CN101881784B - Position and speed measuring device based on inductosyn or rotary transformer - Google Patents

Abstract

The invention discloses a position and speed measuring device based on an inductosyn or a rotary transformer. The device adopts a phase discrimination type processing method and comprises an exciting circuit, an inductosyn or a rotary transformer, a signal processing circuit and a central processing unit. An exciting signal is generated by the exciting circuit and subject to power amplification, a power signal passes through the inductosyn or the rotary transformer, the signal processing circuit carries out filtering, amplification, phase shift and reshaping on the power signal, and then the processed signal is sent to the central processing unit. The central processing unit obtains position information through the phase information of a signal, obtains speed information through a position information difference, obtains acceleration information through a speed information difference, and carries out dynamic error correction on phase or speed according to speed and acceleration information. The invention is simple to realize, has low cost and high accuracy, is suitable for both a high-speed operation condition and a low-speed operation condition, and can be applied to the fields of aerospace, military, industry and the like to realize dynamic position and speed measurement.

Description

A kind of position and velocity measuring device based on inductosyn or rotary transformer

Technical field

The present invention relates to the measuring technique of speed, position, more particularly, the present invention relates to a kind of position and velocity measuring device based on inductosyn or rotary transformer.

Background technology

Inductosyn and rotary transformer all are to utilize electromagnetic principle to convert displacement of the lines or angular displacement the sensor of electric signal to, and the collocation driving circuit can go out positional information by the amplitude or the phase extraction of electric signal, and two kinds of sensors have identical principle of work.Because the principle of work of inductosyn and rotary transformer is based under the static situation designed, therefore no matter be according to the amplitude discrimination type or the measurement mechanism of phase demodulating type method manufacturing, in mobile or rotation process, all dynamic error can be occurred, even miscount can be caused.

Wherein the phase demodulating type method is as " Harbin University of Science and Technology's journal " method that February in 2003, the 8th volume the 1st phase 12-14 page or leaf " systematic study of a phase demodulating type inductosyn dynamic measuring angle " literary composition was discussed, according to the measurement mechanism of its principle manufacturing as shown in Figure 1, comprise exciting circuit, inductosyn or rotary transformer, signal processing circuit, CPU (central processing unit), excited signal is by the exciting circuit generation and carry out power amplification, after power signal passes through inductosyn or rotary transformer, carry out filtering by signal processing circuit, amplify, phase shift, the signal that obtains is shown in (1) formula:

$e=k(1-\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})\sin ({\mathrm{\ω}}_{0}t-{\mathrm{\θ}}_t)---\left(1\right)$

Wherein k is a constant, θ _tBe signal phase, represented the mechanical angle that turns in segmentation cycle, ω _tBe the speed of inductosyn or rotary transformer motion, ω ₀Be the excited signal angular frequency.(1) signal e is identified its phase information and is obtained positional information by CPU (central processing unit) through after the shaping in the formula, and obtains velocity information by the positional information difference, and the process flow diagram of CPU (central processing unit) internal program computing as shown in Figure 2.According to the measurement mechanism of this method manufacturing, its advantage is: need not produce tracking signal or carry out multiplying, therefore than amplitude discrimination type method realize convenient, cost is low.Weak point is: because the mechanical angle of motion itself also is a variations per hour, when rotating, can cause the variation of signal frequency, and shown in (2) formula,

$e=k(1-\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})\sin [{\mathrm{\ω}}_{0}t-{\mathrm{\ω}}_{t}t-{{\mathrm{\θ}}_t}^{\′}]$

$=k(1-\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})\sin [(1-\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}){\mathrm{\ω}}_{0}t-{{\mathrm{\θ}}_t}^{\′}]---\left(2\right)$

The phase information θ that identifies by (2) formula _t' and the mechanical angle of actual process is also unequal, brings dynamic error, can make that simultaneously saltus step appears in measurement result when inductosyn turns over zero point in segmentation cycle, as shown in Figure 3, and makes speed calculation mistake occur in this sampling period, as shown in Figure 4.

Summary of the invention

Based on the problems referred to above, the objective of the invention is to design a kind of simple in structure, with low cost, position and velocity measuring device that accuracy is high based on inductosyn or rotary transformer.

