A method and apparatus for correcting three-dimensional ultrasonic wind speed measurement data

Abstract

The application provides a three-dimensional ultrasonic wind speed measurement data correction method and device, computer equipment and storage medium, including obtaining initial measurement data measured by a target measurement device in a preset time period, inputting the initial measurement data into a pre-constructed wind speed network model to obtain an initial wind speed, and inputting the initial measurement data into a pre-constructed wind direction network model to obtain an initial wind direction, using pre-calculated output wind speed time correlation data and a first Pearson correlation coefficient of the wind speed to perform weighted processing on the initial wind speed to obtain a final wind speed, and using pre-calculated output wind direction time correlation data and a second Pearson correlation coefficient of the wind direction to perform weighted processing on the initial wind direction to obtain a final wind direction. The application improves the accuracy of the corrected data by introducing long-time, multi-parameter (such as wind speed, wind direction, rainfall, temperature, humidity, air pressure, etc.) correlation analysis, thereby improving the accuracy of the measured wind speed and wind direction.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, computer equipment, and storage medium for correcting three-dimensional ultrasonic wind speed measurement data. Background Technology

[0002] An automatic weather station is an unattended meteorological data acquisition system that integrates data collection, storage, transmission, and management. Under electronic equipment or computer control, it automatically conducts meteorological observations and data collection and transmission, playing an irreplaceable role in meteorological monitoring and early warning, and disaster prevention. However, traditional automatic weather stations are large and complex to install, making them unsuitable for special applications such as high-speed rail meteorology, shipboard marine meteorology, and mobile emergency observation. Therefore, meteorological departments are vigorously developing miniature automatic weather stations that are small in size, lightweight, low in power consumption, quick to install, and easy to transport and carry.

[0003] On the one hand, currently available miniature automatic weather stations can only provide two-dimensional wind speed and direction measurements in a planar direction. However, two-dimensional wind monitoring is insufficient to meet the demand for refined wind speed and direction information in many situations, especially regarding vertical wind. For example, along high-speed rail lines, excessive vertical wind can cause the overhead power contact network to vibrate, leading to poor contact between the train's pantograph and the contact network, resulting in power supply failures. In overhead transmission lines, excessive vertical wind can cause high-frequency, low-amplitude micro-wind vibrations in the vertical plane. Prolonged exposure to this repeated vibration can reduce the conductor's fatigue strength, leading to wire breakage, hardware wear, and damage to tower components. Furthermore, the resulting fatigue breakage and other accidents are highly concealed and extremely dangerous.

[0004] On the other hand, the time-of-flight method for calculating wind speed based on ultrasound is affected by external factors such as rainfall, temperature, humidity, and air pressure, thus requiring high-standard equipment to correct the data. Traditional correction methods, such as the classic Kalman filter, are only affected by one or two of these states. However, the observation period for a single wind event is often several seconds or more, far exceeding the milliseconds of two data periods (i.e., two states in a Kalman filter). Therefore, the Kalman filter fails to observe the long-term correlation of the entire wind.

[0005] In conclusion, there is an urgent need to propose a new data correction method to solve the above problems. Summary of the Invention

[0006] To address the problems of existing technologies, this invention provides a novel method, apparatus, computer equipment, and storage medium for correcting three-dimensional ultrasonic wind speed measurement data. By introducing long-term, multi-parameter correlation analysis (such as wind speed, wind direction, rainfall, temperature, humidity, and air pressure), the accuracy of the corrected data is improved, thereby enhancing the precision of the measured wind speed and direction.

[0007] To solve one or more of the above-mentioned technical problems, the technical solution adopted in this application is:

[0008] Firstly, a method for correcting three-dimensional ultrasonic wind speed measurement data is provided, the method comprising:

[0009] Acquire initial measurement data obtained by the target measuring device within a preset time period. The initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure.

[0010] The initial measurement data is input into a pre-built wind speed network model to obtain the initial wind speed, and the initial measurement data is input into a pre-built wind direction network model to obtain the initial wind direction;

[0011] The initial wind speed is weighted using the pre-calculated time-related data of output wind speed and the first Pearson correlation coefficient of wind speed to obtain the final wind speed. The initial wind direction is weighted using the pre-calculated time-related data of output wind direction and the second Pearson correlation coefficient of wind direction to obtain the final wind direction.

[0012] In a specific embodiment, the calculation process of the first and second Pearson correlation coefficients includes:

[0013] The target measurement device is used to acquire target measurement data according to a preset rule, and the reference measurement device is used to acquire reference measurement data according to the preset rule. The target measurement data and the reference measurement data both include at least wind speed, wind direction, rainfall, temperature, humidity and air pressure.

[0014] Using the reference measurement data, calculate the first Pearson correlation coefficient between the output wind speed and the wind speed, and the second Pearson correlation coefficient between the output wind speed and the wind direction.

[0015] In a specific embodiment, the construction process of the wind speed network model and the wind direction network model includes:

[0016] Using the target measurement data as input and the wind speed in the reference measurement data as output, a wind speed network model is constructed based on the learning network.

[0017] Using the target measurement data as input and the wind direction in the reference measurement data as output, a wind direction network model is constructed based on the learning network.

[0018] In a specific embodiment, the step of calculating the first Pearson correlation coefficient between the output wind speed and the wind speed, and the second Pearson correlation coefficient between the output wind speed and the wind direction using the reference measurement data includes:

[0019] The reference measurement data is sampled to obtain sampled data;

[0020] Based on the sampled data, calculate the first Pearson correlation coefficient between the output wind speed and the wind speed, and the second Pearson correlation coefficient between the output wind speed and the wind direction.

