A method, apparatus and electronic device for measuring rainfall

Abstract

This application discloses a rainfall measurement method, apparatus, and electronic device, relating to the field of meteorological monitoring technology. The method includes: acquiring a first count value of a first-range tipping bucket corresponding to a first rainfall sensitivity within a first time period; and acquiring a second count value of a second-range tipping bucket corresponding to a second rainfall sensitivity within the first time period, wherein the second rainfall sensitivity is greater than the first rainfall sensitivity; determining whether the second count value of the second-range tipping bucket increases during the first time period; in response to an increase in the second count value, resetting the first count value of the first-range tipping bucket to zero and recounting; and determining a target rainfall amount based on the re-reset first count value and the second count value. This application can improve the resolution of rainfall measurement and the compatibility with different rainfall intensities.

Description

Technical Field

[0001] This application relates to the field of meteorological monitoring technology, and in particular to a method, apparatus and electronic device for measuring rainfall. Background Technology

[0002] The tipping bucket rain gauge is a widely used automatic rainfall measurement instrument in meteorological and hydrological fields. Its basic working principle is as follows: rainwater is collected through a receiving inlet and then fed through a funnel into a measuring bucket with two independent chambers. When one chamber receives a predetermined volume of rainwater (i.e., the sensitivity, such as the rainfall amount corresponding to 0.1 mm or 0.2 mm), it tilts due to gravity, emptying the water from that chamber and simultaneously placing the other chamber in the receiving position. Each tilting of the bucket drives a switch (such as a reed switch) to generate a switching signal. By recording the number of switching signals, the total rainfall can be calculated.

[0003] Traditional multi-flip bucket structures exist, but when calculating rainfall, existing multi-flip bucket structures either include rainfall data from multiple ranges in the calculation or require manual selection of the range, regardless of the actual rainfall amount. Either way, the data from different ranges will interfere with each other, reducing the accuracy of rainfall measurements. Summary of the Invention

[0004] This application aims to provide a rainfall measurement method, device, and electronic device to solve the technical problems in the prior art where small-range sensors have a small rainfall intensity range, large-range sensors have insufficient resolution, and multi-range sensors have difficulty in selecting the range.

[0005] To achieve the above objectives, a first aspect of this application provides a rainfall measurement method. The method includes: acquiring a first count value of a first-range tipping bucket corresponding to a first rainfall sensitivity within a first time period; and acquiring a second count value of a second-range tipping bucket corresponding to a second rainfall sensitivity within the first time period, wherein the second rainfall sensitivity is greater than the first rainfall sensitivity; determining whether the second count value of the second-range tipping bucket increases during the first time period; in response to an increase in the second count value, resetting the first count value of the first-range tipping bucket to zero and recounting; and determining a target rainfall amount based on the re-recounted first count value and the second count value. This scheme constructs a basic rainfall measurement framework through the collaborative operation of tipping buckets with high and low sensitivity and a count resetting mechanism.

[0006] In one possible implementation of the first aspect, the method further includes obtaining the flipping time series corresponding to multiple flips of the first-range tipping bucket. By introducing time dimension information, a data foundation is laid for subsequent dynamic analysis, solving the problem of relying solely on count values ​​and having a single information dimension.

[0007] In one possible implementation of the first aspect, after determining the target rainfall amount, the method further includes: determining a series of event occurrence rates of the first-range tipping bucket based on the flipping time series; determining dynamic calibration coefficients for correcting the target rainfall amount using a preset measurement event response spectrum model based on the series of event occurrence rates; and correcting the target rainfall amount using the dynamic calibration coefficients to obtain the final rainfall amount. This scheme, by constructing an innovative measurement event response spectrum model, can identify and quantify the impact of nonlinear disturbances based on the dynamic characteristics (event occurrence rate) of rainfall events and generate dynamic calibration coefficients. This solves the problem that existing technologies cannot dynamically calibrate the measurement process, resulting in a significant improvement in measurement accuracy. Furthermore, to a certain extent, it can identify and compensate for measurement deviations that may occur in large-sensitivity tipping buckets under extreme conditions through dynamic pattern analysis of small-sensitivity tipping bucket event sequences, further enhancing the robustness of the entire system.

