Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings and examples.
Fig. 1 and 2 respectively show partial flowcharts of an embodiment of a method for suppressing fluctuation of a rotational speed of an air conditioner compressor according to the present invention. Specifically, the method of suppressing the rotation speed fluctuation of the embodiment includes two processes: one is a process for controlling the compressor according to the real-time angular velocity, and the flow chart is shown in fig. 1; one is a process of controlling the compressor according to the torque, and the flowchart is shown in fig. 2. Specific implementations of these two processes are described below based on a control block diagram shown in fig. 1 and 2 in conjunction with fig. 3.
Referring to fig. 1, a partial flowchart of an embodiment of a method for suppressing the rotation speed fluctuation of an air conditioner compressor based on the present invention, specifically a flowchart for controlling the compressor according to the real-time angular velocity, is shown, and the embodiment employs a process including the following steps to control the compressor according to the real-time angular velocity:
step 11: a shaft error Delta theta reflecting a deviation between an actual position and an estimated position of a compressor rotor is acquired.
In the control of the compressor, the phase of the compressor rotor can be locked to the target phase by a phase-locked loop (PLL) control technique, the control block of which is shown in fig. 3. In the prior art, a phase-locked loop regulator, typically a proportional-integral regulator, is included in the phase-locked loop of the compressor, see K of fig. 3P_PLLAnd KI_PLLand/S. The axis error Δ θ is used as an input of the PLL regulator, and specifically, the axis error Δ θ is subtracted from a target angular fluctuation amount (0 shown in fig. 3), and the difference is input to the PLL regulator, and the output of the PLL regulator is an output angular velocity Δ ω _ PLL. Based on the output angular velocity Δ ω _ PLL of the phase-locked loop regulator, the phase-locked loop outputs a real-time angular velocity ω 1 for compressor control, and the rotor position is controlled using the real-time angular velocity ω 1.
The shaft error Δ θ, which reflects the deviation between the actual position and the estimated position of the compressor rotor, can be calculated by the following equation:
in the formula, the first step is that,
and
respectively a d-axis voltage set value and a q-axis voltage set value of the compressor, I
dAnd I
qReal-time d-axis current and real-time q-axis current, r, of the compressor, respectively
*Is the resistance of the motor of the compressor,
is the q-axis inductance, omega, of the compressor
1Is the real-time angular frequency of the compressor. Among the parameters, I
d、I
qAnd ω
1The detection is carried out in real time by the detection means in the prior art, and other parameter values are known values.
Step 12: and filtering the axis error delta theta to obtain a corrected axis error delta theta' after at least part of axis error fluctuation is filtered.
Since the shaft error is used as an input to the phase locked loop, the real-time angular velocity of the compressor at the output of the phase locked loop is affected. If the shaft error fluctuation is large, the real-time angular speed output by the phase-locked loop is unstable, so that the rotor phase locking is unstable, and further, the compressor has faults of overcurrent, step loss and the like.
After the axis error Δ θ is obtained in step 11, filtering is performed on the axis error Δ θ to filter at least a part of fluctuation components, so as to obtain a corrected axis error Δ θ' after filtering at least a part of axis error fluctuation. The method for filtering the shaft error can be implemented by adopting the prior art, and more preferably, the filtering process is described in the following preferred embodiments.
Step 13: the angular velocity compensation amount P _ out is obtained from the corrected shaft error Δ θ'.
This step can be implemented in a manner of obtaining the angular velocity according to the angle in the prior art. The more preferable processing manner is described in the following preferable embodiment.
The implementation of the above step 12 and step 13 is reflected in the control block diagram of fig. 3, and an axis error fluctuation filtering algorithm is adopted to obtain the angular velocity compensation amount P _ out.
Step 14: the angular velocity compensation amount P _ out is compensated to the output angular velocity delta omega _ PLL of the phase-locked loop regulator in the phase-locked loop for controlling the compressor, and the compensated angular velocity output amount delta omega' is obtained. Specifically, the compensated angular velocity output amount Δ ω' ═ P _ out + Δ ω _ PLL.
