Generator motor control method and generator motor control system

Abstract

In this method for controlling a generator motor, a motor torque command value is generated on the basis of a motor rotational speed command value and a motor rotational speed estimated value of the generator motor such that the generator motor rotated by an engine rotates following the motor rotational speed command value. In a method for controlling a generator motor in which a voltage command value for driving the generator motor is generated on the basis of a motor torque command value and a current detection value of the generator motor, the current detection value is input to a state observer that represents a voltage equation by a [gamma] [delta] axis having a prescribed phase difference with respect to a dq axis. A phase error estimation value and a motor rotation speed estimation value are calculated by inputting a voltage command value as a terminal voltage and calculating a rotor magnetic flux estimation value of the generator motor, the voltage equation representing the relationship between the current detection value and the terminal voltage of the generator motor through the dq axis of the generator motor, and the voltage command value representing the relationship between the current detection value and the terminal voltage of the generator motor through the dq axis of the generator motor. The parameter or the phase error estimation value constituting the state observer is corrected so that the deviation between the motor torque command value and the engine torque command value becomes small.

Description

Technical Field

[0001] This invention relates to a control method for a generator motor and a control system for a generator motor. Background Technology

[0002] JP2009-219338A discloses the following technology: for an electric motor that does not have a sensor for detecting rotor phase (electric angle), the magnetic flux of the rotor is estimated based on the detected current value flowing through the motor and the voltage command value for driving the motor under specified conditions, and the magnetic pole position of the magnet is estimated based on the estimated magnetic flux of the rotor. Summary of the Invention

[0003] In JP2009-219338A, because the structure sets the estimated electromagnetic characteristics (motor parameters) of the motor to a specified value, if the characteristics change due to the motor's operating state (motor speed), the estimated accuracy of the motor's rotor phase decreases, and the torque accuracy decreases. Therefore, when JP2009-219338A is applied to a structure that rotates a generator motor via an engine, the efficiency decreases due to this decrease in torque accuracy.

[0004] Therefore, the object of the present invention is to provide a control method and a control system for a generator motor that improves efficiency by increasing the estimation accuracy of the rotor phase of the generator motor that rotates based on the engine.

[0005] According to one aspect of the present invention, there is a control method for a generator-motor, wherein an engine that drives the generator-motor to rotate is driven based on an engine torque command value, such that the generator-motor rotates in accordance with a motor speed command value. A motor torque command value is generated based on the motor speed command value and an estimated motor speed value of the generator-motor. A voltage command value for driving the generator-motor is generated based on the motor torque command value and a current detection value of the generator-motor. In this control method, the current detection value is input to a state observer that represents a voltage equation via a γδ axis having a predetermined phase difference relative to the dq axis of the generator-motor, and the voltage command value is input as the terminal voltage of the generator-motor. An estimated rotor flux value of the generator-motor is calculated, thereby calculating an estimated phase error value. An estimated motor speed value is calculated based on the estimated phase error value. The voltage equation represents the relationship between the current detection value and the terminal voltage of the generator-motor via the dq axis, thereby correcting the parameters constituting the state observer or the estimated phase error value in a way that reduces the deviation between the motor torque command value and the engine torque command value. Attached Figure Description

[0006] Figure 1 This is an overall structural diagram of the generator system of the generator motor control system using the first embodiment.

[0007] Figure 2 This is a block diagram of the control system of the generator motor according to the first embodiment.

[0008] Figure 3 yes Figure 2 Block diagram of the phase / speed estimation section.

[0009] Figure 4 This is a flowchart of the control system of the generator motor in the first embodiment.

[0010] Figure 5 This is a timing diagram showing the error between the motor torque command value, q-axis inductance, and estimated electrical angle phase value and their actual values ​​when the q-axis inductance is corrected by the control system of the generator motor in the first embodiment.

[0011] Figure 6 This is a block diagram of the phase / speed estimation section of the control system of the generator motor in the second embodiment.

[0012] Figure 7 This is a flowchart of the control system of the generator motor in the second embodiment.

[0013] Figure 8 This is a timing diagram showing the error between the estimated values ​​of the motor torque command, q-axis inductance, and electrical angle phase and their actual values ​​when the phase error estimate is corrected by the control system of the generator motor in the second embodiment.

[0014] Figure 9 This is a flowchart of the control system of the generator motor in the third embodiment. Detailed Implementation

[0015] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0016] [First Implementation]

[0017] <Overall Structure of the Generator System>

[0018] Figure 1 This is an overall structural diagram of the generator system of the control system of the generator motor 6 using the first embodiment.

[0019] The engine 9 transmits the driving force used for power generation to the generator motor 6 via gear 7 and damper 8.

[0020] The generator motor 6 operates regeneratively by receiving the driving force of the engine 9, thereby generating electricity. Additionally, electricity can be consumed by rotating the crankshaft when the engine 9 is started, or by driving the generator motor 6.

[0021] Inverter 5 is connected to generator motor 6 and battery 4, converting the AC power generated by generator motor 6 into DC power to charge battery 4. Additionally, it converts the DC power supplied from battery 4 into AC power to supply generator motor 6.

[0022] Battery 4 is charged with regenerative power supplied from generator motor 6 and discharged with power used to drive generator motor 6.

[0023] Engine controller 3 receives the engine torque command value Teng from system controller 1. In order to achieve the engine torque command value Teng Based on signals such as engine speed or temperature, the throttle valve, ignition timing, and fuel injection quantity of engine 9 are adjusted.

[0024] Generator controller 2 is responsible for implementing the motor speed command value N sent from system controller 1. The inverter 5 is switched on and off based on the speed or voltage of the generator motor 6.

[0025] System controller 1 determines the battery 4's SOC (State of Charge), temperature, and other information, or obtains it from... Figure 1 The signal sent from outside the generator system is used by the generator controller 2 to set the engine torque command value Teng. and motor speed command value N (Based on engine torque command value Teng) The engine torque command value (Teng) is calculated and sent in the engine controller 3. The information is sent while performing calculations. Examples of external signals sent to system controller 1 include information related to various devices that are the target of power supply in the power generation system. For example, when power is supplied to the drive motor of a hybrid vehicle, information indicating the vehicle's status, such as the amount of accelerator pedal operation, vehicle speed, and road gradient, is sent to system controller 1 as an external signal.

[0026] <Generator Controller 2>

[0027] Figure 2 This is a block diagram of the control system (generator controller 2) of the generator motor 6 in the first embodiment.

[0028] The generator controller 2 drives the generator motor 6 using a sensorless design (a structure without sensors to detect the rotor phase (electric angle) of the generator motor 6), thereby realizing the control from the system controller 1. Figure 1 The motor speed command value N sent. The generator controller 2 includes a speed control unit 201, a first voltage command generation unit 202, a second voltage command generation unit 203, a voltage command determination unit 204, a final voltage command generation unit 205, a first coordinate conversion unit 206, a PWM conversion unit 207, a voltage sensor 209, a current sensor 210, a current command generation unit 211, a phase / speed estimation unit 212, and a second coordinate conversion unit 213.

