Acoustic precipitation sensor

Abstract

The invention relates to an acoustic precipitation sensor (10) having at least one element (12) which is arranged such that a precipitation to be detected strikes the element (12), wherein the element (12) is designed such that the striking precipitation generates an acoustic signal, and having at least one measuring element (14) which is arranged in relation to the element (12) in order to detect the acoustic signal; further having a signal processing unit (16) which is designed to receive and process a measurement signal generated by the measuring element (14) in response to the acoustic signal in order to determine one or more properties of the precipitation, for example in real time, based on the measurement signal.

Description

[0001] Acoustic precipitation sensor

[0002] Description

[0003] Embodiments of the present invention relate to an acoustic precipitation sensor and a system comprising one or more acoustic precipitation sensors and a photovoltaic system. In general, one embodiment of the present invention lies in the field of precipitation detection. Embodiments relate to an acoustic precipitation sensor, e.g., an acoustic intelligent rain sensor for plate structures or the use of acoustic sensors on photovoltaic modules or other objects, e.g., which contain plate structures, for the temporally and spatially high-resolution determination of weather data.

[0004] Weather data often includes information about temperature, wind speed, and / or precipitation. Weather data is used, among other things, to develop early warning systems for the detection of natural disasters. Precipitation sensors, in particular, can be used to provide the necessary data for the early detection of severe, heavy rain. Other possible applications for precipitation sensors include research into and / or real-time display of precipitation.

[0005] Conventional rain gauges (so-called ombrometers) are only capable of determining the amount of precipitation, not its type. To determine this, a rain gauge (so-called distrometer) is also required. Both measuring systems are generally unsuitable for installation on mobile devices due to their often large size, high weight, and high energy consumption. Furthermore, such measuring systems are unsuitable for integration into flat or encapsulated elements such as PV modules.

[0006] There are also acoustic rain sensors (e.g., Vaisala RAINCAP® Technology | Vaisala or Rain sensor RHD (sommer.at)) that feature a special exposed element, such as a hemisphere, as a separate "sensor surface." However, such sensors are incompatible with applications such as PV modules. The sensor would have to be mounted above the structure, thus altering the corresponding structure. With PV modules, the corresponding area is obscured, leading to power losses. Current precipitation sensors therefore have disadvantages not only in terms of size and installation space, but also in terms of functionality. Therefore, there is a need for an improved approach.

[0007] In addition to the precipitation sensors discussed above, the state of the art also includes precipitation radar systems, stationary weather stations, precipitation collection containers for aggregated determination of precipitation amounts (distrometers), and optical sensors for glass panes, e.g., windshields or skylights. None of these additional state-of-the-art variants overcome the disadvantages or combinations of disadvantages discussed above.

[0008] The object of the present invention is to create a concept for a precipitation sensor system that creates an improved compromise between flexibility in terms of installation space and weight as well as functional diversity, cost and energy efficiency.

[0009] The problem is solved by the subject matter of the independent patent claims.

[0010] Embodiments of the present invention provide an acoustic precipitation sensor with at least one element arranged such that precipitation to be detected impinges on the element, at least one measuring element, and a signal processing unit. The element is part of an existing object or existing structure and is designed such that the impinging precipitation generates an acoustic signal. For example, the acoustic signal can have a characteristic vibration pattern that is characteristic of one or more properties of the precipitation. The measuring element is arranged with respect to the element to detect the acoustic signal. The signal processing unit is designed to receive and process a measurement signal generated by the measuring element in response to the acoustic signal in order to determine one or more properties of the precipitation based thereon, for example in real time.

[0011] According to the exemplary embodiments, the special feature is that the sensor is integrated into existing objects with a plate-like structure, such as solar modules, vehicles, window panes, etc. An existing plate-like structure of an existing object is thus used as a quasi-sensor surface, whereby the object was not primarily developed with the goal of serving as the sensor surface of the acoustic rain sensor, but essentially fulfills a different purpose.

[0012] According to the embodiments, the properties include one or more of the following:

[0013] • Amount of precipitation,

[0014] • Precipitation rate,

[0015] • Type of precipitation,

[0016] • Drop shape,

[0017] • Drop size, drop size distribution

[0018] • Drop velocity, number of drops.

[0019] Embodiments of the present invention are based on the finding that by using an acoustic precipitation sensor, the functionality of an ombrometer and a distrometer can be combined in a single system that provides the relevant information to draw conclusions about the meteorological properties of precipitation events. This advantageously makes it possible to create a classification of raindrop size and number. In advantageous embodiments, the acoustic precipitation sensor comprises an oscillating element, e.g., a disk, plate, sensor surface, or other oscillatable surfaces, e.g., a (plate-shaped) photovoltaic module, and a corresponding measuring element, e.g., one or more microphones and / or vibration sensors for determining a sound signal, such as airborne and / or structure-borne sound.The acoustic signal recorded by one or more sensors is then fed into a signal processing system that analyzes the acoustic signal. For example, the received measurement signal or the combined measurement signal when combining multiple sensors (multiple microphones and / or structure-borne sound sensors or a combination of structure-borne and airborne sound sensors) can be compared with one or more signal patterns, e.g., using an algorithm trained using machine learning. One or more properties of the precipitation can be assigned to the one or more signal patterns, so that one or more properties of the precipitation can be determined based on the comparison. In other words, this means that a classification of the precipitation based on the one or more properties is possible.Embodiments of the present invention have the advantage that the combination of an element onto which the precipitation impinges and a measuring element for detecting an acoustic signal enables precipitation properties that were previously unmeasurable or only measurable with great effort, such as droplet size distribution, number of drops per second / area, etc. Local precipitation detection is possible with low latency and in real time. Furthermore, this ensures vertical detection and evaluation of weather elements. The sensor technology is also cost-effective to manufacture, low-maintenance, and energy-efficient. Since only one or more sound sensors are added to the normally existing components / surfaces of the device, e.g., a body or a housing, the sensor technology created in this way has advantages in terms of installation space and weight.Surfaces of a PV system or a building (building outer shell) or even a mobile object such as an aircraft can also be used.

