Reactance parameter robustness optimization method and system of alternating current excitation motor

Abstract

The invention discloses a reactance parameter robustness optimization method and system for an alternating current excitation motor. The method comprises the following steps: S1, obtaining an alternating current excitation motor design scheme for determining slot matching, pole pair number, rated voltage, rated capacity, stator slot type, rotor slot type and winding based on engineering preset requirements; s2, establishing a deterministic motor reactance parameter optimization model based on the AC excitation motor reactance parameters; s3, based on the deterministic motor reactance parameter optimization model and the manufacturing tolerance of the motor size parameters, establishing a robustness optimization model of the motor reactance parameters; step S4, using an AC excitation motor to design normal distribution conforming to size parameters and a Taguchi method to generate noise; and S5, optimizing the robustness optimization model of the motor reactance parameter based on a noise and optimization algorithm, outputting a motor reactance parameter robustness optimization scheme, and completing the reactance parameter robustness optimization method of the AC excitation motor.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of parameter calculation and optimization of an electric machine, and particularly relates to a robustness optimization method and system for reactance parameters of an alternating-current field excitation electric machine. BACKGROUND

[0002] Reactance parameters in an equivalent circuit of an alternating-current field excitation electric machine not only determine a time constant of the electric machine, but also reflect rationality of electromagnetic design of the electric machine to a certain extent; in the field of dynamic simulation of a power system, a key point of realizing simulation of a large-capacity alternating-current field excitation electric machine by using a simulation electric machine lies in that equivalent circuit parameter units of the simulation electric machine need to be consistent with the large-capacity alternating-current field excitation electric machine, and it can be seen that the reactance parameters in the equivalent circuit of the alternating-current field excitation electric machine are key parameters of the alternating-current field excitation electric machine. At present, many scholars have proposed accurate calculation methods for the reactance parameters in the equivalent circuit of the alternating-current field excitation electric machine, and the accuracy of the calculation of the reactance parameters has reached a high level. However, these methods are only used in the simulation design stage before the electric machine is manufactured, and when the electric machine is manufactured by a machine manufacturing factory after the design of the electric machine is completed, uncertainty in the manufacturing process will cause a certain error between a final manufacturing size and a design size, and the reactance parameters are extremely sensitive to the size of the electric machine. The uncertainty in the manufacturing process may cause a large error between the reactance parameters of a final prototype and a design value, and the reactance parameters of the alternating-current field excitation electric machine affected by the manufacturing tolerance may not meet the design requirements, so that the control performance and the electromagnetic performance of the prototype are lower than the design target, or the simulation electric machine affects simulation of the large-capacity electric machine. Therefore, it is necessary to perform robustness parameter optimization on the reactance parameters of the electric machine in the parameter design stage of the electric machine to reduce the sensitivity of the electric machine parameters to the manufacturing error and improve the accuracy of the parameter design. SUMMARY

[0003] In order to solve the problem that the sensitivity of a deterministic alternating-current field excitation electric machine reactance parameter design scheme to probabilistic errors in the manufacturing process is low, the parameters of the alternating-current field excitation electric machine manufactured by the machine factory may not meet the design requirements, so that the electromagnetic performance and the control performance are reduced, or the simulation electric machine affects simulation of the large-capacity electric machine, the application provides a robustness optimization method for reactance parameters of an alternating-current field excitation electric machine, and the method specifically includes the following steps.

[0004] Step S1: obtaining an alternating-current field excitation electric machine design scheme based on preset engineering requirements, and the design scheme includes slot matching, number of pole pairs, rated voltage, rated capacity, stator slot type, rotor slot type and winding;

[0005] Step S2: establishing a deterministic electric machine reactance parameter optimization model based on the reactance parameters of the alternating-current field excitation electric machine;

[0006] Step S3: establishing a robustness optimization model for the reactance parameters of the electric machine based on the deterministic electric machine reactance parameter optimization model and the manufacturing tolerance of the size parameters of the electric machine;

[0007] Step S4, using the normal distribution of the AC excitation design size parameter and the Taguchi method to generate noise;

[0008] Step S5, based on the noise and the optimization algorithm, optimizing the robustness optimization model of the motor reactance parameter, outputting the motor reactance parameter robustness optimization scheme, and completing the reactance parameter robustness optimization method of the AC excitation motor.

[0009] Optionally, the deterministic motor reactance parameter optimization model includes a target function and a constraint function.

[0010] Optionally, the target function is specifically:

[0011]

[0012] Wherein, E sm is the difference between the AC excitation motor excitation reactance parameter design value and the AC excitation motor excitation reactance parameter target value, E sσs is the difference between the stator leakage reactance parameter design value and the AC excitation motor stator leakage reactance parameter target value, E sσr is the difference between the rotor leakage reactance parameter design value and the AC excitation motor rotor leakage reactance parameter target value, f1 is the optimization target function of the stator leakage reactance and the rotor leakage reactance with consistent order of magnitude, f2 is the optimization target function of the excitation reactance, λ1 and λ2 are weight coefficients of E sσs and E sσr reactance error, x i is a design variable parameter, X m is the design value of the AC excitation motor excitation reactance parameter, X σs is the design value of the AC excitation motor stator leakage reactance parameter, X σr is the design value of the AC excitation motor excitation reactance rotor leakage parameter, is the target value of the AC excitation motor excitation reactance parameter, is the target value of the AC excitation motor stator leakage reactance parameter, is the target value of the AC excitation motor excitation reactance rotor leakage parameter.

[0013] Optionally, the constraint function is specifically:

[0014]

[0015] x i =[bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ];

[0016] Wherein, g1 is the constraint condition of motor performance, Xv (x i ) and X m (x i ) are the stator harmonic leakage reactance parameters and excitation reactance parameters in the optimization process, X νm is the ratio of stator harmonic leakage reactance to excitation reactance when the harmonic distortion rate of the motor stator back EMF is 5%, x i is the motor design parameter vector, bt1, hs0, hs1, hs2, Bs0, bt2, hr0, hr1, hr2, Br0, D s With D r They represent stator tooth width, stator slot height, stator slot shoulder height, stator slot depth, stator slot width, rotor tooth width, rotor slot height, rotor slot shoulder height, rotor slot depth, rotor slot width, stator core inner diameter and rotor core outer diameter respectively.