The present invention is the improvement of phase demodulating type measurement mechanism, comprise exciting circuit, inductosyn or rotary transformer, signal processing circuit, CPU (central processing unit), excited signal is by the exciting circuit generation and carry out power amplification, after power signal passes through inductosyn or rotary transformer, undertaken sending into CPU (central processing unit) after filtering, amplification, phase shift, the shaping by signal processing circuit, the treatment step of CPU (central processing unit) internal program is:

1) phase information of identification signal;

2) judge whether to carry out the phase place correction, revise, then revise with following formula as needs:

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})$

Or

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})$

Or

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t[1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}+{\mathrm{\θ}}_t\frac{a{\mathrm{\ω}}_t}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}]$

θ wherein _t' be revised phase information, θ _tBe the phase information before revising, ω ₀Be excited signal angular frequency, ω _tBe present speed, a is a current acceleration;

3) obtain positional information by phase information, and set up storehouse, the degree of depth of storehouse is more than 2 or 2;

4) judge whether to carry out position correction, as need not to revise, current location information is pushed on, revise, then revise according to the following steps as needs, its process flow diagram as shown in Figure 5:

A) judge the direction that moves or rotate, identical with the phase place growing direction as direction of motion then is forward, otherwise then is reverse;

B) when counter motion, judge that whether this sampling period is through zero crossing, as without zero point, then the positional information that this sampling period is extracted pushes on, as through zero crossing, then two positional informations are pushed on successively, first is a pairing positional value at this zero point, and second is the positional value that extracts in the current sampling period;

C) when positive movement, judge whether this sampling period extract phase information, if can extract, then the positional information that this sampling period is extracted pushes on, as can not extract phase information, and this sampling period is not done any operation;

5) obtain velocity information by the positional information difference in the storehouse;

6) judge whether to carry out the speed correction, revise, then revise with following formula as needs:

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})+a{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_0}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^2}$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})+a{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_0^2-{\mathrm{\ω}}_0{\mathrm{\ω}}_t+2{\mathrm{\ω}}_t^2}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}$

ω wherein _t' be revised speed, ω _tBe the speed before revising;

7) the velocity information difference that was obtained by a present speed information and a last sampling period obtains acceleration information, and degree of will speed up information is used for the correction to phase place and speed.

CPU (central processing unit) to the Signal Processing flow process as shown in Figure 6.

The present invention has inherited existing phase demodulating type apparatus structure advantage simple, with low cost, also has higher dynamic accuracy simultaneously, and the speed calculation mistake can not occur through zero point in segmentation cycle the time.

It below is the proof of correction algorithm.For better understanding the principle of correction algorithm, as shown in Figure 7 with the signal waveform modeling that observed on the oscillograph.It is ω with speed that the inductosyn output signal is regarded as one _tThe motion object, and sampling thief initially return in each cycle once sampled result and with ω ₀Speed begin to catch up with it, when the two meets, with the position of meeting as sampled result, the quantification progression D=f of sampled result _c/ f ₀, promptly each DN value is represented the electric segmentation angle of 2 π/D, wherein ω _tBe inductosyn speed, ω ₀Be excited signal angular frequency, f _cBe sample frequency, f ₀Be the excited signal frequency.Like this, system just has been transformed into the problem of meeting of two objects to signals sampling, and sample frequency then just causes quantization error to the differentiation of Encounter Time, and we claim this model to be the model that meets.Be without loss of generality, the acceleration of supposing inductosyn is α, and the even rotation of quickening is done by system in the same sampling period.In addition, owing to quantize, the full scale sampling has only (D-1)/D cycle, thus system forward and reverse pass through inductosyn during zero point situation slightly different.

When antiport, suppose T ₀In time, is engraved in the initial position in certain sampling period and samples first rising edge, owing to should just just begin in the sampling period this moment, and system will just can return sampled result after the sampling period finishes, therefore behind one-period nearly promptly at T ₀+ Δ T returns the result 0 of this time sampling constantly and begins new round sampling.This moment, inductosyn passed through-ω _tΔ T-α Δ T ²/ 2 displacement, wherein Δ T=2 π/ω ₀Be the sampling period.So, according to the model that meets, the time that rising edge experienced that next time samples signal satisfies:

$2\mathrm{\π}-{\mathrm{\ω}}_t\mathrm{\ΔT}-\frac{1}{2}\mathrm{\α\Δ}{T}^2=({\mathrm{\ω}}_0+{\mathrm{\ω}}_t+\mathrm{\α\ΔT})t+\frac{1}{2}\mathrm{\α}{t}^2$

Separate this equation and omit high-order term and get:

$t_1=\frac{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0+{\mathrm{\ω}}_t}\mathrm{\ΔT}-\frac{16\mathrm{\α}{\mathrm{\π}}^2}{{({\mathrm{\ω}}_0+{\mathrm{\ω}}_t)}^3}$

$\≈\mathrm{\ΔT}---\left(3\right)$

From (3) formula as can be seen, acceleration alpha only has influence on the cubic term of result of calculation, therefore can temporarily ignore situation about only considering at the uniform velocity.In this time, the current sampling period finishes soon, and system will return when time sampled result at once.That is to say, at T ₀+ 2 Δ T constantly, when system returned for the second time sampled result, inductosyn reality had turned over-2 ω approximately _tThe angle of Δ T, and system can only judge by accident-ω _tThe angle of Δ T, error-ω _tΔ T.The model of system is shown in Fig. 7 (a).

When forward turns round, suppose at T ₀Constantly, sample the rising edge of signal in the last moment of certain one-period, and at T ₀+ Δ t returns the D-1 as a result of this time sampling constantly, begins new round sampling simultaneously.This moment, inductosyn passed through ω _tΔ t+ α Δ t ²/ 2 displacement, wherein Δ t=Δ T/D is a sampling interval.So, according to the model that meets, sample the needed time of system's rising edge satisfies next time:

$2\mathrm{\π}\frac{D-1}{D}+{\mathrm{\ω}}_t\mathrm{\Δt}+\frac{1}{2}\mathrm{\α}{\mathrm{\Δt}}^2=({\mathrm{\ω}}_0-{\mathrm{\ω}}_t-\mathrm{\α\Δt})t-\frac{1}{2}\mathrm{\α}{t}^2$

Separate this equation and omit high-order term and get:

$t_2=\frac{{\mathrm{\ω}}_0}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}\mathrm{\ΔT}+\mathrm{\Δt}+\frac{4\mathrm{\α}{\mathrm{\π}}^2}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}$

$>\mathrm{\ΔT}---\left(4\right)$

From (4) formula as can be seen, angular acceleration still in the cubic term of result of calculation, can temporarily be ignored, and only considers situation at the uniform velocity.This moment, the inductosyn signal entered next cycle, can not meet with sampled signal in this cycle.Changing a kind of saying is exactly that the inductosyn output signal frequency is greater than f ₀, and the zero clearing again of this hour counter.Sampling thief may return last sampled result D-1 like this, also may return a result 0, and this depends on the setting of system itself.Supposing the system is to return last sampled result, and will judge inductosyn so by accident is static at this moment, and in fact inductosyn has turned over about ω _tThe angle of Δ T, error ω _tΔ T.At this moment system model is shown in Fig. 7 (b).

According to the analysis of front, for the system of uniform rotation, dynamic error can be accumulated to maximal value and reach (p/360) (ω when being about to by zero point _t/ ω ₀), wherein p is the inductosyn number of pole-pairs.In all the other angles, error should have following form so:

$\mathrm{\Δ}{{\mathrm{\θ}}_{t1}}^{\′}=-{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}---\left(5\right)$

(5) formula is a modifying factor of dynamic error, and (2) formula then becomes like this:

$e=k\sin [(1-\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}){\mathrm{\ω}}_{0}t-{\mathrm{\θ}}_t\left(t\right)+\underset{0}{\overset{t}{\∫}}{\mathrm{\θ}}_t\left(t\right)\frac{{\mathrm{\ω}}_t\left(t\right)}{{\mathrm{\ω}}_0}\mathrm{dt}]---\left(6\right)$

Can release the error of correction algorithm by (3), (4) and (5) formula, at first consider the situation of uniform rotation:

$\left|\mathrm{\Δ}{{\mathrm{\θ}}_t}^{\′\′}\right|\≤|\mathrm{\ΔT}-\frac{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0+{\mathrm{\ω}}_t}\mathrm{\ΔT}|\·2\mathrm{\π}\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}/2\mathrm{\ΔT}$

