An uncertainty evaluation method and device based on adaptive Monte Carlo method

Abstract

The application discloses a kind of based on the uncertainty evaluation method and device of adaptive Monte Carlo method, it is related to uncertainty evaluation technical field, the method includes: the initial test sample size of the uncertainty of three-dimensional positive ultrasonic anemometer to be measured is respectively obtained;And the execution number of the uncertainty of three-dimensional positive ultrasonic anemometer to be measured is evaluated;Determine the target test sample size of current execution number and total test sample size;Sample value based on target test sample size is obtained by preset rule to obtain the uncertainty evaluation result of current three-dimensional positive ultrasonic anemometer to be measured, and the standard deviation of the uncertainty evaluation result of current three-dimensional positive ultrasonic anemometer to be measured is obtained;When judging that the standard deviation of evaluation result is all less than preset numerical tolerance, then according to total test sample size determine final evaluation result.The application solves the problem of low efficiency in the prior art when evaluating the uncertainty of three-dimensional positive ultrasonic anemometer.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of uncertainty evaluation, in particular to a Monte Carlo method based uncertainty evaluation method and device. BACKGROUND

[0002] In the measurement process, all measurement results inevitably have a certain degree of uncertainty. Uncertainty is an important parameter representing the dispersion of measurement results, and is closely related to the measurement results. In order to obtain more accurate and reliable measurement data, it is particularly important to evaluate the uncertainty of the three-dimensional ultrasonic anemometer. Through uncertainty evaluation, the possible error sources in the measurement process can be identified, and then appropriate measures can be taken to reduce these errors and improve the accuracy of measurement.

[0003] In the uncertainty evaluation method, the Monte Carlo method (MCM) is preferred due to its high flexibility and accuracy in dealing with complex models. However, the existing Monte Carlo method for evaluating measurement uncertainty, such as single batch Monte Carlo method and improved single batch Monte Carlo method, has certain limitations. For example, the single batch Monte Carlo method usually requires a larger sample size to ensure the accuracy of the evaluation results. Although this method is accurate, it requires a large amount of calculation and high storage space. Another method uses a more flexible strategy. Initially, a smaller fixed sample size is selected, then data is gradually accumulated by repeatedly executing single batch MCM (each sample space is a fixed sample size), and the process is incremented. After each increment, the output result is checked to see if it has reached statistical stability. Once the stable state is reached, the corresponding evaluation result is output.

[0004] However, the improved single batch Monte Carlo method also has certain limitations. Since it uses a monotonically increasing method, when a large number of cycles are required to reach statistical stability, the recursive depth increases with the increase in the number of cycles, and the consumption of system stack resources also increases. Especially when facing the three-dimensional ultrasonic anemometer, the data required for measurement uncertainty is more than that of other sensors, resulting in low efficiency of uncertainty evaluation of the three-dimensional ultrasonic anemometer. SUMMARY

[0005] Therefore, the purpose of the present application is to provide a Monte Carlo method based uncertainty evaluation method and device, which aims to solve the problem of low efficiency in the prior art when evaluating the three-dimensional positive ultrasonic anemometer.

[0006] One purpose of the present application is to provide a Monte Carlo method based uncertainty evaluation method, which comprises:

[0007] respectively obtain an initial test sample size of the uncertainty of the to-be-tested three-dimensional positive ultrasonic anemometer and an execution number in which the uncertainty of the to-be-tested three-dimensional positive ultrasonic anemometer is evaluated;

[0008] determine a target test sample size of a current execution number according to the execution number and the initial test sample size, and determine a total test sample size of all participations in evaluating the uncertainty of the to-be-tested three-dimensional positive ultrasonic anemometer;

[0009] cycle through the sample values of the target test sample size to obtain the uncertainty evaluation result of the to-be-tested three-dimensional positive ultrasonic anemometer according to a preset rule, and obtain a standard deviation of the uncertainty evaluation result of the to-be-tested three-dimensional positive ultrasonic anemometer;

[0010] When it is judged that the standard deviations of the evaluation results are all less than a preset numerical tolerance, a final evaluation result is determined according to the total test sample size of all participations in the cycle;

[0011] The calculation formula of the target test sample size is:

[0012] ;

[0013] The calculation formula of the total test sample size is:

[0014] ;

[0015] wherein, N is an initial test sample space value, h is the execution number.

[0016] Further, the uncertainty evaluation method based on the adaptive Monte Carlo method, wherein the step of obtaining the uncertainty evaluation result of the to-be-tested three-dimensional positive ultrasonic anemometer according to the sample values of the target test sample size through the preset rule comprises:

[0017] establishing an actual measurement model of the three-dimensional positive ultrasonic anemometer, and obtaining a probability distribution function of a model input quantity in the actual measurement model;

[0018] randomly extracting the sample values of the target test sample size according to the sample values generated by the probability distribution function of the model input quantity, and calculating the sample values of the target test sample size through the actual measurement model to obtain discrete values of corresponding output quantities;

[0019] performing non-decreasing sorting on the discrete values of the output quantities to obtain a distribution function of the sorted output quantities, so as to determine the evaluation result of the measurement uncertainty of the three-dimensional positive ultrasonic anemometer according to the discrete representation of the distribution function of the output quantities.

[0020] Further, the uncertainty evaluation method based on the adaptive Monte Carlo method, wherein the step of establishing the actual measurement model of the three-dimensional positive ultrasonic anemometer comprises:

[0021] respectively obtaining the theoretical measurement model of the indication error of the wind speed and the indication error of the wind direction of the three-dimensional positive ultrasonic anemometer;

[0022] establishing a three-dimensional positive ultrasonic anemometer test system composed of a wind tunnel device, a standard wind speed device, a standard wind direction device, and the three-dimensional positive ultrasonic anemometer to be evaluated;

[0023] obtaining the influence of the instability of the wind tunnel performance of the wind tunnel device in the three-dimensional positive ultrasonic anemometer test system, and the influence of the structure of the standard wind speed device, to establish the actual measurement model of the indication error of the wind speed of the three-dimensional positive ultrasonic anemometer according to the theoretical measurement model of the indication error of the wind speed of the three-dimensional positive ultrasonic anemometer;

[0024] obtaining the influence of the instability of the wind tunnel performance of the wind tunnel device in the three-dimensional positive ultrasonic anemometer test system, to establish the actual measurement model of the indication error of the wind direction of the three-dimensional positive ultrasonic anemometer according to the theoretical measurement model of the indication error of the wind direction of the three-dimensional positive ultrasonic anemometer.

[0025] Further, the uncertainty evaluation method based on the adaptive Monte Carlo method, wherein the influence of the instability of the wind tunnel performance of the wind tunnel device comprises the influence of the fluctuation of the airflow in the wind tunnel, and the influence of the airflow deflection angle and the turbulence degree;

[0026] the influence of the structure of the standard wind speed device comprises the influence of the airflow blockage in the wind tunnel caused by the structure of the standard wind speed device, and the influence of the installation of the standard wind speed device;

[0027] the expression of the actual measurement model of the indication error of the wind speed of the three-dimensional positive ultrasonic anemometer is:

[0028]

[0029] wherein, v is the measured wind speed of the three-dimensional positive ultrasonic anemometer, is the standard wind speed measured by the standard wind speed device, is the fluctuation coefficient of the wind tunnel, is the airflow deflection angle coefficient of the wind tunnel, gamma is the turbulence degree coefficient of the airflow of the wind tunnel, is the blockage error coefficient of the standard wind speed device, is the installation error coefficient of the standard wind speed device.