[0021] In a specific embodiment, the steps of acquiring target measurement data using the target measuring device according to a preset rule and acquiring reference measurement data using the reference measuring device according to the preset rule include:

[0022] The target measurement device is used to collect target measurement data under multiple wind speeds at multiple different wind angles within a preset time period, and the reference measurement device is used to collect reference measurement data under multiple wind speeds at multiple different wind angles within the preset time period.

[0023] In one specific embodiment, the preset time period is not less than 10 minutes.

[0024] In one specific embodiment, the wind speed network model and / or the wind direction network model includes the LRCN network model.

[0025] Secondly, corresponding to the above method, this application also provides a correction device for three-dimensional ultrasonic wind speed measurement data, the device comprising:

[0026] The acquisition module is used to acquire initial measurement data obtained by the target measuring device within a preset time period. The initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure.

[0027] The calculation module is used to input the initial measurement data into a pre-built wind speed network model to obtain the initial wind speed, and to input the initial measurement data into a pre-built wind direction network model to obtain the initial wind direction;

[0028] The correction module is used to perform weighted processing on the initial wind speed using the pre-calculated output wind speed time-related data and the first Pearson correlation coefficient of the wind speed to obtain the final wind speed, and to perform weighted processing on the initial wind direction using the pre-calculated output wind direction time-related data and the second Pearson correlation coefficient of the wind direction to obtain the final wind direction;

[0029] The calculation process for the first and second Pearson correlation coefficients includes:

[0030] The target measurement device is used to acquire target measurement data according to a preset rule, and the reference measurement device is used to acquire reference measurement data according to the preset rule. The target measurement data and the reference measurement data both include at least wind speed, wind direction, rainfall, temperature, humidity and air pressure.

[0031] Using the reference measurement data, calculate the first Pearson correlation coefficient between the output wind speed and the wind speed, and the second Pearson correlation coefficient between the output wind speed and the wind direction.

[0032] Thirdly, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and when the computer program is executed by the processor, a method for correcting the three-dimensional ultrasonic wind speed measurement data is implemented.

[0033] Fourthly, a computer-readable storage medium is also provided, wherein a computer program is stored therein, and when the computer program is executed, a method for correcting the three-dimensional ultrasonic wind speed measurement data is implemented.

[0034] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0035] The present invention provides a method, apparatus, computer equipment, and storage medium for correcting three-dimensional ultrasonic wind speed measurement data. The method includes acquiring initial measurement data obtained by a target measuring device within a preset time period. The initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure. The initial measurement data is input into a pre-constructed wind speed network model to obtain an initial wind speed, and the initial measurement data is input into a pre-constructed wind direction network model to obtain an initial wind direction. The initial wind speed is weighted using a first Pearson correlation coefficient between pre-calculated output wind speed time-related data and wind speed to obtain a final wind speed. The initial wind direction is weighted using a second Pearson correlation coefficient between pre-calculated output wind direction time-related data and wind direction to obtain a final wind direction. This application improves the accuracy of the corrected data by introducing long-term, multi-parameter (such as wind speed, wind direction, rainfall, temperature, humidity, air pressure, etc.) correlation analysis, thereby improving the accuracy of the measured wind speed and wind direction. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the hardware structure of the three-dimensional ultrasonic anemometer provided in an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of the external structure of the three-dimensional ultrasonic anemometer provided in an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of the structure of the rain sensor in the three-dimensional ultrasonic anemometer provided in this embodiment of the invention;

[0040] Figure 4 This is a hardware structure diagram of the piezoelectric rain sensor provided in an embodiment of the present invention;

[0041] Figure 5 This is a hardware circuit architecture diagram of the three-dimensional ultrasonic anemometer provided in the embodiment of the present invention;

[0042] Figure 6 This is a schematic diagram of the installation of a set of ultrasonic probes in the three-dimensional ultrasonic anemometer provided in this embodiment of the invention;

[0043] Figure 7 This is a schematic diagram of the vertical projection of three wind speeds in the horizontal direction in the three-dimensional ultrasonic wind speed measurement method provided in the embodiment of the present invention;

[0044] Figure 8 This is a flowchart of the method for correcting three-dimensional ultrasonic wind speed measurement data provided in an embodiment of the present invention;

[0045] Figure 9 This is a schematic diagram of the structure of the calibration device for three-dimensional ultrasonic wind speed measurement data provided in an embodiment of the present invention;

[0046] Figure 10 This is an architecture diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] As described in the background section, existing correction methods, such as the classic Kalman filter, take into account the influence of one or two of the external factors such as rainfall, temperature, humidity, and air pressure. However, the observation period of a single wind is often several seconds or more, which is often far beyond the millisecond level of two data periods (i.e., two states in the Kalman filter). Therefore, the long-term correlation of the entire wind is not observed.

[0049] To address one or more of the aforementioned problems, this application presents a novel method for correcting three-dimensional ultrasonic wind speed measurement data. Compared to existing correction methods, this method improves the accuracy of the corrected data by introducing long-term, multi-parameter correlation analysis (such as wind speed, wind direction, rainfall, temperature, humidity, and air pressure), thereby enhancing the precision of the measured wind speed and direction.

[0050] The solution of this application will now be described in detail with reference to the accompanying drawings and various embodiments.