$=\frac{2\mathrm{\π}{\mathrm{\ω}}_t^2}{({\mathrm{\ω}}_0+{\mathrm{\ω}}_t){\mathrm{\ω}}_0}---\left(7\right)$

(7) formula is to be obtained by (3) formula and (5) formula, because actually when inductosyn is discussed oppositely by zero point considered two sampling periods, and therefore will be divided by 2 Δ T.Simultaneously, can obtain by (4) formula and (5) formula

$\left|\mathrm{\Δ}{{\mathrm{\θ}}_t}^{\′\′}\right|\≤|\mathrm{\ΔT}-\frac{{\mathrm{\ω}}_0}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}\mathrm{\ΔT}-\mathrm{\Δt}|\·2\mathrm{\π}\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}/\mathrm{\ΔT}$

$=|\frac{2\mathrm{\π}{\mathrm{\ω}}_t^2}{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t){\mathrm{\ω}}_0}+2\mathrm{\π}\frac{{\mathrm{\ω}}_t}{D{\mathrm{\ω}}_0}|---\left(8\right)$

If sample frequency is enough high, can regard D → ∞ as, (8) formula then becomes so

$\left|\mathrm{\Δ}{{\mathrm{\θ}}_t}^{\′\′}\right|\≤\frac{2\mathrm{\π}{\mathrm{\ω}}_t^2}{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t){\mathrm{\ω}}_0}---\left(9\right)$

After considering direction, (7) formula has identical form with (9) formula.The error that adds this two formula, (5) formula should be modified to:

$\mathrm{\Δ}{{\mathrm{\θ}}_{t2}}^{\′}=-{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}-{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_t^2}{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t){\mathrm{\ω}}_0}$

$=-{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}---\left(10\right)$

Under the situation of variable motion, need to add the cubic term in (3) formula and (4) formula.Because (3) formula has been analyzed the situation in two sampling periods, square after and the cubic term of (4) formula 4 times relation is arranged.Therefore, in conjunction with the form of (7), (9) and (10) formula, the modifying factor in the time of can releasing variable motion is:

$\mathrm{\Δ}{{\mathrm{\θ}}_{t3}}^{\′}=-{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}-{\mathrm{\θ}}_t^2\frac{a{\mathrm{\ω}}_t}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}---\left(11\right)$

(5) formula, (10) formula and (11) formula be respectively the phase place dynamic error once, secondary and three modifying factors.Respectively to (5) formula, (10) formula and (11) formula differentiate can obtain the speed dynamic error once, secondary and three modifying factors are:

$\mathrm{\Δ}{\mathrm{\ω}}_{t1}=-{\mathrm{\ω}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}-{\mathrm{\θ}}_t\frac{a}{{\mathrm{\ω}}_0}$

$\≈-{\mathrm{\ω}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0}---\left(12\right)$

$\mathrm{\Δ}{\mathrm{\ω}}_{t2}=-{\mathrm{\ω}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}-{\mathrm{\θ}}_t\frac{a{\mathrm{\ω}}_0}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^2}$

$\≈-{\mathrm{\ω}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}---\left(13\right)$

$\mathrm{\Δ}{\mathrm{\ω}}_{t3}=-{\mathrm{\ω}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}-{\mathrm{\θ}}_t\frac{a{\mathrm{\ω}}_0}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^2}-2{\mathrm{\ω}}_t{\mathrm{\θ}}_t\frac{a{\mathrm{\ω}}_t}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}-{\mathrm{\θ}}_t^2\frac{a^2({\mathrm{\ω}}_0+2{\mathrm{\ω}}_t)}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^4}$

$\≈-{\mathrm{\ω}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}-a{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_0^2-{\mathrm{\ω}}_0{\mathrm{\ω}}_t+2{\mathrm{\ω}}_t^2}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}$

$\≈-{\mathrm{\ω}}_t\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}-a\frac{{\mathrm{\θ}}_t{\mathrm{\ω}}_0}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^2}$

(14)

