A permanent magnet motor sliding mode model predictive control method and system

Abstract

This invention discloses a sliding mode model predictive control method and system for permanent magnet motors. Based on sliding mode control theory, the invention establishes a model predictive control cost function that satisfies the definition of a Lyapunov function. This cost function can adjust the current error by selecting a voltage vector, thereby enabling the cost function to converge within a finite time. The cost function designed in this invention not only undergoes rigorous stability proofs, but when applied to model predictive control strategies, it selects the voltage vector with the minimum cost function as the optimal voltage vector for motor control. This process eliminates the need for current prediction; it directly evaluates the control error of the voltage vector using the reference current, actual current, and voltage vector, thus simplifying the control process and avoiding the influence of motor parameters on control performance. Compared to vector control, this invention retains the advantages of model predictive control in terms of speed and flexible multi-objective control.

Description

Technical Field

[0001] This invention belongs to the field of motor control technology, and more specifically relates to a sliding mode model predictive control method and system for permanent magnet motors. Background Technology

[0002] In the field of permanent magnet motor control, model predictive control (MMC) has attracted much attention due to its simple design, excellent dynamic response, and ability to simultaneously optimize multiple control objectives. Its rapid dynamic response characteristics perfectly meet the requirements of permanent magnet motors in high-precision, high-speed operation scenarios. However, MMC is a model-based control method, and its control performance is highly dependent on the accuracy of the motor model. Once motor parameters (inductance, resistance, flux linkage, etc.) change or become mismatched, the control error will increase significantly, leading to a decrease in system performance. This extreme sensitivity to parameters greatly limits the robustness and reliability of MMC in practical applications.

[0003] Therefore, improving the parameter robustness of model predictive control (MMC) methods in permanent magnet motors has become a key research focus. Introducing model-free predictive control strategies (a type of MMC, significant for avoiding the influence of motor parameters on the predictive model) has become a common option in this field. Currently, the mainstream model-free predictive control method employs a first-order hyperlocal model, equating quantities independent of the voltage vector to "lumped disturbances," and using an observer method to observe and represent them. This significantly reduces the negative impact of model parameter mismatch on control performance, thereby enhancing system stability and adaptability. This approach expands the application scope of MMC and fully taps the enormous potential of permanent magnet motors in high-performance drive systems.

[0004] Furthermore, although numerous experiments have verified the stability of model predictive control, a rigorous theoretical proof is still lacking, which limits its application. The main reason for this is that the cost function does not satisfy the form defined by the Lyapunov function. The stability analysis of the Lyapunov function is the design basis for sliding mode systems and is widely used and accepted; therefore, further improvements are needed. Summary of the Invention

[0005] This invention aims to construct a cost function that satisfies the definition of a Lyapunov function and apply it to sliding mode model predictive control of permanent magnet motors, fundamentally solving the aforementioned problems in existing technologies. Based on the sliding mode concept, a new cost function is redesigned, which avoids the use of motor parameters. Therefore, this invention provides a sliding mode model predictive control method and system for permanent magnet motors. The constructed cost function not only undergoes rigorous stability proof but also includes candidate voltage vectors and current errors. The voltage vector that minimizes the cost function is selected for motor control. Therefore, when applied to model predictive control strategies, there is no need to predict the current; the control error of the voltage vector is directly evaluated using the reference current, actual current, and voltage vector. This eliminates the current prediction process in model predictive control, simplifying the control flow. Compared with traditional model predictive control, this invention's method exhibits smaller voltage control errors and a simpler control flow when parameters are mismatched.

[0006] On the one hand, the present invention provides a sliding mode model predictive control method for a permanent magnet motor, comprising the following steps:

[0007] The current error along the αβ axis in a two-phase stationary coordinate system is selected as the sliding surface. The derivative of the current error sliding surface is then obtained, and based on Lyapunov theory, the necessary and sufficient conditions for the convergence of the current error sliding surface are designed.

[0008] Based on this necessary and sufficient condition, construct a cost function that satisfies the definition of a Lyapunov function;

[0009] The cost function is applied to the sliding mode predictive control of the permanent magnet motor, that is, in each control cycle, each selectable voltage vector is... u αβ Substituting the cost function, the voltage vector with the minimum cost function value is taken as the optimal voltage vector. Then, in the next control cycle, the corresponding inverter switching signal is generated and applied to the inverter of the permanent magnet motor for drive control.