[0051] Example 1

[0052] To achieve the above method, this application provides a three-dimensional ultrasonic anemometer (i.e., target measuring device) capable of measuring three-dimensional wind, referring to... Figure 1 As shown, in terms of hardware, this three-dimensional ultrasonic anemometer adopts a design scheme with one master controller and one slave controller. As a preferred example, in this embodiment, the master controller can be a microcomputer host and an STM32F103 series chip, and the slave controller can also be an STM32F103 series chip. The microcomputer host in the master controller is mainly used for data correction, the STM32F103 series chip in the master controller is mainly used for processing temperature, humidity, pressure data, and wind speed and direction data, and the slave controller is mainly used for processing rainfall data and sending the processed rainfall data to the master controller for unified output. Further reference... Figure 1 As shown, in actual deployment, the hardware of the 3D ultrasonic anemometer also includes an SPI interface, which will not be elaborated here.

[0053] The external structure of the three-dimensional ultrasonic anemometer provided in this application embodiment is referenced. Figure 2As shown, it mainly includes a rain sensor, multiple ultrasonic probes, a temperature and humidity sensor, and a barometric pressure sensor. Further reference... Figure 2 As shown, the rain sensor 7 can be a piezoelectric rain sensor, which is installed on the top of the three-dimensional ultrasonic anemometer; there are 6 ultrasonic probes, and each pair is arranged opposite to each other; the temperature and humidity sensor includes, but is not limited to, the SHT75, and the air pressure sensor includes, but is not limited to, the MS5540C.

[0054] Further reference Figure 2 As shown, the six ultrasonic probes are arranged in three groups at a 120° angle in space, and each probe support has a fixed angle θ with the central vertical axis of the entire instrument housing. Ultrasonic probes (also called transducers) 1 and 4 form one group, ultrasonic probes 2 and 5 form another group, and ultrasonic probes 3 and 6 form yet another group.

[0055] The three-dimensional ultrasonic anemometer provided in this application involves multiple power supplies with different voltage values. In specific implementations, the SHT75 temperature and humidity sensor, MS5540C pressure sensor, and STM32F103 all support 3.3V voltage, the integrated operational amplifier in the amplifier circuit supports ±5V, and the ultrasonic wind speed and direction sensor supports 12V voltage. This application embodiment can use a unified 24V power supply input, and then convert the input voltage to 3.3V, 5V, and 12V respectively to power different components. Specifically, the 24V to 12V voltage regulator circuit can be designed using the LM317BT power regulator chip, the 24V to ±5V power supply circuit can be designed using the K78X5-1000(L) chip, and the 5V to 3.3V power supply circuit can be designed using the AMS1117-3.3 conversion chip, etc., which will not be elaborated further here.

[0056] The rain sensor in this embodiment is a piezoelectric rain sensor, referring to... Figure 3 As shown, the piezoelectric rain sensor includes a piezoelectric ceramic plate and an arc-shaped steel cover fixed together. The piezoelectric ceramic plate is positioned above the arc-shaped steel cover. When raindrops impact the arc-shaped steel cover, the impact force is transmitted to the piezoelectric ceramic plate. Under the influence of this force, the piezoelectric ceramic plate becomes polarized, causing charge movement. Positive and negative electrodes are drawn from the piezoelectric ceramic plate to output charge. In this embodiment, the control chip for the piezoelectric rain sensor can be an STM32F103ZET6, and the peripheral circuitry can be designed to form a minimum system.

[0057] Reference Figure 4As shown, in terms of hardware, the piezoelectric rain sensor in this embodiment mainly uses a charge-to-voltage signal conversion circuit, a normalization amplifier circuit, a filtering circuit, a detector follower circuit, and an AD conversion circuit to acquire voltage. All related circuits can be implemented using existing circuits in the prior art, and will not be described in detail here. It should be noted that, considering the extremely weak charge signal generated by the piezoelectric rain sensor, it is not convenient to acquire directly. The charge signal needs to be converted into a proportional voltage signal to facilitate filtering and analog-to-digital conversion in subsequent circuits. Therefore, this embodiment uses a charge amplifier as the charge-to-voltage signal conversion method. The normalization amplifier circuit in the piezoelectric rain sensor section ensures that the voltage output is consistent after normalization amplification when different piezoelectric sensors are used, thus increasing the circuit's practicality and versatility. The filtering circuit can use a combination of active low-pass filtering and passive high-pass filtering to form a bandpass filter with a bandpass frequency of 1Hz to 1kHz. The detector circuit in this embodiment is mainly used to cut off the negative half-axis signal of the voltage signal, completely preserving the waveform of the positive half-axis. In a specific design, diodes in opposite directions can be used to form a rectifier circuit to cut off the waveform of the upper half-axis. Then, the waveform of the lower half-axis can be reversed to the upper half-axis through voltage reversal, facilitating stable acquisition via AD conversion. Since the acquired voltage signals are all analog signals, AD conversion is required to convert the analog quantity into a digital quantity. In this embodiment, the internal ADC of an STM32 chip can be used, with a sampling frequency of 50000Hz, which can capture weak signals.

[0058] The main controller of the 3D ultrasonic anemometer uses the STM32F103 series chip. The peripheral circuitry is drawn to form a minimum system, referencing... Figure 5 As shown, its hardware circuit mainly includes an ultrasonic transmitting circuit, an ultrasonic receiving circuit, a communication circuit, and a power supply circuit.