[0030] ​Furthermore, in the above uncertainty assessment method based on the adaptive Monte Carlo method, the expression for the measured wind speed of the three-dimensional positive definite ultrasonic anemometer is:

[0031]

[0032] The three-dimensional positive-deterministic ultrasonic anemometer includes distributed ultrasonic sensors. a Ultrasonic sensors b Ultrasonic sensors c Ultrasonic sensors d Ultrasonic sensors e and ultrasonic sensors f Among them, ultrasonic sensors a Ultrasonic sensors b As a group, used to measure in x Wind speed in the axial direction, ultrasonic sensor c Ultrasonic sensors d As a group, used to measure in y Wind speed in the axial direction, ultrasonic sensor e and ultrasonic sensors f As a group, used for measurement z Wind speed along the axial direction, where, x shaft and y The axes together form a horizontal plane. x The axis is oriented east-west. y The axis is oriented north-south;

[0033] L ab For ultrasonic sensors a and ultrasonic sensors b The distance between them t ab To obtain from ultrasonic sensors a Sending ultrasonic signals, ultrasonic sensor b The time elapsed to receive the ultrasonic signal. t ba To obtain from ultrasonic sensors b Sending ultrasonic signals, ultrasonic sensor a The time elapsed to receive the ultrasonic signal. L cd For ultrasonic sensors c and ultrasonic sensors d The distance between them t cd To obtain from ultrasonic sensors c Sending ultrasonic signals, ultrasonic sensor d The time elapsed to receive the ultrasonic signal. tdc for transmitting an ultrasonic signal from an ultrasonic sensor d for transmitting an ultrasonic signal from an ultrasonic sensor c for receiving a time taken for the ultrasonic signal to travel between the ultrasonic sensor L ef for transmitting an ultrasonic signal from an ultrasonic sensor e for transmitting an ultrasonic signal from an ultrasonic sensor f for receiving a distance between the ultrasonic sensor t ef for transmitting an ultrasonic signal from an ultrasonic sensor e for transmitting an ultrasonic signal from an ultrasonic sensor f for receiving a time taken for the ultrasonic signal to travel between the ultrasonic sensor t fe for transmitting an ultrasonic signal from an ultrasonic sensor f for transmitting an ultrasonic signal from an ultrasonic sensor e for receiving a time taken for the ultrasonic signal to travel between the ultrasonic sensor

[0034] Further, the uncertainty evaluation method based on the adaptive Monte Carlo method, wherein the influence of the instability of the wind tunnel device includes the influence of the fluctuation of the airflow in the wind tunnel, the influence of the angle of attack of the airflow, and the influence of the turbulence intensity of the airflow.

[0035] The expression of the actual measurement model of the wind direction indication error of the three-dimensional positive ultrasonic anemometer is:

[0036] ;

[0037] wherein, D is the wind direction indication of the three-dimensional positive ultrasonic anemometer, is the standard wind direction measured by the standard wind direction device, is the fluctuation coefficient of the wind tunnel, is the angle of attack coefficient of the airflow in the wind tunnel, is the turbulence intensity coefficient of the airflow in the wind tunnel.

[0038] Further, the uncertainty evaluation method based on the adaptive Monte Carlo method, wherein the expression of the wind direction indication of the three-dimensional positive ultrasonic anemometer is:

[0039] ;

[0040] wherein, L ab for transmitting an ultrasonic signal from an ultrasonic sensor a for transmitting an ultrasonic signal from an ultrasonic sensor b for receiving a distance between the ultrasonic sensor t ab for transmitting an ultrasonic signal from an ultrasonic sensor a for transmitting an ultrasonic signal from an ultrasonic sensor b for receiving a time taken for the ultrasonic signal to travel between the ultrasonic sensor tba to transmit an ultrasonic signal from an ultrasonic sensor b to transmit an ultrasonic signal from an ultrasonic sensor a to receive a time taken for an ultrasonic signal to travel between the ultrasonic sensor L cd to transmit an ultrasonic signal from an ultrasonic sensor c to receive a time taken for an ultrasonic signal to travel between the ultrasonic sensor d to receive a time taken for an ultrasonic signal to travel between the ultrasonic sensor t cd to transmit an ultrasonic signal from an ultrasonic sensor c to transmit an ultrasonic signal from an ultrasonic sensor d to receive a time taken for an ultrasonic signal to travel between the ultrasonic sensor t dc to transmit an ultrasonic signal from an ultrasonic sensor d to transmit an ultrasonic signal from an ultrasonic sensor c to receive a time taken for an ultrasonic signal to travel between the ultrasonic sensor

[0041] Another object of the present application is to provide an uncertainty evaluation device based on adaptive Monte Carlo method, which comprises:

[0042] an acquisition module, configured to acquire an initial test sample size of a three-dimensional positive ultrasonic anemometer to be measured and an execution number of evaluation of uncertainty of the three-dimensional positive ultrasonic anemometer to be measured, respectively;

[0043] a determination module, configured to determine a target test sample size of a current execution number according to the execution number and the initial test sample size, and determine a total test sample size of all participating in evaluation of uncertainty of the three-dimensional positive ultrasonic anemometer to be measured;

[0044] a circulation module, configured to circulate sample values based on the target test sample size to acquire an uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be measured by a preset rule, and acquire a standard deviation of the uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be measured;

[0045] an evaluation module, configured to determine a final evaluation result according to the total test sample size of all participating in circulation to acquire the evaluation result when judging that the standard deviations of the evaluation results are all less than a preset numerical tolerance;

[0046] wherein, a calculation formula of the target test sample size is:

[0047] ;

[0048] a calculation formula of the total test sample size is:

[0049] ;

[0050] wherein, N is an initial test sample space value,h the number of executions

[0051] Another object of the present application is to provide a readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any one of the preceding claims.

[0052] Another object of the present application is to provide an electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, the processor implementing the steps of the method according to the preceding claim when executing the program.

[0053] The present application obtains the initial test sample size of the uncertainty of the three-dimensional positive ultrasonic anemometer to be measured and the number of executions of the uncertainty of the three-dimensional positive ultrasonic anemometer to be measured by respectively acquiring the initial test sample size of the uncertainty of the three-dimensional positive ultrasonic anemometer to be measured and the number of executions of the uncertainty of the three-dimensional positive ultrasonic anemometer to be measured; the target test sample size of the current number of executions is determined according to the number of executions and the initial test sample size, and the total test sample size participating in the evaluation of the uncertainty of the three-dimensional positive ultrasonic anemometer to be measured is determined; the uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be measured is obtained by the preset rule based on the sample value of the target test sample size in a loop, and the standard deviation of the uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be measured is obtained; when it is judged that the standard deviations of the evaluation results are all less than the preset numerical tolerance, the final evaluation result is determined according to the total test sample size participating in the loop to obtain the evaluation result; the sample space value in the loop body is half of the total test sample value from the number of loop executions h =3, the sample space value in the loop body is half of the total test sample value when the number of loop executions increases by 1 each time, by this method, the total sample value obtained can effectively reduce the total experimental sample space number, reduce the demand for data storage capacity, quickly improve the execution efficiency of the evaluation program, and at the same time can reduce the number of loop executions, effectively improve the stability of the program. And since the conditions for statistical significance remain unchanged, the accuracy of the final evaluation result can still be effectively guaranteed. The problem of low efficiency in the prior art when performing three-dimensional positive ultrasonic anemometers is solved. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 A flowchart of the uncertainty evaluation method based on the adaptive Monte Carlo method provided by an embodiment of the present application;

[0055] Figure 2 A structural schematic diagram of a three-dimensional positive ultrasonic anemometer test system in the uncertainty evaluation method based on the adaptive Monte Carlo method in the first embodiment of the present application;

[0056] Figure 3 A structural schematic diagram of a three-dimensional positive ultrasonic anemometer in the uncertainty evaluation method based on the adaptive Monte Carlo method in the first embodiment of the present application;

[0057] Figure 4 This is a schematic diagram of the principle of the three-dimensional positive definite ultrasonic anemometer in the uncertainty assessment method based on the adaptive Monte Carlo method in the first embodiment of the present invention.

[0058] Figure 5 This is a structural block diagram of the uncertainty assessment device based on the adaptive Monte Carlo method in the third embodiment of the present invention.