[0010] Optionally, the cost function is:

[0011] ;

[0012] In the formula, k Indicates the gain parameter; α-axis current error Δ i α =i α -i αref β-axis current error Δ i β =i β -i βref; ∫ represents the integral symbol, i α , i β These are the stator current vectors. i αβ α-axis components and β-axis components; i αref , i βref These are the α and β axis reference currents, respectively. i αβref α-axis components and β-axis components; u α , u β These are the output voltage vectors. u αβ The α-axis and β-axis components.

[0013] Alternatively, the process of constructing a cost function that satisfies the definition of a Lyapunov function based on the necessary and sufficient conditions is as follows:

[0014] Construct the following current error sliding surface model:

[0015] ;

[0016] In the formula, x α and x β These represent the current error sliding surfaces corresponding to the α and β axes, respectively. k Indicates the gain parameter; α-axis current error Δ i α =i α -i αref β-axis current error Δ i β =i β -i βref ∫ represents the integral symbol. i α , i β These are the stator current vectors. i αβ The α-axis components and β-axis components, i αref , i βref These are the α and β axis reference currents, respectively. i αβref α-axis components and β-axis components;

[0017] After differentiating the current error sliding surface, the following exists:

[0018] ;

[0019] In the formula, t represents time;

[0020] According to Lyapunov's theory, the necessary and sufficient condition for the convergence of the designed sliding surface is:

[0021] ;

[0022] Construct a cost function that satisfies the definition of a Lyapunov function;

[0023] ;

[0024] In the formula, , These are the costs corresponding to axes a and b, respectively. Let be the cost function.

[0025] Optionally, the cost function can be simplified to:

[0026] ;

[0027] Due to the existence of permanent magnet motor control system , Therefore, by further simplifying the cost function and combining it with the state equation of the permanent magnet motor, we obtain:

[0028] ;

[0029] In the formula, R This represents the stator resistance of a permanent magnet motor; L This represents the stator inductance of a permanent magnet motor; e α and e β Represents the back electromotive force along the α and β axes; u α , u β These are the output voltage vectors. u αβ α-axis components and β-axis components;

[0030] Considering the current within a controller i α , i β and back electromotive force e α , e β For a constant value and current error Δ i α Δ i βIf the values ​​are constant, they will not affect the cost function. g α , g β Given the relative magnitudes, the cost function can be further simplified to:

[0031] ;

[0032] In the formula, 1 / L After removing the proportional term, the cost function is further transformed into... .

[0033] Secondly, the present invention provides a sliding mode model predictive control method for permanent magnet motors, which performs the following steps in each control cycle:

[0034] T1, Sample the rotational speed of the permanent magnet motor at the current moment. oh e ( t Phase angle i ( t and three-phase stator current i abc ( t );

[0035] T2, change the three-phase stator current i abc ( t Transforming the coordinates to the two-phase stationary coordinate system αβ yields the αβ-axis stator current. i αβ ( t );

[0036] T3, Given d-axis reference current i dref =0 and reference speed oh ref ; reference speed oh ref and rotational speed oh e ( t The difference is used to obtain the q-axis reference current through a PI controller. i qref ; through phase angle i ( t ) d-axis reference current i dref and q-axis reference current i qref Transform to the αβ axis to obtain the αβ axis reference current. i αβref ;

[0037] T4, for each selectable voltage vector u αβ, to the voltage vector u αβ Stator current with αβ axis i αβ ( t ) and reference current i αβref Substitute into the cost function g This is used to assess control errors;

[0038] Wherein, the cost function g The method involves selecting the current error along the αβ axis in a two-phase stationary coordinate system as the sliding surface, differentiating the current error sliding surface, and designing the necessary and sufficient conditions for the convergence of the current error sliding surface based on Lyapunov theory. A cost function that satisfies the definition of a Lyapunov function is then constructed based on these necessary and sufficient conditions.

[0039] T5. Select the voltage vector with the minimum cost function value as the optimal voltage vector. u αβ ( t +1), and in the next control cycle, generate the corresponding inverter switching signal, and then apply it to the inverter of the permanent magnet motor for drive control.

[0040] Optionally, the cost function is:

[0041] ;

[0042] In the formula, k Indicates the gain parameter; α-axis current error Δ i α =i α -i αref β-axis current error Δ i β =i β -i βref ; ∫ represents the integral symbol, i α , i β These are the stator current vectors. i αβ α-axis components and β-axis components; i αref , i βref These are the α and β axis reference currents, respectively. i αβref α-axis components and β-axis components; u α , u β These are the output voltage vectors.u αβ The α-axis and β-axis components.