[0020] With regard to signal processing, it should be noted that, according to exemplary embodiments, the signal processing is configured to classify or detect precipitation based on a time and / or frequency representation or signals, in particular time and / or frequency signals, e.g., a classification of raindrop sizes, raindrop counts in the case of precipitation in the liquid state. It should be noted at this point that the time-frequency behavior forms a very distinct signal pattern, so that, based on these two dimensions, different properties of the precipitation can be easily identified and thus derived.According to embodiments, the signal processing unit comprises an analysis algorithm, wherein the analysis algorithm is configured to adapt the evaluation performed by the signal processing unit to one or more environmental conditions and / or to the location of the precipitation sensor and / or to a mounting position of the measuring element relative to the element. This advantageously allows signal characteristics resulting from factors independent of the precipitation to be taken into account and thus do not negatively influence the determination or classification of the precipitation.

[0021] With regard to the measuring element, it should be noted that, according to exemplary embodiments, it comprises one or more microphones which are designed to detect a sound signal, e.g. airborne sound and / or structure-borne sound (microphones tend to detect airborne sound; however, airborne sound can also result from structure-borne sound) and / or wherein the measuring element comprises one or more vibration sensors which are designed to detect a vibration signal and / or a structure-borne sound signal. Both acoustic signals, e.g. in the audible range or in the inaudible range, and structure-borne sound signals can be detected cost-effectively, easily and reliably. Because the sensor system can also be arranged on the inside of one of the elements (e.g. integrated into a module or on the edge), it is advantageously possible to protect the sensor system from environmental influences, such as precipitation.

[0022] According to embodiments, one element can have a special material or a special geometry or special dimensions in order to generate the acoustic signal in such a way that one or more properties of the precipitation can be easily determined. Based on these influencing factors, a characteristic vibration pattern is then formed by a corresponding precipitation, so that the properties of the corresponding precipitation can be characterized based on this vibration pattern. Possible shapes are defined by the existing structure that is used as an oscillating body and can be, for example, a plate-shaped element (flat plate), a dome shape, a key shape, a corrugated sheet shape, the shape of a cavity, the shape of a resonance body, a liquid surface, e.g. water, or the shape of at least part of a body of a vehicle (e.g.of a land vehicle, aircraft, watercraft or spacecraft), etc. Other conceivable implementations are a plate, pane, glass plate, dome, housing surface, outer wall or another oscillating system. A particular embodiment can be the use of a plate-shaped photovoltaic module, for example, as the element (other shapes, such as cylindrical PV modules, would also be possible). Likewise, the use of a roof surface (roof panels or sheet metal) or generally a surface as an oscillating structure would also be conceivable. In this case, it is advantageously possible to use a photovoltaic module or any (arbitrary) surface thereof as the oscillating element and to create an acoustic precipitation sensor by supplementing it with a measuring element and signal processing.

[0023] According to embodiments, the existing object or structure utilized by the precipitation sensor can comprise a stationary structure, e.g., a PV module or the outer shell of a building, and / or a mobile structure, e.g., a vehicle. Advantageously, the described precipitation sensor thus provides a retrofit option for existing systems, such as stationary systems (e.g., PV systems or skylights) or mobile systems (e.g., a vehicle).

[0024] According to exemplary embodiments, the element can be arranged such that the impinging precipitation leaves the element again and is drained away. An inclined arrangement would be conceivable, for example. Furthermore, the element can also be heated, for example, to melt precipitation in the form of snow and allow it to drain off the element. Thus, the precipitation sensor automatically clears itself of past precipitation and can thus detect new precipitation again.

[0025] According to embodiments, the element and the measuring element are mounted in a vibration-decoupled manner. This means that, according to embodiments, a vibration-decoupling element can be provided in the acoustic precipitation sensor, which is arranged between the element and the measuring element. Such vibration decoupling manages to reduce interference with the signal. This is particularly interesting for airborne sound receivers.

[0026] According to embodiments, the precipitation sensor may comprise an amplification element configured to amplify the acoustic signal generated by the element. According to embodiments, the signal processing unit receives the amplified acoustic signal. According to embodiments, the signal processing device may also comprise means for digitizing the measurement signal (amplified acoustic signal or acoustic signal) in order to then further process it as digital signals.

[0027] As already explained above, precipitation comes in different forms, e.g. in the solid or liquid state, depending on the outside temperature, pressure, etc. The precipitation is preferably a state of matter released from the atmosphere. The acoustic precipitation sensor is designed to detect both solid (frozen) and / or liquid precipitation. According to embodiments, the precipitation sensor can also have a plurality of elements, e.g. elements arranged spatially or locally distributed. For this purpose, one or more measuring elements can be provided, for example one measuring element per element. A distribution of a plurality of acoustic precipitation sensors at several locations, such as in the case of several solar cells, each with an acoustic sensor on several PV modules or buildings, would also be conceivable.The advantage is that this creates a sensor network that delivers high-resolution data for improved weather observation and forecasting. This enables cost-effective, low-maintenance, and energy-efficient sensor technology to create a dense sensor network. Using a photovoltaic module as an oscillating element makes it possible to create a dense energy infrastructure network that is predestined for the sensor task, which, according to further embodiments, supplies itself with energy and is constantly available. This advantageously enables always-on sensor technology for weather data for permanent data acquisition. Thus, according to embodiments, the acoustic precipitation sensor can be designed to be activated permanently or only at defined times or for predetermined periods.Furthermore, human, mechanical, and weather-related measurement inaccuracies are reduced through high local sensor density and intelligent evaluation. Thus, according to embodiments, a system is created with a plurality of spatially distributed acoustic precipitation sensors, e.g., arranged on different roofs. Furthermore, the system comprises a unit connected to all precipitation sensors that aggregates the local events of the signal processing unit of the precipitation sensors and / or improves an analysis algorithm of one or more of the signal processing units of the precipitation sensors. Another embodiment provides a photovoltaic system with at least one plate-shaped photovoltaic module, a measuring element, and a signal processing unit.Advantageously, the operation of the measuring element and / or the signal processing unit can be realized with energy provided by the photovoltaic system.

