Apparatus for driving and controlling synchronous motor using permanent magnets as its field system

0001The present invention relates to a driving and controlling apparatus for controlling the output torque of a synchronous motor using permanent magnets as its field system (hereafter referred to as PM motor) so as to attain a reference torque by controlling the vector of a motor current. Particularly, it relates to the control of a PM motor which can be suitably mounted on an electric vehicle and can execute field weakening control.

0002A PM motor is a synchronous motor using permanent magnets as its field system. Therefore, it is characterized by having a large magnetomotive force per unit volume. This characteristic is effective for increasing the output of a motor and reducing the size of a motor. Particularly, the PM motor is preferable as the drive motor of an electric vehicle.

0003FIG. 20 shows an equivalent circuit per phase of a PM motor. In FIG. 20, the magnetomotive force of permanent magnets used as a field system, that is, the main magnetic flux of the PM motor, is shown by the symbol E₀. When assuming the axial angular velocity of the PM motor as ω (=2πN, N: rotational speed of the motor), the counter electromotive force generated during operation of the PM motor can be expressed as ωE_o. When the PM motor is driven by converting the discharge output of a battery into alternating current by an inverter and using the AC power, the terminal voltage V of the PM motor is obtained by multiplying a battery voltage V_dc by the voltage conversion ratio of the inverter. In FIG. 20, symbol R represents the primary resistance per phase of the PM motor, L represents the inductance per phase, and I represents the primary current (phase current).

0004The motor current I of the PM motor can be decomposed into the vector components, i.e., the field current I_d and the torque current I_q. The field current I_d is a component for generating a field flux in the PM motor and the torque current I_q is a component for generating torque by intersecting with the field flux of permanent magnets, that is, the main magnetic flux E₀. When the PM motor is provided with saliency, the field current I_d also generates torque. The torque generated when the torque current I_q intersects with the main magnetic flux E₀ is known as magnetic torque and the torque caused by the saliency is known as reluctance torque. The following equation (1) is an expression showing the torque T of the PM motor, in which the first term at the right side shows magnetic torque and the second term shows reluctance torque. Therefore, the second term does not appear when the PM motor is a non-salient-pole motor.<maths id="math0001" num="(1)"><math display="block"><mrow><msub><mrow><mtext>T = E</mtext></mrow><mrow><mtext>0</mtext></mrow></msub><msub><mrow><mtext> + (L</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext> - L</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext>)I</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>q</mtext></mrow></msub></mrow></math><img file="EP0638457B1_D0001.tif" /></maths>

0005L_d and L_q in the equation (1) are called d-axis inductance and q-axis inductance, respectively, which are the d-axis component and q-axis component of the inductance L of the PM motor, respectively. The terminal voltage V of the PM motor can also be decomposed into the d-axis component and q-axis component. The d-axis component and q-axis component of the terminal voltage V, that is, d-axis voltage V_d and q-axis voltage V_q, can be expressed as shown below by using the field current I_d, torque current I_q, d-axis inductance L_d, and q-axis inductance L_q when the number of pole pairs is 1.<maths id="math0002" num="(2)"><math display="block"><mrow><mtable><mtr><mtd><mrow><mtable><mtr><mtd><mrow><msub><mrow><mtext>V</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext> = RI</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext> + jωL</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>q</mtext></mrow></msub></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext>V</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext> = jωL</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext> + RI</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext> + ωE</mtext></mrow><mrow><mtext>0</mtext></mrow></msub></mrow></mtd></mtr><mtr><mtd><mrow><mtext>j: An imaginary component</mtext></mrow></mtd></mtr></mtable></mrow></mtd></mtr></mtable></mrow></math><img file="EP0638457B1_D0002.tif" /></maths>

0006Moreover, the following relationship exists between d-axis voltage V_d and q-axis voltage V_q.<maths id="math0003" num="(3)"><math display="block"><mrow><msup><mrow><mtext>|V|</mtext></mrow><mrow><mtext>2</mtext></mrow></msup><msub><mrow><mtext> = |V</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msup><mrow><mtext>|</mtext></mrow><mrow><mtext>2</mtext></mrow></msup><msub><mrow><mtext> + |V</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msup><mrow><mtext>|</mtext></mrow><mrow><mtext>2</mtext></mrow></msup></mrow></math><img file="EP0638457B1_D0003.tif" /></maths>

0007In general, because the primary resistance R of the PM motor is much smaller than the inductance, it is possible to ignore the resistance R. When the primary resistance R is ignored, the equation (2) is expressed as shown below. FIG. 21 shows a vector diagram of the PM motor drawn in accordance with the equation (4).<maths id="math0004" num="(4)"><math display="block"><mrow><mtable><mtr><mtd><mrow><mtable><mtr><mtd><mrow><msub><mrow><mtext>V</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext> = jωL</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>q</mtext></mrow></msub></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext>V</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext> = jωL</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext> + ωE</mtext></mrow><mrow><mtext>0</mtext></mrow></msub></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext>V = ωE</mtext></mrow><mrow><mtext>0</mtext></mrow></msub><msub><mrow><mtext> + jωL</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msub><mrow><mtext> + jωL</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>d</mtext></mrow></msub></mrow></mtd></mtr></mtable></mrow></mtd></mtr></mtable></mrow></math><img file="EP0638457B1_D0004.tif" /></maths>

0008To control the PM motor having the above characteristics, it is preferable to control the vector of the motor current I. For example, it is preferable to decompose the motor current I into the field current component I_d and the torque current component I_q and control the absolute value of each component. To control the vector of the motor current I of the PM motor, the controller receives a command for a vector to be output from the PM motor, that is, a reference torque T^* from a supervisor controller. The controller receiving the reference torque T^* determines the field current I_d and the torque current I_q which are control targets so that the reference torque T^* is realized. In this case, it is not necessary to excite the PM motor with the field current I_d, because permanent magnets generate the main magnetic flux. Therefore, the field current I_d is normally controlled to 0 or some other specified value. When the field current I_d is kept constant, it is possible to control the torque T with the torque current I_q as shown by the equation (1). Therefore, the reference torque T* is exclusively used to determine the torque current I_q. The controller outputs the determined field current I_d and torque current I_q to a motor controller as a reference field current I_d* and a reference torque current I_q*, respectively. The motor controller controls a power converter such as an inverter so that the field current I_d and the torque current I_q corresponding to the reference field current I_d* and the reference torque current I_q*, respectively, flow through the PM motor. Though the motor current I can be decomposed into the vector components of the field current I_d and the torque current I_q as shown above, it is also possible to decompose the motor current I into an absolute value component |I| and a phase component argI.

0009To control the vector of the current I of the PM motor, it is preferable to perform the field weakening control at the same time. The field weakening control is defined as the control for weakening a field flux when the PM motor rotates at a high speed.

