An optimization fitting method for flow pressure drop curve of flame arrester
By applying mean filtering and cubic polynomial fitting to the flow and pressure drop data of flame arresters, and optimizing the residuals using a cosine function, the problem of poor fitting of the flow and pressure drop curve of flame arresters was solved, achieving more accurate fitting and prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 青岛安工装备科技有限公司
- Filing Date
- 2026-02-10
- Publication Date
- 2026-06-05
AI Technical Summary
The existing power function fitting of the flow-pressure drop curve of flame arresters has a large deviation, which cannot meet the requirements of accurate calculation and optimization design in actual operation.
The flow rate and pressure drop scatter data were obtained by mean filtering, fitted with a cubic polynomial function, and the residual function was optimized by a cosine function to obtain an optimized fitting function, which was used to fit the flow rate and pressure drop scatter data.
It significantly improves the fitting effect of the flow rate and pressure drop curve, reduces the influence of noise signals, and facilitates the prediction of the flame arrester's status during actual operation.
Smart Images
Figure CN122152270A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of flame arrester technology, and in particular to an optimized fitting method for the flow-pressure drop curve of a flame arrester. Background Technology
[0002] Currently, the flow rate and pressure drop of flame arresters are typically expressed using a power function. The power function is used to represent the flow rate and pressure drop, and it is represented as a straight line in a logarithmic coordinate system. However, comparing this function curve with actual flow rate and pressure drop scatter plot data reveals a certain deviation between the curve and the actual data, and this deviation increases with the pressure drop of the flame arrester. Therefore, the power function expression is only an approximate fit to the flow rate and pressure drop curve and cannot meet the needs of precise calculation and optimized design of flame arresters in actual operation. Summary of the Invention
[0003] To address the problem of poor fitting between the current function curve and the actual flow rate pressure drop curve, this invention provides an optimized fitting method for the flow rate pressure drop curve of a flame arrester.
[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: An optimized fitting method for the flow-pressure drop curve of a flame arrester includes the following steps: S1, performing flow and pressure drop tests on the flame arrester, and obtaining flow-pressure drop scatter data through mean filtering; S2, performing curve fitting on the obtained flow-pressure drop scatter data to obtain a fitting function, and calculating the residual function between the fitting function and the power function; S3, plotting the residual function, selecting an appropriate optimized residual function based on the graph to refit the residual, and obtaining an optimized fitting function by combining the optimized residual function and the power function; S4, fitting the obtained flow-pressure drop scatter data using the optimized fitting function, testing the fitting effect, and comparing the fitting effect of the optimized fitting function with the fitting effect of the power function to verify the fitting effect of the optimized fitting function.
[0005] Furthermore, in S1, multiple flow and pressure drop tests are performed on the same flame arrester within the same flow variation range to obtain at least five sets of experimental data. The average value of the at least five sets of pressure drop data corresponding to the same flow is then taken to obtain scatter plot data of flow and pressure drop.
[0006] Furthermore, in S2, a cubic polynomial function is selected as the fitting function to fit the scatter plot data of flow rate and pressure drop.
[0007] Furthermore, in S3, the residuals are fitted using a cosine function to obtain the optimized residual function: ; In the formula, Represents the residual; Indicates coefficient; Indicates flow rate; , , These represent the coefficients of the function terms.
[0008] Furthermore, the expression for the power function is: ; In the formula, Indicates pressure drop; Indicates the flow coefficient; Indicates the pressure drop index; The expression for the optimized fitting function is: .
[0009] Furthermore, in S4, during the optimization fitting process... , This remains unchanged, thus simplifying the optimization fitting process, including: Take the average value of the obtained flow rate and pressure drop scatter plot data; The obtained average flow rate and pressure drop data are fitted using a power function to obtain... and ; Will get and Substitute the optimized fitting function into the data, and then refit the obtained flow rate and pressure drop scatter data using the optimized fitting function. Verify the fitting effect of the optimized fitting function.
[0010] The beneficial effects of this invention are as follows: This invention preprocesses the original flow and pressure drop data by means of mean filtering, which reduces the influence of noise signals on the data fitting effect. Furthermore, it obtains a cosine function based on the curve of the cubic polynomial fitting function to simplify the residual function. This not only simplifies the fitting function and facilitates calculation, but also significantly improves the fitting effect between the function curve and the actual flow and pressure drop curve, making it easier to predict the operating status of the flame arrester in actual operation. Attached Figure Description
[0011] Figure 1 The diagram shown is a flowchart of the present invention.
[0012] Figure 2 The figure shown is a graph of fitting results of different functions to experimental data of a pair of DN25 flame arresters in an embodiment of the present invention.
