PROCEEDINGS
OF THE TOMSK POLYTECHNIC INSTITUTE
OF THE ORDER OF THE OCTOBER REVOLUTION AND THE ORDER
OF THE RED BANNER OF LABOR
named after S. M. KIROV

Volume 285　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　1975

RELAY ACS OF A TRACTION ELECTRIC MOTOR WITH VARIABLE STRUCTURE

A. P. ZAITSEV, V. A. PODLYAGIN

(Presented by the scientific-technical seminar of the Scientific Research Institute of AEM)

Synchronization of self-oscillations in a relay automatic control system ensures a constant switching frequency of the relay element while maintaining high static and dynamic characteristics of the control process. In [1] the principle of operation is considered of a relay ACS of the current of a traction electric motor, in which synchronization of self-oscillations is ensured by supplying periodic pulses to switch off the regulating thyristor switch.

A disadvantage of such an ACS of the traction electric-motor current is the limited value of the maximum repetition frequency of the synchronization pulses. At a certain critical value of the repetition frequency of the synchronization pulses, at the beginning of the control process self-oscillations with a constant frequency are possible, caused by the comparatively slow decrease of the current to the prescribed level.

Below a relay ACS of the traction-motor current is considered that ensures constancy of the switching frequency at higher synchronization frequencies. A substantial increase in the limiting possible synchronization frequency is due to an automatic change in the structure of the ACS in such a way that the controlled current reaches the prescribed level along steeper sections of the curves of its rise or fall. In the process of accelerating the electric motor, when the relative on-time is \(\gamma < 0.5\), the controlled thyristor switch is opened by synchronizing pulses following with a constant frequency, and is closed at the moments when the current rises to the prescribed level; and when the relative on-time is \(\gamma > 0.5\), the control logic is changed: the thyristor switch is closed by periodic synchronizing pulses, and is opened at the moments when the controlled current decreases to the prescribed level.

Let us consider the operation of the system with variable structure according to the block diagram of Fig. 1 and the time diagrams of Fig. 2.

At the beginning of motor starting, \(\gamma < 0.5\); the contacts of the switching device PU are in the lower position, as shown in Fig. 1. The voltage pulses \(U_1\) of the setting generator ZG, which is a blocking oscillator, arrive simultaneously at the shaper of blocking pulses FBI and at the shaper of delayed pulses FZI. The voltage pulses \(U_3\) formed by FZI are shifted relative to the voltage pulses \(U_1\) by a certain time \(\Delta t\) (Fig. 2). Through the closed contact of the switching device PU they arrive at the lower arm of the trigger Tr and switch it into one of its stable states. On the trailing edge of the voltage \(U_7\), taken from ...

Fig. 1.

Fig. 2.

from the lower arm of the flip-flop \(T_p\), the amplifier \(U_2\) is started, which forms at its output a voltage pulse \(U_9\), opening the thyristor switch \(TK\). From this moment on, the regulated armature current \(I\) begins to increase along the steep segment up to the specified level \(I_0\). At the instant when the currents \(I\) and \(I_0\) are compared, the voltage \(U_d\) of the armature-current sensor \(DTYA\) becomes equal to the reference voltage \(U_0\), and the comparison device \(CC\)

will cease generating the voltage pulses \(U_5\). This removes the prohibition on operation of the “NOT” inversion device, and a voltage pulse \(U_4\) will appear at the output of the latter. The voltage pulse \(U_4\), through the closed contact of the switching device PU, is applied to the upper arm of the trigger \(Tr\). On the trailing edge of the voltage \(U_6\), amplifier \(U_1\) is started, forming a voltage pulse \(U_8\), which acts to close the thyristor switch TK. Thus, the system maintains at a constant level the average value of the controlled current \(I\). Under these conditions the controlled current \(I\) in no case exceeds the specified value \(I_0\).

The DIS structure-change sensor, to whose inputs the voltages \(U_6\) and \(U_7\) from the arms of the trigger \(Tr\) are continuously applied, continuously monitors the relative on-duration \(\gamma\) and, when during acceleration of the electric motor \(\gamma \geq 0.5\), issues a signal in the form of voltage \(U_{10}\) for switching PU. PU switches its contacts to the position opposite to that shown in Fig. 1, thereby changing the control logic of the thyristor switch.

After PU switches, the synchronizing pulses arriving at the thyristor switch act to close it, while the opening pulses arrive from the output of the comparison device SS when the controlled current \(I\) decreases to a certain value \(I_0\). Under these conditions the controlled current \(I\) in no case falls below the specified value \(I_0\).

