Aircraft noise ground monitoring data optimization method for linear flight segment

Abstract

The invention provides an aircraft noise ground monitoring data optimization method for a linear flight segment, relates to the technical field of environment monitoring and noise control, and solves the problem that the noise evaluation accuracy is limited due to the fact that the influence of the flight state of an aircraft on noise propagation is not considered in an existing method. The method comprises the following steps: acquiring an aircraft track and ground key coordinates, and converting the aircraft track and the ground key coordinates into a three-dimensional rectangular coordinate system after Gaussian coordinate system conversion; determining an aircraft noise event, matching track data corresponding to the event with noise data, dynamically adjusting a confidence interval, and correcting the noise data to a confidence boundary; polynomial fitting is carried out, the ratio of the sound level slope to the flight time is used as a constraint, the minimum global error quadratic sum of noise data fitting is used as an optimization objective function, and a model is constructed; and solving model parameters by adopting a genetic algorithm to finish an optimization process. According to the method, the optimization processing mode of the aircraft noise sound level data is perfected, and the influence of the aircraft noise around the airport can be accurately judged.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring and noise control, and in particular to a method for optimizing ground monitoring data of aircraft noise in a straight flight segment. Background Art

[0002] While the growth of air transport is driving economic development and population concentration around airports, aircraft noise has also become increasingly prominent. Countries are increasingly focusing on noise pollution around airports, using ground-based monitoring as a basis for noise prediction and control. As a core method for assessing the impact of aircraft noise, the accuracy of noise prediction is directly linked to the scientific nature of tasks such as aircraft noise control and surrounding land planning. All aircraft noise prediction models require calibration and verification using ground-based monitoring data to adapt to local meteorological conditions and enhance model accuracy.

[0003] For airport aircraft noise monitoring, environmental impact assessments, and system development, ground measurement locations are primarily selected based on the locations of sensitive ground points, including schools, hospitals, and residences. These sensitive points are often near human and vehicle activities. Furthermore, wind field fluctuations significantly impact aircraft noise levels, causing ground-monitored aircraft noise levels to fluctuate. Directly calculating key evaluation indicators such as effective perceived sound level, exposure level, and equivalent sound level can result in significant deviations in the final prediction results. Therefore, it is necessary to adopt scientific and rational methods to optimize and process ground-based noise data, reducing noise errors at the source and laying an excellent foundation for more scientific aircraft noise prediction and assessment.

[0004] Currently, aircraft ground-based noise data is typically processed using single-monitoring-point data modeling and localized multi-monitoring-point comparative analysis. Single-monitoring-point data modeling typically suppresses ambient noise interference through static background noise modeling, frequency-domain filtering, and statistical regression analysis. However, this approach struggles with dynamically changing non-target noise, such as gusts of wind and sudden traffic noise. Localized multi-monitoring-point comparative analysis can reduce local interference through cross-comparison, but typically only a small number of monitoring points are used for comparison and calibration, resulting in limited improvement in overall noise prediction accuracy. Furthermore, when experiencing varying interferences simultaneously, it can be difficult to determine which monitoring point to use as a benchmark.

[0005] Existing methods also generally ignore the direct impact of aircraft flight conditions, such as distance, altitude, and speed, on noise propagation, and fail to incorporate the dynamic position of aircraft sound sources and the changing patterns of sound levels into their calculations. The isolated processing of noise data results in a significant amount of residual ambient noise interference, impacting the accuracy of subsequent noise prediction and assessment. Therefore, a more appropriate method for optimizing the processing of aircraft noise data is crucial. Summary of the Invention

[0006] Based on the current state of the art, the present invention aims to address the problem that existing methods for processing aircraft ground noise monitoring data fail to consider the impact of the aircraft's flight status on noise propagation, resulting in limited noise assessment accuracy. Therefore, a method for optimizing aircraft noise ground monitoring data for straight-line flight segments is proposed. By constraining the slope of the sound level curve at the moment the aircraft is closest to the monitoring point and the ratio of flight time before and after a certain decibel threshold for maximum sound level, the present invention improves the optimization processing of aircraft noise level data, thereby facilitating accurate assessment of the impact of aircraft noise around airports.