[0029] Speed ​​control unit 201 receives motor speed command value N from system controller 1 The motor speed estimate N', estimated by the phase / speed estimation unit 212, is used as input, and a PI controller (not shown) or similar device is used to set the speed of the generator motor 6 (rotor) to the motor speed command value N. Motor torque command value Tgen Perform the calculation.

[0030] The current command generation unit 211 generates the motor torque command value Tgen The DC voltage Vdc of battery 4 and the estimated motor speed N' calculated by phase / speed estimation unit 212 are used as inputs. Referring to these inputs and the dq-axis current table calculated in advance through experiments or analysis, the d-axis current command value id is generated. q-axis current command value iq .

[0031] The first voltage command generation unit 202 generates the motor torque command value Tgen. The DC voltage Vdc of battery 4, the estimated motor speed N' estimated by phase / speed estimation unit 212, and the d-axis current command value id generated by current command generation unit 211. and q-axis current command value iq The d-axis current detection value id and the q-axis current detection value iq of the generator motor 6 are used as inputs to input the first d-axis voltage command value Vd1. The voltage command value Vq1 for the 1st q axis Perform the calculation.

[0032] The voltage command value Vd1 for the 1st d-axis The voltage command value Vq1 for the 1st q axis For example, as shown in mathematical formulas (1) and (2), PI control is performed based on the deviation between the current command value and the current detection value, and the interference voltage V along the d-axis is used. d_dcpl q-axis interference voltage V q_dcpl The d-axis interference voltage V is calculated by adding them together. d_dcpl q-axis interference voltage V q_dcpl It is the motor torque command value Tgen The DC voltage Vdc and the estimated motor speed N' calculated by the phase / speed estimation unit 212 are used as inputs, and the generator is generated with reference to the dq axis interference voltmeter calculated in advance through experiments or analysis.

[0033] [Formula 1]

[0034] [Equation 2]

[0035] Here, K p1 It is the proportional gain, K i1 It is the integral gain.

[0036] The second voltage command generation unit 203 generates the motor torque command value Tgen. The second d-axis voltage command value Vd2 is input to the DC voltage Vdc, the estimated motor speed N' estimated by the phase / speed estimation unit 212, the d-axis current detection value id and the q-axis current detection value iq of the generator motor 6. The voltage command value Vq2 for the 2nd q axis Perform the calculation.

[0037] The second voltage command generation unit 203, for example, is based on the DC voltage Vdc and the modulation rate command value MF. The standard voltage command value Va is obtained by using the following mathematical formula (3). .

[0038] [Formula 3]

[0039] The second voltage command generation unit 203 takes the d-axis current detection value id, the q-axis current detection value iq, and the motor speed estimation value N' as inputs, and uses the correspondence diagram obtained in advance through experiments or analysis to calculate the motor torque estimation value Test.

[0040] The second voltage command generation unit 203 is based on the motor torque estimation value Test and the motor torque command value Tgen. The deviation between them is calculated using the following mathematical formula (4), for example, by performing FB control such as PI control, to obtain the voltage phase correction value αfb. .

[0041] [Formula 4]

[0042] Here, K p2 It is the proportional gain, K i2 It is the integral gain.

[0043] The second voltage command generation unit 203 generates the motor torque command value Tgen. The DC voltage Vdc and the estimated motor speed N' derived from the phase / speed estimation unit 212 are used as inputs. Referring to a table obtained in advance through experiments or analysis, the target voltage phase value αff is generated. .

[0044] The second voltage command generation unit 203 generates the voltage phase correction value αfb. With voltage phase target value αff The voltage phase command value α is generated by adding them together. Perform the vector transformation shown in the following mathematical formulas (5) and (6) to obtain the voltage command value Vd2 of the 2d axis. The voltage command value Vq2 for the 2nd q axis Perform the calculation.

[0045] [Formula 5]

[0046] [Formula 6]

[0047] The voltage command determination unit 204 determines whether to select the output of the first voltage command generation unit 202 or the output of the second voltage command generation unit 203, and outputs a signal Msw indicating the selected output.

[0048] One signal generation method is as follows: if the final d-axis voltage command value Vd is generated based on the output of the final voltage command generation unit 205... The final q-axis voltage command value Vq If the modulation rate MF calculated from the DC voltage Vdc and the following mathematical formula (7) is less than a specified value, then the output is the first d-axis voltage command value Vd1 generated by the first voltage command generation unit 202. and the voltage command value Vq1 of the 1st q axis If the selected signal is greater than or equal to a specified value, the output is the second d-axis voltage command value Vd2 generated by the second voltage command generation unit 203. and the voltage command value Vq2 of the 2q axis The signal is used for selection. At this point, to prevent jitter, the switching threshold is usually made to have hysteresis.

[0049] [Formula 7]

[0050] The final voltage command generation unit 205 selects either the output of the first voltage command generation unit 202 or the output of the second voltage command generation unit 203 as the final d-axis voltage command value Vd based on the signal Msw from the voltage command determination unit 204. The final q-axis voltage command value Vq And the output.

[0051] The first coordinate transformation unit 206 will convert the final d-axis voltage command value Vd The final q-axis voltage command value Vq The electrical angle phase estimate θ' (a value corresponding to the magnetic pole position) estimated by the phase / speed estimation unit 212 is used as input and converted into a three-phase voltage command value (Vu) based on the following mathematical formula (8). Vv Vw ).

[0052] [Formula 8]

[0053] The PWM converter 207 inputs the three-phase voltage command value (Vu). Vv Vw The three-phase voltage command value (Vu) is generated by performing known processing such as dead-time compensation or voltage utilization improvement on the DC voltage Vdc and DC voltage Vdc. Vv Vw The corresponding power element drive signal (Duu) of inverter 5 Dul Dvu Dvl Dwu Dwl The inverter 5 is connected to the battery 4, and converts the DC voltage Vdc output from the battery 4 into analog AC voltages (Vu, Vv, Vw) for output. Furthermore, the DC voltage Vdc is detected by the voltage sensor 209.

[0054] The current sensor 210 detects, for example, the U-phase current iu and the V-phase current iv of the generator motor 6. Furthermore, in principle, the W-phase current iw of the generator motor 6 can be calculated using the following mathematical formula (9).

[0055] [Formula 9]

[0056] The second coordinate transformation unit 213 takes the U-phase current iu and V-phase current iv detected by the current sensor 210 and the electrical angle phase estimation value θ' estimated by the phase / speed estimation unit 212 as inputs, and converts them into the d-axis current detection value id and the q-axis current detection value iq based on the following mathematical formula (10).

[0057] [Formula 10]

[0058] <Phase / Speed ​​Estimation Section 212>

[0059] Figure 3 yes Figure 2 A block diagram of the phase / speed estimation unit 212. The phase / speed estimation unit 212 includes a rotor flux estimation unit 301, a PLL control unit 302, and an Lq correction calculation unit 303.