[0008] In one possible implementation of the first aspect, the measured event response spectrum model includes at least one valid response interval and at least one invalid response interval. By dividing the event occurrence rate into different response intervals, this scheme can effectively distinguish between real rainfall signals and noise signals generated by physical disturbances, providing a basis for differentiated calibration and solving the problem of how to effectively identify signal quality.

[0009] In one possible implementation of the first aspect, the method further includes performing an adaptive model convergence process to update the measurement event response spectrum model. This application, by introducing a learning mechanism, enables the measurement system to self-optimize the calibration model based on measured data, solving the static problem of existing calibration methods having fixed parameters and being unable to adapt to environmental changes, thus bringing beneficial effects such as improved system adaptability and long-term measurement stability.

[0010] In one possible implementation of the first aspect, the adaptive model convergence process includes adjusting the boundaries of the response interval using a model plasticity parameter. This scheme, by introducing a plasticity parameter, controls the rate and magnitude of model updates, ensuring the stability and robustness of the convergence process and addressing the problem of potentially overly drastic model adjustments or sluggish responses.

[0011] In one possible implementation of the first aspect, the method further includes selecting an initial measurement event response spectrum model based on external meteorological data. This approach, by introducing a feedforward control mechanism, enables the system to pre-adjust to the state most suitable for the upcoming rainfall type, solving the problem of system response lag when dealing with sudden weather changes and bringing beneficial effects on measurement preparedness and extreme weather response capabilities.

[0012] The second aspect of this application provides a rainfall measurement device capable of performing the method described in any possible implementation of the first aspect.

[0013] A third aspect of this application provides an electronic device including a processor and a memory, wherein the memory stores an embedded program, and the processor is configured to, when executing the embedded program, implement the method described in any possible implementation of the first aspect. Attached Figure Description

[0014] Figure 1 A schematic flowchart of Embodiment 1 of the rainfall measurement method provided for some embodiments of this application;

[0015] Figure 2 A schematic flowchart of Embodiment 2 of the rainfall measurement method provided for some embodiments of this application;

[0016] Figure 3 A schematic diagram of the structure of a rainfall measurement method system provided for some embodiments of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0018] In the following description, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0019] In this application, unless otherwise expressly specified and limited, the term "connection" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral part; it can be a direct connection or an indirect connection through an intermediate medium. Furthermore, the term "electrical connection" can refer to the manner in which an electrical connection is used to achieve signal transmission.

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be described in detail below. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0021] This embodiment provides a method for measuring rainfall. The method is applied to a rainfall measurement system integrating two tipping buckets with different sensitivity ranges. Physically, the system includes a water collector, a main metering tipping bucket, and two secondary tipping buckets located below the main metering tipping bucket. The first tipping bucket has a smaller first rainfall sensitivity, for example, 0.1 mm; the second tipping bucket has a larger second rainfall sensitivity, for example, 0.5 mm. Rainfall collected by the water collector first drives the main metering tipping bucket to flip, and the discharged water then passes sequentially through the first and second tipping buckets. The core of this method lies not only in the coordinated processing of the count values ​​from the two tipping buckets, but also in the introduction of an adaptive dynamic calibration mechanism based on event time series analysis. This mechanism can significantly improve measurement accuracy in complex environments.

[0022] Example 1

[0023] The execution entity of this method can be a microcontroller (MCU) embedded in the rain gauge, or an external data processing terminal connected to the rain gauge via wired or wireless connection. The steps of this method will be described in detail below. Figure 1 As shown:

[0024] S100: Obtain the double-flipping bucket count value and execute the basic zeroing logic.

[0025] At the beginning of each measurement cycle (e.g., every minute), or during continuous system monitoring, the processing module first needs to acquire the count values ​​of the two measuring ranges of the tipping bucket. Specifically, the processing module accumulates the first count value by reading the pulse signals generated by sensors (e.g., reed switches or Hall effect sensors) associated with the first and second measuring ranges of the tipping bucket. Second count value First count value The second count value corresponds to the number of times the bucket flips in the first measuring range (sensitivity of 0.1mm). The number of times the bucket is flipped corresponding to the second range (sensitivity of 0.5 mm).

