CN102680173A - Control method for improving tone quality of single-rotor compressor of inverter air conditioner - Google Patents

Abstract

The invention discloses a control method for improving tone quality of a single-rotor compressor of an inverter air conditioner. When the working rotating speed of a rotor is higher than a first-order critical rotating speed, a rigid rotor becomes a flexible rotor, and the high-speed flexible rotor is balanced by adopting a limit balance weight-based least square influence coefficient dynamic balance method. Based on many actual vibration testing data, the size of a balance weight required when dynamic balance measurement value of the compressor flexible rotor is more than the balanced surface value is figured out by adopting a least square method and using a linear relation between the correction quantity and the measured value in a linear system, and the optimal balance position is found according to a constraint condition, therefore, the unbalance response caused by bending deformation of the high-speed rotor is eliminated. According to the control method, the balance of force and torque is taken into account and the unbalance response caused by bending deformation of the flexible rotor is taken into account, thus the size of unbalance quantity, the unbalance position and the like of the flexible rotor under the change of the rotating speed can be effectively determined; and the control method has the characteristics of flexibility in operation and low low-frequency noise.

Description

Improve the control method of the single-rotor compressor tonequality of transducer air conditioning

Technical field

The present invention relates to a kind of control method of improving the single-rotor compressor tonequality of transducer air conditioning.

Background technology

The DC voltage that the DC frequency-changeable compressor of direct current varied-frequency air conditioner is given motor through change changes rotating speed of motor; There is not the electromagnetic induction effect of stator rotating magnetic field to rotor in DC frequency-changeable compressor; The electromagnetic noise and the rotor loss of AC frequency conversion compressor have been overcome; Have than invariable frequency compressor efficient is high and hang down characteristics with noise, the efficient of the efficiency ratio invariable frequency compressor of DC frequency-changeable compressor high about 30%; Although direct current varied-frequency air conditioner has outstanding energy saving and comfortableness,, when the noise test of the off-premises station of transducer air conditioning, find that also noise A weighted sound pressure level value is not high, " roars of laughter roars of laughter " sound of its lower frequency is irksome.Studying carefully its main cause is because the eccentric wheel of single-rotor compressor and the barycenter of rotary piston do not overlap with the rotation center of main shaft; Thereby in the compressor operation process, can produce rotary centrifugal force, the low frequency that the uneven vibration that produces of rotor-support-foundation system causes " roars of laughter roars of laughter " sound.

The method that tradition is subdued rotary centrifugal force is to adorn a suitable counterbalance weight at the reverse direction of eccentric mass; But because the structure singularity of single-rotor compressor; Can't counterbalance weight be installed at the reverse direction of eccentric mass; So take other rotatable parts beyond eccentric wheel to add counterbalance weight at present, promptly on rotor, install the first counterbalance weight m additional ₁With the second counterbalance weight m ₂Give balance, see shown in the accompanying drawing 1.

Can set up the power of this power system and the balance equation of moment is according to mechanics principle:

$\left\{\begin{array}{c}{F}_{m_1}+F_{m_e}=F_{m_2}\\ F_{m_1}{L}_1=F_{m_2}{\mathrm{<figure-callout\; id="2"\; label="L"\; filenames="HDA00001619772500011.png"\; state="\{\{state\}\}">L</figure-callout>}}_2\end{array}\right.$ Formula one,

In the formula,

Be the first counterbalance weight m ₁Rotary centrifugal force (N),

Be the second counterbalance weight m ₂Rotary centrifugal force (N),

Be eccentric mass m _eRotary centrifugal force (N), r ₁, r ₂Be respectively the first counterbalance weight m ₁, the second counterbalance weight m ₂Radius of turn (m); E is eccentric mass m _eRadius of turn (m); L ₁, L ₂Be respectively

, F _mDistance

Length (m).