[0060] The output of the final voltage command generation unit 205, namely the final d-axis voltage command value Vd, is input to the rotor flux estimation unit 301. and the final q-axis voltage command value Vq The outputs of the second coordinate transformation unit 213, namely the d-axis current detection value id and the q-axis current detection value iq, are shown below, and the estimated values ​​of rotor magnetic flux (Φmd' (=Φmγ), Φmq' (=Φmδ)) are calculated.

[0061] Here, the relationship between the d-axis current detection value id and the q-axis current detection value iq and the d-axis terminal voltage vd and the q-axis terminal voltage vq can be expressed by the voltage equation (mathematical formula (11)) expressed on the dq axis of the generator motor 6.

[0062] [Equation 11]

[0063] Here, R is the winding resistance [Ω], Ld is the d-axis inductance [H], Lq is the q-axis inductance [H], s is the differential operator, ω is the rotational angular velocity [rad / s], and Φ is the magnetic flux [Wb] of the rotor of the generator motor 6.

[0064] The rotor flux estimation unit 301 has a state observer (mathematical formula (14)) which re-expresses the above voltage equation (mathematical formula (11)) through the γδ axis with a phase difference of θγ relative to the dq axis (mathematical formula (12)) and converts it into a state space expression with rotor flux Φm as the state variable (mathematical formula (13)).

[0065] [Equation 12]

[0066] [Equation 13]

[0067] [Formula 14]

[0068] Here, G is a 2x2 observer gain matrix, and I is a 2x2 identity matrix.

[0069] The rotor flux estimation unit 301 calculates and outputs the estimated rotor flux values ​​(Φmγ, Φmδ) using mathematical formula (14). Here, vγ, vδ, iγ, and iδ in mathematical formula (14) correspond to the input (Vd) of the rotor flux estimation unit 301. Vq , id, iq), the Φmγ and Φmδ in mathematical formula (14) correspond to the outputs (Φmd', Φmq') of the rotor flux estimation unit 301. That is, the final d-axis voltage command value Vd and the final q-axis voltage command value Vq They are respectively input as terminal voltages.

[0070] The PLL control unit 302 takes the output of the rotor flux estimation unit 301, i.e. the rotor flux estimation value (Φmd', Φmq'), as input, and performs arctan processing of the mathematical formula (15) shown below to calculate the phase error estimation value θγ'.

[0071] [Formula 15]

[0072] The PLL control unit 302 takes the estimated phase error value θγ' as input and performs PI control according to the following mathematical formula (16) to calculate the estimated electrical angular velocity value ω'.

[0073] [Formula 16]

[0074] Here, K p_pll It is the proportional gain, K i_pllIt is the integral gain. Additionally, the estimated motor speed N' is obtained by converting the estimated electrical angular velocity ω' to [rpm].

[0075] The PLL control unit 302 calculates the electrical angular phase estimate θ' by integrating the electrical angular velocity estimate ω' as shown in the following mathematical formula (17).

[0076] [Equation 17]

[0077] Here, in the rotor flux estimation unit 301, the impact of the error between the motor parameters set in the observer and the actual value on the sensorless control performance is explained. When the value of the winding resistance or inductance set in mathematical formula (14) has an error relative to the actual value of the generator motor 6, the rotor flux estimation value has a stable error relative to the actual value. As a result, the final output of the phase / speed estimation unit 212, namely the electrical angle phase estimation value θ', also has a stable error relative to the actual value. In particular, the setting error of the q-axis inductance relative to the actual value has a large impact on the phase estimation accuracy of the observer. However, on the other hand, when "the q-axis inductance setting value Lq > the actual q-axis inductance value", the sensorless drive system may become unstable. Therefore, in order to avoid this instability, the q-axis inductance setting value Lq needs to be set to be less than or equal to the actual q-axis inductance value. Furthermore, since the actual value of the q-axis inductance varies depending on the q-axis current (the larger the q-axis current, the smaller the q-axis inductance), the q-axis inductance setting value Lq is set to be less than the value that the q-axis inductance may take in all operating regions of the generator motor 6.

[0078] Furthermore, in the aforementioned situation, in order to achieve the motor torque command value Tgen, the electrical angle phase estimation value θ' is used. The calculated final d-axis voltage command value Vd The final q-axis voltage command value Vq When voltage is applied to generator motor 6, the actual value of the generated motor torque differs from the motor torque command value Tgen. Inconsistency leads to decreased torque accuracy. Consequently, when aiming for efficient control at a target operating point (Maximum Torque Per Ampere), the control deviates from this target operating point, resulting in decreased efficiency (generation efficiency, electricity cost efficiency).

[0079] Therefore, the Lq correction calculation unit 303 corrects the q-axis inductance as shown below, thereby increasing the motor torque command value Tgen. It approaches its true value, thereby improving torque accuracy.

[0080] Lq correction calculation unit 303 calculates the output of speed control unit 201, i.e., motor torque command value Tgen. and the engine torque command value Teng sent from system controller 1 to generator controller 2 As input, the correction value L is assigned to the q-axis inductance setting value Lq of the mathematical formula (14) of the state observer of the opposing rotor flux estimation unit 301. q_comp The calculations are performed and the result is output.

[0081] The Lq correction operation unit 303, for example, performs the correction value L as shown in mathematical formula (18). q_comp Perform the calculation.

[0082] [Formula 18]

[0083] Lq correction calculation unit 303 calculates the motor torque command value Tgen. The motor torque command value Tgen after removing vibration components through a low-pass filter (LPF) The average value T¯gen The Lq correction calculation unit 303 performs calculations on the engine torque command value Teng. With motor torque command value Tgen The average value T¯gen The deviation is integrated and multiplied by a gain K determined to achieve the level of correction desired by the designer. Lq_comp The obtained value is used as L q_comp And set as the output of Lq correction calculation unit 303. In mathematical formula (18), the motor torque command value Tgen on the regenerator side is taken into account. The average value T¯gen When the sign is negative, use T¯gen. absolute value | T¯gen | Calculate the torque deviation. At this point, in situations such as... Figure 1 In this way, when the generator motor 6 and the engine 9 are connected via gear 7, the gear ratio K is used. gear The torque deviation on this shaft is calculated. Additionally, by adjusting the torque... With T¯gen The deviation has an offset value T offset (T) offset Integrate the value of >0) to obtain the correction value L q_comp Perform calculations to make Teng With T¯gen The deviation becomes T offset .

[0084] In this embodiment, when the engine 9 outputs engine torque and the generator motor 6 generates electricity at a constant rotation, the correction implemented by mathematical formula (18) is applied. Therefore, the engine torque command value Teng It is not 0, and the motor torque command value Tgen The average value T¯gen If the absolute value of the time change rate is less than or equal to the specified value, it is determined that the generator motor 6 is in a state of generating electricity with constant rotation, and the mathematical formula (18) is calculated to start the correction.