[0026] While acquiring the count value, a crucial basic processing logic is executed. The processing module continuously monitors the count value of the second-range tipping bucket. Whether an increment has occurred. Because of its larger sensitivity, each tilt of the second-range tipping bucket can be considered a relatively stable and reliable "baseline event," indicating that a determined 0.5mm rainfall has accumulated. When the processing module detects... The value from Become Time (of which) (If it is a non-negative integer), then it is determined that the second count value has increased.

[0027] In response to this benchmark event, the processing module immediately performs a zeroing operation: resets the first count value corresponding to the first range tipping bucket. The value is forcibly set to zero. The underlying logic of this operation is that, theoretically, one 0.5mm flip event contains five 0.1mm flip events. Therefore, after the 0.5mm bucket flips, the previously accumulated 0.1mm count can be considered to have been "absorbed" and "summarized" by this 0.5mm count, and should therefore be reset to zero to avoid double counting. This mechanism ensures that rainfall is not double-counted within a statistical period, forming the basis of the entire measurement system.

[0028] After performing any possible zeroing operations, the processing module bases its count on the current (potentially already zeroed) first count value. Second count value Calculate a nominal, uncalibrated target rainfall. The calculation formula is as follows: in, This is the first rainfall measurement, with a value of 0.1 mm; This is the second rainfall sensitivity reading, with a value of 0.5 mm. This can serve as a preliminary measurement result, easily understood through the analysis of the first count value. The zeroing process allows the device to automatically select the measurement range regardless of rainfall intensity. Specifically, when rainfall is heavy, the first count will repeatedly reset to zero; similarly, when rainfall is light, the second count value will reset. The value can be 0, thus avoiding interference from data of different ranges, greatly improving the resolution and accuracy of rainfall measurement, and eliminating the need for staff to select the range, reducing manual labor and avoiding errors caused by manual range selection.

[0029] For example, suppose that during a continuous monitoring process, the processing module records the following events and count changes. Time is in seconds, from... start.

[0030] The bucket flips over for the first time with a 0.1mm tilt. At this point... , .

[0031] The bucket flips for the second time by 0.1mm. At this point... , .

[0032] The bucket flips for the third time, with a diameter of 0.1mm. At this point... , .

[0033] The bucket flips for the fourth time, by 0.1mm. At this point... , .

[0034] The bucket flips for the fifth time, with a diameter of 0.1mm. At this point... , .

[0035] Immediately afterwards, the bucket tilted 0.5mm. The processing module detected this. The change from 0 to 1 represents an increment. The system immediately performs a zeroing operation, setting the value to 1. The value is set from 5 to 0. At this point, the system state is... , .

[0036] The sixth 0.1mm tipping action (this is the first after the 0.5mm tipping action). At this point... , If in If the target rainfall is calculated once, then This process fully demonstrates the complete logic of data acquisition, incremental judgment, zeroing and recounting, and basic rainfall calculation in step S100. It constitutes the core framework of the entire rainfall measurement method, ensuring the orderliness and non-repetition of the counting.

[0037] Example 2

[0038] The difference from Example 1 is that, as Figure 2 As shown, after completing the rainfall calculation in step S100, the following steps are also included to further improve the accuracy of the rainfall calculation:

[0039] S200: Obtain the flipped time series and determine the event occurrence rate.

[0040] To achieve depth calibration beyond simple counting, this step introduces a time dimension. The processing module not only records the number of times the bucket flips in the first range, but also precisely records the timestamp of each flip event. These timestamps constitute a flipping time series. ,in It is the first The absolute or relative time of each flip.

[0041] Obtaining this time series is fundamental to all subsequent dynamic analysis. The precision of the timestamps is crucial to the analysis results; therefore, the system typically uses a time resolution at the millisecond (ms) level. This timestamp information is stored in the memory of the processing module, forming a dynamically updated queue or list.

[0042] After obtaining the flipped time series, the core task of the processing module is to calculate the event occurrence rate. The event occurrence rate is a key indicator characterizing the instantaneous changes in rainfall intensity. In this embodiment, a series of instantaneous event occurrence rates... It was calculated. Among them, the first... Instantaneous event occurrence rate Based on the The second flip and the first The time interval between each flip is determined by the following formula: The physical meaning of this value is the number of flips per unit time, and its dimension is... (e.g., Hz or times / second). A high event rate corresponds to a short overturn interval, indicating a large instantaneous rainfall intensity; conversely, a low event rate corresponds to light rain or intermittent rainfall.

