to be stored;
[0034] S108: Continue to analyze the correlation between the data to be stored DATA7 and the existing stored data, exclude the data in DATA7 that can be quantized and reconstructed by the existing data, and add the quantized and reconstructed mathematical relationship as storage data to DATA7 to form new data to be stored DATA8.
[0035] Furthermore, in S101, for the reactor core, one or more of the four types of regional objects, namely, point, line, surface, and body, are set as regional objects for core data sampling and analysis. The specific type of regional object selected and the size and number of each type of regional object can be set according to the amount of computing resources, but all important power and temperature monitoring positions in the reactor core need to be arranged to obtain the state characteristics of important spatial positions and support safety and performance analysis. In addition, for the setting of body-type regional objects, the number can be controlled and spatially arranged continuously to cover the entire core. By changing the size of the body region, performance monitoring of the core with different spatial resolutions can be obtained.
[0036] Furthermore, the important parameters in S102 include temperature, pressure and flow rate, and the weighted integral calculation includes line weighted integral calculation, area weighted integral calculation and volume weighted integral calculation.
[0037] Furthermore, in the process of refined numerical iteration, according to the set line, surface and volume analysis area range, and for the important parameter data (temperature, pressure, velocity, etc.) within the fine grid, weighted integral calculations of different important parameters are carried out on the fine grids on the line, surface and volume area objects, respectively, to obtain the weighted average value of the important parameters in each range; for the set point spatial position, the value grid of the important parameters is determined according to the point location, and the important parameter values of the grid are used as the values of the important parameters at each point position.
[0038] The main purpose of this content is to be used for subsequent rapid status analysis, and to use low-resolution spatiotemporal data instead of refined data for data storage and analysis to reduce the amount of data stored when the needs are met.
[0039] Furthermore, the specific content of S103 is: compare the values and distribution of important parameters in the fine data DATA1 or data DATA2 and the spatiotemporal gradients of important parameters with the preset parameter extreme values and gradient extreme values to determine whether there is an abnormality. If there is no abnormality, the data will not be stored or stored at a low resolution. If it is an abnormality, the data will be stored and stored at a high resolution to obtain the storage data of data DATA3.
[0040] Furthermore, the main work of performing anomaly and importance analysis on the important parameter values and distribution and the important parameter spatiotemporal gradients in the reactor core data DATA1 or DATA2 is as follows:
[0041] 1) In the numerical iterative calculation, the average values of important parameters in different point, line, surface, and volume data sampling areas are analyzed and compared. At the same time, for some important parameters (one or more of pressure, temperature, and flow rate), the spatial gradients of important parameters between different regions are calculated, and the time gradients of important parameters at different time intervals are calculated. The gradient can be calculated using refined data or low-resolution spatiotemporal data. The selection principle is still based on the research accuracy requirements and the amount of computing resources.
[0042] 2) Perform macroscopic distribution analysis on important parameters of reactor core refined calculation data, learn the distribution characteristics of important parameters over time and space, and determine whether the non-uniformity of parameter distribution will affect safety and operational performance; determine whether the spatiotemporal distribution of important parameters is important enough for operation and safety research. When judging macroscopic distribution and trends, low-resolution spatiotemporal reconstruction data can be used for analysis to improve analysis efficiency.
[0043] 3) Screen and identify the extreme values and gradient extreme values of each important parameter, determine whether the values of important parameters are abnormal, determine whether the gradient of the spatial and temporal distribution of parameters is abnormal, and then determine whether the analyzed working condition will damage safety or significantly affect the performance level of the reactor, that is, whether the analyzed working condition needs to be stored for detailed research. The parameter extreme value screening judgment can screen the refined data or the average values of different low-resolution areas. The selection principle is also based on the research accuracy requirements and the amount of computing resources.
[0044] The main tasks of online analysis and timely feedback of reactor core refined data are:
[0045] 1) When computing resources are sufficient, ensure that the refined data that is dynamically updated is calculated while performing fine calculations, and perform low-resolution reconstruction, macro trend analysis, extreme value screening, anomaly judgment, and importance analysis.
[0046] 2) If computing resources are insufficient, low-resolution reconstruction of fine-scale computing data, macro trend analysis, extreme value screening, anomaly judgment, and importance analysis are completed after the calculation is completed and before the data is saved.
[0047] In this part of the study, the meaning of "online" refers to the process state of data calculation, processing, and analysis on computing resources, and should be the process state before the data is stored.
[0048] Furthermore, the specific content of S104 is: according to the storage requirements and important parameters, confirm the spatial scope that affects the core safety and operating performance level, narrow the scope of the data storage area, and at the same time confirm the time period that affects the core performance level, narrow the time range of data storage, and obtain data DATA4 with reduced data storage volume.