For position and velocity accuracy the system of requirement is arranged all, only need that phase place is carried out the dynamic error correction and get final product, with position and speed that revised phasometer is calculated, its dynamic error is corrected.And not high but to position accuracy demand to the demanding system of velocity accuracy, can only carry out the dynamic error correction and do not need the position is revised speed.Traditional rate algorithm only writes down a positional information that extracts in the past, and obtains speed with the current positional information difference that extracts, and this method does not have immunity to position sudden change shown in Figure 2, causes the speed calculation mistake.For obtaining continuous position information, need carry out interpolation processing, also need to write down at least simultaneously the positional information that two or more extract in the past.Interpolation processing is equivalent to the zero degree correction to speed, is particularly useful for this system that is bad at the chip of floating-point operation as central processing element with FPGA, and angle information reduces greatly through its dynamic error of velocity information that calculates after the interpolation processing.The present invention uses the method recording angular information of storehouse, so the degree of depth of storehouse is more than 2 or 2, and the used positional information of computing velocity is extracted in storehouse, the processing that once pushes on of cycle of promptly extracting phase place or position in each sampling period.The step of interpolation algorithm and being contemplated that:

A) judge the direction that moves or rotate, identical with the phase place growing direction as direction of motion then is forward, otherwise then is reverse.

B) when counter motion, because of signal frequency becomes greatly, the cycle shortens, and need insert a positional information in a segmentation cycle, this positional information is pairing position at this segmentation zero point in cycle.Therefore need whether judge this sampling period that as without zero point, then the positional information with the extraction of this sampling period pushes on through zero crossing; As through zero crossing, then two positional informations are pushed on successively, first is a pairing positional value at this zero point, second is the positional value that extracts in the current sampling period.

C) when positive movement, because of signal frequency diminishes, the cycle is elongated, last interior sampling period of segmentation cycle can not extracted phase information, and the positional information of register memory storage is the positional information of extracting in a last sampling period, need throw away this wrong positional information.Therefore need judge whether this sampling period extract phase information, if can extract that then the positional information with the extraction of this sampling period pushes on; As can not extract phase information, and expression is about to through zero crossing, and this sampling period is not done any operation, promptly is equivalent to throw away the positional information of a mistake.

The process flow diagram of the interpolation and the algorithm that pushes on as shown in Figure 4.

Description of drawings

Fig. 1 is based on the measurement mechanism basic structure block diagram of inductosyn or rotary transformer.

Fig. 2 tradition phase demodulating type measurement mechanism CPU (central processing unit) program flow diagram.

Positional information saltus step synoptic diagram during Fig. 3 tradition phase demodulating type measurement mechanism zero crossing.

Speed miscount synoptic diagram during Fig. 4 tradition phase demodulating type measurement mechanism zero crossing.

Fig. 5 interpolation stack algorithm of the present invention process flow diagram.

Fig. 6 CPU (central processing unit) program flow diagram of the present invention.

Fig. 7 dynamic error correction algorithm of the present invention model.

Embodiment

The present invention includes exciting circuit, inductosyn or rotary transformer, signal processing circuit, CPU (central processing unit), excited signal is by the exciting circuit generation and carry out power amplification, after power signal passes through inductosyn or rotary transformer, undertaken sending into CPU (central processing unit) after filtering, amplification, phase shift, the shaping by signal processing circuit, the treatment step of CPU (central processing unit) internal program is:

1) phase information of identification signal;

2) judge whether to carry out the phase place correction, also there is requirement the position as system when the kinetic measurement, then need earlier phase place to be revised, correction formula is derived by (5), (10), (11) formula:

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})---\left(15\right)$

Or

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})---\left(16\right)$

Or

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t[1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}+{\mathrm{\θ}}_t\frac{a{\mathrm{\ω}}_t}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}]---\left(17\right)$

θ wherein _t' be revised phase information, θ _tBe the phase information before revising, ω ₀Be excited signal angular frequency, ω _tBe present speed, a is a current acceleration, and above formula is selected one according to the system design index, the error maximum of (15) formula, the error minimum of (17) formula;

3) obtain positional information by phase information, and set up storehouse, the degree of depth of storehouse is more than 2 or 2;

4) judge whether to carry out position correction,, then need not to carry out again position correction as phase place being revised before; As need not to revise, current location information is pushed on, revise as needs, then revise according to the following steps, its process flow diagram as shown in Figure 4:

A) judge the direction that moves or rotate, identical with the phase place growing direction as direction of motion then is forward, otherwise then is reverse;

