CN105319389A - High-precision and wide-range ultrasonic wind speed measuring system and method - Google Patents

Abstract

The invention discloses a high-precision and wide-range ultrasonic wind speed measuring system and method. The wind speed measuring system comprises four ultrasonic probes, an analog switch module, an AD sampling module, a microprocessor module and a communication module, wherein every two of the four ultrasonic probes are orthogonal; each ultrasonic probe is connected with a receiving and transmitting module; the control end of each receiving and transmitting module is connected with the microprocessor module; and the output end of the analog switch module is connected to the input end of the AD sampling module. The method comprises the steps of firstly, initializing all the modules; secondly, detecting whether the communication module receives a control command or not; thirdly, carrying out time delay estimation on four groups of digital quantity of receiving and transmitting signals; and fourthly, obtaining the wind speeds in the directions of two vertically corresponding ultrasonic probes and the wind speeds in the directions of two horizontally corresponding ultrasonic probes. The high-precision and wide-range ultrasonic wind speed measuring system and method have the beneficial effects that the wind speed vector measuring precision of the instrument is also greatly improved, and the important practical significance is achieved.

Description

A kind of high precision wide range ultrasound wind system and method

Technical field

The present invention relates to a kind of wind measuring system and method, particularly a kind of high precision wide range ultrasound wind system and method.

Background technology

Current, ultrasonic wind speed and direction method of measuring oneself have the developing history of decades, successively propose the many measuring methods such as time difference method, frequency-difference method, phase difference method, Doppler method and correlation method.Wherein, time difference method, frequency-difference method, phase difference method and Doppler method are because measuring circuit is complicated, be subject to the reasons such as Environmental Noise Influence, practical application in the measurement of high-precision ultrasonic wind speed and direction is less, and series of advantages the becomes measuring method that current high-precision ultrasonic anemoscope generally adopts such as the measuring circuit that correlation method has with it is simple, antijamming capability is strong.

Correlation technique is actually the time delay estimation method based on signal statistics correlation theory.In this research field, existing a large amount of achievements in research.Be white Gaussian noise situation for measurement ground unrest, Knapp etc. propose broad sense cross-correlation method, though the method principle is simple, calculated amount is little, and estimated accuracy is not high; Maximum-likelihood method is a kind of time delay estimation method of the best, but the method needs the probability density of known signal, and this point is difficult to accomplish exactly, and therefore the method seldom uses in practice; The time delay estimation method based on cyclic autocorrelation function that Gardner and Chen etc. propose, due to can suppress to be different from signal frequency any noise and extremely people attract attention, but the method is only applicable to transmitting and receiving for homogenous frequency signal situation.Due to the Doppler shift that high wind speed certainly leads to, therefore cannot use in the ultrasonic wind meter of Wide measuring range; For measurement ground unrest be gaussian colored noise situation, Higher-Order Cumulants can be used, because Higher Order Cumulants has extremely strong rejection ability for gaussian colored noise, the time delay estimation method therefore based on Higher Order Cumulants can reach very high estimated accuracy.At present, domestic and international high-precision ultrasonic anemometer is all adopt this method substantially.But when ground unrest is non-Gaussian noise, particularly when containing pulse shock noise in ground unrest, the time delay estimadon precision of the method sharply declines.

Pulse shock noise is a kind of non-Gaussian noise with obvious pulse shock character, as discharged the atmospheric noise caused in space, the ignition noise of motor car engine, the switching noise of electrical equipment, Wireless Telecom Equipment harass noise etc., all belong to pulse shock noise.Thus pulse shock noise is a kind of common noise form in sonication times delay measurements environment.Because pulse shock noise meets α Stable distritation, pulse shock noise is therefore generally claimed to be α Stable distritation noise or referred to as α noise.A maximum feature of α noise is that it does not exist limited variance, and therefore, those time delay estimation methods based on generalized related function, maximum-likelihood method, Circular correlation method and Higher-Order Cumulants lost efficacy.This is that one of key factor of measurement data instability often appears in the ultrasonic wind meter of carrying out time delay estimadon based on said method.

In recent years, for α noise, people have carried out large quantifier elimination.Nikias points out, α Stable distritation exists Fractional Lower Order Moments, theoretical according to this, under scholars proposes many α noise backgrounds in succession, based on the time delay estimation method of Fractional Lower Order Moments.But, in a large amount of practical applications, inherent shortcoming based on the time delay estimation method of Fractional Lower Order Moments also comes out gradually: first, Fractional Lower Order Moments is a kind of nonlinear method, particularly there is not half unchangeability in it, namely two mutual statistical independently stochastic variable and Fractional Lower Order Moments be not equal to the Fractional Lower Order Moments of respective stochastic variable and, this makes us cannot carry out signal to be separated with the effective of noise.In addition, the Fractional Lower Order Moments perseverance of α noise and Gaussian noise is non-vanishing, and this illustrates that the noise inhibiting ability of fractional lower-order Moment Methods is not strong.Thus the time delay estimadon precision based on Fractional Lower Order Moments is generally poor.Because α Stable distritation noise is often all mixed in together with Gaussian noise, therefore, people are in the urgent need to a kind of time delay estimation method α noise and Gaussian noise all to extremely strong rejection ability.