[0007] The present invention adopts the following technical solutions to achieve the purpose:

[0008] A method for optimizing ground monitoring data of aircraft noise in a straight flight segment comprises the following steps:

[0009] S1. Obtain the aircraft track and key ground coordinates, convert them into a Gaussian coordinate system, and then convert the Gaussian coordinate system data into a three-dimensional rectangular coordinate system constructed with the airport reference point as the origin;

[0010] S2. Determine an aircraft noise event based on the converted data in the three-dimensional rectangular coordinate system, and match the flight path data corresponding to the event with the noise data;

[0011] S3. After the matching is completed, the noise data is preprocessed by dynamically adjusting the confidence interval to correct the noise data to the confidence boundary;

[0012] S4. After the preprocessing is completed, a preset proportion of noise data is selected for polynomial fitting, with the ratio of sound level slope to flight time as a constraint and the minimum sum of squared errors of the noise data fitting as the optimization objective function, to construct an aircraft noise data processing model;

[0013] S5. Use genetic algorithm to solve the parameters of the aircraft noise data processing model, obtain the optimal objective function, output the polynomial fitting parameters and the final fitting noise data, and complete the data optimization.

[0014] Specifically, in step S1, the aircraft track is obtained by parsing the aircraft ADS-B data or radar track data, and includes the aircraft's latitude and longitude coordinates, altitude, speed, heading and time; the ground key coordinates include the latitude and longitude coordinates of the ground monitoring point and the runway center coordinates; the aircraft's latitude and longitude coordinates, the ground monitoring point's latitude and longitude coordinates and the runway center coordinates are first converted to a Gauss-Krüger projection coordinate system, and then converted to a three-dimensional rectangular coordinate system.

[0015] Specifically, the runway center coordinate point is used as the airport reference point, that is, the origin of the three-dimensional rectangular coordinate system, the direction parallel to the runway on the horizontal plane is used as the X-axis direction, the direction perpendicular to the runway is used as the Y-axis direction, and the height direction perpendicular to the horizontal plane where the X-axis and Y-axis are located is used as the Z-axis direction to construct a three-dimensional rectangular coordinate system; the data in the Gauss-Krüger projection coordinate system is converted into the three-dimensional rectangular coordinate system.

[0016] Preferably, in step S2, aircraft noise data corresponding to multiple ground monitoring points of ground key coordinates are obtained from a three-dimensional rectangular coordinate system, the aircraft noise data are traversed and sorted in a preset order, and candidate event data exceeding a preset noise event threshold are identified therefrom, and then the effective noise events of the aircraft are determined based on preset screening conditions.

[0017] Preferably, after identifying candidate event data based on a preset noise event threshold, events whose duration is within a preset interval and whose maximum noise value is higher than a limit value based on the preset noise event threshold are screened as effective noise events of the aircraft; the aircraft track corresponding to the effective noise event is matched with the time point in the noise data through cubic spline interpolation, and the aircraft track and noise data are stored correspondingly in chronological order.

[0018] Preferably, in step S3, for aircraft noise events, the noise data therein are preliminarily fitted by multiple polynomials, the residuals are calculated, and the confidence level is dynamically selected based on their mean and standard deviation; wherein, if the absolute mean plus the standard deviation exceeds a preset decibel threshold, the first confidence level is adopted, otherwise the second confidence level is adopted; based on the selected confidence level, the identified residual points that exceed the confidence interval are regarded as outliers, and the outliers are modified to the confidence boundary.

