An adaptive wind farm clutter recognition and suppression method for air traffic control primary radar

Abstract

The invention discloses an air traffic control primary radar adaptive wind farm clutter identification and suppression method, comprising: receiving and storing primary radar original echo, primary radar point track data and secondary radar track data; performing azimuth distance gridding in the primary radar detection area, analyzing the quality of primary radar data in the grid, and identifying abnormal areas; utilizing the radar original echo data and radar point track data to perform feature extraction in the abnormal area and identify the wind farm clutter area; selecting a wind farm clutter suppression technology according to the wind farm area characteristics, and adaptively optimizing clutter suppression parameters; quantitatively evaluating the optimization effect, iteratively optimizing strategies and parameters until the radar detection performance index requirements are met. The invention adaptively selects a wind farm clutter suppression means, adjusts the clutter suppression parameters, and evaluates the radar detection performance after clutter suppression, thereby greatly improving the radar detection performance under the wind farm clutter interference.

Description

Technical Field

[0001] The present invention relates to the technical field of air traffic control radars, and in particular to an air traffic control primary radar adaptive wind farm clutter recognition and suppression method. Background Art

[0002] With the development of urbanization, the supply of traditional fossil energy such as coal has become increasingly tight, resulting in increasingly prominent environmental problems. As a pollution-free, low-cost power industry, wind power generation has attracted attention from countries around the world. At the same time, with the development of modern technology, wind power generation technology has gradually matured and has the conditions for large-scale development and commercial development. In recent years, the installed capacity of wind turbines in the world has grown exponentially. However, wind turbines in wind farms are usually composed of blades, nacelles and masts. The height of wind turbines is up to 100 meters, with a large radar cross-sectional area (RCS). At the same time, the rotating blades will produce continuously changing Doppler speeds in the frequency domain, which will cause target shielding and false targets to the radar, resulting in an increase in the probability of false alarms or a decrease in the probability of detection.

[0003] As a low-resolution two-coordinate radar, air traffic control radar has difficulty identifying wind farm clutter through height and distance information. Traditional clutter suppression methods of air traffic control radar, such as pulse compression, AMTD, and automatic constant false alarm, cannot suppress wind farm clutter. Target detection above or near wind farms will be affected by wind farm clutter, resulting in a decrease in radar detection performance, which will seriously affect air traffic safety. The existence of wind farms affects the performance of existing radar stations on the one hand, and brings difficulties to the site selection of new radar stations on the other.

[0004] The traditional wind farm clutter suppression method for air traffic control radar requires manual marking of wind farm locations within the radar detection range. As the number of wind farms increases, the difficulty of marking wind farm locations increases. The degree of impact of wind farms on radar is related to the following conditions:

[0005] 1. The scale and number of wind farms;

[0006] 2. The relative position between the wind farm and the radar;

[0007] 3. Radar parameters.

[0008] Wind farm clutter suppression methods involve the entire radar processing process. Depending on the degree of impact, different processing optimization parameters are involved, and manual settings are complex. Summary of the invention

[0009] The purpose of the present invention is to provide an air traffic control primary radar adaptive wind farm clutter identification and suppression method, by analyzing radar data quality and extracting wind farm clutter characteristic parameters, identifying wind farm clutter areas and impact types, quantifying the impact of wind farms on radar performance, adaptively iterating the optimization parameters of wind farm suppression means to alleviate the impact of wind farm clutter and improve radar detection performance.

[0010] The purpose of the present invention can be achieved through the following technical solutions:

[0011] An air traffic control primary radar adaptive wind farm clutter identification and suppression method comprises the following steps:

[0012] S1: Receive and store primary radar raw echo data, primary radar point data, primary radar track data and secondary radar track data;

[0013] Among them, the coordinate system with the fusion center as the reference point synchronizes the detection data of each sensor to the processing cycle of the fusion processing, completing the spatiotemporal alignment of the sensor's measurement data and the fusion center processing;

[0014] S2: Gridding the azimuth and distance in the primary radar detection area, analyzing the quality of the primary radar data in the grid, and identifying areas with abnormal radar detection performance;

[0015] S2-1, gridding the primary radar detection range according to M distance units and N azimuth sectors, and establishing index values ​​for the grid units;

[0016] S2-2. Perform data quality analysis on the divided grid cells, calculate the type performance index within the grid area, and select the evaluation indicators of detection probability, false alarm rate and track error to identify abnormal areas in view of the impact of wind farm clutter on radar performance;

[0017] S2-3. Perform a comprehensive regional quality score based on the calculated detection probability Pd, false alarm rate Pf and track error D, set the detection probability qualification index Pdth, false alarm rate qualification index Pfth, and track error qualification index Dth. When the detection probability in the grid area is greater than the index Pdth, the false alarm rate is less than the index Pfth, and the track error is less than the index Dth, it is judged as a normal area, and the rest of the areas are judged as abnormal areas.