[0085] This explains the effect of the correction achieved by mathematical formula (18) when the generator motor 6 is generating electricity at a constant rotation. In this operating state, both the engine 9 and the generator motor 6 generate torque, which is transmitted through… Figure 2 The speed control unit 201 of the generator controller 2 operates to maintain constant rotation while the torque of the engine 9 and the torque of the generator motor 6 are in balance. That is, the engine torque command value Teng... With motor torque command value Tgen The absolute value is obtained through the gear ratio K gear And become the same value.

[0086] However, when the parameters of the generator motor 6, such as the q-axis inductance, set by the state observer in the phase / speed estimation unit 212 deviate from their true values, the phase estimation accuracy deteriorates. As a result, the motor torque command value Tgen... The torque generated by the generator motor 6 deviates from the actual torque output, resulting in decreased torque accuracy. In this situation, the speed control unit 201 ensures constant rotation to balance the torque generated by the generator motor 6 with the torque of the engine 9, but the motor torque command value Tgen... Becomes relative to the engine torque command value Teng And values ​​that deviate from the balance.

[0087] Therefore, through mathematical formula (18), the engine torque command value Teng is obtained. With motor torque command value Tgen The absolute value deviation is reduced by assigning a correction value to the q-axis inductance value of the state observer's mathematical formula (14), thereby reducing the phase estimation error and increasing the motor torque command value Tgen. The error between the torque generated by the actual generator motor 6 and the torque accuracy is also reduced, thus improving torque accuracy.

[0088] Here, mathematical formula (18) uses the motor torque command value Tgen when calculating the torque deviation. T¯gen after LPF processing The torque actually generated by engine 9 includes a vibration component, and the torque of generator-motor 6 is controlled in a way that balances this component. Therefore, the motor torque command value Tgen is... It also includes vibration components. Engine torque command value Teng Since it does not contain vibration components, the torque deviation is calculated using the motor torque command value Tgen. The value after removing the vibrational component is T¯gen .

[0089] Furthermore, mathematical expression (18) is based on |T¯gen | With Teng The deviation is not 0 but T offset The formula for correction is as follows. For the parameters of the generator motor 6 set in the mathematical formula (14) of the state observer, if not only the q-axis inductance, but also the winding resistance or d-axis inductance has errors relative to the true value, even if only the q-axis inductance is corrected to the true value, the phase estimation error will not become 0, and the torque deviation will not become 0. That is, it is possible to perform q-axis inductance correction until "q-axis inductance setting value Lq>q-axis inductance true value", and the sensorless drive system becomes unstable. Therefore, considering the existence of setting errors of parameters other than q-axis inductance, an offset value T is set for deviation 0. offset .

[0090] <Flowchart of the first embodiment>

[0091] Figure 4 This is a flowchart of the control system for the generator motor 6 in the first embodiment. Based on Figure 4 The flowchart illustrates the process of correcting the q-axis inductance setting value Lq implemented by the Lq correction calculation unit 303 of the above structure. In the initial state, the correction value L... q_comp It is zero.

[0092] In Step 101, the Lq correction calculation unit 303 obtains Teng and Tgen .

[0093] In Step 102, the Lq correction calculation unit 303 calculates the Tgen. Motor torque command value Tgen with vibration component removed by applying LPF The average value T¯gen .

[0094] In Step 103, the Lq correction calculation unit 303 calculates according to L... q_comp The value of L is used to determine whether correction began at the previous time point. Here, if L... q_comp If the value is 0 (initial value), it is determined that calibration has not yet started, and proceed to Step 104. Otherwise, it is determined that calibration has started, and proceed to Step 107.

[0095] In Step 104, the Lq correction arithmetic unit 303 adjusts T¯gen. The rate of change over time is calculated.

[0096] In Step 105, the Lq correction calculation unit 303 determines whether the generator motor 6 is in a state of constant rotation for power generation. Specifically, if "T¯gen The absolute value of the time rate of change is less than or equal to the specified value and "Teng If the value is not 0, the Lq correction calculation unit 303 determines that the generator motor 6 is in a state of generating electricity at a constant rotation, and proceeds to Step 107. Otherwise, proceeds to Step 106.

[0097] In Step 106, the Lq correction calculation unit 303 does not perform q-axis inductance correction, and instead... q_comp The value is set to 0 (initial value).

[0098] In Step 107, the Lq correction operation unit 303 corrects L using mathematical formula (18). q_comp The value is calculated.

[0099] In Step 108, the Lq correction calculation unit 303 assigns the calculated correction value Lq to the q-axis inductance setting value Lq in the mathematical formula (14) of the state observer. q_comp The q-axis inductance is corrected in the form of [formula / method].

[0100] <Timing Diagram of Implementation Method 1>

[0101] Figure 5 The motor torque command value Tgen is obtained when the q-axis inductance is corrected by the control system of the generator motor 6 in the first embodiment. Timing diagram (curve of correction effect) showing the error between the estimated value θ' of the q-axis inductance and the actual value of the electrical angle phase.

[0102] Before correction (before time t0), the actual value of the q-axis inductance has a large error compared to the set value (Lq), and the estimated electrical angle phase θ' also has a large error compared to the actual value. As a result, the motor torque command value Tgen... The average value T¯gen The large deviation between the absolute value of the torque and the absolute value (average value) of the actual torque (T) generated by the generator motor 6 results in poor torque accuracy.

[0103] Then, at time t0, if it is determined that the generator motor 6 is generating electricity at a constant rotation, correction begins. The correction value L is added to the q-axis inductance set value Lq by the Lq correction calculation unit 303. q_comp This is close to the true value, therefore, the phase error is reduced. Additionally, T¯gen By approximating the absolute value (average value) of the actual generated torque (T), torque accuracy is improved. Therefore, when control is desired at an efficient operating point such as MTPA, it is possible to approach the target operating point through correction, thus improving efficiency.

[0104] [Second Implementation]

[0105] Figure 6 This is a block diagram of the phase / speed estimation unit 212 of the control system of the generator motor 6 in the second embodiment. In the first embodiment, the q-axis inductance setting value Lq of the mathematical formula (14) of the state observer was corrected, but in addition to the q-axis inductance, the phase error estimation value θγ' can also be corrected by the PLL control unit 602, thereby improving the phase estimation accuracy.

[0106] like Figure 6 As shown, the phase / speed estimation unit 212 includes a rotor flux estimation unit 601, a PLL control unit 602, and an θγ' correction calculation unit 603.

[0107] The structure of the rotor flux estimation unit 601 and Figure 3 The rotor flux estimation unit 301 is the same, and the structure of the PLL control unit 602 is the same. Figure 3 It is the same as the PLL control unit 302.

[0108] The θγ' correction calculation unit 603 calculates the output of the speed control unit 201, i.e., the motor torque command value Tgen. and the engine torque command value Teng sent from system controller 1 to generator controller 2 As input, the correction value θγ' is assigned to the phase error estimate θγ' calculated by the PLL control unit 602 using mathematical formula (15). _comp The calculations are performed and the result is output.

[0109] Correction value θγ' _comp For example, it is calculated as follows. First, for the motor torque command value Tgen... The motor torque command value Tgen, which has had vibration components removed by LPF. The average value T¯gen Calculations are performed on the engine torque command value Teng. With motor torque command value Tgen The absolute value of the average | T¯gen The torque deviation is calculated until it is less than or equal to the specified value T. error Up to this point, the correction value θγ' is adjusted in each control cycle. _comp Add a certain fixed value θγ' _const When the torque deviation becomes less than or equal to the specified value T error Then, the correction value θγ' _comp Once the value is fixed, the calibration ends.

[0110] In the second embodiment, similarly to the first embodiment, the above correction is performed when the engine 9 outputs engine torque and the generator motor 6 generates electricity at a constant rotation. Therefore, the engine torque command value Teng is... It is not 0 and the motor torque command value Tgen The average value T¯gen If the absolute value of the time change rate is less than or equal to the specified value, it is determined that the generator motor 6 is generating electricity at a constant rotation, and the above correction is initiated.

[0111] By performing the above correction method while the generator motor 6 is generating electricity at a constant rotation, the phase estimation error is reduced, similar to the first embodiment, thereby improving torque accuracy. Furthermore, by setting it to only apply the determined fixed value θγ'... _const A simple calculation involving addition to the correction value, and when the torque deviation is less than or equal to the specified value T. error The correction value θγ' will be used. _comp Fixing the operation terminates the correction calculation, thereby reducing the computational load.

[0112] Here, the correction ends when the torque deviation is less than or equal to a specified value T that is not zero. error And the determination. When attempting to control the generator motor 6 at the MTPA operating point, regardless of whether the estimated electrical angle phase value θ' has an error of being ahead or behind the actual value, it is related to the motor torque command value Tgen. Conversely, the actual torque generated by the generator motor 6 decreases. That is, regardless of the sign of the phase estimation error, it becomes the "motor torque command value Tgen". ≥ Engine torque command value Teng "When correcting θγ', if the phase estimation error crosses 0 and the sign reverses, the correction is performed in the direction of increasing torque deviation. Therefore, the torque deviation is less than or equal to the specified value T, which has a margin without crossing 0." error When the time is right, the calibration is considered complete.

[0113] <Flowchart of the second embodiment>

[0114] Figure 7 This is a flowchart of the control system for the generator motor 6 in the second embodiment. Based on Figure 7 The flowchart illustrates the process of correcting the phase error estimate θγ' implemented by the θγ' correction calculation unit 603 of the above structure. In the initial state, θγ' _comp It is zero.

[0115] In Step 201, the θγ' correction calculation unit 603 obtains Teng and Tgen .

[0116] In Step 202, the θγ' correction calculation unit 603 determines whether θγ' has been corrected up to the previous time. _comp Fix and end the calibration. If the calibration is not yet finished, proceed to Step 203. If the calibration is finished, proceed to Step 211.

[0117] In Step 203, the θγ' correction calculation unit 603 is for Tgen The torque command value Tgen of the motor after applying LPF to remove vibration components. The average value T¯gen Perform the calculation.

[0118] In Step 204, the θγ' correction calculation unit 603 calculates based on θγ' _comp The value of θγ' is used to determine whether correction began at the previous time point. Specifically, if θγ' _comp If the value is 0, it is determined that calibration has not yet started, and proceed to Step 205. Otherwise, it is determined that calibration has started, and proceed to Step 207.

[0119] In Step 205, the θγ' correction operation unit 603 adjusts T¯gen The rate of change over time is calculated.

[0120] In Step 206, the θγ' correction calculation unit 603 determines whether the generator motor 6 is in a state of constant rotation for power generation. Specifically, if "T¯gen The absolute value of the time rate of change is less than or equal to the specified value and "Teng If the value is not 0, it is determined that the generator motor 6 is generating electricity at a constant rotation, and proceeds to Step 208. Otherwise, proceed to Step 209.

[0121] In Step 207, the θγ' correction calculation unit 603 adjusts the engine torque command value Teng. With motor torque command value Tgen The absolute value of the average | T¯gen The torque deviation is calculated. When the torque deviation is less than or equal to the specified value T... error In the case of [condition], proceed to Step 210; otherwise, proceed to Step 208.

[0122] In Step 208, the θγ' correction calculation unit 603 corrects the value θγ'. _comp Add a fixed value θγ' _const .

[0123] In Step 209, the θγ' correction calculation unit 603 calculates the correction value θγ'. _comp Set to 0 (initial value).

[0124] In Step 210, the θγ' correction calculation unit 603 calculates the correction value θγ'. _comp If the value is fixed at the current value, it is determined that the correction has ended.

[0125] In Step 211, the θγ' correction calculation unit 603 adds the correction value θγ' to the phase error estimate θγ' calculated by the PLL control unit 602 based on mathematical formula (15). _comp The estimated phase error value θγ' is corrected in the form of [formula / formula].

[0126] <Timing Diagram of Implementation Method 2>

[0127] Figure 8 The motor torque command value Tgen is obtained when the estimated phase error θγ' is corrected by the control system of the generator motor 6 in the second embodiment. Timing diagram (curve of correction effect) showing the error between the estimated value θ' of the q-axis inductance and the actual value of the electrical angle phase.

[0128] Before correction (before time t0), the actual value of the q-axis inductance has a large error compared to the set value, and the estimated electrical phase angle θ' also has a large error compared to the actual value. As a result, the motor torque command value Tgen... The absolute value of the average | T¯gen | A large deviation between the absolute value of the actual generated torque (T) of the generator motor 6 and the actual generated torque is considered a state of poor torque accuracy.

[0129] Then, at time t0, if the θγ' correction calculation unit 603 determines that the generator motor 6 is in a state of generating electricity at a constant rotation, it starts to correct. The phase error estimation value θγ' is corrected by the θγ' correction calculation unit 603 to be close to the true value. Therefore, the electrical angle phase estimation value θ' is also close to the true value, the phase estimation error is reduced, and the torque accuracy is improved.

[0130] Then, at time t1, if the torque deviation is less than or equal to the specified value T error Then the correction value θγ' _comp The value is fixed, and the calibration ends. Furthermore, in this case, since the q-axis inductance setting is not calibrated, the error between the actual q-axis inductance and the setting (Lq) remains large, but the phase estimation error is reduced. Therefore, it can be confirmed that calibrating the phase error estimation value θγ' also has an effect.

[0131] Furthermore, the mathematical formula (14) of the state observer is constructed on the dq axis of the generator motor 6, but it is also possible to construct the observer on the αβ axis (fixed coordinate axis). In this case, the estimated electrical angular phase value θ' estimated by the observer constructed on the αβ axis corresponds to a value equivalent to the magnetic pole position.

[0132] [Third Implementation]

[0133] In the third embodiment, the correction value L of the Lq correction calculation unit 303 is changed compared to the first embodiment. q_comp The calculation method is the same as that in the first embodiment. The structure of the rotor flux estimation unit 301 and the PLL control unit 302 is the same as that in the first embodiment.

[0134] Lq correction calculation unit 303 calculates the output of speed control unit 201, i.e., motor torque command value Tgen. and the engine torque command value Teng sent from system controller 1 to generator controller 2 As input, the correction value L is assigned to the q-axis inductance setting value Lq of the mathematical formula (14) of the state observer of the opposing rotor flux estimation unit 301. q_comp The calculations are performed and the result is output.

[0135] Correction value L q_comp For example, it is calculated as follows. First, for the motor torque command value Tgen... The motor torque command value Tgen, which has had vibration components removed by LPF. The average value T¯gen Calculations are performed on the engine torque command value Teng. With motor torque command value Tgen The absolute value of the average | T¯eng The torque deviation is calculated. For each torque deviation, a specific correction value L is pre-created. q_comp The corresponding diagram will show the correction value L determined based on the torque deviation. q_comp The correction value is fixed by adding the q-axis inductance setting value Lq of the state observer of the rotor flux estimation unit 301 in a stepwise manner.

[0136] In the third embodiment, similarly to the first embodiment, the above correction is applied when the engine torque is output and the generator motor 6 is generating electricity at a constant rotation. Therefore, the engine torque command value Teng It is not 0 and the motor torque command value Tgen The average value T¯gen If the absolute value of the time change rate is less than or equal to the specified value, it is determined that the generator motor 6 is generating electricity at a constant rotation, and then the above correction is performed.

[0137] By performing the above correction method while the generator motor 6 is generating electricity at a constant rotation, the phase estimation error is reduced, similar to the first embodiment, thereby improving torque accuracy. Furthermore, since the correction is performed all at once based on the correction value determined by the torque deviation, the correction can be completed quickly.

[0138] <Flowchart of the third embodiment>

[0139] Figure 9 This is a flowchart of the control system for the generator motor 6 in the third embodiment. Based on Figure 9 The flowchart below explains the process of correcting the q-axis inductance setting value Lq by the Lq correction calculation unit 303 of the above structure.

[0140] In Step 301, the Lq correction calculation unit 303 obtains Teng and Tgen .

[0141] In Step 302, the Lq correction calculation unit 303 determines whether the correction value L has been determined up to the previous time. q_comp If the decision has been made, proceed to Step 303; otherwise, proceed to Step 304.

[0142] In Step 303, the Lq correction calculation unit 303 calculates the correction value L. q_comp Set it to be the same as the previous value.

[0143] In Step 304, the Lq correction calculation unit 303 performs a correction for Tgen. The torque command value Tgen of the motor after applying LPF to remove vibration components. The average value T¯gen Perform the calculation.

[0144] In Step 305, the Lq correction arithmetic unit 303 corrects T¯gen. The rate of change over time is calculated.

[0145] In Step 306, the Lq correction calculation unit 303 determines whether the generator motor 6 is in a state of constant rotation for power generation. Specifically, if "T¯gen The absolute value of the time rate of change is less than or equal to the specified value and "Teng If the value is not 0, it is determined that the generator motor 6 is generating electricity at a constant rotation, and proceeds to Step 307. Otherwise, proceed to Step 308.

[0146] In Step 307, the Lq correction calculation unit 303 determines the correction value L corresponding to the torque deviation by referring to the corresponding diagram. q_comp The value of .

[0147] In Step 308, the Lq correction calculation unit 303 calculates the correction value L. q_comp The value is set to 0.

[0148] In Step 309, the Lq correction calculation unit 303 adds the calculated L to the q-axis inductance setting value Lq of the mathematical formula (14) of the state observer. q_comp The q-axis inductance is corrected in the form of [formula / form].

[0149] In the third embodiment, if it is determined that the generator motor 6 is generating electricity at a constant rotation, then referring to the corresponding diagram, a correction value L corresponding to the torque deviation at that time point is determined. q_comp The q-axis inductance setting value Lq is added to the mathematical formula (14) of the state observer of the rotor flux estimation unit 301 as a step change, and then the correction value is fixed. As a result, the error between the q-axis inductance setting value Lq and the true value is reduced, the electrical angle phase estimation value θ' is also close to the true value, the phase estimation error is reduced, and the torque accuracy is improved. Since the correction value is changed stepwise in one go, the time consumed by the correction can be shortened.

[0150] [Modifications of the first embodiment]

[0151] In a variation of the first embodiment, the correction value θγ' of the second embodiment... _comp Similarly, the Lq correction arithmetic unit 303 ( Figure 3 Set the correction value L q_comp .

[0152] Correction value L q_comp For example, it is calculated as follows. First, for the motor torque command value Tgen... The motor torque command value Tgen, which has had vibration components removed by LPF. The average value T¯gen Calculations are performed on the engine torque command value Teng. With motor torque command value Tgen The absolute value of the average | T¯gen The torque deviation is calculated until it becomes less than or equal to the specified value T. error Up to this point, the correction value L is adjusted in each control cycle. q_comp Add a certain fixed value Lq _const When the torque deviation becomes less than or equal to the specified value T error Then, the correction value L q_comp Once the value is fixed, the calibration ends.

[0153] Furthermore, the flowchart of the modified example of the first embodiment is equivalent to... Figure 7 Lieutenant General (θγ') _comp ,θγ' _const Replace with (L) q_comp L q_const ), replace Step211 with Figure 4 Step 108.

[0154] [Modifications of the Second Embodiment]

[0155] In a variation of the second embodiment, the correction value L is the same as that in the first embodiment. q_comp Similarly, θγ' correction operation unit 603 ( Figure 6 Set the correction value θγ' _comp .

[0156] The θγ' correction operation unit 603, for example, performs a correction operation on the correction value θγ' as shown in mathematical expression (19). _comp Perform the calculation.

[0157] [Formula 19]

[0158] The θγ' correction calculation unit 603 calculates the motor torque command value Tgen. The motor torque command value Tgen, which has had vibration components removed by a low-pass filter (LPF). The average value T¯gen The calculation is performed. The θγ' correction calculation unit 603 calculates the engine torque command value Teng. With motor torque command value Tgen The average value T¯gen The deviation is integrated and multiplied by a gain K determined to achieve the level of correction desired by the designer. θγ’_comp The obtained value is used as θγ' _comp The output is then sent to the PLL control unit 602. Furthermore, similar to the first embodiment, by adjusting the Teng... With T¯gen The deviation has an offset value T offset (T) offset Integrate the values ​​after >0 to obtain the correction value θγ' _comp Perform calculations to make Teng With T¯gen The deviation becomes T offset .

[0159] Furthermore, the flowchart of the modified example of the second embodiment is equivalent to... Figure 4 Lieutenant General L q_comp Replace with θγ' _comp Replace Step108 with Figure 7 Step 211.