[0028] At this point, it should be noted that both in photovoltaic systems and in other elements, such as roof windows, the acoustic precipitation sensor can be advantageously used according to embodiments in order to control these devices, in particular for regulation and protection.

[0029] Embodiments of the present invention are explained below with reference to the accompanying drawings. In the drawings: Fig. 1 shows a schematic representation of an acoustic rain sensor according to a basic embodiment;

[0030] Fig. 2a is a schematic representation of a rain sensor utilizing an existing dome-shaped structure (e.g., skylight) with optional features according to extended embodiments;

[0031] Fig. 2b / c schematic representations to illustrate the principle of acoustic decoupling inside the rain sensor according to embodiments;

[0032] Fig. 2d is a schematic representation of an acoustic rain sensor according to an extended embodiment;

[0033] Fig. 3 is a schematic representation of an irrigation system for evaluating a rain sensor according to embodiments;

[0034] Fig. 4 is a schematic diagram illustrating variations in the falling velocity of raindrops as a function of distance travelled (K. Wang and HR Pruppacher. “Acceleration to terminal velocity of cloud and raindrops”. In: Journal of Applied Meteorology 16.3 (1977), pp. 275-280);

[0035] Fig. 5 is a schematic table explaining measurement data from rain sensors;

[0036] Fig. 6a-d schematic spectrograms as an explanation of the acoustic signals for processing at rain sensors associated with the measurement data from Fig. 5; and

[0037] Fig. 7a-g schematic representations of possible geometries for an element of the rain sensor according to embodiments.

[0038] Before exemplary embodiments of the present invention are explained below with reference to the accompanying drawings, it should be noted that elements and structures with the same function are provided with the same reference numerals, so that the description of them is applicable to one another or interchangeable.

[0039] Fig. 1 shows an acoustic rain sensor 10 with the two central elements 12 and 14. The element 12 can be an existing element, such as a pane, glass pane, plate, dome, housing surface, outer wall, body surface or another type of vibrating surface, which is arranged such that a precipitation 17 to be detected impinges on the element 12, wherein the element 12 is designed such that the impinging precipitation 17 generates an acoustic signal 19.

[0040] Element 14 is a measuring element, such as a microphone or vibration sensor (the vibration sensor could be attached directly to the plate structure, i.e., without any spatial separation), which is arranged with respect to element 12 in such a way as to detect the acoustic signal 19. An acoustic signal can be, for example, a sound signal, e.g., in the audible or inaudible range (ultrasound, infrasound, etc.), or a structure-borne sound signal. The microphone and / or vibration sensor detects this acoustic signal 19 and converts it into a measurement signal. This measurement signal is then forwarded (directly, amplified, or preprocessed) to a signal processing device 16, which evaluates it. Alternatively, the evaluation could also be performed externally, e.g., on a server or in the cloud.

[0041] During evaluation or processing by the unit 16, one or more properties of the precipitation, such as a precipitation amount, a precipitation rate, a precipitation type, a drop shape, a drop size, a drop distribution, a drop velocity, a drop number, etc., can be determined based on the measurement signal. Such a determination is possible in real time according to embodiments. According to embodiments, the signal processing unit can be designed to compare the received measurement signal with one or more signal patterns. The one or more signal patterns can be assigned to one or more properties. A graphically displayed signal pattern for different precipitation events, ie, precipitation with different properties, is shown in Figs. 6a-d. Figs. 6a-d show different time-frequency diagrams, iei.e., spectrograms that can be assigned to different rainfall events, such as those with different droplet velocities and different water quantities. According to embodiments, the measurement signal can be compared with one or more signal patterns using a Kl algorithm. In this respect, the signal processing for evaluation comprises an algorithm trained using machine learning. Based on the comparison, the one or more properties can be determined or the precipitation can be generally classified. According to embodiments, one or more properties of the precipitation can be grouped into classes for the classification and assigned to characteristic signal patterns. Machine learning can advantageously be used to determine such a database. The determination of the database and the testing of the precipitation sensor 10 is carried out using an irrigation system 80, as shown in Fig.3 is possible.

[0042] The irrigation system 80 according to Fig. 3 comprises a water reservoir 81 that simulates rainfall 17'. For this purpose, water is pumped from the collection basin 82 into the water reservoir 81 by means of a pump 83. The water in the reservoir 81 can generate different droplet shapes of the rainfall 17' using a droplet generator 84. An overflow can also be present. The sensor 10 or the surface of the sensor 12 is provided between the droplet generator 84 or the water reservoir 81 and the collection basin 82. With this device 80, it can be demonstrated that there are differences in the acoustic characteristics of various rainfall scenarios and that these can be classified using the developed rainfall sensor 10. The water reservoir 81 feeds the droplet generator 84. The generation consists of several drops 17'.The drip speed (i.e., number of drops per unit of time), number of drops, or droplet size can be adjusted according to the respective precipitation scenarios generated. The water droplets 17' fall onto the sensor surface 12 and, through their impact, generate an acoustic signal. The runoff water is then collected by a gutter and directed into the collection basin 82. This prevents disturbing noises caused by water dripping from the edge of the sensor surface. Since droplet generation is based on the principle of gravitational pressure of water, a consistently high water column is created in the water reservoir 81 by the pump 83. To this end, the pump 83 continuously pumps water from the collection basin 82 into the water reservoir 81. The excess water generated there can be compensated for by the overflow, which feeds the excess water back into the collection basin 82.The connecting hose between the overflow and the collecting basin is not vertical, but spiral-shaped. This results in a reduced flow velocity, which in turn reduces flow noise. Fig. 4 shows the variation in the falling velocity of raindrops as a function of the distance traveled (K. Wang and HR Pruppacher. "Acceleration to terminal velocity of cloud and raindrops". In: Journal of Applied Meteorology 16.3 (1977), pp. 275-280). This illustrates that the length of the droplet fall path is an important factor. This is determined by the distance between the sensor surface 12 and the droplet generator 84. A minimum fall path of 2.5 meters was defined. As can be seen from Fig. 4, this corresponds roughly to the path required by an average raindrop (larger than 0.5 mm in diameter) to reach its final velocity of approximately 5.8 m / s.