Summary of the invention

The object of the invention is to solve a kind of high precision wide range ultrasound wind system and method that problems that existing ultrasound wind system and method exist provide.

High precision wide range ultrasound wind system provided by the invention includes four ultrasonic probes, analog switch module, AD sampling module, microprocessor module and communication module, wherein four ultrasonic probes are orthogonally set between two, each ultrasonic probe is all connected with transceiver module, four transceiver modules are in parallel, the output terminal of each transceiver module is all connected with the input end of analog switch module, the control end of each transceiver module is all connected with microprocessor module, the output terminal of analog switch module is connected to the input end of AD sampling module, the control end of analog switch module is connected to microprocessor module, the output terminal of AD sampling module is connected to microprocessor module, microprocessor module is connected with communication module, for exporting measured wind speed and receiving control command.

High precision wide range ultrasound wave wind detection method provided by the invention, its method is as described below:

After step one, system electrification, initialization is carried out to each module;

After step 2, initialization complete, detect communication module and whether receive control command, if the control command of receiving, then microprocessor module calls control command process function and processes this order, returns and continue detection after process; If do not receive control command, four transceiver modules that then microprocessor module controls to be connected with four ultrasonic probes successively drive ultrasonic probe to send ultrasound wave, again control simulation switch module successively gating four ultrasonic probes as receiving transducer, and the signal transmission received is carried out analog to digital conversion to AD sampling module, its result will pass to microprocessor module and preserve, and altogether obtain the receiving and transmitting signal of four groups of digital quantities;

Step 3, adopt the ultrasonic signal delay time estimation method of the high precision wide range based on fractional order semi-invariant to carry out time delay estimation to the receiving and transmitting signal of four groups of digital quantities respectively, obtain four time delay estimated value t1, t2, t3, t4;

Step 4, establish the distance between two relative ultrasonic probes to be d, then the wind speed obtaining wherein about two corresponding ultrasonic probe directions by relative time error method is the wind speed in about two corresponding ultrasonic probe directions is

Step 5, obtain actual wind speed by orthogonal synthesis and be wind angle is

Step 6, to be exported by communication module at the wind speed and direction angle obtained in step 5, return step 2 afterwards, so circulation obtains real-time wind speed and direction.

The ultrasonic signal delay time estimation method of the high precision wide range based on fractional order semi-invariant described in step 3, its concrete grammar is as follows:

1) determination of fractional order semi-invariant function and standard

(1) determination of fractional order semi-invariant function:

If the fundamental function that Φ (u) is stochastic variable X, have

${C_{RL}}^{kp}={e^{-j\frac{\mathrm{\π}kp}{2}}\frac{{d_{RL}}^{kp}}{{\mathrm{dt}}^{kp}}ln\mathrm{\Φ}\left(u\right)|}_{u=0}$

In formula: for left Riemann-Liouville Fractional Derivative, 0 < p≤1, k is arbitrary integer, claims ^rLc ^kpfor the fractional order semi-invariant of stochastic variable X, fractional order semi-invariant ^rLc ^kpalso can be designated as ^rLcum ^kp();

(2) determination of fractional order semi-invariant standard:

Determine that fractional order semi-invariant standard is as follows:

Standard 1: establish a ₁, a ₂..., a _kfor constant, X (k)=[x ₁, x ₂..., x _k] be stochastic variable, then

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;a_1{x}_1^{p_1},a_2{x}_2^{p_2},\mathrm{...},a_k{x}_k^{p_k}\&rsqb;\\ =a_1{a}_2...a_{k}c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;\end{array}$

In formula: kp=p ₁+ p ₂+ ... + p _k

Standard 2: fractional order semi-invariant is symmetrical to its independent variable, their value and the order of independent variable have nothing to do in other words, namely

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;\\ =c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_{i_1}^{p_{i_1}},x_{i_2}^{p_{i_2}},\mathrm{...},x_{i_k}^{p_{i_k}}\&rsqb;\end{array}$

Wherein, i ₁, i ₂..., i _k1,2 ..., an arrangement of k;

Standard 3: if k stochastic variable { x _ia subset and other parts independent, then

$c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;\&equiv;0$

Standard 4: if stochastic variable collection [x ₁, x ₂..., x _k] and [y ₁, y ₂..., y _k] be independently, then have