[0017] Optionally, in step S3, the content of establishing the robustness optimization model of the motor reactance parameters is specifically:

[0018] Get the manufacturing tolerance Δx of the motor size parameters i ;

[0019] Under the premise that all design parameters obey normal distribution, that is, N(x i ,σ 2 xi), based on the deterministic motor reactance parameter optimization model and the manufacturing tolerance, a robust optimization model of the motor parameters is established using the Six Sigma Design method:

[0020]

[0021] f1(x i )=0.5E sσs +0.5E sσr

[0022] f2(x i )=E sm

[0023]

[0024] Among them, n is the sigma level, μ and σ are the mean and standard deviation of the corresponding objects respectively, σ(x i ) is the standard deviation of the motor design parameters, μ g1 (x i ) is the mean of the constraint function values, is the standard deviation of the constraint function value, x i is a design parameter variable. As shown in the embodiment of the present invention and the second embodiment, the value of n is 6.

[0025] Optionally, the manufacturing tolerance Δx of the motor size parametersi The acquisition method of the motor design parameters and the calculation method of the standard deviation sigma (x i ) of the motor design parameters are as follows:

[0026] According to the actual processing mode of the motor, the manufacturing tolerance Delta x i of the motor design size parameter is determined:

[0027] Delta x i =[Delta b t1 Delta h s0 Delta h s1 Delta h s2 Delta B s0 Delta b t2 Delta h r0 Delta h r1 Delta h r2 Delta B r0 Delta D si Delta D ri ];

[0028] The standard deviation sigma xi of the motor slot size and the air gap length is calculated according to the following formula:

[0029] Sigma x i =Delta x i / 3.

[0030] The application further discloses an AC excitation motor reactance parameter robustness optimization system, which comprises:

[0031] A parameter acquisition module is configured to obtain an AC excitation motor design scheme of slot matching, pole pair number, rated voltage, rated capacity, stator slot type, rotor slot type and winding based on engineering preset requirements;

[0032] An initial parameter optimization model construction module is configured to establish a deterministic motor reactance parameter optimization model based on the AC excitation motor reactance parameter;

[0033] A robust optimization parameter model construction module is configured to establish a robust optimization model of the motor reactance parameter based on the deterministic motor reactance parameter optimization model and the manufacturing tolerance of the motor size parameter;

[0034] A noise generation module is configured to generate noise using a normal distribution in line with the AC excitation motor reactance parameter and a Taguchi method;

[0035] An algorithm optimization module is configured to optimize the robust optimization model of the motor reactance parameter based on noise and an optimization algorithm, output a motor reactance parameter robustness optimization scheme, and complete an AC excitation motor reactance parameter robustness optimization method.

[0036] Optionally, the deterministic motor reactance parameter optimization model comprises a target function and a constraint function.

[0037] Optionally, the target function is specifically:

[0038]

[0039] Wherein, E smE is a difference between a design value of a field reactance parameter of the AC excited motor and a target value of the field reactance parameter of the AC excited motor sσs E is a difference between a design value of a stator leakage reactance parameter of the AC excited motor and a target value of the stator leakage reactance parameter of the AC excited motor sσr E is a difference between a design value of a rotor leakage reactance parameter of the AC excited motor and a target value of the rotor leakage reactance parameter of the AC excited motor, f1 is an optimization target function of the stator leakage reactance and the rotor leakage reactance with the same order of magnitude, f2 is an optimization target function of the field reactance, λ1 and λ2 are respectively a weight coefficient of the reactance error of the stator leakage reactance and the rotor leakage reactance, and λ3 is a weight coefficient of the reactance error of the field reactance sσs and E sσr is a weight coefficient of the reactance error, x i is a design variable parameter, X m is the design value of the field reactance parameter of the AC excited motor, X σs is the design value of the stator leakage reactance parameter of the AC excited motor, X σr is the design value of the field reactance rotor leakage reactance parameter of the AC excited motor, is the target value of the field reactance parameter of the AC excited motor, is the target value of the stator leakage reactance parameter of the AC excited motor, is the target value of the field reactance rotor leakage reactance parameter of the AC excited motor.

[0040] Optionally, the constraint function is specifically:

[0041]

[0042] x i = [bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ];

[0043] wherein g1 is a constraint condition of motor performance, X v (x i ) and X m (x i ) are respectively the stator harmonic leakage reactance parameter and the field reactance parameter in the optimization process, X νm is a ratio of the stator harmonic leakage reactance to the field reactance when the harmonic distortion rate of the motor back electromotive force is 5%, x i is a motor design parameter phasor, bt1, hs0, hs1, hs2, Bs0, bt2, hr0, hr1, hr2, Br0, D s and D r respectively represent the stator tooth width, the stator slot opening height, the stator slot shoulder height, the stator slot depth, the stator slot opening width, the rotor tooth width, the rotor slot opening height, the rotor slot shoulder height, the rotor slot depth, the rotor slot opening width, the inner diameter of the stator core and the outer diameter of the rotor core.