Step 15: and correcting the real-time angular speed omega 1 for controlling the compressor according to the compensated angular speed output quantity, and controlling the compressor according to the corrected real-time angular speed omega 1.
Specifically, the method of determining the real-time angular velocity corresponding to the target angular velocity fluctuation amount of 0 in the following velocity loop control is: referring to fig. 3, the compensated angular velocity output amount Δ ω' is added to the angular velocity command ω × in, and the real-time angular velocity ω 1 for controlling the compressor is output. The angular velocity command ω _ in is a given angular velocity value of the compressor control system, and the determination method of the value of the given angular velocity command ω _ in is implemented by using the prior art. The target angular velocity fluctuation quantity of the speed loop is 0, and the real-time angular velocity is determined based on the output angular velocity delta omega _ PLL of the phase-locked loop regulator and the given angular velocity command omega _ in, so that the compressor is controlled more accurately and stably.
Referring to fig. 2, a partial flowchart of an embodiment of a method for suppressing the speed fluctuation of an air conditioner compressor according to the present invention, specifically a flowchart for controlling the compressor according to the torque, is shown, and the embodiment employs a process including the following steps:
step 21: and calculating the difference between the target angular velocity fluctuation amount and the feedback angular velocity amount to obtain a first angular velocity difference value. The feedback angular velocity amount is the sum of the direct current component P _ DC of the angular velocity compensation amount P _ out and the compensated angular velocity output amount Δ ω'.
In compressor control, the rotational speed of the compressor rotor can be controlled to approach a set rotational speed by a speed loop (ASR) control technique. Referring to the block diagram of FIG. 3, the speed loop includes a speed loop toneNode-makers, typically proportional-integral-regulators, see K in FIG. 3P_ASRAnd KI_ASR/S。
In this step, both the direct current component P _ DC of the angular velocity compensation amount P _ out and the compensated angular velocity output amount Δ ω' are used as inputs to the velocity loop. Specifically, the direct current component P _ DC of the angular velocity compensation amount P _ out is extracted, and the sum of the direct current component P _ DC and the compensated angular velocity output amount Δ ω 'is calculated to obtain the feedback angular velocity amount Δ ω 1, where Δ ω 1 is P _ DC + Δ ω'. The extraction of the DC component P _ DC of the angular velocity compensation amount P _ out can be realized by using a conventional technique for extracting a DC component, for example, by using a low-pass filter to extract the DC component P _ DC of the angular velocity compensation amount P _ out.
Then, a difference between the target angular velocity fluctuation amount and the angular velocity output amount Δ ω' is calculated, and the difference therebetween is determined as a first angular velocity difference Δ ω 2. Here, the target angular velocity fluctuation amount is a desired angular velocity fluctuation amount and is a known input amount. As a preferred embodiment, in this example, the target angular velocity fluctuation amount is 0.
Step 22: filtering the first angular velocity difference to obtain a filtered angular velocity at least part of which is filtered out of angular velocity fluctuation, and inputting the filtered angular velocity as an input quantity to a speed ring regulator in a speed ring for controlling the compressor to obtain an output torque of the speed ring regulator; and meanwhile, executing moment compensation based on the first angular velocity difference to obtain moment compensation amount corresponding to part of angular velocity fluctuation in the first angular velocity difference.
The first angular velocity difference Δ ω 2 is used as an input to the speed loop regulator, and affects the output torque at the speed loop output. If the first angular speed difference value fluctuates greatly, the fluctuation of the output torque is large, and further the fluctuation of the rotating speed of the compressor is large. After the first angular velocity difference is obtained in step 21, a filtering process is performed on the first angular velocity difference to filter out at least part of the angular velocity fluctuation component, so as to obtain a filtered angular velocity Δ ω _ K. The method for filtering the angular velocity can be implemented by adopting a filtering mode in the prior art, and more preferably, the filtering process is described in the following preferred embodiment. Then, the filtered angular velocity Δ ω _ K is input to the speed loop regulator as an input amount, and the output torque τ _ ASR of the speed loop regulator is obtained.