[0059] In one specific embodiment, the ultrasonic signal transmitting circuit mainly consists of two parts: a transmitting signal gating circuit and an ultrasonic transducer driving circuit. Specifically, a 74F805 chip can be used to design a six-channel transmitting signal selection switch circuit. This chip has six inputs and is a NOR driver integrating six NOR gates. Six-channel gating and transmitting can be achieved by controlling the high and low levels of the pins connecting the 74F805 chip to the main controller. The transmitting signal gating circuit can use relevant circuits in the prior art, which will not be elaborated here. The ultrasonic transducer driving circuit can use a BOOST boost method, and the driving boost circuit can use relevant circuits in the prior art, which will not be elaborated here. In this embodiment, the ultrasonic transducer is a transceiver integrated design. When the ultrasonic transducer is used as a receiver, the received ultrasonic signal will enter the subsequent signal processing circuit.

[0060] The ultrasonic signal receiving circuit is mainly used to process the received ultrasonic signal. In a specific embodiment, it can be divided into five parts: a receiving signal gating circuit, a signal amplification circuit, a signal filtering circuit, a rectification and inversion circuit, and a signal comparison circuit. As a preferred example, a CD4051 can be used as the gating switch in the ultrasonic signal receiving circuit. The signal amplification circuit can use a 5532 operational amplifier chip as the operational amplifier. The signal filtering circuit can set the low-pass filter cutoff frequency to 220kHz and the high-pass filter cutoff frequency to 180kHz, and the two filters combined form a bandpass filter with a center frequency of 200kHz and a bandwidth of 40kHz. The rectification and inversion circuit is included to facilitate the conversion of the sine wave into a square wave signal that the microcontroller can directly recognize in the later circuit. It can cut off the signal on the negative half-axis of the sine wave signal for conversion by the subsequent comparison circuit, and then invert the signal to the positive half-axis through the voltage inversion circuit. The signal comparison circuit is designed to facilitate the microcontroller's recognition of ultrasonic signals. It transforms the sinusoidal analog signal on the positive half-axis into a regular square wave signal with a frequency consistent with the center frequency of the ultrasonic signal. In practice, a comparator can be used for signal conversion. A voltage comparator can convert a continuous analog signal into a discrete digital level signal with only high and low states.

[0061] The temperature and humidity sensor is mainly used to collect temperature and humidity data. In this embodiment, Sensirion's SHT75 digital temperature and humidity sensor can be used. The barometric pressure sensor can be Intersema's MS5540C sensor.

[0062] The communication circuit in this embodiment can be designed to support both RS485 long-distance communication and RS232 short-distance communication modes. These two communication methods can also be integrated into a DTU. By inserting a SIM card with GPRS functionality into the DTU, serial port data can be converted into IP data and uploaded to the data terminal via the SIM card network, which is very convenient.

[0063] When rainfall occurs, raindrops impact the curved steel cover on the surface of the piezoelectric rain sensor. The steel cover transmits the impact force to the piezoelectric ceramic plate, which is polarized, generating a charge signal proportional to the impact force. The weak charge value is converted into a voltage value that is easy to measure through a charge-to-voltage signal conversion circuit. The voltage signal is then amplified by a normalization circuit, filtered, and subjected to detection and follow-up processing. Finally, the analog signal is converted into a digital signal through an AD conversion circuit and sent to the main control chip for data processing.

[0064] In the three-dimensional ultrasonic anemometer section, the main controller transmits a 3.3V 200kHz PWM wave signal, which, through the transmit gating circuit, reaches the ultrasonic transducer drive boost circuit to drive the ultrasonic transducer to emit ultrasonic signals. After receiving the ultrasonic signal, the echo signal sequentially passes through the receive gating circuit, signal amplification circuit, rectification and inversion circuit, filter circuit, and comparison circuit. Finally, it is identified by the main controller, triggering a timer interrupt to calculate the propagation time of one ultrasonic wave. Once the propagation times of all six ultrasonic waves are calculated, the actual wind speed and direction can be determined. The three-dimensional ultrasonic anemometer in this invention establishes a wind measurement scheme using the time difference method and a non-orthogonal array.

[0065] Corresponding to the aforementioned three-dimensional ultrasonic anemometer, this application embodiment also provides a three-dimensional ultrasonic wind speed measurement method. Assume that one set of ultrasonic probes (also called transducers) in the aforementioned three-dimensional ultrasonic anemometer are transducer C and transducer D, with the installation method referring to... Figure 6 As shown. Assume the wind speed along the straight line between transducers C and D is V, and the distance between transducers C and D is L, meaning the propagation distance of the ultrasonic wave between transducers C and D is L. If the time it takes for the ultrasonic wave to travel from transducer C to transducer D is t... cd The time taken from when the ultrasonic wave is emitted by transducer D to when it is received by transducer C is t. dc In air, the speed of ultrasonic waves in a windless state is V0. Taking the direction of the ultrasonic signal propagating from transducer C to transducer D as the positive direction, the wind speed V in the CD direction can be obtained according to the time-of-flight method. CD for:

[0066]

[0067] Similarly, this method is used to calculate the wind speed V1 along the line containing ultrasonic transducers 1 and 4, the wind speed V2 along the line containing ultrasonic transducers 2 and 5, and the wind speed V3 along the line containing ultrasonic transducers 3 and 6. After calculation, these three wind speeds are vertically projected onto the horizontal direction, and one of them is aligned with the true north direction of the geographic coordinate system, such as... Figure 7 As shown, the horizontal wind speed Vx perpendicular to due north, the horizontal wind speed Vy parallel to due north, and the vertical wind speed Vz can be calculated respectively:

[0068]

[0069]

