Lidar measurement

Abstract

A lidar method (110) for measuring atmospheric properties comprises providing a lidar system (112) and operating (114) an optical emitter to emit an optical beam into atmosphere, and obtaining (118) backscattering signals, wherein the method (110) further comprises modulating (114) the frequency of the optical beam emitted by the optical emitter, and analysing (122) the backscattering signals to derive a value representative of a modulation-based frequency shift. The method may be used to resolve direction ambiguity of wind speed measurements, to discern (120) backscattering from focus regions and out-of-focus regions, to determine a cloud ceiling, for focus calibration, and / or to determine (126) a wind speed from a focal region without cloud contributions, and for several or all of the foregoing.

Description

[0001] Lidar measurement

[0002] Field of the Invention

[0003] The present invention relates to a wind speed measurement system and method using Lidar. Specifically, the present invention relates to methods that use Lidar for wind speed measurements, and that derive information in addition to wind speed, such as wind direction, from Lidar measurements. More specifically, the present invention is concerned with methods to better characterise wind behaviour within a measurement volume or stack.

[0004] Background

[0005] Coherent continuous-wave (CW) Lidar (Light detection and ranging) is an established technology, and Lidar-based Doppler measurements have become a mature method to measure wind speeds at different altitudes above ground. As one example, the wind speed above ground, at altitude regions of 50 to 250 metres, and sometimes higher, is of interest for assessing the suitability of potential wind farm sites, and also for appropriately controlling wind turbine operation in an operational wind farm.

[0006] Lidar measurements rely, inherently, on back-scattering. Therefore, Doppler signals obtained from the atmosphere are stronger the denser the atmosphere is populated with natural aerosols such as dust, pollutants, droplets etc. It will be appreciated that completely “clean” air may not create an appropriate amount of backscattering. Herein, the concentration or amount of backscattering particles, aerosols, etc., shall be referred to as atmospheric density.

[0007] A particular aspect of atmospheric measurements, i.e., In the absence of hard targets, is that the atmospheric density may vary at different height levels and at different times of the day, generating fluctuations in signal to noise ratio, or power function, which can affect reliability of measurements and can make their interpretation difficult.

[0008] The present disclosure seeks to provide additional options to improve the quality of, and / or confidence in, measurements obtained from atmospheric Lidar systems.

[0009] Summary of the Invention

[0010] In accordance with a first aspect of the invention, there is disclosed a method as defined in claim 1 , for measuring atmospheric properties, the method comprising using an optical emitter to emit an optical beam into atmosphere, using an optical sensor to obtain backscattering signals of the beam from the atmosphere, and recording the backscattering signals, wherein the method further comprises modulating the frequency of the optical beam emitted by the optical emitter, recording

[0011] Version 2025-03-28 backscattering signals corresponding to modulated frequencies, and analysing the backscattering signals to derive a value representative of a modulation-based frequency shift.

[0012] It will be understood that for conventional continuous wave (CW) wind lidar measurements, a lidar system comprises a focusing mechanism of an optical emitter that is controlled to focus the optical beam at one of different focus distances, i.e. at different altitudes, also referred to as altitude slices, which may typically be at a height of interest for assessing the suitability of potential wind farms, and also for controlling and adjusting the operation of wind turbines. Although Lidar devices may be capable of measuring distances from a few meters to several 100 metres, typical heights of interest for such turbines are in the region of 50 to 250 metres.

[0013] The optical emitter may be a continuous wave laser operating in the near infrared region. As will be appreciated, by measuring backscattering signals from a focal volume of the beam, certain optical properties of the focal volume can be derived. A typical atmospheric property determined from the backscattering signals is wind speed that may be derived from the presence and amount of a frequency shift of the backscattered signal received at the optical sensor. Specifically, by evaluating a frequency shift in the backscattering signal relative to the emitted signal, a Doppler measurement can be performed, permitting wind speed to be determined.

[0014] A lidar system configured in this manner may be referred to as wind lidar. Wind lidars are, typically, ground-based systems that may be freestanding or mounted to a surface, including for marine environments, measuring generally upward to scan different altitudes of the atmosphere. In some configurations, a lidar may be side facing, such that its measurement cone axis is close to horizontal. In that case, the focus distance will be understood to relate to the horizontal distance from the lidar source.

[0015] The wavelength, or frequency, of the optical beam from the optical emitter may be altered by modulating it, for at least some periods of its operation.

[0016] A return signal received at the optical sensor may be analysed for a Doppler shift, indicative of wind speed. An appreciation underlying the present invention was that the addition of a modulated frequency of the emitted beam adds a modulation-dependent component to a frequency shift of the returned, backscattered signal. Unless specified otherwise, “modulation” herein refers to modulation of, i.e. changing, the frequency of the emitted optical beam, e.g. by ramping, or other suitable methods. As such, it will be appreciated that modulation can be used to effect rapidly varying changes in emitter frequency (as opposed to a fixed frequency shift). The modulation-dependent or modulated-based component can be derived, e.g. by deconvoluting the frequency shift, to distinguish Doppler-based shift and modulation-based shift. Specifically, the magnitudes of the shifts may differ as the modulation-based shift is range-dependent, whereas the Doppler-based shift is wind-speed dependent. The ability to discern modulation-based shift allows information in addition to wind speed to be derived from the measurement volume.

[0017] Version 2025-03-28 In some embodiments, the method comprises coordinating modulating the frequency of the optical beam and modulating the frequency of a reference beam used for deriving a value representative of the modulation-based frequency shift.

[0018] Modulation of the optical beam may be achieved via an external modulator, such as an acousto-optic modulator or an electro-optic modulator. In this case, a corresponding, coordinated modulation of the frequency of the reference beam, e.g. the local oscillator, and the emitted optical beam, may be required to allow the method to be able to determine a range-dependent frequency shift from a backscattering signal that is both range-dependent and modulation-dependent.

[0019] In some embodiments, the method comprises modulating a driving voltage of the optical emitter thereby to modulate the frequency of the optical beam emitted by the optical emitter.

[0020] An appreciation underlying some embodiments was that modulation of the optical beam, e.g. via modulation of the driving voltage, prior to splitting the beam into an emitted part and a reference part (e.g. for a local oscillator), achieves that the same modulated emitter beam is used as the source for the backscattering signal and for the reference signal. As such, a modulation of the optical emitter via the driving voltage achieves a synchronised modulation of the frequency of the optical beam and of the reference beam. For instance, the rate of change of the frequency modulation will be understood to be identical, because the emitted beam and reference signal originate from the same, modulated, optical source. Further, modulation of the optical emitter frequency via the driving voltage may achieve that a wider bandwidth, and, hence, a shallower ramp gradient, can be used for modulation than might otherwise be easily achievable.

[0021] In some embodiments, the method comprises determining, via analysis of the modulation-based frequency shift, a parameter indicative of a backscatter region outside a focal region of the optical beam.