[0043] To further explain the results, the following definitions are introduced: In meteorology, the term precipitation is generally defined as the release of water from the atmosphere. This can occur in the solid and / or liquid state and be observed or measured on the ground. A distinction is also made between different types of precipitation. There are falling, stirred up, deposited, and settled precipitation. Falling precipitation is caused by the release of water from clouds and can be in liquid or solid form. There are basically three causes for its formation: condensation, sublimation, or collision of cloud particles. Types of falling precipitation include rain, freezing rain, snow, sleet, and hail.

[0044] Rain is defined as precipitation in liquid form. Raindrops range in diameter from 0.5 to a maximum of 5 mm [source: German Weather Service]. Showers can reach diameters of up to 6 mm [source: German Weather Service]. Larger drops are also possible, of course.

[0045] The measurement setup described below was used to record the sound, or rather the vibration, of the noise generated by the drops as they hit the impact surface of the precipitation sensor. This setup consists of two measurement microphones from Microtech Gefell, a 12AQ preamplifier from GRAS, a measurement interface from HEIM (consisting of the PWAC, DIC6B, and LMF2FE modules), and a laptop with the associated Sirecord recording software. One measurement microphone consists of a MKS 221 microphone capsule and an MV212 microphone amplifier. The two microphones were each suspended from the rear and left edges, at a height of 1 meter, a distance of 50 cm from the center of the impact surface, and in the same plane as the inclined impact surface. This corresponds to an effective distance of 1.25 meters.

[0046] Based on this, measurement data can be generated, which are shown, for example, in Fig. 5. Fig. 5 shows a table with four precipitation intensities (levels 1-4). Each precipitation intensity is defined by a drip rate per dripper and a water quantity per 0.01 m 2 defined. According to embodiments, the sensor system can differentiate between the four levels or between more (higher resolution of precipitation intensities) or fewer levels.

[0047] Before each measurement data collection, the measurement microphones were calibrated using the recording software at 1000 kHz, 94 dß(SPL) and a gain factor of 20 dB set on the preamplifier. The recording was made at a sampling rate of 96 kHz. The files were converted to .wav format for further processing. The first audio channel is assigned to the microphone on the rear side and the second channel to the microphone on the left side of the impact surface. The drip speed was set manually for each stage. For this purpose, the time interval between two drops was measured for each dropper and adjusted until it was within the error tolerance. The drip speed was also defined individually for each stage. For each setting, 11 hours of audio data were recorded. This will later serve as a training dataset for a machine-based learning algorithm as part of the internal research project.For the generation and annotation of the data sets, e.g., by means of an irrigation system with a defined flow rate and controlled droplet size, see Fig. 3, an image-based recording of precipitation events can be used as an alternative reference for the acoustic recording (exposing a defined plate structure to real precipitation events and recording the precipitation using standardized methods (e.g., using reference measurement technology at weather measuring stations).

[0048] To evaluate the results, a spectrogram was created for each measurement stage (Fig. 5) (Fig. 6a-d). This shows the temporal progression of the logarithmically displayed frequency spectrum in a range from 0 to 40 kHz and over a duration of 20 seconds. As can be seen in the individual spectrograms, the impact of a water droplet on the impact surface of the precipitation sensor is expressed by a brief amplitude maximum that extends across the entire frequency range under consideration. This characteristic feature makes it possible to precisely identify the individual impacting drops. Different droplet sizes also differ in their general appearance in the spectrogram. Characteristics here include, for example, the temporal decay behavior, the maximum frequency, and the maximum sound amplitude.

[0049] The following can be seen in the spectrogram of stage 1, which represents the lowest drip velocity (Fig. 6a). The maxima, which occur at a greater temporal interval than the other stages, clearly represent the slow dripping of the irrigation system. The frequency band visible in the lower frequency range (0-260 Hz) tends to be due to the noise of the measurement setup and can also be found in the other three spectrograms. Comparing the spectrograms of stages 1-4, one can also observe a correlation between the increasing drip velocity and the increasing number of amplitude maxima.

[0050] In summary, it can be said that the precipitation sensor makes it possible to distinguish between different rain intensities based on the information contained in the spectrograms. Thus, it can be concluded that the rain sensor could fulfill the requirement outlined in the task description of enabling conclusions to be drawn about the characteristics of different precipitation events. Furthermore, in view of the results, it can be assumed that it should also be possible to analyze and classify other types of precipitation, such as snow, using this approach.

[0051] In addition, it would be conceivable to vary the sensor surface using different shapes (e.g. dome-shaped) and materials (e.g. plastic or aluminum), as individual designs are particularly well suited for specific types of precipitation.

[0052] Referring to Fig. 7a-g, different surface structures or geometries of the element onto which the precipitation impinges are discussed, along with their advantages and disadvantages as well as preferred applications. Fig. 7a shows a cavity resonator with a dome-shaped surface 12a. A cavity is created inside the cavity resonator. This cavity influences the acoustic signal when picked up by the microphone 14. Reflections of different frequencies depend on the room geometry. In this respect, this geometry can enable acoustic optimization, e.g., by pre-filtering the frequency spectrum. Another dependent factor is the material thickness, which influences the sensitivity. To take this into account, different dome shapes can be used simultaneously, as shown in Fig. 7b.

[0053] Both the embodiments of Fig. 7a and the embodiments of Fig. 7b have a curved surface 12b, which advantageously allows the impinging precipitation to run off laterally. This reduces the influence of standing water on the frequency spectrum. The shape also influences interference signals, such as those originating from wind or splashing water, which overlay the acoustic signal to be evaluated. According to embodiments, the surface can also be coated to prevent standing water.