[0043] Optionally, the cost function g is the cost function of the current tracking control objective. If the control objective of the permanent magnet motor sliding mode model predictive control is one or more control objectives in addition to current tracking, then the cost function of the permanent magnet motor sliding mode model predictive control system is the sum of the cost function g and the weighted cost functions of other control objectives.

[0044] The weights of each control objective are adjusted by changing the size of the weighting system.

[0045] Optionally, in step T4, the voltage vector u αβ The acquisition model is:

[0046] ;

[0047] in, S a , S c and S c This indicates the switching state of the inverter; "0" indicates that the upper bridge arm is off and the lower bridge arm is on, and "1" indicates that the upper bridge arm is off and the lower bridge arm is on. u dc Indicates the DC bus voltage; u α , u β These are the output voltage vectors. u αβ The α-axis and β-axis components.

[0048] In three aspects, the present invention provides a control system based on the above method, comprising:

[0049] The feedback module is used to acquire the three-phase stator signals of the permanent magnet motor within the current control cycle. i abc ( t ) and rotational speed oh e ( t ), and phase angle i ( t and voltage vector u αβ ;

[0050] d-axis current control module, setting d-axis current reference value i dref =0;

[0051] The q-axis current control module is used to control the reference speed. oh ref With rotational speed oh e ( t The difference between the two values ​​is used for PI control to obtain the q-axis current reference value. i qref ;

[0052] The coordinate transformation module is used to transform the three-phase stator signals. i abc ( t (Transform to the αβ axis to obtain the αβ axis stator current) i αβ ( t It is also used to measure phase angles. i ( t ) d-axis current reference value i dref and q-axis current reference value i qref Convert to the αβ axis to obtain the αβ axis current reference value. i αβref ;

[0053] The prediction error evaluation module is used to calculate each voltage vector. u αβ The corresponding cost function; where, for each voltage vector u αβ , to the voltage vector u αβ Stator current with αβ axis i αβ ( t ) and reference current i αβref Substitute into the cost function g The magnitude of the cost function represents the control error of the voltage vector. g It is a cost function that satisfies the definition of a Lyapunov function;

[0054] The optimal vector selection module is used to select the voltage vector with the minimum cost function value as the optimal voltage vector. In the next control cycle, it generates the corresponding inverter switching signal so as to apply it to the inverter used to control the permanent magnet motor.

[0055] In four aspects, the present invention provides a permanent magnet motor system, comprising:

[0056] Permanent magnet synchronous motor;

[0057] The inverter has its three-phase bridge arm midpoints connected to the three-phase windings of the permanent magnet motor, respectively.

[0058] A control system is connected to the permanent magnet motor and the inverter, respectively.

[0059] Fifthly, the present invention provides a computer-readable storage medium storing a computer program, which is called by a processor to implement the steps of the above-mentioned sliding mode model predictive control method for a permanent magnet motor.

[0060] Compared with the prior art, the present invention achieves the following effects:

[0061] The technical solution of this invention establishes a model predictive control cost function that satisfies the definition of Lyapunov function based on sliding mode control theory. This cost function is a completely new cost function redesigned based on the idea of ​​sliding mode. It can adjust the current error by selecting the voltage vector, thereby enabling the cost function to converge in a finite time. Furthermore, the cost function can avoid the use of motor parameters.

[0062] The cost function designed in this invention not only allows for rigorous stability verification, but also, when applied to model predictive control strategies, selects the voltage vector with the minimum cost function as the optimal voltage vector for motor control. This process eliminates the need for current prediction; it directly evaluates the control error of the voltage vector using the reference current, actual current, and voltage vector. This simplifies the control process, avoids the influence of motor parameters on control performance, and eliminates the current prediction process required in model predictive control. Compared to existing technologies such as CN18677317A, which require the use of a sliding mode observer to observe lumped disturbances and select the optimal voltage vector, this invention does not require setting additional observation values. Attached Figure Description

[0063] Figure 1 The flowchart of the sliding mode model predictive control method for permanent magnet motors provided in the embodiments of the present invention is shown below.

[0064] Figure 2 A block diagram of a sliding mode model predictive control method for a permanent magnet motor provided in an embodiment of the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. The technical features involved in the various embodiments of the invention described below can be combined with each other as long as they do not conflict with each other.