[0022] The modulation-based frequency shift may indicate a region, such as a peak, of a backscattering signal outside the focal volume of the optical beam. A further appreciation underlying this embodiment was that the backscattering expected from atmospheric measurements from clear atmosphere, with relatively low atmospheric density, is usually weak, such that a sufficient signal-to- noise ratio is obtained primarily from a focal region of the optical beam. However, in the event of denser atmospheric conditions, such as cloud conditions, there may be other backscatter contributions, originating from outside the focal region, that are strong enough to be detected. In that case, a modulation-based frequency shift is expected to show a more pronounced offset for larger- range backscattering, compared to an unmodulated or other reference modulated beam. As a practical example, a large range frequency shift may be indicative of a cloud ceiling creating strong backscatter originating outside the focal region of a beam. As will be appreciated, this may allow determining not only the presence of cloud conditions, but also the altitude of a cloud ceiling,

[0023] Version 2025-03-28 specifically, how low the cloud ceiling is, and / or how far the cloud ceiling is from the focal distance. The cloud ceiling parameter may be referenced as a ceilometric parameter.

[0024] In some embodiments, the method comprises determining, via analysis of the modulation-based frequency shift, presence of cloud-contaminated data.

[0025] Cloud-contaminated data may be considered data obtained from focal regions underneath the cloud ceiling that includes components or contributions from cloud measurements. Cloud-contaminated data may be considered data obtained from focal regions within an altitude area underneath the cloud ceiling, e.g. 50m, 100m, 150m or 200m immediately underneath the cloud ceiling. The inventors appreciated that knowledge about the height of the cloud ceiling provides a mechanism to distinguish data originating from underneath the cloud ceiling, which can be assumed to be more likely to be affected by the presence of cloud conditions the closer the cloud ceiling. Cloud- contaminated data may be rejected from further analysis. Alternatively, a cloud correction method may be applied to cloud-contamination data to remove cloud contributions from a measurement.

[0026] In some embodiments, the method comprises determining, via analysis of the modulation-based frequency shift, a wind direction.

[0027] In this regard, the method allows wind direction ambiguity, or the “sign” of the wind direction, to be resolved. This may be achieved by comparing the sign of the Doppler-based shift component relative to the sign of a modulation-based shift component.

[0028] In some embodiments, the method comprises determining a magnitude of the modulation-based frequency shift.

[0029] By magnitude, it is meant how much the modulation-based shift is offset relative to the backscattering obtained, or expected, from a focal region. Underlying this embodiment is an appreciation that the magnitude of the modulation-based shift is proportional to the distance, or range, of the backscattering event.

[0030] In some embodiments, the method comprises determining, via analysis of the modulation-based frequency shift, a fog parameter indicative of fog-contaminated data.

[0031] Fog conditions may be defined as conditions in which clouds extend down to ground level, specifically to below a target measurement height. During fog conditions, measurements can be affected by attenuation of the emitted beam before it reaches the desired measurement height.

[0032] Similar to detecting the presence of a cloud ceiling, a detection of strong backscattering from lower altitudes, e.g. lower than 200, 150, 100, or 50 metres above ground / above the optical emitter, may be indicative of fog conditions. A distinction between cloud conditions and fog conditions may be

[0033] Version 2025-03-28 made in that cloud conditions may permit correction and / or data measurements from underneath the cloud ceiling, whereas there may not currently be a method for readily correcting measurements obtained during fog conditions.

[0034] In some embodiments, the method comprises removing cloud spectrum contributions from a wind spectrum.

[0035] In some embodiments, the method comprises determining a corrected wind speed from a focal region by altering or shifting an uncorrected wind speed, affected by cloud speed contributions, by a correction value for reducing or removing a cloud speed contribution affecting the focal region.

[0036] For simplicity, a cloud spectrum may be understood as data including cloud speed data from a cloud region, and a wind spectrum may be understood as data intended to show wind speed data from a focal region. As may be imagined, wind speed measurements from cloud spectra may affect wind speed measurements from wind spectra. By using a frequency modulated signal, the contribution of the cloud spectra can be estimated, which in turn allows a ‘cloud-corrected’ wind speed measurement to be obtained from which cloud spectra contributions are removed, or suppressed.

[0037] In some embodiments, the method comprises determining, via analysis of the modulation-based frequency shift, a focus calibration value.

[0038] An appreciation underlying this aspect was that focus calibration may be carried out without a need for a reference target, by focusing a beam into sufficiently clear atmosphere region, such that a backscattering signal is not expected from beyond the focus region, and is expected only from the focus region. A difference between a calculated modulation-based frequency shift and an expected reference value may indicate a need to calibrate device optics and / or adjustment mechanisms. The calibration may be carried out gradually, in increments, to reduce the difference between the calculated modulation-based frequency shift and the expected reference value, until the difference reaches a pre-determined threshold or zero.

[0039] In some embodiments, the method comprises determining, via analysis of the modulation-based frequency shift, two or more of a wind direction, magnitude of the modulation-based frequency shift, ceilometric parameter, cloud parameter, fog parameter, and / or focus calibration value.

[0040] As will be appreciated, the use of frequency modulation enables the option of obtaining multiple parameters using the same measurement device. To this end, the device may be configured to follow different measurement protocols depending on results of different parameters or different test criteria, and / or to adjust measurement protocols. As one example, the method may comprise making a determination about the presence of fog conditions before continuing to make a determination about the presence of cloud conditions, in circumstances in which a presence of cloud conditions is irrelevant once fog conditions have been determined.

[0041] Version 2025-03-28 As another example, the method may comprise determining the absence of cloud and fog conditions before proceeding to carry out a focus calibration relying on clear atmosphere.

[0042] As yet another example, the method may comprise making a determination of cloud conditions before proceeding to determining an altitude of a cloud ceiling, before continuing to apply cloud correction procedures. Different cloud correction algorithms may have different effect depending on the type of cloud conditions. As such, the method may comprise selecting a cloud correction procedure, correction values or estimates, or combinations thereof, that are appropriate for a particular cloud ceiling height, cloud ceiling speed, difference between cloud ceiling and target measurement altitude, and / or difference between cloud wind speed and target altitude wind speed. The method may comprise applying the selected cloud correction procedure.

[0043] In some embodiments, the method comprises operating the optical emitter in a first modulation mode and in a second modulation mode differing from the first modulation mode.

[0044] The first and second modes may be alternated. It will be appreciated that more than two modes may be used.

[0045] In some embodiments, the method comprises comparing measurements from the first and second modulation modes to characterise the frequency shift component.

[0046] In this manner, comparison of data obtained using different modulation modes may be used to determine a frequency shift component caused by frequency modulation, and a continuous mode component caused by a Doppler shift.

[0047] In some embodiments, one of the first and second modulation modes is a mode without frequency modulation, and the other is a mode with frequency modulation.