Meanwhile, a moment compensation algorithm is adopted, and moment compensation is executed based on the first angular velocity difference delta omega 2, so that moment compensation amount tau _ out corresponding to part of angular velocity fluctuation in the first angular velocity difference delta omega 2 is obtained. For the torque compensation algorithm, all the possible solutions existing in the prior art may be adopted as long as it is ensured that the obtained torque compensation amount τ _ out corresponds to a part of the angular velocity fluctuation in the first angular velocity difference Δ ω 2. The preferred moment compensation algorithm is described in the following preferred embodiments.
Step 23: and compensating the torque compensation amount to the output torque of the speed ring regulator to obtain the compensated output torque.
Specifically, the torque compensation amount τ _ out is added to the output torque τ _ ASR of the speed loop regulator to obtain the compensated output torque τM:τM=τ_out+τ_ASR。
Step 24: and controlling the air-conditioning compressor according to the compensated output torque. The specific control process refers to the prior art.
By adopting the method of the embodiment formed by the above-mentioned fig. 1 and fig. 2, the double loop control of the speed loop and the phase-locked loop of the compressor is realized. In addition, in the phase-locked loop control, the fluctuation filtering is carried out on the shaft error delta theta reflecting the deviation between the actual position and the estimated position of the compressor rotor, the angular speed compensation quantity corresponding to the corrected shaft error after at least part of the shaft error fluctuation is filtered is compensated to the output angular speed of the phase-locked loop regulator, the compensated angular speed output quantity is obtained, the real-time angular speed of the compressor is corrected according to the compensated angular speed output quantity, and when the compressor is controlled by the corrected real-time angular speed, the variation quantity and the phase of the target rotating speed can be close to the variation quantity and the phase of the actual rotating speed, so that the operation of the compressor tends to be stable. Moreover, because the fluctuation of the shaft error is a front end direct factor causing speed fluctuation, the periodical fluctuation of the shaft error is reduced by filtering the fluctuation of the shaft error at the front end, the speed fluctuation can be directly and quickly inhibited, and the effectiveness of speed control is improved. In the control of the speed loop, the difference value between the feedback angular velocity quantity and the target angular velocity fluctuation quantity is subjected to filtering processing, and the filtered angular velocity with at least part of angular velocity fluctuation filtered out is input into the speed loop regulator as an input quantity, so that the fluctuation of the output torque of the speed loop regulator can be reduced; obtaining a moment compensation amount based on a difference value between the output angular speed of the phase-locked loop regulator and the target angular speed fluctuation amount, compensating the moment compensation amount into the output moment of the speed loop regulator to obtain a compensated output moment, wherein the compensated output moment can reduce the difference moment between the motor moment and the load moment; then, when the compressor is controlled according to the compensated output torque, the fluctuation of the rotational speed of the compressor can be significantly reduced based on the output torque of the speed loop regulator with reduced fluctuation and the compensated output torque with reduced difference torque, so that the operation of the compressor tends to be smooth. The compressor operates stably, the technical effects of energy conservation and vibration reduction can be achieved, and the operation performance of the compressor is further improved.
In some other embodiments, the filtering the axis error Δ θ to obtain a corrected axis error Δ θ' after at least part of the axis error fluctuation is filtered, specifically includes: and filtering the axis error delta theta to filter at least the first harmonic component in the delta theta and obtain a corrected axis error delta theta' in which at least the first harmonic component is filtered. In a more preferred embodiment, the filtering process is performed on the axis error Δ θ, and includes filtering the first harmonic component and the second harmonic component in Δ θ to obtain a corrected axis error Δ θ' with the first harmonic component and the second harmonic component filtered. Most of fluctuation components in the delta theta can be filtered out by filtering out the first harmonic component or the first harmonic component and the second harmonic component in the delta theta, the calculated amount is moderate, and the filtering speed is high.