[0059] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0060] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0061] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0063] Example 1

[0064] Please see Figure 1 The figure shows an uncertainty assessment method based on adaptive Monte Carlo method in the first embodiment of the present invention, the method including steps S10 to S13.

[0065] Step S10: Obtain the initial test sample size for the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test and the execution number for evaluating the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test.

[0066] Wherein, when the measurement uncertainty is evaluated, multiple tests are needed to obtain multiple sample data, the initial test sample size is the amount of sample data needed when the measurement uncertainty is evaluated for the first time, that is, the number of tests when the measurement uncertainty is evaluated, in the specific implementation, the existing test sample size can be used as the initial test sample size, for example, the reference test sample space value is taken as Z =10 4 .

[0067] Since the embodiment of the application is mainly aimed at adaptive measurement uncertainty, the adaptive measurement uncertainty is mainly through the continuous cycle of single batch measurement uncertainty until the system reaches statistical stability, and then the final evaluation result is determined according to all sample data, the execution number is the number of cycles of adaptive measurement uncertainty, for example, when the third cycle of measurement uncertainty is performed, the execution number is 3.

[0068] Step S11, according to the execution number and the initial test sample size, the target test sample size of the current execution number is determined, and the total test sample size of all participants in the evaluation of the uncertainty of the three-dimensional positive ultrasonic anemometer under test is determined.

[0069] Wherein, the calculation formula of the target test sample size is:

[0070] ;

[0071] The calculation formula of the total test sample size is:

[0072] ;

[0073] Wherein, N is the initial test sample space value, h is the execution number.

[0074] Exemplarily, from the cycle execution number h =3, the cycle number h increases by 1 each time, and the test sample size in the loop body is half of the total test sample size (i.e. h =1, the target test sample size is N ; h =2, the target test sample size is N , and the total test sample size is the sum of the test sample sizes of the previous two cycles, that is, 2× N ; h =3, the test sample size is half of the total test sample size (2× N ), that is, N , so the total test sample size is the sum of the test sample sizes of the previous three times, that is, 3× N ;h =4, the test sample size is half of the total test sample size (3x N ), i.e. the test sample size is 1.5 N , so the total test sample size is 4.5x N ; and so on. In this way, the total test sample size can be effectively reduced, the data storage requirement can be reduced, the execution efficiency of the evaluation program can be improved, the number of loop executions can be reduced, and the stability of the program can be effectively improved. Since the conditions for statistical significance remain unchanged, the accuracy of the final evaluation result can still be effectively guaranteed.

[0075] In step S12, the sample values based on the target test sample size are cycled through a preset rule to obtain the uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be measured, and the standard deviation of the uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be measured is obtained.

[0076] In the embodiment of the present application, the improved adaptive Monte Carlo method (Monte Carlo method, referred to as MCM) is used to obtain the evaluation result of the three-dimensional positive ultrasonic anemometer uncertainty, and the corresponding standard deviation is obtained. The method selects a certain number of experimental samples each time. By sequentially executing h single batch MCM method, and through various output results of the test samples participating in the evaluation, it is determined whether a stable state is reached. Therefore, the embodiment of the present application is based on the sample values of the target test sample size to perform a loop test to obtain the corresponding measurement uncertainty evaluation result.

[0077] The single batch MCM method for evaluating measurement uncertainty is based on the law of propagation of distribution (LPD) of the probability density distribution function, and is suitable for measurement models with any number of probability density functions (PDF) characteristics.

[0078] Specifically, an actual measurement model of the three-dimensional positive ultrasonic anemometer is established, and the probability distribution function of the model input quantity in the actual measurement model is obtained.

[0079] The sample values of the target test sample size are randomly extracted from the sample values generated according to the probability distribution function of the model input quantity, and the sample values of the target test sample size are calculated through the actual measurement model to obtain the discrete values of the corresponding output quantity.

[0080] The discrete values ​​of the output quantity are sorted in non-decreasing order to obtain the distribution function of the sorted output quantity. The measurement uncertainty of the three-dimensional positive definite ultrasonic anemometer is then evaluated based on the discrete representation of the distribution function of the output quantity.

[0081] The multiple discrete output values ​​are then sorted in strictly ascending (or non-descending) order to obtain the sorted output values. Y Discrete representation of the distribution function G ;pass G Calculate the best estimate of the corresponding output result. y Standard uncertainty u ( y The shortest containment interval under the agreed probability [ y low , y high ].

[0082] Specifically, in order to determine whether statistical stability has been achieved, a standard deviation is introduced. The standard deviation is confirmed by each evaluation result. The evaluation result of each measurement uncertainty includes the best estimate, the standard uncertainty, and the left and right endpoints of the shortest coverage interval under the agreed probability.

[0083] Step S13: When it is determined that the standard deviation of the evaluation results is less than the preset numerical tolerance, the final evaluation result is determined based on the total number of test samples that participated in the cycle to obtain the evaluation results.

[0084] Specifically, when the standard deviation of the evaluation results is less than the preset numerical tolerance, the final evaluation result is determined by all sample values, and the final evaluation result is calculated by the total number of test samples that participated in the cycle to obtain the evaluation results.

[0085] The preset numerical tolerance is calculated by taking the standard uncertainty from all sample values ​​during the evaluation process. For example, it is calculated from all sample values. u ( y ), soon u ( y ) represents c ×10 l form( c for n dig (a decimal integer), thus determining 10 l The specific value, and This allows us to determine the preset numerical tolerance, where, l express u ( y The converted index, in practical implementation, has a preset numerical tolerance of 1 / 2. When any one of the standard deviations of the evaluation results is greater than 1 / 2 , it indicates that the statistical stable state has not been reached, and the single batch MCM test is continued to be performed. When all the standard deviations of the evaluation results are less than 1 / 2 , it indicates that the stable state is reached. At this time, the shortest inclusion interval y, u ( y ) and the agreed probability p of the sample values participating in the evaluation are calculated. y low , y high .

[0086] In addition, it should be noted that the preset numerical tolerance can also be adjusted according to actual conditions. The adjustment of the preset numerical tolerance can be determined according to the requirement for the result accuracy, but generally the preset numerical tolerance ≤ , when the requirement for the measurement result accuracy is not high, the preset numerical tolerance = , and when the requirement for the measurement result accuracy is higher, the preset numerical tolerance = / 5.

[0087] In addition, in some optional embodiments, in order to effectively reduce the total experimental sample space number, reduce the data storage amount, and reduce the number of loop executions under the premise of ensuring the statistical stability, a double termination threshold can be set. One of the double termination thresholds is the threshold of the standard deviation of the evaluation result accuracy, and the other is the sample convergence threshold, that is, the "sample information entropy difference upper limit" of two consecutive loops. The core is to quantify the stability of the sample distribution through "information entropy". The closer the entropy value is, the more stable the interval and density covered by the sample are, and there is no need to further expand the sample space through the loop. Specifically, a sample information entropy calculation function is defined. The entropy value of the sample information is calculated after each loop. The entropy difference of two consecutive times is calculated. It is judged whether the entropy difference and the evaluation result accuracy meet the conditions. If one of them does not meet the conditions, the test sample amount in the loop body is multiplied by the total test sample amount for each increase of 1 in the number of loops h to make fine adjustment.