[0070] V z =V1 cosθ + V2 cosθ + V3 cosθ

[0071] After determining Vx, Vy, and Vz, the final actual wind speed VT and horizontal wind angle β can be calculated:

[0072]

[0073]

[0074] The wind direction angle is further determined based on the horizontal wind speed value: First, the following formula is used... After obtaining the absolute value of the wind direction angle, the final wind direction angle is determined based on the values ​​of the horizontal wind speeds Vx and Vy. In practice, the X and Y coordinate systems in the horizontal plane can be divided into four quadrants. When Vx > 0 and Vy > 0, the wind direction angle is in the first quadrant; when Vx < 0 and Vy > 0, it is in the second quadrant; when Vx < 0 and Vy < 0, it is in the third quadrant; and when Vx > 0 and Vy < 0, it is in the fourth quadrant. The specific wind direction angle is determined using this method, which will not be elaborated further here.

[0075] Corresponding to the above-mentioned three-dimensional ultrasonic anemometer and three-dimensional ultrasonic wind speed measurement method, this application embodiment also provides a correction method for three-dimensional ultrasonic wind speed measurement data, so as to take into account the influence of factors such as wind speed, wind direction, rainfall, temperature, humidity, and air pressure on the three-dimensional ultrasonic anemometer during the measurement process, improve the accuracy of the corrected data, and thus improve the accuracy of the measured wind speed and wind direction.

[0076] In practical implementation, the correction method provided in this application mainly includes three steps: establishing a dataset, establishing a model and learning, and correcting the output. The algorithm can be deployed on the main controller and executed using a microcomputer host.

[0077] 1. Create a dataset

[0078] A set of the aforementioned three-dimensional ultrasonic anemometers A and a set of high-standard miniature automated weather stations B are combined to form a learning and comparison kit. Using 10 such learning and comparison kits, experiments are conducted according to the China Meteorological Administration's standard QX / T568-2020. In this embodiment of the application, the recorded data during the experimental operation must at least meet the following requirements:

[0079] (1) The wind direction is arbitrarily spherical, so it is a sphere. In the embodiment of this application, the data is recorded as longitude 360 ​​degrees and latitude 360 ​​degrees, that is, the possible direction angles are 360*360=129600.

[0080] (2) At each azimuth angle, at each of the following 13 wind speed levels, at least 100 data points of 10 minutes in length must be recorded, and at least 5 minutes of these 10 minutes of data must be data points of the following 13 levels.

[0081] Level 0: Calm wind 0-0.2, smoke rises straight up;

[0082] Level 1: Gentle wind 0.3-1.5. Smoke can indicate wind direction, but a wind vane cannot indicate wind direction.

[0083] Level 2: Light breeze 1.6-3.3. You can feel the wind on your face, the leaves rustle slightly, and a regular wind vane can move with the wind.

[0084] Level 3: Light breeze 3.4-5.4, leaves and twigs swaying incessantly, flags unfurling;

[0085] Level 4: Gentle breeze 5.5-7.9, dust and shredded paper are stirred up, and small branches sway.

[0086] Level 5: Fresh and strong winds, 8.0-10.7. Small leafy trees begin to sway, and waves form on inland waters.

[0087] Level 6: Strong winds of 10.8-13.8, large tree branches swaying, power lines making a whooshing sound, making it difficult to open an umbrella;

[0088] Level 7: Gale-force winds of 13.9-17.1, causing the entire tree to sway and making walking against the wind difficult;

[0089] Level 8: Gale-force winds of 17.2-20.7, causing tree branches to break and hindering travel;

[0090] Level 9: Gale-force winds of 20.8-24.4, causing minor building damage (chimney pipes and roof tiles blown off);

[0091] Level 10: Gale-force winds of 24.5-28.4 (rare inland), which uproot trees and damage numerous buildings when encountered.

[0092] Level 11: Storms with a power level of 28.5-32.6 are extremely rare and are accompanied by widespread destruction;

[0093] Category 12: Hurricane ≥ 32.7.

[0094] To quickly achieve the above data results, in this embodiment of the application, a set of common 3R-type robotic arms can be used to fix a set of the above-mentioned three-dimensional ultrasonic anemometers A at the end of the 3R-type robotic arms. Under stable wind speed and wind direction in the external environment, the angle of the above-mentioned three-dimensional ultrasonic anemometers A can be adjusted to quickly obtain 100 sets of data from 129,600 different angles.

[0095] In this way, the measurement data DA of the three-dimensional ultrasonic anemometer A and the measurement data DB of the high-standard micro-automatic weather station are obtained respectively.

[0096] 2. Model building and learning

[0097] A multi-parameter fusion scheme is adopted to perform multi-parameter fusion. The multi-parameters include: wind speed, wind direction, rainfall, temperature, humidity, and air pressure. The entire process consists of two steps: calculating weights based on the Pearson correlation coefficient and model learning.

[0098] 2.1 Using DB data, calculate the Pearson correlation coefficients between the five parameters (wind direction, rainfall, temperature, humidity, and air pressure) and wind speed according to the following formula (1) when outputting wind speed. Where P represents the Pearson correlation coefficient, v represents wind speed, 1 represents level 1 data, pm is any integer from 1 to 6, and from 1 to 6 represent wind speed, wind direction, rainfall, temperature, humidity, and air pressure, respectively. The Pearson correlation coefficient between itself (such as wind speed and wind speed) is calculated as 1.