[0013] Figure 3 The image shows a comparison of the fitting results of cubic polynomial functions and power functions.
[0014] Figure 4 The figure shown is a graph of the residual function.
[0015] Figure 5 The figure shown is a comparison of the fitting effects of the residual function and the optimized residual function.
[0016] Figure 6 The figure shows the fitting effect of the optimized fitting function on the scatter data of flow rate and pressure drop.
[0017] Figure 7 The image shows the fitting effect after further optimization of the optimized fitting function.
[0018] Figure 8 The figure shown is a graph of the fitting results of different functions on the experimental data of the DN32 flame arrester in Embodiment 2 of the present invention.
[0019] Figure 9 The image shows a comparison of the fitting results of cubic polynomial functions and power functions.
[0020] Figure 10 The figure shown is a graph of the residual function.
[0021] Figure 11 The figure shown is a comparison of the fitting effects of the residual function and the optimized residual function.
[0022] Figure 12 The figure shows the fitting effect of the optimized fitting function on the scatter data of flow rate and pressure drop.
[0023] Figure 13 The image shown is a graph illustrating the fitting effect after further optimization of the optimized fitting function. Detailed Implementation
[0024] An optimized fitting method for the flow-pressure drop curve of a flame arrester includes the following steps: S1, performing flow and pressure drop tests on the flame arrester, and obtaining flow-pressure drop scatter data through mean filtering; S2, performing curve fitting on the obtained flow-pressure drop scatter data to obtain a fitting function, and calculating the residual function between the fitting function and the power function; S3, plotting the residual function, selecting an appropriate optimized residual function based on the graph to refit the residual, and obtaining an optimized fitting function by combining the optimized residual function and the power function; S4, fitting the obtained flow-pressure drop scatter data using the optimized fitting function, testing the fitting effect, and comparing the fitting effect of the optimized fitting function with the fitting effect of the power function to verify the fitting effect of the optimized fitting function.
[0025] Furthermore, in S1, multiple flow and pressure drop tests are performed on the same flame arrester within the same flow variation range to obtain at least five sets of experimental data. The average value of the at least five sets of pressure drop data corresponding to the same flow is then taken to obtain scatter plot data of flow and pressure drop.
[0026] Furthermore, in S2, a cubic polynomial function is selected as the fitting function to fit the scatter plot data of flow rate and pressure drop.
[0027] Furthermore, in S3, the residuals are fitted using a cosine function to obtain the optimized residual function: ; In the formula, Represents the residual; Indicates coefficient; Indicates flow rate; , , These represent the coefficients of the function terms.
[0028] Furthermore, the expression for the power function is: ; In the formula, Indicates pressure drop; Indicates the flow coefficient; Indicates the pressure drop index; The expression for the optimized fitting function is: .
[0029] Furthermore, in the process of optimizing the fit , This remains unchanged, thus simplifying the optimization fitting process, including: Take the average value of the obtained flow rate and pressure drop scatter plot data; The obtained average flow rate and pressure drop data are fitted using a power function to obtain... and ; Will get and Substitute the optimized fitting function into the data, and then refit the obtained flow rate and pressure drop scatter data using the optimized fitting function. Verify the fitting effect of the optimized fitting function.
[0030] This invention discloses an optimized fitting method for the flow rate and pressure drop curve of a flame arrester. The following describes one embodiment of this invention in detail with reference to the accompanying drawings.
[0031] like Figure 1 As shown, an optimization fitting method for the flow rate and pressure drop curve of a flame arrester specifically includes the following steps: The first step is to conduct multiple flow and pressure drop tests on the same flame arrester within the same flow variation range to obtain at least five sets of experimental data. Then, the average value of the at least five sets of pressure drop data corresponding to the same flow rate is taken to obtain a set of average filtered flow and pressure drop scatter data.
[0032] The second step is to perform curve fitting on the obtained flow rate and pressure drop scatter data. A cubic polynomial is selected as the fitting function to fit the flow rate and pressure drop scatter data, and the residual function between the fitting function and the power function is calculated.
[0033] The third step is to apply the obtained residual function to the flow rate. The curves within the range of variation are plotted to visually represent the changes in the residuals. Then, based on the curve graph, a cosine function is selected as the optimization residual function to refit the residuals.
[0034] First, the residuals are refitted using a cosine function to obtain an optimized residual function. The expression for the optimized residual function is: ; In the formula, Represents the residual; Indicates coefficient; Indicates flow rate; , , These represent the coefficients of the function terms.
