A method for controlling a switched reluctance motor using a hybrid torque distribution function
By controlling the switched reluctance motor using a hybrid torque distribution function, combined with online compensation and an exponential torque distribution function, the problems of excessive current spikes and torque ripple in traditional methods are solved, resulting in more stable motor operation.
Patent Information
- Application Number
- CN202311190805.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-15
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-09-15
AI Technical Summary
Traditional torque distribution function control of switched reluctance motors suffers from excessive current spikes and torque ripple, affecting the smooth operation of the motor.
A hybrid torque distribution function control method is adopted, which is a control system composed of a fractional sliding mode controller, an angle corrector, a hybrid torque distribution function, a cost function calculator and an inverter. By combining online compensation and an exponential torque distribution function, the torque distribution process is optimized, current spikes are reduced and torque ripples are suppressed.
It effectively reduced the peak current and decreased torque ripple by 14.5%, while the peak current was reduced by 25%, thus improving the motor's operational stability.
Smart Images

Figure CN117498727B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the control of switched reluctance motors, and in particular to a method for controlling switched reluctance motors using a hybrid torque distribution function. Background Technology
[0002] Switched reluctance motors (SRMs) offer advantages such as simple structure and control, stable performance, high cost-effectiveness, and high operating efficiency. However, the doubly salient pole structure of the SRM stator and rotor, coupled with its highly saturated magnetic flux, leads to inherent problems such as torque pulsation and large current fluctuations, hindering its development in a wider range of fields.
[0003] In the study of torque ripple and phase current in SRM stators and rotors, the commonly used Torque Distribution Function (TSF) can theoretically effectively suppress the torque ripple effect generated by the SRM stator and rotor during commutation. The principle of the traditional TSF is to use a specific distribution function to distribute the reference torque output by the speed controller to each phase winding to obtain the reference torque of each phase. Then, the torque controller is used to make the torque of each phase winding track the reference torque value of each phase winding and convert it into a reference current to control each phase winding of the motor.
[0004] However, during the commutation of the SRM phase winding, the increase or decrease in torque between the turned-on and turned-off phase windings cannot always cancel each other out to zero, causing fluctuations in the combined torque during this period. This also generates obvious current spikes, which are usually caused by changes in the nonlinear inductance of the circuit and may lead to temporary current overload. Current spikes may cause damage or malfunctions to the circuit and related equipment, so they require special attention and handling.
[0005] In the existing methods, reference [1] proposes an improved TSF method based on motor model calculation. By analyzing the static characteristics of the motor using the finite element method, the ideal shape of the TSF is determined, so that it has better current and torque tracking performance.
[0006] Reference [2] proposes an improved torque distribution function. The turn-on angle and turn-off angle of the improved torque distribution function are optimized by a genetic algorithm, which effectively reduces the torque ripple and copper loss during the operation of the switched reluctance motor.
[0007] Although references [1] and [2] both used torque distribution function (TSF) to reduce the torque ripple of SRM to varying degrees, they both generated large current spikes in the initial stage of commutation. The peak value of the current spike was about twice the stable value, which greatly wasted the load capacity of the switching device and caused hidden dangers to the smooth operation of the motor system.
[0008] References[1]Hamouda M,Szamel L.Compensation of switched reluctancemotor torque ripple based on TSF strategy for electric vehicle applications[C] / / International Conference on Power Generation Systems and RenewableEnergy Technologies(PGSRET).Islamabad,Pakistan,2018:1-6.
[0009] References [2] Chen Hao, Dong Linjie, Pan Jianfei, et al. Improved torque distribution function control method for switched reluctance motor [P]. Jiangsu Province: CN114826035A, 2022-07-29. Summary of the Invention
[0010] This invention addresses the problems of excessive current spikes and torque ripple in traditional TSF-controlled switched reluctance motors by designing a method for controlling switched reluctance motors using a hybrid torque distribution function. The aim of this method is to effectively suppress excessive current spikes while reducing the impact of torque ripple in traditional TSF-controlled switched reluctance motors.
[0011] The technical solution adopted in this invention is as follows: A method for controlling a switched reluctance motor using a hybrid torque distribution function employs a control system composed of a switched reluctance motor, a fractional-order sliding mode controller, an angle corrector, a hybrid torque distribution function, a cost function calculator, a torque-angle characteristic table, an inverter, and a subtraction operator. The control process involves the following steps:
[0012] Step 1: Differentiate the actual rotational speed ω obtained by differentiating the electrical angle θ fed back from the switched reluctance motor using a differentiator, and compare it with the reference rotational speed ω. ref Input the subtraction operator and output the rotational speed error.
[0013] Step 2: Input the speed error obtained in Step 1 into the fractional-order sliding mode controller, and output the reference torque T. ref .
[0014] Step 3: Input the electrical angle θ fed back from the switched reluctance motor into the angle corrector to obtain the rotor angle θ(n) in the nth cycle.
[0015] Step 4: Design a hybrid torque distribution function, which will distribute the reference torque T obtained in Step 2. refThe rotor angle θ(n) obtained in step 3 within the nth cycle is input into the hybrid torque distribution function, based on the commutation zone separation point θ. eq The hybrid torque distribution function is divided into segmented intervals: segment I, segment II, segment III, and segment IV; then, during the commutation stage, θ... ov The cosine torque distribution function is optimized using an online compensation method in segments II and III of the segmented interval. The cosine torque distribution function is compensated using an exponential torque distribution function in segments I and IV of the segmented interval, resulting in a hybrid torque distribution function that outputs the actual control torque T(n). Finally, the online compensated hybrid torque distribution function f(θ) is obtained.
[0016] Step 5: Construct the cost function. Determine the weighting coefficients of the cost function according to the torque formula to obtain a new cost function J′, which is then used by the cost function calculator for calculation.
[0017] Step 6: Input the current signal i, electrical angle θ and flux linkage signal ψ fed back in real time from the switched reluctance motor into the torque-angle characteristic table, and output the current i(n) and rotor angle θ(n) in the nth cycle, and the phase current i(n+1) and rotor angle θ(n+1) in the (n+1)th cycle.
[0018] Step 7: Input the torque T(n) obtained in Step 4, the phase current i(n) and rotor angle θ(n) in n cycles obtained in Step 6, and the phase current i(n+1) and rotor angle θ(n+1) in n+1 cycles into the cost function calculator described in Step 5 to calculate and output the optimal current control quantity.
[0019] Step 8: Input the optimal current control value obtained in Step 7 into the inverter to control the three-phase switch, and then control the switched reluctance motor to start rotating, so as to suppress torque pulsation and reduce current spikes.
[0020] The beneficial effects of this invention are as follows: Addressing the problems of excessive current spikes and torque ripple in traditional torque distribution function (TSF) control of switched reluctance motors, this invention employs a designed hybrid torque distribution function (HTSF). By introducing an exponential torque distribution function, it improves torque tracking accuracy while compensating for torque errors to obtain torque T(n). Then, a cost function is designed and weighting coefficients are determined. Using predictive control, with torque T(n) as a reference, the optimal current control quantity is selected to control the switched reluctance motor, thereby reducing current spikes and suppressing torque ripple. Compared with existing technologies, this method significantly reduces the peak value of current spikes while effectively reducing torque ripple. Attached Figure Description
[0021] Figure 1 This is a block diagram of the control system principle of the present invention;
[0022] Figure 2 for Figure 1 Output results and segmented interval diagram of the mixed torque distribution function;
[0023] Figure 3 This is a flowchart illustrating the calculation of the optimal current control quantity in the implementation of this invention;
[0024] Figure 4 A schematic diagram of the output torque data for a traditional torque distribution function-controlled SRM.
[0025] Figure 5 This is a schematic diagram of the output torque data of the hybrid torque distribution function control SRM in an embodiment of the present invention;
[0026] Figure 6 The current waveform diagram for SRM controlled by the traditional torque distribution function;
[0027] Figure 7 This is a current waveform diagram of the hybrid torque distribution function controlling the SRM in an embodiment of the present invention. Detailed Implementation
[0028] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0029] Terminology Explanation: 1. Switched Reluctance Motor (SRM): Compared to traditional DC motors or induction motors, SRMs do not have permanent magnets; instead, they utilize magnetic reluctance to generate torque. SRMs offer advantages in certain applications, such as those operating in high-temperature environments or those requiring high reliability.
[0030] 2. Hybrid Torque Distribution Function (HTSF): The hybrid torque distribution function is a control strategy used to determine how much torque each phase (stator winding) should output. Due to the special structure and working principle of the SRM, the goal of the hybrid torque distribution function is to balance the torque output of each phase throughout the entire rotor rotation cycle to achieve smooth operation, i.e., a relatively stable control current.
[0031] 3. Current spike: refers to the phenomenon in a circuit where the instantaneous current suddenly increases to a higher peak value.
[0032] like Figure 1 , 2 As shown in Figure 3, this method employs a control system consisting of a switched reluctance motor, a fractional-order sliding mode controller (FOSMC), an angle corrector, a hybrid torque distribution function (HTSF), a cost function calculator, a torque-angle characteristic table, an inverter, and a subtraction operator. The control process involves the following steps:
[0033] Step 1: Differentiate the actual rotational speed ω obtained by differentiating the electrical angle θ fed back from the switched reluctance motor using a differentiator, and compare it with the reference rotational speed ω. refInput the subtraction operator and output the rotational speed error.
[0034] Step 2: Input the speed error obtained in Step 1 into the fractional-order sliding mode controller, and output the reference torque T. ref .
[0035] Step 3: Input the electrical angle θ fed back from the switched reluctance motor into the angle corrector to obtain the rotor angle θ(n) in the nth cycle.
[0036] Step 4: Design the Hybrid Torque Distribution Function (HTSF) to distribute the reference torque T obtained in Step 2. ref The rotor angle θ(n) obtained in step 3 within the nth cycle is input into the hybrid torque distribution function, based on the commutation zone separation point θ. eq The hybrid torque distribution function is divided into segmented intervals: segment I, segment II, segment III, and segment IV; then, during the commutation stage, θ... ov The cosine torque distribution function is optimized using an online compensation method in segments II and III of the segmented interval. The cosine torque distribution function is compensated using an exponential torque distribution function in segments I and IV of the segmented interval, resulting in a hybrid torque distribution function that outputs the actual control torque T(n). Finally, the online compensated hybrid torque distribution function f(θ) is obtained.
[0037] like Figure 2 As shown, in step 4, different segmented intervals are divided during the commutation stage to perform online compensation and cosine-type compensation respectively, resulting in the hybrid torque distribution function f(θ):
[0038]
[0039] In equation (1), τ r It is a single-phase period; θ on θ is the commutation zone opening angle; off The turn-off angle in the commutation region; Δf * The online compensation value; Δf h This is the compensation value for the exponential torque distribution function.
[0040] Then the reference torque T obtained in step 2 ref The rotor angle input and mixed torque distribution function in step 2 are used to calculate the actual control output. The operating principle of the switched reluctance motor is to generate torque by switching the energized phase windings inside the motor to produce a rotating magnetic field, which drives the motor to rotate. Let the currently energized phase winding be f. k (θ); f k-1 (θ) represents the previous phase winding; f k+1 (θ) represents the next phase winding, and the process of switching the power on and off between the two phases is the commutation stage. ov θ ov The overlap angle between the two-phase on / off states also represents the commutation stage of the two-phase windings.
[0041] During the commutation stage θ ov With the separation point θ eq Divide the system into different segmented intervals I, II, III, and IV, with the separation point θ during the commutation stage. eq This is the point where the actual torque equals the reference torque.
[0042] First, for sections II and III, the difference in the rate of change of inductance between the open-phase winding and the closed-phase winding is used as the compensation Δf. * This allows the compensation gain to be adaptively adjusted based on the relative positions of the on-phase and off-phase windings, achieving online compensation through the improved torque distribution function.
[0043] Next, the compensation value Δf is corrected online. * This improves torque ripple but introduces small current spikes. Since exponential functions are sensitive to parameter changes, and to reduce the actual torque tracking error caused by nonlinear changes in inductance in segments I and IV, an exponential function-type TSF is introduced as positive compensation for segment I and negative compensation value Δf for segment IV. h This improves the real-time torque tracking accuracy and uses fractional sliding mode control to optimize the compensation value, thus mitigating current spikes.
[0044] Step 5: Construct the cost function. Determine the weighting coefficients m, r, s, and t of the cost function according to the torque formula to obtain a new cost function J′, which is then used by the cost function calculator for calculation.
[0045] In step 5, according to the torque formula The constructed cost function J is:
[0046]
[0047] In equation (2): m, r, s, and t are the weight coefficients of the cost function;
[0048] i k (θ n This represents the current in the current phase winding at the current moment, in amperes (A).
[0049] i k (θ n+1 () represents the phase current of the current phase winding at the next moment, in A;
[0050] i k+1 (θ n () represents the phase current of the next phase winding at the current moment, in A;
[0051] i k+1 (θ n+1() represents the phase current of the next phase winding at the next moment, in A.
[0052] The torque generated during the commutation phase is the sum of the torques of the two adjacent phases, which is the reference torque T. ref :
[0053]
[0054] In equation (3): T ref Reference torque, unit: Nm;
[0055] L(θ,i k (θ n+1 L(θ,i) k+1 (θ n+1 These are the inductance windings of the previous phase and the next phase at the next moment, respectively, in units of H;
[0056] i k (θ n+1 ), i k+1 (θ n+1 These are the phase currents of the current phase winding and the next phase winding at the next moment, respectively, in A.
[0057] Combining equations (2) and (3), let the Lagrange function be F:
[0058]
[0059] In equation (4), λ is the Lagrange multiplier in the Lagrange function F. For the phase current i of the current phase at the next moment in equation (4)... k (θ n+1 ), the phase current i of the next phase at the current moment k+1 (θ n After taking the partial derivatives of the electrical angle θ and setting them to zero, we obtain formulas (5), (6), and (7):
[0060]
[0061]
[0062]
[0063] Assuming the electrical angle difference Δθ between the current moment and the next moment is sufficiently small, then the phase current i of the current phase winding at the current moment in formula (4) is... k (θ n ) and the current phase current i at the next moment k (θ n+1 Approximate equality can be expressed as Formulas (5)(6)(7) give us formulas (8)(9)(10):
[0064]
[0065]
[0066]
[0067] The above formulas are all functions of the electrical angle θ, meaning the weighting coefficients of the cost function change with the electrical angle θ. When m = r = 1, the new cost function J′ is obtained as:
[0068]
[0069] In equation (11): λ = 2;
[0070] Step 6: Input the torque-angle characteristic table using the real-time feedback current signal i, electrical angle θ and flux linkage signal ψ of the switched reluctance motor, and output the current i(n) and rotor angle θ(n) in the nth cycle, and the phase current i(n+1) and rotor angle θ(n+1) in the (n+1)th cycle.
[0071] Step 7: Input the torque T(n) obtained in Step 4, the phase current i(n) and rotor angle θ(n) in n cycles obtained in Step 6, and the phase current i(n+1) and rotor angle θ(n+1) in n+1 cycles into the cost function calculator in Step 5 to calculate and output the optimal current control quantity.
[0072] like Figure 3 As shown, in step 7, the process for calculating the optimal current control quantity is as follows: The phase current i(n), rotor angle θ(n), and torque T(n) of the switched reluctance motor in the nth cycle are sampled and acquired. The optimal current control quantity is then calculated using the torque-angle characteristic table and the cost function calculator. First, in the nth control cycle, for the torque T(n) output by the hybrid torque distribution function, 2α / β alternative input torques T(n) are generated, centered on T(n), with [-α, α] as the boundary and β as the increment. * Then, the current state of the switched reluctance motor is sampled to obtain the phase current i(n) and rotor angle θ(n) information within the nth cycle; then, under different alternative input torques T(n)... *By combining the phase current i(n) and rotor angle θ(n) information in the nth cycle, the phase current i(n+1) and rotor angle θ(n+1) in the n+1th cycle are obtained; then, based on the data of the phase current i(n+1) and rotor angle θ(n+1) in the torque-angle characteristic table corresponding to the n+1th cycle, the torque T(n+1) in the n+1th cycle is obtained; finally, the obtained phase current i(n) and rotor angle θ(n) in the nth cycle and the phase current i(n+1) and rotor angle θ(n+1) in the n+1th cycle are input into the cost function calculator to calculate and output the optimal current control quantity.
[0073] Step 8: Input the optimal current control value obtained in Step 7 into the inverter to control the three-phase switch, and then control the switched reluctance motor to start rotating, so as to suppress torque pulsation and reduce current spikes.
[0074] The example data is as follows:
[0075] 1. Different reference speeds ω of the switched reluctance motor ref The switched reluctance motor is controlled by the input hybrid torque distribution function (HTSF). In this embodiment, the data provides different reference speeds ω for the switched reluctance motor. ref The values are 200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800, 2000, 2200, 2400, and 2600, respectively, with the unit being revolutions per minute (r / min).
[0076] 2. The reference torque T given in this embodiment ref The value is 8, and the unit is Newton-meter (Nm). The output torque T(n) is 8.005, 8.008, 8.013, 8.017, 8.022, 8.024, 8.032, 8.04, 8.017, 8.047, 8.045, and 8.032, in Newton-meter (Nm).
[0077] 3. Then, the optimal current control quantity is obtained through sampling and cost function calculation, resulting in the final reduced current peak value I. peak The values are 36.52, 35.07, 33.96, 33.22, 32.61, 32, 31.73, 31.37, 31.15, 30.81, 30.75, 30.58, and 30.32, respectively, in amperes (A).
[0078] This embodiment uses torque ripple and current spikes to represent system performance. A comparison of the torque and current control effects of two control methods—a hybrid torque distribution function and a traditional torque distribution function—for controlling a switched reluctance motor is presented. Figure 4 , Figure 5 , Figure 6 , Figure 7 As shown.
[0079] The hybrid torque distribution function method for controlling a switched reluctance motor proposed in this invention effectively reduces torque ripple while significantly lowering current peak values. The data results are shown in Table 1. Compared to the traditional torque distribution function method, the hybrid torque distribution function reduces torque ripple by 14.5% while lowering current peak values by 25%.
[0080] Table 1 Performance Comparison
[0081]
Claims
1. A method for controlling a switched reluctance motor using a hybrid torque distribution function, characterized in that, A control system consisting of a switched reluctance motor, a fractional-order sliding mode controller, an angle corrector, a hybrid torque distribution function, a cost function calculator, a torque-angle characteristic table, an inverter, a subtractor, and a differentiator controls the switched reluctance motor using the hybrid torque distribution function. The control process includes the following steps: Step 1, the actual speed ω obtained by differentiating the electrical angle θ fed back by the switched reluctance motor through a differentiator is compared with the reference speed ω ref The speed error is input into the subtractor, and the speed error is output. Step 2: Input the speed error obtained in Step 1 into the fractional-order sliding mode controller, and output the reference torque T. ref ; Step 3: Input the electrical angle θ fed back from the switched reluctance motor into the angle corrector to obtain the rotor angle θ(n) in the nth cycle; Step 4: Design a hybrid torque distribution function, which will distribute the reference torque T obtained in Step 2. ref The rotor angle θ(n) obtained in step 3 within the nth cycle is input into the hybrid torque distribution function, based on the commutation zone separation point θ. eq The hybrid torque distribution function is divided into segmented intervals: segment I, segment II, segment III, and segment IV; then, during the commutation stage, θ... ov The cosine torque distribution function is optimized using an online compensation method in segments II and III of the segmented interval. The cosine torque distribution function is compensated using an exponential torque distribution function in segments I and IV of the segmented interval, resulting in a hybrid torque distribution function that outputs the actual control torque T(n). Finally, the online compensated hybrid torque distribution function f(θ) is obtained. Step 5: Construct the cost function. Determine the weighting coefficients m, r, s, and t of the cost function according to the torque formula to obtain a new cost function J′, which is then used by the cost function calculator for calculation. Step 6: Input the current signal i, electrical angle θ and flux linkage signal ψ fed back in real time from the switched reluctance motor into the torque-angle characteristic table, and output the current i(n) and rotor angle θ(n) in the nth cycle, and the phase current i(n+1) and rotor angle θ(n+1) in the (n+1)th cycle. Step 7: Input the torque T(n) obtained in Step 4 and the phase current i(n) and rotor angle θ(n) in n cycles and the phase current i(n+1) and rotor angle θ(n+1) in n+1 cycles obtained in Step 6 into the cost function calculator described in Step 5 to calculate and output the optimal current control quantity; Step 8: Input the optimal current control value obtained in Step 7 into the inverter to control the three-phase switch, and then control the switched reluctance motor to start rotating, so as to suppress torque pulsation and reduce current spikes.
2. The method for controlling a switched reluctance motor using a hybrid torque distribution function according to claim 1, characterized in that, In step 4, during the commutation stage, different segmented intervals are divided for online compensation and cosine-type compensation, resulting in the hybrid torque distribution function f(θ): In equation (1), f(θ) represents the torque distribution function; τ r It is a single-phase period; θ on θ is the commutation zone opening angle; off The commutation region turn-off angle; Δf * The online compensation value; Δf h The compensation value is for the exponential torque distribution function; Then the reference torque T obtained in step 2 ref Combine the rotor angle input with the mixed torque distribution function from step 2 to calculate the actual control torque value T(n).
3. The method for controlling a switched reluctance motor using a hybrid torque distribution function according to claim 2, characterized in that, In step 5, according to the torque formula The constructed cost function J is: In equation (2): m, r, s, and t are the weight coefficients of the cost function; i k (θ n This represents the current in the current phase at the current moment, in amperes (A). i k (θ n+1 () represents the phase current of the current phase at the next moment, in A; i k+1 (θ n () represents the phase current of the next phase at the current moment, in A; i k+1 (θ n+1 () represents the phase current of the next phase at the next moment, in A. The torque generated during the commutation phase is the sum of the torques of the two adjacent phases, which is the reference torque T. ref : In equation (3): T ref Reference torque, unit: Nm; L(θ,i k (θ n+1 L(θ,i) k+1 (θ n+1 These are the inductance windings of the previous phase and the next phase at the next moment, respectively, in units of ohms (H). i k (θ n+1 ), i k+1 (θ n+1 These are the phase currents of the current phase winding and the next phase winding at the next moment, respectively, in A; Combining equations (2) and (3), let the Lagrange function be F: For the phase current i of the current phase at the next moment in equation (4) k (θ n+1 ), the phase current i of the next phase at the current moment k+1 (θ n After taking the partial derivatives of the electrical angle θ and setting them to zero, we obtain formulas (5), (6), and (7): Assuming the electrical angle difference Δθ between the current time and the next time is sufficiently small, then the phase current i of the current phase at the current time in formula (4) is... k (θ n ) and the current phase current i at the next moment k (θ n+1 Approximate equality can be expressed as Formulas (5)(6)(7) give us formulas (8)(9)(10): The weighting coefficients of the cost function change with the electrical angle θ. When m = r = 1, the new cost function J′ is obtained as follows: In equation (11): λ = 2; 4. The method for controlling a switched reluctance motor using a hybrid torque distribution function according to claim 3, characterized in that, In step 7, the process for calculating the optimal current control quantity is as follows: The phase current i(n), rotor angle θ(n), and torque T(n) of the switched reluctance motor in the nth cycle are sampled and acquired. The optimal current control quantity is then calculated using the torque-angle characteristic table and the cost function calculator. First, in the nth control cycle, for the torque T(n) output by the hybrid torque distribution function, 2α / β alternative input torques T(n) are generated, centered at T(n), with [-α, α] as the boundary and β as the increment. * Then, the current state of the switched reluctance motor is sampled to obtain the phase current i(n) and rotor angle θ(n) information within the nth cycle; then, different alternative input torques T(n) are selected. * By combining the phase current i(n) and rotor angle θ(n) information in the nth cycle, the phase current i(n+1) and rotor angle θ(n+1) in the n+1th cycle are obtained; then, based on the data of the phase current i(n+1) and rotor angle θ(n+1) in the torque-angle characteristic table corresponding to the n+1th cycle, the torque T(n+1) in the n+1th cycle is obtained; finally, the obtained phase current i(n) and rotor angle θ(n) in the nth cycle and the phase current i(n+1) and rotor angle θ(n+1) in the n+1th cycle are input into the cost function calculator to calculate and output the optimal current control quantity.
Citation Information
Patent Citations
Improved torque distribution function control method for switched reluctance motor
CN114826035A