0010As shown in the equation (4) and FIG. 21, when the rotational speed N of the PM motor, namely, the axial angular velocity ω of the PM motor increases, the counter electromotive force ωE₀ increases in proportion to the increase of the axial angular velocity ω and the terminal voltage V of the PM motor increases with the increase of the counter electromotive force ωE₀. When the terminal voltage V of the PM motor exceeds a value corresponding to the battery voltage V_dc, a voltage corresponding to the difference between the terminal voltage V and the battery voltage V_dc is applied to an inverter provided between the PM motor or the power source. This voltage causes the inverter or the like to be damaged. Hereafter, the rotational speed N of the PM motor when the terminal voltage V equals a value corresponding to the battery voltage V_dc is referred to as a base rotational speed N_B. The region of rotational speed equal to or higher than the base rotational speed N_B is referred to as a high rotation region, the region of rotational speed equal to or lower than the base rotational speed N_B is referred to as a low rotation region, and the region of rotational speed close to the base rotational speed N_B is referred to as a medium rotation region.

0011In the PM motor, a field flux is mainly generated by permanent magnets. However, when the vector of the motor current I is controlled as described above, it is possible to generate a field flux having a necessary intensity by utilizing the field current I_d, and also to generate a field flux with an intensity which partially cancels the field flux (main magnetic flux) E₀ generated by the permanent magnets. The field weakening control can specifically be executed as the control for generating a field flux with an intensity which partially cancels the main magnetic flux E₀, by controlling the field current I_d to a negative value. When this type of control is performed, the absolute value of the terminal voltage V decreases as shown by the equation (4) and FIG. 21. Therefore, when performing the field weakening control by aiming at the field current I_d with a proper intensity, it is possible to control the terminal voltage V in the high rotation region to the battery voltage V_dc or lower. The field current I_d when the field weakening control is executed is referred to as a field weakening current.

0012To execute the field weakening control, the value of the field weakening current I_d is normally determined in accordance with the rotational speed N of the PM motor and the requested output torque (= reference torque T*) by assuming that the battery voltage V_dc is constant. For example, as shown in FIG. 22, the field current I_d is controlled to 0 in the low rotation region. As the rotational speed N increases exceeding the base rotational speed N_B, the effective amplitude of the field weakening current I_d is varied from -30 A to -50 A, to -100 A,...in that order.

0013The field weakening control is disclosed in "Variable Speed Drive System of Permanent Magnet Synchronous Motors with Flux-weakening Control", Keita HATANAKA et al., at the 1991 General Conference No. 74 of the Industrial Application Division of The Institute of Electrical Engineers of Japan, pp 310-315.

0014However, because the actual battery voltage V_dc greatly depends on its state of charge (SOC) and load state, the field weakening current I_d may become too large or too small. This causes a problem that the efficiency lowers or the reference torque T* is not accurately realized.

0015In this case, a value obtained by converting the battery voltage V_dc, assumed to determine the field weakening current I_d, into a value comparable with the terminal voltage V is referred to as an allowable power-source voltage V_B. When the allowable power-source voltage V_B is higher than the actual battery voltage V_dc, the field weakening current I_d with an absolute value excessively larger than reference torque T^* flows. In other words, the absolute value of the field current I_d increases, which is a current component not contributing to generation of torque. This means that the efficiency lowers. Conversely, when the allowable power-source voltage V_B is lower than the actual battery voltage V_dc, it is impossible to provide sufficient field weakening current I_d to control the terminal voltage V to the allowable power-source voltage V_B or lower, or sufficient torque current I_q to realize the reference torque T*.

0016Document "IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS AND CONTROL", vol. 35, no. 4, November 1988, New York US, pages 544 - 559, B.K. BOSE ET ALK "A Microcomputer-Based Control and Simulation of an advanced IPM Synchronous Machine Drive for Electric Propulsion" discloses a simulation system of a digitally controlled IPM synchronous machine which constitutes part of an electric vehicle drive system, wherein the synchronous machine (synchronous motor) is provided with permanent magnets as its field system and is supplied with a necessary drive current by a power source in the form of a battery. A control means (inverter) controllably converts the d.c. voltage of the battery into a variable-frequency variable-voltage alternating current as a power supply for the synchronous machine, whereby the drive system is operable in all four quadrants and regenerative braking is performed by returning the generated energy back to the battery for recharging the battery. When the synchronous machine reaches higher speeds causing a dominant armature reaction effect, the drive system enters into a field weakening control mode thereby defining a field weakening current component for partially cancelling the field flux provided by the permanent magnets. This field weakening control is provided on the basis of the monitored rotational speed of the synchronous machine and further predetermined operating conditions. Furthermore, reference values of control parameters are determined by simulation process (calculations and estimations), which enables the control of the synchronous machine by feedback signal processing. However, when carrying out the field weakening control in the above driving system, the varying battery voltage is not taken into consideration.

0017It is the first object of the present invention to make it possible to control the field weakening current I_d in accordance with a change of the actual battery voltage V_dc and to execute optimum field weakening control.

0018It is the second object of the present invention to secure the efficiency and favorably achieve the reference torque T*.

0019It is the third object of the present invention to secure the efficiency and achieve the reference torque T* while considering the operation of an inverter, when one is used.

0020It is the fourth object of the present invention to secure the accelerating performance from standstill and to simultaneously prevent the efficiency from lowering due to decrease of the battery voltage V_dc caused by degradation of SOC (relative remaining capacity) of the battery.

0021It is the fifth object of the present invention to improve the drive feeling of an electric vehicle.

0022It is the sixth object of the present invention to allow a user to select the drive in consideration of efficiency in consideration of output, and thus, it is possible to flexibly fulfill the user's request.

0023It is the seventh object of the present invention to favorably secure the output of a loaded motor.

0024It is the eighth object of the present invention to realize an efficient control performance.

0025It is the ninth object of the present invention to minimize the storage capacity necessary for control.

0026According to the present invention the above objects are accomplished by a driving and controlling apparatus as claimed in the appended independent claim 1, and in particular, according to a first embodiment of the present invention by a driving and controlling apparatus for use in a drive unit having a synchronous motor using permanent magnets as its field system which can be excited by a field flux generated by a field current component in a primary current; a battery as a source of power; and power conversion means for converting power discharged from the battery into alternating current and feeding the alternating current as the primary current to the synchronous motor; said apparatus comprising: <ul id="ul0001" list-style="none" compact="compact"><li>current control means for controlling the primary current of the synchronous motor in accordance with a reference current; and</li><li>current condition determination means for carrying out a field weakening being characterized in that</li><li>said current condition determination means is provided for minimizing the reference current in accordance with a voltage of the battery under predetermined conditions and at least partially cancelling the field flux generated by the permanent magnet with the field flux generated by the field current component so that a counter electromotive force of the synchronous motor does not exceed the voltage of the battery and so that the field flux generated by the field current component varies as the voltage of the battery varies.</li></ul>

