Das Projekt "Ein Laser-Doppler-Anemometer fuer die Messung der Windgeschwindigkeiten und sein Einsatz bei Windstromerzeugern" wird vom Umweltbundesamt gefördert und von Universität Erlangen-Nürnberg, Institut für Verfahrenstechnik, Lehrstuhl für Strömungsmechanik durchgeführt. Objective: The present report summarizes the work carried out in this project during the year 1988. The aim in this period was to complete the wind measuring system. The main work has concentrated on two points. First to finish the mounting of the optical system. Laser-Doppler anemometer for wind velocity measurements is nearly completed. The second point of interest was to refine the signal processing in order to get an effective technique for wind velocity measurements. General Information: The technique employed requires photon detection of the scattered light from small naturally available particles. Instead of the variation of an analogue voltage at the output of the photomultiplier, the Doppler frequency is coded as a changing density of photon events. The resulting pulse train is amplitude normalized and fed into a digital correlator (Malvern K7026) which yields its temporal autocorrelation function. Due to the long measurement distance the intensity of the scattered light is much smaller as compared to that of the daylight. In order to differentiate these two light sources, an option has been provided to specify a clip level. All photon counts below the clip level are then suppressed. This has the effect that the autocorrelation function is evaluated only for those time intervals in which a sufficiently large particle giving a slightly higher than average intensity crosses the measurement volume. In this manner the autocorrelation function of the pulse train is continuously accumulated in the buffer of the correlator. The contents of this buffer are periodically transmitted to a microcomputer via the fast Access/DMA port and the correlator-buffer is reset. During the time the data gets accumulated for building up a new correlogram in the correlator the evaluation of velocity for the previous set of data is carried out by the microcomputer. A built-in hardware-FFT-card performs a fast fourier transform on the data to evaluate the spectrum of the autocorrelation function, which generally has a sharp peak at the main frequency. To estimate the signal quality, a signal-to-noise-ratio (SNR) is calculated by dividing the height of the main peak by the average level of the other frequency-components. If this value exceeds a certain preassigned number, the autocorrelation function, together with the corresponding parameters like time, position of the measurement volume, velocity, overall light intensity, etc , is stored on the disk. Thereafter the computer is ready to receive the next correlogram and the whole process is repeated. The stored values of the various parameters may be used later to apply more sophisticated but time-consuming frequency evaluation methods.