[0048] In some embodiments, the method comprises focusing the optical beam at different focus distances, and modulating the frequency at several or each of the focus distances.

[0049] The focus distance may relate to an altitude, or altitude slice, for upward-facing setups. The focus distance may relate to the measurement distance for sideways-facing setups. By modulating the frequency for the same focus distance, measurements using two or more modulation modes, such as an unmodulated mode and a modulated frequency mode, may be obtained from the same focus distance to be available for comparison.

[0050] In some embodiments, the method comprises alternating between the first and second modulation modes at each focus distance.

[0051] Version 2025-03-28 In some embodiments, the method comprises controlling the optical beam to repeat a measurement path multiple times per focus distance.

[0052] It will be understood that the focus distance, in this context, corresponds to a focus setting of the optical beam, which is moved along a path of loop form, e.g., a circular path, along an altitude slice or a path perpendicular to a sideways-facing measurement cone.

[0053] In some embodiments, the method comprises operating the optical beam in a continuous wave mode for at least a full round of the measurement path and operating the frequency modulated mode for at least a full round of the measurement path.

[0054] In some embodiments, the measurement path comprises a loop.

[0055] As will be appreciated, the measurement path may be a closed loop so as to be repeatable, for multiple measurement rounds per focus distance.

[0056] In some embodiments, modulating the frequency of the optical beam comprises ramping.

[0057] The ramping may be a repeat pattern in the manner of a sawtooth profile.

[0058] In some embodiments, modulating the frequency of the optical beam comprises repeating a periodic pattern comprising a predetermined number of ramp units and a predetermined period of constant frequency without modulation.

[0059] In some embodiments, repeated ramp units have the same ramp gradient.

[0060] Data obtained from ramps of same gradient is believed to be better suited for averaging data from multiple measurements.

[0061] In some embodiments, the number and / or direction of measurement path loops, and / or the beam modulation during a measurement loop, is determined using statistical analysis and / or machine learning.

[0062] In some embodiments, the modulation pattern is determined based on historic data for a given location, for a given installation, and / or for a given system type, or combinations of two or more thereof.

[0063] The method may comprise carrying out measurements focused at different focus distances in sequence and generating a profile sequence therefrom, and calculating an output value representative of multiple profile sequences measured over a pre-determined period of time. A profile sequence may be called a stack. For instance, a typical predetermined period of time may be

[0064] Version 2025-03-28 10 minutes, for a profile sequence, the time covering multiple measurements per focus distance (e.g., altitude) and refocusing to different focus distances (e.g., altitudes). A typical profile sequence may comprise measurements from a number between 5 and 15 different focus distance levels.

[0065] In accordance with a second aspect of the invention, there is disclosed a lidar system for measuring atmospheric properties, the system comprising an optical emitter configured to emit an optical beam into atmosphere, an optical sensor arrangement configured to obtain backscattering signals of the beam from the atmosphere, wherein the system comprises a controller to modulate the frequency of the optical beam emitted by the optical emitter, and is configured to record backscattering signals corresponding to modulated frequencies, and to analyse the backscattering signals to derive a value representative of a modulation-based frequency shift.

[0066] One or more embodiments of the second aspect may comprise a configuration to carry out the methods of any one or more embodiments of the first aspect.

[0067] In this manner, a single Lidar device may be configured to determine several atmospheric properties, such as wind direction, ceilometric parameter, cloud parameter, fog parameter, and / or a focus calibration value, as described in relation to the first aspect.

[0068] The quality or confidence of a measurement may be characterised by an output value indicating the absence of cloud conditions or by a confirmed distance of the cloud ceiling above the focal plane.

[0069] In some embodiments, a lidar system, such as the system according to the second aspect, comprises a processor and software instructions implemented by the processor, the software instructions arranged to carry out the method according to any one of the embodiments of the first aspect. The software instructions may be provided on a non-transitory storage medium.

[0070] Description of the Figures

[0071] Exemplary embodiments of the invention will now be described with reference to the Figures, in which:

[0072] Figure 1 illustrates a scanning arrangement;

[0073] Figure 2 illustrates an exemplary driving scheme;

[0074] Figure 3 is a graph visualising an exemplary measurement;

[0075] Figures 4A and 4B show illustrative graphs to visualise exemplary measurements;

[0076] Figures 5A, 5B, 5C and 5D show graphs visualising a convolution method;

[0077] Figure 6 shows exemplary steps of a measurement method;

[0078] Figure 7 shows exemplary steps of another measurement method; and

[0079] Figure 8 shows exemplary steps of yet another measurement method.

[0080] Version 2025-03-28 Description

[0081] Figure 1 illustrates a scanning setup for measuring wind speeds at different altitudes, using a continuous wave wind lidar such as the Applicant’s ZX 300 or ZX 300M system (ZX Lidars, Willow End, Blackmore Park Road, Welland, Malvern WR13 6BD, United Kingdom). A lidar system 10 provides an infrared detector and an infrared radiation source directed as a beam 12 upward to scan the atmosphere 1. The radiation source may be a laser, specifically a continuous wave laser, operated at a near infrared frequency.

[0082] Using appropriate optics, the beam 12 is focused at a focal point 14 at a height 16A in the region typically of 10 metres up to several 100 metres. For a wind measurement, the beam 12 is operated to follow a measurement path 18, here illustrated as a circular path, whereby the focal point 14 is maintained at the height 16A, and therefore follows, generally, the path 18. It will be appreciated that the circular path is just one practical example, and other path geometries may be used. The measurement path 18 may be centred about a measurement axis A.

[0083] A proportion of light is backscattered from atmospheric particles within the volume of air in close proximity to the focal point 14 and reaches the infrared detector. Using an appropriate detector arrangement and evaluation system, typically embodied as software, backscattered light can be detected and analysed, for instance for its wavelength properties. A change in backscattered wavelength (or frequency) relative to the emitted wavelength (or frequency, respectively) may be indicative of a Doppler shift, allowing Doppler measurements to be made from different positions along the path 18, to thereby determine wind speed characteristics at the height 16A determined by the focal point 14. As such, it is understood that the lidar system 10 is a wind lidar type configured to obtain wind speed measurements, e.g. by way of Doppler calculations.

[0084] A stack profile may be obtained by adjusting the optics of the laser 10 to focus the beam 12 at different altitudes, e.g., at a second height 16B, at a third height 16C, and so forth, and repeating the measurement path 18 at the different altitudes. A wind lidar of this type may comprise an adjustment mechanism to ensure it is set up in a sufficiently level, or vertical, orientation to ensure the measurement path 18 is, for practical purposes, extending horizontally, so as to be representative for a given altitude. The system is understood to remain positioned in place, or stationary, throughout the performance of the wind measurement at each of the heights. As will be appreciated, the measurement paths may increase in diameter with increasing distance from the optical source.

