# Analysis and Solution of Sector Transition Defects in Permanent Magnet Motor Drives

Analysis and solution of sector transition defects in permanent magnet motor inverters of electric drive automation CHEN Huang-xing i Bao-hui (College of Electrical Engineering, Zhejiang University, Hangzhou 310027, Zhejiang, China) Principle of pulse width modulation, analyzes the reasons for these problems, and gives Possible solutions.

1 Overview Since space vector pulse width modulation (SVPWM) has considerable advantages in current harmonic suppression and voltage utilization, it has been widely used in AC speed control systems. This paper applies SVPWM technology and develops a three-phase variable-frequency power supply based on TMS32F240DSP, so as to realize the variable frequency control of the permanent magnet synchronous motor. However, in the actual operation process, there is a defect that can not be ignored in the sector transition of SVPWM. This paper analyzes the multiple causes of this engineering problem from the principle of space vector pulse width modulation and gives a feasible solution. .

2 Derivation of Formulas for Space Vector Pulse Width Modulation According to the three-phase inverter circuit shown, it is possible to infer the relationship between the state of the three upper limb switch tubes 1, 23 and the a and å© axes, and then to obtain the switch tube 1 The relationship between the state of the 2, 3, and the space voltage vector is as shown. As can be seen, the switch tubes 1, 23 combined into eight switch states correspond to eight kinds of space voltage vectors, where: V, V are zero vectors, and these voltage vectors divide the space into 6 sectors.

The basic principle of space vector pulse width modulation is to decompose any one space voltage vector into two voltage vectors and one zero vector adjacent to the sector where it is located. The size of each component after decomposing represents the action time of the voltage vector.

(1) Calculation of the sector where the space voltage vector is located For the variable frequency circuit of the permanent magnet synchronous motor, the control strategy with id = 0 is used here. After id and iq are adjusted by the PI, the ud and Uq are output according to the version and Uq component of the voltage vector. The position of the sector can be calculated directly. The calculation method is as follows: "Axis position; 9 is the position of the rotor relative to the "phase axis; 9 is the angle between the voltage vector and the d axis."

Therefore, the sector number where any space voltage vector is equal to 9 is divided by 60 and the remainder is added by 1. (2) Calculation of the space voltage vector on-time Md, Uq are obtained through Parker inverse transformation M, M, 3 Now suppose that V is at the position shown, the on-times T4 and T6 of the decomposed voltage vectors V4 and V can be obtained. Among them, K=Uc, T is the PWM carrier period value.

By analogy, the general formula for calculating the conduction time of the space voltage vector in any sector is: Sector 1: Sector Sector 6: Space.

Analysis of the causes of the 3 sector transition defect and its solution In theory, the above formula for calculating the space voltage vector conduction time has no problem. However, during the test process, it is found that the calculated ti and t2 often make mistakes; and when the motor operating frequency is lower, The more errors there are. After careful observation of its error rule, it was found that such errors always occur when the voltage vector sector transitions. The reason is mainly due to the following three special conditions of the inverter.

(1) The sector in which the actual voltage vector V is located is erroneously judged as an adjacent sector, and thus the voltage vector resulting from the calculation of V. U., U3 also falls within this misjudged sector, and thus the calculated conduction The time does not appear negative, as shown. In a physical sense, the solution is that the voltage vector V originally required to be generated is replaced by another voltage vector. This deviation caused by purely arithmetic error can be tolerated.

(2) Compute the algorithm error of the arctangent transform and the divide-by-arrow error causes the spatial voltage vector actual sector to be misjudged as an adjacent sector. Therefore, the voltage vector is divided according to the two switching voltage vectors of the adjacent sectors. It can be seen that the voltage vector V (9 in the figure is a very small angle) in the first sector should be decomposed into T2V, and fV4. But the result of the calculation is the second sector, so the decomposition result is and TV2. In physics In the sense, since there is no negative number of times, it is not possible to generate V2 as shown by the voltage vector V2 and the actually calculated T1 appears negative. At the same time by the introduction: T2 because IV> l is not:: iV2i, after the error on-time after on-time Tr and t / min according to the cause of the error can be corrected, because the included angle between the sectors are 60 Â° From the radial relationship we can draw: Tr = 0, that is, only conduction, V2 or V4 are not conductive.

Therefore, according to the above analysis of the sector transition problem, the first case belongs to the allowable deviation and is not included in the discussion; the latter two cases have the problem that the conduction time is less than 0 in the calculation process. However, it is impossible to discern what kind of situation is in the program, so the two situations can be summarized and solved in a simple and easy way. Here, we propose a solution to the third case.

In this way, the actual control program needs to add a correction procedure for the voltage vector sector transition. The flow is: where T/, is the calculation result before correction, and Ti, T2 are the corrected values.

In this case, since it is impossible to know how large the deviation between V and V is, it is always around the voltage vector V6. Therefore, it is possible to use a vector whose magnitude is the same as j/, and whose direction is V6. Because 9 is small, V and V have the same amplitude, so when the calculated conduction time is less than 0 at 890Â°, it indicates that one of the two types of sector transition defects indicated in this paper has appeared. At this time, the corrective procedure is called. can.

4 Conclusions Tests have shown that when a voltage vector sector transition correction procedure is added, the sector transition is observed from the data collected in real time by the simulator, and the result is completely correct. This avoids sudden jumps in the voltage vector due to calculation errors, thus ensuring the smoothness of the torque of the permanent magnet motor and perfecting the control technology of the SVPWM to the permanent magnet motor.

1.Our Polycarbonate Sheet,tiles will not be broken within 10 years under the weather and impact strength. Yellowing index and transmittance loss will not exceed the value defined by warranty;

2.We guarantee for the surface of pc sheet not caused collateral damage due to hail impact, but our guarantee not including the damage caused by other impacts such as scratches, abrasions, exposure, self-storage, transportation, handling and improper installation, physical and chemical changes occur.

Polycarbonate Awning,Awning Bracket,Pc Awning,Polycarbonate Door Awnings

FOSHAN KASO BUILDING MATERIAL CO.,LTD , https://www.kasocover.com