Can obtain the first counterbalance weight m according to equilibrium condition ₁, the second counterbalance weight m ₂Value be:

$\left\{\begin{array}{c}{m}_1=\frac{{\mathrm{EL}}_2}{r_1{L}_1}{m}_e\\ {\mathrm{<figure-callout\; id="2"\; label="m"\; filenames="HDA00001619772500011.png"\; state="\{\{state\}\}">m</figure-callout>}}_2=\frac{e}{r_2}(1+\frac{L_2}{L_1})m_e\end{array}\right.$ Formula two,

Formula two for present single rotor frequency-changeable compressor in order to subdue the uneven method that is adopted of rotor, can find out that from this method the single rotor frequency-changeable compressor is not considered at present: 1. rotor is a flexible rotor.Usually calling rigid rotator being operated in the rotor of rotating speed less than first critical speed 70%, for the balance of rigid rotator, mainly is the equilibrium problem of solution power and moment; But; For transducer air conditioning working rotor rotating speed is 60 ~ 9000rpm, has been higher than first critical speed during high rotating speed, and the rotor of this moment is called flexible rotor; Balance for flexible rotor; Not only will consider the balance of power and moment, also will consider flexible rotor because the unbalance response that causes is out of shape in rotor bow, size, the phase place of flexible rotor amount of unbalance all change along with change in rotational speed.2. stator and the air gap between the rotor in the compressor is inhomogeneous.The radial electromagnetic force of motor stator and rotor and tangential electromagnetic torque are inhomogeneous, make that air gap is inhomogeneous, and uneven air gap causes the bigger unbalance dynamic of rotor.

In order to subdue the uneven shortcoming of frequency conversion single-rotor compressor effectively, normal at present employing the: 1. frequency conversion two-spool compressor, 180 ° of symmetrical distributions of its two rolling pistons, twice of each suction and discharge in changeing of compressor; Balance rotating inertial force, rotatory inertia moment, particularly cold media gas power effectively because the transient equilibrium characteristic is good, make compressor to turn round effectively from low speed to high speed, and its vibration, noise are all relatively little.But, relative frequency conversion single-rotor compressor, its manufacturing process is complicated, cost is higher.2. adopt the multipole motor of multiple-grooved, reduce the unevenness of electromagnetic force and electromagnetic torque, still, so also increase winding, thereby cause the raising of cost.So seek a kind ofly to subdue frequency conversion single-rotor compressor unbalance dynamic method simply and effectively, and then improve its design, improve its quality, to the new market of developing transducer air conditioning with consolidate the existing market and have very important meaning.

Summary of the invention

The object of the invention aims to provide the control method of a kind of flexible operation, the single-rotor compressor tonequality of improving transducer air conditioning that low-frequency noise is low, to overcome weak point of the prior art.

A kind of control method of improving the single-rotor compressor tonequality of transducer air conditioning by this purpose design is characterized in that the rotor-bearing system to a linearity, and under rotational speed omega, rotor is at the vibration A at i point place _i, with the amount of unbalance U at j point place _jBetween exist following relation:

A _i=α _Ij(ω) U _jFormula three,

In the formula, α _Ij(ω) be shadow coefficient under the rotational speed omega, the relation between the amount of unbalance at j point place vibrates with i point place on the expression rotor; Wherein, i ≠ j,

For asking influence coefficient α _Ij(ω), at first under certain rotating speed, measure the vibration A at the i point place that the rotor original imbalance causes _I0(amplitude and phase place) adds a known amount of unbalance U then on plane j point _j, measure the vibration A at i point place again _IjSo influence coefficient is:

${\mathrm{\&alpha;}}_{\mathrm{Ij}}\left(\mathrm{\&omega;}\right)=\frac{A_{\mathrm{Ij}}-A_{i0}}{U_j}$ Formula four,

The vibration of supposing rotor-bearing system meets linear conditions, and is promptly proportional and phasing degree that vibrate is a constant between amplitude under certain rotating speed and little aequum; Therefore, put irrelevantly with increasing the weight of test mass size and angular phasing on the face, have nothing to do with epitrochanterian original unbalance condition by the value of the influence coefficient of formula four definition; Also promptly under certain rotating speed, in the range of linearity, influence coefficient α _IjValue (ω) is a constant; Increase the weight of face and be the facet that the j point adds counterbalance weight,