[0018] S2-4, storage area index, attributes and area performance index parameters;

[0019] S3: For areas with abnormal detection performance, the original echo data of the radar and the radar point track data are used to extract features in the abnormal areas and identify the clutter areas of the wind farm;

[0020] S3-1. If the wind farm is not within the radar line of sight, it will not affect the radar. The radar line of sight is calculated based on the altitude of the geographical location and the size of the wind turbine within the radar coverage area. The abnormal area is marked and stored within the line of sight for the next step of identification. Otherwise, it is considered not to be a clutter area of ​​the wind farm.

[0021] S3-2, perform saturation detection on the abnormal area within the line of sight range. If there is no saturation in the area, mark it and store it for the next step of identification. If the area is saturated, mark the saturated distance unit in the area as a saturated area, and set the STC curve for attenuation according to the saturation situation;

[0022] S3-3. Extract the original echo features from the unsaturated abnormal area, perform spectrum analysis on each distance unit echo in the area after accumulating multiple frames, identify the area with broadened echo Doppler spectrum as the wind farm clutter area, and identify the distance unit where the wind farm clutter is located;

[0023] S4. Adaptively select wind farm clutter suppression strategies and parameters based on the comprehensive performance indicators of the identified wind farm clutter area to suppress the wind farm clutter;

[0024] Wind farm clutter suppression methods involve radar signal processing, radar point trace processing and radar data processing. The available clutter suppression methods for wind farm clutter include adaptive scanning technology, wind farm signal removal algorithm, sub-channel high-resolution clutter map, enhanced CFAR, point trace filtering algorithm and track adaptive tracking filtering algorithm.

[0025] According to the different impacts and impact degrees of wind farm clutter on radar, a wind farm clutter suppression strategy table is established. Six wind farm clutter suppression methods are arranged and combined to form an impact type-suppression strategy table. According to the comprehensive performance indicators of the wind farm clutter area, the corresponding suppression strategy is selected.

[0026] Wind farm clutter removal algorithm, high-resolution clutter map for each channel, and point-trace filtering algorithm are required options;

[0027] S5. Perform performance evaluation on the clutter area of ​​the wind farm using anti-interference suppression measurement. The evaluation indicators are detection probability improvement rate, false alarm probability improvement rate and track error improvement rate. When the improved performance reaches the set threshold, the optimization is terminated. Otherwise, the optimized measurement and parameters are optimized to iterate the anti-wind farm interference effect.

[0028] Beneficial effects of the present invention:

[0029] (1) The present invention analyzes radar data quality and extracts wind farm clutter characteristic parameters to identify wind farm clutter areas and impact types, quantifies the impact of wind farms on radar performance, and adaptively iterates the optimization parameters of wind farm suppression methods to alleviate the impact of wind farm clutter and improve radar detection performance;

[0030] (2) The software processing method of the present invention realizes the suppression of wind farm clutter without changing the hardware;

[0031] (3) The present invention evaluates the quality of radar data, analyzes and detects abnormal areas, extracts regional characteristic parameters, and automatically identifies wind farm clutter areas;

[0032] (4) The present invention adaptively selects wind farm suppression optimization parameters, performs iterative comparison operations on the optimization results, and improves radar performance, thereby reducing user manual operations and improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below in conjunction with the accompanying drawings.

[0034] Figure 1 It is a system processing flow chart of the present invention.

[0035] Figure 2 The present invention is a flowchart of data quality assessment area division.

[0036] Figure 3 It is a grid diagram of the detection range of the present invention.

[0037] Figure 4 This is a flow chart of wind farm clutter area identification according to the present invention.

[0038] Figure 5 This is a schematic diagram of the air traffic control radar area affected by the wind farm of the present invention.

[0039] Figure 6 This is a flow chart of the wind farm signal clearing algorithm of the present invention. DETAILED DESCRIPTION

[0040] 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.

[0041] See also Figure 1 As shown, the present invention is an air traffic control primary radar adaptive wind farm clutter identification and suppression method, comprising the following steps:

[0042] S1: Receive and store primary radar raw echo data, primary radar point data, primary radar track data and secondary radar track data.