[0043] Through this step, the original, discrete reversal events are transformed into a continuous, quantified sequence of event occurrences. This sequence contains far more information than the original counts; it not only reflects the total amount of rainfall but, more importantly, depicts the "morphology" and "rhythm" of the rainfall process. For example, a rapid, high-occurrence sequence of events may be associated with showers or heavy rain, while a sparse, low-occurrence sequence corresponds to drizzle. More importantly, atypical reversals caused by physical disturbances (such as gusts) will appear in this event occurrence sequence as anomalous, isolated peaks or irregular fluctuations, which facilitates subsequent identification and calibration.

[0044] For example, continuing with the example in S100, we supplement it with millisecond-level timestamp information. Assume the obtained flipped time series... (Unit: seconds) is: (Note that only the 0.1mm tipping bucket overturning event timestamp from the previous example is used here.) The processing module will calculate the event occurrence rate sequence based on this sequence. :

[0045]

[0046]

[0047]

[0048]

[0049]

[0050] The final generated event occurrence rate sequence is This sequence clearly demonstrates the dynamic process of rainfall intensity first increasing and then decreasing. If any of these reversals were caused by gusts, for example... and A fake flip event is suddenly inserted between them. Then the calculated and It will be and This unusually high 1.9 Hz value became a identifiable signal of "nonlinear measurement disturbance".

[0051] S300: Dynamic calibration is performed using a measurement event response spectrum model.

[0052] In this step, the system no longer treats all 0.1mm flip events equally, but introduces a complex measurement event response spectrum model to analyze the event occurrence sequence generated by S200. The analysis and interpretation are performed, and a dynamic calibration coefficient is ultimately generated to correct the target rainfall calculated in S100. .

[0053] This event response spectrum model is essentially a predefined or learned function or set of rules. It takes the event occurrence rate as input. Mapped to a corresponding calibration factor The core idea of ​​this model is that different event occurrence rate intervals correspond to different physical processes and signal credibility. Therefore, the model is internally divided into at least two types of intervals: valid response intervals and invalid response intervals.

[0054] The Effective Response Interval covers the range of event occurrence rates considered representative of a true, stable rainfall event. Event occurrence rates falling within this interval indicate that the corresponding bucket overturning was effective and reliable. Therefore, calibration factors are assigned to these events. It is usually close to 1.0, or slightly greater than 1.0 (to compensate for possible minor losses under high rainfall intensity).

[0055] The Invalid Response Interval covers the range of event occurrence rates considered to represent nonlinear measurement disturbances (i.e., noise). For example, an extremely low occurrence rate (close to 0) might correspond to electronic noise from the instrument itself or spurious signals from sensor drift; while an extremely high occurrence rate (far exceeding physical possibility) might correspond to false flips caused by gusts of wind, severe vibration, or water splashes. Events falling into this interval have very low reliability. Therefore, the calibration factors assigned to these events... This will be a value with a strong inhibitory effect, such as being much less than 1.0, or even 0.

[0056] In the specific implementation, the processing module will process the event occurrence rate sequence. Each value in Perform a matching operation to determine which interval it falls into, and then query or calculate the corresponding calibration factor accordingly. After processing the entire sequence, the system will obtain a calibration factor sequence. .

[0057] Subsequently, the processing module needs to aggregate this calibration factor sequence to generate a single dynamic calibration coefficient that applies throughout the entire measurement cycle. There are many ways to aggregate data, for example:

[0058] Average method: ,in It is the length of the sequence.

[0059] Weighted average method: Different weights can be assigned to different calibration factors based on the occurrence rate of events or other indicators.

[0060] Product method: That is, geometric mean.

[0061] In this embodiment, the averaging method is preferred.

[0062] Finally, using the generated dynamic calibration coefficients The target rainfall amount obtained in S100 Make corrections to obtain the final, deeply calibrated rainfall amount. :

[0063] This final result Because it eliminates or suppresses the effects of nonlinear disturbances, its accuracy and reliability are compared to It will be significantly improved.