0027Further embodiments of the present invention are recited in the appended subclaims. <ul id="ul0002" list-style="none" compact="compact"><li>FIG. 1 is a block diagram showing the system constitution of an electric vehicle to which each embodiment of the present invention is suitable;</li><li>FIG. 2 is a flow chart showing the operation of a current condition computing section in the first embodiment of the present invention;</li><li>FIG. 3 is an illustration showing the relationship between field weakening current, terminal voltage, and efficiency converted into values per phase;</li><li>FIG. 3A is an illustration showing the relationship between field weakening current, terminal voltage, maximum motor-applied voltage and motor rotational speed;</li><li>FIG. 4 is an illustration conceptually showing the field current limiting method in the first embodiment, in which symbol N_max represents the maximum rotational speed of a PM motor;</li><li>FIG. 5 is an illustration conceptually showing the output limiting method in the first embodiment;</li><li>FIG. 6 is an illustration conceptually showing an output limiting method according to the prior art;</li><li>FIG. 7 is a timing chart showing a battery-voltage fluctuation due to loaded operation of a motor;</li><li>FIG. 8 is an illustration showing the relation between motor rotational speed, reference torque value and battery voltage, and optimum field current component for maximizing the motor efficiency;</li><li>FIG. 9 is an illustration showing the relation between motor rotational speed, reference torque value and battery voltage, and optimum torque current component for optimizing the motor efficiency about a salient-pole motor;</li><li>FIG. 10 is an illustration showing the linear equation approximation of the relationship between optimum field current component and battery voltage;</li><li>FIG. 11 is an illustration showing the linear equation approximation of the relationship between optimum torque current component and battery voltage;</li><li>FIG. 12 is a flow chart showing the entire operation of the current condition computing section in the second embodiment of the present invention;</li><li>FIG. 13 is a flow chart showing the reference current value correcting operation of the current condition computing section in the second embodiment;</li><li>FIG. 14 is a timing chart showing the control timing in the second embodiment;</li><li>FIG. 15 is a flow chart showing the reference minimum voltage determining operation in the third embodiment of the present invention;</li><li>FIG. 16 is an illustration showing the method for constructing a reference minimum voltage determining table in the third embodiment of the present invention;</li><li>FIG. 17 is a flow chart showing the reference current value correcting operation of the current condition computing section in the fourth embodiment of the present invention;</li><li>FIG. 18 is an illustration showing the interpolating operation executed in the fourth embodiment;</li><li>FIG. 19 is an illustration showing the relationship between V_dc1 and V_dc2 in FIG. 18;</li><li>FIG. 20 is a circuit diagram showing an equivalent circuit per phase of a PM motor;</li><li>FIG. 21 is a vector diagram of a PM motor; and</li><li>FIG. 22 is an illustration showing the relationship between motor rotational speed and torque when field weakening control is performed.</li></ul>

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

0028Preferred embodiments of the present invention are described below by referring to the accompanying drawings.

(a) System constitution

0029FIG. 1 shows a system environment suitable for executing each embodiment of the present invention. The system shown in FIG. 1 is an electric vehicle drive system and its controller. In the system shown in FIG. 1, the electric discharge output of a battery 10 is converted from DC to 3-phase AC by an inverter 12 and fed to a PM motor 14.

0030The power conversion operation in the inverter 12 is vector-controlled by a current condition computing section 15 and a current control section 18. The current condition computing section 16 determines a field reference current I_d* and reference torque current I_q* in accordance with a reference torque T* having a value corresponding to an amount of depression of an accelerator or brake pedal by a driver given from a supervisor control section (vehicle controller, regeneration controller, or the like), and feeds the reference currents to the current control section 18.

0031The current control section 18 inputs not only these reference currents I_d* and I_q* but also the angular position θ of the rotor of the motor 14 detected by a motor rotational-position sensor 20 affixed to the motor 14. The current control section 18 performs the rotational transform of the current vector coordinate in correspondence to the position θ, in order to realize the alternation of current, and simultaneously generates U, V, and W phase reference currents I_U*, I_V*, and I_W* in accordance with the reference currents I_d* and I_q* to feed them to the inverter 12.

0032The inverter 12 generates, for example, PWM (pulse width modulation) signals in accordance with the fed reference currents I_U*, I_V* and I_W*. The inverter 12 comprises, for example, a plurality of switching devices such as an IGBTs (insulated gate bipolar transistors). These switching devices are switched by the PWM signals generated by conversion so that currents I_U, I_V, and I_W with values corresponding to the reference currents I_U*, I_V*, and I_W* flow through the U, V, and W phases of the motor 14. When this type of switching is performed, the field current I_d* and torque current I_q* having values which correspond to the reference currents I_d* and I_q* flow through the U, V, and W phases of the motor 14 and these currents generate the alternating magnetic field. The torque T outputted from the motor 14 as the result of the above current control comes to a value corresponding to the reference torque T*.

0033The current condition computing section 16 normally controls the reference field current I_d* to a constant value, for example, 0. Therefore, the field flux in the motor 14 is kept constant. The current condition computing section 16 changes the reference torque current I_q* in accordance with a change of the reference torque T*. Thus, since the time constant for responding to the torque current component I_q* is small, it is possible to rapidly change the output torque of the motor 14.

0034Moreover, the current condition computing section 16 executes the field weakening control. Therefore, the current condition computing section 16 detects the rotational speed N of the motor 14 with the motor rotational-position sensor 20.

0035In the present invention, the value of the field weakening reference current I_d* is changed in accordance with the battery voltage V_dc. Thus, it is possible to favorably execute the field weakening control in accordance with a change of the battery voltage V_dc. In the system shown in FIG. 1, a voltage sensor 22 is affixed to the battery 10 to realize the above control, and the voltage V_dc between both terminals of the battery 10 is detected by the voltage sensor 22. It is also possible to estimate the battery voltage V_dc in accordance with an SOC of the battery 10 or a vehicle state such as a road load. In FIG. 1, a member 24 set to the battery 10 is an SOC sensor for detecting an SOC of the battery 10.

0036The present invention is not restricted to the system constitution of FIG. 1.

(b) First embodiment

0037FIG. 2 shows the flow of operation of the current condition computing section 16 in the first embodiment of the present invention.

0038The current condition computing section 16 receives the reference torque T* from a not-illustrated supervisor controller and the motor rotational speed N from the motor rotational-position sensor 20, respectively (step 100). The current condition computing section 16 compares the inputted motor rotational speed N with the base rotational speed N_B of the PM motor 14 (102).

0039When the rotational speed N is lower than the base rotational speed N_B, the current condition computing section 16 computes the torque current I_q by substituting "<maths id="math0005" num=""><math display="inline"><mrow><mtext>ω=2πN</mtext></mrow></math><img file="EP0638457B1_D0005.tif" /></maths>", "I_d=0" and "<maths id="math0006" num=""><math display="inline"><mrow><mtext>T=T*</mtext></mrow></math><img file="EP0638457B1_D0006.tif" /></maths>" for the equation (1) (step 104) and outputs the obtained torque current I_q as the reference torque current I_q^* (step 106). Because the main magnetic flux E₀ is already known, the above computation is possible. In this case, the current condition computing section 16 outputs 0 or a constant value as the field reference current I_d*. Normal field control is performed in the above manner.

0040When the rotational speed N is equal to or higher than the base rotational speed N_B, the current condition computing section 16 performs the field weakening control. This embodiment is characterized by the improved field weakening control.