[0043] The received track data is parsed by protocol, the target information is extracted, and the radar coordinates are used as reference points to unify the primary radar track data and the secondary radar track data into the same coordinate system and synchronize the processing cycle, complete the data time and space calibration, and fuse the primary radar track data with the secondary radar track;

[0044] S2: Figure 2 As shown, the azimuth and distance gridding is performed in the primary radar detection area, the quality of the primary radar data in the grid is analyzed, and the abnormal area is identified;

[0045] S2-1, such as Figure 3 As shown, the detection range of a radar is gridded according to M distance units and N azimuth sectors, and index values ​​are established for the grid units;

[0046] S2-2. Perform data quality analysis on the divided grid cells, calculate the type performance index within the grid area, and select the evaluation indicators of detection probability, false alarm rate and track error to identify abnormal areas in view of the impact of wind farm clutter on radar performance;

[0047] The detection probability Pd is described by the ratio of the total number of fused tracks in N frames within the data statistical period to the total number of fused tracks and the total number of single and secondary tracks:

[0048]

[0049] in:

[0050] Pd——false alarm rate;

[0051] N——Number of evaluation data frames;

[0052] MS——number of single secondary tracks in a single frame;

[0053] MT——Number of fused tracks in a single frame.

[0054] The false alarm rate Pf is evaluated by the ratio of the number of single tracks to the total number of fused tracks within N frames within the data statistical time period;

[0055]

[0056] in:

[0057] Pf——false alarm rate;

[0058] N——Number of evaluation data frames;

[0059] MF——Number of single-frame single-time tracks;

[0060] MT——Number of fused tracks in a single frame.

[0061] The track error D is evaluated by the mean of the total track error of N frames within the data statistical time period;

[0062]

[0063] in:

[0064] D——track error;

[0065] N——Number of evaluation data frames;

[0066] Dr——average track error of the i-th frame;

[0067]

[0068] in:

[0069] Dr——average track error of the rth frame;

[0070] k——the total number of tracks in the rth frame;

[0071] R2——the distance of the secondary radar track in the fused track;

[0072] ΔR i ——The distance error of the i-th point in the r-th frame fused track is the primary radar track distance minus the secondary track distance;

[0073] Δθ i ——The azimuth error of the i-th point in the r-th frame fused track, which is the primary radar track azimuth minus the secondary track azimuth;

[0074] σ R ——Distance system error, 60 meters is selected in this example;

[0075] σ θ ——Azimuth system error, 1 degree is selected in this example.

[0076] S2-3. Perform a comprehensive regional quality assessment based on the calculated detection probability Pd, false alarm rate Pf and track error D, and set the detection probability qualification index Pdth, false alarm rate qualification index Pfth, and track error qualification index Dth. When the detection probability in the grid area is greater than the index Pdth, the false alarm rate is less than the index Pfth, and the track error is less than the index Dth, it is judged as a normal area, and the rest of the area is judged as an abnormal detection performance area.

[0077] S2-4, storage area index, attributes and area performance indicator parameters.

[0078] S3: Figure 4As shown, for the abnormal detection performance area, the original echo data of the radar and the radar point track data are used to extract features in the abnormal area and identify the clutter area of ​​the wind farm;

[0079] S3-1, such as Figure 5 As shown, if the wind farm is not within the radar line of sight, it will not affect the radar;

[0080] Among them, with the radar as the center, the radar sight range is calculated based on the geographical location altitude and the size of the wind turbine within the radar coverage area. The abnormal area is marked and stored within the sight range for the next step of identification, otherwise it is considered not to be a wind farm clutter area;

[0081] S3-2, perform saturation detection on the abnormal area within the line of sight range. If there is no saturation in the area, mark it and store it for the next step of identification. If the area is saturated, mark the saturated distance unit in the area as a saturated area, and set the STC curve for attenuation according to the saturation situation;

[0082] S3-3, extract the original echo features of the unsaturated abnormal area, and perform spectrum analysis on the echo of each distance unit in the area after accumulating multiple frames. The area with broadened echo Doppler spectrum is marked as the wind farm clutter area, and the distance unit where the wind farm clutter is located is marked.

[0083] S4. Adaptively select wind farm clutter suppression strategies and parameters based on the comprehensive performance indicators of the identified wind farm clutter area to suppress the wind farm clutter;

[0084] The wind farm clutter suppression method involves radar signal processing, radar trace processing and radar data processing;

[0085] The clutter suppression methods available for wind farm clutter include adaptive scanning technology, wind farm filter removal algorithm, sub-channel high-resolution clutter map, enhanced CFAR, point-track filter algorithm and track adaptive tracking filter algorithm;

[0086] According to the different impacts and impact degrees of wind farm clutter on radar, a wind farm clutter suppression strategy table is established. Six wind farm clutter suppression methods are arranged and combined to form an impact type-suppression strategy table. According to the comprehensive performance indicators of the wind farm clutter area, the corresponding suppression strategy is selected.