B) when counter motion, judge that whether this sampling period is through zero crossing, as without zero point, then the positional information that this sampling period is extracted pushes on, as through zero crossing, then two positional informations are pushed on successively, first is a pairing positional value at this zero point, second is the positional value that extracts in the current sampling period

C) when positive movement, judge whether this sampling period extract phase information, if can extract, then the positional information that this sampling period is extracted pushes on, as can not extract phase information, and this sampling period is not done any operation;

5) obtain velocity information by the positional information difference in the storehouse;

6) judge whether to carry out the speed correction,, then need not speed is revised as phase place being revised before; As correction being carried out in the position before, then can revise speed, can not revise yet, depend on the system design index, high as the speed dynamic accuracy is required, then need to carry out the speed correction; Revise as needs, then revise with following formula:

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})---\left(18\right)$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})---\left(19\right)$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})+a{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_0}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^2}---\left(20\right)$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})+a{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_0^2-{\mathrm{\ω}}_0{\mathrm{\ω}}_t+2{\mathrm{\ω}}_t^2}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}---\left(21\right)$

ω wherein _t' be revised speed, ω _tBe the speed before revising, above formula is selected one according to the system design index, the error maximum of (18) formula, the error minimum of (21) formula;

7) the velocity information difference that was obtained by a present speed information and a last sampling period obtains acceleration information, and degree of will speed up information is used for the correction to phase place and speed.

CPU (central processing unit) to the Signal Processing flow process as shown in Figure 5.

Claims (1)

1. position and speed measurement method based on an inductosyn or a rotary transformer, adopt the phase demodulating type disposal route, comprise exciting circuit, inductosyn or rotary transformer, signal processing circuit, CPU (central processing unit), excited signal is by the exciting circuit generation and carry out power amplification, after power signal passes through inductosyn or rotary transformer, undertaken sending into CPU (central processing unit) after filtering, amplification, phase shift and the shaping by signal processing circuit, it is characterized in that: the step of described CPU (central processing unit) internal program operation is:

1) phase information of identification signal;

2) judge whether to carry out the phase place correction, revise, then revise with following formula as needs:

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})$

Or

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})$

Or

${{\mathrm{\θ}}_t}^{\′}={\mathrm{\θ}}_t[1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t}+{\mathrm{\θ}}_t\frac{a{\mathrm{\ω}}_t}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}]$

θ wherein _t' be revised phase information, θ _tBe the phase information before revising, ω ₀Be excited signal angular frequency, ω _tBe present speed, a is a current acceleration;

3) obtain positional information by phase information, and set up storehouse, the degree of depth of storehouse is more than 2;

4) judge whether to carry out position correction, as need not to revise, current location information is pushed on, revise, then revise according to the following steps as needs:

A) judge the direction that moves or rotate, identical with the phase place growing direction as direction of motion then is forward, otherwise then is reverse;

B) when counter motion, judge that whether this sampling period is through zero crossing, as without zero point, then the positional information that this sampling period is extracted pushes on, as through zero crossing, then two positional informations are pushed on successively, first is a pairing positional value at this zero point, and second is the positional value that extracts in the current sampling period;

C) when positive movement, judge whether this sampling period extract phase information, if can extract, then the positional information that this sampling period is extracted pushes on, as can not extract phase information, and this sampling period is not done any operation;

5) obtain velocity information by the positional information difference in the storehouse;

6) judge whether to carry out the speed correction, revise, then revise with following formula as needs:

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0})$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})+a{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_0}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^2}$

Or

${{\mathrm{\ω}}_t}^{\′}={\mathrm{\ω}}_t(1+\frac{{\mathrm{\ω}}_t}{{\mathrm{\ω}}_0-{\mathrm{\ω}}_t})+a{\mathrm{\θ}}_t\frac{{\mathrm{\ω}}_0^2-{\mathrm{\ω}}_0{\mathrm{\ω}}_t+2{\mathrm{\ω}}_t^2}{{({\mathrm{\ω}}_0-{\mathrm{\ω}}_t)}^3}$

ω wherein _t' be revised speed, ω _tBe the speed before revising;

7) the velocity information difference that was obtained by a present speed information and a last sampling period obtains acceleration information, and degree of will speed up information is used for the correction to phase place and speed.

Images