[0035] The expression for the power function is: ; In the formula, P represents pressure drop; k represents flow coefficient; and n represents pressure drop exponent. Then, the optimized fitting function is obtained by optimizing the residual function and the power function. The expression of the optimized fitting function is: .
[0036] The fourth step is to use an optimized fitting function to fit the obtained flow rate and pressure drop scatter data, test the fitting effect, and compare the fitting effect of the optimized fitting function with that of the power function to verify the fitting effect of the optimized fitting function.
[0037] Fifth, if the fitting effect of the optimized fitting function is worse than that of the power function, or if the fitting effect of the optimized fitting function needs to be further improved, the optimized fitting function can be used as a basis to refine the fitting function. , , , as well as Further optimization can be performed to improve the fitting effect.
[0038] Step 6: The optimized fitting function has a good fitting effect on the flow rate and pressure drop curve of the flame arrester, and during the optimization fitting process... , Since this remains unchanged, the optimization fitting process can be simplified, allowing for further prioritization, including: first, averaging the obtained flow-pressure drop scatter data; second, fitting the obtained average flow-pressure drop data using a power function to obtain... and ; Again, the obtained and The optimized fitting function was then used to refit the obtained flow rate and pressure drop scatter data; finally, the fitting effect of the optimized fitting function was verified.
[0039] The following describes Embodiment 1 of the present invention in detail with reference to the DN25 flame arrester: 1. Mean filtering: Perform multiple flow and pressure drop tests on the same DN25 flame arrester within the same flow variation range to obtain at least five sets of experimental data. Take the average value of the at least five sets of pressure drop data corresponding to the same flow rate to obtain a set of mean-filtered flow and pressure drop scatter data, as shown in Table 1.
[0040] Table 1. Flow rate and pressure drop scatter plot data for DN25 flame arresters .
[0041] 2. Curve Fitting and Finding the Optimal Solution: For the scatter plot data of flow rate and pressure drop of the above DN25 flame arrester, curve fitting with different functions is performed to obtain... Figure 2 The results shown are from Figure 2 As can be seen, the cubic polynomial function provides the best fit. The cubic polynomial function, substituted with the above data, is as follows: ; Substituting the above data into the power function, the result is as follows: ; By plotting the expressions for the cubic polynomial function and the power function using the same data, and displaying them on the same graph, we can obtain the following: Figure 3 The results are shown. From Figure 3 The goodness of fit of the cubic polynomial function shown Goodness of fit of power functions This shows that the fitting effect of the cubic polynomial function is greater than that of the power function.
[0042] 3. Generate residual functions and curves: Based on the above analysis, generate residuals. The expression is as follows: ; The graph of the residual function is as follows Figure 4 As shown, Figure 4 The curve in the graph is approximated by a cosine function.
[0043] 4. Cosine function fitting residuals: Since the residual function obtained from the cubic polynomial is quite complex, and combined with... Figure 4 As shown in the function curve, the residuals can be refitted based on the cosine function to obtain the optimized residual function as follows: ; The optimized residual function will be plotted as a curve, and compared with the curve of the residual function. Figure 5 As shown.
[0044] 5. Optimized flow rate and pressure drop curve: The power function is optimized using the optimized residual function to obtain the optimized fitting function. The optimized fitting function is expressed as follows: .
[0045] 6. Verify the fitting results: Fit the flow rate and pressure drop scatter data using the optimized fitting function, and obtain the following results: Figure 6 The fitting result is shown. From Figure 6 The goodness of fit of the optimized fitting function shown The goodness of fit is derived from the power function expression above. Upgraded to This indicates that the residual optimization effect is good.
[0046] 7. Optimize the fitting function depth: Optimize the parameters and constants of the residual part in the optimized fitting function to obtain the fitting curve as shown below. Figure 7 As shown, the expression for the optimized fitting function after depth optimization is: ; from Figure 7 As can be seen from the data, the goodness of fit before and after deep optimization is... No significant changes were observed, but the root mean square error (RMSE) further decreased, indicating a significant improvement in the fitting effect.
[0047] The following describes Embodiment 2 of the present invention in detail with reference to the DN32 flame arrester: 1. Mean filtering: Perform multiple flow and pressure drop tests on the same DN32 flame arrester within the same flow variation range to obtain at least five sets of experimental data. Take the average value of the at least five sets of pressure drop data corresponding to the same flow rate to obtain a set of mean-filtered flow and pressure drop scatter data, as shown in Table 1.
[0048] Table 2. Flow rate and pressure drop scatter plot data for DN32 flame arresters .