Fig. 4 is a logic block diagram showing a specific example of the axis error fluctuation filtering algorithm in fig. 3, specifically, a logic block diagram showing a specific example of obtaining the angular velocity compensation amount P _ out corresponding to the corrected axis error Δ θ' after filtering the first harmonic component and the second harmonic component in the axis error Δ θ. The specific acquisition process is as follows:
firstly, the axis error Delta theta is subjected to Fourier series expansion to obtain the axis errorDelta theta with respect to mechanical angle thetamIs used for the functional expression of (1). The method comprises the following specific steps:
in the formula,. DELTA.theta.
DCIs the direct component of the axis error, θ
d_n=θ
peak_ncosφ
n,θ
q_n=θ
peak_nsinφ
n,
Δθ
peak_nFor the n harmonic axis error fluctuation amplitude, theta
m1、θ
m2Is the first harmonic mechanical angle. And second harmonic mechanical angle theta
m2Expressed as: theta
m2=2θ
m1。
And then, extracting a first harmonic component and a second harmonic component from the function expression, and filtering the first harmonic component and the second harmonic component by adopting an integrator to obtain a filtering result.
Specifically, the first harmonic component and the second harmonic component can be extracted from the above functional expression by using a low-pass filtering method or an integration method. With particular reference to FIG. 4, the functional expressions are each related to cos θm1And cos θm2After multiplication, a low-pass filter is used for filtering or an integrator is used for taking an integral average value in a period, and a d-axis component of a first harmonic and a d-axis component of a second harmonic of an axis error delta theta are extracted; respectively comparing the function expressions with-sin thetam1And-sin θm2After multiplication, the q-axis component of the first harmonic and the q-axis component of the second harmonic of the axis error delta theta are extracted by filtering through a low-pass filter or taking an integral average value in a period through an integrator. Then, the d-axis component and the q-axis component of the first harmonic and the d-axis component and the q-axis component of the second harmonic are respectively subtracted from 0, and the resultant is input to an integrator KI_PAnd performing integral filtering treatment in the step S to obtain a filtering result for filtering the first harmonic component and the second harmonic component, wherein the filtering result is changed into the angular velocity.
Then, each filtering result is subjected to inverse fourier transform, and an angular velocity compensation amount P _ out corresponding to the correction axis error Δ θ' in which the first harmonic component and the second harmonic component are filtered is obtained. Specifically, the filtering result of the d-axis component for filtering the first harmonic and the filtering result of the q-axis component for filtering the first harmonic are respectively subjected to the sum of results after inverse fourier transform, so as to form an angular velocity compensation quantity P _ out1 corresponding to the correction axis error for filtering the first harmonic component; the filtering result of the d-axis component for filtering the second harmonic and the filtering result of the q-axis component for filtering the second harmonic are respectively subjected to the sum of results after Fourier inverse transformation, and an angular velocity compensation quantity P _ out2 corresponding to the correction axis error for filtering the second harmonic component is formed; the sum of the two angular velocity compensation amounts forms an angular velocity compensation amount P _ out of P _ out1+ P _ out2 corresponding to the correction axis error Δ θ' in which the first harmonic component and the second harmonic component are filtered out.