[0066] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0067] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0068] The present invention provides a sliding mode model predictive control method for permanent magnet motors, which fundamentally solves the technical obstacle that the cost function does not satisfy the definition of the Lyapunov function. The constructed cost function can not only be rigorously proven for stability, but when applied to the model predictive control strategy, it does not require current prediction. Instead, it directly evaluates the control error of the voltage vector through the reference current, the actual current, and the voltage vector, thus eliminating the current prediction process in model predictive control and simplifying the control process.

[0069] This will be explained in detail below.

[0070] The process of constructing the cost function for the technical solution of this invention is as follows:

[0071] (1);

[0072] In the formula, x α and x β These represent the current error sliding surfaces corresponding to the α and β axes, respectively. k Indicates the gain parameter; α-axis current error Δ i α =i α -i αref β-axis current error Δ i β =i β -i βref ∫ represents the integral symbol. i α , i β These are the stator current vectors. i αβ The α-axis components and β-axis components, i αref , iβref These are the α and β axis reference currents, respectively. i αβref The α-axis and β-axis components.

[0073] Differentiating the current error sliding surface, we have:

[0074] (2);

[0075] According to Lyapunov's theory, the necessary and sufficient condition for the convergence of the designed sliding surface is:

[0076] (3);

[0077] To achieve optimal switching state control of the permanent magnet motor, sliding mode control requires selecting a suitable voltage vector. Therefore, a cost function satisfying the definition of a Lyapunov function is established as follows:

[0078] (4);

[0079] In the formula, , These are the costs corresponding to axes a and b, respectively. Let be the cost function.

[0080] After simplification, it becomes:

[0081] (5);

[0082] In permanent magnet motor control systems, the sampling frequency of the actual stator current is much higher than the motor's operating frequency, while the reference current frequency is consistent with the motor's operating frequency. Furthermore, the actual stator current sampled by the control system... i a and i b It exhibits high-frequency ripple related to the switching frequency, while the reference current... i aref and i bref The harmonic content is extremely low. Considering the above conditions, therefore, in the presence of... , Therefore, it can be simplified to:

[0083] (6);

[0084] Further simplification using the state equations of a permanent magnet motor:

[0085] (7);

[0086] (8);

[0087] in, R This represents the stator resistance of a permanent magnet motor; L This represents the stator inductance of a permanent magnet motor; e α and e β Represents the back electromotive force along the α and β axes; u α , u β These are the output voltage vectors. u αβ The α-axis and β-axis components.

[0088] Considering the current within a controller i α , i β and back electromotive force e α , e β It is a constant value and does not affect the cost function. g α , g β The relative magnitude of the values ​​has no impact on the selection of the optimal voltage vector, and can be simplified again to:

[0089] (9);

[0090] Similarly, the current Δ within a controller i α Δ i β It is a constant value and does not affect the cost function. g α , g β Given their relative size, it can be further simplified to:

[0091] (10);

[0092] Furthermore, 1 / L The proportional term can also be removed, so the final cost function is designed as follows:

[0093] (11).

[0094] It should be understood that, regardless of the above formula or other simplified cost function formulas, they are all considered cost functions that satisfy the definition of the Lyapunov function, and can be proven to be stable. They can fundamentally solve the technical problem that the cost function in traditional model predictive control does not satisfy or does not consider the definition of the Lyapunov function. Therefore, the above cost functions can meet the requirements of the technical solution of this invention. In practical applications, the preferred formula (11) is preferred because when evaluating the control error based on this cost function formula, its calculation process not only does not require current prediction, but directly evaluates the control error of the voltage vector through the reference current, actual current and voltage vector, which can save the current prediction process of model predictive control and simplify the control process; moreover, it is independent of motor parameters. Therefore, when the motor parameters (inductance, resistance, flux linkage, etc.) change or become mismatched, it will not cause a large change in the control error.

[0095] Based on the above theoretical analysis, it can be summarized that the technical idea of ​​the sliding mode model predictive control method for permanent magnet motors provided by the present invention is as follows:

[0096] S1: Select the current error of the αβ axis in the two-phase stationary coordinate system as the sliding surface, then differentiate the current error sliding surface and design the necessary and sufficient conditions for the convergence of the current error sliding surface according to Lyapunov theory.