[0160] [Modifications of the Third Embodiment]

[0161] In a variation of the third embodiment, the correction value L is the same as that in the third embodiment. q_comp Similarly, θγ' correction operation unit 603 ( Figure 6 Set the correction value θγ' _comp .

[0162] Correction value θγ' _comp For example, it is calculated as follows. First, for the motor torque command value Tgen... The motor torque command value Tgen, which has had vibration components removed by LPF. The average value T¯gen Calculations are performed on the engine torque command value Teng. With motor torque command value Tgen The absolute value of the average | T¯eng The torque deviation is calculated. For each torque deviation, a specific correction value θγ' is pre-created. _comp The corresponding diagram will show the correction value θγ' determined based on the torque deviation. _compThe correction value is fixed by adding the estimated phase error value θγ' calculated by the PLL control unit 602 through mathematical formula (15) in a stepwise manner.

[0163] Furthermore, the flowchart of the modified example of the third embodiment is equivalent to... Figure 9 In the middle, L q_comp Replace with θγ' _comp Replace Step309 with Figure 7 Step 211.

[0164] [Effects of this implementation method]

[0165] The control method for the generator motor 6 in this embodiment is based on the engine torque command value Teng. The engine 9 drives the generator motor 6 to rotate, so that the generator motor 6 follows the motor speed command value N. The rotation method is based on the motor speed command value N. Based on the aforementioned estimated motor speed N' of the generator motor, the motor torque command value Tgen is generated. Based on the motor torque command value Tgen Based on the current detection values ​​(id, iq) of generator motor 6, a voltage command value (Vd) is generated to drive generator motor 6. Vq In the control method of the generator motor 6, the current detection values ​​(id, iq) are input to the state observer (mathematical formula (14)) which represents the voltage equation (mathematical formula (11)) with a specified phase difference (θγ) relative to the dq axis, and the voltage command value (Vd) is input to the state observer (mathematical formula (14)). Vq The input is the terminal voltage (vd, vq), which is used to calculate the estimated values ​​of the rotor flux (Φmd' (=Φmγ), Φmq' (=Φmδ)) of the generator motor 6, thereby calculating the estimated value of the phase error θγ', and the estimated value of the motor speed N' is calculated based on the estimated value of the phase error θγ'. The voltage equation (mathematical formula (11)) represents the relationship between the current detection value (id, iq) and the terminal voltage (vd, vq) of the generator motor 6 through the dq axis of the generator motor 6, so that the motor torque command value Tgen is obtained. With engine torque command value Teng The deviation (torque deviation) between them is reduced in such a way that the parameters (e.g., q-axis inductance) or the estimated phase error θγ' of the state observer (mathematical formula (14)) are corrected.

[0166] By using the above method, the engine torque command value Teng is made possible. With motor torque command value Tgen By reducing torque deviation, parameters (such as q-axis inductance) or the estimated phase error θγ' are corrected, thereby making the estimated rotor flux (Φmd' (=Φmγ), Φmq' (=Φmδ)) (estimated values ​​of magnetic pole position) closer to the true value, and improving the phase estimation accuracy (accuracy of the estimated phase error θγ'). Furthermore, as a result, torque accuracy is improved; therefore, when controlling the generator-motor 6 at a highly efficient operating point such as MTPA, control can be performed at the target operating point, improving efficiency.

[0167] In this embodiment, the q-axis inductance, which constitutes the parameters, is corrected.

[0168] Using the above method, among the parameters of the generator motor 6 that need to be set in order to estimate the magnetic pole position, the value of the q-axis inductance has a significant impact on the phase estimation accuracy. Therefore, by correcting the q-axis inductance value to be close to the true value, the phase estimation accuracy can be improved.

[0169] In this embodiment, when correcting parameters (e.g., q-axis inductance), a first correction value (L) is added to the parameter. q_comp This corrects the parameters and makes the first correction value (L) so that the deviation (torque deviation) becomes the specified deviation. q_comp ) Increase (mathematical formula (18)) based on the deviation (torque deviation) for the first correction value (L) q_comp The amount of increase is set (mathematical formula (18)). When correcting the estimated phase error value θγ', the second correction value (θγ') is added to the estimated phase error value θγ'. _comp This corrects the estimated phase error value θγ', and makes the second correction value (θγ') such that the deviation (torque deviation) becomes the specified deviation. _comp Increase (mathematical formula (19)) based on the deviation (torque deviation) for the second correction value (θγ') _comp The increase in ) is set (mathematical formula (19)).

[0170] By using the above method, the correction value is calculated by inputting the torque deviation into the integrator (mathematical formula (18) or mathematical formula (19)), so that the larger the torque deviation, the greater the change in the correction value, and thus the rapid improvement of the phase estimation accuracy can be achieved.

[0171] In this embodiment, when correcting parameters (e.g., q-axis inductance), a first correction value (L) is added to the parameter. q_compThis corrects the parameters, and in a manner that the deviation (torque deviation) becomes the specified deviation, so that the first correction value (L) is achieved. q_comp Starting from zero, gradually increase (gradually add a fixed value Lq). _const When correcting for the estimated phase error θγ', the second correction value (θγ') is added to the estimated phase error θγ'. _comp This corrects the estimated phase error value θγ', and makes the second correction value (θγ') so that the deviation (torque deviation) becomes the specified deviation. _comp Starting from zero, gradually increase (gradually add a fixed value θγ'). _const ).

[0172] By using the above method, and by repeatedly performing the correction step (L) until the torque deviation reaches the specified deviation, the process is repeated at regular intervals. q_comp ,θγ' _comp ) plus a fixed value (Lq) _const ,θγ' _const And set it as the new correction value (L) q_comp ,θγ' _comp This allows for simpler operations, thus reducing the computational load.

[0173] In this embodiment, when correcting parameters (e.g., q-axis inductance), a first correction value (L) is added to the parameter. q_comp This allows for parameter correction, and the first correction value (L) is set in such a way that the deviation (torque deviation) becomes a specified deviation. q_comp When correcting for the estimated phase error θγ', the second correction value (θγ') is added to the estimated phase error θγ'. _comp This corrects the estimated phase error value θγ', and sets a second correction value (θγ') in such a way that the deviation (torque deviation) becomes a specified deviation. _comp ).

[0174] By using the above method, the correction value corresponding to the torque deviation can be set as a step value in a single calculation, thereby shortening the time spent on correction.

[0175] In this embodiment, the specified deviation (T) will be defined. offset Set to a value other than zero (e.g., T) offset >0).

[0176] Using the above method, when the parameter setting values ​​of the phase / speed estimation unit 212, excluding the q-axis inductance, deviate from the true values, even if the q-axis inductance is set to the true value, the phase estimation error remains, and the torque deviation will not become zero. Therefore, even after correcting the q-axis inductance setting value Lq to the true value, further correction will occur, causing it to deviate from the true value. Furthermore, if the q-axis inductance setting value Lq becomes greater than the true q-axis inductance value, the sensorless drive system may become unstable. Therefore, by setting the predetermined deviation that will become the target of the torque deviation to a non-zero value with an offset (T...),... offset This allows us to avoid the aforementioned phenomena.

