A common-mode voltage suppression method based on a switching function

Abstract

The present invention discloses a common-mode voltage suppression method based on a switching function. Step 1: Establish a voltage vector set; Step 2: Disable zero vectors in the voltage vector set established in Step 1; Step 3: Tune the switching function to achieve common-mode voltage suppression. The present invention is an optimization for the problem of common-mode voltage spikes caused by the freewheeling of switching tubes in the dead zone during motor vector control. By tuning the switching function, each switching tube can self-turn off when the corresponding freewheeling occurs, thereby blocking the freewheeling path of the common-mode voltage spike and achieving the suppression of the common-mode voltage spike.

Description

Technical Field

[0001] The present invention relates to the field of motor control, and particularly to a common-mode voltage suppression method based on a switching function. Background Art

[0002] Common-mode voltage commonly exists in two-level inverters, and its harms mainly include:

[0003] 1. The high-frequency signal of the common-mode voltage generates high-frequency coupling current, which promotes the aging of the motor insulation.

[0004] 2. The high-frequency change rate of the common-mode voltage will generate shaft current, increasing the mechanical wear between bearings, and the life of the bearings may be reduced accordingly.

[0005] 3. The leakage current generated by the high-frequency common-mode voltage flows into the ground through the winding and the machine shell, and then flows into the power grid through the conductor, causing electromagnetic interference and affecting the normal operation of other electrical equipment.

[0006] 4. The high-frequency leakage current generated by the high-frequency common-mode voltage will cause misoperation of the ground current relay protection device, which is extremely harmful to the motor drive system and must be suppressed.

[0007] Currently, the research on common-mode voltage suppression mainly starts from two aspects. One is to optimize the inverter topology, but it often requires adding additional hardware, increasing the cost. The other is to disable relevant non-zero vectors, but this will affect the stability of the control system.

[0008] Therefore, it is very valuable to propose an optimized method for common-mode voltage suppression that can be achieved without adding additional hardware and disabling non-zero vectors. Summary of the Invention

[0009] The technical problem to be solved by the present invention is to provide a common-mode voltage suppression method based on a switching function in view of the deficiencies of the above-mentioned prior art. The common-mode voltage suppression method based on a switching function can, without adding additional hardware and disabling non-zero vectors, block the freewheeling path of the common-mode voltage spike by adjusting the switching function so that each switching tube can self-turn off when the corresponding freewheeling occurs, thereby achieving the suppression of the common-mode voltage spike.

[0010] To solve the above technical problem, the technical solution adopted by the present invention is:

[0011] A common-mode voltage suppression method based on a switching function includes the following steps.

[0012] Step 1: Establish a voltage vector set S w : For a two-level three-leg inverter, establish a voltage vector set S w ; The voltage vector set S w has 8 voltage vectors, specifically:

[0013]

[0014] Voltage vector set S w Among them, the three elements in each voltage vector respectively represent the current flow directions of the three phases of current in the frequency converter; the element "0" represents the current flowing into the frequency converter; the element "1" represents the current flowing out of the frequency converter.

[0015] Step 2: Construct voltage vector set S' w : Disable the zero vectors U0 and U7 in the voltage vector set S w to form a new voltage vector set S' w , specifically:

[0016]

[0017] In the voltage vector set S' w , U1, U3, U2, U6, U4, and U5 are arranged in a clockwise order, with a difference of 60° between each two. Let two voltage vectors with a difference of 60° be called adjacent vectors, two voltage vectors with a difference of 120° be called separated vectors, and two voltage vectors with a difference of 180° be called opposite vectors.

[0018] Step 3: Common-mode voltage suppression: Each phase leg of the frequency converter has two switching tubes, and the three-phase legs have a total of 6 switching tubes; in the voltage vector set S' w , when the separated vectors are switched, only one switching tube remains on during the dead time. The current directions of the three-phase legs are different, and the common-mode voltage is high; at this time, using the switching function, the one switching tube that remains on is also turned off, thereby reducing or suppressing the common-mode voltage.