[0019] Furthermore, in step S4, the noise data is recorded as D = [t i ,L i ]2×N,t i is the i-th time point, L i is the noise value at the i-th time point, N is the length of the noise data; when the noise data of the preset ratio is selected, the selected point is recorded as PN×1, where P is a 0 or 1 vector, that is, when the data is selected, it is 1, otherwise it is 0, thus forming a new fitting noise data D new =D(P); Based on this recording method, a polynomial model is constructed and fitted using a k-order polynomial, as shown below:

[0020]

[0021] In the formula, θ = {θ0, θ1, θ2,…,θ k}, θ is the coefficient to be solved, and m is the degree of the polynomial; the coefficient is solved by minimizing the sum of squared errors, which is the following matrix form:

[0022] θ=(W T W) -1 W T L

[0023]

[0024] Where, (…) T represents the transpose of the matrix; L = L i , which represents the noise value of the corresponding noise data.

[0025] Specifically, in step S5, when the genetic algorithm is used to solve the parameters of the aircraft noise data processing model, a random vector chromosome s from 0 to 1 is generated. i (i=1,2,…,n),s i The value greater than 0.5 is 1, otherwise it is 0; the vector length is the same as the length of the noise data, and it is judged whether the preset ratio requirement is met, and the random vector chromosomes that meet the requirements are formed into the initial population S0; then, based on the constraints of step S4 and the optimization objective function, the objective function value of each individual is calculated, and the individuals with objective function values less than the optimization threshold are selected as parents, and crossover and mutation operations are performed according to the preset probability to generate new individuals and populations, and this is iterated until the termination condition is reached; when the optimal objective function value cannot be found, the data selection ratio requirement is reduced, and the above process is repeated, and finally the optimal solution that meets the constraints is output to obtain the optimal objective function.

[0026] In summary, due to the adoption of this technical solution, the beneficial effects of the present invention are as follows:

[0027] The proposed method fully considers the actual flight status of an aircraft during a straight-line flight segment, including factors such as its relative distance from the monitoring point, altitude, and speed. By simultaneously collecting aircraft track and noise data and establishing a three-dimensional rectangular coordinate system with the airport reference point as the coordinate origin, it achieves dynamic modeling of the sound source location. This method effectively improves the accuracy and stability of ground monitoring data in complex environmental noise interference, reducing the impact of errors at the source of the data.

[0028] This invention further optimizes the noise data identification and processing logic by introducing a constraint on the slope of the sound level curve and a mechanism that determines the ratio of flight time before and after a certain decibel threshold for maximum sound level. This allows for more accurate extraction of aircraft noise events and avoids the assessment bias caused by traditional methods that ignore flight state changes. This not only improves the calibration accuracy of noise prediction models but also provides more scientific data foundation for land planning and noise control around airports.

[0029] The method of the present invention also has strong adaptability and scalability, can be compatible and integrated with existing airport noise monitoring systems, and improve the intelligence level of the overall system. It is of great significance to promote the development of aircraft noise prediction and assessment technology, and has good prospects for promotion and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram briefly describing the overall process of the method of the present invention;

[0031] Figure 2 Schematic diagram of Gaussian transformation of ground key points and part of the aircraft track coordinates in the present invention;

[0032] Figure 3 This is an example diagram of data of effective noise events after screening and storage in the present invention;

[0033] Figure 4 This is an example diagram of the noise data confidence correction in the present invention;

[0034] Figure 5 Schematic diagram of the relationship between ground monitoring points and aircraft positions in the present invention;

[0035] Figure 6 This is an example diagram comparing the aircraft noise fitting results and polynomial fitting in the present invention. DETAILED DESCRIPTION

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of 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. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0037] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0038] Example

[0039] A method for optimizing ground monitoring data of aircraft noise in straight flight segments. Figure 1 The overall process of the method is briefly described in the following figure, which can be viewed simultaneously. The key steps of the method can be summarized as follows:

[0040] S1. Obtain the aircraft track and key ground coordinates, convert them into a Gaussian coordinate system, and then convert the Gaussian coordinate system data into a three-dimensional rectangular coordinate system constructed with the airport reference point as the origin;

[0041] S2. Determine an aircraft noise event based on the converted data in the three-dimensional rectangular coordinate system, and match the flight path data corresponding to the event with the noise data;

[0042] S3. After the matching is completed, the noise data is preprocessed by dynamically adjusting the confidence interval to correct the noise data to the confidence boundary;

[0043] S4. After the preprocessing is completed, a preset proportion of noise data is selected for polynomial fitting, with the ratio of sound level slope to flight time as a constraint and the minimum sum of squared errors of the noise data fitting as the optimization objective function, to construct an aircraft noise data processing model;

[0044] S5. Use genetic algorithm to solve the parameters of the aircraft noise data processing model, obtain the optimal objective function, output the polynomial fitting parameters and the final fitting noise data, and complete the data optimization.

