Research On Data Processing And Application Of Wind Profiler

To improve the detection performance and data quality of wind profiler, an whole methods about to set radar working mode, analyse echo spectrum, detect target position, calculate wind profile, and control data quality are suggested. The method is being applied to operation, and its capacity is discussed. The main contents and conclusions are as follows:First, for to know the feasibility that the modern spectrum analysis ways are applied in wind profiler, the wavelet technology,maximum entropy method (MEM) and Fast Fourier Translate(FFT) are contrasted by using simulation data and radar measurement data respectively. The result shows when the radar echo is strong, the effect of three methods are equivalent. But when the echo is weak, the MEM spectra are better than others. The MEM can powerfully remove the ground clutter contaminant. Furthermore, the effect of wavelet technology has close relationship with its basic wavelet which is selected. The MEM spectra are smooth, so it can be used to reduce white noise influence also. The iterative steps in MEM have some influence to the spectrum. The step calculated by Final Prediction Error (FPE) rule is less. Using 15 steps in MEM can get a better result.Second, based on studying of objective target detection method,maximal chain detection Algorithm and NCAR Improved Moments Algorithm(NIMA), a new synthetic identification method is suggested for turbulence target detection in wind profiler. The primary steps in the algorithm are the folowings:1)part and average the spectrum, then get minimum as the spectrum noise level; 2)classify the spectrum shape to remove ground clutter;3) synthetic identify the turbulence target position in spectrum; 4) integrate the section of target spectrum to calculate moments. Using radar field test data and simulation data, the method accuracy is studied. The result shows when the radar echo spectra are clean, the objective and the synthetic methods are the same for targeting. But when the spectra are disorder, or severely contaminated, the effect of synthetic method is better and the error is less.Third, through researching and contrasting the different effect whether consensus average is used or not, whether three-beam or five-beam is used to calculate wind, the Algorithm using five-beam data to get wind are given. And then the detail way of data quality control in wind profiler is produced. Comparing with pibal data, the method is effective. The Algorithm using five-beam data to get wind can improve the data quality in wind profiler.Four, the whole method is applied to operation of troposphere wind profiler. The result shows:Data availability is above 90% below 6000m. and then the data availability decrease along with high increase. Data availability is still 87% at 11850m. The data that observed by different mode at same altitude is consistent. Individual example shows that profiler data can indicate the weather changing clearly and particularly. The changing that profiler data displayed correspond with surface observation.