0041For the field weakening control, the current condition computing section 16 first computes the maximum motor-applied voltage V_MAX in accordance with the battery voltage V_dc detected by the voltage sensor 22 (108). In more detail, the current condition computing section 16 converts the present battery voltage V_dc from an input-side value to an output-side value, of the inverter 12. An obtained converted value is used as the maximum motor-applied voltage V_MAX. In this case, the current condition computing section 16 assumes the PWM modulation factor of the inverter 12 to be 100%. Therefore, the maximum motor-applied voltage V_MAX comes to a value representing the highest inverter output voltage obtained from the present battery voltage V_dc. Particularly, when a triangular wave is used as the carrier wave of a PWM signal in the inverter 12, the current control section 18 gives the reference currents I_U*, I_V*, and I_W* to the inverter 12 as sine-wave signals, and the inverter 12 generates the PWM signals by comparing these sine-wave signals with the carrier wave, respectively, the arithmetic equation evaluated in the step 108 is shown as follows:<maths id="math0007" num="(5)"><math display="block"><mrow><msub><mrow><mtext>V</mtext></mrow><mrow><mtext>MAX</mtext></mrow></msub><msup><mrow><mtext> = 3</mtext></mrow><mrow><mtext>1/2</mtext></mrow></msup><msub><mrow><mtext>V</mtext></mrow><mrow><mtext>B</mtext></mrow></msub><msup><mrow><mtext>/(2∗2</mtext></mrow><mrow><mtext>1/2</mtext></mrow></msup><mtext>)</mtext></mrow></math><img file="EP0638457B1_D0007.tif" /></maths>

0042Needless to say, this is just one example of PWM signal generating methods. In the present invention, it is also possible to use a wave other than the triangular wave as the carrier wave. In this case, although the above equation (5) cannot directly be used, a person skilled in the art can easily modify the equation (5) based on the disclosure of this application.

0043The current condition computing section 16 computes the field weakening current I_d and the torque current I_q by substituting "<maths id="math0008" num=""><math display="inline"><mrow><mtext>ω=2πN</mtext></mrow></math><img file="EP0638457B1_D0008.tif" /></maths>", "<maths id="math0009" num=""><math display="inline"><mrow><mtext>T=T*</mtext></mrow></math><img file="EP0638457B1_D0009.tif" /></maths>" and "<maths id="math0010" num=""><math display="inline"><mrow><msub><mrow><mtext>V=V</mtext></mrow><mrow><mtext>MAX</mtext></mrow></msub></mrow></math><img file="EP0638457B1_D0010.tif" /></maths>" for the equations (1) to (3) (110). That is, because L_d, L_q, R, and E_o are already known, it is possible to solve these equations for I_d and I_q by applying ω, T and V to these equations (1) to (3). Because the term of I_dI_q is included in the equation (1), the operation is complicated. To avoid this problem, it is recommended to prepare and store a map for obtaining I_d and I_q corresponding to N, T and V by referring to N, T and V as keys. Thus, it is possible to execute the step 110 simply by referring to the map. It is also possible to use the equation (4) instead of the equation (2).

0044Then, the current condition computing section 16 judges whether a mode command given from an operator through a supervisor controller indicates selection of an output mode or an efficiency mode (112). When the section 16 judges that the mode command indicates "output mode", it executes the step 118 by skipping the steps 114 and 116. When the section 16 judges that the mode command indicates "efficiency mode", it executes the step 118, after executing the steps 114 and 116. In the step 114, the current condition computing section 16 judges whether an SOC detected by the SOC sensor 24 is the predetermined value SOC₀ or less, that is, whether the SOC has degraded. When the section 16 judges that the SOC has not degraded, it skips the step 116. When the section 16 judges that the SOC has degraded, it limits the value of the field weakening current I_d computed in the step 110 to a predetermined value I_d0 (116).

0045The field current I_d and torque current I_q thus obtained are outputted to the inverter 12 as the field weakening reference current I_d* and the reference torque current I_q* (118). The field weakening control in this embodiment is performed in the above manner.

0046For this embodiment, as described above, the actual battery voltage V_dc is considered to determine the field weakening reference current I_d* and the reference torque current I_q*. For example, when the battery voltage V_dc fluctuates due to a change of the SOC of the battery 10 or depending on a load of the PM motor 14, the maximum motor-applied voltage V_MAX changes in correspondence with the fluctuation of the battery voltage V_dc. Thereby, the field weakening reference current I_d* and the reference torque current I_q* computed in accordance with the maximum motor-applied voltage V_MAX are also changed. This embodiment realizes the optimum field weakening control for preventing too-large or too-small a field weakening current I_d from flowing, due to the above control.

0047As shown in FIG. 3, by increasing the absolute value of the field weakening current I_d, the terminal voltage V is lowered and also the efficiency η decreases. That is, when the field weakening current I_d increases, the primary current <maths id="math0011" num=""><math display="inline"><mrow><msub><mrow><mtext>I=(I</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msup><mrow><mtext>*</mtext></mrow><mrow><mtext>2</mtext></mrow></msup><msub><mrow><mtext>+I</mtext></mrow><mrow><mtext>q</mtext></mrow></msub><msup><mrow><mtext>*</mtext></mrow><mrow><mtext>2</mtext></mrow></msup><msup><mrow><mtext>)</mtext></mrow><mrow><mtext>1/2</mtext></mrow></msup></mrow></math><img file="EP0638457B1_D0011.tif" /></maths> of the PM motor 14 increases, the copper loss increases due to the increase of the current I, and resultingly the efficiency η decreases. Moreover, it is found from FIG. 3 that the efficiency η is maximized at the terminal voltage V equal to or lower than the maximum motor-applied voltage V_MAX when the field weakening current I_d at <maths id="math0012" num=""><math display="inline"><mrow><msub><mrow><mtext>V=V</mtext></mrow><mrow><mtext>MAX</mtext></mrow></msub></mrow></math><img file="EP0638457B1_D0012.tif" /></maths> is used as the field weakening reference current I_d*. Therefore, when the maximum motor-applied voltage V_MAX is computed in accordance with the battery voltage V_dc, and the reference field current I_d* is determined by using the V_MAX, deterioration of the efficiency η due to excessive flow of the field weakening current I_d or output deficiency due to insufficient field weakening current I_d does not occur.

0048In more detail, the reason why the reference field current I_d* in this embodiment should be set in response to the SOC and therefore to the maximum motor-applied voltage V_MAX is presented in the relationship shown in FIG. 3A. As is shown in this figure, the terminal voltage V decreases as the absolute value of the field weakening current I_d increases, and it also decreases as the revolution speed N of the motor 14 decreases. The maximum motor-applied voltage V_MAX decreases as the SOC of the battery 10 decreases.