[0085] Figure 2 illustrates a simplified driving scheme for a continuous wave laser, as may be used in the lidar system 10. A voltage source 20 is operated, for instance by a controller, to create a driving voltage 22 to drive a laser source 24, which emits a beam 26 with a frequency depending on the driving voltage 22. As will be appreciated, by modulating the driving voltage 22, the frequency of the beam 26 emitted by the laser source 24 is modulated. Likewise, by maintaining the driving voltage 22 constant, the frequency of the beam 26 is understood to remain constant. As such, the system

[0086] Version 2025-03-28 may be operated in a frequency modulated mode and in a constant frequency (non-modulated) mode, and may switch between those modes. The system 10 is configured to direct the beam 26 via a focusing arrangement and scanning arrangement, indicated by numeral 25, such as appropriately adjustable optical components, diverters and / or motor elements, to provide a probing beam 12 that is emitted to interrogate the atmosphere 1 at a focus point indicated at an altitude 16. Backscatter is recorded as a backscatter signal 28 at a sensor 30, such as a photodiode or other sensor, and further recorded and analysed as wind speed data for further use, either as raw data output and / or for subsequent calculations. The arrangement comprising the sensor 30 will typically be understood as comprising a local oscillator, to allow a frequency shift to be determined from a difference between backscatter signal and a local oscillator wave. An appreciation contributing to the invention was that, by modulating the frequency of the optical beam by modulating the voltage (e.g., the current driving the laser source), the reference frequency shift of the local oscillator is matched to the frequency shift of the modulated light source, thereby inherently achieving synchronisation of modulation patterns and sequence are synchronised.

[0087] The illustration of Figure 2 is to be understood as an illustrative example of one of several frequency modulation setups, here to modulate beam frequency of the optical emitter via modulating driving voltage. Other methods may be known to a skilled person to create a frequency-modulated optical beam, such as passing the laser output through a separate modulator component. The frequency may be modulated by upramping, downramping, and combinations thereof.

[0088] As will be appreciated, the backscattering signal detected at the sensor 30, when measured using a continuous wave spectrum, is stronger from the focal region of the probing beam, and so readjustment of the focus of the probing beam allows different distances, i.e., altitude slices, to be interrogated, based on the assumption that backscattering from out-of-focus regions is typically negligible in atmosphere. A further appreciation underlying several embodiments of the invention was that, despite the stronger signal being expected from a focal beam region, the presence of a strong scatterer outside the focal region may also create a sufficiently strong backscattering signal from out-of-focus regions along the optical beam.

[0089] It was found that dense atmosphere, such as cloud or fog, may provide sufficiently strong backscattering to form a measurable backscattering signal, even when unfocused. To provide an illustrative example, conventional constant frequency continuous wave measurements, focused at a focal distance of 100m in low-density atmosphere, may provide a backscattering signal from an atmosphere slice at 100m, and simultaneously an overlapping backscattering signal originating from a divergent region of the beam, at a cloud ceiling that may be at 300m. In that case, a combined frequency shift may be indicative of either a wind speed of 15m / s at the focal height, or of a cloud movement of similar speed, such as 16m / s measured at the height of the cloud ceiling. Without further separation of the wind speed components, it may be difficult to distinguish the individual contributions of similar wind speeds (here: 15m / s and 16m / s, respectively) at the focal distance height from a cloud movement at the cloud ceiling. Furthermore, clouds may move relatively faster

[0090] Version 2025-03-28 than lower region wind, and may move in different directions at the same time at different altitudes. As such, measurements in cloudy conditions may be unreliable.

[0091] The use of a suitable frequency modulated optical beam creates a range-dependent frequency shift that allows an indication to be determined of the range from which a backscattering signal is obtained. This is the case even if the backscattering signal is originating from outside (e.g., further away, or closer, than) the focal region of the optical beam.

[0092] Figure 3 illustrates a polar plot indicating the Doppler component obtained from a circular path measurement at an altitude 16, here at an example height of 300m. Assuming constant wind conditions and wind direction around the measurement disk for the duration of a measurement loop, it will be appreciated that Doppler components in the measurement plane depend on the alignment of the optical beam with the horizontal wind direction, whereas the beam comes into and out of alignment as it is rotated. In this manner, the lidar measures a wind component along the beam axis, or line-of-sight wind speed. As such, the Doppler components of the backscattering signal may cover a range from a maximum, when the beam is aligned along the wind axis, to a minimum of zero, when the beam is perpendicular to the wind axis. Consequently, the Doppler value presents as a characteristic Figure-of-Eight, comprising two lobes separated by a waist, the waist indicating a zero velocity component along an axis perpendicular to wind direction, and the lobe direction indicating a velocity component along the wind axis. Using conventional Doppler frequency shift, it is possible to determine the wind speed and wind axis, indicated by the wind axis W through (in this example) the 150° and 330° direction. However, it is not straightforward to distinguish the sign, leading to an unresolved direction ambiguity.

[0093] Figure 3 illustrates four Figure-of-Eight graphs, here to illustrate two measured graphs (dotted lines 32 and 36) and two simplified graphs (solid lines 34 and 38). To obtain two measured graphs, the system 10 may be operated to obtain measurements in a frequency modulated mode, indicated by the dotted line 32, and in a constant mode, indicated by the dotted line 36. As such, the system 10 may be operating in a constant continuous wave mode for a first period of time, e.g. for the duration required to loop through a path 18, and in a frequency modulated continuous wave mode for a second period of time, e.g. for the duration required to loop through the path 18 a second time. In this manner, backscattering data from the same focal distance can be obtained for different frequency modulation modes and / or for an unmodulated (constant) mode. The simplified graphs 34 and 38 may be obtained via a suitable fitting method from the data 32 and 36, respectively. For instance, a simple fitting method may use the maximum peak of each spectrum, indicating strongest backscatter, and may exclude outlier points. An appreciation by the inventors was that, as a simplifying approximation, the atmospheric conditions and wind pattern can be assumed to remain constant for the duration of successive loops. As such, different frequency modulation measurements and unmodulated measurements from the same slice can be compared to each other.

[0094] Version 2025-03-28 In the presence of air flow, an unmodulated beam may comprise only a Doppler shift component, whereas a frequency modulated beam may comprise both a Doppler shift component and a frequency-based shift component. The frequency-based component may be determined as a difference (or remainder) between the unmodulated measurement and frequency modulated measurement.

[0095] The measurements 32 and 36 may be used to determine a fitted curve for the frequency modulated (FM) component 34 and a fitted curve for the CW (non-modulated, or constant, continuous wave) component 38, respectively. Underlying the calculation is an appreciation that the average radius of the FM graph and the constant graph is the same, but that the extent may be offset, or “stretched”. As such, one lobe 34b of the two lobes 34a, 34b is smaller than the reference continuous mode lobe 38b, and the other lobe 34a of the two lobes 34a, 34b is larger than the reference continuous mode lobe 38a. In this example, the lobe 34a is expanded whereas the lobe 34b shrinks, relative to the respective continuous mode lobes 38a and 38b, and so the orientation of the lobe 34a indicates that the wind direction is pointing in the 150° orientation, indicated by arrow 39, rather than the 330° orientation.