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;{(x_1+y_1)}^{p_1},{(x_2+y_2)}^{p_2},...,{(x_k+y_k)}^{p_k}\&rsqb;\\ {=}^{RL}{\mathrm{cum}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;{+}^{RL}{\mathrm{cum}}^{kp}\&lsqb;y_1^{p_1},y_2^{p_2},\mathrm{...},y_k^{p_k}\&rsqb;\end{array}$

But

$\begin{array}{c}{\mathrm{mom}}^{kp}\&lsqb;{(x_1+y_1)}^{p_1},{(x_2+y_2)}^{p_2},...,{(x_k+y_k)}^{p_k}\&rsqb;\\ \&NotEqual;{\mathrm{mom}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;+{\mathrm{mom}}^{kp}\&lsqb;y_1^{p_1},y_2^{p_2},\mathrm{...},y_k^{p_k}\&rsqb;\end{array}$

Standard 5: for 2p rank fractional order semi-invariant ^rLc ^kp(τ), when τ=0, maximal value is had, namely

| ^RLC ^kp(τ)|≤ ^RLC ^kp(0)

2) fractional order semi-invariant is to the rejection ability of α noise and Gaussian noise and suppressing method:

α Stable distritation is a kind of generalized Gaussian distribution, and the fundamental function of standard α Stable distritation is:

Φ(u)＝exp{-γ|u| ^α}

In formula: parameter γ > 0 is called dispersion coefficient; Parameter alpha ∈ (0,2] be called characteristic exponent, when characteristic exponent α=2, α Stable distritation deteriorates to Gaussian distribution;

About fractional order semi-invariant to the rejection ability of α noise and Gaussian noise and suppressing method, there is following theorem:

Theorem 1: the fundamental function of the accurate α Stable distritation of bidding, as shown in above formula, makes m be the minimum positive integer being more than or equal to p, then as p > 0 and α > 0 time, the p rank fractional order semi-invariant of standard α Stable distritation is:

(1) when α-p is not integer,

${C_{RL}}^p=\left\{\begin{array}{cc}0,& \mathrm{\&alpha;}-p>0\\ -\mathrm{\&gamma;}\mathrm{\&Gamma;}(\mathrm{\&alpha;}+1)e^{-j\frac{\mathrm{\&pi;}k\mathrm{\&alpha;}}{2}},& \mathrm{\&alpha;}=p\\ \mathrm{\&infin;},& \mathrm{\&alpha;}-p<0\end{array}\right.$

(2) when 1≤p-α≤m is integer,

^RLC ^p＝0

Visible, for the p rank fractional order semi-invariant of standard α Stable distritation signal, when getting p < α, or when 1≤p-α≤m is integer, its p rank fractional order semi-invariant exists and is zero, due to the special case that Gaussian distribution is in standard α Stable distritation when α=2, therefore, fractional order semi-invariant is still set up gaussian signal, this namely fractional order semi-invariant to the rejection condition of α and Gaussian noise and suppressing method, because the fractional order semi-invariant of α noise and Gaussian noise is zero, namely as p < α, mean the suppression completely to these two kinds of noises, therefore, fractional order semi-invariant has extremely strong rejection ability to α noise and Gaussian noise,

3) based on the ultrasonic signal delay time estimation method of fractional order semi-invariant

For ultrasound wind system, due to the impact by ultrasonic probe and spatial electromagnetic interference, hyperacoustic transmitting and receiving signal is all containing noisy, transmit if hyperacoustic into

x ₁(k)＝s(k)+n _α1(k)+n _g1(k)

The Received signal strength of ultrasonic sensor is

x ₂(k)＝βs(k-D)+n _α2(k)+n _g2(k)

In upper two formulas, s (k), s (k-D) are that D is the time delay of Received signal strength without making an uproar transmitting and receiving signal; β is decay factor; n _{α 1}(k), n _{α 2}(k) and n _g1(k), n _g2k () is respectively the adjoint zero-mean α noise of transmitting and receiving signal and Gaussian noise, n _{α 1}(k), n _{α 2}(k), n _g1(k), n _g2k () is separate between two and with transmitting and receiving signal s (k), s (k-D) is separate;

To x ₁(k) and x ₁k (), gets 2p (2p < α≤2) rank fractional order semi-invariant, by canonical function 1,3,4 and the theorem 1 of fractional order semi-invariant, have