[0049] 2. Curve Fitting and Finding the Optimal Solution: For the scatter plot data of flow rate and pressure drop of the above DN32 flame arrester, curve fitting with different functions is performed to obtain... Figure 8 The results shown are from Figure 8 As can be seen, the cubic polynomial function provides the best fit. The cubic polynomial function, substituted with the above data, is as follows: ; Substituting the above data into the power function, the result is as follows: ; By plotting the expressions for the cubic polynomial function and the power function using the same data, and displaying them on the same graph, we can obtain the following: Figure 9 The results are shown. From Figure 9 The goodness of fit of the cubic polynomial function shown Goodness of fit of power functions This shows that the fitting effect of the cubic polynomial function is greater than that of the power function.
[0050] 3. Generate residual functions and curves: Based on the above analysis, generate residuals. The expression is as follows: ; The graph of the residual function is as follows Figure 10 As shown, Figure 10 The curve in the graph is approximated by a cosine function.
[0051] 4. Cosine function fitting residuals: The residual function described above is quite complex and combines... Figure 10 As shown in the function curve, the residuals can be refitted based on the cosine function to obtain the optimized residual function as follows: ; The optimized residual function will be plotted as a curve, and compared with the curve of the residual function. Figure 11 As shown.
[0052] 5. Optimized flow rate and pressure drop curve: The power function is optimized using the optimized residual function to obtain the optimized fitting function. The optimized fitting function is expressed as follows: .
[0053] 6. Verify the fitting results: Fit the flow rate and pressure drop scatter data using the optimized fitting function, and obtain the following results: Figure 12 The fitting result is shown. From Figure 12 The goodness of fit of the optimized fitting function shown The goodness of fit improved from 0.9993 for the power function expression to 0.9994, indicating that the residual optimization effect was good.
[0054] 7. Optimize the fitting function depth: Optimize the parameters and constants of the residual part in the optimized fitting function to obtain the fitting curve as shown below. Figure 13 As shown, the expression for the optimized fitting function after depth optimization is: ; from Figure 13 As can be seen from the data, the goodness of fit before and after deep optimization is... No significant changes were observed, but the root mean square error (RMSE) further decreased, indicating a significant improvement in the fitting effect.
[0055] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.
Claims
1. An optimized fitting method for the flow-pressure drop curve of a flame arrester, characterized in that, Includes the following steps: S1. Perform flow and pressure drop tests on the flame arrester, and obtain scatter plot data of flow and pressure drop through mean filtering; S2. Perform curve fitting on the obtained flow rate and pressure drop scatter data to obtain the fitting function, and calculate the residual function between the fitting function and the power function. S3. Plot the residual function, select an appropriate optimization residual function based on the graph to refit the residual, and obtain the optimal fitting function by optimizing the residual function and the power function. S4. Use the optimized fitting function to fit the obtained flow rate and pressure drop scatter data, test the fitting effect, and compare the fitting effect of the optimized fitting function with the fitting effect of the power function to verify the fitting effect of the optimized fitting function.
2. The method for optimizing and fitting the flow-pressure drop curve of a flame arrester according to claim 1, characterized in that: In S1, multiple flow and pressure drop tests are performed on the same flame arrester within the same flow variation range to obtain at least five sets of experimental data. The average value of the at least five sets of pressure drop data corresponding to the same flow is then taken to obtain scatter plot data of flow and pressure drop.
3. The method for optimizing and fitting the flow rate and pressure drop curve of a flame arrester according to claim 2, characterized in that: In S2, a cubic polynomial function is selected as the fitting function to fit the scatter plot data of flow rate and pressure drop.
4. The method for optimizing and fitting the flow-pressure drop curve of a flame arrester according to claim 3, characterized in that: In S3, the residuals are fitted using a cosine function to obtain the optimized residual function: ； In the formula, Represents the residual; Indicates coefficient; Indicates flow rate; , , These represent the coefficients of the function terms.
5. The method for optimizing and fitting the flow-pressure drop curve of a flame arrester according to claim 4, characterized in that: The expression for the power function is: ； In the formula, Indicates pressure drop; Indicates the flow coefficient; Indicates the pressure drop index; The expression for the optimized fitting function is: 。 6. The optimized fitting method for the flow-pressure drop curve of a flame arrester according to claim 5, characterized in that: In S4, during the optimization fitting process , This remains unchanged, thus simplifying the optimization fitting process, including: Take the average value of the obtained flow rate and pressure drop scatter plot data; The obtained average flow rate and pressure drop data are fitted using a power function to obtain... and ; Will get and Substitute the optimized fitting function into the data, and then refit the obtained flow rate and pressure drop scatter data using the optimized fitting function. Verify the fitting effect of the optimized fitting function.