An optimized design method for a starting motor based on engine load matching

Abstract

The present invention provides an optimized design method for a starting motor based on engine load matching, which includes the following steps: ① Preliminary modeling: Establish a preliminary model of a brushless DC motor according to the parameter requirements of the engine electrical system; ② Determine the no-load condition: Analyze the no-load condition of the motor by using the electromagnetic finite element field-circuit coupling method according to the rated voltage value; ③ Optimize the torque coefficient: Maximize the torque coefficient by controlling the no-load speed in the electromagnetic scheme near the maximum disengagement speed n T to optimize the volume and weight to the greatest extent; ④ Set the relationship between load torque and time: Calculate the limit value of the phase current, fit and calculate the relationship parameters between the engine turbine load data and the speed, and thus calculate the engine turbine load torque; ⑤ Calculate the starting performance. The present invention can achieve the optimal volume and weight of the starting motor design scheme under the condition of satisfying that the bus current does not exceed the limit, the starting time and the whole-process performance curve can be predicted, and the starting reliability is improved.

Description

Technical Field

[0001] The present invention relates to an optimized design method for a starting motor based on engine load matching. Background Art

[0002] The current international situation is extremely severe, turbulent and unstable. The reliability of national defense weapons not only largely determines the outcome of a war, but also is the main source of people's sense of security. Whether it is an aircraft or a hypersonic weapon, their main power source is a gas turbine engine. And the turbine engine needs a starting motor to drive it to the ignition speed during each ignition starting process, and there are strict requirements for the starting time. The starting processes of traditional gas turbine APUs and compressed air starters are uncontrollable, with low starting efficiency, large and complex structures, and it is even more difficult to guarantee reliability, which easily leads to the failure of engine ignition starting at critical moments, causing irreparable consequences. With the next-generation multi-electric / full-electric upgrade of turbine engines, as an efficient and reliable auxiliary power device for engine ignition starting, the starting motor has gradually become an essential key device for international advanced turbine engines, conforming to the mainstream development trend of the electrification of the next-generation advanced aircraft. However, the existing aviation starting motors have the following problems:

[0003] 1) Independently designed regardless of load characteristics, with low starting efficiency / utility rate

[0004] The existing design of starting motors is in its infancy, without normative documents for guidance. Generally, it is independently designed according to the rated operating point (a speed-torque point), and it is impossible to achieve performance matching with the engine turbine. Its optimal operating point is the rated speed, while the speed and load during the ignition starting process of the turbine engine are constantly changing. The optimal operating point designed for the starting motor often deviates from the ignition speed, resulting in low starting efficiency and performance utilization rate of the starting motor, and often occupying more volume and weight to ensure the starting effect.

[0005] 2) The starting time cannot be predicted, and the starting success rate is low

[0006] The existing design mode of starting motors is difficult to calculate the starting time after driving the engine. And the ignition starting process of the turbine engine is a complex non-linear process. During the engine starting process, the speed of the starting motor is constantly increasing, and the turbine load is constantly changing. Since the battery voltage is constant, the output torque of the motor continuously decreases during the starting process. At the critical moment of engine ignition, starting failure often occurs due to insufficient speed.

[0007] 3) It is difficult to calculate the performance data throughout the starting process, and the optimization target cannot be obtained

[0008] During actual starting operation, when the starting motor drives the large-inertia engine turbine to accelerate, the rotational speed rises slowly, and the turbine load also changes continuously. Therefore, during the starting process, data such as the current and output torque of the starting motor are largely determined by the engine turbine load and the current rotational speed. However, the traditional design mode can only consider the rated point performance when the starting motor operates independently, unable to obtain the performance data of the entire starting process, lacking optimization objectives and targeted optimization methods.

[0009] How to achieve the matching design of the starting motor and the engine turbine load characteristics, make the volume and weight of the starting motor optimal, predict the performance parameter curves during the entire starting process to obtain optimization objectives, and achieve optimization through adjusting the control strategy and the motor structure size to improve the ignition starting success rate is still a technical problem faced in the current industry. Summary of the Invention

[0010] To solve the above technical problems, the present invention provides an optimized design method for a starting motor based on engine load matching. The optimized design method for a starting motor based on engine load matching can achieve the optimal volume and weight of the starting motor design scheme under the condition of ensuring that the bus current does not exceed the limit, the starting time and the performance curve of the whole process can be predicted, and the starting reliability is improved.

[0011] The present invention is achieved through the following technical solutions.

[0012] An optimized design method for a starting motor based on engine load matching provided by the present invention includes the following steps:

[0013] ① Preliminary modeling: Establish a preliminary model of a brushless DC motor according to the parameter requirements of the engine electrical system;

[0014] ② Determine the no-load condition: Analyze the no-load condition of the motor by using the electromagnetic finite element field-circuit coupling method according to the rated voltage value;

[0015] ③ Optimize the torque coefficient: Judge the ratio of the no-load speed n0 to the maximum disengagement speed n of the starting motor during the engine ignition starting process. If the ratio is less than the preset low value or greater than the preset high value, return to step ② after trying to change the parameters. If the ratio is between the preset low value and the preset high value, enter step ④; changing the parameters means changing the magnetic conductive material and / or adjusting the main electromagnetic parameters; T ④ Set the relationship between load torque and time: Calculate the phase current limit value, fit and calculate the relationship parameters between the engine turbine load data and the rotational speed, thereby calculate the engine turbine load torque, and obtain the characteristics of the engine turbine load torque changing with the rotational speed;

[0016]

[0017] ​⑤ Starting performance calculation: Use finite element simulation to calculate starting performance parameters and verify them to obtain the determined main electromagnetic parameters and the limit value of phase current.

[0018] The adjustment of the main electromagnetic parameters is carried out by at least one of the following methods: reducing the number of slots, reducing the number of poles, reducing the number of turns of the coil, reducing the armature diameter, increasing the air gap, and shortening the armature length.

[0019] In step ⑤, the starting performance parameters include the engine turbine speed value, the motor busbar current value, the turbine load value, and the motor output electromagnetic torque value.

[0020] In step ④, calculate the limit value of phase current. By calculating the phase current limit value when the motor thermal load is less than 960A 2 mm 3 the highest output torque is obtained.

[0021] In step ④, the fitting calculation is carried out by polynomial fitting.

[0022] The engine turbine speed value is calculated by integrating the mechanical angular acceleration of the engine turbine with respect to time.

[0023] The motor busbar current value is equivalently calculated by the following formula:

[0024]

[0025] where U is the rated voltage of the cabin battery, R DC is the cable resistance, R1 and L1 are the phase resistance and phase inductance of the motor respectively, f k is the switching frequency of the power transistor, p is the number of pole pairs of the motor, and Ψ m is the magnetic flux of the motor phase winding.

[0026] The polynomial fitting is carried out by interpolation method to smooth the curve and then fit, and the fitting order is above the third order.

[0027] The preset low value is 1.05 and the preset high value is 1.10.

[0028] The beneficial effects of the present invention are as follows: Based on the preliminary design of the permanent magnet synchronous motor by combining the field and circuit, the key function variables affecting the starting performance can be analyzed. While meeting the busbar current limit and the engine disengagement speed, the starting ability can be maximally improved, and the volume and weight of the starting motor can be optimized. At the same time, by using the method of solving the nonlinear equations or software simulation, the curves of starting speed - time, electromagnetic torque - time, busbar voltage - time, busbar current - time, etc. can be calculated, so that the starting performance of the whole system can be predicted, and the starting reliability of the whole system can be effectively improved. Brief Description of the Drawings

[0029] Figure 1 is a schematic flow chart of at least one embodiment of the present invention;

[0030] Figure 2 is a performance curve graph of the starting process of Scheme 1 for comparison in at least one embodiment of the present invention;

[0031] Figure 3 is a performance curve graph of the starting process of Scheme 2 for comparison in at least one embodiment of the present invention;

[0032] Figure 4 is a simulation model framework diagram for the calculation in the implementation process of the present invention;

[0033] Figure 5 is a three-phase current curve graph of Scheme 1 under amplitude limiting control;

[0034] Figure 6 is an electromagnetic torque - time curve graph of the starting process of Scheme 1. Specific Embodiments

[0035] The technical solution of the present invention will be further described below, but the claimed scope is not limited thereto.

[0036] Example 1

[0037] As Figures 1 to 6 shown, an optimized design method for a starting motor based on engine load matching includes the following steps:

[0038] ① Preliminary modeling: Establish a preliminary model of a brushless DC motor according to the parameter requirements of the engine electrical system;

[0039] ② Determine the no-load condition: Analyze the no-load condition of the motor by using the electromagnetic finite element field - circuit coupling method according to the rated voltage value;

[0040] ③ Optimize the torque coefficient adjustment: Judge the ratio of the no-load speed n0 to the maximum disengagement speed n T of the starting motor during the engine ignition starting process. If this ratio is less than a preset low value or greater than a preset high value, then return to step ② after trying to change the parameters. If this ratio is between the preset low value and the preset high value, then enter step ④; Changing the parameters means changing the magnetic conductive material and / or adjusting the main electromagnetic parameters;

[0041] ④ Set the load torque - time relationship: Calculate the phase current limit value, fit and calculate the relationship parameters between the engine turbine load data and the speed, thereby calculate the engine turbine load torque, and obtain the variation characteristics of the engine turbine load torque with the speed;

[0042] ⑤ Calculate the starting performance: Use finite element simulation to calculate and verify the starting performance parameters to obtain the determined main electromagnetic parameters and the phase current limit value.

[0043] Example 2

[0044] Based on Example 1, adjust the main electromagnetic parameters by at least one of the following methods: reducing the number of slots, reducing the number of poles, reducing the number of turns of the coil, reducing the armature diameter, increasing the air gap, and shortening the armature length.

[0045] Example 3

[0046] Based on Example 1, in step ⑤, the starting performance parameters include the engine turbine speed value, the motor bus current value, the turbine load value, and the motor output electromagnetic torque value.

[0047] Example 4

[0048] Based on Example 1, in step ④, calculate the phase current limit value, which is obtained by calculating the phase current limit value with the highest output torque when the motor thermal load is less than 960A 2 mm 3 at the time.

[0049] Example 5

[0050] Based on Example 1, in step ④, the fitting calculation is performed by polynomial fitting.

[0051] Example 6

[0052] Based on Example 3, the engine turbine speed value is calculated by integrating the mechanical angular acceleration of the engine turbine with respect to time.

[0053] Example 7

[0054] Based on Example 3, the motor bus current value is calculated equivalently using the following formula:

[0055]

[0056] where U is the rated voltage of the nacelle battery, R DC is the cable resistance, R1 and L1 are the motor phase resistance and phase inductance respectively, f k is the power transistor switching frequency, p is the number of motor pole pairs, Ψ m is the motor phase winding magnetic flux, J e is the engine turbine moment of inertia.

[0057] Example 8

[0058] Based on Example 5, the polynomial fitting is performed by interpolation to smooth the curve and then fitting, and the fitting order is above the third order.

[0059] Example 9

[0060] Based on Example 1, the preset low value is 1.05 and the preset high value is 1.10.

[0061] Example 10

[0062] Based on the above embodiments, the following steps are specifically adopted:

[0063] Step 1: According to the specific requirements put forward by the engine electrical system for the starting motor, the brushless DC motor is preliminarily designed by the magnetic circuit method. The type of the starting motor is selected as the brushless DC motor, and the driving mode is selected as the star three-phase six-state square wave drive. Under the 28VDC low-voltage power supply system specified in GJB181B "Aircraft Power Supply System", the starting motor system under this motor type and driving mode has the highest power supply voltage utilization rate and the best starting performance;

[0064] Step 2: According to the rated voltage value of the cabin battery, the electromagnetic finite element field-circuit coupling method is used to analyze the no-load condition of the motor under the star three-phase six-state square wave drive (no-load speed n0, no-load back electromotive force E0, phase resistance R1, phase inductance L1, air-gap magnetic density B δ etc.);

[0065] Step 3: Design the no-load speed of the starting motor to be at the maximum disengagement speed n T of the accessory during the engine ignition starting process, which can improve the output performance and accelerate the starting speed; Compare the no-load speed n0 obtained in step S2 with the maximum disengagement speed n T of the starting motor during the engine ignition starting process. If, under the rated voltage, the no-load speed of the current motor scheme is close to 1.05 times the maximum disengagement speed n T during the engine ignition starting process, then jump to step 4. If the no-load speed is greater than 1.1 times n T , then jump to step 3.1. If the no-load speed n0 of the current scheme is less than n T , then jump to step 5; (This process is to ensure the starting performance of the starting motor in the high-speed section, avoid becoming an engine load and affecting the engine ignition success rate, and at the same time make the ratio of the back electromotive force coefficient to the torque coefficient of the motor reach the best)

[0066] Step 3.1: Optimize the magnetic conductive material and adjust the main electromagnetic parameters (by methods such as increasing the number of slots, increasing the number of poles, increasing the number of coil turns, increasing the armature diameter, reducing the air gap, increasing the armature length, etc.) to increase the torque coefficient and improve the torque-current ratio. After completing the scheme optimization, return to step 2;

[0067] Step 3.2: Change the magnetic conductive material and adjust the main electromagnetic parameters (by methods such as reducing the number of slots, reducing the number of poles, reducing the number of coil turns, reducing the armature diameter, increasing the air gap, shortening the armature length, etc.) to reduce the no-load back electromotive force coefficient, reduce the winding inductance and resistance values. After completing the scheme optimization, return to step 2;

[0068] Step 4: The starting motor is driven and controlled by a square wave. To meet the requirements of the bus DC current limit and prevent the power tubes from burning out, it is necessary to limit the phase current. The specific adjustment method is phase current hysteresis control, which controls the phase current value between the upper and lower limit amplitudes. When the phase current exceeds the upper limit amplitude, the voltage is controlled by reducing the duty cycle of the PWM to bring the phase current back within the required range. When the phase current is lower than the lower limit amplitude, the voltage is controlled by increasing the duty cycle of the PWM to bring the phase current back within the required range. The following method is used to determine the phase current limit amplitude:

[0069] Step 4.1: Determine the maximum allowable current I that the power tube can pass through according to the model manual M , and the initial limit amplitude I max Take The hysteresis width is taken as 0.02I max ; Calculate the motor thermal load according to the following formula:

[0070]

[0071] In the formula: A is the motor electric load, J is the current density, m is the number of motor phases, N is the total number of conductors in the armature winding, I max is the current limit amplitude, a is the number of parallel branches of the motor, D i1 is the inner diameter of the armature, N t is the number of parallel windings, and d is the wire diameter.

[0072] If the thermal load is greater than 960A 2 mm 3 Then reduce the phase current limit amplitude until it approaches this value (this step is to maximize the effective value of the phase current and increase the output torque);

[0073] Step 5: Since the engine load torque changes non-linearly and dynamically, the polynomial fitting method is used to fit the provided engine turbine load data to the speed to obtain the engine turbine load torque as:

[0074] T L (t) = an(t) 3 + bn(t) 2 + cn(t) + T0 (2)

[0075] Where T L is the engine turbine load torque, a, b, and c are the polynomial coefficients obtained by fitting, and T0 is the static braking torque of the turbine. This data can be measured by using a torque-speed sensor during the cold operation of the generator. The fitting method can use interpolation to make the curve smoother and then perform the fitting. The fitting order includes but is not limited to 3 times, and the fitting method includes but is not limited to polynomial fitting. Perform the corresponding fitting according to the actually collected turbine load torque data.

[0076] Step 6: Starting performance calculation

[0077] Under the two-phase conduction star-connected three-phase six-state drive strategy, the brushless permanent magnet starting motor outputs an electromagnetic torque T EM The magnitude is:

[0078]

[0079] In the formula: T EM is the electromagnetic torque of the motor, K av is the chopping coefficient, p is the number of pole pairs of the motor, Ψ m is the magnetic flux of the motor phase winding, and I(t) is the phase current varying with time.

[0080] When adopting the PWM phase-limited current amplitude control strategy, due to the influence of the switching frequency, the motor phase current waveform is not a standard rectangular wave, and the motor output torque will be corrected according to the chopping coefficient K av as follows:

[0081]

[0082] In the formula: K av is the chopping coefficient, R1 and L1 are the motor phase resistance and phase inductance respectively, f k is the control switching frequency of the power transistor.

[0083] The rotational speed during the starting process is the integral of the mechanical angular acceleration of the engine turbine with respect to time:

[0084]

[0085] In the formula: n(t) is the motor rotational speed varying with time, Ω is the motor mechanical angular velocity, K av is the chopping coefficient, p is the number of pole pairs of the motor, Ψ m is the magnetic flux of the motor phase winding, I(t) is the bus current varying with time, T L (t) is the engine turbine load torque varying with time, and J e is the engine turbine moment of inertia.

[0086] The bus current value during the starting process is equivalently calculated according to the following formula:

[0087]

[0088] Among them, U is the rated voltage of the cabin battery, R DC is the cable resistance, R1 and L1 are the motor phase resistance and phase inductance respectively, f k is the power transistor switching frequency, p is the number of pole pairs of the motor, Ψ m is the magnetic flux of the motor phase winding, and J e is the engine turbine moment of inertia.

[0089] By solving the non - linear equations by simultaneously considering equations (2), (3), (4), (5), and (6), the engine turbine speed values, motor bus current values, turbine load values, and motor output electromagnetic torque values corresponding to each moment from 0s to the maximum allowable starting time \(t\) can be calculated, and the dynamic curves can be plotted to directly determine the starting time and the peak value of the bus current. s The above - mentioned starting process can be calculated by using finite - element and other simulation programs to build a control - circuit module with the following structure. For example,

[0090] As shown: Taking the TwinBuilder program as an example, a voltage - source module is used to simulate the storage battery, a resistor module is used to simulate the cable resistance and winding resistance, an inductor module is used to simulate the motor phase inductance, a motor module is used to simulate the starting motor, the load - torque module and the inertia module simulate the change of the engine turbine load through formula (2) for function simulation, a three - phase inverter - bridge module is used to simulate the inverter, an ammeter module is used to simulate the current sensor, a hysteresis - control module limits the phase - current, a position - information module simulates the rotor - position sensor, and a commutation module realizes the commutation of the current with the rotor position. Through software - simulation methods, the starting process is further calculated in detail to obtain a smoother data curve. Figure 4 Shown as follows: Taking the TwinBuilder program as an example, use the voltage - source module to simulate the storage battery, the resistor module to simulate the cable resistance and winding resistance, the inductor module to simulate the motor phase inductance, the motor module to simulate the starting motor, the load - torque module and the inertia module to simulate the change of the engine turbine load through formula (2) for function simulation, the three - phase inverter - bridge module to simulate the inverter, the ammeter module to simulate the current sensor, the hysteresis - control module to limit the phase - current, the position - information module to simulate the rotor - position sensor, and the commutation module to realize the commutation of the current with the rotor position. Through software - simulation methods, the starting process is further calculated in detail to obtain a smoother data curve.

[0091] Step 7: According to the calculation results of the starting time and the peak value of the bus current in the above - mentioned starting process, re - verify and design some electromagnetic parameters, and the above results can effectively assist in completing the design of the aviation starting motor.

[0092] Embodiment 11

[0093] Based on the above - mentioned embodiment, taking the WZ - XX starting motor for a certain type of aero - engine as an example, the bus voltage is 28 VDC, the bus - current limit is 260 A, and it is required to start up to 9500 r / min within 20 s. By repeating steps 1 - 3 to repeatedly optimize the electromagnetic scheme, on the premise of meeting the bus - current limit and the disengagement speed, the no - load speed of the motor is controlled near 1.1 times the maximum disengagement speed of the motor, and the output torque of the motor is increased to the maximum. Two electromagnetic schemes that meet the preliminary requirements are obtained as shown in Table 1 below:

[0094] Table 1 Optimized motor electromagnetic scheme

[0095]

[0096] The optimal phase - current limit value of 372 A is obtained through step 4, the coefficient values of the turbine - load - torque function are obtained by fitting through step 5, and finally, through step 6 combined with the simulation program, the calculation results are obtained, including the bus - current curve, starting - speed curve, motor - terminal - voltage curve, starting time to reach the ignition speed, electromagnetic - torque curve, phase - current curve, etc. during the starting process.

[0097] From the calculation results, both schemes meet the starting time requirements, but Scheme 1 has a faster starting speed. At the same time, the maximum value of the busbar current is accurately designed near 260 A, which meets the current limit requirements and maximally improves the output capacity. Therefore, Scheme 1 is selected for production.