[0087] Among them, the wind farm clutter removal algorithm, the high-resolution clutter map for each channel, and the point trace filtering algorithm are required options;

[0088] Adaptive scanning technology, according to the location information of the clutter area of ​​the wind farm, adaptively adjusts the timing of the transmitted pulse, changes the splicing position of long and short pulses, and uses narrow pulses and high repetition rate pulse trains for distance coverage. ATC primary radar usually uses long and short pulse splicing to cover the distance. The long pulse width is usually 100us in ASR and 300us in ARSR. For wind farms, the pulse pressure of long pulses is farther from the side lobe, which is not conducive to target detection. The impact of wind farms on radars is concentrated at a relatively short distance. By changing the splicing position of long and short pulses, short pulses are used to the maximum extent for close-range coverage to reduce the impact of wind farms on radars.

[0089] When there is wind farm clutter within the detection range, the wind turbines in the wind farm have a large RCS, the echo signal amplitude is strong, and the rotation of the wind turbine blades leads to the broadening of the signal spectrum. When the target signal and the wind farm echo signal are in the same or adjacent distance unit, the wind farm echo signal will cover the target signal. The wind farm echo signal is inverted, and the clearing algorithm is used to clear the wind farm signal in the original echo to achieve the wind farm suppression effect;

[0090] like Figure 6 As shown in the figure, the wind farm clutter removal algorithm extracts the amplitude and phase of the original echo signal of the distance unit where the wind farm clutter is located according to the wind farm clutter area identified in the abnormal area analysis and the distance unit identification where the wind farm clutter is located, and inverts the wind farm signal. After the original echo signal is subtracted from the inverted signal, it is determined whether the pulse pressure side lobe is reduced. When the side lobe is reduced, the next step is carried out. If it is not reduced, the signal inversion and removal are continued.

[0091] Channel-by-channel high-resolution clutter map. For each scanning cycle, each channel output of AMTD is used to update the adaptive clutter map. The estimated clutter level in the clutter map is used as the threshold value at the output of the AMTD filter to ensure the specified false alarm rate. It is mainly used for channel-by-channel filter target echo super-clutter detection. The resolution of the adaptive clutter map is one pulse group multiplied by one range quantization unit.

[0092] The traditional CFAR algorithm compares the detection unit with the average value or the larger value of the surrounding reference units to obtain the detection result. When the wind farm clutter exists, the echo amplitude of the wind farm clutter is large, which will increase the detection threshold of CFAR, so that the relatively weak target signal cannot pass the threshold and is missed. The enhanced CFAR matches and counts the reference unit whose amplitude is greater than a certain threshold. When the N / M criterion is met, the CFAR is calculated and the average noise level is used instead. N is usually selected as 3 frames and M is usually selected as 5 frames.

[0093] The point trace filtering algorithm is aimed at the point trace characteristics of wind farm echoes, and performs wind farm point trace suppression to eliminate the residual wind farm clutter in the abnormal area. The location information (distance and azimuth) of the wind farm clutter point traces is basically unchanged or changes very little, and it does not appear in every frame. In addition, the echoes that form the point traces have the characteristics of Doppler velocity broadening and low altitude. In view of such characteristics, inter-frame matching is performed based on the regional RAG map in the abnormal area. If the match can be achieved, inter-frame counting is performed, and then the criteria are extracted according to the inter-frame sliding window method. The clutter targets that meet a certain range of criteria can be determined as fixed clutter, specifically:

[0094] Step 1: Divide the abnormal area into several processing units according to distance and orientation;

[0095] Step 2: For each target point trace in the processing unit, the point trace quality is evaluated based on the relative amplitude of the point trace, the spectrum information of the point trace, and the height estimation value of the point trace;

[0096] Step 3: Determine whether the target point track parameters meet the characteristics of wind farm clutter. The points that meet the requirements are marked as "wind farm echo points" and are not used for subsequent track-related processing.

[0097] When there is clutter in the wind farm, the number of residual points in the wind farm increases and the number of target points decreases, which causes the target to deviate from the track or even start again when it passes through the wind farm area. The track adaptive tracking model refers to the use of enhanced correlation algorithm and multi-interactive tracking model to process tracks in the clutter area of ​​the wind farm, so as to improve the multi-target tracking performance in the clutter area of ​​the wind farm.

