Katsuhiro Izumi

A sensorless vector control system for induction motors using q-axis flux with stator resistance identification

This paper presents a sensorless vector control system for general-purpose induction motors, which is based on the observer theory and the adaptive control theories. The proposed system includes a rotor speed estimator using a q-axis flux and stator resistance identifier using the d-axis flux. The advantages of the proposed system are simplicity and avoidance of problems caused by using only a voltage model. Since the mathematical model of this system is constructed in a synchronously rotating reference frame, a linear model is easily derived for analyzing the system stability, including the influence of the observer gain, motor operating state, and parameter variations. In order to obtain stable low-speed operation and speed control accuracy, an algorithm for compensating for the deadtime of the inverter and correcting the nonideal features of an insulated gate bipolar transistor was developed. The effectiveness of the proposed system has been verified by digital simulation and experimentation.

A simplified MRAS based sensorless vector control method of induction motor

This paper presents a new sensorless vector control system that is based on the model reference adaptive system (MRAS) theory. The estimation of stator currents and rotor flux is constructed in a synchronously rotating reference frame, a simple slip frequency control is obtained and a linear model is easily derived. By a constant flux control and estimation of q-axis current only, a very simple sensorless scheme is proposed too. By computing the poles and zeros of the speed transfer function, the system stability is discussed. The effectiveness of the proposed system has been investigated by digital simulation and experimentation.

A novel parameter identification of vector controlled induction motor using a phase lag current control

This paper presents a novel identification of stator and rotor resistances for the vector controlled induction motor system. A current controller that has a phase lag element instead of conventional PI controller is used for the parameter identification without losing current control property. Both stator and rotor resistances are tuned by driving the errors between the references of d- and q-axis currents and their actual values to zero. Therefore, the proposed system is simpler than the conventional MRAS based parameter identification. In the motoring and regenerating modes, the system stability is guaranteed by computing the trajectories of poles from a linear model. The effectiveness of proposed system is verified by digital simulation and experimentation. The improved steady-state torque characteristics are demonstrated.

Stability improvement of speed sensorless induction motor vector control system using q-axis flux with stator resistance identification

We have proposed a speed sensorless vector control system for induction motors using q-axis flux with parameter identification. In this paper, we go further to discuss the system stability in both motoring and regenerative operation by computing the pole-zero trajectories of the linear model, with the emphasis on the error influence of stator resistance and its identification gain. The stability improvement is significant by changing the sign of the identification gain, according to the motor operation state. The effectiveness is verified by digital simulation and experiment.

A speed sensorless induction motor vector control system using q-axis flux with parameter identification

In this paper, a speed sensorless vector control system for induction motors is presented, which includes rotor speed estimation and stator resistance identification. A linear model of the system is proposed to analyze the changes in observer and speed estimation gains. By computing the trajectories of poles and zeros and the transient responses, the system stability is discussed. The effectiveness is verified by digital simulation and experimental results.

Model following servo control of CSI-IM vector control system

An application of model-following servo (MFS) control to the current source inverter (CSI)-fed induction motor (IM) vector control system is studied. The MFS control is a kind of two degree-of-freedom control, and the controller is designed systematically by using optimal regulator theory. For the speed control of IM, a comparison between the MFS control, conventional PI (proportional plus integral) control, and PI control with a load torque observer is made under the assumption of ideal vector control. A linear model in which a PI current controller and the change of rotor resistance are considered is presented. The robustness of the MFS control is discussed by examining the trajectories of poles and zeros and the transient responses.<<ETX>>

Stability analysis of a CSI-fed inductor motor with digital vector controller

A linear sampled-data model for a current source inverter (CSI)-fed induction-motor vector control system which is controlled digitally by a microprocessor is derived. The system stability is discussed from the viewpoints of sampling period, microprocessor computation time, and feedback parameters by examining the loci of dominant eigenvalues of the state transition matrix and the transient responses. A comparison between the sampled-data model and a continuous system model is given.<<ETX>>

Active power filter with optimal servo controller

This paper presents a reduction method of higher harmonic currents using an active filter at the AC side of a condenser input-type three-phase diode bridge rectifier. As an active filter, a three-phase PWM inverter is connected in parallel to rectifier. The principle of the active filter is to cancel higher harmonic currents contained in load current by injecting reversed phase harmonic currents into the power source side. In this system, an optimal digital servo controller with dead time compensation is used to reduce the higher harmonics at the current controller. The root locus method and the simulation are used to obtain optimal weighting factor of a performance criterion and to select the control method. Finally, power spectra of a load side current and a source side one are shown. Experimental results are shown to verify the usefulness of this system.

Improvement of fuzzy auto-tuning method of DC chopper system using manipulated value

An auto-tuning using fuzzy reasoning is discussed for the I-PD control systems. We have previously (1993) proposed one method about same topic. In this paper, we have proposed two new methods of reasoning and compared with conventional ones. In the fuzzy auto-tuning, the characteristic values are calculated with the reference, the controlled, and the manipulated value at each sampling time. Next, new gains are set by multiplication with adjustment coefficients determined by fuzzy reasoning and last gain values. The fuzzy reasoning is obtained by a product-sum compositional rule of inference, and then defuzzified by the height procedure to generate a nonfuzzy value. In the simulation, the settling time by using these was shorter than one using a conventional value, which only uses controlled value and reference. An appropriate control response is obtained by tuning of several times on DSP control system. In the experiment, the tuning time using the new proposed characteristic value was shorter than that using the previous method. In the result, the effectiveness of this method is shown.<<ETX>>

Dead beat control of DC chopper system with nonlinear load

Two-parameter deadbeat control is applied to the current control of a DC chopper system with lamp load. The two-parameter deadbeat control of the DC current is achieved using a single-chip digital signal processor (DSP). The control program for the DSP has a short computation time. In the actual system, the robustness against the change of controlled system parameter is confirmed.<<ETX>>