[0177] In this embodiment, based on the motor torque command value Tgen Command values ​​and engine torque command values ​​with high-frequency components removed (Teng) The deviation (torque deviation) is calculated.

[0178] Even if the engine torque command value Teng The torque generated by the engine 9 is actually a constant value, but it also includes a vibration component. The motor torque command value Tgen is calculated by the speed control unit 201 of the generator controller 2. It also exhibits pulsating components. Therefore, by applying LPF and other parameters when calculating torque deviation using the method described above, the motor torque command value Tgen can be obtained. The average value after removing vibration components and the engine torque command value Teng By making comparisons, more accurate corrections can be achieved.

[0179] The control system of the generator motor 6 in this embodiment includes: the generator motor 6; and an engine 9, which is based on the engine torque command value Teng. Driven, the generator motor 6 rotates; the speed control unit 201 causes the generator motor 6 to follow the motor speed command value N. The rotation method is based on the motor speed command value N. The estimated motor speed N' of generator motor 6 is used to generate the motor torque command value Tgen. ; and a voltage generation unit (final voltage command generation unit 205), which is based on the motor torque command value Tgen Based on the current detection values ​​(id, iq) of generator motor 6, a voltage command value (Vd) is generated to drive generator motor 6. Vq In the control system of the generator motor 6, an estimation unit (phase / speed estimation unit 212) is also included, which inputs current detection values ​​(id, iq) to a state observer (mathematical formula (14)) representing the voltage equation (mathematical formula (11)) with a predetermined phase difference (θγ) relative to the dq axis, and outputs voltage command values ​​(Vd). Vq The input is the terminal voltage (vd, vq), and the estimated values ​​of the rotor flux of the generator motor 6 (Φmd' (=Φmγ), Φmq' (=Φmδ)) are calculated. The estimated value of the phase error θγ' is calculated, and the estimated value of the motor speed N' is calculated based on the estimated value of the phase error θγ'. The voltage equation (mathematical formula (11)) represents the relationship between the current detection value (id, iq) and the terminal voltage (vd, vq) of the generator motor 6 through the dq axis. The estimation unit (phase / speed estimation unit 212) makes the motor torque command value Tgen With engine torque command value Teng The deviation (torque deviation) between them is reduced in such a way that the parameters (e.g., q-axis inductance) or the estimated phase error θγ' of the state observer (mathematical formula (14)) are corrected.

[0180] Through the above structure, the engine torque command value Teng is made possible. With motor torque command value Tgen By correcting parameters (e.g., q-axis inductance) or the estimated phase error θγ' in a way that reduces torque deviation, the estimated rotor flux (Φmd' (=Φmγ), Φmq' (=Φmδ)) (estimated values ​​of pole position) can be made closer to the true value, and the phase estimation accuracy (accuracy of the estimated phase error θγ') can be improved. Furthermore, as a result, torque accuracy is improved; therefore, when controlling the generator-motor 6 at a highly efficient operating point such as MTPA, control can be performed at the target operating point, thus improving efficiency.

[0181] The embodiments of the present invention have been described above. However, the above embodiments only show a part of the application examples of the present invention and are not intended to limit the technical scope of the present invention to the specific structure of the above embodiments.

Claims

1. A control method for a generator-motor, wherein, An engine that drives a generator motor to rotate based on engine torque command values. In a manner that causes the generator-motor to rotate in accordance with the motor speed command value, a motor torque command value is generated based on the motor speed command value and the estimated motor speed value of the generator-motor. Based on the motor torque command value and the current detection value of the generator motor, a voltage command value for driving the generator motor is generated. In the control method of this generator motor, The current detection value is input to a state observer that represents the voltage equation via a γδ axis having a predetermined phase difference relative to the dq axis of the generator motor, and the voltage command value is input as the terminal voltage of the generator motor. An estimated rotor flux value for the generator motor is calculated, thereby calculating an estimated phase error value. Based on the estimated phase error value, an estimated motor speed value is calculated. The voltage equation represents the relationship between the current detection value and the terminal voltage via the dq axis. The parameters constituting the state observer or the estimated phase error value are corrected in a manner that reduces the deviation between the motor torque command value and the engine torque command value.

2. The control method for a generator motor according to claim 1, wherein, The q-axis inductance that constitutes the parameters is corrected.

3. The control method for a generator motor according to claim 1, wherein, When correcting the parameter, the parameter is corrected by adding a first correction value, and the first correction value is increased in such a way that the deviation becomes a predetermined deviation. The amount of increase in the first correction value is set based on the deviation. When correcting the estimated phase error value, the estimated phase error value is corrected by adding a second correction value to the estimated phase error value, and the second correction value is increased in such a way that the deviation becomes the specified deviation, and the amount of increase of the second correction value is set based on the deviation.

4. The control method for a generator motor according to claim 1, wherein, When correcting the parameter, the parameter is corrected by adding a first correction value, and the first correction value is increased in stages from zero in a manner that makes the deviation a predetermined deviation. When correcting the estimated phase error, the estimated phase error is corrected by adding a second correction value to the estimated phase error, and the second correction value is increased in stages from zero in such a way that the deviation becomes the specified deviation.

5. The control method for a generator motor according to claim 1, wherein, When correcting the parameter, the parameter is corrected by adding a first correction value, and the first correction value is set in a manner that makes the deviation a predetermined deviation. When correcting the estimated phase error value, the estimated phase error value is corrected by adding a second correction value, and the second correction value is set in such a way that the deviation becomes the specified deviation.

6. The control method for a generator-motor according to any one of claims 3 to 5, wherein, The specified deviation is set to a value other than zero.

7. The control method for a generator-motor according to any one of claims 1 to 5, wherein, The deviation is calculated based on the command value obtained by removing the high-frequency component from the motor torque command value and the engine torque command value.

8. A control system for a generator motor, comprising: Generator motor; An engine, driven by an engine torque command value, causes the generator motor to rotate; A speed control unit generates a motor torque command value based on the motor speed command value and an estimated motor speed value of the generator motor, such that the generator motor rotates in accordance with the motor speed command value; and The voltage generation unit generates a voltage command value for driving the generator motor based on the motor torque command value and the current detection value of the generator motor. In the control system of this generator motor, It also includes an estimation unit that inputs the current detection value to a state observer representing the voltage equation via a γδ axis having a predetermined phase difference relative to the dq axis of the generator motor, and inputs the voltage command value as the terminal voltage of the generator motor, calculates an estimated value of the rotor flux of the generator motor, thereby calculating an estimated value of the phase error, and calculates an estimated value of the motor speed based on the estimated value of the phase error, wherein... The voltage equation represents the relationship between the current detection value and the terminal voltage via the dq axis. The estimation unit corrects the parameters constituting the state observer or the estimated phase error value in a manner that reduces the deviation between the motor torque command value and the engine torque command value.

Images