[0019] In Step 3, the two switching tubes on each phase leg are respectively the upper switching tube and the lower switching tube; let the switching signals of the three upper switching tubes at the current moment be S1, S3, and S5, and the switching control signals of the three upper switching tubes at the next moment be S'1, S'3, and S'5; the switching signals of the three lower switching tubes at the current moment are S2, S4, and S6, and the switching control signals of the three lower switching tubes at the next moment are S'2, S'4, and S'6. Then the expression of the switching function is:

[0020]

[0021] In the formula, the symbol represents rounding up.

[0022] In step 3, the switching signals S1, S2, S3, S4, S5, and S6 of the six switching tubes at the current moment are obtained by controlling with the SVPWM vector control model; the SVPWM vector control model selects the optimal voltage vector at the current moment from the voltage vector set S′ w based on the current control model, and then obtains the switching signals S1, S2, S3, S4, S5, and S6 of the six switching tubes at the current moment.

[0023] The SVPWM vector control model includes a current control model and a cost function g; among them, the expression of the cost function g is:

[0024]

[0025] In the formula, is the set current of the stator d-axis; is the stator q-axis current, which is obtained by PI adjustment of the stator q-axis current of the motor at the set speed and the stator q-axis current of the motor at the feedback speed at the current k moment; i d (k + 2) and i q (k + 2) are the stator d-axis current and stator q-axis current at the k + 2 moment respectively, and are predicted by the current control model.

[0026] At the current k moment, the corresponding cost function g value is calculated for each voltage vector in the voltage vector set S′ w , and the voltage vector corresponding to the minimum g value is used as the optimal voltage vector at the current moment.

[0027] i d (k + 2) and i q (k + 2) prediction method includes the following steps:

[0028] Step 31, calculate i d (k) and i q (k): i d (k) and i q (k) are the stator d-axis current and stator q-axis current at the current k moment respectively, and are obtained by Park transformation of the three-phase currents i a , i b , and i c of the motor at the current k moment.

[0029] Step 32, calculate i d (k + 1) and i q (k + 1): i d (k + 1) and i q (k + 1) are the stator d-axis current and stator q-axis current at the k + 1 moment respectively, and are predicted by the stator current, and the specific prediction formula is:

[0030]

[0031] Wherein, u d (k) and u q (k) are the stator d-axis voltage and the stator q-axis voltage at time k, respectively.

[0032] L d and L q are the motor d-axis inductance and the motor q-axis inductance, respectively, obtained by measurement.

[0033] R is the per-phase load value of the motor, which is a known constant value; T s is the sampling period, which is a set value.

[0034] Step 33: Calculate i d (k + 2) and i q (k + 2), and the specific calculation formula is:

[0035]

[0036] Wherein, u d (k + 1) and u q (k + 1) are the stator d-axis voltage and the stator q-axis voltage at time k + 1, respectively.

[0037] In Step 31, the Park transformation formula is:

[0038]

[0039] Wherein, θ is the electrical angle of the motor, i0 is the zero-sequence current; P is the Park transformation matrix.

[0040] In Step 32, at the current time k, assume that the voltage vector being calculated is the nth row element of the voltage vector set S′ w and is denoted as S w (n, :); where, 1 ≤ n ≤ 6, then u d (k) and u q (k) are calculated as:

[0041]

[0042] Wherein, U represents the voltage output vector matrix at time k; U dc represents the DC power supply voltage; U(1) represents the first element of the matrix U; U(2) represents the second element of the matrix U.

[0043] In Step 33, at time k + 1, assume that the voltage vector being calculated is the mth row element of the voltage vector set S′ w and is denoted as S w(m, :); where 1 ≤ m ≤ 6, then u d (k + 1) and u q (k + 1) is calculated as follows:

[0044]

[0045] In the formula, U′ represents the voltage output vector matrix at time k + 1; U′(1) represents the first element of matrix U′; U′(2) represents the second element of matrix U′.

[0046] The present invention has the following beneficial effects:

[0047] 1. The present invention does not add extra control hardware and can be widely applied to two-level three-leg inverters to achieve the purpose of reducing the common-mode voltage.