[0064] For example, suppose the preset measurement event response spectrum model is defined as follows:

[0065] Invalid response range 1 (too low): The corresponding calibration factor This range is used to suppress extremely low-frequency noise such as the flipping of residual water droplets after evaporation due to disturbance.

[0066] Valid response range: The corresponding calibration factor This range represents the normal rainfall range, from drizzle to torrential rain.

[0067] Invalid response range 2 (too high): The corresponding calibration factor This range is used to suppress physically impossible ultra-high frequency flips caused by severe instrument vibration or direct splashing of water droplets into the tipping bucket.

[0068] After completing the above construction, this model can be applied to process the event occurrence rate sequence generated in S200. .

[0069] If the response falls within the effective response range, a calibration factor is assigned. .

[0070] If the response falls within the effective response range, a calibration factor is assigned. .

[0071] If the response falls within the effective response range, a calibration factor is assigned. .

[0072] If the response falls within the effective response range, a calibration factor is assigned. .

[0073] If the response falls within the effective response range, a calibration factor is assigned. The generated calibration factor sequence is The dynamic calibration coefficients were obtained by averaging polymerization. Assuming the result calculated in S100 The final rainfall In this example, since all events are considered valid, the calibration factor is 1.

[0074] Now consider an example that includes a perturbation. Suppose the sequence is... .

[0075] ; It falls within the effective response range. (Here it is assumed that 1.9Hz is still within the effective range).

[0076] Now, assuming a more refined model, the invalid interval is defined as... ,but Falling into the invalid interval .

[0077]

[0078]

[0079]

[0080] The new calibration factor sequence is Aggregation yields Final rainfall This result effectively suppressed the exaggerated effect of the abnormally high frequency event, making the measurement results closer to the true value.

[0081] S400: Executes the adaptive model convergence process.

[0082] While static measurement event response spectrum models are effective, their interval boundaries are fixed and cannot adapt to the environmental characteristics of specific installation locations (such as specific airflow effects) or the aging of the instrument itself. To address this issue, this step introduces an adaptive model convergence process. This process endows the system with the ability to "learn" and "evolve," enabling it to dynamically and continuously optimize its internal measurement event response spectrum model based on long-term measured data.

[0083] The core driving force of this process remains the tipping event of the second-range tipping bucket (0.5mm sensitivity), which is considered the calibration signal for "ground truth." This convergence process is triggered when a 0.5mm tipping event occurs. It backtracks and examines a series of 0.1mm tipping events that occurred prior to the 0.5mm tipping (typically the most recent 4 to 6). The system analyzes the event occurrence pattern of this short 0.1mm tipping sequence and correlates it with the "result" of the 0.5mm tipping.

[0084] If the occurrence pattern of this series of 0.1mm events, after being processed by the current response spectrum model, can well "predict" or "explain" this 0.5mm flip (e.g., the cumulative calibrated count value is exactly close to 5), then the system considers the current response spectrum model to be effective. It will "reward" the current model by slightly consolidating or expanding the boundaries of the effective response intervals that have been triggered, making them easier to match in the future.

[0085] Conversely, if the pattern of the series of 0.1mm events deviates significantly from the result of the 0.5mm flip (e.g., a 0.5mm flip is triggered after only 3 flips, or after 7 flips), the system considers the current response spectrum model to be flawed. It will "penalize" the current model by adjusting the boundaries of the triggered response interval. For example, if the occurrence rate of an event that should be in the valid interval causes prediction bias, the system may fine-tune the boundary near that occurrence rate towards the "invalid" side.

[0086] The magnitude and rate of this adjustment are determined by a key model plasticity parameter. This is used to control it. This parameter is a value between 0 and 1. During the initial system installation... The initial value will be set relatively high (e.g., 0.8) to allow the model to learn quickly from the initial data, resulting in significant adjustments to the boundaries of the response spectrum. As time progresses or the amount of collected data increases, the system will gradually reduce the initial value. The value (e.g., through an exponential decay function) This allows the model to gradually stabilize and solidify after learning enough information, avoiding excessive influence from subsequent random noise events. This process is similar to the annealing of metal, slowly cooling from a high-energy, unstable state to a low-energy, ordered, and stable optimal state.