[0054] A hydrophobic coating and / or a lotus blossom effect could be conceivable to increase precipitation dissipation. This can increase sensitivity, especially for low-level precipitation that is difficult to detect. Fig. 7g shows a further development of Fig. 7a, namely a cavity resonator with three domes. To explain the cavity resonator: The plate of the resonator represents a vibrator, while the raindrop represents the generator. The use of a resonator creates amplification. The resonance chamber influences the frequency response. Frequency components closer to the natural resonances are transmitted with less attenuation than those further away from the natural frequency. In other words, a mechanical filter is created. For example, resonance amplification can be achieved by specifically adjusting the resonances.

[0055] According to exemplary embodiments, an adaptation of the resonance is conceivable using different materials and / or thicknesses and / or shapes. For example, different materials are used instead of or in combination with the different thicknesses to create different acoustic properties. The different acoustic properties result in different acoustic characteristics. The combined consideration allows conclusions to be drawn about droplet size and quantity. For example, it should be mentioned that thinner membrane thicknesses are better suited for analyzing light rain than thicker ones, while thicker membrane thicknesses are suitable for analyzing heavy rain because they have higher attenuation and lower sensitivity and / or thus filter out certain frequencies.

[0056] Overall, it should be noted that high sensitivity would be desirable for such low-intensity weather events, whereas high sensitivity may be disadvantageous for heavy rain. According to exemplary embodiments, it would also be conceivable to use different sensors or sensor configurations in combination to selectively detect different rainfall events. In this respect, the exemplary embodiment in Fig. 7b with multiple membranes, here three, with, for example, different material thicknesses and / or different dimensions, etc., represents an advantageous compromise for detecting different weather events with different sensitivities.

[0057] According to embodiments, the surface can also deliberately have a fluid, such as water, which changes the vibration behavior. A water surface, e.g. a pool, can also be used directly. This means that instead of a solid surface, a fluid can be used according to embodiments. This causes the effect of volume pulsation. This creates a bubble radius of, for example, 0.15 mm to 15 cm and thus generates a tonality between 20 Hz and 20,000 Hz. The noise depends on the interfacial tension of the water. Depending on this, certain drop sizes (volume) and certain falling heights with certain kinetic energy can be easily detected. The water surface as a vibrating membrane transmits sound very effectively to the air. The fill level or water surface membrane has an influence on side effects, such as,The formation of secondary bubbles emanating from the main bubble, which can lead to a distortion of the frequency spectrum. The advantage of using such fluidic membranes is that the energy of secondary droplets is usually insufficient to generate a typical sound, so that the use of a fluidic membrane eliminates noise in the form of splashing water.

[0058] It should also be noted that the angle of impact, and thus also the shape, has a significant influence on the behavior. In this respect, one or more angles of impact can advantageously be defined in various embodiments, e.g., by using a corrugated sheet or similar. Figs. 7c and d show the use of wave-shaped surfaces, such as a micro-corrugated sheet (see Fig. 7d).

[0059] The dome shapes shown in Figs. 7a and 7b can be either a one-dimensional dome (basic cubic shape) as shown in Fig. 7e or a three-dimensional dome as shown in Fig. 7f (basic cylindrical shape). The embodiment shown in Fig. 7e is simpler to manufacture, while the embodiment shown in Fig. 7f provides similar impact angles in all directions.

[0060] This ultimately means that surface and object properties, such as geometric shape, surface area, material type, thickness, thickness gradient (in the sense of varying thickness), coating, etc., influence the frequency spectrum and, above all, the amplitude, so that filtering and / or amplification effects can be achieved. The base area (round, square, ...) or size can also have an influence. These factors can also reduce interference factors such as splash water or wind. A hydrophone, for example, is used with fluidic membranes, although conventional microphones are also conceivable (example: fluid film on a skylight, which uses a microphone or structure-borne sound receiver to monitor the signal transmitted from the water film to the skylight). Sound in water is typically recorded as a near-field signal.In the ocean, it has been shown that the peak is typically above 13.5 kHz, or above 12 kHz, or above 14 kHz, or above 15 kHz. In contrast, drops on a solid membrane exhibit a peak of approximately 7 kHz (range from 4.5 to 9.5 kHz, or 6 to 8 kHz, or 6.6 to 7.4 kHz). Therefore, the frequency range to be evaluated is highly dependent on the materials used and the membrane type.

[0061] All of these influencing factors can be taken into account during signal processing. Depending on the oscillating element used, signal processing can be carried out in a training or calibration process (e.g., automated). The evaluation algorithm can also be expanded during operation using machine learning.

[0062] Fig. 2a shows a sensor 10' with, for example, a dome 12b, e.g., a dome of a skylight, as the vibrating element. This dome can be made of glass, aluminum, plastic, or even a coated material, and is curved on two sides. Inside the dome, a microphone 14m serves as a signal pickup. According to exemplary embodiments, this microphone is mounted in a vibration-decoupled manner by a vibration-decoupling element 15s. An insulating material, such as insulating wool, can also be provided within this vibration-decoupling element 15s or generally around the microphone 14m. The vibration-decoupling element and the insulating material have the effect of preventing interference. Additionally or alternatively, a vibration pickup 14s can also be provided as a measurement signal pickup. This is, for example, mechanically coupled directly to the element 12b.As an alternative to a microphone, a so-called hydrophone can also be used. One possible application would be, for example, the arrangement shown in Fig. 2c.

[0063] In addition to the two sensors discussed, additional sensors can also be provided, such as temperature sensors 15t1 and 15t2. 15t1 is located on the dome and thus measures the outside temperature, while 15t2 is located inside the dome. Based on the temperatures, e.g., based on the temperature difference between the two sensor signals 15t1 and 15t2, a heating coil for heating the dome can be controlled.

[0064] According to embodiments, sensor values ​​such as temperature information or other sensor values ​​can also be used for evaluation, e.g., to adapt the signal processing. These one or more signals can be used as an additional input signal in the evaluation by means of K1.

[0065] In addition, the acoustic sensor shown here also has a data processing device 16. This device is powered externally, for example, and can transmit the sensor data externally via a data cable or via a WLAN module or radio module 16f. Encryption with the public and / or private key to the receiver or server is possible.