As a preferred embodiment, the control of harmonic filtering can also be achieved by adding an enable switch. Specifically, in the block diagram of fig. 4, Gain _1 and Gain _2 are enable switches for determining whether to turn on/off the filtering algorithm function. In the case where the enable switch states of Gain _1 and Gain _2 are the functions of filtering the first harmonic and filtering the second harmonic, the angular velocity compensation amount P _ out corresponding to the correction axis error Δ θ' of filtering the first harmonic component and the second harmonic component is obtained as P _ out1+ P _ out 2. If the enable switch states of Gain _1 and Gain _2 are the functions of filtering the first harmonic and the second harmonic, the whole axis error filtering function is turned off, and the angular velocity compensation amount P _ out cannot be output. If one of the enable switches is in the state of turning on the filtering algorithm function, and the other enable switch is in the state of turning off the filtering algorithm function, the obtained angular velocity compensation quantity P _ out is only the angular velocity compensation quantity for filtering the first harmonic (the Gain _1 enable switch is in the state of turning on the filtering first harmonic function, and the Gain _2 enable switch is in the state of turning off the filtering second harmonic function), or is only the angular velocity compensation quantity for filtering the second harmonic (the Gain _1 enable switch is in the state of turning off the filtering first harmonic function, and the Gain _2 enable switch is in the state of turning on the filtering second harmonic function).
In the embodiment of filtering only the first harmonic component, the process of extracting the first harmonic component and filtering the first harmonic component in fig. 4 may be directly adopted. Of course, in the embodiment of filtering only the first harmonic component, the control of filtering the first harmonic component may also be implemented by adding an enable switch, and the specific implementation manner is also referred to fig. 4 and will not be repeated herein.
As a preferred embodiment, the filtering processing is performed on the first angular velocity difference Δ ω 2 to obtain a filtered angular velocity Δ ω _ K after at least part of the angular velocity fluctuation is filtered, and the method specifically includes: and extracting partial angular velocity fluctuation K _ out in the first angular velocity difference value delta omega 2 by adopting a velocity fluctuation extraction algorithm, and calculating the difference value between the first angular velocity difference value delta omega 2 and the partial angular velocity fluctuation K _ out, wherein the difference value is determined as the filtering angular velocity delta omega _ K.
In some other preferred embodiments, a speed fluctuation extraction algorithm is used to extract a partial angular velocity fluctuation in the first angular velocity difference, and a difference between the first angular velocity difference and the partial angular velocity fluctuation is calculated, where the difference is determined as a filtered angular velocity, and the method specifically includes: and adopting a speed fluctuation extraction algorithm to extract at least a first harmonic component in the first angular speed difference value as part of angular speed fluctuation, calculating the difference value between the first angular speed difference value and the first harmonic component, and determining the difference value as a filtering angular speed for filtering at least the first harmonic component. As a more preferable embodiment, the extracting a partial angular velocity fluctuation from the first angular velocity difference by using a velocity fluctuation extraction algorithm, and calculating a difference between the first angular velocity difference and the partial angular velocity fluctuation, where the difference is determined as a filtered angular velocity specifically includes: and extracting a first harmonic component and a second harmonic component in the first angular velocity difference value by adopting a velocity fluctuation extraction algorithm, taking the sum of the first harmonic component and the second harmonic component as part of angular velocity fluctuation, calculating the difference value between the first angular velocity difference value and the sum of the first harmonic component and the second harmonic component, and determining the difference value as the filtering angular velocity after the first harmonic component and the second harmonic component are filtered. Most of fluctuation components in the first angular velocity difference value can be filtered out by filtering out the first harmonic component in the first angular velocity difference value or filtering out the first harmonic component and the second harmonic component in the first angular velocity difference value, the calculated amount is moderate, and the filtering speed is high.
Fig. 5 is a logic block diagram showing a specific example of the speed fluctuation extraction algorithm in fig. 3, specifically, a logic block diagram showing a specific example of extracting the first harmonic component and the second harmonic component from the first angular velocity difference value to form a partial angular velocity fluctuation. Referring to fig. 5, this specific example obtains a partial angular velocity fluctuation containing a first harmonic component and a second harmonic component by the following method:
firstly, a Fourier series expansion is carried out on the first angular velocity difference delta omega 2 to obtain the first angular velocity difference delta omega 2 relative to the mechanical angle thetamIs used for the functional expression of (1). This process can be implemented using existing technology and is not described in detail here.