If increase the weight of face j=1,2 ..., M, the vibration measuring point i=1 that chooses, 2 ..., N comprises all measuring points under the different balancing speeds, should make each balanced surface add the counterweight amount under the ideal situation or deducts weight Q ₁, Q ₂..., Q _MAfter, each vibration of vibrating measuring point is zero, can get:

$\underset{j=1}{\overset{M}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_{\mathrm{Ij}}{Q}_j+A_{i0}=0$ I=1,2 ..., N formula five,

Write as matrix form:

$\left[\begin{array}{cccc}{\mathrm{\&alpha;}}_{11}& {\mathrm{\&alpha;}}_{12}& ...& {\mathrm{\&alpha;}}_{1M}\\ {\mathrm{\&alpha;}}_{21}& {\mathrm{\&alpha;}}_{22}& ...& {\mathrm{\&alpha;}}_{2M}\\ .& .& & .\\ .& .& & .\\ .& .& & .\\ {\mathrm{\&alpha;}}_{N1}& {\mathrm{\&alpha;}}_{\mathrm{<figure-callout\; id="2"\; label="N"\; filenames="HDA00001619772500011.png"\; state="\{\{state\}\}">N</figure-callout>}2}& ...& {\mathrm{\&alpha;}}_{\mathrm{Nm}}\end{array}\right]\left\{\begin{array}{c}{Q}_1\\ {\mathrm{<figure-callout\; id="2"\; label="Q"\; filenames="HDA00001619772500011.png"\; state="\{\{state\}\}">Q</figure-callout>}}_2\\ .\\ .\\ .\\ Q_M\end{array}\right\}+\left\{\begin{array}{c}{A}_{10}\\ A_{20}\\ .\\ .\\ .\\ A_{N0}\end{array}\right\}=0$ Formula six,

When M=N, and N=ab, wherein a is the number of balancing speed, and b is a measuring point number under the rotating speed, and N is total measuring point number, and formula five has exact solution, and promptly the correcting value on each plane is:

Q _j=-α _Ij ^-1A _I0Formula seven,

Usually increase the weight of and limited, promptly M N, formula five are the overdetermined equation group, can only try to achieve approximate solution, utilize least square method to find the solution:

Q _j=-((α _Ij ^*) ^Tα _Ij) ^-1(α _Ij ^*) ^TA _I0Formula eight,

Wherein, α _Ij ^*Be α _IjGrip matrix altogether;

The residual oscillation of each measuring point is:

δ _j=α _IjQ _j+ A _I0Formula nine,

The counterweight amount is unit usually with g, because in actual engineering, the counterweight amount that the high speed flex rotor allows to install usually is restricted, therefore, transient equilibrium should reach satisfied balance result with as far as possible little balance mass;

Because least square method puts undue emphasis on the minimization residual oscillation to the scope of counterweight amount addition of constraints not; Therefore; Use the least square influence coefficient method, just obtain the method for influence coefficient, in application, can cause balance mass bigger with least square method; Can't realize that in some occasion this need impose restriction to the counterweight amount;

To consider that also residual oscillation can not be too big when the counterweight amount meets the demands certainly, could obtain satisfied result;

According to the minimized thought of weighted iteration least square method root mean square, make the quadratic sum of counterweight amount minimum, thereby set up the objective function constraints:

$\mathrm{Min}p=\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_j{\left|Q_j\right|}^2$ Formula 11,

$s.t.\left\{\begin{array}{c}\max \left|Q_j\right|<Q_{\lim }\\ \max \left|{\mathrm{\&delta;}}_j\right|<{\mathrm{\&delta;}}_{\lim }\end{array}\right.$

Wherein, Q _LimBe the permissible value of maximum counterweight amount, δ _LimBe the permissible value of maximum residual oscillation, C _jIt is the weight coefficient of the balance of weights face of j balanced surface;

Least square method is a kind of statistic algorithm, and weighted iteration least square method root mean square is least square method a kind of algorithm wherein,

This method is applicable to that the counterweight amount that can install additional on the balanced surface is very limited, but the residual oscillation value is required not high situation, promptly in the constraint condition

First formula: max|Q _j|<q _LimBe main determining factor, should control its upper limit according to the actual conditions strictness, and