[0096] A further appreciation underlying the invention was that the magnitude 31 of the frequency modulated shift is range dependent, i.e., the magnitude of the shift is proportional to the range of the focal region, or, more generally, the backscattering event. In the presence of cloud conditions or fog conditions, additional backscattering signals can be observed from a cloud ceiling or fog ceiling, originating even outside the focal region. In that case, a modulation-based frequency shift will include a component of backscattering signals outside the focal point, closer to the system 10, whereas the magnitude of the shift corresponds to the height of the cloud ceiling. Based on this appreciation, the system 10 may be used as ceilometer, to measure the height of the cloud ceiling. In a variation of the method, the optical beam is not focused, e.g. focused beyond several hundred metres such that the beam has negligible focus waist, and is for practical purposes collimated. In that case, backscatter is expected from all possible scattering sources along the probing beam, and hence will be dominated by contributions from the height with the strongest scatterers reached by non-attenuated portions of the emitted beam, such as the cloud ceiling.

[0097] The measured range and backscatter from the cloud measurement can be used to estimate the level of cloud contamination expected in wind measurements. For example, if a cloud base is detected at 1 ,000 metres altitude and with low backscatter, wind measurements at 100 metres, much below the cloud base, are less likely to be cloud contaminated. However, if the cloud base is detected at 200 metres, wind measurements above 200 metres may be flagged as being of poor quality because wind speeds cannot be measured above the cloud layer with lidars.

[0098] As such, the height of clouds, i.e., their proximity to ground level, can be used to make a determination about the likelihood of measurements being affected by the presence of cloud conditions in a relevant altitude. In accordance with embodiments, the presence of cloud conditions

[0099] Version 2025-03-28 in relatively higher altitudes may be ignored if it can be determined that the measurement altitudes are sufficiently below the cloud ceiling.

[0100] Likewise, if no relevant backscatter is measured from an out-of-focus region, this can be taken as an indication of the absence of cloud conditions or other dense atmosphere conditions in the path of the optical beam.

[0101] Figures 4A and 4B illustrate plots that may be obtained without, and with cloud backscatter, respectively. Figure 4A illustrates a graph 40a conceptually corresponding to simplified line representations 34 and 38 of Figure 3, comprising a first graph 42 and a second graph 44. The first and second graphs 42, 44 represent graphs calculated from backscattering data obtained from a focal region during a circular scan.

[0102] The first graph 42 depicts a continuous wave Doppler graph comprising a first lobe 42a and a second lobe 42b. The second graph 44 depicts a frequency-shifted graph comprising a first lobe 44a and a second lobe 44b. The focal shift 41 of the first lobes 42a and 44a relative to each other indicates the wind direction.

[0103] Figure 4B illustrates a graph 40b that incorporates the first and second graphs 42, 44 of Figure 4. In addition, the graph 40b shows a third graph 46 and a fourth graph 48. The third and fourth graphs 46, 48 are obtained during the circular scan, but based on backscatter originating outside the focal region, and illustrated as representative of backscatter originating from moving cloud ceiling.

[0104] The third graph 46 depicts a continuous wave Doppler graph comprising a first lobe 46a and a second lobe 46b. The fourth graph 48 depicts a frequency-shifted graph comprising a first lobe 48a and a second lobe 48b. A cloud shift 45 of the first lobes 46a, 48a relative to each other indicates the wind (cloud) direction. In this example, the cloud shift 45 is illustrated in the same direction as the focal shift 41 . The magnitude (length) of the cloud shift 45 is larger than the magnitude of the focal shift 41 , which indicates that the cloud ceiling is higher than the focal region. Furthermore, the larger average radius of the cloud spectral lobes 46a, 46b in comparison to the average radius of the wind spectral lobes 42a, 42b indicates that the speed of the cloud ceiling is larger than the wind speed in the focal region.

[0105] As will be appreciated, if a measurement is taken using a collimated beam, or a beam with practically no focus, backscattering signals are obtained only in the presence of a dense atmosphere region, e.g. from a cloud ceiling if present. In that case, a graph such as Figure 4B would comprise only the third graph 46 and the fourth graph 48, without focus spectra in the form of graphs 42 and 44.

[0106] The ability to determine the relative height of a cloud ceiling is advantageous, because cloud conditions tend to provide backscatter even from unfocused beam distances, that may be relatively strong compared to lower-altitude atmospheric wind speed measurements. It is therefore relatively

[0107] Version 2025-03-28 likely, with conventional methods, that a determination of cloud conditions leads to measurements having to be disregarded as unreliable. In contrast, if the cloud ceiling is known, and known to be sufficiently high above a target measurement region, the presence of cloud conditions may be tolerable. For instance, a measurement may yield that the cloud ceiling is at 600 metres, yet the interrogation area of interest may be at altitudes between 100 and 250 meters, which may be considered sufficiently far underneath the cloud ceiling. As a consequence, it is believed that the method disclosed herein allows a larger proportion of useful measurements to be obtained in the presence of cloud conditions.

[0108] Alternatively or in addition, a cloud correction algorithm may be applied to data obtained during cloud measurements. The cloud correction algorithm may determine a correction value to account for influences from cloud speed. The cloud correction value may be determined for each measurement. For simplicity, herein, a speed measurement from a cloud ceiling may be referenced as “cloud spectrum” or “cloud speed”, and a speed measurement from the target altitude slice may be referenced as “wind spectrum” or “wind speed”.

[0109] An appreciation underlying this embodiment was that the wind speed measurement may comprise an error component in the form of a contribution from a contemporaneously measured cloud signal. In practical terms, a fit to the entire graph may return a wind speed value which has been skewed by the presence of cloud spectrum components in the signal. Illustrating this with reference to Figure 4B, the fitted wind speed 44 as well as the fitted wind speed shift 41 may be distorted by contributions from the cloud spectral lobes and the cloud shift 45, to an extent depending on several factors such as altitude of cloud ceiling, relative wind speeds and others. The cloud contribution may affect both magnitude and direction.

[0110] Figures 5A to 5D depict graphs to illustrate a method proposed herein to improve the quality of wind speed data by removing contributions from cloud data to the wind speed measurement. It will be understood that the Figures are intended to visualise steps of an exemplary method, and that the method does not necessarily require visualisation or the use of graphs, respectively. Figures 5A to 5C show different convolution steps of the same measurement data. Figure 5D shows three graphs corresponding to amplitudes of a representative spectrum of the Figure-of-Eight graphs of Figures 5A to 5C. For ease of reference, the same numerals are used throughout Figures 5A to 5C, using different suffixes -A, -B, and -C, respectively, to distinguish different convolution states.