[0088] In summary, the uncertainty evaluation method based on the adaptive Monte Carlo method in the above embodiments of the present application, by respectively acquiring the initial test sample size of the uncertainty of the to-be-tested three-dimensional positive ultrasonic anemometer and the execution number of the uncertainty evaluation of the to-be-tested three-dimensional positive ultrasonic anemometer, determining the target test sample size of the current execution number according to the execution number and the initial test sample size, and determining the total test sample size of all participating in the uncertainty evaluation of the to-be-tested three-dimensional positive ultrasonic anemometer, the sample values based on the target test sample size are acquired through the preset rule to obtain the uncertainty evaluation result of the current to-be-tested three-dimensional positive ultrasonic anemometer, and the standard deviation of the uncertainty evaluation result of the current to-be-tested three-dimensional positive ultrasonic anemometer is acquired; when it is judged that the standard deviations of the evaluation results are all less than the preset numerical tolerance, the final evaluation result is determined according to the total test sample size of all participating in the loop to obtain the evaluation result; starting from the loop execution number h=3, the sample space value in the loop body is half of the total test sample value every time the loop execution number increases by 1, through this method, the total sample value can effectively reduce the total experimental sample space number, reduce the demand for data storage capacity, quickly improve the execution efficiency of the evaluation program, and at the same time, the loop execution number is reduced, and the stability of the program is effectively improved. And since the conditions for the statistical significance are unchanged, the accuracy of the final evaluation result can still be effectively guaranteed. The problem of low efficiency in the prior art when performing the three-dimensional positive ultrasonic anemometer is solved.

[0089] Embodiment two

[0090] The embodiment two of the present application also proposes an uncertainty evaluation method based on the adaptive Monte Carlo method, which is different from the uncertainty evaluation method based on the adaptive Monte Carlo method proposed in the embodiment one of the present application in that:

[0091] The step of establishing the actual measurement model of the three-dimensional positive ultrasonic anemometer includes:

[0092] The theoretical measurement model of the wind speed indication error and the wind direction indication error of the three-dimensional positive ultrasonic anemometer is acquired respectively;

[0093] A three-dimensional positive ultrasonic anemometer test system constructed by a wind tunnel device, a standard wind speed device, a standard wind direction device and the to-be-evaluated three-dimensional positive ultrasonic anemometer is established;

[0094] The influence of the unstable wind tunnel performance of the wind tunnel device in the three-dimensional positive ultrasonic anemometer test system and the influence of the structure of the standard wind speed device are acquired to establish the actual measurement model of the wind speed indication error of the three-dimensional positive ultrasonic anemometer according to the theoretical measurement model of the wind speed indication error of the three-dimensional positive ultrasonic anemometer;

[0095] The influence of unstable wind tunnel performance of a wind tunnel device in a three-dimensional positive ultrasonic anemometer test system is acquired to establish an actual measurement model of wind direction indication error of the three-dimensional positive ultrasonic anemometer according to a theoretical measurement model of the wind direction indication error of the three-dimensional positive ultrasonic anemometer.

[0096] According to the indication error theory of the wind speed and the wind direction, a theoretical measurement model of the wind speed indication error and the wind direction indication error of the three-dimensional positive ultrasonic anemometer is established, and specifically, the theoretical measurement model of the wind speed indication error is:

[0097] ;

[0098] Wherein, △ v is the wind speed indication error of the anemometer, and the unit is m / s; v is the actual measured wind speed of the anemometer, and the unit is m / s; v s is the standard wind speed, and the unit is m / s.

[0099] The theoretical measurement model of the wind direction indication error is:

[0100] ;

[0101] Wherein, is the wind direction indication error value of the anemometer, D is the actual measured wind direction value of the anemometer, D s is the standard wind direction value.

[0102] Wherein, as Figure 2As shown, the three-dimensional positive ultrasonic anemometer test system includes a wind tunnel device, a three-dimensional positive ultrasonic anemometer placed on a placement table built in the wind tunnel device, and a standard wind speed device (standard wind speed measuring device) and an angle encoder (standard wind direction measuring device). In order to test the wind speed and wind direction of the three-dimensional positive ultrasonic anemometer, first of all, a wind tunnel device capable of providing a stable airflow environment needs to be selected. Before testing, the anemometer, the standard wind speed device (standard wind speed measuring device), and the angle encoder (standard wind direction measuring device) are placed in a stable wind speed environment (wind speed test section of the wind tunnel device) at the same time. And align the wind direction of the anemometer with the orientation of the angle encoder. When working, the host computer controls the frequency converter through the wind speed control module to make the wind tunnel device produce stable airflow. After the anemometer contacts the stable airflow, the host computer reads the wind speed and wind direction values of the measured anemometer through the measured wind speed and wind direction module. Then compare the wind speed value of the measured anemometer with the wind speed value (standard wind speed value) of the standard wind speed device to calculate the deviation of the wind speed value of the anemometer from the standard wind speed value (i.e. wind speed indication error); At the same time, compare the wind direction value of the measured anemometer with the wind direction value (value measured by the angle encoder) of the standard wind direction device to calculate the deviation of the wind direction value of the anemometer from the standard wind direction device (i.e. wind direction indication error).

[0103] Wherein, after establishing the theoretical model, the standard wind speed in the theoretical model needs to be obtained. Specifically, a corresponding three-dimensional positive ultrasonic anemometer test system is established. Considering the instability of wind tunnel performance and the influence of standard wind speed device structure on standard wind speed. The standard wind speed can be expressed as:

[0104] ;

[0105] Wherein, is the measured value of the wind speed of the standard wind speed device, with the unit of m / s; △ M is the influence of the instability of the wind tunnel performance, with the unit of m / s; △ N is the influence of the structure of the standard wind speed device, with the unit of m / s.

[0106] Due to the influence of the instability of the wind tunnel performance on the standard wind speed value, the main source is the fluctuation of the airflow in the wind tunnel and the influence of the airflow deflection angle and turbulence degree, so △ M can be expressed as:

[0107] ;

[0108] Wherein, is the wind tunnel fluctuation coefficient, is the wind tunnel airflow deflection angle coefficient, is the wind tunnel airflow turbulence degree coefficient.

[0109] Due to the influence of the standard wind speed device structure on the standard value of the wind speed, the main source is the airflow blockage in the wind tunnel caused by the standard wind speed device structure itself, and the influence caused by the standard wind speed device structure, so N It can be expressed as:

[0110] ;

[0111] Wherein, is the blockage error coefficient of the standard wind speed device, is the installation error coefficient of the standard wind speed device.

[0112] Finally, the actual measurement model of the wind speed indication error of the three-dimensional positive ultrasonic anemometer is obtained:

[0113] ;

[0114] Wherein, v is the measured wind speed of the three-dimensional positive ultrasonic anemometer, is the standard wind speed measured by the standard wind speed device, is the wind tunnel fluctuation coefficient, is the wind tunnel airflow deflection angle coefficient, is the wind tunnel airflow turbulence coefficient, is the blockage error coefficient of the standard wind speed device, is the installation error coefficient of the standard wind speed device.

[0115] In addition, since the present application is to evaluate the uncertainty of the three-dimensional positive ultrasonic anemometer, unlike the two-dimensional ultrasonic wind speed sensor, the core structural characteristics and working mechanism of the three-dimensional positive ultrasonic anemometer are deeply excavated in the embodiment of the application, so that the measured wind speed of the three-dimensional positive ultrasonic anemometer is systematically analyzed and obtained.

[0116] Specifically, as shown in Figures 3 to 4 The three-dimensional positive ultrasonic anemometer is composed of three groups of ultrasonic sensors which are intersected with each other in space, each group of ultrasonic sensors has two corresponding ultrasonic sensors, and the ultrasonic sensor a and the ultrasonic sensor b are a group, which are used to measure the wind speed in the x axis direction v x Similarly, the ultrasonic sensor c and the ultrasonic sensor d are used to measure the wind speed in the y axis direction v y The ultrasonic sensor e and f are used to measure the wind speed in the z axis directionv z ,in x axis, y shaft and z The axes intersect each other perpendicularly, forming a complete three-dimensional coordinate system in space. x shaft and y The axes together form a horizontal plane. x The axis is generally oriented east-west. y The axis is oriented north-south.

[0117] A wind speed v measurement model is established by synthesizing wind speeds from three different directions:

[0118] ;

[0119] From ultrasonic sensors a Sending ultrasonic signals, ultrasonic sensor b The time taken to receive the ultrasonic signal is t ab :

[0120] ;

[0121] in, L ab Indicates ultrasonic sensor a and ultrasonic sensors b The distance between them; c The speed of sound propagation; v x Wind speed v exist x The wind speed vector value of the axis.