0049As has been described hereinbefore, since the situation V>V_MAX must not occur, the field weakening current I_d must be controlled so that the terminal voltage V is equal to or less than V_MAX. The filed weakening current I_d becomes I_dmin when <maths id="math0013" num=""><math display="inline"><mrow><msub><mrow><mtext>V=V</mtext></mrow><mrow><mtext>MAX</mtext></mrow></msub></mrow></math><img file="EP0638457B1_D0013.tif" /></maths>. I_dmin is determined depending on the revolution speed N and the SOC (or V_MAX), as shown in FIG. 3A. In this embodiment, the reference field current I_d* is set in response to the maximum motor-applied voltage V_MAX, on the basis of the above described relationship. In other words, the reference field current I_d* is controlled so that it is set larger when the SOC or the maximum motor-applied voltage V_MAX decreases, because the absolute value of I_dmin becomes large. Thus, by using such an I_d* control adaptive to the SOC, the minimum weakening field current I_dmin is obtained, which results in the improved efficiency.

0050In this embodiment, operations of the current condition computing section 16 are changed in accordance with a mode command. Therefore, driving stressing the output power or driving stressing the efficiency can be carried out, depending on the preference of a vehicle driver.

0051A vehicle driver can limit the maximum value of the absolute value of the field weakening reference current I_d* to -I_d0 by giving, for example, a mode command for selecting the efficiency mode to the current condition computing section 16. That is, when the battery voltage V_dc lowers in accordance with the degradation of an SOC, the maximum motor-applied voltage V_MAX computed in the step 108 lowers and, as shown in FIG. 4, the absolute value of the field weakening current I_d increases in accordance with the decrease of the voltage V_MAX. If the field weakening current I_d with an increased absolute value is directly outputted as the field weakening reference current I_d*, the efficiency decreases because a current component (I_d) that is not contributing to generation of torque, increases. In this embodiment, the value of the field weakening reference current I_d* is limited to the predetermined value I_d0 in accordance with the degradation of an SOC when the efficiency mode is selected. Therefore, a vehicle driver, particularly a driver stressing the efficiency, can execute driving with the emphasis on efficiency, in accordance with his preference.

0052A vehicle driver can obtain a PM-motor output not limited by I_d0 in the high rotation region by issuing a mode command for selecting the output mode to the current condition computing section 16 independently of degradation of an SOC. That is, when the above field current limitation is executed, the output range of torque T is narrowed to the low-torque side and as a result output limitation occurs since the motor output is determined by TN. Therefore, this embodiment uses the output mode for omitting the steps 116 and 118 so that the output torque T can be secured.

0053Moreover, the output limitation resulting from by the field current limitation does not cause the problem seen in, for example, Japanese Patent Laid-Open No. Hei 5 - 38003. In this official gazette, the output torque T of a PM motor is limited in accordance with degradation of an SOC as shown in FIG. 6. In FIG. 6, the maximum output torque in the high rotation region is changed from an isopower line with a comparatively high power (line where the PM-motor output power obtained as a product of the output torque T and rotational speed N is constant) to an isopower line with less power in accordance with the degradation of an SOC. In the case of this output limitation, the output torque T is not limited even in the high rotation region as long as the reference torque T* is low. In this case, however, a loss occurs because field weakening control is performed. Moreover, output is limited even in the medium rotation region in which a high efficiency is to be obtained. In this embodiment, the output limitation due to the field current limitation executed in accordance with degradation of an SOC functions even in the high-rotation low-torque region as shown in FIG. 5. Therefore, the efficiency is prevented from being lowered in spite of the increase of the absolute value of the field current I_d caused by degradation of the SOC, and moreover the accelerating performance from standstill is secured and the drive feeling is favorably secured. Moreover, because the output limitation is not performed in the medium rotation region, a high efficiency is also obtained from this point of view.

0054In the above description, the only case which is shown is one in which the current vector I of the motor 14 is decomposed into the field current I_d and the torque current I_q to perform vector control. However, it is also possible to perform vector control by decomposing the current I into the absolute value |I| and the phase argI of the motor current I. This is also applicable to the other embodiments.

(c) Principle of second to fourth embodiments

0055In the first embodiment, the field weakening reference current I_d* is determined in accordance with the battery voltage V_dc. However, the battery voltage V_dc is also caused to fluctuate by applying a load to the PM motor 14 which is a driving object. That is, even if the vector of the motor current I is controlled in accordance with the battery voltage V_dc detected at the time t=0, the battery voltage V_dc is caused to fluctuate due to the load of the PM motor 14 and frequently converges to a value V_s lower than the value at the time t=0 as shown in FIG. 7. Therefore, when control is performed in accordance with the battery voltage V_dc at the time t=0 so that the best efficiency is obtained, necessary current may not be obtained from the battery 10 and thereby output may be insufficient for the reference torque T*.

0056The second to fourth embodiments described below settle the fluctuation of the battery voltage V_dc due to the load of the PM motor 14. That is, the embodiments make it possible to operate the PM motor 14 under conditions for maximizing the efficiency by using the approximation set out below as a main principle.

0057FIGs. 8 and 9 show the field current I_d and torque current I_q for maximizing the efficiency of the PM motor 14, that is, the optimum field current I_dp and the optimum torque current I_qp. Particularly, FIG. 9 shows a case in which a salient-pole motor is used as the PM motor 14. As shown in FIGs. 8 and 9, the optimum field current I_dp and the optimum torque current component I_qp, when using a salient-pole motor as the PM motor 14, depend on the motor rotational speed N, reference torque T*, and battery voltage V_dc. The field current I_d is negative. Moreover, attention should be paid to the fact that for simplification only, T* is drawn up to 100% in the case of N=0 and <maths id="math0014" num=""><math display="inline"><mrow><msub><mrow><mtext>N=N</mtext></mrow><mrow><mtext>1</mtext></mrow></msub></mrow></math><img file="EP0638457B1_D0014.tif" /></maths> (>0) and up to 40% in the case of <maths id="math0015" num=""><math display="inline"><mrow><msub><mrow><mtext>N=N</mtext></mrow><mrow><mtext>2</mtext></mrow></msub></mrow></math><img file="EP0638457B1_D0015.tif" /></maths> (>N₁).

0058As shown in FIG. 8, when the motor rotational speed N equals 0, the value of the optimum field current I_dp is constant, independent of the battery voltage V_dc but it varies in accordance with the value of the reference torque T*. When the motor rotational speed N increases and reaches N₁, and moreover reaches N₂, the value of the optimum field current I_dp changes in a region where the battery voltage V_dc is low. That is, the characteristic of the optimum field current I_dp to the battery voltage V_dc is inclined. This is because the counter electromotive force of the motor 14 exceeds the battery voltage V_dc if the normal field control is performed and therefore the above described field weakening control is executed to prevent the force from exceeding the voltage.

0059Moreover, as shown in FIG. 9, when the motor rotational speed N is 0, the optimum torque current component I_qp is constant independently of the battery voltage V_dc. However, when the motor rotational speed N reaches N₁, and moreover reaches N₂, the component I_qp is inclined in a region where the battery voltage V_dc is low. This is because the motor 14 is a salient-pole motor and thereby generates reluctance torque, that is, torque depending on the field current I_d {see the second term of the right side of the equation (1)} and therefore it is necessary to decrease the torque current I_q when increasing the absolute value of the field current I_d in a region where the battery voltage V_dc is low in order to maintain an amplitude of the primary current I and to perform field weakening control.