[0049] Furthermore, the main work of reducing the time range and space range of the required saved data DATA3 is as follows:
[0050] 1) According to the demand and parameter calculation and judgment, confirm the spatial range that affects the safety and operating performance level of the core, narrow the data storage area, and at the same time confirm the time period that affects the performance level of the core, and narrow the time range of data storage. That is: confirm that the fine calculation data or low-resolution data in time and space within a specific time range and a specific spatial range are the data to be stored DATA4, avoid fine data storage in the full time domain and full space domain, and achieve a reduction in the amount of data storage.
[0051] 2) This part of the content is equivalent to: no data storage planning is made for data in non-critical working conditions and spatial areas.
[0052] It should be noted that the parameter data required to be saved in different time and space domains can be the same or different.
[0053] Furthermore, in S105, all or part of the data in the data DATA4 is converted into basic common data and quantization conversion relationship data.
[0054] Furthermore, the amount of data to be stored can be reduced by converting all or part of the target storage data DATA4 into basic common data and quantitative conversion relationship data, that is, when the target storage data can be decomposed into data D(1), data D(2)...data D(N), if one or more groups of data such as D(N2) or D(N3) can be reconstructed based on the data D(N1), then D(N1) is the basic common data (1≤N1≤N; 1≤N2≤N; 1≤N3≤N). By converting part or all of the target storage data DATA4 into basic common data and quantitative conversion relationship data as much as possible, the amount of data storage can be reduced. After the judgment and conversion of this part, the data to be saved becomes DATA5.
[0055] Furthermore, in S106, quantitative configuration characteristic analysis and conversion are performed on part or all of the data in the data DATA5.
[0056] Furthermore, part or all of the required stored data DATA5 is subjected to quantitative configuration characteristic analysis and conversion to reduce the amount of data to be stored.
[0057] The converted mathematical formula data is the following about pressure P, speed V in three directions x 、V y 、V z , temperature T, and other special considerations in specific studies Data formula data between time and space coordinates (x, y, z, t).
[0058]
[0059] That is, part or all of the data to be saved is converted into mathematical formula data as much as possible, and it is necessary to ensure that the data before conversion or similar data can be reconstructed through the coordinate position and the saved mathematical formula, and the degree of similarity can be limited by the relative error and absolute error defined by the user. After the judgment and conversion of this part, the data to be saved becomes DATA6.
[0060] Furthermore, the specific content of S107 is: perform a correlation analysis between data DATA6 and the data stored in the existing database, determine whether all or part of the data of DATA6 is already in the existing database, and if so, establish an associated calling relationship and save the calling relationship data, while excluding the data in DATA6 that can be quantified and reconstructed by the existing data, and add the quantified and reconstructed mathematical relationship to DATA6 as storage data to form new data DATA7 that needs to be stored.
[0061] Furthermore, the specific work of analyzing the correlation between the data DATA6 to be saved and the data stored in the existing database and establishing a correlation calling relationship to reduce the amount of data to be stored is as follows:
[0062] Analyze and confirm whether all or part of the data DATA6 to be saved in the above-mentioned content exists in the saved database or other data carriers, or can be constructed through the stored data. If so, an associated calling relationship is established, and the calling relationship data + data without an associated relationship are saved. The sum of these two parts of data is used as the data DATA7 to be saved, and the data that can be converted is not stored, so as to organize the storage data and reduce the amount of data storage.
[0063] Furthermore, the method further includes S109: dividing the space where the data DATA8 is stored into multiple partitions as needed, and implementing parallel storage for data to be stored in different partitions.
[0064] Furthermore, after the above-mentioned analysis and confirmation process, the establishment of quantitative formulas, and the establishment of associated calling relationships, the data DATA8 that needs to be saved in the core is finally divided into multiple partitions according to the space where the data is located, and parallel storage is implemented for the data that needs to be saved in different partitions, further improving data storage efficiency.
[0065] The characteristics of the present invention are mainly embodied in:
[0066] 1) Through online analysis and low-resolution data reconstruction, the core status is quickly and timely (supported by low-resolution data reconstruction) fed back during the reactor core fine calculation process (supported by online analysis), and based on this, it is determined whether the calculated fine calculation data needs to be stored, or what time and space range and what resolution level of data needs to be stored, thereby reducing data storage and resource consumption.
[0067] 2) Converting the target stored data into basic data and quantitative conversion relationship data, quantitative formulas, and establishing quantitative associations with the stored data can extract the most essential data reflecting the physical mechanism of the reactor core, which is equivalent to concentrating and refining the core data, deeply reducing the storage of invalid and redundant data, and significantly reducing data storage and resource consumption.
[0068] 3) The spatial multi-partition distribution and parallel storage of the core multi-resolution data can be carried out, which can perform differentiated data storage at different levels of detail according to the different spatial state characteristics of the core, avoiding the waste of resources caused by over-detailed data storage.
[0069] The present invention is based on the multi-dimensional (point, line, surface and solid) and multi-resolution data analysis area division, data processing and online analysis of the reactor core refined calculation. The present invention analyzes low-resolution data to quickly recognize the performance status of different positions in the core, determine the value of refined data storage and analysis, and guide whether to store data in each area of the core and what time and space resolution to use for data storage. The amount of data storage is continuously reduced by reducing the time and space scope of data storage, establishing a quantitative relationship between the data to be stored and the basic data and replacing the data to be stored, converting the data to be stored into a quantitative formula for storage, analyzing the quantitative association between the data to be stored and the stored data and replacing the data to be stored, and multi-partition multi-resolution distributed parallel data storage, so as to reduce the consumption of storage hardware resources and data processing time.
[0070] The above description of the disclosed embodiments is provided in a progressive manner to enable those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those 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 data slimming method based on online analysis of reactor core refined calculation data, characterized in that: The following steps are involved: S101: setting one or more of point, line, surface, and volume data sampling and analysis areas at different spatial positions in the reactor core to obtain fine data DATA1; S102: during analysis or after analysis is paused or stopped, and before the fine data DATA1 is stored, weighted integral calculation of important parameters of different core partitions is performed to complete low-resolution reconstruction of the fine data DATA1, and obtain data DATA2; S103: analyzing whether the performance status of each spatial position of the core is abnormal and important, determining whether data is stored in each area and what time and space resolution is used for storage, and determining that the stored data becomes DATA3; S104: reducing the data to be stored in the core to a specific time and space range, reducing the data DATA3 to data DATA4 within the specific time and space range; S105: converting part of the data in the data DATA4 into common basic data and quantitative conversion relationship data, and arranging the unconverted data in the planned storage data DATA4 and the quantitative conversion relationship data formed after the conversion in DATA4 into the data to be stored DATA5; S106: transforming the data DATA5 to be stored by a quantized mathematical formula, and replacing the corresponding transformed data with the transformed mathematical formula data to form the data DATA6 to be stored; S107: Perform correlation analysis between the data DATA6 to be stored and the existing stored data, exclude the data in DATA6 that overlap with the existing data, and determine the new data DATA7 to be stored; S108: Continue to analyze the correlation between the data to be stored DATA7 and the existing stored data, exclude the data in DATA7 that can be quantized and reconstructed by the existing data, and add the quantized and reconstructed mathematical relationship as storage data to DATA7 to form new data to be stored DATA8.
2. The data slimming method based on online analysis of reactor core refined calculation data according to claim 1 is characterized in that: The important parameters in S102 include temperature, pressure and flow rate, and the weighted integral calculation includes line weighted integral calculation, area weighted integral calculation and volume weighted integral calculation.
3. The data slimming method based on online analysis of reactor core refined calculation data according to claim 1 is characterized in that: The specific content of S103 is: compare the important parameter values and distribution in the fine data DATA1 or data DATA2 and the important parameter spatiotemporal gradients with the preset parameter extreme values and gradient extreme values to determine whether there is an abnormality. If there is no abnormality, the data will not be stored or stored at a low resolution. If it is an abnormality, the data will be stored and stored at a high resolution to obtain the storage data of data DATA3.
4. The data slimming method based on online analysis of reactor core refined calculation data according to claim 1 is characterized in that: The specific content of S104 is: according to the storage requirements and important parameters, confirm the spatial scope that affects the safety and operating performance level of the core, narrow the scope of the data storage area, and at the same time confirm the time period that affects the performance level of the core, narrow the time range of data storage, and obtain data DATA4 with reduced data storage volume.
5. The data slimming method based on online analysis of reactor core refined calculation data according to claim 1, characterized in that: In S105, all or part of the data in the data DATA4 is converted into basic common data and quantization conversion relationship data.
6. The data slimming method based on online analysis of reactor core refined calculation data according to claim 1 is characterized in that: In S106, quantitative configuration characteristic analysis and conversion are performed on part or all of the data in DATA5.
7. The data slimming method based on online analysis of reactor core refined calculation data according to claim 1 is characterized in that: The specific content of S107 is: perform a correlation analysis between data DATA6 and the data stored in the existing database, determine whether all or part of the data of DATA6 is already in the existing database, and if so, establish an associated calling relationship and save the calling relationship data. At the same time, exclude the data in DATA6 that can be quantified and reconstructed by the existing data, and add the quantified and reconstructed mathematical relationship to DATA6 as storage data to form new data DATA7 that needs to be stored.
8. The data slimming method based on online analysis of reactor core refined calculation data according to any one of claims 1 to 7, characterized in that: The process also includes S109: dividing the space where the data DATA8 is stored on demand into multiple partitions, and performing parallel storage for the data to be stored in different partitions.
Citation Information
Patent Citations
Multiscale reactor core thermal hydraulic analysis method
CN108875213A
Method for parallel calculation of reactor core thermal hydraulic distribution
CN108897907A
Sub lattice cell scale reactor thermal hydraulic control body division method
CN110807246A
Reactor core assembly drainage basin rapid CFD calculation method based on fine flow field reconstruction
CN110826178A
Reactor core model update system
US5024801A