[0122] Similarly, from ultrasonic sensors b Sending ultrasonic signals, ultrasonic sensor a The time taken to receive the ultrasonic signal is t ba :

[0123] ;

[0124] Calculate the corresponding wind speed v exist x The wind speed vector of the axis is magnitude v x :

[0125] ;

[0126] Similarly, ultrasonic sensors c and ultrasonic sensors d ultrasonic sensor e and ultrasonic sensorsf the distance between the two is L cd and L ef the corresponding wind speed v at the y axis is of size v y at the z axis is of size v z

[0127] ;

[0128] ;

[0129] the final wind speed v :

[0130] ;

[0131] wherein L ab is the distance between the ultrasonic sensor a and the ultrasonic sensor b , t ab is the time taken for the ultrasonic signal to be transmitted from the ultrasonic sensor a and to be received by the ultrasonic sensor b , t ba is the time taken for the ultrasonic signal to be transmitted from the ultrasonic sensor b and to be received by the ultrasonic sensor a , L cd is the distance between the ultrasonic sensor c and the ultrasonic sensor d , t cd is the time taken for the ultrasonic signal to be transmitted from the ultrasonic sensor c and to be received by the ultrasonic sensor d , t dc is the time taken for the ultrasonic signal to be transmitted from the ultrasonic sensor d and to be received by the ultrasonic sensor c , L ef is the distance between the ultrasonic sensor e and the ultrasonic sensor f , t ef is the time taken for the ultrasonic signal to be transmitted from the ultrasonic sensore transmitting ultrasonic signals, ultrasonic sensor f the time taken to receive the ultrasonic signals, t fe for from ultrasonic sensor f transmitting ultrasonic signals, ultrasonic sensor e the time taken to receive the ultrasonic signals.

[0132] wherein the actual measurement model of the wind direction indication error is established, and the influence of the unstable wind tunnel performance is considered, and the standard wind direction value D s Generally, it will be affected by the fluctuation of air flow and the air flow angle and turbulence degree, therefore D s can be expressed as:

[0133] D s ;

[0134] ;

[0135] wherein, D is the wind direction indication of the three-dimensional positive ultrasonic anemometer, is the standard wind direction measured by the standard wind direction device, is the wind tunnel fluctuation coefficient, is the wind tunnel air flow angle coefficient, is the wind tunnel air flow turbulence degree coefficient.

[0136] wherein, the wind direction in each direction can be obtained by calculating the speed difference of the anemometer in each direction, and then obtaining the wind speed component in three directions, in the implementation of the present application, the horizontal wind direction is targeted, and it is assumed that the angle between the environmental wind and the horizontal north direction is D , then the wind direction value is:

[0137] ;

[0138] wherein v x , v y respectively represent x axis and y axis, i.e. the wind speed vector of east-west direction and north-south direction.

[0139] the wind direction indication of the three-dimensional positive ultrasonic anemometer is obtained as:

[0140] ;

[0141] the actual measurement model of the wind direction indication error is finally obtained as:

[0142] ;

[0143] wherein, is a standard wind direction measured by a standard wind direction device, is a wind tunnel fluctuation coefficient, is a wind tunnel airflow deflection angle coefficient, gamma is a wind tunnel airflow turbulence coefficient.

[0144] The uncertainty evaluation method based on the adaptive Monte Carlo method in the above embodiment of the present application, by respectively acquiring the initial test sample size of the uncertainty of the to-be-tested three-dimensional positive ultrasonic anemometer and the execution number of the uncertainty evaluation of the to-be-tested three-dimensional positive ultrasonic anemometer, determining the target test sample size of the current execution number according to the execution number and the initial test sample size, and determining the total test sample size of all participating in the uncertainty evaluation of the to-be-tested three-dimensional positive ultrasonic anemometer, circulating the sample values based on the target test sample size to acquire the uncertainty evaluation result of the current to-be-tested three-dimensional positive ultrasonic anemometer through a preset rule, and acquiring the standard deviation of the uncertainty evaluation result of the current to-be-tested three-dimensional positive ultrasonic anemometer, when judging that the standard deviations of the evaluation results are all less than a preset numerical tolerance, then determining the final evaluation result according to the total test sample size of all participating in the circulation to acquire the evaluation result, from the circulation execution number h=3, the circulation execution number increases by 1 each time, the sample space value in the loop body is half of the total test sample value, through this method, the total sample value obtained can effectively reduce the total experimental sample space number, reduce the demand for data storage capacity, quickly improve the execution efficiency of the evaluation program, at the same time, the circulation execution number is reduced, and the stability of the program is effectively improved. And since the conditions for statistical significance remain unchanged, the accuracy of the final evaluation result can still be effectively guaranteed. The problem of low efficiency in the prior art when performing the three-dimensional positive ultrasonic anemometer is solved.

[0145] Embodiment three

[0146] Please refer to Figure 5 , which is the uncertainty evaluation device based on the adaptive Monte Carlo method proposed in the third embodiment of the present application, the device comprises:

[0147] The acquisition module 100 is used for respectively acquiring the initial test sample size of the uncertainty of the to-be-tested three-dimensional positive ultrasonic anemometer and the execution number of the uncertainty evaluation of the to-be-tested three-dimensional positive ultrasonic anemometer.

[0148] The determination module 200 is used for determining the target test sample size of the current execution number according to the execution number and the initial test sample size, and determining the total test sample size of all participating in the uncertainty evaluation of the to-be-tested three-dimensional positive ultrasonic anemometer.

[0149] The circulating module 300 is used for circulating the sample value based on the target test sample amount to obtain the current uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be tested through the preset rule, and obtain the standard deviation of the current uncertainty evaluation result of the three-dimensional positive ultrasonic anemometer to be tested;

[0150] The evaluation module 400 is used for determining the final evaluation result according to the total test sample amount of all participating cycles for obtaining the evaluation result when it is judged that the standard deviations of the evaluation results are all less than the preset numerical tolerance;

[0151] The calculation formula of the target test sample amount is as follows:

[0152] ;

[0153] The calculation formula of the total test sample amount is as follows:

[0154] ;

[0155] The total test sample amount is calculated as follows: N The initial test sample space value is h The execution number is

[0156] The functions or operation steps realized when the above modules are executed are basically the same as those of the above method embodiments, and will not be described here.

[0157] In addition, in order to further more clearly and specifically describe the specific implementation process of the embodiments of the present application, the following will be described through different evaluation methods in a specific embodiment to obtain different results, so as to prove the superiority of the embodiments of the present application.

[0158] (1) Determine the basic information of the input amount

[0159] By querying the original test data of the three-dimensional positive ultrasonic anemometer, it is known that the corresponding L ab , L cd , L ef The original test data is shown in Table 1.

[0160] Table 1

[0161]

[0162] When the wind tunnel system wind speed is set to 6 m / s, the three-dimensional positive ultrasonic anemometer x axis, y axis and z axis direction ultrasonic time parameters are read by the upper computer software, and the actual measured value of the standard wind speed device is read As shown in Table 2.

[0163] Table 2

[0164]

[0165] The experiment uses a WZ860070-E straight open wind tunnel. According to the technical parameters of the inquirer, the fluctuation coefficient of the wind tunnel is a = ± 0.1%; the airflow deflection angle is less than ± 1°, and the corresponding deflection angle coefficient is b = ± (1° / 360°) x 100% = ± 0.28%; and the turbulence degree of the wind tunnel is c = ± 0.2%.

[0166] The wind speed standard device used in the experiment is a laser Doppler velocimeter (LDV). Since the LDV velocity measurement principle is to measure the Doppler shift generated by the scattering of moving particles, this method is a non-contact measurement method (when LDV is used as a wind speed standard device, it does not need to be installed in the wind tunnel), so LDV will not affect the airflow in the wind tunnel, and the blockage error coefficient e = 0 and the installation error coefficient f = 0.