0060The relationships shown in FIGs. 8 and 9, that is, the relationship between the optimum field current I_dp, optimum torque current I_qp, and battery voltage V_dc, can be approximated as shown in FIGs. 10 and 11.

0061First, the optimum field current I_dp, as shown in FIG. 10, can be approximated by the straight line of inclination k_d (I_d1* in FIG. 10) in a region where the battery voltage V_dc is low and by the value of the optimum field current I_dp (I_d2* in FIG. 10) when the motor rotational speed N is 0 in a region where the battery voltage V_dc is high. Then, the optimum torque current I_qp, as shown in FIG. 11, can be approximated by the straight line of inclination k_q (I_q1* in FIG. 11) in a region where the battery voltage V_dc is low and by the value of the torque current component I_q (I_q2* in FIG. 11) when the motor rotational speed N is 0 in a region where the battery voltage V_dc is high. The inclination vector (k_d, k_q) has a value corresponding to the reference torque T* and motor rotational speed N. These approximate equations can be expressed as follows:<maths id="math0016" num="(6)"><math display="block"><mrow><mtable><mtr><mtd><mrow><mtable><mtr><mtd><mrow><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>dp</mtext></mrow></msub><msub><mrow><mtext> = I</mtext></mrow><mrow><mtext>d1</mtext></mrow></msub><msub><mrow><mtext>* for ΔV</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><mtext> &lt; 0</mtext><mspace linebreak="newline" /><msub><mrow><mtext> = I</mtext></mrow><mrow><mtext>d2</mtext></mrow></msub><msub><mrow><mtext>* for ΔV</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><mtext> ≥ 0</mtext></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>qp</mtext></mrow></msub><msub><mrow><mtext> = I</mtext></mrow><mrow><mtext>q1</mtext></mrow></msub><msub><mrow><mtext>* for ΔV</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><mtext> &lt; 0</mtext><mspace linebreak="newline" /><msub><mrow><mtext> = I</mtext></mrow><mrow><mtext>q2</mtext></mrow></msub><msub><mrow><mtext>* for ΔV</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><mtext> ≥ 0</mtext></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>d1</mtext></mrow></msub><msub><mrow><mtext>* = I</mtext></mrow><mrow><mtext>dmin</mtext></mrow></msub><msub><mrow><mtext> + k</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext>∗ΔV</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext>I</mtext></mrow><mrow><mtext>q1</mtext></mrow></msub><msub><mrow><mtext>* = I</mtext></mrow><mrow><mtext>qmin</mtext></mrow></msub><msub><mrow><mtext> + k</mtext></mrow><mrow><mtext>d</mtext></mrow></msub><msub><mrow><mtext>∗ΔV</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub></mrow></mtd></mtr><mtr><mtd><mrow><msub><mrow><mtext>ΔV</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><msub><mrow><mtext> = V</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><msub><mrow><mtext> - V</mtext></mrow><mrow><mtext>dcmin</mtext></mrow></msub><mspace linebreak="newline" /><msub><mrow><mtext> V</mtext></mrow><mrow><mtext>dcmin</mtext></mrow></msub><msub><mrow><mtext>: A minimum value of V</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><mtext> which can be outputted from the battery 10 (Reference minimum voltage)</mtext><mspace linebreak="newline" /><msub><mrow><mtext> I</mtext></mrow><mrow><mtext>dmin</mtext></mrow></msub><msub><mrow><mtext>: I</mtext></mrow><mrow><mtext>dp</mtext></mrow></msub><msub><mrow><mtext> for V</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><msub><mrow><mtext> = V</mtext></mrow><mrow><mtext>dcmin</mtext></mrow></msub><mspace linebreak="newline" /><msub><mrow><mtext> I</mtext></mrow><mrow><mtext>qmin</mtext></mrow></msub><msub><mrow><mtext>: I</mtext></mrow><mrow><mtext>qp</mtext></mrow></msub><msub><mrow><mtext> for V</mtext></mrow><mrow><mtext>dc</mtext></mrow></msub><msub><mrow><mtext> = V</mtext></mrow><mrow><mtext>dcmin</mtext></mrow></msub><mtext></mtext></mrow></mtd></mtr></mtable></mrow></mtd></mtr></mtable></mrow></math><img file="EP0638457B1_D0016.tif" /></maths>

(d) Second embodiment

0062FIG. 12 shows an overall flow of operation of the current condition computing section 16 in the second embodiment of the present invention. This flow is executed at a predetermined frequency.

0063In this embodiment, the current condition computing section 16 first receives the reference torque T* from the supervisor controller (200) and detects the motor rotational speed N with the motor rotational-position sensor 20 (202).

0064Thereafter, the current condition computing section 16 sets the minimum value of the battery voltage V_dc which can be outputted from the battery 10 to the reference minimum voltage V_dcmin (204). The current condition computing section 16 obtains the values of the optimum field current I_dp and the optimum torque current I_qp when the battery voltage V_dc equals the reference minimum voltage V_dcmin, that is, the intercept vector (I_dmin, I_qmin) of the equation (6) in accordance with the set reference minimum voltage V_dcmin. In this case, the current condition computing section 16 refers to a previously prepared table 300. This table makes the motor rotational speed N and the reference torque T* correspond to the intercept (I_dmin, I_qmin). This table can be prepared in accordance with the relationships shown in FIGs. 8 and 9.

0065Thereafter, the current condition computing section 16 judges whether the reference torque T* and the motor rotational speed N change by as much as a predetermined value or more (208). That is, the section 16 computes a change ΔT* in reference torque T* and a change ΔN in motor rotational speed N between the reference torque T* and the motor rotational speed N respectively, computes this time and the previous time, and judges whether the obtained changes ΔT* and ΔN exceed the predetermined values ΔT₀ and ΔN₀ respectively. As the result of the judgment, when either of the inequalities ΔT*>ΔT₀ and ΔN>ΔN₀ is effected, the current condition computing section 16 commences step 210 by assuming that the load of the motor 14 increases, while it commences the step 212 for cases other than the above.

0066In the step 210, the current condition computing section 16 outputs the intercept (I_dmin, I_qmin) computed in the step 206 to the current control section 18 as the reference current (I_d*, I_q*). Thereby, the vector of the current I of the motor 14 is controlled in accordance with the intercept (I_dmin, I_qmin), and in turn, the reference minimum voltage V_dcmin of the battery voltage V_dc. Under the above state, the efficiency of the motor 14 is not maximized. However, the output power of the motor 14 obtained by using the intercept (I_dmin, I_qmin) as the reference current (I_d*, I_q*) comes to a value equivalent to the value obtained by multiplying the reference torque T* by the motor rotational speed N. Therefore, even if the battery voltage V_dc fluctuates due to the load of the motor 14 as shown in FIG. 7, at least the requested output power of the motor 14 can be secured.

0067Thereafter, the current condition computing section 16 waits until the battery voltage V_dc is converged (214, 216). Once the battery voltage V_dc is sufficiently converged, the current condition computing section 16 corrects the values of the reference current (I_d*, I_q*) in accordance with a converged battery voltage V_s (212) and feeds the corrected reference current (I_d*, I_q*) to the current control section 18 (212). Thereby, the motor 14 is brought into a state in which the maximum efficiency is obtained.