[0097] S2: Construct a cost function that satisfies the definition of a Lyapunov function based on sufficient and necessary conditions;

[0098] S3: Apply the cost function to the sliding mode predictive control of the permanent magnet motor, that is, in each control cycle, the cost function is applied to each selectable voltage vector. u αβ Substituting the cost function, the voltage vector with the minimum cost function value is taken as the optimal voltage vector. Then, in the next control cycle, the corresponding inverter switching signal is generated and applied to the inverter of the permanent magnet motor for drive control.

[0099] In practical applications, the sliding mode model predictive control method for permanent magnet motors provided by this invention performs the following steps in each control cycle:

[0100] T1, Sample the rotational speed of the permanent magnet motor at the current moment. oh e ( t Phase angle i ( t and three-phase stator current i abc ( t ).

[0101] In step T1 of this embodiment, the αβ axis current of the permanent magnet motor during the current control cycle... iαβ ( t ) and rotational speed oh e ( t The methods for obtaining ) include:

[0102] The phase current of the permanent magnet motor during the current control cycle is collected by a current sensor. i a and i b c-phase current i c =- i a - i b ;

[0103] Acquire the angle signal of the permanent magnet linear synchronous motor within the current control cycle. i ( t ), and in accordance with Calculate rotational speed oh e ( t ), To control / sample period length, This is the angle signal corresponding to time t-1;

[0104] It is easy to understand that the phase current signal of a permanent magnet synchronous motor can be easily acquired using current sensors and speed sensors. i a , i b and position signal i ( t ).

[0105] T2, change the three-phase stator current i abc ( t Transforming the coordinates to the two-phase stationary coordinate system αβ yields the αβ-axis stator current. i αβ ( t );

[0106] T3, Given d-axis reference current i dref =0 and reference speed oh ref ; reference speed oh ref and rotational speed oh e ( t The difference is calculated and passed through a PI controller to obtain the q-axis reference current. i qref ; through phase angle i ( t) d-axis reference current i dref and q-axis reference current i qref Transform to the αβ axis to obtain the αβ axis reference current. i αβref ;

[0107] T4, for each selectable voltage vector u αβ , to the voltage vector u αβ Stator current with αβ axis i αβ ( t ) and reference current i αβref Substitute the cost function that satisfies the definition of the Lyapunov function g This is used to assess control errors;

[0108] T5. Select the voltage vector with the minimum cost function value as the optimal voltage vector. u αβ ( t +1), and in the next control cycle, generate the corresponding inverter switching signal, and then apply it to the inverter of the permanent magnet motor for drive control.

[0109] In some embodiments, step T2 uses the following transformation formula to transform the three-phase stator signals. i abc ( t Transform to the two-phase stationary coordinate system αβ to obtain i αβ ( t ):

[0110] (12);

[0111] In the formula, i a , i c and i c Represents current i abc Current components along the a, b, and c axes i α and i β Represents current i αβ Current components on the α and β axes.

[0112] In some embodiments, the q-axis reference current in step T3 i qref and αβ axis reference currenti αβref The calculation formula is as follows:

[0113] (13);

[0114] in, i αref , i βref These are the reference currents for the α and β axes, respectively. i αβref The α-axis components and β-axis components, For the phase angle of the motor, This refers to the motor speed. For reference speed, k p and k i These represent the parameters of the speed loop PI controller. s This represents a differential operator.

[0115] In some embodiments, voltage vector u αβ The calculation formula is as follows:

[0116] (14);

[0117] in, S a , S c and S c This indicates the switching state of the inverter; "0" indicates that the upper bridge arm is off and the lower bridge arm is on, and "1" indicates that the upper bridge arm is off and the lower bridge arm is on. u dc Indicates the DC bus voltage; u α , u β These are the output voltage vectors. u αβ The α-axis and β-axis components.

[0118] It should be understood that in other feasible embodiments, formulas (12)-(14) above can be selectively replaced with reference to the prior art, and the present invention does not impose specific limitations on this.

[0119] In summary, this invention has found that when the cost function in this embodiment is designed to conform to the definition of a Lyapunov function, not only can rigorous stability proofs be performed, but the current prediction process of model predictive control can also be eliminated, simplifying the control flow. Furthermore, the control error of the voltage vector can be evaluated through the cost function without motor parameters, improving the parameter robustness of model predictive control and enhancing the control performance of the motor under parameter mismatch.