Then, the first harmonic component and the second harmonic component are extracted from the functional expression, respectively.
Specifically, as shown in FIG. 5, the functional expression is related to cos θ
m1After multiplication, pass through a low-pass filter
Filtering to obtain d-axis component omega of first harmonic before inverse transformation
d1Then, carrying out inverse Fourier transform to obtain the d-axis component of the inverse transformed first harmonic; multiplying the functional expression by-sin θ
m1After multiplication, pass through a low-pass filter
Filtering to obtain q-axis component omega of the first harmonic before inverse transformation
q1Then carrying out inverse Fourier transform to obtain the q-axis component of the inverse transformed first harmonic; then, the d-axis component and the q-axis component of the inverse-transformed first harmonic are added to obtain a first harmonic component K _ out1 in the first angular velocity difference value. Similarly, the functional expression is related to cos θ
m2After multiplication, pass through a low-pass filter
Filtering to obtain d-axis component omega of second harmonic before inverse transformation
d2Then, carrying out inverse Fourier transform to obtain the d-axis component of the second harmonic after inverse transform; will be provided withFunction expression and-sin theta
m2After multiplication, pass through a low-pass filter
Filtering to obtain q-axis component omega of second harmonic before inverse transformation
q2Then carrying out inverse Fourier transform to obtain q-axis components of the second harmonic after inverse transform; then, the d-axis component and the q-axis component of the inverse-transformed second harmonic are added to obtain a second harmonic component K _ out2 in the first angular velocity difference value. Finally, the first harmonic component K _ out1 is added to the second harmonic component K _ out2, and the resulting sum forms part of the angular velocity fluctuation K _ out. Wherein, theta
m1Mechanical angle of first harmonic, theta, in a functional expression developed as a Fourier series
m2Mechanical angle of the second harmonic in a functional expression developed as a Fourier series, and θ
m2=2θ
m1,T
_PD_filterIs the time constant of the low pass filter.
After obtaining the partial angular velocity fluctuation K _ out containing the first harmonic component and the second harmonic component, calculating a difference between the first angular velocity difference Δ ω 2 and the partial angular velocity fluctuation K _ out as a filtered angular velocity Δ ω _ K, where the filtered angular velocity Δ ω _ K is the filtered angular velocity after the first harmonic component and the second harmonic component are filtered out.
As a preferred embodiment, the control of the harmonic extraction can also be achieved by adding an enable switch. Specifically, in the block diagram of fig. 5, Gain _1 and Gain _2 are enable switches for determining whether to turn on/off the extraction algorithm function. Under the condition that the enabling switch states of the Gain _1 and the Gain _2 are on, the functions of extracting the first harmonic and extracting the second harmonic are obtained, and partial angular velocity fluctuation formed by the first harmonic component and the second harmonic component is obtained: k _ out is K _ out1+ K _ out 2. If the enable switch states of Gain _1 and Gain _2 are the functions of extracting the first harmonic and extracting the second harmonic, the whole speed fluctuation extraction algorithm function is turned off, and part of the angular speed fluctuation is 0. If one of the enable switches is in the state of opening the extraction algorithm function, and the other enable switch is in the state of closing the extraction algorithm function, the obtained part of the angular speed fluctuation is only a first harmonic component in the first angular speed difference (the state of the Gain _1 enable switch is in the state of opening the extraction first harmonic function, and the state of the Gain _2 enable switch is in the state of closing the extraction second harmonic function) or only a second harmonic component in the first angular speed difference (the state of the Gain _1 enable switch is in the state of closing the extraction first harmonic function, and the state of the Gain _2 enable switch is in the state of opening the extraction second harmonic function).
In the embodiment of extracting only the first harmonic component, the process of extracting the first harmonic component in fig. 5 may be directly employed; of course, the control of the first harmonic extraction may also be implemented by adding an enable switch, and the specific implementation manner is also shown in fig. 5, which is not repeated herein.