0068FIG. 13 shows the details of the step 212, that is, the details of correction of the reference current (I_d*^, I_q*) in this embodiment.

0069In this embodiment, the current condition computing section 16 first computes the inclination (k_d, k_q) to correct the reference current (I_d*, I_q*) (220). In this case, the current condition computing section 16 uses a table 302 for making the reference torque T* and the motor rotational speed N correspond to the inclination (k_d, k_q). This table can be prepared in accordance with the relationships shown in FIGs. 8 and 9. The current condition computing section 16 substitutes the obtained inclination (k_d, k_q) and the intercept (I_dmin, I_qmin) for the equation (6) and also substitutes the converged battery voltage V_s for the V_dc of the equation (6) to compute an approximate current vector (I_d*₁, I_q*₁) (222). The approximate current (I_d*₁, I_q*₁) thus obtained can be used as the reference current (I_d*, I_q*) when the battery voltage V_dc is low.

0070Thereafter, the current condition computing section 16 reads the vector of the motor current I for realizing the maximum efficiency when the motor rotational speed N is 0, that is, the approximate current vector (I_d*₂, I_q*₂) shown in FIGs. 10 and 11, from a built-in memory (224). The approximate current (I_d*₂, I_q*₂) read in at step 224 can be used as the reference current (I_d*, I_q*) in a region where the battery voltage V_dc is high as shown in FIGs. 10 and 11.

0071In the steps 226 and 228, the current condition computing section 16 compares the approximate current (I_d*₁, I_q*₁) obtained in the step 222 with the approximate current (I_d*₂, I_q*₂) read in the step 224.

0072In step 226, the judgement relating to the field current is made. In the case of "I_d*₁ &lt; I_d*₂", it can be considered that the present battery voltage or the converged battery voltage V_s is included in the left region in FIG. 10. That is, it can be considered that the converged battery voltage V_s is in a region where the value of the optimum field current I_dp depends on the value of the battery voltage V_dc. Therefore, the current condition computing section 16 sets the approximate field current I_d*₁ obtained in the step 222 to the reference field current I_d* to be outputted in the step 218 (230).

0073In the case of "I_d*₁ ≥ I_d*₂", however, it can be considered that the present battery voltage V_s is included in the right region in FIG. 10. That is, it can be considered that the converged battery voltage V_s is in the region where the value of the optimum field current I_dp does not depend on the value of the battery voltage V_dc. Therefore, the current condition computing section 16 sets the approximate field current I_d*₂ read in the step 224 to the reference field current I_d* to be outputted in the step 218 (232).

0074Moreover, the current condition computing section 16 judges in the step 228 whether I_q*₁ &lt; I_q*₂ or not. When the current condition computing section 16 judges that "I_q*₁ &lt; I_q*₂", it can be considered that the value of the optimum torque current component I_qp is included in the left region (the region where the value of the optimum torque current I_qp depends on the value of the battery voltage V_dc) in FIG. 11. Therefore, the current condition computing section 16 sets the approximate torque current I_q*₁ obtained in the step 222 to the reference torque current I_q* to be outputted in the step 218 (234). When the current condition computing section 16 judges that "I_q*₁ ≥ I_q*₂", however, it can be considered that the battery voltage V_s is included in the right region in FIG. 11, that is, the region where the value of the optimum torque current component I depends on the value of the battery voltage V_dc. In this case, the current condition computing section 16 sets the approximate torque current I_q*₂ read in the step 224 to the reference torque current I_q* to be outputted in the step 218 (236).

0075Thus, the reference current (I_d*, I_q*) in this embodiment is corrected.

0076Therefore, in this embodiment, because the reference torque T* and motor rotational speed N are inputted and detected and then the vector of the motor current I is initially controlled in accordance with the reference minimum voltage V_dcmin, the problem does not occur that necessary output power cannot be obtained from the motor 14 even when the battery voltage V_dc fluctuates due to the load of the motor 14.

0077Moreover, in this embodiment, it is possible to quickly control the efficiency of the motor 14 to the maximum value because the above initial control is performed and the battery voltage V_dc is converged and thereafter the vector of the motor current I is controlled in accordance with the converged battery voltage V_s.

0078Furthermore, the maximum efficiency control of the motor 14 is performed by approximating the relation between the optimum current (I_dp, I_qp) for maximizing the efficiency of the motor 14 and the battery voltage V_dc with the linear approximation equation in the equation (6) and correcting the initial reference current {= intercept (I_dmin, I_qmin)} with the linear approximation equation. Therefore, it is possible to realize the maximum efficiency control of the motor 14 without storing a lot of data showing the relationship between the reference torque value T*, motor rotational speed N, and battery voltage V_dc. That is, it is only necessary to store the inclination (k_d, k_q) of the linear approximation equation in a region where the battery voltage V_dc is low and the values of the approximate current (I_d*₂, I_q*₂) when the motor rotational speed N is 0 in a region where the battery voltage V_dc is high. Therefore, it is not necessary to use an expensive CPU having a large memory capacity in order to perform the maximum efficiency control of the motor 14.

0079FIG. 14 shows the control timing in this embodiment.

0080As shown in FIG. 14, the reference torque T* and motor rotational speed N are inputted and detected every predetermined sampling period. When the reference torque T* is inputted and the motor rotational speed N is also inputted, the processing ("initial computation" in FIG. 14) in the step 206 is performed. When the intercept (I_dmin, I_qmin) is obtained, the vector of the motor current I is controlled in accordance with the intercept. After this control is executed, the battery voltage V_dc changes due to the load of the motor 14. When the fluctuation of the value of the battery voltage V_dc gradually decreases and, for example, dV_dc/dt becomes smaller than a predetermined microvalue δ (see FIG. 7), the current condition computing section 16 judges that the battery voltage V_dc has converged to V_s and executes the processing ("corrective computation" in FIG. 14) in the step 212. After the resulting reference current (I_d*, I_q*) is outputted, the efficiency of the motor 14 is maximized. In this case, symbol T in FIG. 14 represents an actual value of the output torque (actual torque) of the motor 14, which becomes equal to the reference torque value T* when the initial computation is completed and (I_dmin, I_qmin) is outputted. In FIG. 14, the first-time initial computation is executed due to a change of the reference torque T* and the second-time initial computation is executed due to a change of the motor rotational speed N.

0081It is also possible to use a non-salient-pole motor as the motor 14. In this case, the optimum torque current I_qp does not decrease in the low rotation region shown in FIG. 9. Moreover, the approximation equation used in the step 212 is not restricted to a linear equation but it is possible to use, for example, a polynomial equation. That is, it is possible to use any equation as an approximation equation in the second embodiment as long as the equation passes through the intercept (I_dmin, I_qmin) and approximates the relationship between the optimum field current I_dp and the optimum torque current I_qp to the battery voltage V_dc. These points also apply to the following embodiments.