[0111] Figure 5A illustrates a graph 50A showing four Figure-of-Eight graphs 52A, 54A, 56A, 58A, the representation corresponding to Figure 4B. A first pair, here the inner graphs 52A, 54A closer to the centre, depict spectra from the focal region, a Doppler-induced frequency shift wind spectrum 52A and a modulation-induced frequency shift wind spectrum 54A. A second pair, here the outer graphs 56A, 58A, further away from the centre, depict spectra from a cloud ceiling, a Doppler-induced frequency shift cloud spectrum 56A, and a modulation-induced frequency shift cloud spectrum 58A. It will be appreciated that the data is idealised and that in practice, it may be difficult to distinguish if a

[0112] Version 2025-03-28 graph is a cloud-induced shift or a focus-region shift 51A. Often, cloud-induced Doppler and modulation shifts are expected to be larger due to generally higher wind speeds and higher cloud ceilings respectively, however this is not necessarily always the case.

[0113] The suggestion made herein is to mathematically adjust the measured wind speed value by a correction value, to identify a point at which the modulation-induced frequency shift and the Doppler- induced frequency shift align. The degree of alignment may be determined iteratively, e.g., by a convolution. Figure 5B shows one example iteration, showing the same graphs as Figure 5A, shifted by a correction value, the numerals being reproduced with a suffix -B. The correction value may be an estimate, or may be a pre-determined seed value. For instance, the wind speed spectra 52B and 56B may be adjusted by an estimate of 0.5 m / s relative to the original wind speed spectra 52A and 56B. Suitable estimate values or seed values may be determined for known measurement heights and typical wind speeds.

[0114] As will be appreciated, the convolution - applied to all spectra - will affect cloud spectrum shift and wind spectrum shift differently, and may reduce or increase the wind spectrum shift 51 B, here illustrated as a wind spectrum shift 51 B that is reduced relative to the wind spectrum shift 51 A.

[0115] An appreciation leading to the development of this aspect was that convolution will eventually determine an overlap depicted in Figure 5C, where the estimated correction value results in a wind spectrum shift that practically overlaps, with small or zero offset 51 C, due to overlapping graphs 52C and 54C, at a point at which the cloud spectra still have a large offset.

[0116] Figure 5D presents the magnitude of convoluted graphs along the dotted line 53A, 53B and 53C of Figures 5A, 5B and 5C, respectively, with a simplified “straight” baseline. Convoluted graphs 53A and 53B show four distinct peaks corresponding to the graphs 52A,52B; 54A,54B; 56A,56B; and 58A,58B, respectively. Once aligned, as illustrated in Figure 5C, the convoluted graph 53C shows a merged peak, or single peak, for graphs 52C, 54C that is considerable larger than the corresponding peaks of graphs 56C, 58C. A further appreciation during the development of the method was that the alignment point, illustrated in Figure 5D by the single peak 52C,54C, can be determined regardless of the magnitude of the, or any, uncorrected cloud peak 58C or 56C. The correction value determined via convolution can be used to determine the cloud-speed corrected wind speed in the focal region. While convolution has been found to be a practical method, it will be appreciated that an end point may be determined in other ways that indicate a correlation between the modulation- induced frequency shift and the Doppler-induced frequency shift. Depending on the manner in which the end point is determined, or the manner in which convolution is applied, the method may apply one or more further correction value estimates to determine whether or not one or more further iteration steps lead to a diverging region at which the degree of convolution decreases, to avoid misinterpretation of a local maximum as absolute maximum.

[0117] Version 2025-03-28 A further appreciation was that the processing of the data, to establish the overlap, can be carried out relatively fast for most data. Specifically, the processing can usually be carried out fast enough to be completed within a measurement cycle and before the system has completed data acquisition of a subsequent measurement loop. As the procedure may be of an iterative nature, or may diverge, it is possible in some scenarios that an end point cannot be found within a predetermined time limit. This may be the case in scenarios in which an initial estimate or seed value was not appropriate. In that case; such data, i.e. scans without end point found within a predetermined time limit, may be rejected.

[0118] The methods disclosed above in relation to cloud detection are believed to be applicable for fog detection.

[0119] In a further application, the comparison of frequency modulated data with continuous wave data obtained from a focal region of the optical beam may be used to calibrate optical components. In the absence of wind movement and in clear air, a modulation-dependent shift that is other than the expected range-dependent shift may be taken as an indication that the focusing system requires calibration. The calibration of a focusing system may be confirmed by the magnitude of a modulation-dependent shift being within expected boundaries corresponding to the range at which the system 10 is focused. As will be appreciated, the calibration may be carried out via adjustment and / or servicing of optical components. Alternatively, or in addition, a calibration may be carried out mathematically, by applying a correction value to measured data.

[0120] Figure 6 shows exemplary steps of a measurement method 60 to resolve a direction ambiguity of wind speed measurements. In step 62, a lidar system is provided. The lidar system may be configured as described in relation to Figures 1 and 2, configured to allow it to emit an optical beam that is from time to time operated in a frequency-modulated mode and a continuous wave mode. In step 64, the lidar system is operated in a frequency modulated mode for at least part of the time. In step 66, the system is used to scan an optical path, such as circular path 18 (Figure 1), at a set focus distance. The focus distance may be representative of an altitude slice 16A.

[0121] In step 68, a backscattering signal is obtained at the optical sensor. It will be appreciated that the backscattering signal may be weak if atmospheric conditions are clear. The backscattering signal may be strong if a high concentration of backscattering particles and / or aerosols is present either in a focusing volume and / or due to dense atmosphere conditions such as cloud conditions.

[0122] In steps 70, 72 and 74, the backscattering signal is analysed to determine a Doppler shift, indicative of air flow movement, and to determine a modulation-based frequency shift corresponding to a shift resulting from the use of a frequency modulated beam. The modulation-based shift component may be determined by comparison, or deconvolution, with measurements obtained from the same path and the same focus distance with a constant (unmodulated) optical beam. Steps 70, 72, and 74 may be carried out in different order and / or simultaneously. For instance, the frequency modulated shift

[0123] Version 2025-03-28 component may be determined by subtracting or deconvoluting, as the case may be, the Dopplerbased shift from the combined Doppler shift and emitter-modulated shift.

[0124] In step 76, the sign of the modulation-induced shift is determined. In step 78, the sign of the modulation-induced shift is taken as indicator of the wind direction, to determine the wind direction from the beam focus area. In this manner, a direction ambiguity can be resolved that would otherwise remain unresolved from analysis of a Doppler-induced shift alone.

[0125] Some or all of the steps of the method 60 may be repeated periodically, and / or for each measurement, during the operation of a Lidar system.