[0045] This embodiment will introduce the details of each step in detail according to the above step sequence, as well as some examples that can be optimized.

[0046] In step S1, the aircraft track is obtained by parsing the aircraft ADS-B data or radar track data, and includes the aircraft's latitude and longitude coordinates, altitude, speed, heading, and time; the ground key coordinates include the latitude and longitude coordinates of the ground monitoring point and the runway center coordinates; the aircraft's latitude and longitude coordinates, the ground monitoring point's latitude and longitude coordinates, and the runway center coordinates are first converted to a Gauss-Krüger projection coordinate system, and then converted to a three-dimensional rectangular coordinate system.

[0047] Figure 2 The figure shows the effect of aircraft track data, longitude and latitude of ground monitoring points and airport runway after conversion to the Gauss-Krüger projection coordinate system, which can be viewed simultaneously. Based on this Gaussian coordinate system, the runway center coordinates can be used as the origin to convert to the corresponding three-dimensional rectangular coordinate system.

[0048] In this embodiment, the runway center coordinate point is used as the airport reference point, that is, the origin of the three-dimensional rectangular coordinate system, the direction parallel to the runway on the horizontal plane is used as the X-axis direction, the direction perpendicular to the runway is used as the Y-axis direction, and the height direction perpendicular to the horizontal plane where the X-axis and Y-axis are located is used as the Z-axis direction to construct a three-dimensional rectangular coordinate system; the data in the Gauss-Krüger projection coordinate system is converted into the three-dimensional rectangular coordinate system.

[0049] In step S2, aircraft noise data corresponding to multiple ground monitoring points at key ground coordinates are obtained from a three-dimensional rectangular coordinate system. After traversing the aircraft noise data and sorting them in a preset order, 90 decibels is taken as the preset noise event threshold, and candidate event data exceeding this threshold is identified. Then, based on the preset screening conditions, valid aircraft noise events are determined.

[0050] The preset screening conditions in this embodiment are: events with a duration of 10 to 60 seconds and a maximum noise value 10 decibels higher than 90 decibels are selected as effective aircraft noise events; aircraft tracks corresponding to effective noise events are matched with time points in noise data through cubic spline interpolation, and aircraft tracks and noise data are stored in chronological order. The relevant data of effective noise events after screening and storage can be found in Figure 3 's hint.

[0051] In step S3, for aircraft noise events, the noise data is preliminarily fitted by multiple polynomials, the residuals are calculated, and the confidence level is dynamically selected based on the mean and standard deviation. If the absolute mean plus the standard deviation exceeds 3 decibels, a 95% confidence level is used, otherwise a 90% confidence level is used. Based on the selected confidence level, the residual points that are identified as exceeding the confidence interval are regarded as outliers, and the outliers are modified to the confidence boundary. The example of the noise data after correction in this embodiment is as follows: Figure 4 shown.

[0052] In step S4, the noise data is recorded as D = [t i ,L i ]2×N,t i is the i-th time point, L i is the noise value at the i-th time point, N is the length of the noise data; initially select noise data with a ratio greater than 50%, and record the selected points as PN×1, where P is a 0 or 1 vector, that is, when the data is selected, it is 1, otherwise it is 0, thus forming a new fitting noise data D new =D(P); Based on this recording method, a polynomial model is constructed and fitted using a k-order polynomial, as shown below:

[0053]

[0054] In the formula, θ = {θ0, θ1, θ2,…,θ k}, θ is the coefficient to be solved, and m is the degree of the polynomial; the coefficient is solved by minimizing the sum of squared errors, which is the following matrix form:

[0055] θ=(W T W) -1 W T L

[0056]

[0057] Where, (…) T represents the transpose of the matrix; L = L i , which represents the noise value of the corresponding noise data.