[0044] Compared with the prior art, the application has the following beneficial effects:

[0045] The application firstly proposes a method of optimizing the ratio of the stator harmonic leakage reactance to the excitation reactance to optimize the stator back EMF harmonic distortion rate of the motor, provides a new idea for reducing the stator back EMF harmonic distortion rate of the motor, and can improve the efficiency of optimizing the stator back EMF harmonic distortion rate; secondly, a robustness optimization method of the reactance parameters of the AC excitation motor is established, which calculates the mean value and standard deviation of the performance of the deterministic scheme based on the orthogonal table generated by the Taguchi method, and compared with the mean value and standard deviation of the performance of the deterministic scheme calculated based on the traditional Monte Carlo method, the efficiency of robustness optimization is improved; finally, there is no robustness optimization model for the reactance parameters of the AC excitation motor before, and the application fills the gap in this regard, and the scheme obtained according to the application has lower sensitivity to manufacturing tolerances and higher accuracy of parameter design compared with the traditional deterministic parameter optimization model. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the application, the following briefly introduces the drawings needed to be used in the embodiments, and obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0047] Figure 1 is a two-dimensional graph of an optimized target motor provided by the embodiments of the application;

[0048] Figure 2 is a flowchart of a parameter optimization method of an AC excitation motor provided by the embodiments of the application;

[0049] Figure 3 is a stator back EMF distortion rate graph of 30 schemes with manufacturing errors in the final robustness optimization scheme;

[0050] Figure 4 is an error graph of the excitation reactance of 30 schemes with manufacturing errors in the final robustness optimization scheme and the design value. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the application will be further described in detail below with reference to the drawings and specific embodiments.

[0052] Embodiment one

[0053] A robustness optimization method of the reactance parameters of an AC excitation motor, as shown in Figure 2 The method comprises:

[0054] Step S1, based on the engineering preset requirements, the design scheme of the AC excitation motor is determined, including slot matching, pole pair number, rated power and stator slot type.

[0055] For details of the motor design scheme, refer to patent document 202411964605.2 "AC excitation variable speed pumping dynamic simulation motor design method".

[0056] The equivalent circuit reactance parameters of the AC excitation dynamic simulation motor need to be consistent with the large-capacity AC excitation motor. Therefore, taking a small-capacity AC excitation dynamic simulation motor with a power of 40 kW as an example, its two-dimensional plan view is shown in Figure 1 The parameter robustness optimization of the small-capacity AC excitation dynamic simulation motor is performed, and the capacity of the large-capacity AC excitation simulation motor is 300 megawatts.

[0057] The initial design scheme of the motor is input, including the scheme obtained according to the engineering preset requirements, including the parameters of the determined slot matching, pole pair number, rated voltage, rated capacity, stator slot type, rotor slot type and winding. Considering that the electromagnetic design scheme of the AC excitation motor has a mature design method, therefore, the present application does not discuss this part, only the method of reactance parameter robustness optimization which has not been studied is discussed, so the initial scheme of the AC excitation motor is obtained.

[0058] The initial design scheme of the motor is input, including the scheme obtained according to the engineering preset requirements, including the parameters of the determined slot matching, pole pair number, rated voltage, rated capacity, stator slot type, rotor slot type and winding. Considering that the electromagnetic design scheme of the AC excitation motor has a mature design method, therefore, the present application does not discuss this part, only the method of reactance parameter robustness optimization which has not been studied is discussed, so the initial scheme of the AC excitation motor is obtained.

[0059] Table 1

[0060] Motor parameters Values Motor parameters Values Number of stator slots 108 Number of rotor slots 126 Number of pole pairs 3 Rated power 40 kW Stator slot type Half closed slot Rotor slot type Half closed slot

[0061] Step S2, based on the reactance parameters of the AC excitation motor, a deterministic motor reactance parameter optimization model is established.

[0062] The per-unit value of the field reactance in the reactance parameters of the 300 megawatt large-capacity AC excitation motor is 2.786, the per-unit value of the stator leakage reactance is 0.133, and the per-unit value of the rotor leakage reactance is 0.154. Therefore, the target values of the per-unit values of the field reactance, stator leakage reactance and rotor leakage reactance of the dynamic simulation motor are 2.786, 0.133 and 0.154 respectively. In terms of performance requirements, the stator back EMF harmonic distortion rate of the AC excitation dynamic simulation motor needs to be less than 5%. When the stator back EMF harmonic distortion rate is 5%, the ratio of the stator harmonic leakage reactance to the field reactance is 0.028, and the motor design variable is represented by the phasor x i i ​The design parameters include stator tooth width bt1, stator slot opening height hs0, stator slot shoulder height hs1, stator slot depth hs2, stator slot opening width Bs0, rotor tooth width bt2, rotor slot opening height hr0, rotor slot shoulder height hr1, rotor slot depth hr2, rotor slot opening width Br0, stator core inner diameter D s and rotor core outer diameter D r The initial scheme of the size parameters of the motor is shown in Table 2.

[0063] Table 2

[0064] Motor size parameters Values Motor size parameters Values stator tooth width bt1 12.6 mm stator slot opening height hs0 30 mm stator slot shoulder height hs1 3.2 mm stator slot depth hs2 68 mm Stator slot opening width Bs0 3.6 mm rotor tooth width bt2 7.8 mm Rotor notch height hr0 28 mm Rotor slot shoulder height hr1 2.3 mm rotor slot depth hr2 110 mm Rotor slot opening width Br0 2.7 mm Stator core inner diameter D s ]]> 853.8 mm Rotor core outer diameter D r ]]> 850 mm

[0065] According to the motor reactance parameter target value and the stator back electromotive force harmonic distortion rate of 5%, the stator harmonic leakage reactance value is established to determine the parameter optimization model as follows,

[0066] E sm = X m -2.786

[0067] E sσs = X σs -0.133

[0068] E sσr = X σr -0.154

[0069]

[0070] x i = [bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ]

[0071] Wherein, E sm , E sσs and E sσr are the difference between the design value and the target value of the excitation reactance of the alternating current excitation motor, the difference between the design value and the target value of the stator leakage reactance, and the difference between the design value and the target value of the rotor leakage reactance, f1 and f2 are target functions, g1 is a constraint function, and x i is a motor design parameter phasor. Considering that the order of magnitude of the excitation reactance is larger than that of the stator and rotor leakage reactance, if E sm , E sσs and E sσr are placed in the same target function for optimization, E sm may affect E sσs and E sσrTherefore, the optimization of the stator and rotor leakage reactance of the same order of magnitude is placed in one objective function f1, and the optimization of the excitation reactance is placed in another objective function f2, and λ1 and λ2 are weight coefficients of the corresponding reactance errors. Considering that the optimization of the stator and rotor leakage reactance parameters is equally important, λ1 and λ2 are taken as 0.5.

[0072] The constraint function construction process of the deterministic parameter optimization model is specifically as follows:

[0073] The leakage reactance parameters of the AC excitation motor include harmonic leakage reactance, slot leakage reactance, end leakage reactance and tooth tip leakage reactance. The size of the harmonic leakage reactance can reflect the size of the harmonic content of the magnetic potential in the air gap, and excessive harmonic content of the magnetic potential can reduce the performance of the motor. Therefore, the performance of the motor can be improved by reasonably designing the size of the harmonic leakage reactance. Among the reactance parameters, the excitation reactance can reflect the size of the fundamental flux linkage, and therefore the ratio of the harmonic leakage reactance to the excitation reactance can be used to map the relationship between the harmonic flux linkage and the fundamental flux linkage in the air gap, and the relationship is as follows:

[0074]

[0075] In the formula, Ψ m is the fundamental flux linkage, ∑Ψ ν is the algebraic sum of the harmonic flux linkages of various orders, the harmonic orders include 5, 7, 11…(6k±1), k=1, 2, 3…, and the maximum k is Z1 / (2p), Z1 is the number of stator slots, p is the number of motor pole pairs, and I is the effective value of the current.

[0076] When the harmonic distortion rate of the stator back electromotive force is 5%, the ratio X νm of the motor stator harmonic leakage reactance to the excitation reactance is taken as the constraint boundary. The constraint condition of the deterministic motor parameter optimization model is established as follows. The constraint condition can be used to optimize the harmonic back electromotive force distortion rate by optimizing the reactance,

[0077]

[0078] x i =[bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ]

[0079] In the formula, g1 is the constraint condition of the motor performance, X ν (x i ) and X m (x i ) are the stator harmonic leakage reactance parameters and the excitation reactance parameters in the optimization process, X νm is the ratio of the stator harmonic leakage reactance to the excitation reactance when the harmonic distortion rate of the stator back electromotive force is 5%, and x ibt1, hs0, hs1, hs2, Bs0, bt2, hr0, hr1, hr2, Br0, D s and D r respectively represent stator tooth width, stator slot opening height, stator slot shoulder height, stator slot depth, stator slot opening width, rotor tooth width, rotor slot opening height, rotor slot shoulder height, rotor slot depth, rotor slot opening width, stator core inner diameter and rotor core outer diameter.

[0080] Step S3, based on the deterministic motor reactance parameter optimization model and the manufacturing tolerance of motor size parameters, a robust optimization model of motor reactance parameters is established.

[0081] The manufacturing tolerance Δx of motor size parameters is obtained i , the robust optimization model of motor reactance parameters is established, including calculating the standard deviation of motor design size parameters according to the actual processing mode of motor, and then the standard deviation σx of motor design parameters can be determined according to the following formula i .

[0082] σx i = Δx i / 3

[0083] Δx i = [Δbt1Δhs0Δhs1Δhs2ΔBs0Δbt2Δhr0Δhr1Δhr2ΔBr0ΔD si ΔD ri ]

[0084] Where Δx i is the manufacturing tolerance vector of motor design parameters, and σx i is the standard deviation of motor design parameters.

[0085] Based on the six sigma design method, the robust optimization model of motor parameters is established according to the deterministic optimization model as follows,

[0086] E sm = X m -2.786

[0087] E sσs = X σs -0.133

[0088] E sσr = X σr -0.154

[0089] f1(x i ) = 0.5E sσs + 0.5E sσr

[0090] f2(x i ) = E sm

[0091]

[0092] wherein, μ and σ are mean and standard deviation of corresponding object respectively, each design parameter related to motor performance is considered to be subject to normal distribution N(x i ,σ 2 xi In the embodiment, n is 6.

[0093] Step S4, using the normal distribution of the design size parameter of the AC excited motor and the noise generated by the Taguchi method.

[0094] Each design size parameter is subject to normal distribution N(x i ,0.0067 2 ), 4 noise levels are generated according to the normal distribution, and 4 noise levels of each parameter are shown in Table 3, and the orthogonal table L 64 (4 12 ) of 4 levels and 12 factors is generated according to 4 noise levels of 12 optimization variables, and the orthogonal table is shown in Table 4. In the table, 1, 2, 3 and 4 under the column of design parameters represent x i According to the normal distribution N(x i ,σ 2 xi) generates 4 different noise levels.

[0095] Table 3

[0096]

[0097] Table 4

[0098]

[0099] According to the noise generation scheme of the normal distribution and the Taguchi method of the optimization variable, Y noise levels are generated according to the normal distribution N(x i of each motor design parameter x i ,σ 2 xi Y can be determined according to the accuracy of robust optimization, generally Y is greater than or equal to 3.

[0100] Step S5, based on the noise and the optimization algorithm, the robust optimization model of the motor reactance parameter is optimized, the robust optimization scheme of the motor reactance parameter is output, and the robust optimization method of the reactance parameter of the AC excited motor is completed.

[0101] NSGA-II optimization algorithm has efficient non-dominated sorting and crowded distance mechanism, taking into account convergence and solution diversity, without preset weight, strong adaptability and easy to implement, is a classic and practical method to solve multi-objective optimization problem, therefore, NSGA-II is selected as the optimization algorithm of the embodiment. In the algorithm setting, the population size is set to 90, the evolution number is set to 50, the crossover probability is set to 0.9, the mutation probability is set to 0.1, the crossover distribution index is set to 20, and the mutation distribution index is set to 20.

[0102] According to the value range of μ(x i ) in the robustness optimization model, the value range of μ(x i ) in the embodiment is as follows: 90 initial populations are randomly generated in the optimization algorithm, the scheme in the population is called a deterministic scheme, 64 noise schemes are generated for each scheme in the population based on the noise orthogonal table generated in the foregoing, the specific generation method of the noise scheme is: on the basis of the design parameters of a deterministic scheme, the parameters in the orthogonal table are added, that is, the noise scheme corresponding to the deterministic scheme is obtained. The performance of the 64 noise schemes of each deterministic scheme in the initial population is calculated, and then the mean and standard deviation of the deterministic scheme objective function and constraint function are calculated according to the performance of the 64 noise schemes, and then the mean and standard deviation are brought into the fitness function and the constraint function for calculation, and finally the final robustness scheme is obtained based on the iterative calculation of the optimization algorithm.

[0103] minx i +0.02≤μ(x i )≤maxx i -0.02i=1,2,3...12

[0104] Wherein, minx i and maxx i are the minimum and maximum values in the value range of the design parameters.

[0105] The design size of the final reactance parameter robustness optimization scheme is shown in Table 5, and the reactance parameter result is shown in Table 6.

[0106] Table 5

[0107] Motor size parameters Values Motor size parameters Values stator tooth width bt1 11 mm stator slot opening height hs0 35 mm stator slot shoulder height hs1 1.2 mm <![CDATA[定子槽深hs2]]> 73 mm Stator slot opening width Bs0 2.3 mm rotor tooth width bt2 7 mm Rotor notch height hr0 32 mm Rotor slot shoulder height hr1 1.3 mm rotor slot depth hr2 116 mm Rotor slot opening width Br0 1.8 mm Stator core inner diameter D s ]]> 853 mm Rotor core outer diameter D r ]]> 850 mm

[0108] Table 6

[0109]

[0110] Figure 3The stator back EMF distortion rates of 30 schemes with manufacturing errors added in the final robustness optimization scheme are given, and it can be seen that the stator back EMF distortion rates of the 30 schemes are within 5%, and the fluctuation range is within 1% to 2%, and the fluctuation range is small. Figure 4 The errors of the excitation reactance values of the 30 schemes with manufacturing errors added in the final robustness optimization scheme are given, and it can be seen that the design errors of the excitation reactance of the 30 schemes are within 2%, and the fluctuation is very small. In summary, the robustness optimization method of the motor reactance parameter proposed in the application can not only optimize the motor reactance parameter to the target value, but also effectively optimize the stator back EMF harmonic distortion rate by means of the optimization of the reactance parameter. In addition, the application also greatly improves the robustness of the reactance parameter to manufacturing errors.

[0111] Example two

[0112] A reactance parameter robustness optimization system of an alternating current excited motor, the system comprises:

[0113] A parameter acquisition module for obtaining an alternating current excited motor design scheme based on engineering preset requirements, including determined slot coordination, pole pair number, rated voltage, rated capacity, stator slot type, rotor slot type and winding.

[0114] For details of the electrode design scheme, refer to patent document 202411964605.2 "A design method of an alternating current excited variable speed pumped storage dynamic simulation motor".

[0115] The equivalent circuit reactance parameters of the alternating current excited dynamic simulation motor need to be consistent with large-capacity alternating current excited motors, so, taking a 40kW alternating current excited dynamic simulation motor as an example, its two-dimensional plan view is as shown in Figure 1 The parameter robustness optimization of the small motor is performed, and the capacity of the large-capacity alternating current excited simulation motor is 300 megawatts.

[0116] The initial design scheme of the motor is input, including a scheme with determined slot coordination, pole pair number, rated voltage, rated capacity, stator slot type, rotor slot type and winding, etc. obtained according to engineering preset requirements. Considering that there is a mature design method for the electromagnetic design scheme of the alternating current excited motor, the application does not discuss this part, and only the reactance parameter robustness optimization method which has not been researched is discussed, so the initial scheme of the alternating current excited motor is obtained.

[0117] The initial design scheme with determined slot coordination, pole pair number, rated power, stator slot type, rotor slot type and other parameters is obtained, and the 40kW motor scheme parameters optimized by the embodiment of the application are shown in Table 7.

[0118] Table 7

[0119] Motor parameters Values Motor parameters Values Number of stator slots 108 Number of rotor slots 126 Number of pole pairs 3 Rated power 40 kW Stator slot type Half closed slot Rotor slot type Half closed slot

[0120] An initial parameter optimization model construction module is configured to establish a deterministic motor reactance parameter optimization model based on the AC excited motor reactance parameter.

[0121] The reactance parameter of the 300 MW large-capacity AC excited motor has an excitation reactance of 2.786, a stator leakage reactance of 0.133, and a rotor leakage reactance of 0.154. Therefore, the target values of the dynamic simulation motor excitation reactance, stator leakage reactance, and rotor leakage reactance are 2.786, 0.133, and 0.154, respectively. The performance requirement is that the stator back EMF harmonic distortion rate of the AC excited dynamic simulation motor needs to be less than 5%. When the stator back EMF harmonic distortion rate is 5%, the ratio of the stator harmonic leakage reactance to the excitation reactance is 0.028. The motor design variable is represented by a phasor x i , x i The design parameters contained include a stator tooth width bt1, a stator slot opening height hs0, a stator slot shoulder height hs1, a stator slot depth hs2, a stator slot opening width Bs0, a rotor tooth width bt2, a rotor slot opening height hr0, a rotor slot shoulder height hr1, a rotor slot depth hr2, a rotor slot opening width Br0, a stator core inner diameter D s , and a rotor core outer diameter D r The initial scheme of the size parameters of the motor is shown in Table 8.

[0122] Table 8

[0123] Motor size parameters Values Motor size parameters Values stator tooth width bt1 12.6 mm stator slot opening height hs0 30 mm stator slot shoulder height hs1 3.2 mm stator slot depth hs2 68 mm Stator slot opening width Bs0 3.6 mm rotor tooth width bt2 7.8 mm Rotor notch height hr0 28 mm Rotor slot shoulder height hr1 2.3 mm rotor slot depth hr2 110 mm Rotor slot opening width Br0 2.7 mm Stator core inner diameter D s ]]> 853.8 mm Rotor core outer diameter D r ]]> 850 mm

[0124] According to the motor reactance parameter target value and the stator back EMF harmonic distortion rate of 5%, the deterministic parameter optimization model is established as follows,

[0125] E sm = X m - 2.786

[0126] E sσs = X σs - 0.133

[0127] E sσr = X σr - 0.154

[0128]

[0129] x i = [bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ]

[0130] wherein E sm , E sσs and E sσr are respectively the difference between the design value and the target value of the excitation reactance of the AC excitation motor, the difference between the design value and the target value of the stator leakage reactance, and the difference between the design value and the target value of the rotor leakage reactance, f1 and f2 are target functions, g1 is a constraint function, x i is a motor design parameter phasor. Considering that the order of magnitude of the excitation reactance is larger than that of the stator and rotor leakage reactance, if E sm , E sσs and E sσr are placed in the same target function for optimization, E sm may affect the optimization accuracy of E sσs and E sσr , therefore, the optimization of the stator and rotor leakage reactance with the same order of magnitude is placed in one target function f1, and the optimization of the excitation reactance is placed in another target function f2, λ1 and λ2 are respectively the weight coefficients of the corresponding reactance error, and considering that the optimization of the stator and rotor leakage reactance parameters are equally important, λ1 = λ2 = 0.5 is taken.

[0131] wherein the constraint function construction process of the deterministic parameter optimization model is specifically as follows:

[0132] The leakage reactance parameters of the AC excitation motor include harmonic leakage reactance, slot leakage reactance, end leakage reactance and tooth tip leakage reactance, wherein the size of the harmonic leakage reactance can reflect the size of the harmonic content of the magnetic potential in the air gap, and the excessive harmonic content of the magnetic potential can reduce the performance of the motor, therefore, the performance of the motor can be improved by reasonably designing the size of the harmonic leakage reactance. Among the reactance parameters, the excitation reactance can reflect the size of the fundamental magnetic flux linkage, therefore, the relationship between the harmonic magnetic flux linkage and the fundamental magnetic flux linkage in the air gap can be mapped through the ratio relationship between the harmonic leakage reactance and the excitation reactance, and the relationship is as follows:

[0133]

[0134]

[0135] In the formula, Ψ m is the fundamental magnetic flux linkage, and ∑Ψ ν is the algebraic sum of the harmonic magnetic flux linkages of various magnetic potentials, the harmonic orders include 5, 7, 11…(6k±1), k = 1, 2, 3…, and the maximum k is Z1 / (2p), Z1 is the number of stator slots, p is the number of motor pole pairs, and I is the effective value of current.

[0136] When the harmonic distortion rate of the stator counter electromotive force is 5%, the ratio X νm of the motor stator harmonic leakage reactance and the excitation reactance is the constraint boundary, the constraint condition of the following deterministic motor parameter optimization model is established, and the constraint condition can be used to optimize the harmonic counter electromotive force distortion rate by optimizing the reactance,

[0137]

[0138] x i = [bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ]

[0139] wherein g1 is the constraint condition of motor performance, X ν (x i ) and X m (x i ) are the stator harmonic leakage reactance parameters and the excitation reactance parameters in the optimization process, X νm is the ratio of the stator harmonic leakage reactance and the excitation reactance when the stator harmonic distortion rate of the motor back electromotive force is 5%, x i is the motor design parameter phasor, bt1, hs0, hs1, hs2, Bs0, bt2, hr0, hr1, hr2, Br0, D s and D r represent the stator tooth width, the stator slot opening height, the stator slot shoulder height, the stator slot depth, the stator slot opening width, the rotor tooth width, the rotor slot opening height, the rotor slot shoulder height, the rotor slot depth, the rotor slot opening width, the inner diameter of the stator core, and the outer diameter of the rotor core.

[0140] A robust optimization parameter model construction module is configured to establish a robust optimization model of motor reactance parameters based on the deterministic motor reactance parameter optimization model and the manufacturing tolerance of motor size parameters.

[0141] The manufacturing tolerance Δx i of the motor size parameters is obtained, and the robust optimization model of the motor reactance parameters is established, including calculating the standard deviation of the motor design size parameters according to the actual processing mode of the motor, and then determining the standard deviation σx i of the motor slot size and the air gap length according to the following formula.

[0142] σx i = Δx i / 3

[0143] Δx i = [Δbt1Δhs0Δhs1Δhs2ΔBs0Δbt2Δhr0Δhr1Δhr2ΔBr0ΔD si ΔD ri ]

[0144] wherein Δx i is the manufacturing tolerance vector of the motor design parameters, and σx i is the standard deviation of the motor design parameters.

[0145] Establish a robust optimization model for motor parameters. Based on the Six Sigma design method, the robust optimization model for motor parameters is established according to the deterministic optimization model as follows:

[0146] E sm =X m -2.786

[0147] E sσs =X σs -0.133

[0148] E sσr =X σr -0.154

[0149] f1(x i )=0.5E sσs +0.5E sσr

[0150] f2(x i )=E sm

[0151]

[0152] Among them, μ and σ are the mean and standard deviation of the corresponding objects, respectively. Each design parameter related to motor performance is considered to obey the normal distribution N(x i ,σ 2 xi ).

[0153] The noise generation module is used to generate noise using the normal distribution of the reactance parameters of the AC excitation motor and the Taguchi method.

[0154] Each design size parameter follows the normal distribution N(x i ,0.0067 2 ), four noise levels are generated according to the normal distribution. The four noise levels of each parameter are shown in Table 9. The orthogonal table L with four levels and twelve factors is generated according to the four noise levels of the twelve optimized variables. 64 (4 12 ), the orthogonal table is shown in Table 10. In the table, 1, 2, 3, and 4 under the design parameter column represent the design parameters x. i According to the normal distribution N(x i ,σ 2 xi) 4 different noise levels generated.

[0155] Table 9

[0156]

[0157] Table 10

[0158]

[0159] According to the normal distribution consistent with the optimization variable and the noise scheme generated by the Taguchi method, including generating Y noise levels according to each motor design parameter x i consistent with the normal distribution N(x i ,σ 2 xi )Y can be determined according to the accuracy of robustness optimization, generally Y greater than or equal to 3.

[0160] The algorithm optimization module is used to optimize the robustness optimization model of the motor reactance parameter based on the noise and the optimization algorithm, output the motor reactance parameter robustness optimization scheme, and complete the reactance parameter robustness optimization method of the AC excitation motor.

[0161] NSGA-II optimization algorithm has efficient non-dominated sorting and crowded distance mechanism, and takes into account convergence and solution diversity, without pre-setting weight, strong adaptability and easy to implement, is a classic and practical method for solving multi-objective optimization problem, therefore, NSGA-II is selected as the optimization algorithm of the embodiment. In the algorithm setting, the population size is set to 90, the evolution number is set to 50, the crossover probability is set to 0.9, the mutation probability is set to 0.1, the crossover distribution index is set to 20, and the mutation distribution index is set to 20.

[0162] According to the value range of μ(x i ) in the robustness optimization model, the value range of μ(x i ) in the embodiment is as follows: 90 initial populations are randomly generated in the optimization algorithm, the scheme in the population is called a deterministic scheme, 64 noise schemes are generated for each scheme in the population based on the noise orthogonal table generated before, and the specific generation method of the noise scheme is: on the basis of a deterministic scheme design parameter, add the parameters in the orthogonal table, that is, obtain the noise scheme corresponding to the deterministic scheme. The performance of the 64 noise schemes of each deterministic scheme in the initial population is calculated, and then the mean and standard deviation of the deterministic scheme objective function and constraint function are calculated according to the performance of the 64 noise schemes, and then the mean and standard deviation are brought into the fitness function and the constraint function for calculation, and finally the final robustness scheme is obtained based on the optimization algorithm iterative calculation.

[0163] minx i +0.02≤μ(x i )≤maxx i -0.02i=1,2,3...12

[0164] Wherein, minx i and maxx i are the minimum and maximum values in the design parameter value range.

[0165] The design size of the final reactance parameter robustness optimization scheme is shown in Table 11, and the reactance parameter results are shown in Table 12.

[0166] Table 11

[0167] Motor size parameters Values Motor size parameters Values stator tooth width bt1 11 mm stator slot opening height hs0 35 mm stator slot shoulder height hs1 1.2 mm stator slot depth hs2 73 mm Stator slot opening width Bs0 2.3 mm rotor tooth width bt2 7 mm Rotor notch height hr0 32 mm Rotor slot shoulder height hr1 1.3 mm rotor slot depth hr2 116 mm Rotor slot opening width Br0 1.8 mm Stator core inner diameter D s ]]> 853 mm Rotor core outer diameter D r ]]> 850 mm

[0168] Table 12

[0169]

[0170] Figure 3 The stator back EMF distortion rates of the 30 schemes with manufacturing errors added in the final robustness optimization scheme are given, and it can be seen that the stator back EMF distortion rates of the 30 schemes are all within 5%, and the fluctuation range is within 1% to 2%, and the fluctuation range is small. Figure 4 The errors of the excitation reactance values of the 30 schemes with manufacturing errors added in the final robustness optimization scheme and the design values are given, and it can be seen that the design errors of the excitation reactance of the 30 schemes are all within 2%, and the fluctuation is small. In summary, the motor reactance parameter robustness optimization method provided by the application can not only ensure that the motor reactance parameters are optimized to the target value, but also achieve the goal of effectively optimizing the stator back EMF harmonic distortion rate by means of the optimization of the reactance parameters. In addition, the application greatly improves the robustness of the reactance parameters to manufacturing errors.

[0171] The above-described embodiments are only descriptions of the preferred modes of the application, and do not limit the scope of the application. Without departing from the design spirit of the application, various modifications and improvements to the technical solutions of the application made by those of ordinary skill in the art should fall within the protection scope of the claims of the application.

Claims

1. A robustness optimization method for reactance parameters of an AC excitation motor, characterized in that: The method comprises: Step S1: obtaining an AC excitation motor design scheme that determines slot matching, pole pair number, rated voltage, rated capacity, stator slot type, rotor slot type, and winding based on preset engineering requirements; Step S2: establishing a deterministic motor reactance parameter optimization model based on the AC excitation motor reactance parameters; Step S3: establishing a robust optimization model for the motor reactance parameters based on the deterministic motor reactance parameter optimization model and the manufacturing tolerances of the motor size parameters; Step S4, generating noise using the normal distribution that the design size parameters of the AC excitation motor conform to and the Taguchi method; Step S5: Optimize the robustness optimization model of the motor reactance parameters based on the noise and optimization algorithm, output the robustness optimization scheme of the motor reactance parameters, and complete the robustness optimization method of the reactance parameters of the AC excitation motor.

2. The method for optimizing the robustness of reactance parameters of an AC excitation motor according to claim 1, wherein: The deterministic motor reactance parameter optimization model includes an objective function and a constraint function.

3. The method for optimizing the robustness of reactance parameters of an AC excitation motor according to claim 2, characterized in that: The objective function is specifically: Among them, E sm E is the difference between the design value of the excitation reactance parameter of the AC excitation motor and the target value of the excitation reactance parameter of the AC excitation motor. sσs E is the difference between the design value of the stator leakage reactance parameter and the target value of the stator leakage reactance parameter of the AC excitation motor. sσr is the difference between the design value of the rotor leakage reactance parameter and the target value of the rotor leakage reactance parameter of the AC excitation motor, f1 is the optimization objective function of the stator leakage reactance and rotor leakage reactance with the same order of magnitude, f2 is the optimization objective function of the excitation reactance, λ1 and λ2 are E sσs and E sσr Weighting factor for reactance error, x i is the design variable parameter, X m is the design value of the excitation reactance parameter of the AC excitation motor, X σs is the design value of the stator leakage reactance parameter of the AC excitation motor, X σr is the design value of the excitation reactance and rotor leakage reactance parameters of the AC excitation motor, is the target value of the excitation reactance parameter of the AC excitation motor, is the target value of the stator leakage reactance parameter of the AC excitation motor, It is the target value of the excitation reactance and rotor leakage reactance parameters of the AC excitation motor.

4. The method for optimizing the robustness of reactance parameters of an AC excitation motor according to claim 3, characterized in that: The constraint function is specifically: x i =[bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ]; Among them, g1 is the constraint condition of motor performance, X v (x i ) and X m (x i ) are the stator harmonic leakage reactance parameters and excitation reactance parameters in the optimization process, X νm is the ratio of stator harmonic leakage reactance to excitation reactance when the harmonic distortion rate of the motor stator back EMF is 5%, x i is the motor design parameter vector, bt1, hs0, hs1, hs2, Bs0, bt2, hr0, hr1, hr2, Br0, D s With D r They represent stator tooth width, stator slot height, stator slot shoulder height, stator slot depth, stator slot width, rotor tooth width, rotor slot height, rotor slot shoulder height, rotor slot depth, rotor slot width, stator core inner diameter and rotor core outer diameter respectively.

5. The method for optimizing the robustness of reactance parameters of an AC excitation motor according to claim 4, characterized in that: In step S3, the content of establishing the robustness optimization model of the motor reactance parameters is specifically as follows: Get the manufacturing tolerance Δx of the motor size parameters i ; Under the premise that all design parameters obey normal distribution, that is, N(x i ,σ 2 xi), based on the deterministic motor reactance parameter optimization model and the manufacturing tolerance, a robust optimization model of the motor parameters is established using the Six Sigma Design method: f1(x i )=0.5E sσs +0.5E sσr f2(x i )=E sm Among them, n is the sigma level, μ and σ are the mean and standard deviation of the corresponding objects respectively, σ(x i ) is the standard deviation of the motor design parameters, is the mean of the constraint function value, is the standard deviation of the constraint function value, x i is the design parameter variable.

6. The method for optimizing the robustness of reactance parameters of an AC excitation motor according to claim 5, characterized in that: The manufacturing tolerance Δx of the motor size parameters i The acquisition method and the standard deviation σ(x i ) is calculated as follows: Determine the manufacturing tolerance Δx of the motor design dimensional parameters based on the actual motor processing method i : Δx i =[Δbt1Δhs0Δhs1Δhs2ΔBs0Δbt2Δhr0Δhr1Δhr2ΔBr0ΔD si ΔD ri ]; The standard deviation σxi of the motor slot size and air gap length is calculated according to the following formula: σx i =Δx i / 3。 7. A system for optimizing the robustness of reactance parameters of an AC excitation motor, the system being used to implement the method according to any one of claims 1 to 6, characterized in that the system include: The parameter acquisition module is used to obtain the AC excitation motor design scheme based on the preset engineering requirements, including slot matching, number of pole pairs, rated voltage, rated capacity, stator slot type, rotor slot type and winding; An initial parameter optimization model building module is used to establish a deterministic motor reactance parameter optimization model based on the AC excitation motor reactance parameters; A robust optimization parameter model building module is used to establish a robust optimization model of the motor reactance parameters based on the deterministic motor reactance parameter optimization model and the manufacturing tolerance of the motor size parameters; A noise generation module is used to generate noise using the normal distribution of the reactance parameters of the AC excitation motor and the Taguchi method; The algorithm optimization module is used to optimize the robustness optimization model of the motor reactance parameters based on noise and optimization algorithm, output the robustness optimization scheme of the motor reactance parameters, and complete the robustness optimization method of the reactance parameters of the AC excitation motor.

8. The reactance parameter robustness optimization system for an AC excitation motor according to claim 7, characterized in that: The deterministic motor reactance parameter optimization model includes objective function and constraint function.

9. The reactance parameter robustness optimization system for an AC excitation motor according to claim 8, characterized in that: The objective function is specifically: Among them, E sm E is the difference between the design value of the excitation reactance parameter of the AC excitation motor and the target value of the excitation reactance parameter of the AC excitation motor. sσs E is the difference between the design value of the stator leakage reactance parameter and the target value of the stator leakage reactance parameter of the AC excitation motor. sσr is the difference between the design value of the rotor leakage reactance parameter and the target value of the rotor leakage reactance parameter of the AC excitation motor, f1 is the optimization objective function of the stator leakage reactance and rotor leakage reactance with the same order of magnitude, f2 is the optimization objective function of the excitation reactance, λ1 and λ2 are E sσs and E sσr Weighting factor for reactance error, x i is the design variable parameter, X m is the design value of the excitation reactance parameter of the AC excitation motor, X σs is the design value of the stator leakage reactance parameter of the AC excitation motor, X σr is the design value of the excitation reactance and rotor leakage reactance parameters of the AC excitation motor, is the target value of the excitation reactance parameter of the AC excitation motor, is the target value of the stator leakage reactance parameter of the AC excitation motor, It is the target value of the excitation reactance and rotor leakage reactance parameters of the AC excitation motor.

10. The reactance parameter robustness optimization system for an AC excitation motor according to claim 9, characterized in that: The constraint function is specifically: x i =[bt1 hs0 hs1 hs2 Bs0 bt2 hr0 hr1 hr2 Br0 D s D r ]; Among them, g1 is the constraint condition of motor performance, X v (x i ) and X m (x i ) are the stator harmonic leakage reactance parameters and excitation reactance parameters during the optimization process, X νm is the ratio of stator harmonic leakage reactance to excitation reactance when the harmonic distortion rate of the motor stator back EMF is 5%, x i are the motor design parameter phasors, bt1, hs0, hs1, hs2, Bs0, bt2, hr0, hr1, hr2, Br0, D s With D r They represent stator tooth width, stator slot height, stator slot shoulder height, stator slot depth, stator slot width, rotor tooth width, rotor slot height, rotor slot shoulder height, rotor slot depth, rotor slot width, stator core inner diameter and rotor core outer diameter respectively.

Images