[0126] Figure 7 shows exemplary steps of a measurement method 80 to determine the presence of fog or a cloud ceiling, and / or to determine a misalignment, and / or magnitude thereof, of focusing optics, to thereby enable a focus calibration. Steps 82 to 84 correspond to steps 62 to 64 of the method 60. I.e., in step 82, a lidar system is provided, capable of operating in a frequency-modulated mode and a continuous wave mode. In step 84, the lidar system is operated in a frequency modulated mode for at least part of the time. In step 86, the system is used to emit an optical beam to interrogate the atmosphere 1. To this end, the system 10 may be focused at a typical measurement distance, e.g. at 50m, 100m, 150m, 200m, 250m, 300m, 350m, or 400m, or any other suitable distance.

[0127] Alternatively, in step 86, the focus distance may be set far, for instance at 800 metres, which for practical purposes may correspond to an infinity focus, meaning that the optical beam has no or practically no pronounced focus waist, and the optical beam is for practical purposes collimated. In that case no, or practically no, signal is expected from clear regions along the near-collimated beam. In such a configuration, backscattering signals are expected only from sufficiently dense atmospheric regions, e.g. cloud or fog conditions.

[0128] During step 86, the optical beam may also be operated to follow an optical path. However, in some scenarios, the beam may be kept stationary for at least some of the duration of a measurement, for instance if no speed information is required.

[0129] In step 88, a backscattering signal is obtained at the optical sensor. Backscattering signals obtained during operation of a constant (unmodulated) optical beam may comprise a Doppler-induced frequency shift component. Backscattering signals obtained during operation of a frequency- modulated optical beam may comprise, in addition to a Doppler-induced frequency shift component, a modulation-based component. An appreciation underlying the method is that the modulationbased component will be range dependent. As such, in step 90, a range dependent shift is determined relative to a signal expected from the focus distance.

[0130] The information can be utilised depending on whether the beam was focused at a large distance, beyond 800 metres, so as to have, for practical purposes, no focal region, described in one

[0131] Version 2025-03-28 procedure covered by steps 92 to 100, or whether the beam was focused in a manner defining a focal region at a typical measurement region, described in another procedure covered by steps 100 and 102.

[0132] In step 92, the range-dependent shift is used to determine the presence and altitude of a dense backscattering layer, by analysing a shift, or offset, between the backscattering signal obtained from a continuous wave beam and the backscattering signal obtained from a modulated beam. In step 94, a determination is made whether or not the backscattering data is indicative of the presence of fog conditions or indicative of the presence of cloud conditions. A lower altitude of the dense backscattering layer may be treated as indicative of the presence of fog conditions. For instance, if the dense backscattering is in a region underneath 100 metres, or underneath 50 metres, this may be interpreted as an indication of the presence of fog conditions.

[0133] The altitude of the dense backscattering layer may be interpreted as cloud ceiling. For instance, if the dense backscattering is in a region of 700 metres or higher, the absence of backscattering underneath the cloud ceiling may indicate clear atmosphere conditions underneath the cloud ceiling. In that case, data obtained from altitudes below a cloud ceiling may be treated as acceptable data. In step 96, the method is used to determine if measurements obtained during cloud conditions may be used if obtained from altitudes underneath the cloud ceiling.

[0134] In an optional step 98, the data are corrected for cloud ceiling movement. An exemplary cloud correction method 110 is described below with reference to Figure 8.

[0135] Continuing the description of Figure 7, after step 90, the method 80 may proceed straight to step 100, without executing steps 92 to 98, to determine the difference between the range-dependent frequency shift from a focal volume. In a variation used in some embodiments, the method 80 may proceed from step 90 to step 92. If, in step 92, it is determined that that no atmospheric out-of-focus backscatter is present, this is interpreted as an indication of the absence of cloud or fog conditions, and the method may proceed from step 92 to step 100 without necessarily carrying out steps 94 to 98.

[0136] In step 100, the range-dependent frequency shift in the backscattering signal is compared to an expected distance or reference distance. A discrepancy between range-dependent frequency shift and an expected distance may be indicative of a misalignment of the optical system, e.g. of the focusing optics. As will be appreciated, if it can be assumed that the Doppler shift is zero, as would be expected in the absence of wind conditions, the method may avoid a need for determining a Doppler shift.

[0137] The data so obtained may be used for focus calibration of the system 10. In step 102, a focus calibration is applied. The focus calibration may be carried out by adjusting optical components either manually or by way of motorised adjustment mechanisms. Several methods of mechanically

[0138] Version 2025-03-28 calibrating optical components will be known to a skilled person and are not disclosed in further detail herein. The system 10 may be adjusted and measurements may be repeated until a difference between the distance calculated from a range-dependent frequency shift and the expected distance (or reference distance) is reduced to no more than a pre-determined tolerance. Steps 100 and 102 may be repeated iteratively until a desired calibration accuracy has been achieved.

[0139] Subsequently, the method may return to step 86 and carry out a further focus calibration at a different focus distance.

[0140] The method steps 100 and 102 may be carried out independently of the method steps 94 to 98. The system 10 may be configured to be able to carry out all of method steps 94 to 102, or to carry out only selected ones of the method steps, depending on system configuration. As will be appreciated, the execution of step 86 may be chosen depending on which ones of the steps 92 to 102 are carried out subsequently, e.g. the absence of cloud conditions may be determined once, in step 92, with a far-distance (infinity) focus set in step 86, before proceeding to set different focus distances for a focus calibration in steps 100 and 102.

[0141] Figure 8 shows exemplary steps of a method 110, to correct wind speed measurements that may be contaminated with cloud measurements. In step 1 12, a lidar system is provided that can be operated in a frequency-modulated mode and a continuous wave mode. In step 114, the lidar system is operated in a frequency modulated mode for at least part of the time. In step 116, the system is used to scan an optical path, at a set focus distance. In step 118, a backscattering signal is obtained at the optical sensor.

[0142] Steps 112 to 118 may correspond to the method steps 62 to 68 described above, and considerations in relation to method 60 apply mutatis mutandis to method 110.

[0143] In step 120, backscattering data from different altitudes are recorded and analysed, the different altitudes including at least a focus region and an out-of-focus region. The out-of-focus region may provide cloud spectra, i.e. backscattering data from a cloud ceiling. In step 122, the modulation- induced frequency shift and the Doppler-induced frequency shift is determined for each of the focus region wind spectra and the out-of-focus region cloud spectra. Steps 120 and 122 may correspond to the procedure described in steps 70, 72, 74, and 76, carried out for a focus region and for an out- of-focus region.

[0144] With reference to Figures 5A and 5B, the data may include both wind spectra, indicative of the wind speed at the measured altitude, and cloud spectra, indicative of the speed of the cloud ceiling. The wind speed measurement may include a component of the cloud ceiling.