[0120] In practical applications, the three-phase windings of the permanent magnet motor are connected to the midpoints of the three-phase bridge arms of the inverter. In the inverter, each phase bridge arm has a switching transistor installed at its upper and lower ends. S a , S b , S c These represent the drive signals of the upper switching transistors of the bridge arms connected to the A, B, and C phase windings of the permanent magnet linear synchronous motor, respectively. 1 represents a high level, and 0 represents a low level. It's easy to understand that a high level is the level that turns on the upper switching transistor and turns off the lower switching transistor, while a low level is the level that turns off the upper switching transistor and turns on the lower switching transistor. Based on the driving state of the switching transistors, there are a total of 8 voltage vectors, as shown in Table 1.

[0121] Table 1 Voltage Vector Relationship Table

[0122]

[0123] The technical solution of this invention addresses the problems of poor parameter robustness and susceptibility to changes in flux linkage parameters in traditional model predictive control of permanent magnet motors. It uses a DC evaluation function of the voltage vector to assess the control performance without motor parameters, eliminating the current prediction process of model predictive control, simplifying the control process, improving the parameter robustness of model predictive control, and enhancing the control performance of the motor under parameter mismatch.

[0124] It should be understood that the above technical solution takes current tracking of the motor control system as the control objective. In other feasible embodiments, if the motor control system has control objectives such as low switching frequency and common-mode voltage suppression in addition to current tracking, other control objectives can be weighted on the basis of the above cost function, and the weight of each control objective can be adjusted by weighting coefficients.

[0125] In a multi-objective permanent magnet motor sliding mode model predictive control system, assuming the motor control system includes two or more control objectives such as current tracking, low switching frequency, and common-mode voltage suppression, the cost function can be designed as follows, and the weights of each control objective can be adjusted by weighting coefficients:

[0126] (15);

[0127] Where G is the cost function of the multi-objective sliding mode predictive control system for permanent magnet motors. , Indicates the weighting coefficient; S a , S c and S c Indicates the switching state of the inverter; u dc This indicates the DC bus voltage.

[0128] This embodiment addresses the complexity of multi-objective control in traditional vector control applications for permanent magnet synchronous motors by directly integrating multiple control objectives into a cost function and adjusting the weights of each control objective through weighting coefficients, thereby achieving multi-objective control of the permanent magnet motor.

[0129] In some embodiments, the present invention also provides a control system based on the above method, including: a feedback module, a d-axis current control module, a q-axis current control module, a coordinate transformation module, a prediction error evaluation module, and an optimal vector selection module connected in sequence or interconnected with each other.

[0130] The feedback module is used to acquire the three-phase stator signals of the permanent magnet motor within the current control cycle. i abc ( t ) and rotational speed oh e ( t ), and phase angle i ( t and voltage vector u αβ .

[0131] d-axis current control module, setting d-axis current reference value i dref =0.

[0132] The q-axis current control module is used to control the reference speed. oh ref With rotational speed oh e ( t The difference between the two values ​​is used for PI control to obtain the q-axis current reference value. i qref .

[0133] The coordinate transformation module is used to transform the three-phase stator signals. i abc ( t (Transform to the αβ axis to obtain the αβ axis stator current) i αβ (t It is also used to measure phase angles. i ( t ) d-axis current reference value i dref and q-axis current reference value i qref Convert to the αβ axis to obtain the αβ axis current reference value. i αβref .

[0134] The prediction error evaluation module is used to calculate each voltage vector. u αβ The corresponding cost function; where, for each voltage vector u αβ , to the voltage vector u αβ Stator current with αβ axis i αβ ( t ) and reference current i αβref Substitute into the cost function g The magnitude of the cost function represents the control error of the voltage vector. g It is a cost function that satisfies the definition of a Lyapunov function.

[0135] The optimal vector selection module is used to select the voltage vector with the minimum cost function value as the optimal voltage vector. In the next control cycle, it generates the corresponding inverter switching signal so as to apply it to the inverter used to control the permanent magnet motor.

[0136] In this embodiment, the specific implementation methods of each module can be referred to the description of the foregoing method embodiments, and will not be repeated here. Furthermore, the above modules are functional modules, and can be implemented in hardware or software.

[0137] In some embodiments, the present invention provides a permanent magnet motor system, including: a permanent magnet synchronous motor, an inverter, and a control system. The midpoints of the three-phase bridge arms of the inverter are respectively connected to the three-phase windings of the permanent magnet motor; the control system is connected to both the permanent magnet motor and the inverter.