Second formula: max| δ _j|<δ _Lim, upper limit broad, it is too little that maximum residual oscillation value needn't limit;

Set the permissible value Q of counterweight amount _LimIf the counterweight amount of j balanced surface calculating gained is greater than Q _Lim, then the weight coefficient on this plane is:

${C_j}^{\left(s\right)}=\frac{\left|Q_j\right|}{Q_j\&times;{10}^6}+{C_j}^{(s-1)}$ Formula 12,

In the formula, s represents iteration the s time, and C _j ⁽⁰⁾=0; At this moment, balanced surface counterweight amount is:

{ Q _j([the α of }=- _Ij ^*] ^T[α _Ij]+C _j) ^-1[α _Ij ^*] ^T{ A _I0Formula 13,

The computation process of the least square method of restriction counterweight is following: at first set counterweight amount permissible value Q _Lim, calculate the counterweight amount according to primary data with least square method, whether check satisfies constraint condition, then need add a weight coefficient and recomputates the counterweight amount if do not satisfy, revise repeatedly weight coefficient up to the counterweight amount less than Q _Lim, and then calculate residual oscillation, if meeting the demands, residual oscillation can finish to calculate; Otherwise, carry out calculating the counterweight amount again behind the weighted iteration, if the counterweight amount is still calculated less than promptly finishing; Otherwise, carry out cycle calculations again according to above-mentioned steps;

Because the counterweight amount reduces, the residual oscillation value can rise to some extent, if the maximum residual permissible value δ that shakes _LimBe provided with too for a short time, will can not jump circulation; Therefore, need only the counterweight amount in specialized range, and residual oscillation is little, can stop iteration, and as net result rotor is carried out transient equilibrium with the result who obtains this moment; Go weight to require under the very little situation at balanced surface, the least square method of restriction counterweight can drop to balanced surface counterweight amount in the allowed band effectively.

The present invention is directed to the counterbalance weight of existing transducer air conditioning single-rotor compressor through having rigid rotator now can't the balance compressor operation in out-of-balance force; Thereby low frequency " roars of laughter roars of laughter " sound, the compressor surface vibration that cause the uneven vibration that produces of compressor rotor system to cause are big; Make the comfortableness of transducer air conditioning weak point such as can not get demonstrating fully; The present invention proposes a kind of when the working rotor rotating speed is higher than first critical speed; Rigid rotator becomes flex rotor, adopts the least square influence coefficient dynamic balance method based on the restriction counterweight that the high speed flex rotor is carried out balancing rotor.

The present invention is based on more actual vibration test data; Utilize correcting value in the linear system and measured between linear relationship; The size of needed counterweight amount when adopting least square method to calculate the compressor flexible rotor dynamic balance to record number and count greater than balanced surface; And seek out the optimum balance position, thereby subdued the unbalance response that high speed rotor causes owing to the rotor bow distortion according to constraint condition.

The present invention compares with existing frequency conversion single-rotor compressor rotor balancing technique, has following characteristics:

One, effectively the high-speed cruising compressor rotor is carried out accurate transient equilibrium.Compare with present rigid rotator transient equilibrium; This balance method not only will be considered the balance of power and moment; Also consider flex rotor because the unbalance response that rotor bow distortion causes, can effectively confirm size, non-equilibrium site etc. with flex rotor amount of unbalance under the rotation speed change.

Two, subdue " roars of laughter roars of laughter " different sound of low frequency that the high-speed cruising rotor unbalance causes.For frequency conversion single-rotor compressor off-premises station noise principal character is the fundamental frequency and the harmonics of running frequency; Running frequency is relatively low, and as when working speed is 5400rpm, its running frequency is 90Hz; The A weighted acoustic pressure of the noise of 90Hz is not high, but produces irksome " roars of laughter roars of laughter " sound.And the employing patented method, the transient equilibrium of the rotor that runs up can effectively be controlled, and the different sound of the low frequency that makes unbalance dynamic cause is able to subdue even eliminate.

Three, reduce the surface vibration of frequency conversion single-rotor compressor effectively.After adopting this patent method, it can control the torsional oscillation of surface normal, axial vibration and the sense of rotation of the housing that rotor unbalance causes; And also reduce the vibration of the air conditioner piping that is connected with compressor effectively, thereby the reliability of lifting transducer air conditioning pipe arrangement.