[0087] For example, suppose the upper boundary of the current valid response interval is Model plasticity parameters .exist At that time, a 0.5mm flip occurred. System backtracking analysis revealed that in a series of preceding 0.1mm flip sequences, an event with an occurrence rate of... The event was initially deemed invalid by the current model. However, comprehensive analysis revealed that the sequence including this event perfectly corresponded to a rainfall of 0.5 mm, indicating that... In the current environment, this might represent a valid event occurrence rate caused by heavy rainfall. The system identifies this as a "penalty" event (model error). Boundary adjustments are needed. Adjustments were made to expand it to include... The amount of adjustment The calculation is as follows: New upper boundary for: Through this learning process, the model expanded the upper boundary of the effective response range to 2.02Hz.

[0088] To concretize the above adjustment process and avoid improper experimentation by those skilled in the art, a preferred adaptive model convergence algorithm may include the following steps: First, after a 0.5mm flipping event occurs, the system extracts the event occurrence rate sequence of N (e.g., N=5) 0.1mm flipping events preceding that event. Second, the system uses the current measurement event response spectrum model to calculate the calibration factors corresponding to these N events and obtain a calibrated equivalent count value. Subsequently, the deviation from the theoretical value of 5 was calculated. . If there is a deviation If the absolute value exceeds a preset threshold (e.g., 0.1), boundary adjustment is initiated. The intensity of the adjustment can be determined by an overall adjustment coefficient. Control, this coefficient is the deviation A function, such as a sigmoid function , where k is the gain constant to achieve nonlinear adjustment. When an underestimation occurs ( This means that some valid events were mistakenly judged as invalid. The system will identify the events that contribute the most to the bias and are judged as invalid. and the boundary of its interval Move in the direction containing the value, the movement amount is Conversely, when overestimation occurs ( This means that some invalid events were mistakenly judged as valid. The system will identify the events that contribute the most to the bias and are judged as valid based on their occurrence rate. The boundary of the interval containing the rain gauge is then shifted in the direction that rejects the value. After thousands of such fine-tunings, the measurement event response spectrum model will increasingly accurately reflect the true physical response characteristics of the rain gauge under specific installation conditions, thereby achieving long-term high accuracy in measurement.

[0089] S500: Optionally, performs advanced feedforward control and active calibration.

[0090] To further enhance the intelligence and reliability of the system, this embodiment also includes the following two mechanisms:

[0091] Based on feedforward control using external meteorological data, this mechanism aims to give the system "predictive" capabilities. The processing module can be configured to access the internet and obtain publicly available weather forecast APIs. Before a rainfall event is about to begin, the system can obtain a prediction of the rainfall type, such as "light rain," "moderate rain," "heavy rain," or "thunderstorm." The system internally stores multiple optimized measurement event response spectrum models for different rainfall types. For example:

[0092] The "Light Rain" model offers more precise segmentation of low-event-occurrence-rate intervals and stronger suppression of extremely low-frequency noise.

[0093] The "Heavy Rain" model has a higher upper boundary for its effective response range to accommodate the extremely high event rate caused by heavy rainfall, while also incorporating special suppression logic for short-interval event pairs (e.g., less than 50ms) that may be caused by splashing. When the system receives a "Heavy Rain" warning, it automatically loads the "Heavy Rain" model as the preset model used in the S300. This feedforward control allows the system to prepare in advance rather than passively adapting after rainfall occurs, significantly improving the measurement accuracy for sudden and extreme weather events.

[0094] This mechanism, calibrated based on response analysis of active perturbations, is primarily used for instrument installation, commissioning, or routine maintenance. Maintenance personnel can use a high-precision, flow-controlled micro-pump to inject a known volume of water into the rain gauge's inlet. By setting different injection rates, rainfall of varying intensities can be accurately simulated. For example, maintenance personnel can set the injection pump to... The system injects water at a rate corresponding to a known, precise rainfall intensity. It records the complete flip-flop time series and counts generated during this process. Because the "input" rainfall is precisely known, the system can interpret the measurement results... This is directly compared to the "true ground value." If a deviation exists, the system can derive a global calibration coefficient in reverse, or further, analyze the event occurrence rate distribution under a specific water injection rate and directly forcefully and precisely correct the interval boundaries and calibration factors of the measured event response spectrum model. For example, a specific correction logic is as follows: First, based on the known water injection rate and the water inlet area of ​​the rain gauge, calculate the theoretical event occurrence rate. Then, the system records the measured average event occurrence rate under this operating condition. If there is a discrepancy between the two, an accurate operating condition calibration factor can be calculated. Finally, the system will use this precisely calibrated data to... As a high-priority anchor point, force updates to the measured event response spectrum model. The calibration factor value within the specified interval, or the entire response spectrum curve fitted and reconstructed based on multiple such anchor points. This process is equivalent to a comprehensive "laboratory-level" calibration of the instrument, enabling the rapid construction of a highly accurate initial model, or efficient recalibration of aging or misaligned instruments.