[0098] S5: Evaluate the performance of the clutter area of ​​the wind farm using anti-interference suppression measurement. The evaluation indicators are the detection probability improvement rate, false alarm probability improvement rate and track error improvement rate. When the improved performance reaches the set threshold, the optimization is terminated. Otherwise, the optimized measurement and parameters are optimized to iterate the anti-wind farm interference effect.

[0099] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An air traffic control primary radar adaptive wind farm clutter identification and suppression method, characterized in that: The following steps are involved: S1: Receive and store primary radar original echo, primary radar point track data and secondary radar track data; S2: Grid the azimuth and distance in the primary radar detection area, analyze the quality of the primary radar data in the grid, and identify abnormal areas; S3: Using the original radar echo data and radar point track data, feature extraction is performed in abnormal areas to identify clutter areas in wind farms; S4: According to the regional characteristics of the wind farm, select the wind farm clutter suppression technology and adaptively optimize the clutter suppression parameters; S5: Quantitatively evaluate the optimization effect, iterate the optimization strategy and parameters until the radar detection performance index requirements are met; The clutter suppression methods selected for wind farm clutter in S4 include adaptive scanning technology, wind farm clutter removal algorithm, sub-channel high-resolution clutter map, enhanced CFAR, point-track filtering algorithm and track adaptive tracking filtering algorithm; According to the different impacts and impact degrees of wind farm clutter on radar, a wind farm clutter suppression strategy table is established. Six wind farm clutter suppression methods are arranged and combined to form an impact type-suppression strategy table. According to the comprehensive performance indicators of the wind farm clutter area, the corresponding suppression strategy is selected. Adaptive scanning technology is to adaptively adjust the timing of transmitting pulses according to the location information of the clutter area of ​​the wind farm, change the distance splicing position of long and short pulses, and use narrow pulses and high repetition rate pulse trains for long-distance coverage; The wind farm clutter removal algorithm extracts the amplitude and phase of the original echo signal of the distance unit where the wind farm clutter is located according to the wind farm clutter area identified in the abnormal area analysis and the distance unit identification where the wind farm clutter is located, and inverts the wind farm signal. After the inversion signal is subtracted from the original echo signal, it is determined whether the pulse pressure side lobe is reduced. If the side lobe is reduced, the next step is carried out. If it is not reduced, the signal inversion and removal are continued. The sub-channel high-resolution clutter map is for each scanning cycle. Each channel output of AMTD is used to update the adaptive clutter map. The estimated clutter level in the clutter map is used as the threshold value at the output of the AMTD filter to ensure the specified false alarm rate. It is mainly used for sub-channel filter target echo super-clutter detection. The resolution of the adaptive clutter map is one pulse group multiplied by one distance quantization unit.

2. According to claim 1, the method for adaptive wind farm clutter identification and suppression of air traffic control primary radar is characterized in that: In S2, the radar detection area is gridded according to M range units and N azimuth sectors, and index values ​​are established for the grid units; The data quality of the divided grid cells is analyzed, the type performance indicators in the grid area are calculated, and the evaluation indicators such as detection probability, false alarm rate and track error are selected to identify abnormal areas.

3. The method for identifying and suppressing clutter in an air traffic control primary radar adaptive to wind farms according to claim 2 is characterized in that: The regional quality is comprehensively scored based on the calculated detection probability Pd, false alarm rate Pf and track error D. The areas that meet the index requirements are judged as normal areas, and the rest are judged as abnormal areas.

4. The method for identifying and suppressing clutter in an air traffic control primary radar adaptive to wind farms according to claim 1 is characterized in that: Enhanced CFAR matches and counts the reference units whose amplitude is greater than a certain threshold. When the N / M criterion is met, CFAR is calculated and the average noise level is used instead.

5. The method for identifying and suppressing clutter in an air traffic control primary radar adaptive to wind farms according to claim 1 is characterized in that: The point trace filtering algorithm performs inter-frame matching based on the regional RAG map in the abnormal area, performs inter-frame counting on the matching, and then extracts the criteria in an inter-frame sliding window manner. The clutter targets that meet a certain range of criteria can be determined as fixed clutter.

6. The method for identifying and suppressing clutter in an air traffic control primary radar adaptive to wind farms according to claim 1 is characterized in that: S5 includes performance evaluation of the clutter area of ​​the wind farm using anti-interference suppression measurement, and the evaluation indicators are detection probability improvement rate, false alarm probability improvement rate and track error improvement rate; That is, when the performance improvement reaches the set threshold, the optimization ends, otherwise the optimization measurements and parameters are optimized and the anti-wind farm interference effect is iterated.

Images