[0066] Fig. 2b shows another form of precipitation sensor 10". Here, an oscillating element 12c, e.g., in plate form, is shown. An acoustic sensor 14 is connected to the membrane 12c via damping elements 15s. As shown here, the sensor 14 is arranged below the membrane 12c, which offers good moisture protection. Regarding the surface shape of the membrane 12c, it should be noted that lateral drops are difficult to detect. The dome shape shown in Figs. 2a and 2d, although more complex, offers both good moisture protection properties and good detectability of lateral drops.

[0067] Fig. 2d shows a simplification of the embodiment shown in Fig. 2a. The dome-shaped membrane 12b creates a cavity, which is further sealed by a damping element 15s. The sound sensor 14 is located inside, e.g., centrally. The dome 12b has, for example, a constant radius over 180 degrees, so that lateral rain can be detected and lateral protection from wind is also ensured.

[0068] Fig. 2c shows another possible arrangement of the membrane 12c in combination with a measurement signal sensor 14, which is arranged above here and determines the acoustic signals reflected from the surface of the membrane 12c. This design is inexpensive and simple and can be used as a retrofit. The disadvantage of a plate-shaped structure, especially in a horizontal arrangement, is that the impacting drops may be sprayed upwards and a renewed impact may be detected. The advantage is that good sensitivity can be achieved with a flat membrane 12c because of the lower rigidity. With an inclined membrane 12c, drainage of the water can also be ensured. At this point, it should be noted that, according to exemplary embodiments, the membrane can also be designed in several parts, so that several elements vibrate. One or more sound sensors per membrane or per arrangement comprising several membrane elements would be conceivable.An angled surface, e.g. 10 degrees, in particular, allows the incoming water to be removed quickly and, in this case, the dirt that influences the acoustic signal to be largely removed through self-cleaning. The lotus effect or a hydrophobic coating, according to further embodiments, support these two effects. The aim here is to optimize the utilization of the Cassie-Baxter state. The wetting of solid surfaces is influenced by the surface structure or surface roughness. In the Cassie state, the droplet is located on the tips of the surface structure, whereby air can be trapped between the droplet and the surface. This effect can be used specifically to design the surface so that it has hydrophobic properties. The Wenzel state describes the opposite effect that occurs in the Cassie state.In the Wenzel state, the water droplet "interlocks" with the rough surface structure, thus creating a hydrophilic effect. (Source: Christian Dorrer and Jürgen Rühe. "Condensation and Wetting Transitions on Microstructured Ultrahydrophobic Surfaces", In: Langmuir 23.7 (2007), pp. 3820-3824) The surface size influences the signal, and in particular the signal strength. If the surface is too small, amplification using an amplifier can be used. Preferably, linear amplification across the entire frequency response is desired. Depending on the exemplary embodiments, nonlinear amplification or filtering to emphasize different frequency components can also be used. This type of filtering can be used to optimize for specific precipitation events, such as droplet sizes, and thus also to achieve selectivity.This means that, according to exemplary embodiments, adaptation to specific precipitation events is possible through the use of filters and / or amplifiers and / or by selecting the resonant frequency of the oscillating element. Thus, it would be conceivable to optimize the acoustic precipitation sensor particularly for weak precipitation events with low water volumes and / or low impact velocities and / or small droplet sizes.

[0069] In summary, it can be stated that optimization of sensitivity or selective sensitivity can be achieved through various factors, such as the material surfaces of the vibrating element, the surface shape of the vibrating element, the surface coating, and / or the targeted use of a surface film by water. This can also be achieved by technical means, such as filters or reinforcement elements. The shape and / or inclination of the vibrating element also has an influence, as they promote or prevent the water from draining away.

[0070] The application also influences the geometry, properties, and shape of the precipitation sensor. Two specific examples are explained below: the rain sensor as used under laboratory conditions and the rain sensor integrated into a photovoltaic module.

[0071] Based on the irrigation system explained above, the rain sensor 10' from Fig. 3, which is used to acquire learning data, is explained. Using the irrigation system from Fig. 3, different droplet combinations are used to determine acoustic signals associated with the droplet combinations as precipitation events. The resulting acoustic signals are recorded as frequency and amplitude spectra. Droplet size, length, and kinetic energy of the water droplets influence the signal, which can be simulated as follows.

[0072] This makes it possible to classify the droplets according to droplet size (subdivision into different diameter widths), number of droplets of different droplet diameters, and / or impact energy / speed. It has been found that the distance between the end dropper and the sensor surface has a particularly significant influence, as already explained with reference to Fig. 4. This is because the droplet height influences the final velocity. Based on this finding, a plate-shaped sensor can be used in a photovoltaic system. The photovoltaic system, or rather the plate-shaped surface, represents the oscillating element. Due to its geometric shape as a plate structure, a PV module is susceptible to the development of oscillation patterns that could be caused by rain or wind.If these patterns develop appropriately, it is conceivable that a (vibro)acoustic sensor combined with an intelligent signal analysis algorithm could create a completely new, self-sufficient, and continuously available system for weather data collection. If valid data analysis is achieved, the vision is to aggregate multiple units to create a more extensive network that can, for the first time, collect high-resolution weather data in real time.

[0073] Extreme weather conditions are considered the greatest risk to the operation of photovoltaic systems and, in the worst case, can lead to spontaneous outages. To minimize the risk to energy grids and grid operators, continuous weather monitoring is necessary to ensure energy sovereignty.

[0074] Furthermore, the collected weather data would contribute to improving meteorological weather modeling and forecasting. Precision agriculture could benefit from this by optimizing yields through resource-efficient irrigation, taking into account a resilient water supply. In terms of civil protection and disaster management, the more precise spatial and temporal weather data can be used not only to protect critical infrastructure but also to deploy emergency personnel more effectively and thus, in an emergency, to save lives and limb through appropriate measures. According to implementation examples, the aggregation of multiple local units to create a more extensive network is possible.