As can be seen from the time diagrams of Fig. 2, the average values of the controlled current \(I\) for different positions of the switching device differ somewhat from one another. Equality of the average current values for different system structures can be ensured by changing the magnitude of the reference voltage at the moment the structure is changed.

For reliable operation of the circuit, the voltage pulses \(U_8\) and \(U_9\), which act to close and open the thyristor switch, must not be located too close to one another; otherwise a failure of TK will occur. This is achieved by placing the pulse \(U_3\) at the center of the voltage pulse \(U_2\) formed by the blocking-pulse shaper FBI. The voltage pulses \(U_2\) of FBI open transistors 1 and 2 (Fig. 1) and thereby, for the duration of their action, prohibit the passage of the voltage pulses \(U_4\) and \(U_5\). The duration of the blocking voltage pulse \(U_2\) is composed of the charge-discharge times of the commutating capacitance of the thyristor switch TK used.

To start the system (when the contacts of the switching device PU are in the lower position), an additional connection is provided, by means of which the voltage pulse \(U_3\) from FZI is applied directly to amplifier \(U_2\). The system is started independently of the initial state of the trigger \(Tr\). To a greater extent, the precise operation of the system depends on the speed of response of the “NOT” inversion circuit, which is controlled by the comparison device SS.

Let us consider the operation of these devices, which fully meet the speed requirements, according to the block diagram of Fig. 3 and the time diagrams of Fig. 4.

The comparison device consists of a sawtooth-voltage shaper FPN and a comparator K. The voltage \(U_d\) from the DTYa is applied to the FPN and is converted by it into a sawtooth voltage \(U_{\mathrm{fip}}\) (Fig. 4). The amplitude of the voltage \(U_{\mathrm{fip}}\), proportional to the half-period of the voltage \(U_d\), is compared by comparator K with a certain reference voltage \(U_0\). At the instant these voltages are compared, the comparator generates voltage pulses \(U_5\), which are applied to the trigger \(Tr\). To the other input of the trigger \(Tr\), from the differentiating circuit DTs, are applied

Fig. 3.

Fig. 4.

voltage pulses \(U_{\mathrm{dc}}\). When comparator \(K\) generates voltage pulses \(U_5\), voltage pulses \(U_{\mathrm{tr}}\) are formed at trigger \(Tr\). The voltage pulses \(U_{\mathrm{tr}}\) at the instants \(t_1\) act on the blocking generator \(ZBG\) and inhibit its operation. In the second half-period (time \(t_2\)), the operation of \(ZBG\) is inhibited by the voltage pulses \(U_i\) coming from the inverter \(I\). At the instants when voltage \(U_5\) disappears, trigger \(Tr\) switches to one of its stable states, permitting the operation of \(ZBG\).

The sensor for a change in the structure of the DCS (block diagram in Fig. 5) consists of integrators \(\Sigma_1\) and \(\Sigma_2\), comparators \(K_1\) and \(K_2\), and trigger \(Tr\).

Voltage pulses \(U_6\) and \(U_7\), whose durations are equal respectively to the on and off times of the thyristor switch \(TK\), are fed from the arms of trigger \(Tr\) (Fig. 1) to integrators \(\Sigma_1\) and \(\Sigma_2\). The voltage levels \(U_{\Sigma 1}\), \(U_{\Sigma 2}\) at the outputs of the integrators will depend on the duty factor of the applied pulses \(U_6\), \(U_7\). The output voltages of the integrators \(U_{\Sigma 1}\) and \(U_{\Sigma 2}\) are compared by comparators \(K_1\) and \(K_2\) with certain reference voltages \(U_{e1}\) and \(U_{e2}\). Depending on

Fig. 5.

Fig. 6.

on the relative duration of switching-on, \(\gamma\), one of the comparators will generate a signal, transferring the trigger \(Tr\) into the corresponding state. The trigger \(Tr\), in turn, controls the operation of the switching device \(PU\), which changes the structure of the system.

It should be noted that the reference voltages \(U_{31}\) and \(U_{32}\) should be set somewhat apart in order to exclude simultaneous generation by the comparators \(K_1\) and \(K_2\), and consequently also uncertainty in the state of the trigger \(Tr\) (Fig. 6).

REFERENCES

1. A. P. Zaitsev, V. A. Podlyagin. A two-phase relay system for regulating the starting current of traction motors. Izvestiya TPI, vol. 262, Tomsk, 1972.