[0095] This embodiment provides a rainfall measurement device, which is the physical implementation of the method described above. This device can be an embedded system integrated within the housing of a tipping bucket rain gauge, or it can be a separate data acquisition and processing unit connected to the rain gauge sensor section. Figure 3 As shown, the device mainly includes the following modules in its structure: an acquisition module, a judgment module, a processing module, a determination module, and a correction module. These modules can be hardware entities or software functional units running on a processor.

[0096] The acquisition module is responsible for interfacing with the rain gauge's sensor. It typically includes sensor interface circuitry to receive pulse signals from reed switches or Hall effect sensors in both the first and second range tipping buckets. Each time a pulse signal is received, the acquisition module timestamps the event with high precision and updates the corresponding count value in its internal memory. or Therefore, the output of this module is the real-time double-tipping bucket count value and the flipping time series of the first-range tipping bucket.

[0097] The judgment module is used to continuously monitor the second count value provided by the acquisition module. It determines this by comparing the current value with the value at the previous moment. Has an increment occurred? Once an increment is detected, it immediately generates a trigger signal and sends it to the processing module.

[0098] The processing module is the core computing unit of the device, typically implemented by a microprocessor (CPU) or microcontroller (MCU). It receives trigger signals from the judgment module. Upon receiving the signal, it executes the process of setting the first count value... The operation of clearing to zero.

[0099] The determination module is also implemented by the processor. Based on the first count value (which may be reset after the processing module's operation) and the current second count value, it applies preset sensitivity values ​​(0.1mm and 0.5mm) to calculate the target rainfall. .

[0100] The correction module is also executed by the processor and implements the complex algorithms described in S200, S300, and S400. Specifically, it first obtains the flipped time series from the acquisition module and calculates the event occurrence rate sequence. Then, it loads a preset or adaptively updated measurement event response spectrum model from memory, processes the event occurrence rate sequence, and generates dynamic calibration coefficients. Finally, this coefficient is used to correct the target rainfall amount calculated by the determination module. This outputs the final high-precision rainfall data. In addition, this module is also responsible for performing the adaptive model convergence process, updating the response spectrum model parameters stored in non-volatile memory (such as Flash) based on the 0.5mm flip event.

[0101] This embodiment provides an electronic device, such as a dedicated meteorological data logger, an industrial computer, or a server. The electronic device includes a processor, a memory, and a bus. The bus is used to enable communication between the processor and the memory, as well as with other components (such as network interfaces and human-machine interfaces). The memory can be, for example, high-speed random access memory (RAM) or stable non-volatile memory, such as a hard disk or flash memory. The memory contains or downloads computer programs (i.e., application software). The processor, such as a central processing unit (CPU), when configured to execute the computer program stored in the memory, can fully implement all the steps of the rainfall measurement method described in Embodiment 1, including data acquisition, baseline zeroing, event rate calculation, calibration using a response spectrum model, and adaptive model convergence.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

[0103] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0104] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0105] The units described as separate components may or may not be physically separate. A component shown as a unit can be one physical unit or multiple physical units; that is, it can be located in one place or distributed in multiple different places. Depending on actual needs, some or all of the units can be selected to achieve the purpose of this embodiment.

[0106] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit described above can be implemented in hardware.