[0075] The approach according to the exemplary embodiments is based on the use of cost-effective, energy-efficient, acoustic sensors on solar panels in the completely new application area of ​​determining high-resolution weather data to strengthen energy sovereignty and better understand climate change. The use in photovoltaic systems is particularly advantageous because an increased expansion of photovoltaic systems is expected in the coming years, allowing the creation of a very effective sensor network for weather measurement. Photovoltaic systems can also use the sensor information directly, e.g., to forecast the expected amount of energy. Another exemplary embodiment relates to a mobile device with a corresponding sensor. As already explained above, the sensor is advantageous in terms of installation space and weight, so that even small, lightweight devices can incorporate such sensors and serve to determine weather data.Many objects have free surfaces that can serve as vibrating elements, so the weather sensor can be completed by adding a sound pickup. The surface material, surface texture, etc., all have an influence that must be taken into account.

[0076] Since the surfaces are often different, according to embodiments, the determination of precipitation data or the general classification can be carried out using different training data. The training data is preferably determined using the sensor or a comparable sensor and assigned to different precipitation events or properties of the precipitation. This leads to the improvement of meteorological weather modeling and forecasts using the recorded weather data. It should be noted that recorded measurement data can be used to measure other parameters through correlation, such as particulate matter pollution (condensation nuclei) or cloud height, or the impact of / on wind turbines.

[0077] Regarding classification: As explained above, different properties of precipitation, such as the total number of raindrops, the precipitation rate, etc., can be detected. The kinetic energy, i.e., the precipitation velocity and / or droplet size, influences the amplitude of the measurement signal. The droplet diameter influences the amplitude on the one hand, and the frequency spectrum on the other. Thus, a holistic evaluation typically allows for the determination of not just one, but several properties of precipitation. This classification allows for detailed analysis of the type and / or composition of precipitation. Furthermore, the spatial and temporal resolution of precipitation measurements and forecasts can be improved.

[0078] According to embodiments, the plate structures can be susceptible to the formation of vibration patterns caused by rain and wind, e.g., PV modules or skylights. According to embodiments, a (vibro)acoustic sensor (e.g., as a pressure and / or pressure gradient receiver) and / or vibration sensor is attached directly to the plate to detect the airborne and / or structure-borne sound. This sensor is combined with an intelligent signal analysis algorithm (possibly with machine learning) to determine the type and amount of precipitation, wind force, and other weather characteristics. To improve accuracy, signal analysis can optionally be adapted to the local sensor. The system can be expanded in conjunction with additional sensors, e.g., temperature, air pressure, wind speed, for example.

[0079] For both PV systems and other surfaces, the measuring element can be arranged above (on the side of the precipitation) or below (on the side facing away from the precipitation or protected from precipitation).

[0080] Regarding the PV module, it should be noted that the surface, for example the top glass surface of the PV mode or the laminated PV module itself, can typically be used as the oscillating surface.

[0081] Furthermore, recorded weather data can be used to control other devices, such as closing skylights or regulating and protecting PV systems. In the worst case, severe weather can lead to spontaneous outages during operation of photovoltaic systems. Continuous weather monitoring can reduce the risk to energy grids and grid operators.

[0082] Technical areas of application for the sensors explained above are

[0083] - Photovoltaic system / modules

[0084] - Roof windows and window panes

[0085] - Body and windows of land, air, water and space vehicles

[0086] - (Industrial) flat roofs, such as corrugated iron roofs

[0087] - Agricultural irrigation management

[0088] - Wind turbine

[0089] Although some aspects have been described in the context of a device, it should be understood that these aspects also represent a description of the corresponding method, so that a block or component of a device can also be understood as a corresponding method step or as a feature of a method step. Analogously, aspects described in the context of or as a method step also represent a description of a corresponding block, detail, or feature of a corresponding device. Some or all of the method steps may be carried out by (or using) a hardware apparatus, such as a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, some or more of the key method steps may be carried out by such an apparatus.

[0090] Depending on specific implementation requirements, embodiments of the invention may be implemented in hardware or software. The implementation may be performed using a digital storage medium, such as a floppy disk, a DVD, a Blu-ray Disc, a CD, a ROM, a PROM, an EPROM, an EEPROM, or a FLASH memory, a hard disk, or other magnetic or optical storage device storing electronically readable control signals that can interact or cooperate with a programmable computer system to perform the respective method. Therefore, the digital storage medium may be computer-readable.

[0091] Some embodiments according to the invention thus comprise a data carrier having electronically readable control signals capable of interacting with a programmable computer system such that one of the methods described herein is carried out.

[0092] In general, embodiments of the present invention may be implemented as a computer program product having a program code, wherein the program code is effective to perform one of the methods when the computer program product is run on a computer.

[0093] The program code can, for example, also be stored on a machine-readable medium.

[0094] Other embodiments include the computer program for performing one of the methods described herein, wherein the computer program is stored on a machine-readable medium. In other words, one embodiment of the method according to the invention is thus a computer program that has program code for performing one of the methods described herein when the computer program is executed on a computer.

[0095] A further embodiment of the method according to the invention is thus a data carrier (or a digital storage medium or a computer-readable medium) on which the computer program for carrying out one of the methods described herein is recorded.

[0096] A further embodiment of the method according to the invention is thus a data stream or a sequence of signals that represents the computer program for carrying out one of the methods described herein. The data stream or the sequence of signals can be configured, for example, to be transferred via a data communication connection, for example, via the Internet.

[0097] A further embodiment comprises a processing device, for example a computer or a programmable logic device, which is configured or adapted to carry out one of the methods described herein.

[0098] A further embodiment comprises a computer on which the computer program for performing one of the methods described herein is installed.

[0099] A further embodiment according to the invention comprises a device or system designed to transmit a computer program for performing at least one of the methods described herein to a receiver. The transmission can be electronic, wireless, or optical, for example. The receiver can be, for example, a computer, a mobile device, a storage device, or a similar device. The device or system can, for example, comprise a file server for transmitting the computer program to the receiver.