[0099]

[0100] In the above formula (1), X and Y represent the parameter data being compared, and n is the total amount of data in the DB. When pm is 2, X is the wind speed in the DB data and Y is the wind direction in the DB data; when pm is 3, X is the wind speed in the DB data and Y is the rainfall in the DB data. These data are data with the same time label. If the time resolution in the DB is 100 milliseconds, then the amount of data is: 360 (angle) * 360 (angle) * 13 (wind level) * 100 (groups) * 10 (minutes) * 60 (seconds) * 10 (10 100 milliseconds per second) = 101088 * 1000000 data points.

[0101] Because the amount of data is very large, if computing resources are insufficient, sampling calculations can be performed, such as reducing 100 sets to 10 sets, 60 seconds to 1 second, 10 to 1, etc.

[0102] After obtaining the original 6 Then, using a normalization method, the six... Transform into Here, 2 represents level 2 data, such that:

[0103]

[0104] 2.2 Using DB data, calculate the Pearson correlation coefficients between the five parameters (wind direction, rainfall, temperature, humidity, and air pressure) and wind direction when outputting wind speed according to the formula (1) above. Then, use the normalization method to convert them into... Where P represents the Pearson correlation coefficient, and ang represents the wind direction, such that:

[0105]

[0106] 2.3 Constructing a learning network using DA and DB data

[0107] 2.3.1 For the wind speed network, the network is a pm-group LRCN network. Its inputs are wind speed, wind direction, rainfall, temperature, humidity, and air pressure from the DA data, and its output is wind speed from the DB data. With a time resolution of 100 milliseconds in the DB, 10 minutes of data includes 10*60*10 = 6000 data points. This embodiment uses a single-parameter input, 1000 bidirectional GRUs per layer, and a two-layer structure to form the LRCN network. This allows for the discovery of the correlation between various DA parameters and DB wind speed data over 100 seconds. Through data comparison and learning, pm well-learned LRCN networks are obtained. The internal parameters of these networks constitute knowledge. These trained networks are programmed into the microcomputer host in the main controller of the aforementioned 3D ultrasonic anemometer hardware architecture for use during calibration. The structure of a single LRCN network is well-known and will not be described further here.

[0108] 2.3.2 For the wind direction network, similar to 2.3.1, a pm-group LRCN network is designed. Its inputs are wind speed, wind direction, rainfall, temperature, humidity, and air pressure from the DA data, and its output is wind direction from the DB data. With a time resolution of 100 milliseconds in the DB, 10 minutes of data includes 10*60*10 = 6000 data points. This application uses a single-parameter input, 1000 bidirectional GRUs per layer, and a two-layer structure to form the LRCN network. This allows for the discovery of the correlation between various DA parameters and DB wind direction data over 100 seconds. Through data comparison and learning, pm well-trained LRCN networks are obtained. The internal parameters of these networks constitute knowledge. These trained networks are programmed into the microcomputer host in the main controller of the aforementioned three-dimensional ultrasonic anemometer hardware architecture for use in calibration.

[0109] 3. Calibrate output

[0110] When calibrating the measurement data, after obtaining 10 minutes of initial data using the aforementioned three-dimensional ultrasonic anemometer, the microcomputer host in the main controller begins to calibrate and output the relevant data.

[0111] For wind speed, input the initial parameters for the 10-minute period into step 2.3.1 respectively to obtain the results. A set of data was obtained Where 0 represents the output value and 1 represents level 1 data.

[0112] Then put As corresponding to all The weights are shown in the formula below, and the weighted output yields the true output value O. v .

[0113]

[0114] Regarding wind direction, input the initial parameters for the 10-minute period into step 2.3.2 respectively to obtain the results. A set of data was obtained Where 0 represents the output value and 1 represents level 1 data.

[0115] Then put As corresponding to all The weights are shown in Equation 11, and the weighted output yields the true output value O. ang .

[0116]

[0117] Example 2

[0118] Corresponding to Embodiment 1 above, this application provides a method for correcting three-dimensional ultrasonic wind speed measurement data. In this embodiment, content that is the same as or similar to that in Embodiment 1 above can be referred to the above description and will not be repeated hereafter. (Refer to...) Figure 8 As shown, the method includes the following steps:

[0119] S101: Obtain initial measurement data obtained by the target measuring device within a preset time period, wherein the initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure;

[0120] S102: Input the initial measurement data into a pre-built wind speed network model to obtain the initial wind speed, and input the initial measurement data into a pre-built wind direction network model to obtain the initial wind direction;

[0121] S103: The initial wind speed is weighted using the pre-calculated output wind speed time-related data and the first Pearson correlation coefficient of the wind speed to obtain the final wind speed; the initial wind direction is weighted using the pre-calculated output wind direction time-related data and the second Pearson correlation coefficient of the wind direction to obtain the final wind direction.

[0122] Specifically, the target measuring device includes, but is not limited to, the three-dimensional ultrasonic anemometer provided in Embodiment 1 above. Setting the initial measurement data acquisition time to a preset time period is mainly to introduce long-term data and avoid failing to observe the long-term correlation of the entire wind due to excessively short observation time. The preset time period can be set according to actual observation needs, and no specific limitation is made here.

[0123] As a preferred implementation method, in this embodiment of the application, the calculation process of the first and second Pearson correlation coefficients includes:

[0124] The target measurement device is used to acquire target measurement data according to a preset rule, and the reference measurement device is used to acquire reference measurement data according to the preset rule. The target measurement data and the reference measurement data both include at least wind speed, wind direction, rainfall, temperature, humidity and air pressure.