[0107] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of measuring rainfall, characterized by, The method comprises: obtaining a first count value of a first range tipping bucket corresponding to a first rainfall intensity in a first time period, and obtaining a second count value of a second range tipping bucket corresponding to a second rainfall intensity in the first time period, wherein the second rainfall intensity is greater than the first rainfall intensity; determining whether the second count value of the second range tipping bucket increases in the first time period; in response to the second count value increasing, resetting and recounting the first count value of the first range tipping bucket; based on the reset and recounted first count value and the second count value, determining a target rainfall; after the step of determining the target rainfall, the method further comprises: obtaining a sequence of tipping times corresponding to multiple tipping of the first range tipping bucket; based on the sequence of tipping times, determining a series of event occurrence rates of the first range tipping bucket; based on the series of event occurrence rates, using a preset measurement event response spectrum model to determine a dynamic calibration coefficient for correcting the target rainfall; using the dynamic calibration coefficient to correct the target rainfall to obtain a final rainfall; wherein the measurement event response spectrum model comprises at least one effective response interval and at least one ineffective response interval; and the step of determining the dynamic calibration coefficient for correcting the target rainfall comprises: matching each event occurrence rate in the series of event occurrence rates with the effective response interval and the ineffective response interval; assigning a first type of calibration factor to the event occurrence rate matched to the effective response interval; assigning a second type of calibration factor to the event occurrence rate matched to the ineffective response interval, the second type of calibration factor having a stronger inhibitory effect than the first type of calibration factor; aggregating multiple calibration factors corresponding to the series of event occurrence rates to generate the dynamic calibration coefficient.

2. The method of claim 1, wherein, The method further comprises: performing an adaptive model convergence process to update the measurement event response spectrum model.

3. The method of claim 2, wherein, The adaptive model convergence process comprises: when the second count value of the second range tipping bucket increases, identifying one or more target event occurrence rates associated with the second count value increasing event; based on the matching result of the one or more target event occurrence rates and the measurement event response spectrum model, and in combination with a preset model plasticity parameter, adjusting the boundaries of the effective response interval and the ineffective response interval.

4. The method of claim 3, wherein, The adaptive model convergence process further comprises: as the number of tipping of the first range tipping bucket accumulates or the measurement time elapses, performing attenuation processing on the model plasticity parameter.

5. The method of claim 1, wherein, The method further comprises: before the measurement starts, obtaining external meteorological data related to the to-be-measured rainfall event; based on the external meteorological data, selecting a model from a plurality of candidate measurement event response spectrum models as the preset measurement event response spectrum model.

6. A rainfall measuring device, characterised in that, The method comprises: The acquisition module is configured to acquire a first count value of a first range tipping bucket corresponding to a first rainfall intensity in a first time period, and acquire a second count value of a second range tipping bucket corresponding to a second rainfall intensity in the first time period, wherein the second rainfall intensity is greater than the first rainfall intensity; The determination module is configured to determine whether the second count value of the second range tipping bucket increases in the first time period; The processing module is configured to, in response to the second count value increasing, clear and re-count the first count value of the first range tipping bucket; The determination module is configured to determine a target rainfall based on the first count value and the second count value after being cleared and re-counted; The acquisition module is further configured to acquire a sequence of tipping times corresponding to multiple tipping of the first range tipping bucket; and the device further comprises: The correction module is configured to determine a series of event occurrence rates of the first range tipping bucket based on the sequence of tipping times; Determine a dynamic calibration coefficient for correcting the target rainfall by using a preset measurement event response spectrum model based on the series of event occurrence rates; and correct the target rainfall by using the dynamic calibration coefficient to obtain a final rainfall; The measurement event response spectrum model comprises at least one effective response interval and at least one ineffective response interval; and the step of determining the dynamic calibration coefficient for correcting the target rainfall comprises: Matching each event occurrence rate in the series of event occurrence rates with the effective response interval and the ineffective response interval; Assigning a first type calibration factor to the event occurrence rate matched to the effective response interval; Assigning a second type calibration factor to the event occurrence rate matched to the ineffective response interval, wherein the second type calibration factor has a stronger inhibitory effect than the first type calibration factor; Aggregating multiple calibration factors corresponding to the series of event occurrence rates to generate the dynamic calibration coefficient.

7. An electronic device, comprising a processor, a memory and a bus, characterized in that: The memory stores a computer program, and the bus is configured to establish a communication connection between the processor and the memory; The processor is configured to execute the computer program stored in the memory, and implement the method according to any one of claims 1 to 5.

Images