(e) Third embodiment

0082FIG. 15 shows a flow of control of the current condition computing section 16, particularly a flow of the processing in step 204. In this embodiment, the current condition computing section 16 detects an SOC of the battery 10 with the SOC sensor 24 (238) and also detects the battery voltage V_dc with the voltage sensor 22 (240). The current condition computing section 16 computes the value of an output reference power P* to the motor 14 by multiplying together the reference torque T* and the motor rotational speed N inputted and detected in the steps 200 and 202, respectively (242).

0083Then, the current condition computing section 16 estimates the converged battery voltage V_s in accordance with the obtained output reference power P* and the detected SOC (244). In this case, the current condition computing section 16 refers to a table 204 showing the relationship between the output P*, SOC, and battery voltage V_s. That is, the converged battery voltage V_s depends on the output P* and SOC as shown in FIG. 16 and this relationship is stored in the current condition computing section 16 as a table 304.

0084The current condition computing section 16 sets the reference minimum voltage V_dcmin in accordance with the estimated converged battery voltage V_s (246). That is, it sets V_dcmin by subtracting a predetermined value ΔV from V_s.

0085When the motor current (I_d, I_q) is controlled in accordance with the reference minimum voltage V_dcmin thus set (210), it is possible to secure the output of the motor 14 even though the battery voltage V_dc fluctuates due to a load as described above. Moreover, when the reference minimum voltage V_dcmin is set by considering an SOC of the battery 10 as is done in this embodiment, it is not necessary to excessively decrease the reference minimum voltage V_dcmin and it is possible to control the efficiency of the motor 14, before convergence of the battery voltage after execution of the step 110, to a reasonably high value, although the value is not maximum.

(f) Fourth embodiment

0086FIG. 10 shows a flow of the operation of the current condition computing section 16 in the fourth embodiment of the present invention, particularly a flow of the reference current correcting operation in step 212.

0087In this embodiment, the current condition computing section 16 computes the reference current (I_d*, I_q*) by using a table 306. The table 306 includes a plurality of unit tables for making the optimum current (I_dp, I_qp) correspond to the reference torque T* and motor rotational speed N. Each of these unit tables is made to correspond to each digitized value of the battery voltage Vdc. To compute the reference current (I_d*, I_q*) by using the table 306, the current condition computing section 16 first selects V_dc1 and V_dc2 which are the closest to the converged battery voltage V_s out of the battery voltages V_dc included in the table 306 (248). The current condition computing section 16 computes the reference currents (I_d*₁, I_q*₁,) and (I_d*₂ and I_q*₂) corresponding to the values V_dc1 and V_dc2 respectively by using not only the reference torque T* and the motor rotational speed N but also V_dc1 and V_dc2 as keys, and referring to the table 306 (250). The current condition computing section 16 computes the reference current (I_d*, I_q*) to be outputted in the step 218 through the interpolation of the obtained reference currents (I_d*₁, I_q*₁) and (I_d*₂, I_q*₂) (252).

0088Therefore, in this embodiment, the reference current (I_d*, I_q*) with the value shown in FIG. 18 is outputted. However, the inclination of (I_d, I_q) is that in the voltage region shown in FIG. 19.

(g) Supplement

0089As described above, the present invention makes it possible to obtain the optimum field weakening current I_d independently of the fluctuation of the SOC of a battery or that of the load of a PM motor because the field weakening current I_d is controlled in accordance with a change of the battery voltage V_dc. Therefore, it is possible to maintain the predetermined efficiency η without causing excessive field weakening current to flow when the battery voltage V_dc is comparatively high and to obtain the current I required to obtain the necessary torque T when the battery voltage V_dc is comparatively low.

0090Moreover, when the allowable power-source voltage V_B is computed in accordance with the battery voltage V_dc and the PWM modulation rate of an inverter, the above advantage can preferably be obtained in controlling the terminal voltage V of a PM motor through the PWM control.

0091Furthermore, by limiting the field weakening current I_d in accordance with the SOC of a battery when the field weakening current I_d is computed in accordance with the battery voltage V_dc, the efficiency η can be prevented from being lowered because a decrease of the battery voltage V_dc due to degradation of the SOC does not cause the field weakening current I_d to increase. Also in this case, an accelerating performance from standstill is secured because the torque T is secured in a low-rotation high-torque region. Therefore, by using the PM motor as the driving mechanism of an electric vehicle, it is possible to secure a preferable drive feeling.

0092Furthermore, when limitation of the field weakening current I_d is executed or stopped in accordance with, for example, a command from a user, the user can select drive in consideration of efficiency or drive in consideration of output. Thus, it is possible to flexibly satisfy the user's preference.

0093Furthermore, in the case where the vector of the motor current I is initially controlled on the assumption that the battery voltage V_dc is the reference minimum voltage V_dcmin and at the time of the convergence of the battery voltage V_dc after this vector control, the vector of the motor current I is controlled in accordance with the converged battery voltage V_s, the motor output power can be prevented from decreasing even if the battery voltage fluctuates due to load after vector control. Furthermore, because the motor efficiency η can be controlled to the maximum value after the battery voltage V_dc is converged, it is possible to obtain a driving and controlling apparatus having a high energy efficiency and suitable for mounting on an electric vehicle or the like.

0094Furthermore, when the reference minimum voltage V_dcmin is set in accordance with the SOC of a battery and the voltage V_dc, the reference minimum voltage V_dcmin can be prevented from being excessively lowered, and a comparatively high motor efficiency η can be realized, even before the battery voltage V_dc is converged.

0095Furthermore, when the relationship between the battery voltage V_dc and the optimum current (I_dp, I_qp) is approximated by a mathematical equation in a region where the optimum current for maximizing the motor efficiency η depends on the battery voltage V_dc to store the coefficient such as the inclination (k_d, k_q) of the approximation equation or the optimum current (I_d*₁, I_q*₁) are stored in a region where the optimum current (I_dp, I_qp) does not depend on the battery voltage Vdc, it is possible to execute vector control after the battery voltage V_dc is converged by using these stored pieces of information. Therefore, it is not necessary to store a complicated relationship between the reference torque value T*, motor rotational speed N, and battery voltage V_dc and an apparatus which can be realized at a low cost is obtained.

0096Furthermore, when the above approximation equation is used as a linear equation and the inclination (k_d, k_q) or intercept of the linear equation are stored as the coefficient, it is possible to make the advantage of decreasing the stored information content more remarkable. An Intercept can use the initial reference current (I_dmin, I_qmin).

0097When a rotational speed (N) of a synchronous motor (14) using permanent magnets as its field system is equal to or higher than a base rotational speed (N_B), a current condition computing section (16) computes a field weakening reference current (I_d*) in accordance with a battery voltage (V_dc), reference torque (T*), and the rotational speed (N). An inverter (12) performs the field weakening control of the synchronous motor (14) in accordance with the field weakening reference current (I_d*) sent from the current condition computing section (16) to prevent the deterioration of efficiency and the deficiency of output due to excessive or insufficient field weakening current (I_d). The current condition computing section (16) determines an optimum reference current (I_d*, I_q*) by using a converged battery voltage (V_s) when the battery voltage (V_dc) is converged after initial control.