[0125] Using the reference measurement data, calculate the first Pearson correlation coefficient between the output wind speed and the wind speed, and the second Pearson correlation coefficient between the output wind speed and the wind direction.

[0126] As a preferred implementation method, in this embodiment of the application, the construction process of the wind speed network model and the wind direction network model includes:

[0127] Using the target measurement data as input and the wind speed in the reference measurement data as output, a wind speed network model is constructed based on the learning network.

[0128] Using the target measurement data as input and the wind direction in the reference measurement data as output, a wind direction network model is constructed based on the learning network.

[0129] As a preferred implementation, in this embodiment of the application, the step of calculating the first Pearson correlation coefficient between the output wind speed and the wind speed, and the second Pearson correlation coefficient between the output wind speed and the wind direction using the reference measurement data, includes:

[0130] The reference measurement data is sampled to obtain sampled data;

[0131] Based on the sampled data, calculate the first Pearson correlation coefficient between the output wind speed and the wind speed, and the second Pearson correlation coefficient between the output wind speed and the wind direction.

[0132] As a preferred implementation, in this embodiment of the application, the step of acquiring target measurement data using the target measuring device according to a preset rule and acquiring reference measurement data using a reference measuring device according to the preset rule includes:

[0133] The target measurement device is used to collect target measurement data under multiple wind speeds at multiple different wind angles within a preset time period, and the reference measurement device is used to collect reference measurement data under multiple wind speeds at multiple different wind angles within the preset time period.

[0134] In a preferred embodiment of this application, the preset time period is not less than 10 minutes.

[0135] As a preferred implementation, in this embodiment of the application, the wind speed network model and / or the wind direction network model includes the LRCN network model.

[0136] Example 3

[0137] Corresponding to Embodiments 1 and 2 above, this application also provides a calibration device for three-dimensional ultrasonic wind speed measurement data. In this embodiment, content that is the same as or similar to Embodiments 1 or 2 above can be referred to the above description and will not be repeated hereafter. (Refer to...) Figure 9 As shown, the device includes:

[0138] The acquisition module is used to acquire initial measurement data obtained by the target measuring device within a preset time period. The initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure.

[0139] The calculation module is used to input the initial measurement data into a pre-built wind speed network model to obtain the initial wind speed, and to input the initial measurement data into a pre-built wind direction network model to obtain the initial wind direction;

[0140] The correction module is used to perform weighted processing on the initial wind speed using the pre-calculated time-related data of the output wind speed and the first Pearson correlation coefficient of the wind speed to obtain the final wind speed, and to perform weighted processing on the initial wind direction using the pre-calculated time-related data of the output wind direction and the second Pearson correlation coefficient of the wind direction to obtain the final wind direction.

[0141] In some embodiments of this application, the three-dimensional ultrasonic wind speed measurement data correction device can also implement the steps corresponding to the method described in Embodiment 1 or 2 when implementing the data correction method. Please refer to the detailed description in Embodiment 1 or 2, which will not be repeated here.

[0142] It should be noted that the three-dimensional ultrasonic wind speed measurement data correction device provided in the above embodiments is only illustrated by the division of the above functional modules when triggering the data correction service. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. In addition, the three-dimensional ultrasonic wind speed measurement data correction device and the three-dimensional ultrasonic wind speed measurement data correction method embodiment provided in the above embodiments belong to the same concept, that is, the device is based on the three-dimensional ultrasonic wind speed measurement data correction method, and its specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0143] Example 4

[0144] Corresponding to embodiments one to three above, the present invention also provides a computer device, including: a processor and a memory, wherein the memory stores a computer program that can run on the processor, and when the computer program is executed by the processor, it executes the correction method for three-dimensional ultrasonic wind speed measurement data provided in any of the above embodiments.

[0145] in, Figure 10 An exemplary computer device 1500 is shown, which may specifically include a processor 1510, a video display adapter 1511, a disk drive 1512, an input / output interface 1513, a network interface 1514, and a memory 1520. The processor 1510, video display adapter 1511, disk drive 1512, input / output interface 1513, network interface 1514, and memory 1520 can communicate with each other via a communication bus 1530.

[0146] The processor 1510 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided by the present invention.

[0147] The memory 1520 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1520 can store the operating system 1521 for controlling the operation of the electronic device, and the basic input / output system (BIOS) 1522 for controlling the low-level operations of the electronic device. Additionally, it can store a web browser 1523, a data storage management system 1524, and a device identification information processing system 1525, etc. The aforementioned device identification information processing system 1525 can be the application program that specifically implements the aforementioned steps in this embodiment of the invention. In summary, when the technical solution provided by this invention is implemented through software or firmware, the relevant program code is stored in the memory 1520 and is called and executed by the processor 1510.

[0148] Input / output interface 1513 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0149] Network interface 1514 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0150] The bus includes a pathway for transmitting information between various components of the device (e.g., processor 1510, video display adapter 1511, disk drive 1512, input / output interface 1513, network interface 1514, and memory 1520).

[0151] In addition, the electronic device can also obtain information on specific claim conditions from the virtual resource object claim condition information database for condition judgment, and so on.

[0152] It should be noted that although the above-described device only shows the processor 1510, video display adapter 1511, disk drive 1512, input / output interface 1513, network interface 1514, memory 1520, bus, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the present invention, and not necessarily all the components shown in the figures.

[0153] Example 5

[0154] Corresponding to embodiments one to four above, this embodiment of the invention also provides a computer-readable storage medium. In this embodiment, the contents that are the same as or similar to those in embodiments one to four above can be referred to the above description and will not be repeated hereafter.

[0155] The computer-readable storage medium has a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0156] Acquire initial measurement data obtained by the target measuring device within a preset time period. The initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure.

[0157] The initial measurement data is input into a pre-built wind speed network model to obtain the initial wind speed, and the initial measurement data is input into a pre-built wind direction network model to obtain the initial wind direction;

[0158] The initial wind speed is weighted using the pre-calculated time-related data of output wind speed and the first Pearson correlation coefficient of wind speed to obtain the final wind speed. The initial wind direction is weighted using the pre-calculated time-related data of output wind direction and the second Pearson correlation coefficient of wind direction to obtain the final wind direction.

[0159] In some implementations of this application, when the computer program is executed by the processor, it can also implement the steps corresponding to the method described in Embodiment 1 or 2. Please refer to the detailed description in Embodiment 1 or 2, which will not be repeated here.

[0160] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0161] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0162] The technical solution provided by this invention has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. Furthermore, those skilled in the art will recognize that, based on the ideas of this invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for correcting three-dimensional ultrasonic anemometer data, characterized in that, The method includes: Acquire initial measurement data obtained by the target measuring device within a preset time period. The initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure. The initial measurement data is input into a pre-built wind speed network model to obtain the initial wind speed, and the initial measurement data is input into a pre-built wind direction network model to obtain the initial wind direction; The initial wind speed is weighted using the pre-calculated time-related data of output wind speed and the first Pearson correlation coefficient of wind speed to obtain the final wind speed; the initial wind direction is weighted using the pre-calculated time-related data of output wind direction and the second Pearson correlation coefficient of wind direction to obtain the final wind direction. The calculation process for the first and second Pearson correlation coefficients includes: The target measurement device is used to acquire target measurement data according to a preset rule, and the reference measurement device is used to acquire reference measurement data according to the preset rule. The target measurement data and the reference measurement data both include at least wind speed, wind direction, rainfall, temperature, humidity and air pressure. The method of calculating the first Pearson correlation coefficient between the output wind speed data and the wind speed, and the second Pearson correlation coefficient between the output wind direction data and the wind direction using the reference measurement data includes: sampling the reference measurement data to obtain sampled data; and calculating the first Pearson correlation coefficient between the output wind speed data and the wind speed, and the second Pearson correlation coefficient between the output wind direction data and the wind direction based on the sampled data. The construction process of the wind speed network model and the wind direction network model includes: Using the target measurement data as input and the wind speed in the reference measurement data as output, a wind speed network model is constructed based on the learning network. Using the target measurement data as input and the wind direction in the reference measurement data as output, a wind direction network model is constructed based on the learning network.

2. The method for correcting three-dimensional ultrasonic wind speed measurement data according to claim 1, characterized in that, The steps of acquiring target measurement data using the target measurement device according to a preset rule and acquiring reference measurement data using the reference measurement device according to the preset rule include: The target measurement device is used to collect target measurement data under multiple wind speeds at multiple different wind angles within a preset time period, and the reference measurement device is used to collect reference measurement data under multiple wind speeds at multiple different wind angles within the preset time period.

3. The method for correcting three-dimensional ultrasonic anemometer data according to claim 1, characterized in that, The preset time period is no less than 10 minutes.

4. The method for correcting three-dimensional ultrasonic anemometer data according to claim 1, characterized in that, The wind speed network model and / or the wind direction network model include the LRCN network model.

5. A calibration device for three-dimensional ultrasonic anemometer data, characterized in that, The device includes at least: The acquisition module is used to acquire initial measurement data obtained by the target measuring device within a preset time period. The initial measurement data includes at least wind speed, wind direction, rainfall, temperature, humidity, and air pressure. The calculation module is used to input the initial measurement data into a pre-built wind speed network model to obtain the initial wind speed, and to input the initial measurement data into a pre-built wind direction network model to obtain the initial wind direction; The correction module is used to perform weighted processing on the initial wind speed using the pre-calculated output wind speed time-related data and the first Pearson correlation coefficient of the wind speed to obtain the final wind speed, and to perform weighted processing on the initial wind direction using the pre-calculated output wind direction time-related data and the second Pearson correlation coefficient of the wind direction to obtain the final wind direction; The calculation process for the first and second Pearson correlation coefficients includes: The target measurement device is used to acquire target measurement data according to a preset rule, and the reference measurement device is used to acquire reference measurement data according to the preset rule. The target measurement data and the reference measurement data both include at least wind speed, wind direction, rainfall, temperature, humidity and air pressure. The method of calculating the first Pearson correlation coefficient between the output wind speed data and the wind speed, and the second Pearson correlation coefficient between the output wind direction data and the wind direction using the reference measurement data includes: sampling the reference measurement data to obtain sampled data; and calculating the first Pearson correlation coefficient between the output wind speed data and the wind speed, and the second Pearson correlation coefficient between the output wind direction data and the wind direction based on the sampled data. The construction process of the wind speed network model and the wind direction network model includes: Using the target measurement data as input and the wind speed in the reference measurement data as output, a wind speed network model is constructed based on the learning network. Using the target measurement data as input and the wind direction in the reference measurement data as output, a wind direction network model is constructed based on the learning network.

6. A computer device, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and when the computer program is executed by the processor, it implements the method for correcting three-dimensional ultrasonic wind speed measurement data according to any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed, it implements the method for correcting three-dimensional ultrasonic wind speed measurement data as described in any one of claims 1 to 4.

Images