[0100] In some embodiments, a programmable logic device (e.g., a field-programmable gate array, an FPGA) may be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field-programmable gate array may interact with a microprocessor to perform any of the methods described herein. In general, in some embodiments, the methods are performed by any hardware device. This may be general-purpose hardware such as a computer processor (CPU) or method-specific hardware such as an ASIC.

[0101] The above-described embodiments are merely illustrative of the principles of the present invention. It is understood that modifications and variations of the arrangements and details described herein will be apparent to others skilled in the art. Therefore, it is intended that the invention be limited only by the scope of the following claims and not by the specific details presented in the description and explanation of the embodiments herein.

Claims

Patent claims 1. An acoustic precipitation sensor (10), comprising at least one element (12) that is part of an existing object or structure and that is arranged such that precipitation to be detected impinges on the element (12), the element (12) being designed such that the impinging precipitation generates an acoustic signal, and at least one measuring element (14) that is arranged with respect to the element (12) to detect the acoustic signal; a signal processing unit (16) that is designed to receive and process a measuring signal generated by the measuring element (14) in response to the acoustic signal in order to determine one or more properties of the precipitation based thereon, for example in real time.

2. Acoustic precipitation sensor (10) according to claim 1, wherein the signal processing unit (16) is designed to determine the one or more properties of the precipitation by evaluation using an algorithm trained by machine learning, which compares the received measurement signal with one or more signal patterns, and / or to determine the one or more properties of the precipitation by comparing the received measurement signal with one or more signal patterns to which the one or more properties of the precipitation are assigned, and / or to classify the precipitation based on the one or more properties of the precipitation; and / or wherein the evaluation (of the measurement signal, the raw data or extracted features) takes place on a server or cloud-based.

3. Acoustic precipitation sensor (10) according to claim 1 or 2; wherein the signal processing unit (16) is designed to classify and / or detect the precipitation based on a time and / or frequency representation or signal, in particular a time and / or frequency signal; and / or wherein the signal processing unit (16) is designed to determine the one or more properties based on one or more patterns of features and / or feature combinations; and / or wherein the signal processing unit (16) is designed to perform feature extraction, in particular in the form of a time-frequency transformation or a transformation into other predefined features; and / or wherein the signal processing unit (16) is designed to provide time data as input for an ML model.

4. Acoustic precipitation sensor (10) according to claim 1, 2 or 3, wherein the one or more properties are selected from a group comprising: • Amount of precipitation, • Precipitation rate, • Type of precipitation, • Drop shape, • Drop size, drop size distribution • Drop velocity, number of drops.

5. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the signal processing unit (16) comprises an analysis algorithm, wherein the analysis algorithm is designed to adapt the evaluation carried out by the signal processing unit (16) to one or more environmental conditions and / or to the location of the precipitation sensor and / or to a mounting position of the measuring element (14) with respect to the element.

6. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the at least one element (12) comprises part of a body of a vehicle, part of an aircraft or part of an object or building; and / or wherein the at least one element (12) comprises a plate, disc, glass pane, windshield, dome, housing surface, outer wall and / or other type of vibrating surface: and / or wherein the element (12) is a photovoltaic module.

7. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the measuring element (14) comprises one or more microphones designed to detect a sound signal, or wherein the measuring element (14) comprises one or more sensors designed to detect a structure-borne sound signal and / or wherein the measuring element (14) comprises a vibration sensor designed to detect a vibration signal or a structure-borne sound signal.

8. The acoustic precipitation sensor (10) according to any one of the preceding claims, wherein a material and / or a geometry and / or a dimension of the element (12) are selected to generate the acoustic signal such that the one or more properties of the precipitation can be determined; and / or wherein the acoustic signal has a characteristic vibration pattern that is characteristic of the one or more properties of the precipitation.

9. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the element (12) is arranged or configured such that the impinging precipitation leaves the element (12); and / or wherein the element (12) is heatable.

10. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the element (12) is designed as follows: dome-shaped, bowl-shaped, corrugated sheet-shaped, in the form of a cavity, in the form of a resonator, as a flat plate.

11. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the element (12) is formed as a liquid surface or surface with a liquid film or wherein the element (12) is covered with a liquid film.

12. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the element (12) and the measuring element (14) are arranged in a vibration-decoupled manner from one another; and / or which has a vibration-decoupling element arranged between the element (12) and the measuring element (14).

13. Acoustic precipitation sensor (10) according to one of the preceding claims, comprising an amplifying element configured to amplify the acoustic signal generated by the element (12), wherein the signal processing unit (16) receives the amplified acoustic signal.

14. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the precipitation to be detected comprises water released from the atmosphere in the solid and / or liquid state.

15. Acoustic precipitation sensor (10) according to one of the preceding claims, with a plurality of elements (12) which are arranged, for example, spatially or locally distributed.

16. Acoustic precipitation sensor (10) according to one of the preceding claims, which is active permanently or only at defined times or for predefined periods of time.

17. Acoustic precipitation sensor (10) according to one of the preceding claims, which is designed to support one or more further devices, in particular for regulating and protecting a photovoltaic system or a roof window.

18. Acoustic precipitation sensor (10) according to one of the preceding claims, wherein the existing object or structure comprises a stationary structure and / or a mobile structure.

19. System comprising a plurality of spatially distributed acoustic precipitation sensors (10) according to one of the preceding claims, e.g. on different roofs, and a unit which is connected to all precipitation sensors and aggregates the local results of the signal processing units (16) of the precipitation sensors and / or effects an improvement of an analysis algorithm of one or more of the signal processing units of the precipitation sensors.

20. A photovoltaic system comprising at least one photovoltaic module, at least one measuring element (14) arranged relative to the element to detect the acoustic signal; and a signal processing unit (16) configured to receive and process a measurement signal generated by the measuring element (14) in response to the acoustic signal in order to determine one or more properties of the precipitation based thereon, for example in real time.

21. Photovoltaic system according to claim 20, wherein the energy for the operation of the measuring element (14) and / or the signal processing unit (16) is provided by the photovoltaic system.

Images