[0138] In some embodiments, the present invention provides a computer-readable storage medium storing a computer program, which is invoked by a processor to implement the steps of the above-described sliding mode model predictive control method for a permanent magnet motor. Specifically, steps S1-S3 or steps T1-T5 are executed.

[0139] Please refer to the explanation of the method above for the specific implementation process of each step.

[0140] The readable storage medium is a computer-readable storage medium, which can be an internal storage unit of the hardware and software device described in any of the foregoing embodiments, such as the hard drive or memory of the controller. The readable storage medium can also be an external storage device of the controller, such as a plug-in hard drive, Smart MediaCard (SMC), Secure Digital (SD) card, or Flash Card equipped on the controller. Further, the readable storage medium can include both internal storage units and external storage devices of the controller. The readable storage medium is used to store the computer program and other programs and data required by the controller. The readable storage medium can also be used to temporarily store data that has been output or will be output.

[0141] Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned readable storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0142] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application refers to flowchart illustrations and / or instructions executed by a processor of a method, apparatus (system), and computer program product according to embodiments of this application to create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams. These computer program instructions may also be stored in a computer-readable storage medium capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowchart illustrations and / or one or more block diagrams. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more blocks of a block diagram.

[0143] It should be emphasized that the examples described in this invention are illustrative rather than limiting. Therefore, this invention is not limited to the examples described in the specific embodiments. Any other embodiments derived by those skilled in the art based on the technical solutions of this invention, without departing from the spirit and scope of this invention, whether modifications or substitutions, are also within the protection scope of this invention.

Claims

1. A sliding mode model predictive control method for permanent magnet motors, characterized in that: Includes the following steps: The current error along the αβ axis in a two-phase stationary coordinate system is selected as the sliding surface. The derivative of the current error sliding surface is then obtained, and based on Lyapunov theory, the necessary and sufficient conditions for the convergence of the current error sliding surface are designed. Construct a cost function that satisfies the definition of a Lyapunov function based on sufficient and necessary conditions; The cost function is applied to the sliding mode predictive control of the permanent magnet motor, that is, in each control cycle, each selectable voltage vector is... u αβ Substituting the cost function, the voltage vector with the minimum cost function value is taken as the optimal voltage vector. Then, in the next control cycle, the corresponding inverter switching signal is generated and applied to the inverter of the permanent magnet motor for drive control. The cost function is: ； In the formula, k Indicates the gain parameter; α-axis current error Δ i α =i α -i αref β-axis current error Δ i β =i β -i βref ; ∫ represents the integral symbol, i α , i β These are the stator current vectors. i αβ α-axis components and β-axis components; i αref , i βref These are the α and β axis reference currents, respectively. i αβref α-axis components and β-axis components; u α , u β These are the output voltage vectors. u αβ The α-axis and β-axis components.

2. The method according to claim 1, characterized in that: The process of constructing a cost function that satisfies the definition of a Lyapunov function based on the necessary and sufficient conditions is as follows: Construct the following current error sliding surface model: ； In the formula, x α and x β These represent the current error sliding surfaces corresponding to the α and β axes, respectively. k Indicates the gain parameter; α-axis current error Δ i α =i α -i αref β-axis current error Δ i β =i β -i βref ∫ represents the integral symbol. i α , i β These are the stator current vectors. i αβ The α-axis components and β-axis components, i αref , i βref These are the α and β axis reference currents, respectively. i αβref α-axis components and β-axis components; After differentiating the current error sliding surface, the following exists: ； In the formula, t represents time; According to Lyapunov's theory, the necessary and sufficient condition for the convergence of the designed sliding surface is: ； Construct a cost function that satisfies the definition of a Lyapunov function; ； In the formula, , These are the costs corresponding to axes a and b, respectively. Let be the cost function.

3. The method according to claim 2, characterized in that: The cost function is simplified to: ； Due to the existence of permanent magnet motor control system , Therefore, by further simplifying the cost function and combining it with the state equation of the permanent magnet motor, we obtain: ； In the formula, R This represents the stator resistance of a permanent magnet motor; L This represents the stator inductance of a permanent magnet motor; e α and e β Represents the back electromotive force along the α and β axes; u α , u β These are the output voltage vectors. u αβ α-axis components and β-axis components; Considering the current within a controller i α , i β and back electromotive force e α , e β For a constant value and current error Δ i α Δ i β If the values ​​are constant, they will not affect the cost function. g α , g β Given the relative size, the cost function simplifies to: ； In the formula, 1 / L After removing the proportional term, the cost function is further transformed into... .