$\begin{array}{c}C_{RL}{{}_{x_1{x}_2}}^{2p}\left(\mathrm{\&tau;}\right)=c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(x_1\right(k),x_2(k+\mathrm{\&tau;})\&rsqb;\\ =c_{RL}{\mathrm{um}}^{2p}\&lsqb;s\left(k\right),s(k-D+\mathrm{\&tau;})\&rsqb;+c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{\mathrm{\&alpha;}1}\right(k),n_{\mathrm{\&alpha;}2}(k+\mathrm{\&tau;})\&rsqb;\\ +c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{\mathrm{\&alpha;}1}\right(k),n_{g2}(k+\mathrm{\&tau;})\&rsqb;+c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{g1}\right(k),n_{\mathrm{\&alpha;}2}(k+\mathrm{\&tau;})\&rsqb;\\ +c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{g1}\right(k),n_{g2}(k+\mathrm{\&tau;})\&rsqb;\\ =C_{RL}{{}_{ss}}^{2p}(\mathrm{\&tau;}-D)\end{array}$

Namely noise is not comprised in last required fractional order semi-invariant, visible the method fully can suppress α noise in measurement environment and Gaussian noise, thus not only there is very high time delay estimadon precision, but also substantially increase the reliability that ultrasonic wind meter works under complex electromagnetic environment, according to the standard 5 of fractional order semi-invariant, as τ-D=0 there is maximal value, therefore have namely after fractional order semi-invariant tried to achieve by microprocessor, it is searched for, find out the τ value that its maximal value is corresponding, the time delay between the ultrasonic transmission/reception signal required by being exactly.

Beneficial effect of the present invention:

1) the fractional order semi-invariant that the present invention proposes is expansion to Higher Order Cumulants and development, and the definition of Higher Order Cumulants is expanded to whole arithmetic number territory by positive integer by it; Fractional order semi-invariant can overcome non-linear existing for fractional lower-order Moment Methods and to α noise and the problem such as Gaussian noise rejection ability is not strong comprehensively;

2) the fractional order semi-invariant that the present invention proposes is in signal transacting field, has the initiative fundamental research of most important theories value and scientific meaning, is an important breakthrough of signal processing theory, has great theory significance and using value;

3) the present invention proposes the time delay estimation method based on fractional order semi-invariant first, and the method can suppress α noise and Gaussian noise on the impact of time delay estimadon precision effectively, has important theory significance and actual application value.

4) the present invention adopts relative time error method to calculate real-time wind speed and direction, the factors such as the temperature of air, humidity, air pressure can be eliminated on the impact of ultrasonic velocity, not only simplify the metering circuit of ultrasonic wind velocity indicator, but also substantially increase the wind vector measuring accuracy of this instrument, there is important practical significance.

Accompanying drawing explanation

Fig. 1 is one-piece construction schematic diagram of the present invention.

Fig. 2 is program circuit schematic diagram of the present invention.

1, ultrasonic probe 2, analog switch module 3, AD sampling module 4, microprocessor module

5, communication module 6, transceiver module.

Embodiment

Refer to shown in Fig. 1 and Fig. 2:

High precision wide range ultrasound wind system provided by the invention includes four ultrasonic probes 1, analog switch module 2, AD sampling module 3, microprocessor module 4 and communication module 5, wherein four ultrasonic probes 1 are orthogonally set between two, each ultrasonic probe 1 is all connected with transceiver module 6, four transceiver modules 6 are in parallel, the output terminal of each transceiver module 6 is all connected with the input end of analog switch module 2, the control end of each transceiver module 6 is all connected with microprocessor module 4, the output terminal of analog switch module 2 is connected to the input end of AD sampling module 3, the control end of analog switch module 2 is connected to microprocessor module 4, the output terminal of AD sampling module 3 is connected to microprocessor module 4, microprocessor module 4 is connected with communication module 5, for exporting measured wind speed and receiving control command.

High precision wide range ultrasound wave wind detection method provided by the invention, its method is as described below:

After step one, system electrification, initialization is carried out to each module;

After step 2, initialization complete, detect communication module 5 and whether receive control command, if the control command of receiving, then microprocessor module 4 calls control command process function and processes this order, returns and continue detection after process; If do not receive control command, four transceiver modules 6 that then microprocessor module 4 controls to be connected with four ultrasonic probes 1 successively drive ultrasonic probe 1 to send ultrasound wave, again control simulation switch module 2 successively gating four ultrasonic probes 1 as receiving transducer, and the signal transmission received is carried out analog to digital conversion to AD sampling module 3, its result will pass to microprocessor module 4 and preserve, and altogether obtain the receiving and transmitting signal of four groups of digital quantities;

Step 3, adopt the ultrasonic signal delay time estimation method of the high precision wide range based on fractional order semi-invariant to carry out time delay estimation to the receiving and transmitting signal of four groups of digital quantities respectively, obtain four time delay estimated value t1, t2, t3, t4;

Step 4, establish the distance between two relative ultrasonic probes 1 to be d, then the wind speed obtaining wherein about two corresponding ultrasonic probe 1 directions by relative time error method is the wind speed in about two corresponding ultrasonic probe 1 directions is

Step 5, obtain actual wind speed by orthogonal synthesis and be wind angle is

Step 6, to be exported by communication module 5 at the wind speed and direction angle obtained in step 5, return step 2 afterwards, so circulation obtains real-time wind speed and direction.