[0098] To verify the accuracy of the design model, the starting motor was installed on the engine casing and an ignition test was carried out on the high-altitude test bench. The comparison data is shown in Table 2 below:

[0099] Table 2 Comparison of test data

[0100]

[0101] It can be found from the comparison of the test data that the calculation error is within the allowable range, which verifies the accuracy of the design model. At the same time, the performance prediction of the whole starting process is realized, and the design reliability is greatly improved.

[0102] Thus, based on the rated voltage and rated current limit of the engine battery, the present invention designs the no-load speed near the maximum disengagement speed during the engine starting process, so that the ratio of the torque coefficient to the back electromotive force coefficient of the starting motor reaches the optimum, which can maximally improve the utilization rate of motor materials and optimize the volume and weight. At the same time, by repeatedly iterating the phase current limit value for the thermal load and the peak value of the busbar current, the optimum current limit value is determined. Finally, through the function fitting of the engine load curve and combining the key parameter variable relationship during the starting process, the key parameter curves during the whole starting process can be drawn by solving a system of multiple time-variable nonlinear equations or by means of software simulation. The starting time and performance characteristics can be intuitively determined. It solves the problems that the performance of the starting motor with the engine turbine load during starting cannot be predicted and the busbar current is prone to exceed the tolerance, and at the same time ensures the starting time and improves the engine ignition success rate.

Claims

1. An optimized design method for a starting motor based on engine load matching, characterized in that: It includes the following steps: ① Preliminary modeling: Establish a preliminary model of the brushless DC motor according to the parameter requirements of the engine electrical system; ② Determine the no-load condition: Analyze the no-load condition of the motor by using the electromagnetic finite element field-circuit coupling method according to the rated voltage value; ③Optimal torque coefficient adjustment: Determine the ratio of the no-load speed n0 to the maximum disengagement speed n of the starting motor during the engine ignition starting process. If this ratio is less than the preset low value or greater than the preset high value, return to step ② after attempting to change the parameters. If this ratio is between the preset low value and the preset high value, proceed to step ④; Changing the parameters means changing the magnetic conductive material and / or adjusting the main electromagnetic parameters; T The ratio is between the preset low value and the preset high value and then proceed to step ④; Changing the parameters means changing the magnetic conductive material and / or adjusting the main electromagnetic parameters; ④ Set the load torque-time relationship: Calculate the phase current limit value, fit and calculate the relationship parameters between the engine turbine load data and the speed, thereby calculate the engine turbine load torque, and obtain the characteristics of the engine turbine load torque varying with the speed; ⑤ Calculate the starting performance: Use finite element simulation to calculate and verify the starting performance parameters, and obtain the determined main electromagnetic parameters and phase current limit value; The adjustment of the main electromagnetic parameters is carried out in at least one of the following ways: reducing the number of slots, reducing the number of poles, reducing the number of turns of the coil, reducing the armature diameter, increasing the air gap, and shortening the armature length; The preset low value is 1.05 and the preset high value is 1.

10.

2. The optimized design method of the starting motor based on engine load matching according to claim 1, characterized in that: In the step ⑤, the starting performance parameters include the engine turbine speed value, the motor bus current value, the turbine load value, and the motor output electromagnetic torque value.

3. The optimized design method of the starting motor based on engine load matching according to claim 1, wherein: In step ④, calculate the phase current limit value by calculating the phase current limit value with the highest output torque when the motor heat load is less than 960 A. 2 mm 3 It is obtained.

4. The optimized design method of the starting motor based on engine load matching according to claim 1, characterized in that: In the step ④, the fitting calculation is carried out by using polynomial fitting.

5. The optimized design method of the starting motor based on engine load matching according to claim 2, characterized in that: The engine turbine speed value is calculated by integrating the mechanical angular acceleration of the engine turbine with respect to time.

6. The optimized design method of the starting motor based on engine load matching according to claim 2, characterized in that: The motor bus current value is equivalently calculated by the following formula: Among them, U is the rated voltage of the cabin battery, and R DC is the cable resistance, R1 and L1 are the phase resistance and phase inductance of the motor respectively, and f k is the switching frequency of the power tube, p is the number of pole pairs of the motor, and Ψ m is the magnetic flux linkage of the motor phase winding, and n(t) is the motor speed varying with time.

7. The optimized design method of the starting motor based on engine load matching according to claim 4, characterized in that: The polynomial fitting is carried out by using the interpolation method to smooth the curve and then fitting, and the fitting order is above the third order.

Images