The present invention has flexible operation, characteristics that low-frequency noise is low.

Description of drawings

Fig. 1 is a transducer air conditioning single-rotor compressor motor balance synoptic diagram.

Embodiment

Below in conjunction with accompanying drawing and embodiment the present invention is further described.

The control method that this improves the single-rotor compressor tonequality of transducer air conditioning is the rotor-bearing system to a linearity, and under rotational speed omega, rotor is at the vibration A at i point place _i, with the amount of unbalance U at j point place _jBetween exist following relation:

A _i=α _Ij(ω) U _jFormula three,

In the formula, α _Ij(ω) be shadow coefficient under the rotational speed omega, the relation between the amount of unbalance at j point place vibrates with i point place on the expression rotor; Wherein, i ≠ j,

For asking influence coefficient α _Ij(ω), at first under certain rotating speed, measure the vibration A at the i point place that the rotor original imbalance causes _I0(amplitude and phase place) adds a known amount of unbalance U then on plane j point _j, measure the vibration A at i point place again _IjSo influence coefficient is:

${\mathrm{\&alpha;}}_{\mathrm{Ij}}\left(\mathrm{\&omega;}\right)=\frac{A_{\mathrm{Ij}}-A_{i0}}{U_j}$ Formula four,

The vibration of supposing rotor-bearing system meets linear conditions, and is promptly proportional and phasing degree that vibrate is a constant between amplitude under certain rotating speed and little aequum; Therefore, put irrelevantly with increasing the weight of test mass size and angular phasing on the face, have nothing to do with epitrochanterian original unbalance condition by the value of the influence coefficient of formula four definition; Also promptly under certain rotating speed, in the range of linearity, influence coefficient α _IjValue (ω) is a constant; Increase the weight of face and be the facet that the j point adds counterbalance weight,

If increase the weight of face j=1,2 ..., M, the vibration measuring point i=1 that chooses, 2 ..., N comprises all measuring points under the different balancing speeds, should make each balanced surface add the counterweight amount under the ideal situation or deducts weight Q ₁, Q ₂..., Q _MAfter, each vibration of vibrating measuring point is zero, can get:

$\underset{j=1}{\overset{M}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_{\mathrm{Ij}}{Q}_j+A_{i0}=0$ I=1,2 ..., N formula five,

Write as matrix form:

$\left[\begin{array}{cccc}{\mathrm{\&alpha;}}_{11}& {\mathrm{\&alpha;}}_{12}& ...& {\mathrm{\&alpha;}}_{1M}\\ {\mathrm{\&alpha;}}_{21}& {\mathrm{\&alpha;}}_{22}& ...& {\mathrm{\&alpha;}}_{2M}\\ .& .& & .\\ .& .& & .\\ .& .& & .\\ {\mathrm{\&alpha;}}_{N1}& {\mathrm{\&alpha;}}_{\mathrm{<figure-callout\; id="2"\; label="N"\; filenames="HDA00001619772500011.png"\; state="\{\{state\}\}">N</figure-callout>}2}& ...& {\mathrm{\&alpha;}}_{\mathrm{Nm}}\end{array}\right]\left\{\begin{array}{c}{Q}_1\\ {\mathrm{<figure-callout\; id="2"\; label="Q"\; filenames="HDA00001619772500011.png"\; state="\{\{state\}\}">Q</figure-callout>}}_2\\ .\\ .\\ .\\ Q_M\end{array}\right\}+\left\{\begin{array}{c}{A}_{10}\\ A_{20}\\ .\\ .\\ .\\ A_{N0}\end{array}\right\}=0$ Formula six,

When M=N, and N=ab, wherein a is the number of balancing speed, and b is a measuring point number under the rotating speed, and N is total measuring point number, and formula five has exact solution, and promptly the correcting value on each plane is:

Q _j=-α _Ij ^-1A _I0Formula seven,

Usually increase the weight of and limited, promptly M N, formula five are the overdetermined equation group, can only try to achieve approximate solution, utilize least square method to find the solution:

Q _j=-((α _Ij ^*) ^Tα _Ij) ^-1(α _Ij ^*) ^TA _I0Formula eight,

Wherein, α _Ij ^*Be α _IjGrip matrix altogether;

The residual oscillation of each measuring point is:

δ _j=α _IjQ _j+ A _I0Formula nine,

The counterweight amount is unit usually with g, because in actual engineering, the counterweight amount that the high speed flex rotor allows to install usually is restricted, therefore, transient equilibrium should reach satisfied balance result with as far as possible little balance mass;

Because least square method puts undue emphasis on the minimization residual oscillation to the scope of counterweight amount addition of constraints not; Therefore; Use the least square influence coefficient method, just obtain the method for influence coefficient, in application, can cause balance mass bigger with least square method; Can't realize that in some occasion this need impose restriction to the counterweight amount;

To consider that also residual oscillation can not be too big when the counterweight amount meets the demands certainly, could obtain satisfied result;

According to the minimized thought of weighted iteration least square method root mean square, make the quadratic sum of counterweight amount minimum, thereby set up the objective function constraints:

$\mathrm{Min}p=\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_j{\left|Q_j\right|}^2$ Formula 11,

$s.t.\left\{\begin{array}{c}\max \left|Q_j\right|<Q_{\lim }\\ \max \left|{\mathrm{\&delta;}}_j\right|<{\mathrm{\&delta;}}_{\lim }\end{array}\right.$

Wherein, Q _LimBe the permissible value of maximum counterweight amount, δ _LimBe the permissible value of maximum residual oscillation, C _jIt is the weight coefficient of the balance of weights face of j balanced surface;

Least square method is a kind of statistic algorithm, and weighted iteration least square method root mean square is least square method a kind of algorithm wherein,

This method is applicable to that the counterweight amount that can install additional on the balanced surface is very limited, but the residual oscillation value is required not high situation, promptly in the constraint condition

First formula: max|Q _j|<q _LimBe main determining factor, should control its upper limit according to the actual conditions strictness, and

Second formula: max| δ _j|＜δ _Lim, upper limit broad, it is too little that maximum residual oscillation value needn't limit;

Set the permissible value Q of counterweight amount _LimIf the counterweight amount of j balanced surface calculating gained is greater than Q _Lim, then the weight coefficient on this plane is:

${C_j}^{\left(s\right)}=\frac{\left|Q_j\right|}{Q_j\&times;{10}^6}+{C_j}^{(s-1)}$ Formula 12,

In the formula, s represents iteration the s time, and C _j ⁽⁰⁾=0; At this moment, balanced surface counterweight amount is:

{ Q _j([the α of }=- _Ij ^*] ^T[α _Ij]+C _j) ^-1[α _Ij ^*] ^T{ A _I0Formula 13,

The computation process of the least square method of restriction counterweight is following: at first set counterweight amount permissible value Q _Lim, calculate the counterweight amount according to primary data with least square method, whether check satisfies constraint condition, then need add a weight coefficient and recomputates the counterweight amount if do not satisfy, revise repeatedly weight coefficient up to the counterweight amount less than Q _Lim, and then calculate residual oscillation, if meeting the demands, residual oscillation can finish to calculate; Otherwise, carry out calculating the counterweight amount again behind the weighted iteration, if the counterweight amount is still calculated less than promptly finishing; Otherwise, carry out cycle calculations again according to above-mentioned steps;

Because the counterweight amount reduces, the residual oscillation value can rise to some extent, if the maximum residual permissible value δ that shakes _LimBe provided with too for a short time, will can not jump circulation; Therefore, need only the counterweight amount in specialized range, and residual oscillation is little, can stop iteration, and as net result rotor is carried out transient equilibrium with the result who obtains this moment; Go weight to require under the very little situation at balanced surface, the least square method of restriction counterweight can drop to balanced surface counterweight amount in the allowed band effectively.

With certain transducer air conditioning experiment is example; The semianechoic room of placing air conditioner the band operating mode carries out the noise contrast test; The working temperature of outside is 7 ℃, and the working temperature of indoor is 20 ℃, air conditioner is set to high wind heats, running frequency is 75Hz; Be that working speed is 4500rpm, adopt B&K vibrating noise testing apparatus that it is gathered and analyzes.Table 1 is the noise figure of certain frequency-change air conditioner outdoor machine noise part third-octave before and after this programme is implemented.

Certain frequency-change air conditioner outdoor machine noise part third-octave contrast level of noise of table 1

Can find out from certain frequency-change air conditioner outdoor machine noise frequency multiplication figure of table 1: this programme is implemented forward and backward its A weighted sound pressure level value 51.1dBA, 51.7dBA respectively, and the level of noise size is suitable basically; But; The different sound of 75Hz low frequency is 31.9dBA (being 54.3dB) at the corresponding 80Hz of third-octave figure before implementing; And implement the corresponding 80Hz in back is 28.4dBA (being 50.8dB); Technical scheme provided by the invention differs 3.5dBA, so can be subdued the different sound of frequency conversion single-rotor compressor low frequency " roars of laughter roars of laughter " effectively after enforcement.

Claims (1)

1. a control method of improving the single-rotor compressor tonequality of transducer air conditioning is characterized in that the rotor-bearing system to a linearity, and under rotational speed omega, rotor is at the vibration A at i point place _i, with the amount of unbalance U at j point place _jBetween exist following relation:

A _i=α _Ij(ω) U _jFormula three,

In the formula, α _Ij(ω) be shadow coefficient under the rotational speed omega, the relation between the amount of unbalance at j point place vibrates with i point place on the expression rotor; Wherein, i ≠ j,

For asking influence coefficient α _Ij(ω), at first under certain rotating speed, measure the vibration A at the i point place that the rotor original imbalance causes _I0(amplitude and phase place) adds a known amount of unbalance U in the plane the j point then _j, measure the vibration A at i point place again _IjSo influence coefficient is:

${\mathrm{\&alpha;}}_{\mathrm{Ij}}\left(\mathrm{\&omega;}\right)=\frac{A_{\mathrm{Ij}}-A_{i0}}{U_j}$ Formula four,

The vibration of supposing rotor-bearing system meets linear conditions, and is promptly proportional and phasing degree that vibrate is a constant between amplitude under certain rotating speed and little aequum; Therefore, put irrelevantly with increasing the weight of test mass size and angular phasing on the face, have nothing to do with epitrochanterian original unbalance condition by the value of the influence coefficient of formula four definition; Also promptly under certain rotating speed, in the range of linearity, influence coefficient α _IjValue (ω) is a constant; Increase the weight of face and be the facet that the j point adds counterbalance weight;

If increase the weight of face j=1,2 ..., M, the vibration measuring point i=1 that chooses, 2 ..., N comprises all measuring points under the different balancing speeds, should make each balanced surface add the counterweight amount under the ideal situation or deducts weight Q ₁, Q ₂..., Q _MAfter, each vibration of vibrating measuring point is zero, can get:

$\underset{j=1}{\overset{M}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_{\mathrm{Ij}}{Q}_j+A_{i0}=0$ I=1,2 ..., N formula five,

Write as matrix form:

$\left[\begin{array}{cccc}{\mathrm{\&alpha;}}_{11}& {\mathrm{\&alpha;}}_{12}& ...& {\mathrm{\&alpha;}}_{1M}\\ {\mathrm{\&alpha;}}_{21}& {\mathrm{\&alpha;}}_{22}& ...& {\mathrm{\&alpha;}}_{2M}\\ .& .& & .\\ .& .& & .\\ .& .& & .\\ {\mathrm{\&alpha;}}_{N1}& {\mathrm{\&alpha;}}_{N2}& ...& {\mathrm{\&alpha;}}_{\mathrm{Nm}}\end{array}\right]\left\{\begin{array}{c}{Q}_1\\ Q_2\\ .\\ .\\ .\\ Q_M\end{array}\right\}+\left\{\begin{array}{c}{A}_{10}\\ A_{20}\\ .\\ .\\ .\\ A_{N0}\end{array}\right\}=0$ Formula six,

When M=N, and N=ab, wherein a is the number of balancing speed, and b is a measuring point number under the rotating speed, and N is total measuring point number, and formula five has exact solution, and promptly the correcting value on each plane is:

Q _j=-α _Ij ^-1A _I0Formula seven,

Usually increase the weight of and limited, promptly M N, formula five are the overdetermined equation group, can only try to achieve approximate solution, utilize least square method to find the solution:

Q _j=-((α _Ij ^*) ^Tα _Ij) ^-1(α _Ij ^*) ^TA _I0Formula eight,

α wherein _Ij ^*Be α _IjGrip matrix altogether;

The residual oscillation of each measuring point is:

δ _j=α _IjQ _j+ A _I0Formula nine,

The counterweight amount is unit usually with g, because in actual engineering, the counterweight amount that the high speed flex rotor allows to install usually is restricted, therefore, transient equilibrium should reach satisfied balance result with as far as possible little balance mass;

Because least square method puts undue emphasis on the minimization residual oscillation to the scope of counterweight amount addition of constraints not; Therefore; Use the least square influence coefficient method, just obtain the method for influence coefficient, in application, can cause balance mass bigger with least square method; Can't realize that in some occasion this need impose restriction to the counterweight amount;

To consider that also residual oscillation can not be too big when the counterweight amount meets the demands certainly, could obtain satisfied result;

According to the minimized thought of weighted iteration least square method root mean square, make the quadratic sum of counterweight amount minimum, thereby set up the objective function constraints:

$\mathrm{Min}p=\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_j{\left|Q_j\right|}^2$ Formula 11,

$s.t.\left\{\begin{array}{c}\max \left|Q_j\right|<Q_{\lim }\\ \max \left|{\mathrm{\&delta;}}_j\right|<{\mathrm{\&delta;}}_{\lim }\end{array}\right.$

Wherein, Q _LimBe the permissible value of maximum counterweight amount, δ _LimBe the permissible value of maximum residual oscillation, C _jIt is the weight coefficient of the balance of weights face of j balanced surface;

This method is applicable to that the counterweight amount that can install additional on the balanced surface is very limited, but the residual oscillation value is required not high situation, i.e. first formula in the constraint condition: max|Q _j|<q _LimBe main determining factor, should control its upper limit according to actual conditions are strict, and second formula: max| δ _j|<δ _LimUpper limit broad, it is too little that maximum residual oscillation value needn't limit;

Set the permissible value Q of counterweight amount _LimIf the counterweight amount of j balanced surface calculating gained is greater than Q _Lim, then the weight coefficient on this plane is:

${C_j}^{\left(s\right)}=\frac{\left|Q_j\right|}{Q_j\&times;{10}^6}+{C_j}^{(s-1)}$ Formula 12,

In the formula, s represents iteration the s time, and C _j ⁽⁰⁾=0; At this moment, balanced surface counterweight amount is:

{ Q _j([the α of }=- _Ij ^*] ^T[α _Ij]+C _j) ^-1[α _Ij ^*] ^T{ A _I0Formula 13,

The computation process of the least square method of restriction counterweight is following: at first set counterweight amount permissible value Q _Lim, calculate the counterweight amount according to primary data with least square method, whether check satisfies constraint condition, then need add a weight coefficient and recomputates the counterweight amount if do not satisfy, revise repeatedly weight coefficient up to the counterweight amount less than Q _Lim, and then calculate residual oscillation, if meeting the demands, residual oscillation can finish to calculate; Otherwise, carry out calculating the counterweight amount again behind the weighted iteration, if the counterweight amount is still calculated less than promptly finishing; Otherwise, carry out cycle calculations again according to above-mentioned steps;

Because the counterweight amount reduces, the residual oscillation value can rise to some extent, if the maximum residual permissible value δ that shakes _LimBe provided with too for a short time, will can not jump circulation; Therefore, need only the counterweight amount in specialized range, and residual oscillation is little, can stop iteration, and as net result rotor is carried out transient equilibrium with the result who obtains this moment; Go weight to require under the very little situation at balanced surface, the least square method of restriction counterweight can drop to balanced surface counterweight amount in the allowed band effectively.

Images