[0145] In step 124, the measured modulation-induced frequency shift is compared to an expected modulation-induced frequency shift. Once the measured modulation-induced frequency shift and the

[0146] Version 2025-03-28 expected modulation-induced frequency shift align, as illustrated in Figure 5C, convolution of the data shows a much stronger response than data that are not aligned.

[0147] In step 126, the accurate shift is determined from the convolution process, and the wind speed without cloud contribution is taken as indicative of the wind speed from the focal region. Some or all of the steps 120 to 126 may be combined into a single step.

[0148] In practice, the convolution approach, is illustrated in Figures 5A to 5C, may yield a wind spectrum peak quickly for most scans, sufficiently fast to be completed within a time window until the next scan in a succession of scans. As will be appreciated, occasionally a spectrum may not yield a conclusive convolution output, or may not lead to a confident result within the available time window until the next scan. Such data may be stored for subsequent processing. Alternatively, such data may be rejected.

[0149] In an optional step 128, the method may comprise applying a time limit to completion of a convolution. The time limit may be determined by the period required to complete a scan. If data processing can be completed, i.e. yields an unambiguous result, within the permitted time limit, the result so achieved is taken as a corrected wind speed measurement. If data processing cannot be completed within the permitted time limit, or if the convolution result is not unambiguous within a permitted time limit, then the data is treated as not processed, e.g. as an outlier, and rejected. Such a criterion allows the admission of corrected data based on whether or not the correction was possible within a predetermined time window. However, other criteria and processing methods may be used.

[0150] Two or more of the methods illustrated in Figures 6, 7 and 8 may be combined in a single system capable of resolving direction ambiguity of wind measurements, range-dependent backscattering information, to determine the presence or absence of fog conditions, presence or absence of cloud conditions, and altitude of a cloud ceiling, to apply focus calibration from time to time in the absence of cloud conditions and / or to improve wind speed measurements in the presence of cloud conditions, or a few of those methods.

[0151] While examples of the description refer to altitude and altitude levels, it will be appreciated that the embodiments of the invention may be applied in any direction, e.g. to side-facing lidars, wherein the axis of the measurement cone faces sideways or nearly horizontally, and operated to measure at different focus distances, for instance if mounted on a nacelle of a wind turbine. The methods described herein may be used in offshore and / or marine applications, for instance on platforms, vessels, boats, or other installations and / or carriers.

[0152] Whilst the principle of the invention has been illustrated using exemplary embodiments, it will be understood that the invention is not so limited, and that the invention may be embodied by other variants defined within the scope of the appended claims.

[0153] Version 2025-03-28

Claims

CLAIMS:

1. A lidar method for measuring atmospheric properties, the method comprising using an optical emitter to emit an optical beam into atmosphere, using an optical sensor to obtain backscattering signals of the beam from the atmosphere, and recording the backscattering signals, wherein the method further comprises modulating the frequency of the optical beam emitted by the optical emitter, recording backscattering signals corresponding to modulated frequencies, and analysing the backscattering signals to derive a value representative of a modulation-based frequency shift.

2. The method according to claim 1 , comprising coordinating modulating the frequency of the optical beam and modulating the frequency of a reference beam used for deriving a value representative of the modulation-based frequency shift.

3. The method according to claim 1 or 2, comprising modulating a driving voltage of the optical emitter thereby to modulate the frequency of the optical beam emitted by the optical emitter.

4. The method according to any one of the preceding claims, comprising determining, via analysis of the modulation-based frequency shift, a parameter indicative of a backscatter region outside a focal region of the optical beam.

5. The method according to any one of the preceding claims, comprising determining, via analysis of the modulation-based frequency shift, presence of cloud-contaminated data.

6. The method according to any one of the preceding claims, comprising determining, via analysis of the modulation-based frequency shift, a wind direction.

7. The method according to any one of the preceding claims, comprising determining a magnitude of the modulation-based frequency shift.

8. The method according to any one of the preceding claims, comprising determining, via analysis of the modulation-based frequency shift, a fog parameter indicative of fog-contaminated data.

9. The method according to any one of the preceding claims, comprising removing cloud spectrum contributions from a wind spectrum.

10. The method according to claim 9, determining a corrected wind speed from a focal region by altering an uncorrected wind speed, affected by cloud speed contributions, by a correction value for reducing or removing a cloud speed contribution affecting the focal region.Version 2025-03-2811. The method according to any one of the preceding claims, comprising determining, via analysis of the modulation-based frequency shift, a focus calibration value.

12. The method according to any one of the preceding claims, comprising determining, via analysis of the modulation-based frequency shift, two or more of a wind direction, magnitude of the modulation-based frequency shift, ceilometric parameter, cloud parameter, fog parameter, and / or focus calibration value.

13. The method according to any one of the preceding claims, comprising operating the optical emitter in a first modulation mode and in a second modulation mode differing from the first modulation mode.

14. The method according to claim 13, comprising comparing measurements from the first and second modulation modes to characterise the frequency shift component.

15. The method according to claim 13 or 14, wherein one of the first and second modulation modes is a mode without frequency modulation, and the other is a mode with frequency modulation.

16. The method according to any one of the preceding claims, comprising focusing the optical beam at different focus distances, and modulating the frequency at several or each of the focus distances.

17. The method according to claim 16, comprising alternating between the first and second modulation modes at each focus distance.

18. The method according to any one of the preceding claims, comprising controlling the optical beam to repeat a measurement path multiple times per focus distance, wherein, optionally, the measurement path comprises a loop.

19. The method according to claim 18, comprising operating the optical beam in a continuous wave mode for at least a full round of the measurement path and operating the frequency modulated mode for at least a full round of the measurement path.

20. The method according to any one of the preceding claims, wherein modulating the frequency of the optical beam comprises ramping.

21. The method according to any one of the preceding claims, wherein modulating the frequency of the optical beam comprises repeating a periodic pattern comprising a predetermined number of ramp units and a predetermined period of constant frequency without modulation.Version 2025-03-2822. The method according to claim 20 or 21 , wherein repeated ramp units have the same ramp gradient.

23. The method according to any one of claims 18 to 22, wherein the number and / or direction of measurement path loops, and / or the beam modulation during a measurement loop, is determined using statistical analysis and / or machine learning.

24. A lidar system for measuring atmospheric properties, the system comprising an optical emitter configured to emit an optical beam into atmosphere, an optical sensor arrangement configured to obtain backscattering signals of the beam from the atmosphere, wherein the system comprises a controller to modulate the frequency of the optical beam emitted by the optical emitter, and is configured to record backscattering signals corresponding to modulated frequencies, and to analyse the backscattering signals to derive a value representative of a modulation-based frequency shift.

25. A lidar system according to claim 24, comprising a processor and software instructions implemented by the processor, the software instructions arranged to carry out the method according to any one of claims 1 to 23,Version 2025-03-28

Images