The ultrasonic signal delay time estimation method of the high precision wide range based on fractional order semi-invariant described in step 3, its concrete grammar is as follows:

1) determination of fractional order semi-invariant function and standard

(1) determination of fractional order semi-invariant function:

If the fundamental function that Φ (u) is stochastic variable X, have

${C_{RL}}^{kp}={e^{-j\frac{\mathrm{\&pi;}kp}{2}}\frac{{d_{RL}}^{kp}}{{\mathrm{dt}}^{kp}}ln\mathrm{\&Phi;}\left(u\right)|}_{u=0}$

In formula: for left Riemann-Liouville Fractional Derivative, 0 < p≤1, k is arbitrary integer, claims ^rLc ^kpfor the fractional order semi-invariant of stochastic variable X, fractional order semi-invariant ^rLc ^kpalso can be designated as ^rLcum ^kp();

(2) determination of fractional order semi-invariant standard:

Determine that fractional order semi-invariant standard is as follows:

Standard 1: establish a ₁, a ₂..., a _kfor constant, X (k)=[x ₁, x ₂..., x _k] be stochastic variable, then

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;a_1{x}_1^{p_1},a_2{x}_2^{p_2},\mathrm{...},a_k{x}_k^{p_k}\&rsqb;\\ =a_1{a}_2...a_{k}c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;\end{array}$

In formula: kp=p ₁+ p ₂+ ... + p _k

Standard 2: fractional order semi-invariant is symmetrical to its independent variable, their value and the order of independent variable have nothing to do in other words, namely

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;\\ =c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_{i_1}^{p_{i_1}},x_{i_2}^{p_{i_2}},\mathrm{...},x_{i_k}^{p_{i_k}}\&rsqb;\end{array}$

Wherein, i ₁, i ₂..., i _k1,2 ..., an arrangement of k;

Standard 3: if k stochastic variable { x _ia subset and other parts independent, then

$c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;\&equiv;0$

Standard 4: if stochastic variable collection [x ₁, x ₂..., x _k] and [y ₁, y ₂..., y _k] be independently, then have

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;{(x_1+y_1)}^{p_1},{(x_2+y_2)}^{p_2},...,{(x_k+y_k)}^{p_k}\&rsqb;\\ {=}^{RL}{\mathrm{cum}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;{+}^{RL}{\mathrm{cum}}^{kp}\&lsqb;y_1^{p_1},y_2^{p_2},\mathrm{...},y_k^{p_k}\&rsqb;\end{array}$

But

$\begin{array}{c}{\mathrm{mom}}^{kp}\&lsqb;{(x_1+y_1)}^{p_1},{(x_2+y_2)}^{p_2},...,{(x_k+y_k)}^{p_k}\&rsqb;\\ \&NotEqual;{\mathrm{mom}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},\mathrm{...},x_k^{p_k}\&rsqb;+{\mathrm{mom}}^{kp}\&lsqb;y_1^{p_1},y_2^{p_2},\mathrm{...},y_k^{p_k}\&rsqb;\end{array}$

Standard 5: for 2p rank fractional order semi-invariant ^rLc ^kp(τ), when τ=0, maximal value is had, namely

| ^RLC ^kp(τ)|≤ ^RLC ^kp(0)

2) fractional order semi-invariant is to the rejection ability of α noise and Gaussian noise and suppressing method:

α Stable distritation is a kind of generalized Gaussian distribution, and the fundamental function of standard α Stable distritation is:

Φ(u)＝exp{-γ|u| ^α}

In formula: parameter γ > 0 is called dispersion coefficient; Parameter alpha ∈ (0,2] be called characteristic exponent, when characteristic exponent α=2, α Stable distritation deteriorates to Gaussian distribution;

About fractional order semi-invariant to the rejection ability of α noise and Gaussian noise and suppressing method, there is following theorem:

Theorem 1: the fundamental function of the accurate α Stable distritation of bidding, as shown in above formula, makes m be the minimum positive integer being more than or equal to p, then as p > 0 and α > 0 time, the p rank fractional order semi-invariant of standard α Stable distritation is:

(1) when α-p is not integer,

${C_{RL}}^p=\left\{\begin{array}{cc}0,& \mathrm{\&alpha;}-p>0\\ -\mathrm{\&gamma;}\mathrm{\&Gamma;}(\mathrm{\&alpha;}+1)e^{-j\frac{\mathrm{\&pi;}k\mathrm{\&alpha;}}{2}},& \mathrm{\&alpha;}=p\\ \mathrm{\&infin;},& \mathrm{\&alpha;}-p<0\end{array}\right.$

(2) when 1≤p-α≤m is integer,

^RLC ^p＝0

Visible, for the p rank fractional order semi-invariant of standard α Stable distritation signal, when getting p < α, or when 1≤p-α≤m is integer, its p rank fractional order semi-invariant exists and is zero, due to the special case that Gaussian distribution is in standard α Stable distritation when α=2, therefore, fractional order semi-invariant is still set up gaussian signal, this namely fractional order semi-invariant to the rejection condition of α and Gaussian noise and suppressing method, because the fractional order semi-invariant of α noise and Gaussian noise is zero, namely as p < α, mean the suppression completely to these two kinds of noises, therefore, fractional order semi-invariant has extremely strong rejection ability to α noise and Gaussian noise,

3) based on the ultrasonic signal delay time estimation method of fractional order semi-invariant

For ultrasound wind system, due to the impact by ultrasonic probe and spatial electromagnetic interference, hyperacoustic transmitting and receiving signal is all containing noisy, transmit if hyperacoustic into

x ₁(k)＝s(k)+n _α1(k)+n _g1(k)

The Received signal strength of ultrasonic sensor is

x ₂(k)＝βs(k-D)+n _α2(k)+n _g2(k)

In upper two formulas, s (k), s (k-D) are that D is the time delay of Received signal strength without making an uproar transmitting and receiving signal; β is decay factor; n _{α 1}(k), n _{α 2}(k) and n _g1(k), n _g2k () is respectively the adjoint zero-mean α noise of transmitting and receiving signal and Gaussian noise, n _{α 1}(k), n _{α 2}(k), n _g1(k), n _g2k () is separate between two and with transmitting and receiving signal s (k), s (k-D) is separate;

To x ₁(k) and x ₁k (), gets 2p (2p < α≤2) rank fractional order semi-invariant, by canonical function 1,3,4 and the theorem 1 of fractional order semi-invariant, have

$\begin{array}{c}C_{RL}{{}_{x_1{x}_2}}^{2p}\left(\mathrm{\&tau;}\right)=c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(x_1\right(k),x_2(k+\mathrm{\&tau;})\&rsqb;\\ =c_{RL}{\mathrm{um}}^{2p}\&lsqb;s\left(k\right),s(k-D+\mathrm{\&tau;})\&rsqb;+c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{\mathrm{\&alpha;}1}\right(k),n_{\mathrm{\&alpha;}2}(k+\mathrm{\&tau;})\&rsqb;\\ +c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{\mathrm{\&alpha;}1}\right(k),n_{g2}(k+\mathrm{\&tau;})\&rsqb;+c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{g1}\right(k),n_{\mathrm{\&alpha;}2}(k+\mathrm{\&tau;})\&rsqb;\\ +c_{RL}{\mathrm{um}}^{2p}\&lsqb;\left(n_{g1}\right(k),n_{g2}(k+\mathrm{\&tau;})\&rsqb;\\ =C_{RL}{{}_{ss}}^{2p}(\mathrm{\&tau;}-D)\end{array}$

Namely noise is not comprised in last required fractional order semi-invariant, visible the method fully can suppress α noise in measurement environment and Gaussian noise, thus not only there is very high time delay estimadon precision, but also substantially increase the reliability that ultrasonic wind meter works under complex electromagnetic environment, according to the standard 5 of fractional order semi-invariant, as τ-D=0 there is maximal value, therefore have namely after fractional order semi-invariant tried to achieve by microprocessor, it is searched for, find out the τ value that its maximal value is corresponding, the time delay between the ultrasonic transmission/reception signal required by being exactly.

Claims (3)

1. a high precision wide range ultrasound wind system, it is characterized in that: include four ultrasonic probes, analog switch module, AD sampling module, microprocessor module and communication module, wherein four ultrasonic probes are orthogonally set between two, each ultrasonic probe is all connected with transceiver module, four transceiver modules are in parallel, the output terminal of each transceiver module is all connected with the input end of analog switch module, the control end of each transceiver module is all connected with microprocessor module, the output terminal of analog switch module is connected to the input end of AD sampling module, the control end of analog switch module is connected to microprocessor module, the output terminal of AD sampling module is connected to microprocessor module, microprocessor module is connected with communication module, for exporting measured wind speed and receiving control command.

2. a high precision wide range ultrasound wave wind detection method, is characterized in that: its method is as described below:

After step one, system electrification, initialization is carried out to each module;

After step 2, initialization complete, detect communication module and whether receive control command, if the control command of receiving, then microprocessor module calls control command process function and processes this order, returns and continue detection after process; If do not receive control command, four transceiver modules that then microprocessor module controls to be connected with four ultrasonic probes successively drive ultrasonic probe to send ultrasound wave, again control simulation switch module successively gating four ultrasonic probes as receiving transducer, and the signal transmission received is carried out analog to digital conversion to AD sampling module, its result will pass to microprocessor module and preserve, and altogether obtain the receiving and transmitting signal of four groups of digital quantities;

Step 3, adopt the ultrasonic signal delay time estimation method of the high precision wide range based on fractional order semi-invariant to carry out time delay estimation to the receiving and transmitting signal of four groups of digital quantities respectively, obtain four time delay estimated value t1, t2, t3, t4;

Step 4, establish the distance between two relative ultrasonic probes to be d, then the wind speed obtaining wherein about two corresponding ultrasonic probe directions by relative time error method is the wind speed in about two corresponding ultrasonic probe directions is

Step 5, obtain actual wind speed by orthogonal synthesis and be wind angle is

Step 6, to be exported by communication module at the wind speed and direction angle obtained in step 5, return step 2 afterwards, so circulation obtains real-time wind speed and direction.

3. a kind of high precision wide range ultrasound wave wind detection method according to claim 2, it is characterized in that: the ultrasonic signal delay time estimation method of the high precision wide range based on fractional order semi-invariant in described step 3, its concrete grammar is as follows:

1) determination of fractional order semi-invariant function and standard

(1) determination of fractional order semi-invariant function:

If the fundamental function that Φ (u) is stochastic variable X, have

${C_{RL}}^{kp}=e^{-j\frac{\mathrm{\&pi;}kp}{2}}\frac{{d_{RL}}^{kp}}{{\mathrm{dt}}^{kp}}ln\mathrm{\&Phi;}\left(u\right){|}_{u=0}$

In formula: for left Riemann-Liouville Fractional Derivative, 0 < p≤1, k is arbitrary integer, claims ^rLc ^kpfor the fractional order semi-invariant of stochastic variable X, fractional order semi-invariant ^rLc ^kpalso can be designated as ^rLcum ^kp();

(2) determination of fractional order semi-invariant standard:

Determine that fractional order semi-invariant standard is as follows:

Standard 1: establish a ₁, a ₂..., a _kfor constant, X (k)=[x ₁, x ₂..., x _k] be stochastic variable, then

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;a_1{x}_1^{p_1},a_2{x}_2^{p_2},...,a_k{x}_k^{p_k}\&rsqb;\\ =a_1{a}_2...a_{k}c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_1^{p_2},...,x_k^{p_k}\&rsqb;\end{array}$

In formula: kp=p ₁+ p ₂+ ... + p _k

Standard 2: fractional order semi-invariant is symmetrical to its independent variable, their value and the order of independent variable have nothing to do, namely

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},...,x_k^{p_k}\&rsqb;\\ =c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_{i_1}^{p_{i_1}},x_{i_2}^{p_{i_2}},...,x_{i_k}^{p_{i_k}}\&rsqb;\end{array}$

Wherein, i ₁, i ₂..., i _k1,2 ..., an arrangement of k;

Standard 3: if k stochastic variable { x _ia subset and other parts independent, then

$C_{RL}{\mathrm{um}}^{kp}(x_1^{p_1},x_2^{p_2},...,x_k^{p_k})\&equiv;0$

Standard 4: if stochastic variable collection [x ₁, x ₂..., x _k] and [y ₁, y ₂..., y _k] be independently, then have

$\begin{array}{c}c_{RL}{\mathrm{um}}^{kp}\&lsqb;{\left(x_1+y_1\right)}^{p_1},{\left(x_2+y_2\right)}^{p_2},...,{\left(x_k+y_k\right)}^{p_k}\&rsqb;\\ =c_{RL}{\mathrm{um}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},...,x_k^{p_k}\&rsqb;+c_{RL}{\mathrm{um}}^{kp}\&lsqb;y_1^{p_1},y_2^{p_2},...,y_k^{p_k}\&rsqb;\end{array}$

But

$\begin{array}{c}{\mathrm{mom}}^{kp}\&lsqb;{\left(x_1+y_1\right)}^{p_1},{\left(x_2+y_2\right)}^{p_2},...,{\left(x_k+y_k\right)}^{p_k}\&rsqb;\\ \&NotEqual;{\mathrm{mom}}^{kp}\&lsqb;x_1^{p_1},x_2^{p_2},...,x_k^{p_k}\&rsqb;+{\mathrm{mom}}^{kp}\&lsqb;y_1^{p_1},y_2^{p_2},...,y_k^{p_k}\&rsqb;\end{array}$

Standard 5: for 2p rank fractional order semi-invariant ^rLc ^kp(τ), when τ=0, maximal value is had, namely

| ^RLC ^kp(τ)|≤ ^RLC ^kp(0)

2) fractional order semi-invariant is to the rejection ability of α noise and Gaussian noise and suppressing method:

α Stable distritation is a kind of generalized Gaussian distribution, and the fundamental function of standard α Stable distritation is:

Φ(u)＝exp{-γ|u| ^α}

In formula: parameter γ > 0 is called dispersion coefficient; Parameter alpha ∈ (0,2] be called characteristic exponent, when characteristic exponent α=2, α Stable distritation deteriorates to Gaussian distribution;

About fractional order semi-invariant to the rejection ability of α noise and Gaussian noise and suppressing method, there is following theorem:

Theorem 1: the fundamental function of the accurate α Stable distritation of bidding, as shown in above formula, makes m be the minimum positive integer being more than or equal to p, then as p > 0 and α > 0 time, the p rank fractional order semi-invariant of standard α Stable distritation is:

(1) when α-p is not integer,

${C_{RL}}^p=\left\{\begin{array}{cc}0,& \mathrm{\&alpha;}-p>0\\ -\mathrm{\&gamma;}\mathrm{\&Gamma;}(\mathrm{\&alpha;}+1)e^{-j\frac{\mathrm{\&pi;}k\mathrm{\&alpha;}}{2}},& \mathrm{\&alpha;}=p\\ \mathrm{\&infin;},& \mathrm{\&alpha;}-p<0\end{array}\right.$

(2) when 1≤p-α≤m is integer,

^RLC ^p＝0

For the p rank fractional order semi-invariant of standard α Stable distritation signal, when getting p < α, or when 1≤p-α≤m is integer, its p rank fractional order semi-invariant exists and is zero, as p < α, to the suppression completely of these two kinds of noises;

3) based on the ultrasonic signal delay time estimation method of fractional order semi-invariant

For ultrasound wind system, due to the impact by ultrasonic probe and spatial electromagnetic interference, hyperacoustic transmitting and receiving signal is all containing noisy, transmit if hyperacoustic into

x ₁(k)＝s(k)+n _α1(k)+n _g1(k)

The Received signal strength of ultrasonic sensor is

x ₂(k)＝βs(k-D)+n _α2(k)+n _g2(k)

In upper two formulas, s (k), s (k-D) are that D is the time delay of Received signal strength without making an uproar transmitting and receiving signal; β is decay factor; n _{α 1}(k), n _{α 2}(k) and n _g1(k), n _g2k () is respectively the adjoint zero-mean α noise of transmitting and receiving signal and Gaussian noise, n _{α 1}(k), n _{α 2}(k), n _g1(k), n _g2k () is separate between two and with transmitting and receiving signal s (k), s (k-D) is separate;

To x ₁(k) and x ₁k (), gets 2p (2p < α≤2) rank fractional order semi-invariant, by canonical function 1,3,4 and the theorem 1 of fractional order semi-invariant, have

$\begin{array}{c}C_{RL}{{}_{x_1{x}_2}}^{2p}\left(\mathrm{\&tau;}\right)=c_{RL}{\mathrm{um}}^{2p}\&lsqb;x_1\left(k\right),x_2(k+\mathrm{\&tau;})\&rsqb;\\ =c_{RL}{\mathrm{um}}^{2p}\&lsqb;s\left(k\right),s(k-D+\mathrm{\&tau;})\&rsqb;+c_{RL}{\mathrm{um}}^{2p}\&lsqb;n_{\mathrm{\&alpha;}1}\left(k\right),n_{\mathrm{\&alpha;}2}(k+\mathrm{\&tau;})\&rsqb;\\ +c_{RL}{\mathrm{um}}^{2p}\&lsqb;n_{\mathrm{\&alpha;}1}\left(k\right),n_{\mathrm{\&alpha;}2}(k+\mathrm{\&tau;})\&rsqb;+c_{RL}{\mathrm{um}}^{2p}\&lsqb;n_{g1}\left(k\right),n_{\mathrm{\&alpha;}2}(k+\mathrm{\&tau;})\&rsqb;\\ +c_{RL}{\mathrm{um}}^{2p}\&lsqb;n_{g1}\left(k\right),n_{g2}(k+\mathrm{\&tau;})\&rsqb;\\ +C_{RL}{{}_{ss}}^{2p}(\mathrm{\&tau;}-D)\end{array}$

Namely noise is not comprised in last required fractional order semi-invariant, according to the standard 5 of fractional order semi-invariant, as τ-D=0, there is maximal value, therefore have $D=\arg \{max\&lsqb;C_{RL}{{}_{x_1{x}_2}}^{2p}\left(\mathrm{\&tau;}\right)\&rsqb;\},$ Namely after fractional order semi-invariant tried to achieve by microprocessor, it is searched for, find out the τ value that its maximal value is corresponding, the time delay between the ultrasonic transmission/reception signal required by being exactly.