[0167] (2) Determine the related parameters of the probability distribution of the input

[0168] According to the actual measurement model, it is necessary to construct the corresponding input L ab , L cd , L ef , t ab , t ba , t cd , t dc , t ef , t fe and , a, b, c, d, e . According to the measured data in Table 1 and Table 2, according to the requirements of JJF1059.2 “Evaluation of Measurement Uncertainty by Monte Carlo Method”, the probability distribution and related parameters of the relevant input are shown in Table 3.

[0169] Table 3

[0170]

[0171] (3) Determine the probability density function (PDF) of the input

[0172] Ultrasonic sensor a and b Distance between L ab According to the construction rule of MCM about PDF (refer to JJF1059.2《Evaluation of measurement uncertainty by Monte Carlo method》), the repetitive measurement data is generally normally distributed, and the mean and standard deviation of repetitive measurement are taken as the expectation and variance of normal distribution. Since L ab The mean and standard deviation of the original factory test data are 74.003 mm and 0.01008 mm respectively, so L ab The corresponding PDF is N (74.003 mm, 0.01008 mm), N The normal distribution symbol is.

[0173] Ultrasonic sensor c and d Distance between L cd The PDF construction method of L ab is similar to that of L cd The original factory test data show that L cd The corresponding PDF is N (74.004 mm, 0.01018 mm);

[0174] Ultrasonic sensor e and f Distance between L ef The PDF construction method of L ab is similar to that of L cd The original factory test data show that L ef The corresponding PDF is N (74.007 mm, 0.01016 mm);

[0175] a Time of flight from b to t ab According to the time of flight data of ultrasonic sensor from a to b read by the upper computer software, it belongs to repetitive measurement data, which is normally distributed, and the mean and standard deviation are 215.98 us and 0.04 us respectively, t abThe corresponding PDF is N (215.98 us, 0.04 us);

[0176] b to a The time of flight from t ba : By the same token, t ba The corresponding PDF is N (220.64 us, 0.07 us);

[0177] c to d The time of flight from t cd : By the same token, t cd The corresponding PDF is N (216.43 us, 0.08 us);

[0178] d to c The time of flight from t dc : By the same token, t dc The corresponding PDF is N (220.90 us, 0.10 us);

[0179] e to f The time of flight from t ef : By the same token, t ef The corresponding PDF is N (215.55 us, 0.14 us);

[0180] f to e The time of flight from t fe : By the same token, t fe The corresponding PDF is N (220.30 us, 0.13 us);

[0181] The measured value of the wind speed of the standard wind speed device : The average value and the standard deviation of the repeated measurement data are taken as the expectation and variance of the normal distribution, so The corresponding PDF is N (6.06 m / s, 0.03 m / s);

[0182] The fluctuation coefficient of the wind tunnel αAccording to the construction rule of PDF of MCM (refer to JJF 1059.2 "Evaluation of measurement uncertainty by Monte Carlo method"), the distribution of the data with the maximum probability at the ends of the interval is generally taken as the inverse sine distribution, and the smaller value of the data with the maximum probability is taken as the lower limit endpoint of the interval, and the larger value is taken as the upper limit endpoint of the interval. Therefore a The corresponding PDF is Asn (-0.1%, 0.1%), Asn is the inverse sine distribution symbol.

[0183] Wind tunnel airflow angle coefficient β According to the construction rule of PDF of MCM (refer to JJF 1059.2 "Evaluation of measurement uncertainty by Monte Carlo method"), the data with the same probability at any value in the interval is generally taken as the uniform distribution, therefore b The corresponding PDF is U (-0.28%, 0.28%), U is the uniform distribution symbol.

[0184] Wind tunnel airflow turbulence coefficient gamma Similarly, the turbulence is taken as the uniform distribution, therefore c The corresponding PDF is U (-0.2%, 0.2%).

[0185] The PDF of the related input quantity is shown in Table 4.

[0186] Table 4

[0187]

[0188] (4) Evaluation of uncertainty

[0189] Single batch MCM

[0190] 1) MCM input: define the output quantity as the error of the three-dimensional ultrasonic anemometer, establish the actual measurement model, set the PDF of the available information as the input quantity (Table 4), and set the test sample space value M = 10 6 .

[0191] 2) MCM propagation and output: extract M sample values from the PDF corresponding to each input quantity, and calculate the M corresponding output quantity discrete values by substituting the sample vector into the actual measurement model, and sort the output quantity distribution function discrete representation in strict increasing order.

[0192] 3) MCM report results: determine the best estimate value △ of the measurement error of the three-dimensional ultrasonic anemometer through the sorted output quantity distribution function discrete representation.v =0.1664 m / s, its standard uncertainty u (△ v )=0.0344 m / s, and the shortest interval of inclusion [△ v low ,△ v high ]=[0.0990 m / s, 0.2338 m / s].

[0193] According to the rounding rule, the final output result is △ v =0.17 m / s 、u (△ v )=0.03 m / s, and the shortest interval of inclusion [△ v low ,△ v high ]=[0.10 m / s, 0.23 m / s].

[0194] Where, a single batch of MCM is performed without loop, and the total number of test sample space values is 10 6 .

[0195] One of the improved Monte Carlo methods

[0196] 1) Set the effective number of significant digits of the standard uncertainty u (△ v )=1, and the probability of the shortest interval of inclusion of the output result n dig =95%; p

[0197] 2) Set the number of loop applications h =1, and the number of MCM executions N= 10 4 ;

[0198] 3) Perform a single batch of MCM test, i.e., substitute the PDFs (the PDFs are shown in Table 4, and the number of test samples N =10 4 ) of the input quantities into the measurement model, perform MCM propagation and output, and obtain the corresponding △ v =0.1664 m / s, u (△ v )=0.0345 m / s, [△ v low ,△ v high ]=[0.0987 m / s, 0.2316 m / s];

[0199] 4) h ​= 1, set h +1, return to step 3, continue to perform single batch MCM test, obtain corresponding v = 0.1671 m / s, u (△ v )= 0.0347 m / s, [△ v low ,△ v high ]=[0.0994 m / s, 0.2349 m / s];

[0200] 5) Calculate the standard deviation of the average value of △ v 、 u (△ v ), △ v low ,△ v high respectively s △v = 0.00065 m / s, s u(△v) = 0.00015 m / s, s △vlow = 0.00073 m / s, s △vhigh = 0.00326 m / s;

[0201] 6) Calculate u (△ v )= 0.0345 m / s by all sample values (2x N sample values), and calculate the numerical tolerance = 0.005 m / s (in order to obtain higher precision measurement results, a smaller value than may be selected according to actual measurement requirements, such as 0.0008 m / s, and let it be △);

[0202] 7) When h is always increasing, until h = 8, 2 s △v = 0.0002215 m / s, 2 s u(△v) = 0.0001839 m / s, 2 △vlow = 0.0008820 m / s, 2 △vhigh = 0.0007490 m / s, 2 s △v , 2 s u(△v) , 2 △vlow , 2 △vhigh of 2 △vlowIn addition, all other values ​​are less than △;

[0203] At this point, the loop continues to execute until... h =9, 2 s △v =0.0002061m / s, 2 s u(△v) =0.0001709m / s, 2 △vlow =0.0007825m / s, 2 △vhigh =0.0006987m / s, 2 s △v 2 s u(△v) 2 △vlow 2 △vhigh If any value in the equation is less than Δ, it indicates that a steady state has been reached. At this point, the 9×10 4 Δ was calculated from the sample values. v= 0.1663m / s u (△ v ) = 0.0342 m / s, [△ v low ,△ v high = [0.0997m / s, 0.2328m / s];

[0204] According to the rounding rules, the final output is △ v =0.17m / s 、u (△ v =0.03m / s, and the shortest containment interval with a conventional probability of 95% [△ v low ,△ v high = [0.10m / s, 0.23m / s].

[0205] In the end, a total of 9 iterations were performed, using a total of 9 × 10⁻⁶ experimental sample space values. 4 .

[0206] Another improved Monte Carlo method for single batches

[0207] The specific execution steps are similar to the method described above, except that the number of iterations is adjusted during loop execution. h Each additional iteration results in a sample space within the loop of ( h- 1)×10 4 The detailed execution steps are not repeated here; only the final result is given. When the program executes the 8th loop, it reaches a stable state. At this point, through 2... (8-1) ×10 4 =128×10 4 Δ was calculated from the sample values.v= 0.1663m / s, u (△ v )=0.0345m / s,[△ v low ,△ v high ]=[0.0986m / s,0.2338m / s]; according to the rounding rule, the final output result is the same as the MCM and AMCM output results.

[0208] Finally, a total of 8 cycles of cycles are performed, and a total of 2 (8-1) ×10 4 =128×10 4 .

[0209] The method adopted by the embodiment of the application

[0210] 1) Set the effective number of bits of (△ u (△ v )=1, and the shortest interval probability of the output result is agreed to be 95%; n dig p

[0211] 2) Set the number of cycle applications h =1, and set the number of MCM executions N= 10 4 .

[0212] 3) Perform a single batch of MCM tests, that is, the PDF (the PDF is shown in Table 4, and the test sample amount N =10 4 ) of each input quantity is substituted into the measurement model, MCM propagation and output are performed, and the corresponding △ v =0.1664m / s, u (△ v )=0.0345m / s,[△ v low ,△ v high ]=[0.0987m / s,0.2316m / s] are obtained.

[0213] 4) h =1, and set h +1, return to step 3, continue to perform a single batch of MCM tests, and the corresponding △ v =0.1671m / s, u (△ v )=0.0347m / s,[△ v low ,△ v high ​​= [0.0994m / s, 0.2349m / s];

[0214] 5) The Δ obtained by calculating the above two cycles v , u (△ v ), △ v low , △ v high The standard deviations of their respective means are respectively s △v =0.00065m / s s u(△v) =0.00015m / s s △vlow =0.00073m / s s △vhigh =0.00326m / s;

[0215] 6) Through all sample values ​​(2× N Calculate from (sample values) u (△ v ) = 0.0345 m / s, and calculate the numerical tolerance. =0.005m / s (Similarly, a ratio can be selected according to actual measurement needs) For smaller values, such as 0.0008 m / s, let it be Δ); in this case, 2 s △v 2 s u(△v) 2 △vlow 2 △vhigh If all values ​​are greater than △, it indicates that a stable state has not yet been reached, so the loop continues.

[0216] 7) When h It keeps increasing (when it continues to increase, when h When ≥3, the experimental sample space value is N ×1.5 （h-3） .For example: h When =3, its experimental sample space value is the total sample value (2×10). 4 0.5 times, that is, 10 4 ; h When =4, its experimental sample space value is the total sample value (3×10). 4 0.5 times that, i.e., 1.5 × 10 4 (and so on), until... h When =6, 2 s △v =0.000315m / s, 2 s u(△v) =0.000128m / s, 2△vlow = 0.000858 m / s, 2 △vhigh = 0.000861 m / s, 2 s △v = 0.000858 m / s, 2 s u(△v) = 0.000861 m / s, 2 △vlow = 0.000858 m / s, 2 △vhigh = 0.000861 m / s, 2 △vlow = 0.000858 m / s, 2 △vhigh = 0.000861 m / s, 2

[0217] At this time, the loop continues to execute, when h = 7, 2 s △v = 0.000279 m / s, 2 s u(△v) = 0.000113 m / s, 2 △vlow = 0.000776 m / s, 2 △vhigh = 0.000732 m / s, 2 s △v = 0.000279 m / s, 2 s u(△v) = 0.000113 m / s, 2 △vlow = 0.000776 m / s, 2 △vhigh = 0.000732 m / s, 2 N Any one of the values of 2 N × 1.5 （7-2） = 15.1875 × v= 0.1666 m / s, u (△ v )= 0.0343 m / s, [△ v low ,△ v high ]=[0.0994 m / s, 0.2335 m / s];

[0218] According to the rounding rule, the final output result is △ v = 0.17 m / s 、u (△ v )= 0.03 m / s, and the shortest containing interval [△ v low ,△ v high ]=[0.10 m / s, 0.23 m / s]. The final output result is the same as the output results of several existing evaluation methods.

[0219] And the method of the embodiment of the application is adopted, a total of 7 loops are performed, and a total of 15.1875 × 10 4 .

[0220] It can be seen that the evaluation method proposed in the embodiments of the present application has significant advantages in calculation efficiency and result stability. Through the growth strategy of dynamically adjusting the sample quantity, the method can reach a stable state with less cycle times and smaller total sample quantity, thereby significantly reducing the calculation cost while ensuring the accuracy of the uncertainty evaluation. The method avoids the problem of too fast growth of the sample quantity, and further improves the calculation efficiency.

[0221] Embodiment Four

[0222] Another aspect of the present application also provides a readable storage medium, which stores a computer program, and the program is executed by a processor to realize the steps of the method described in Embodiment One and Embodiment Two.

[0223] Embodiment Five

[0224] Another aspect of the present application also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor executes the program to realize the steps of the method described in Embodiment One and Embodiment Two.

[0225] The technical features of each of the above embodiments can be combined arbitrarily, and to make the description concise, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0226] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a list of executable instructions for implementing logical functions, which can be embodied in any computer readable storage medium for use by or in conjunction with an instruction execution system, device or apparatus, such as a computer-based system, a system including a processor or other system that can fetch and execute instructions from an instruction execution system, device or apparatus. For the present specification, the "computer readable storage medium" can be any device that can contain, store, communicate, propagate or transport programs for use by or in conjunction with an instruction execution system, device or apparatus, or in conjunction with these instruction execution systems, devices or apparatus.

[0227] More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires (electrical, optical, and the like) a portable computer diskette (magnetic, or optical, e.g., Blu-ray® disk, etc.) a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). Additionally, the computer readable storage medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example via an optical scanner, then compiled, interpreted, or otherwise processed, using suitable tools, and then stored in a computer memory.

[0228] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the embodiments described above, various steps or methods can be implemented, for example, by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following techniques, which are well known in the art, can be used to implement the application: a hybrid of the techniques mentioned above; a combination of one or more of the techniques mentioned above; or one or more other techniques suitable for use in the computer hardware devices described above.

[0229] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in one or more embodiments or examples.

[0230] The above-described embodiments only express several implementation manners of the application, which are described in a more specific and detailed manner, but cannot be understood as a limitation on the patent scope of the application. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the application, which are all within the protection scope of the application. Therefore, the patent protection scope of the application should be subject to the appended claims.

Claims

1. An uncertainty assessment method based on adaptive Monte Carlo method, characterized in that, The method includes: The initial test sample size of the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test and the execution number at which the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test is evaluated are obtained respectively. The target test sample size for the current execution number is determined based on the execution number and the initial test sample size, as well as the total test sample size for all tests involved in evaluating the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test. The uncertainty assessment result of the current three-dimensional positive definite ultrasonic anemometer under test is obtained by using the sample values ​​based on the target test sample size and a preset rule, and the standard deviation of the uncertainty assessment result of the current three-dimensional positive definite ultrasonic anemometer under test is also obtained. When it is determined that the standard deviation of the evaluation results is less than the preset numerical tolerance, the final evaluation result is determined based on the total sample size of all test samples that participated in the cycle to obtain the evaluation results. The formula for calculating the target experimental sample size is as follows: ； The formula for calculating the total experimental sample size is: ； in, N These are the initial experimental sample space values. h For the current execution number, For the target experimental sample size, T h The total sample size in the experiment; The steps for obtaining the uncertainty assessment result of the current three-dimensional positive definite ultrasonic anemometer based on the sample values ​​of the target test sample size through preset rules include: Establish a three-dimensional positive definite ultrasonic anemometer actual measurement model and obtain the probability distribution function of the model input quantity in the actual measurement model; The sample values ​​of the target test sample size are randomly extracted from the sample values ​​generated by the probability distribution function of the model input quantity, and the discrete values ​​of the corresponding output quantity are obtained by calculating the sample values ​​of the target test sample size through the actual measurement model. The discrete values ​​of the output quantity are sorted in non-decreasing order to obtain the distribution function of the sorted output quantity. The measurement uncertainty of the three-dimensional positive definite ultrasonic anemometer is then evaluated based on the discrete representation of the distribution function of the output quantity.

2. The uncertainty assessment method based on adaptive Monte Carlo method according to claim 1, characterized in that, The steps for establishing the actual measurement model of the three-dimensional positive definite ultrasonic anemometer include: Theoretical measurement models for the wind speed indication error and wind direction indication error of the three-dimensional positive definite ultrasonic anemometer were obtained respectively. A three-dimensional positive definite ultrasonic anemometer testing system was established, consisting of a wind tunnel device, a standard wind speed device, a standard wind direction device, and a three-dimensional positive definite ultrasonic anemometer to be evaluated. The effects of wind tunnel performance instability and standard anemometer structure on the wind tunnel device in the three-dimensional positive definite ultrasonic anemometer testing system are obtained, so as to establish an actual measurement model of the wind speed indication error of the three-dimensional positive definite ultrasonic anemometer based on the theoretical measurement model of the wind speed indication error of the three-dimensional positive definite ultrasonic anemometer. The influence of wind tunnel performance instability in the wind tunnel device of the three-dimensional positive definite ultrasonic anemometer testing system is obtained, so as to establish an actual measurement model of the wind direction indication error of the three-dimensional positive definite ultrasonic anemometer based on the theoretical measurement model of the wind direction indication error of the three-dimensional positive definite ultrasonic anemometer.

3. The uncertainty assessment method based on adaptive Monte Carlo method according to claim 2, characterized in that, The effects of unstable wind tunnel performance include fluctuations in airflow within the wind tunnel, as well as the effects of airflow deflection angle and turbulence. The effects of the standard anemometer structure include the obstruction of airflow inside the wind tunnel caused by the standard anemometer structure, as well as the effects of the installation of the standard anemometer structure. The expression for the actual measurement model of the wind speed indication error of a three-dimensional positive definite ultrasonic anemometer is as follows: ； in, v The measured wind speed is from a three-dimensional positive definite ultrasonic anemometer. The standard wind speed measured by a standard anemometer. This is the wind tunnel variability coefficient. This is the wind tunnel airflow deflection coefficient. The turbulence coefficient of the wind tunnel airflow. The blocking error coefficient of the standard wind speed device. The installation error coefficient for the standard anemometer. This represents the wind speed indication error of a three-dimensional positive definite ultrasonic anemometer.

4. The uncertainty assessment method based on adaptive Monte Carlo method according to claim 3, characterized in that, The expression for the measured wind speed of a three-dimensional positive definite ultrasonic anemometer is: The three-dimensional positive-deterministic ultrasonic anemometer includes distributed ultrasonic sensors. a Ultrasonic sensors b Ultrasonic sensors c Ultrasonic sensors d Ultrasonic sensors e and ultrasonic sensors f Among them, ultrasonic sensors a Ultrasonic sensors b As a group, used to measure in x Wind speed in the axial direction, ultrasonic sensor c Ultrasonic sensors d As a group, used to measure in y Wind speed in the axial direction, ultrasonic sensor e and ultrasonic sensors f As a group, used for measurement z Wind speed along the axial direction, where, x shaft and y The axes together form a horizontal plane. x The axis is oriented east-west. y The axis is oriented north-south; v The measured wind speed is from a three-dimensional positive definite ultrasonic anemometer. L ab For ultrasonic sensors a and ultrasonic sensors b The distance between them t ab To obtain from ultrasonic sensors a Sending ultrasonic signals, ultrasonic sensor b The time elapsed to receive the ultrasonic signal. t ba To obtain from ultrasonic sensors b Sending ultrasonic signals, ultrasonic sensor a The time elapsed to receive the ultrasonic signal. L cd For ultrasonic sensors c and ultrasonic sensors d The distance between them t cd To obtain from ultrasonic sensors c Sending ultrasonic signals, ultrasonic sensor d The time elapsed to receive the ultrasonic signal. t dc To obtain from ultrasonic sensors d Sending ultrasonic signals, ultrasonic sensor c The time elapsed to receive the ultrasonic signal. L ef For ultrasonic sensors e and ultrasonic sensors f The distance between them t ef To obtain from ultrasonic sensors e Sending ultrasonic signals, ultrasonic sensor f The time elapsed to receive the ultrasonic signal. t fe To obtain from ultrasonic sensors f Sending ultrasonic signals, ultrasonic sensor e The time elapsed to receive the ultrasonic signal.

5. The uncertainty assessment method based on adaptive Monte Carlo method according to claim 2, characterized in that, The effects of unstable wind tunnel performance include fluctuations in airflow within the wind tunnel, as well as the effects of airflow deflection angle and turbulence. The expression for the actual measurement model of the wind direction indication error of a three-dimensional positive definite ultrasonic anemometer is as follows: ； in, This represents the wind direction indication error of a three-dimensional positive definite ultrasonic anemometer. D The wind direction reading is from a three-dimensional positive definite ultrasonic anemometer. The standard wind direction is measured by a standard wind vane. This is the wind tunnel variability coefficient. This is the wind tunnel airflow deflection coefficient. γ This is the turbulence coefficient of the wind tunnel airflow.

6. The uncertainty assessment method based on adaptive Monte Carlo method according to claim 5, characterized in that, The expression for the wind direction reading of a three-dimensional positive definite ultrasonic anemometer is: ； in, D The wind direction reading is from a three-dimensional positive definite ultrasonic anemometer. L ab For ultrasonic sensors a and ultrasonic sensors b The distance between them t ab To obtain from ultrasonic sensors a Sending ultrasonic signals, ultrasonic sensor b The time elapsed to receive the ultrasonic signal. t ba To obtain from ultrasonic sensors b Sending ultrasonic signals, ultrasonic sensor a The time elapsed to receive the ultrasonic signal. L cd For ultrasonic sensors c and ultrasonic sensors d The distance between them t cd To obtain from ultrasonic sensors c Sending ultrasonic signals, ultrasonic sensor d The time elapsed to receive the ultrasonic signal. t dc To obtain from ultrasonic sensors d Sending ultrasonic signals, ultrasonic sensor c The time elapsed to receive the ultrasonic signal.

7. An uncertainty assessment device based on the adaptive Monte Carlo method, characterized in that, The apparatus for implementing the uncertainty assessment method based on the adaptive Monte Carlo method according to any one of claims 1 to 6 includes: The acquisition module is used to acquire the initial test sample size of the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test and the execution number of the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test. The determination module is used to determine the target test sample size for the current execution number based on the execution number and the initial test sample size, as well as the total test sample size for all test samples involved in evaluating the uncertainty of the three-dimensional positive definite ultrasonic anemometer under test. The loop module is used to iterate through the sample values ​​based on the target test sample size to obtain the uncertainty evaluation result of the current three-dimensional positive definite ultrasonic anemometer under test through preset rules, and to obtain the standard deviation of the uncertainty evaluation result of the current three-dimensional positive definite ultrasonic anemometer under test. The evaluation module is used to determine the final evaluation result based on the total sample size of all test samples that participated in the cycle to obtain the evaluation results when the standard deviation of the evaluation results is found to be less than the preset numerical tolerance. The formula for calculating the target experimental sample size is as follows: ； The formula for calculating the total experimental sample size is: ； in, N These are the initial experimental sample space values. h For the current execution number, For the target experimental sample size, T h This represents the total sample size in the experiment.

8. A readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 6.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor, when executing the program, implements the steps of the method as described in any one of claims 1 to 6.

Images