Fig. 6 is a logic diagram showing a specific example of the torque compensation algorithm in fig. 3, specifically, a logic diagram showing a specific example of obtaining torque compensation amounts corresponding to the first harmonic component and the second harmonic component in the first angular velocity difference. Referring to fig. 6, this embodiment obtains the torque compensation amounts corresponding to the first harmonic component and the second harmonic component in the first angular velocity difference by using the following method:
firstly, a Fourier series expansion is carried out on the first angular velocity difference delta omega 2 to obtain the first angular velocity difference delta omega 2 relative to the mechanical angle thetamIs used for the functional expression of (1). This process can be implemented using existing technology and is not described in detail here.
Then, the d-axis correlation quantity and the q-axis correlation quantity of the first harmonic and the d-axis correlation quantity and the q-axis correlation quantity of the second harmonic are obtained from the functional expression. Specifically, the function expressions are respectively related to cos θm1And-sin θm1Multiplying to obtain d-axis correlation quantity and q-axis correlation quantity of the first harmonic in the first angular velocity difference delta omega 2; respectively connecting the function expressions with cos thetam2And-sin θm2And multiplying to obtain the d-axis correlation quantity and the q-axis correlation quantity of the second harmonic in the first angular speed difference value delta omega 2. Thetam1And thetam2The same as above.
Then, the d-axis related quantity and the q-axis related quantity of the first harmonic and the d-axis related quantity and the q-axis related quantity of the second harmonic are converted into d-axis torque and q-axis torque respectively.
Specifically to this example, as a preferred implementation, two steps of conversion to torque are employed: first, use the 1/T integratorIS is converted, TIRespectively converting d-axis correlation quantity and q-axis correlation quantity of the first harmonic and d-axis correlation quantity and q-axis correlation quantity of the second harmonic into d-axis initial torque delta tau 'of the first harmonic as a time constant of an integrator'd1And primary harmonic q-axis initial moment delta tau'q1D-axis initial moment delta tau 'of second harmonic'd2And q-axis initial moment delta tau 'of the second harmonic'q2. And then, respectively carrying out proportion adjustment on the d-axis initial moment and the q-axis initial moment, and determining the result after the proportion adjustment as the required d-axis moment and the q-axis moment. Specifically, the d-axis number f (ω)d1) D-axis initial moment delta tau 'to first harmonic'd1The proportion is adjusted to obtain the d-axis moment delta tau of the first harmonicd1. d number of axes f (ω)d1) From the d-axis component ω of the first harmonicd1And primary harmonic d-axis initial torque delta tau'd1And (4) determining. Wherein the d-axis component ω of the first harmonicd1The d-axis correlation quantity of the first harmonic is determined, and specifically, the d-axis correlation quantity of the first harmonic is obtained after being filtered by a low-pass filter (see the description of fig. 5). According to q-axis number f (ω)q1) Q-axis initial moment delta tau 'to first harmonic'q1The proportion is adjusted to obtain the q-axis moment delta tau of the first harmonicq1. q axial number f (ω)q1) Q-axis component omega from the first harmonicq1And primary harmonic q-axis initial moment delta tau'q1And (4) determining. Wherein the q-axis component ω of the first harmonicq1Is determined according to the q-axis correlation quantity of the first harmonic, and specifically is obtained after the q-axis correlation quantity of the first harmonic is filtered by a low-pass filter (see the description of fig. 5). According to d-axis number f (ω)d2) D-axis initial moment delta tau 'to second harmonic'd2The d-axis torque delta tau of the second harmonic is obtained by proportion adjustmentd2. d number of axes f (ω)d2) From the d-axis component ω of the second harmonicd2And d-axis initial moment delta tau 'of the second harmonic'd2And (4) determining. Wherein the d-axis component ω of the second harmonicd2Is determined from the d-axis correlation of the second harmonic, and in particular willThe d-axis correlation of the second harmonic is obtained after filtering by a low-pass filter (see description of fig. 5). According to q-axis number f (ω)q2) Q-axis initial moment delta tau 'to second harmonic'q2The proportion is adjusted to obtain the q-axis moment delta tau of the second harmonicq2. q axial number f (ω)q2) From the q-component ω of the second harmonicq2And q-axis initial moment delta tau 'of the second harmonic'q2And (4) determining. Wherein the q-axis component ω of the second harmonicq2Is determined according to the q-axis correlation quantity of the second harmonic, and specifically is obtained after the q-axis correlation quantity of the second harmonic is filtered by a low-pass filter (see the description of fig. 5). In other embodiments, the d-axis related quantity and the q-axis related quantity can be directly converted into the corresponding d-axis moment and q-axis moment by only an integrator without proportional adjustment.