4. A sliding mode model predictive control method for permanent magnet motors, characterized in that: In each control cycle, the following steps are performed: T1, Sample the rotational speed of the permanent magnet motor at the current moment. ω e ( t Phase angle θ ( t and three-phase stator current i abc ( t ); T2, change the three-phase stator current i abc ( t Transforming the coordinates to the two-phase stationary coordinate system αβ yields the αβ-axis stator current. i αβ ( t ); T3, Given d-axis reference current i dref =0 and reference speed ω ref ; reference speed ω ref and rotational speed ω e ( t The difference is used to obtain the q-axis reference current through a PI controller. i qref ; through phase angle θ ( t ) d-axis reference current i dref and q-axis reference current i qref Transform to the αβ axis to obtain the αβ axis reference current. i αβref ; T4, for each selectable voltage vector u αβ , to the voltage vector u αβ Stator current with αβ axis i αβ ( t ) and reference current i αβref Substitute into the cost function g This is used to assess control errors; Wherein, the cost function g The method involves selecting the current error along the αβ axis in a two-phase stationary coordinate system as the sliding surface, differentiating the current error sliding surface, and designing the necessary and sufficient conditions for the convergence of the current error sliding surface based on Lyapunov theory. A cost function that satisfies the definition of a Lyapunov function is then constructed based on these necessary and sufficient conditions. T5. Select the voltage vector with the minimum cost function value as the optimal voltage vector. u αβ ( t +1), and in the next control cycle, generate the corresponding inverter switching signal, and then apply it to the inverter of the permanent magnet motor for drive control; The cost function is: ； In the formula, k Indicates the gain parameter; α-axis current error Δ i α =i α -i αref β-axis current error Δ i β =i β -i βref ; ∫ represents the integral symbol, i α , i β These are the stator current vectors. i αβ α-axis components and β-axis components; i αref , i βref These are the α and β axis reference currents, respectively. i αβref α-axis components and β-axis components; u α , u β These are the output voltage vectors. u αβ The α-axis and β-axis components.

5. The method according to claim 4, characterized in that: The cost function g is the cost function of the current tracking control objective. If the control objective of the permanent magnet motor sliding mode model predictive control is one or more control objectives in addition to current tracking, then the cost function of the permanent magnet motor sliding mode model predictive control system is the sum of the cost function g and the weighted cost functions of other control objectives. The weights of each control objective are adjusted by changing the size of the weighting system.

6. A control system based on the method of any one of claims 4-5, characterized in that: include: The feedback module is used to acquire the three-phase stator signals of the permanent magnet motor within the current control cycle. i abc ( t ) and rotational speed ω e ( t ), and phase angle θ ( t and voltage vector u αβ ; d-axis current control module, setting d-axis current reference value i dref =0; The q-axis current control module is used to control the reference speed. ω ref With rotational speed ω e ( t The difference between the two values ​​is used for PI control to obtain the q-axis current reference value. i qref ; The coordinate transformation module is used to transform the three-phase stator signals. i abc ( t (Transform to the αβ axis to obtain the αβ axis stator current) i αβ ( t It is also used to measure phase angles. θ ( t ) d-axis current reference value i dref and q-axis current reference value i qref Convert to the αβ axis to obtain the αβ axis current reference value. i αβref ; The prediction error evaluation module is used to calculate each voltage vector. u αβ The corresponding cost function; where, for each voltage vector u αβ , to the voltage vector u αβ Stator current with αβ axis i αβ ( t ) and reference current i αβref Substitute into the cost function g The magnitude of the cost function represents the control error of the voltage vector. g It is a cost function that satisfies the definition of a Lyapunov function; The optimal vector selection module is used to select the voltage vector with the minimum cost function value as the optimal voltage vector. In the next control cycle, it generates the corresponding inverter switching signal so as to apply it to the inverter used to control the permanent magnet motor.

7. A permanent magnet motor system, characterized in that, include: Permanent magnet synchronous motor; The inverter has its three-phase bridge arm midpoints connected to the three-phase windings of the permanent magnet motor, respectively. The control system of claim 6 is connected to the permanent magnet motor and the inverter respectively.

8. A computer-readable storage medium, characterized in that: The computer program is stored and is invoked by the processor to implement: The steps of the sliding mode model predictive control method for a permanent magnet motor as described in any one of claims 1-5.

Images