[0058] The polynomial parameters obtained above are only for the selected data D new In order to ensure that it does not deviate too much from the noise data D and cause the fitting data distortion, the noise curve D after fitting is calculated using polynomial parameters when fitting in this embodiment. fit , and find the sum of square errors between it and the noise data D, and take the minimum sum of square errors as the objective function, that is:

[0059] Objective function = minimize sum(D fit -D) 2

[0060] At the same time, the objective function is made to satisfy the preset constraints, as follows:

[0061]

[0062] Where x m D fit The moment when the maximum value is obtained, x i 、x j D fit The time interval value [x i ,x j ], t d 、R min is the time and distance when the aircraft is closest to the ground monitoring point (R min See Figure 5 ), α and β represent the corresponding preset constraint thresholds; Formula t d +R min / 340 represents the moment when the aircraft noise is transmitted to the ground monitoring point.

[0063] In this embodiment, the preset constraint threshold β represents the noise curve D when it is close to the maximum value of the noise. fit The slope limit of β is used to characterize the situation where the aircraft noise level changes from rising to falling, and its theoretical value is close to 0. It can be taken as 0.1 according to the actual noise data and the collection environment conditions.

[0064] In this embodiment, see Figure 5The positional relationship between the ground monitoring points and the aircraft is shown in the figure. The preset constraint threshold α is α=max(1,r1 / r2), where r1 and r2 are the distances from the ground monitoring points p1 and p2 in front and behind the aircraft parallel to the track to the ground monitoring point p0 perpendicular to the track. When the sound levels of the aircraft noise data received by the ground monitoring points P1 and P2 are equal, assuming that the aircraft flight speed is approximately unchanged, then (x m -x i ) / (x j -x m )≈r1 / r2, let r1 be different values, calculate the corresponding r2 and r1 / r2 ratio, take the minimum value of r1 / r2 as the constraint threshold, so as to constrain the time ratio of the left and right sides of the noise data; this embodiment mainly selects the threshold points of 3 / 6 / 10 dB below the peak as the constraint points.

[0065] Furthermore, r1 and r2 in this embodiment satisfy the following relationship:

[0066]

[0067] Where M = v / 340, v is the aircraft speed; sqrt(…) is the square root function; and η is the atmospheric absorption attenuation coefficient. The first two terms represent the difference in sound levels in front and behind due to Doppler efficiency, the third term represents the noise attenuation with distance, and the fourth term represents the attenuation due to atmospheric absorption. The atmospheric absorption attenuation coefficient η typically ranges from 0.001 to 0.05. However, smaller values of η make the above equation difficult to satisfy. Therefore, this embodiment recommends a larger value of 0.05 to provide a lower constraint.

[0068] In step S5, when the genetic algorithm is used to solve the parameters of the aircraft noise data processing model, a random vector chromosome s from 0 to 1 is generated. i (i=1,2,…,n),s i The value of the vector greater than 0.5 is 1, otherwise it is 0; the length of the vector is the same as the length of the noise data, and it is determined whether the preset ratio requirement is met, and the random vector chromosomes that meet the requirement are formed into the initial population S0; then, based on the constraints of step S4 and the optimization objective function, a 6th-order polynomial fit is performed on the selected data to calculate the global objective function value; it is determined whether the slope of the curve at the moment when the aircraft is closest to the monitoring point and the ratio of the time to the left and right sides of the 3 / 6 / 10 decibel threshold points below the maximum value of the fitted sound level curve are met; in this embodiment, additional points can be added as needed. According to the requirements of the relevant aircraft noise regulations, generally, only data within 10 decibels below the maximum sound level need to be considered.

[0069] When the judgment result is not satisfied, add a certain penalty value to the objective function value to reduce its inheritance probability and output the objective function value under the polynomial fitting parameters; select individuals with smaller objective function values as parents, and perform crossover and mutation according to a certain probability to generate new individuals and populations, and iterate until the termination condition, repeat the above steps, and cycle multiple times to find the optimal one; when the optimal objective function value cannot meet the requirements, reduce the data selection ratio, for example, reduce it by 10% each time, and repeat the above process.