And finally, carrying out inverse Fourier transform on the moment to obtain a moment compensation quantity. Specifically, the d-axis moment and the q-axis moment of the first harmonic are respectively related to cos (theta)m1+θshift-K1) And-sin (theta)m1+θshift-K1) Summing the multiplication results after Fourier inverse transformation to form a moment compensation amount tau _ out1 corresponding to the first harmonic fluctuation in the first angular velocity difference delta omega 2; the d-axis moment and the q-axis moment of the second harmonic are respectively related to cos (theta)m2+θshift-K2) And-sin (theta)m2+θshift-K2) The results of the multiplication and the inverse fourier transform are summed to form a torque compensation amount τ _ out2 corresponding to the second harmonic fluctuation in the first angular velocity difference Δ ω 2. The sum of the two torque compensation amounts forms a torque compensation amount τ _ out corresponding to the first harmonic component and the second harmonic component τ _ out1+ τ _ out 2. Wherein, thetashift-K1And thetashift-K2The phase compensation angle of the first harmonic and the phase compensation angle of the second harmonic are respectively, and the angle number of the two phase compensation angles is determined according to the angular velocity phase in the given angular velocity command. The torque compensation quantity is obtained in a phase compensation mode, the torque phase can be deviated based on the compensated output torque obtained by the torque compensation quantity, the torque phase deviates to the load torque of the compressor, the difference torque of the motor torque and the load torque is further reduced, and the purpose of restraining the fluctuation of the rotating speed of the compressor is achievedAnd (5) preparing.
As a preferred embodiment, the control of the moment compensation can also be achieved by adding an enabling switch. Specifically, in the block diagram of fig. 6, Gain _1 and Gain _2 are enable switches for determining whether to turn on/off the torque compensation algorithm function. Under the condition that the enabling switch states of the Gain _1 and the Gain _2 are the first harmonic moment compensation and the second harmonic moment compensation, moment compensation quantities corresponding to the first harmonic component and the second harmonic component are obtained: τ _ out is τ _ out1+ τ _ out 2. If the enabling switch states of the Gain _1 and the Gain _2 are the conditions that the first harmonic moment compensation function and the second harmonic moment compensation function are closed, the whole moment compensation algorithm function is closed, and the moment compensation quantity is 0. If one of the enable switches is in the on-moment compensation algorithm function and the other enable switch is in the off-moment compensation algorithm function, the obtained moment compensation quantity is only the moment compensation quantity corresponding to the first harmonic component in the first angular speed difference (the Gain _1 enable switch is in the on-moment compensation function, and the Gain _2 enable switch is in the off-moment compensation function) or is only the moment compensation quantity corresponding to the second harmonic component in the first angular speed difference (the Gain _1 enable switch is in the off-moment compensation function, and the Gain _2 enable switch is in the on-moment compensation function).
In the embodiment of obtaining the torque compensation amount corresponding to the first harmonic component only, the process of obtaining the torque compensation amount corresponding to the first harmonic component in fig. 6 may be directly adopted; of course, the control of the first harmonic moment compensation can also be realized by adding an enable switch, and the specific implementation manner is also shown in fig. 6, which is not described in additional detail herein.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions.