Method for constructing PI (Proportional Integral) improved self-adaptive fuzzy controller of permanent magnet synchronous motor

Abstract

The invention discloses a permanent magnet synchronous motor PI improved adaptive fuzzy controller construction method, and belongs to the technical field of permanent magnet synchronous motor intelligent control. Constructing a permanent magnet synchronous motor control system structure, wherein a PI-like improved adaptive fuzzy controller is adopted by a rotating speed outer ring of the permanent magnet synchronous motor control system structure; designing the overall structure of the PI-like improved adaptive fuzzy controller, and building a PI-like fuzzy controller without an adaptive mechanism; selecting an error e and a normalized acceleration rv (k) thereof as observed quantities, building an adaptive mechanism with global adaptive adjustment characteristics, deducing a clearness calculation formula of the rv (k), and designing and improving an adaptive adjustment rule; and finally, controller parameter optimization is carried out, and the design of the PI-like improved adaptive fuzzy controller of the PMSM is completed. The invention provides a PI-like improved adaptive fuzzy controller of a permanent magnet synchronous motor and a self-adjusting rule design method thereof. The PI-like improved adaptive fuzzy controller has the characteristics of relatively high response speed, relatively small overshoot, relatively high load disturbance resistance and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of permanent magnet synchronous motors, and is particularly suitable for control applications of permanent magnet synchronous motor drive equipment requiring fast response speed, small overshoot, and strong load disturbance resistance. Background Art

[0002] Permanent magnet synchronous motors are often used as the best actuators for speed control systems due to their small size, high efficiency, low torque pulsation, high power density, and easy maintenance. For high-order, nonlinear, strongly coupled, and multivariable permanent magnet synchronous motor systems, traditional PID control strategies cannot meet the requirements of high-performance dynamic decoupling control. Fuzzy control, as a nonlinear mapping, has the characteristic of being independent of the mathematical model of the controlled object, and has become a research hotspot in the field of permanent magnet synchronous motor control.

[0003] The results of literature and patent search show that the research on fuzzy controllers at home and abroad mainly focuses on the combination of fuzzy controllers and classical PID controllers, the adaptive adjustment of fuzzy controller parameters, the design of fuzzy controller structures and adaptive laws, and the integration of fuzzy control with other advanced control algorithms. In view of the characteristics of simple design, strong nonlinearity, and easy DSP implementation (table lookup method), adaptive fuzzy controllers are very suitable for speed control of permanent magnet synchronous motors. The adaptive fuzzy PID controller, which is widely studied at present, is essentially to adjust the PID parameters online through fuzzy control algorithms. In the control process, the PID controller still bears the main control task. It is not a pure fuzzy controller, so it cannot give full play to the advantages of fuzzy control itself. As a two-dimensional fuzzy controller, the PI-like fuzzy controller does not have fixed proportional and integral gains and has stronger nonlinearity. Adding adaptive adjustment characteristics to the PI-like fuzzy controller itself can better play its superior control function. However, the input observations selected for adaptive adjustment are usually errors e and ec, which can only adjust a certain working point of the system, which has certain limitations. In order to further improve the dynamic and static performance of the permanent magnet synchronous motor system, selecting appropriate adaptive observation variables, improving the controller structure and adaptive law, and designing a permanent magnet synchronous motor PI-like improved adaptive fuzzy controller algorithm with adaptive global adjustment capability has important theoretical and technical application value. Summary of the invention

[0004] The purpose of the present invention is to provide a method for constructing a permanent magnet synchronous motor PI-like improved adaptive fuzzy controller. The permanent magnet synchronous motor control system using the PI-like improved adaptive fuzzy controller has the characteristics of faster response speed, smaller overshoot, and stronger load disturbance resistance.

[0005] To achieve the above object, the present invention provides a method for constructing a permanent magnet synchronous motor type PI adaptive fuzzy controller, comprising the following steps:

[0006] S1. Construct the control system structure of permanent magnet synchronous motor, whose speed outer loop adopts PI-like improved adaptive fuzzy controller;

[0007] S2. Design the overall structure of the PI-like improved adaptive fuzzy controller, and build a PI-like fuzzy controller without adaptive mechanism, select the error e and its normalized acceleration r v (k) As an observation, an adaptive mechanism with global adaptive adjustment characteristics is constructed, and r is derived v ( k )’s clear calculation formula and design improvement adaptive adjustment rules;

[0008] S3. Optimize controller parameters and complete the design of PI-like improved adaptive fuzzy controller for PMSM.

[0009] Preferably, in step S1, the specific steps of constructing the permanent magnet synchronous motor control system structure are as follows:

[0010] A dual closed-loop control method is adopted to construct the control system structure of the permanent magnet synchronous motor, in which: the "speed error" is sent to the PI-like adaptive fuzzy controller of the speed outer loop, and its "normalized acceleration dynamic variable" is obtained through real-time calculation. The adaptive fuzzy controller outputs the q-axis torque current given value in the rotor flux oriented dq coordinate system; the current is dynamically adjusted in the dq coordinate system, and the d-axis electric excitation current closed loop and the q-axis torque current closed loop use ordinary PI regulators.

[0011] Preferably, in step S2, the specific steps of designing the overall structure of the PI-like improved adaptive fuzzy controller are as follows:

[0012] S21, using a two-dimensional incremental PI fuzzy controller as the main speed controller;

[0013] S22, connecting a dual-input single-output auxiliary fuzzy controller in parallel to the incremental PI fuzzy controller in step S21 as a self-adjusting mechanism to construct a "relative rate observer";

[0014] S23. Adaptively adjust the output increment of fuzzy controller 1 through the self-adjustment factor α of the auxiliary fuzzy controller in step S22.

[0015] Preferably, in step S21, the incremental PI fuzzy controller adopts a dual-input single-output structure, and its input quantity e N and △e N The speed error e and error change rate △e are respectively related to the corresponding quantization factor Ge , G Δe After multiplication, the output is the control increment △u N .

[0016] Preferably, in step S22, the "relative rate observer" takes the error change rate Δe as input, and its output is the normalized (relative) error acceleration r v (k);

[0017] r v (k) and e N Together they serve as the input of the auxiliary fuzzy controller; the output of the auxiliary fuzzy controller is α N With the output factor G α After multiplication, the self-adjustment factor α is obtained.

[0018] Preferably, in step S23, the output increment of the quasi-incremental PI fuzzy controller is adaptively adjusted by the self-adjusting factor α output by the auxiliary fuzzy controller. Specifically, the output Δu of the quasi-incremental PI fuzzy controller is N Its output proportional factor G Δu After multiplication, it needs to be multiplied by (1+α) again to obtain the total output U of the speed loop incremental PI adaptive fuzzy controller.

[0019] Preferably, in step S21, the specific steps of constructing a quasi-incremental PI fuzzy controller are as follows:

[0020] Firstly, the fuzzy set is selected according to the actual speed control accuracy of the permanent magnet synchronous motor;

[0021] Select seven levels of negative large NB, negative medium NM, negative small NS, zero ZO, positive small PS, positive medium PM, and positive large PB as fuzzy sets; define the input quantity e of fuzzy controller 1 N , Δe N And the output Δu N The fuzzy sets are: {NB, NM, NS, ZO, PS, PM, PB}, the membership functions are all symmetrical, evenly distributed, semi-overlapping triangles, and the normalized domain is [-1,1];

[0022] Select the quantization factor G of the incremental PI fuzzy controller e , G Δe and the scaling factor G Δu The reference formula for selecting relevant factors is:

[0023]

[0024] Among them [e min ,e max ]、[Δemin ,Δe max ] and [Δu min ,Δu max ] are respectively the dynamic ranges of input and output variables during the actual operation of the system. In the process of speed control of the permanent magnet synchronous motor, the upper and lower limits of the system dynamic range are equal, e max =-e min , formula (1) is simplified to:

[0025]

[0026] Secondly, the output variable Δu is designed according to the expected waveform of the system dynamic response and the characteristics of the classic incremental PI controller. N The fuzzy rule table of

[0027] Finally, the fuzzy output is obtained according to Mamdani's min-max reasoning method, and then the precise output of the fuzzy controller is obtained after defuzzification by the centroid method. The calculation formula is:

[0028]

[0029] Preferably, in step S22, the normalized error acceleration r containing the system response speed information is v (k) and speed error e N As the two input quantities of the auxiliary fuzzy controller, the normalized error acceleration r v The calculation design of (k) is as follows:

[0030] Define the normalized error acceleration r v (k) The formula is as follows:

[0031]

[0032] Where de(k) and dde(k) are the error velocity and error acceleration of the system at time k, respectively, and their definitions are:

[0033] de(k)=Δe=e(k)-e(k-1) (5)

[0034] dde(k)=de(k)-de(k-1) (6)

[0035] The definition of de(m) is:

[0036]

[0037] Substituting formula (7) into formula (4), we get r v The clarified formula of (k) is:

[0038]

[0039] From formula (8), we can see that r v (k) Quantitatively describe the speed of system response, and its changing rules are as follows:

[0040] When the system response speed is fast enough, r v The value of (k) is close to 1;

[0041] When the system response speed is slow, r v The value of (k) is close to -1;

[0042] When the system rises or falls at a certain speed, r v The value of (k) approaches 0;

[0043] When the values ​​of de(k) and de(k-1) are both positive or negative, and the values ​​of |de(k)| and |de(k-1)| are not both 0, r v The range of (k) is [-1,1].

[0044] Preferably, in step S23, the specific steps of designing an adaptive adjustment rule and designing a fuzzy rule table of the self-adjusting factor α accordingly are as follows:

[0045] According to the observation error e and normalized error acceleration r at the input of the auxiliary fuzzy controller v (k) The changing law of the system divides the system into different response stages;

[0046] When the auxiliary fuzzy controller makes real-time adaptive adjustment to the output of the quasi-incremental PI fuzzy controller, the fuzzy rule table can be determined according to the following rules:

[0047] When the system error is large and the response speed is fast, the output of the auxiliary fuzzy controller is increased to increase the total output U of the controller in a positive direction, and the system will quickly approach the set value;

[0048] When the system error is small and the response speed is constant, adjust the output of the auxiliary fuzzy controller to reduce the total output U of the controller to avoid excessive overshoot;

[0049] When the system speed starts to drop and the response speed is slow, the total output U of the controller should be further reduced to quickly approach the steady state;

[0050] When the system falls back to the vicinity of the set point and the response speed is constant, fine-tune the output of the auxiliary fuzzy controller to increase the total output U of the controller in advance and reduce the reverse overshoot;

[0051] When the system falls back excessively, regardless of the speed of response, the output of the auxiliary fuzzy controller is adjusted to make the total output U of the controller continue to increase, suppress overshoot and speed up the response speed;

[0052] A fuzzy rule table is designed based on the above adaptive rules.

[0053] Preferably, in step S23, the auxiliary fuzzy controller in the PI-like improved adaptive fuzzy controller has an output variable and an output factor G α The variation law and range of the self-adjusting factor α obtained after multiplication are determined by the two input quantities of the auxiliary fuzzy controller, the fuzzy rules and reasoning methods used, and its output factor G α Joint decision making;

[0054] The size of the self-adjustment factor α will dynamically affect the degree of adaptive adjustment of the system. The adjustment algorithm for the total output U(k) of the system is designed as follows:

[0055] At the current time k of the system, the output of the fuzzy controller 1 is:

[0056] ΔU(k1)=Δu N (k)*G Δu (9)

[0057] When introducing the self-adjusting mechanism:

[0058] ΔU(k)=Δu N (k)*G Δu (1+α(k)) (10)

[0059] Among them, since the output range of the two fuzzy controllers involved in the design process is [-1,1], the range of α(k) is [-G α ,G α ], the range of ΔU(k) ​​is [ΔU(k1)*(1-G α ),ΔU(k1)*(1+G α )];

[0060] In addition, according to the structure of the PI-like fuzzy controller:

[0061] U(k)=U(k-1)+ΔU(k) ​​(11)

[0062] Therefore, substituting equation (11) into equation (10), the total output U(k) of the control system at time k can be expressed as:

[0063] U(k)=U(k-1)+Δu N (k)*G Δu (1+α(k)) (12)

[0064] Therefore, the present invention adopts a method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller of the above structure, which has the following beneficial effects:

[0065] (1) The present invention uses an adaptive mechanism that can globally adjust the control system to adjust the output variables of the PI-like fuzzy controller, and proposes a design method for the self-adjustment law of a PI-like improved adaptive fuzzy controller for a permanent magnet synchronous motor; the permanent magnet synchronous motor control system using a PI-like adaptive fuzzy controller has the characteristics of faster response speed, smaller overshoot, and stronger load disturbance resistance.

[0066] (2) The technology of the present invention has certain reference significance for the design of PI-like adaptive fuzzy controllers for other AC motors, and can be extended to be applied to advanced adaptive fuzzy control systems of other types of AC motors.

[0067] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a schematic diagram of the principle of a permanent magnet synchronous motor adaptive fuzzy control system using a permanent magnet synchronous motor type PI adaptive fuzzy controller construction method of the present invention;

[0069] Figure 2 The present invention is a schematic diagram of the principle of a quasi-(incremental) PI improved adaptive fuzzy controller using a method for constructing a quasi-PI adaptive fuzzy controller for a permanent magnet synchronous motor. DETAILED DESCRIPTION

[0070] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0071] Unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0072] Example

[0073] The present invention provides a method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller, comprising the following steps:

[0074] S1. Construct a PI adaptive fuzzy control system for a permanent magnet synchronous motor. Use a double closed-loop control method to construct a permanent magnet synchronous motor control system structure, such as Figure 1 shown. Figure 1 The structure diagram of the PMSM control system using an improved adaptive PI fuzzy controller is shown in Figure 2. d =0 control mode, both current inner loops are designed as PI control, and the speed outer loop adopts PI-like improved adaptive fuzzy controller to improve the dynamic and static performance of the system. Among them: after the "speed error" is sent to the PI-like improved adaptive fuzzy controller of the speed outer loop, its "normalized acceleration dynamic variable" is obtained through real-time calculation, and the adaptive fuzzy controller outputs the q-axis torque current given value in the rotor flux orientation dq coordinate system; the current is dynamically adjusted in the dq coordinate system, and the d-axis electric excitation current closed loop and the q-axis torque current closed loop adopt ordinary PI regulators.

[0075] S2. Construct a PI-like adaptive fuzzy speed controller. Figure 2 As shown ( Figure 2 The structure diagram of the improved PI-like adaptive fuzzy controller. The overall control structure of two fuzzy controllers in parallel is adopted; fuzzy controller 1 plays a leading control role, and fuzzy controller 2 is an auxiliary controller that can adjust the output of the former in real time. The incremental PI-like adaptive fuzzy controller is a speed controller. The specific steps are:

[0076] First, a two-dimensional incremental PI fuzzy controller (fuzzy controller 1) is used as the main speed controller; the controller adopts a dual-input single-output structure, and its input quantity e N and △e N The speed error e and error change rate △e are respectively related to the corresponding quantization factor G e , G Δe After multiplication, the output is the control increment △u N .

[0077] Then, a dual-input single-output auxiliary fuzzy controller (fuzzy controller 2) is connected in parallel to fuzzy controller 1 as a self-adjusting mechanism; in order to improve the global adaptive characteristics of the controller, a "relative rate observer" is constructed to observe the relative speed of the system response speed in real time. The "relative rate observer" takes the error change rate △e as input, and its output is the normalized (relative) error acceleration r v (k); r v (k) and e NTogether they serve as the input of fuzzy controller 2; the output of fuzzy controller 2 is α N With the output factor G α After multiplication, the self-adjustment factor α is obtained.

[0078] Finally, the self-adjustment factor α output by fuzzy controller 2 is used to realize the adaptive adjustment of the output increment of fuzzy controller 1. Specifically, the output Δu of fuzzy controller 1 N Its output proportional factor G Δu After multiplication, it needs to be multiplied by (1+α) again to obtain the total output U of the speed loop (incremental) PI adaptive fuzzy controller.

[0079] (1) Constructing a quasi-(incremental) PI fuzzy controller (fuzzy controller 1)

[0080] Firstly, the fuzzy set is selected according to the actual speed control accuracy of the permanent magnet synchronous motor;

[0081] Select seven levels including negative big (NB), negative middle (NM), negative small (NS), zero (ZO), positive small (PS), positive middle (PM), and positive big (PB) as fuzzy sets; define the input quantity e of fuzzy controller 1 N , Δe N And the output Δu N The fuzzy sets are: {NB, NM, NS, ZO, PS, PM, PB}. The shapes of the membership functions are symmetrical, evenly distributed, semi-overlapping triangles, and the normalized domain is [-1,1].

[0082] Then, select the quantization factor G of fuzzy controller 1 e , G Δe and the scaling factor G Δu The reference formula for selecting relevant factors is:

[0083]

[0084] Among them [e min ,e max ]、[Δe min ,Δe max ] and [Δu min ,Δu max ] are the dynamic ranges of input and output variables when the system is actually running. Considering that in the process of speed control of the permanent magnet synchronous motor, the upper and lower limits of the system dynamic range are equal, that is, e max =-e min , so formula (1) can be simplified to:

[0085]

[0086] Secondly, the output variable Δu is designed according to the expected waveform of the system dynamic response and the characteristics of the classic incremental PI controller. N The fuzzy rule table is shown in Table 1.

[0087] Table 1 shows the output Δu of the PI-like improved adaptive fuzzy controller N In the actual application process, 49 fuzzy rules can be fine-tuned to obtain a more ideal system dynamic response waveform.

[0088] Table 1Δu N Fuzzy rule table

[0089]

[0090] Table 1 shows the output variable Δu when the error e and the error change rate Δe are fuzzified and take different language values N The fuzzy rule table determines the size of the controller's output increment at the current moment, which in turn affects the controller's total output.

[0091] Finally, the fuzzy output is obtained according to Mamdani's min-max reasoning method, and then the precise output of the fuzzy controller is obtained after defuzzification by the centroid method. The calculation formula is:

[0092]

[0093] (2) Build an improved PI-like adaptive fuzzy controller adaptive mechanism (fuzzy controller 2)

[0094] The normalized error acceleration r containing the system response speed information v (k) and speed error e N As two input quantities of the self-adjusting mechanism (fuzzy controller 2), the various stages of the system response process are dynamically identified; the definition of the fuzzy subsets of the input and output variables of fuzzy controller 2, the shape of the membership function, the division of the domain, and the reasoning and defuzzification methods are consistent with those of fuzzy control 1; the construction steps of fuzzy controller 2 are similar to those of the classic incremental fuzzy PI controller, and its main difference is that it needs to perform "normalized error acceleration r v The calculation of "(k)" and the design of adaptive adjustment rules are as follows:

[0095] 1) Calculate the normalized error acceleration r v (k).

[0096] First, define the normalized error acceleration r v (k) The formula is as follows:

[0097]

[0098] Where de(k) and dde(k) are the error velocity and error acceleration of the system at time k, respectively, and their definitions are:

[0099] de(k)=Δe=e(k)-e(k-1) (5)

[0100] dde(k)=de(k)-de(k-1) (6)

[0101] The definition of de(m) is:

[0102]

[0103] Then, substitute equation (7) into equation (4) to obtain r v The clarified formula of (k) is:

[0104]

[0105] From formula (8), we can see that r v (k) The speed of system response can be quantitatively described, and its changing rules are as follows:

[0106] ①When the system response speed is fast enough, r v The value of (k) is close to 1;

[0107] ②When the system response speed is very slow, r v The value of (k) is close to -1;

[0108] ③When the system rises or falls at a certain speed, r v The value of (k) approaches 0;

[0109] ④ When the values ​​of de(k) and de(k-1) are both positive or negative, and the values ​​of |de(k)| and |de(k-1)| are not both 0, r v The range of (k) is [-1,1]. These two restrictions need to be taken into account when the system is actually running.

[0110] 2) Design the adaptive adjustment rules and design the fuzzy rule table of the self-adjustment factor α accordingly

[0111] First, according to the observed error (i.e., speed error) e at the input of the self-adjusting mechanism and the normalized error acceleration r v (k) changes, and the system is divided into different response stages.

[0112] Then, appropriate fuzzy rules are formulated for different system response stages based on expert knowledge and experience. Figure 2When the fuzzy controller 2 in the embodiment performs real-time adaptive adjustment on the output of the fuzzy controller 1, the fuzzy rule table can be determined according to the following rules:

[0113] ① When the system error is large and the response speed is fast, increase the output of fuzzy controller 2 to increase the total output U of the controller in a positive direction, and the system will quickly approach the set value;

[0114] ② When the system error is small and the response speed is constant, adjust the output of fuzzy controller 2 to reduce the total output U of the controller to avoid excessive overshoot;

[0115] ③ When the system speed starts to drop and the response speed is slow, the total output U of the controller should be further reduced to quickly approach the steady state;

[0116] ④ When the system falls back to the vicinity of the set point and the response speed is constant, fine-tune the output of fuzzy controller 2 to increase the total output U of the controller in advance and reduce the reverse overshoot;

[0117] ⑤ When the system falls back excessively, regardless of the response speed, adjust the output of fuzzy controller 2 so that the total output U of the controller continues to increase, suppress overshoot and speed up the response speed.

[0118] Finally, according to the above adaptive rules, a fuzzy rule table as shown in Table 2 is designed.

[0119] Table 2 is a fuzzy rule table of the self-adjusting factor α, which can be corrected according to the actual response waveform of the permanent magnet synchronous motor control system so that the system can obtain correct adjustment action in a certain response stage.

[0120] Table 2 Fuzzy rules table of self-adjusting factor α

[0121]

[0122]

[0123] For example, the linguistic variable of the error e corresponding to rule ③ is NS, and the normalized error acceleration r v The linguistic variable of (k) is NM or NS, and the fuzzy output should be PM or PB. Figure 2 When the self-adjusting factor α takes a positive value, and the output Δu of fuzzy controller 1 is N A negative value means that the total output U of the current controller will decrease at a faster rate, allowing the system to reach a steady state quickly.

[0124] 3) Optimize controller parameters and complete the design of PMSM PI-like improved adaptive fuzzy controller. Design the adjustment algorithm of the self-adjusting factor α to the total output U(k) of the system. The output variable of the fuzzy controller 2 in the PI-like adaptive fuzzy controller is related to its output factor G α The variation law and variation range of the self-adjusting factor α obtained after multiplication are determined by the two input quantities of the fuzzy controller 2, the fuzzy rule used and its output factor G α The size of the self-adjusting factor α will dynamically affect the degree of adaptive adjustment of the system. The adjustment algorithm for the total output U(k) of the system is designed as follows:

[0125] First, at the current time k of the system, for the output end of the fuzzy controller 1 without adding a self-adjusting mechanism, there is:

[0126] ΔU(k1)=Δu N (k)*G Δu (9)

[0127] Secondly, when introducing the self-adjusting mechanism:

[0128] ΔU(k)=Δu N (k)*G Δu (1+α(k)) (10)

[0129] Among them, the range of α(k) is [-G α ,G α ], the range of ΔU(k) ​​is [ΔU(k1)*(1-G α ),ΔU(k1)*(1+G α )]. In addition, according to the structure of the PI-like fuzzy controller:

[0130] U(k)=U(k-1)+ΔU(k) ​​(11)

[0131] Therefore, substituting equation (11) into equation (10), the total output U(k) of the control system at time k can be expressed as:

[0132] U(k)=U(k-1)+Δu N (k)*G Δu (1+α(k)) (12)

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller, characterized in that: The following steps are involved: S1. Construct the control system structure of permanent magnet synchronous motor, whose speed outer loop adopts PI-like improved adaptive fuzzy controller; S2. Design the overall structure of the PI-like improved adaptive fuzzy controller, and build a PI-like fuzzy controller without adaptive mechanism, select the error e and its normalized acceleration r v (k) As an observation, an adaptive mechanism with global adaptive adjustment characteristics is constructed, and r is derived v ( k )’s clear calculation formula and design improvement adaptive adjustment rules; S3. Optimize controller parameters and complete the design of PI-like improved adaptive fuzzy controller for PMSM.

2. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 1, characterized in that: In step S1, the specific steps of constructing the permanent magnet synchronous motor control system structure are as follows: The control system structure of permanent magnet synchronous motor is constructed by double closed-loop control, in which: after the "speed error" is sent to the PI-like adaptive fuzzy controller of the speed outer loop, its "normalized acceleration dynamic variable" is obtained through real-time calculation. The output of the adaptive fuzzy controller is the q-axis torque current given value in the rotor flux orientation dq coordinate system; the current is dynamically adjusted in the dq coordinate system, and the d-axis electric excitation current closed loop and the q-axis torque current closed loop use ordinary PI regulators.

3. The method for constructing a permanent magnet synchronous motor type PI adaptive fuzzy controller according to claim 1, characterized in that: In step S2, the specific steps of designing the overall structure of the PI-like improved adaptive fuzzy controller are as follows: S21, using a two-dimensional incremental PI fuzzy controller as the main speed controller; S22, connecting a dual-input single-output auxiliary fuzzy controller in parallel to the incremental PI fuzzy controller in step S21 as a self-adjusting mechanism to construct a "relative rate observer"; S23. Adaptively adjust the output increment of fuzzy controller 1 through the self-adjustment factor α of the auxiliary fuzzy controller in step S22.

4. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 3, characterized in that: In step S21, the incremental PI fuzzy controller adopts a dual-input single-output structure, and its input quantity e N and △e N The speed error e and error change rate △e are respectively related to the corresponding quantization factor G e , G Δe After multiplication, the output is the control increment △u N .

5. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 4, characterized in that: In step S22, the relative rate observer takes the error change rate Δe as input, and its output is the normalized relative error acceleration r v (k); r v (k) and e N Together they serve as the input of the auxiliary fuzzy controller; The output of the auxiliary fuzzy controller α N With the output factor G α After multiplication, the self-adjustment factor α is obtained.

6. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 5, characterized in that: In step S23, the output increment of the quasi-incremental PI fuzzy controller is adaptively adjusted by the self-adjusting factor α output by the auxiliary fuzzy controller, specifically, the output Δu of the quasi-incremental PI fuzzy controller N Its output proportional factor G Δu After multiplication, it needs to be multiplied by (1+α) again to obtain the total output U of the speed loop incremental PI adaptive fuzzy controller.

7. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 3, characterized in that: In step S21, the specific steps of constructing a quasi-incremental PI fuzzy controller are as follows: Firstly, the fuzzy set is selected according to the actual speed control accuracy of the permanent magnet synchronous motor; Select seven levels of negative large NB, negative medium NM, negative small NS, zero ZO, positive small PS, positive medium PM, and positive large PB as fuzzy sets; define the input quantity e of fuzzy controller 1 N , Δe N And the output Δu N The fuzzy sets are: {NB, NM, NS, ZO, PS, PM, PB}, the membership functions are all symmetrical, evenly distributed, semi-overlapping triangles, and the normalized domain is [-1,1]; Select the quantization factor G of the incremental PI fuzzy controller e , G Δe and the scaling factor G Δu , the reference formula for selecting the relevant factors is: Among them [e min ,e max ]、[Δe min ,Δe max ] and [Δu min ,Δu max ] are respectively the dynamic ranges of input and output variables during the actual operation of the system. In the process of speed control of the permanent magnet synchronous motor, the upper and lower limits of the system dynamic range are equal, e max =-e min , formula (1) is simplified to: Secondly, the output variable Δu is designed according to the expected waveform of the system dynamic response and the characteristics of the classic incremental PI controller. N The fuzzy rule table of Finally, the fuzzy output is obtained according to Mamdani's min-max reasoning method, and then the precise output of the fuzzy controller is obtained after defuzzification by the centroid method. The calculation formula is:

8. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 3, characterized in that: In step S22, the normalized error acceleration r containing the system response speed information is converted into v (k) and speed error e N As the two input quantities of the auxiliary fuzzy controller, the normalized error acceleration r v The calculation design of (k) is as follows: Define the normalized error acceleration r v (k) The formula is as follows: Where de(k) and dde(k) are the error velocity and error acceleration of the system at time k, respectively, and their definitions are: de(k)=Δe=e(k)-e(k-1)(5) dde(k)=de(k)-de(k-1)(6) The definition of de(m) is: Substituting formula (7) into formula (4), we get r v The clarified formula of (k) is: From formula (8), we can see that r v (k) Quantitatively describe the speed of system response, and its changing rules are as follows: When the system response speed is fast enough, r v The value of (k) is close to 1; When the system response speed is slow, r v The value of (k) is close to -1; When the system rises or falls at a certain speed, r v The value of (k) approaches 0; When the values ​​of de(k) and de(k-1) are both positive or negative, and the values ​​of |de(k)| and |de(k-1)| are not both 0, r v The range of (k) is [-1,1].

9. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 3, characterized in that: In step S23, the specific steps of designing an adaptive adjustment rule and designing a fuzzy rule table of the self-adjusting factor α accordingly are as follows: According to the observation error e and normalized error acceleration r at the input of the auxiliary fuzzy controller v (k) The changing law of the system divides the system into different response stages; When the auxiliary fuzzy controller makes real-time adaptive adjustment to the output of the incremental PI fuzzy controller, the fuzzy rule table is determined according to the following rules: When the system error is large and the response speed is fast, the output of the auxiliary fuzzy controller is increased to increase the total output U of the controller in a positive direction, and the system will quickly approach the set value; When the system error is small and the response speed is constant, adjust the output of the auxiliary fuzzy controller to reduce the total output U of the controller to avoid excessive overshoot; When the system speed starts to drop and the response speed is slow, the total output U of the controller should be further reduced to quickly approach the steady state; When the system falls back to the vicinity of the set point and the response speed is constant, fine-tune the output of the auxiliary fuzzy controller to increase the total output U of the controller in advance and reduce the reverse overshoot; When the system falls back excessively, regardless of the speed of response, the output of the auxiliary fuzzy controller is adjusted to make the total output U of the controller continue to increase, suppress overshoot and speed up the response speed; A fuzzy rule table is designed based on the above adaptive rules.

10. The method for constructing a permanent magnet synchronous motor type PI improved adaptive fuzzy controller according to claim 3, characterized in that: In step S23, the auxiliary fuzzy controller in the PI-like adaptive fuzzy controller has its output variable and its output factor G α The variation law and range of the self-adjusting factor α obtained after multiplication are determined by the two input quantities of the auxiliary fuzzy controller, the fuzzy rules and reasoning methods used, and its output factor G α Joint decision making; The size of the self-adjustment factor α will dynamically affect the degree of adaptive adjustment of the system. The adjustment algorithm for the total output U(k) of the system is designed as follows: At the current time k of the system, the output of the fuzzy controller 1 is: ΔU(k1)=Δu N (k)*G Δu (9) When introducing the self-adjusting mechanism: ΔU(k)=Δu N (k)*G Δu (1+α(k))(10) Among them, since the output ranges of the two fuzzy controllers involved in the design process are both [-1,1], the range of α(k) is [-G α ,G α ], the range of ΔU(k) ​​is [ΔU(k1)*(1-G α ),ΔU(k1)*(1+G α )]; In addition, according to the structure of the PI-like fuzzy controller: U(k)=U(k-1)+ΔU(k)(11) Substituting equation (11) into equation (10), the total output U(k) of the control system at time k can be obtained as follows: U(k)=U(k-1)+Δu N (k)*G Δu (1+α(k))(12)。

Images