[0070] Finally, output the optimal solution that meets the conditions; see Figure 6 The example data shown, the final fitting parameters of the 6th order polynomial for a noise event, from high to low order, are: [-7.10×10 -8 ,1.21×10 -5 ,-7.54×10 -4 ,0.02,-0.26,2.05,44.72], the optimal curve after fitting is Figure 6 Shown in; from Figure 6 As can be seen, when conventional general polynomial fitting is used, when the ambient noise lasts for a long time and fluctuates greatly, the fitted noise level will be more biased towards the ambient noise, resulting in a larger deviation in the subsequent aircraft noise index calculation. The optimal curve obtained by the method of this embodiment avoids this deviation problem, thereby achieving data optimization.

Claims

1. A method for optimizing ground monitoring data of aircraft noise in a straight flight segment, characterized in that: The steps include: S1. Obtain the aircraft track and key ground coordinates, convert them into a Gaussian coordinate system, and then convert the Gaussian coordinate system data into a three-dimensional rectangular coordinate system constructed with the airport reference point as the origin; S2. Determine an aircraft noise event based on the converted data in the three-dimensional rectangular coordinate system, and match the flight path data corresponding to the event with the noise data; S3. After the matching is completed, the noise data is preprocessed by dynamically adjusting the confidence interval to correct the noise data to the confidence boundary; S4. After the preprocessing is completed, a preset proportion of noise data is selected for polynomial fitting, with the ratio of sound level slope to flight time as a constraint and the minimum sum of squared errors of the noise data fitting as the optimization objective function, to construct an aircraft noise data processing model; S5. Use genetic algorithm to solve the parameters of the aircraft noise data processing model, obtain the optimal objective function, output the polynomial fitting parameters and the final fitting noise data, and complete the data optimization.

2. The method for optimizing aircraft noise ground monitoring data according to claim 1, characterized in that: In step S1, the aircraft track is obtained by parsing the aircraft ADS-B data or radar track data, and includes the aircraft's latitude and longitude coordinates, altitude, speed, heading, and time; the ground key coordinates include the latitude and longitude coordinates and altitude of the ground monitoring point and the runway center; the aircraft's latitude and longitude coordinates, the ground monitoring point's latitude and longitude coordinates, and the runway center coordinates are first converted to a Gauss-Krüger projection coordinate system, and then converted to a three-dimensional rectangular coordinate system.

3. The method for optimizing aircraft noise ground monitoring data according to claim 2, characterized in that: The runway center coordinate point is used as the airport reference point, that is, the origin of the three-dimensional rectangular coordinate system. The direction parallel to the runway on the horizontal plane is used as the X-axis direction, the direction perpendicular to the runway is used as the Y-axis direction, and the height direction perpendicular to the horizontal plane where the X-axis and Y-axis are located is used as the Z-axis direction to construct a three-dimensional rectangular coordinate system; the data in the Gauss-Krüger projection coordinate system is converted into the three-dimensional rectangular coordinate system.

4. The method for optimizing aircraft noise ground monitoring data according to claim 1, characterized in that: In step S2, aircraft noise data corresponding to multiple ground monitoring points at key ground coordinates are obtained from a three-dimensional rectangular coordinate system. After traversing the aircraft noise data and sorting them in a preset order, candidate event data that exceeds a preset noise event threshold is identified, and then effective noise events of the aircraft are determined based on preset screening conditions.

5. The method for optimizing aircraft noise ground monitoring data according to claim 4, characterized in that: After identifying candidate event data based on the preset noise event threshold, screen the events whose duration is within the preset interval and whose maximum noise value exceeds the limit value based on the preset noise event threshold as valid aircraft noise events; The aircraft tracks corresponding to the effective noise events are matched with the time points in the noise data through cubic spline interpolation, and the aircraft tracks and noise data are stored in chronological order.

6. The method for optimizing aircraft noise ground monitoring data according to claim 1, characterized in that: In step S3, for aircraft noise events, the noise data are preliminarily fitted by multiple polynomials, the residuals are calculated, and the confidence level is dynamically selected based on their mean and standard deviation; if the absolute mean plus the standard deviation exceeds the preset decibel threshold, the first confidence level is adopted, otherwise the second confidence level is adopted; based on the selected confidence level, the identified residual points that exceed the confidence interval are regarded as outliers, and the outliers are modified to the confidence boundary.

7. The method for optimizing aircraft noise ground monitoring data according to claim 1, characterized in that: In step S4, the noise data is recorded as D = [t i ,L i ]2×N,t i is the i-th time point, L i is the noise value at the i-th time point, and N is the length of the noise data; When selecting noise data of a preset ratio, the selected points are recorded as PN×1, where P is a 0 or 1 vector, that is, when the data is selected, it is 1, otherwise it is 0, thus forming a new fitting noise data D new =D(P); Based on this recording method, a polynomial model is constructed and fitted using a k-order polynomial, as shown below: In the formula, θ = {θ0, θ1, θ2,…,θ k }, θ is the coefficient to be solved, and m is the degree of the polynomial; the coefficient is solved by minimizing the sum of squared errors, which is the following matrix form: θ=(W T W) -1 W T L Where, (…) T represents the transpose of the matrix; L = L i , which represents the noise value of the corresponding noise data.

8. The method for optimizing aircraft noise ground monitoring data according to claim 7, characterized in that: When fitting, the noise curve D after fitting is calculated using polynomial parameters. fit , and find the sum of square errors between it and the noise data D, and take the minimum sum of square errors as the objective function, that is: Objective function = minimize sum(D fit -D) 2 At the same time, the objective function is made to satisfy the preset constraints, as follows: Where x m D fit The moment when the maximum value is obtained, x i 、x j D fit The time interval value [x i ,x j ], t d 、R min are the time and distance when the aircraft is closest to the ground monitoring point, and α and β represent the corresponding preset constraint thresholds.

9. The method for optimizing aircraft noise ground monitoring data according to claim 8, characterized in that: The preset constraint threshold β represents the noise curve D when it is close to the maximum value of the noise fit The slope limit is set; the preset constraint threshold α is α=max(1,r1 / r2), where r1 and r2 are the distances from the ground monitoring points P1 and P2 in front and behind the aircraft parallel to the track to the ground monitoring point P0 perpendicular to the track; when the sound levels of the aircraft noise data received by the ground monitoring points P1 and P2 are equal, assuming that the aircraft flight speed is approximately unchanged, then (x m -x i ) / (x j -x m )≈r1 / r2, let r1 be different values, find the corresponding r2 and r1 / r2 ratio, and take the minimum value of r1 / r2 as the constraint threshold to constrain the time ratio of the left and right sides of the noise data; r1 and r2 satisfy the following relationship: Where, M = v / 340, v is the aircraft speed; sqrt(…) is the square root function; η is the atmospheric absorption attenuation coefficient.

10. The method for optimizing aircraft noise ground monitoring data according to claim 1, characterized in that: In step S5, when the genetic algorithm is used to solve the parameters of the aircraft noise data processing model, a random vector chromosome s from 0 to 1 is generated. i (i=1,2,…,n),s i The value of the vector greater than 0.5 is 1, otherwise it is 0; the length of the vector is the same as the length of the noise data, and it is judged whether it meets the preset ratio requirement, and the random vector chromosomes that meet the requirements are formed into the initial population S0; Subsequently, based on the constraints and optimization objective function of step S4, the objective function value of each individual is calculated, and individuals with objective function values less than the optimization threshold are selected as parents. Crossover and mutation operations are performed according to the preset probability to generate new individuals and populations. This is repeated until the termination condition is reached. When the optimal objective function value cannot be found, the data selection ratio requirement is lowered, and the above process is repeated. Finally, the optimal solution that meets the constraints is output to obtain the optimal objective function.

Images