[0048] 2. The present invention does not disable non-zero vectors, which makes this method not sacrifice the motor operation stability, and the signal generation method is simple, and the control system is simple and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a schematic diagram of the topology structure of a two-level three-leg inverter.

[0050] Figure 2 is a block diagram of the system model predictive control.

[0051] Figure 3 is a schematic diagram of 8 voltage vectors corresponding to a two-level three-leg inverter.

[0052] Figure 4 is a switching state diagram in the dead zone when the adjacent vector U4 switches to U2.

[0053] Figure 5 is a freewheeling diagram of the common-mode voltage spike generation when the adjacent vector U4 switches to U2.

[0054] Figure 6 is a freewheeling diagram after S6 is turned off in the dead zone when the adjacent vector U4 switches to U2

[0055] Figure 7 is a switching state diagram of each switch before and after switching when the adjacent vector U4 switches to U2.

[0056] Figure 8 is the simulation result of the common-mode voltage under ordinary predictive control.

[0057] Figure 9 is the simulation result of the common-mode voltage under predictive control using the switching function. DETAILED DESCRIPTION OF THE INVENTION

[0058] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific preferred embodiments.

[0059] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "left side", "right side", "upper part", "lower part", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. "First", "second", etc. do not represent the importance of the components, so it cannot be understood as a limitation to the present invention. The specific dimensions adopted in this embodiment are only for illustrating the technical solution and do not limit the protection scope of the present invention.

[0060] As Figure 1 shown, a two-level three-leg inverter includes a DC power supply U dc , three legs and a three-phase load.

[0061] The three legs are all connected in parallel on both sides of the DC power supply U dc , and are respectively the first leg, the second leg and the third leg.

[0062] The two switching tubes on each phase leg are respectively the upper switching tube and the lower switching tube; the switching signals of the three upper switching tubes at the current moment are S1, S3 and S5 respectively, and the switching signals of the three lower switching tubes at the current moment are S2, S4 and S6 respectively.

[0063] Each switching tube includes a switch and a freewheeling diode.

[0064] The three terminals of the three-phase load are respectively connected between the two switching tubes of the three legs, and the connection points are respectively points a, b and c.

[0065] Each phase load includes a load R and an inductor L.

[0066] Furthermore, the inverter further includes two series-connected voltage stabilizing capacitors. The middle connection point of the two voltage stabilizing capacitors is grounded, and the two ends of the two series-connected voltage stabilizing capacitors are connected in parallel across the DC power supply U dc ends.

[0067] As Figure 2 shown, a common-mode voltage suppression method based on a switching function includes the following steps.

[0068] Step 1, establish a voltage vector set S w

[0069] As Figure 3 shown, for a two-level three-leg inverter, establish a voltage vector set S w ; the voltage vector set S w has 8 voltage vectors, specifically:

[0070]

[0071] Voltage vector set S w Among them, the three elements in each voltage vector respectively represent the current flow directions of the three phases in the frequency converter; the element "0" represents that the current flows into the frequency converter; the element "1" represents that the current flows out of the frequency converter.

[0072] Step 2: Construct the voltage vector set S' w : Disable the zero vectors U0 and U7 in the voltage vector set S w to form a new voltage vector set S' w , specifically:

[0073]

[0074] In the voltage vector set S' w , U1, U3, U2, U6, U4, and U5 are arranged in a clockwise order, with a difference of 60° between each two. Let two voltage vectors with a difference of 60° be called adjacent vectors, two voltage vectors with a difference of 120° be called separated vectors, and two voltage vectors with a difference of 180° be called opposite vectors.

[0075] Step 3: Common-mode voltage suppression: Each phase bridge arm of the frequency converter has two switching tubes, and the three-phase bridge arm has a total of 6 switching tubes; in the voltage vector set S' w , when the separated vectors are switched, only one switching tube remains on during the dead time. The current directions of the three-phase bridge arms are different, and the common-mode voltage is high; at this time, using the switching function, the one switching tube that remains on is also turned off, thereby reducing or suppressing the common-mode voltage.

[0076] Let the switching control signals of the three upper switching tubes at the next moment be S'1, S'3, and S'5 respectively; the switching control signals of the three lower switching tubes at the next moment be S'2, S'4, and S'6 respectively, then the expression of the switching function is:

[0077]

[0078] In the formula, the symbol represents rounding up.

[0079] Taking the switching from vector U4(100) to vector U2(010) as an example in the present invention, as Figure 4 shown: During the dead time of the switching from U4 to U2, only S6 remains on, S1 and S4 are turned off during the dead time, and S2 and S3 are turned on after the dead time. At this time, these four switching tubes all rely on diodes for freewheeling. During this stage, if the current direction flowing out of the frequency converter is defined as 1 and flowing into the frequency converter is defined as 0, then when i a 、ib 、i c If the current direction is (110), a freewheeling path in Figure 5 will be generated. At this time, the common-mode voltage of -U / 2 is induced in all three phases a, b, and c, generating a common-mode voltage spike. dc / 2 of the common-mode voltage, generating a common-mode voltage spike.

[0080] Therefore, the present invention adopts an SVPWM vector control model to control the switching signals of the six switching tubes at the current moment. The principle of the SVPWM vector control model is as follows: Based on the current control model, the optimal voltage vector at the current moment is selected from the voltage vector set S′ w to obtain the switching signals S1, S2, S3, S4, S5, and S6 of the six switching tubes at the current moment.

[0081] The SVPWM vector control model includes a current control model and a cost function g; among them, the expression of the cost function g is:

[0082]

[0083] In the formula, is the set current of the stator d-axis; is the stator q-axis current, which is obtained by PI regulation of the stator q-axis current of the motor at the set speed and the stator q-axis current of the motor at the current feedback speed at the kth moment; i d (k + 2) and i q (k + 2) are the stator d-axis current and stator q-axis current at the (k + 2)th moment respectively, which are predicted through the current control model.

[0084] At the current kth moment, the corresponding cost function g value is calculated for each voltage vector in the voltage vector set S′ w , and the voltage vector corresponding to the minimum g value is used as the optimal voltage vector at the current moment.

[0085] i d (k + 2) and i q (k + 2) The prediction method includes the following steps:

[0086] Step 31. Calculate i d (k) and i q (k): i d (k) and i q (k) are the stator d-axis current and stator q-axis current at the current kth moment respectively, which are obtained by Park transformation of the three-phase currents i a , i b and i c of the motor at the current kth moment.

[0087] Among them, the Park transformation formula is:

[0088]

[0089] Wherein, θ is the electrical angle of the motor, i0 is the zero-sequence current; P is the Park transformation matrix.

[0090] Step 32: Calculate i d (k + 1) and i q (k + 1): i d (k + 1) and i q (k + 1) are respectively the stator d-axis current and the stator q-axis current at the moment of k + 1, which are obtained by predicting the stator current. The specific prediction formula is:

[0091]

[0092] Wherein, u d (k) and u q (k) are respectively the stator d-axis voltage and the stator q-axis voltage at the moment of k.

[0093] L d and L q are respectively the d-axis inductance of the motor and the q-axis inductance of the motor, which are obtained by measurement.

[0094] R is the load value of each phase of the motor, which is a known constant value; T s is the sampling period, which is a set value.

[0095] In the present invention, at the current moment of k, assume that the voltage vector being calculated is the nth row element of the voltage vector set S′ w in, denoted as S w (n, :); wherein, 1 ≤ n ≤ 6, then the calculation formulas of u d (k) and u q (k) are:

[0096]

[0097] Wherein, U represents the voltage output vector matrix at the moment of k; U dc represents the DC power supply voltage; U(1) represents the first element of the matrix U; U(2) represents the second element of the matrix U.

[0098] Step 33: Calculate i d (k + 2) and i q (k + 2), and the specific calculation formula is:

[0099]

[0100] Wherein, u d (k + 1) and u qu_d(k + 1) and u_q(k + 1) are the stator d-axis voltage and stator q-axis voltage at the (k + 1)-th moment respectively.

[0101] In the present invention, at the (k + 1)-th moment, the voltage vector being calculated is the m-th row element of the voltage vector set S', denoted as S(m, :); where 1 ≤ m ≤ 6, then the calculation formulas for u_d(k + 1) and u_q(k + 1) are: w in S' w (m, :); where 1 ≤ m ≤ 6, then u d (k + 1) and u q (k + 1) are:

[0102]

[0103] In the formula, U' represents the voltage output vector matrix at the (k + 1)-th moment; U'(1) represents the first element of the matrix U'; U'(2) represents the second element of the matrix U'.

[0104] In the present invention, due to the use of the switching function, in the dead time, since S1, S2, S3, and S4 are all off and only S6 is on, at this time, S1 = S2 = S3 = S4 = 0. Therefore, according to formula (1), S6' will also be equal to zero, that is, the switching function will turn off the switch tube S6, and then its freewheeling path will become Figure 6 the situation shown. Since S6 is turned off, the current on this bridge arm turns to freewheel through the diode of S5, and the common-mode voltage generated at this time is -U / 6, and the common-mode voltage spike is weakened. dc / 6, and the common-mode voltage spike is weakened.

[0105] Figure 7 Fig. is the schematic diagram of the switching state when the voltage vector U4 switches to U2 under the control of the switching function. It can be seen that during the switching, S1, S2, S3, and S4 are all in the off state (signal value is 0) in the dead time. According to the switching function, the input signal of S6 becomes 0 at this time, realizing the function of turning off the corresponding switch tube in the dead time. Similarly, when other adjacent vectors switch, the corresponding switch tubes can also be turned off according to the switching function.

[0106] Since there is only a situation where the four switch tubes on two bridge arms are all turned off in the dead time during the adjacent vector switching, the switching function can well solve the problem of generating common-mode voltage in the dead time during the adjacent vector switching. At the same time, since the action only occurs in the dead time and the signals of each switch tube remain the original vector after the dead time, the switching function will not affect the vector action of the control system.

[0107] The switching function essentially binds the signals of each switching tube to the signals of four switching tubes on the other two bridge arms. This is because only when the adjacent vectors are switched, there will be a situation where the four switching tubes on the two bridge arms are all turned off within the dead time. At this time, turning off the switched-on switching tube on the other bridge arm can block the freewheeling path. At the same time, since the action only occurs within the dead time and returns to normal after the dead time ends, it will not affect the change of the actual operating vector. The mathematical relationships of the respective switching signals set by the switching function can well achieve this purpose.

[0108] Figure 8 and Figure 9 The common-mode voltage simulation results without using the switching function and with using the switching function are shown. It can be seen that after using the switching function control, the common-mode voltage spikes are effectively eliminated.

[0109] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A common-mode voltage suppression method based on a switching function, characterized in that: It includes the following steps: Step 1: Establish the voltage vector set S w : For a two-level three-leg frequency converter, establish the voltage vector set S w ; The voltage vector set S w contains 8 voltage vectors, specifically: Voltage vector set S w Among them, the three elements in each voltage vector respectively represent the current flow directions of the three phases in the frequency converter; the element "0" represents that the current flows into the frequency converter; the element "1" represents that the current flows out of the frequency converter; Step 2: Construct the voltage vector set S' w : Disable the zero vectors U0 and U7 in the voltage vector set S w to form a new voltage vector set S' w , specifically: In the voltage vector set S′ w , U1, U3, U2, U6, U4, and U5 are arranged in a clockwise order, with a difference of 60° between each pair. Let two voltage vectors with an angle difference of 60° be called adjacent vectors, those with a difference of 120° be called separated vectors, and those with a difference of 180° be called opposite vectors. Step 3, common-mode voltage suppression: Each phase leg of the frequency converter has two switching tubes, and there are a total of six switching tubes for the three-phase legs; in the voltage vector set S' w when the adjacent vectors are switched to each other, only one switching tube remains on during the dead time. The current directions of the three-phase legs are different, and the common-mode voltage is high. At this time, the switching function is used to turn off the one switching tube that remains on, thereby reducing or suppressing the common-mode voltage; The two switching tubes on each phase leg are respectively the upper switching tube and the lower switching tube; assume that the switching signals of the three upper switching tubes at the current moment are S1, S3, and S5 respectively, and the switching control signals of the three upper switching tubes at the next moment are S1′, S3′, and S5′ respectively; the switching signals of the three lower switching tubes at the current moment are S2, S4, and S6 respectively, and the switching control signals of the three lower switching tubes at the next moment are S2′, S4′, and S6′ respectively. Then the expression of the switching function is: In the formula, the symbol represents rounding up.

2. The common-mode voltage suppression method based on a switching function according to claim 1, characterized in that: In step 3, the switching signals S1, S2, S3, S4, S5, and S6 of the six switching transistors at the current moment are obtained by controlling with the SVPWM vector control model; the SVPWM vector control model, based on the current control model, selects the optimal voltage vector at the current moment from the voltage vector set S' w to obtain the switching signals S1, S2, S3, S4, S5, and S6 of the six switching transistors at the current moment.

3. The common-mode voltage suppression method based on a switching function according to claim 2, wherein: The SVPWM vector control model includes a current control model and a value function g; among them, the expression of the value function g is: Wherein, is the set current of the stator d-axis; is the stator q-axis current, which is obtained by PI regulation of the stator q-axis current of the motor at the set speed and the stator q-axis current of the motor at the feedback speed at the current k-th moment; i d (k + 2) and i q (k + 2) are the stator d-axis current and the stator q-axis current at the (k + 2)-th moment respectively, which are predicted by the current control model; At the current k moment, for each voltage vector in the voltage vector set S′ w calculate the corresponding value function g value. When the g value is the smallest, the corresponding voltage vector is used as the optimal voltage vector at the current moment.

4. The common-mode voltage suppression method based on a switching function according to claim 3, characterized in that: i d (k + 2) and i q (k + 2) prediction method, comprising the following steps: Step 31, calculate i d (k) and i q (k): i d (k) and i q (k) are the stator d-axis current and the stator q-axis current at the current k-th moment respectively, and are obtained by performing Park transformation on the three-phase currents i a , i b and i c of the motor at the current k-th moment; Step 32, calculate i d (k + 1) and i q (k + 1):i d (k + 1) and i q (k + 1) are the stator d-axis current and the stator q-axis current at the (k + 1)th moment, respectively, which are obtained by predicting the stator current. The specific prediction formula is as follows: where, u d (k) and u q (k) are the stator d-axis voltage and the stator q-axis voltage at the k-th moment, respectively; L d and L q are the d-axis inductance and q-axis inductance of the motor, respectively, obtained by measurement; R is the load value per phase of the motor, which is a known constant value; T s is the sampling period, which is a set value; Step 33, calculate i d (k + 2) and i q (k + 2), and the specific calculation formula is as follows: where, u d (k + 1) and u q (k + 1) are the stator d-axis voltage and the stator q-axis voltage at the (k + 1)-th moment, respectively.

5. The common-mode voltage suppression method based on a switching function according to claim 4, characterized in that: In step 31, the Park transformation formula is: In the formula, θ is the electrical angle of the motor, i0 is the zero-sequence current; P is the Park transformation matrix.

6. The method for suppressing common-mode voltage based on a switching function according to claim 5, wherein: In step 32, at the current k moment, the voltage vector being calculated is the nth row element in the voltage vector set S′ w , denoted as Sw(n, :); where 1 ≤ n ≤ 6, then u d (k) and u q (k) are calculated as follows: wherein, U represents the voltage output vector matrix at the k-th moment; U dc represents the DC power supply voltage; U(1) represents the first element of the matrix U; U(2) represents the second element of the matrix U.

7. The common-mode voltage suppression method based on a switching function according to claim 5, characterized in that: In step 33, at the (k + 1)-th moment, the voltage vector being calculated is the m-th row element in the voltage vector set S′ w denoted as Sw(m, :); where 1 ≤ m ≤ 6, then u d (k + 1) and u q (k + 1) are calculated by the following formulas: In the formula, U′ represents the voltage output vector matrix at the k + 1 moment; U′(1) represents the first element of the matrix U′; U′(2) represents the second element of the matrix U′.

Images