Permanent magnet synchronous motor model prediction control method and system for air compressor

Abstract

The invention provides a permanent magnet synchronous motor model prediction control method and system for an air compressor. The system comprises an improved rotating speed controller based on rotating speed loop disturbance observation, a target current corrector, a double-step ideal voltage calculator and a double-step current prediction controller based on current loop disturbance observation. The improved rotating speed controller based on rotating speed loop disturbance observation outputs target quadrature-axis current; the target current corrector corrects the target current value in real time; the two-step ideal voltage calculator calculates an ideal voltage vector considering two-step prediction; according to the two-step current prediction controller based on current loop disturbance observation, current loop disturbance and change of rotating speed in a sampling period are added into a prediction equation, and two-step prediction is carried out. According to the method, a series of problems caused by parameter mismatch due to reasons such as ultrahigh rotating speed and temperature rise of the air compressor driving motor are effectively solved, the fluctuation of the current of the motor is finally reduced, and accurate control over the rotating speed of the air compressor is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of electromechanical coupling control of air compressors for fuel cell systems, and in particular to a model predictive control method and system for a permanent magnet synchronous motor for an air compressor. Background Art

[0002] Turbine air compressors are widely used in the industrial field due to their high efficiency, large flow, low vibration, low noise, and strong stability. Turbine air compressors accelerate the gas through the high-speed rotating impeller, increase the kinetic energy of the gas, and convert the kinetic energy into pressure energy through the diffuser, thereby achieving the purpose of compressing the gas.

[0003] The drive motor of a turbine air compressor is usually an ultra-high-speed permanent magnet synchronous motor. Under normal circumstances, ultra-high-speed permanent magnet synchronous motors have the characteristics of extremely low inductance and extremely small magnetic flux. In actual working processes, due to changes in speed, increased temperature, external interference, fatigue and aging, ultra-high-speed permanent magnet synchronous motors often have obvious parameter mismatches, which in turn leads to increased speed fluctuations and reduced control accuracy, greatly affecting the adaptability of the air compressor control algorithm to working conditions. Therefore, how to improve the control algorithm's response to parameter mismatches of air compressor drive motors, and then adapt to complex working conditions and improve control accuracy has become a hot topic of current research. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a model predictive control method and system for a permanent magnet synchronous motor for an air compressor, which overcomes the influence of speed error and current fluctuation caused by parameter mismatch problems, improves the robustness of the system, improves the control accuracy of the system, alleviates system fatigue caused by current fluctuations, and extends the life of the system.

[0005] To achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a model predictive control method for a permanent magnet synchronous motor for an air compressor, an improved speed controller based on speed loop disturbance observation receives the quadrature-axis current signal and speed error signal of the current sensor at the current moment, and based on the improved sliding mode observation algorithm, simulates the parameter mismatch problem generated by the air compressor drive motor as a load torque, adds it to the improved sliding mode control equation, and outputs the target quadrature-axis current value.

[0006] In a preferred embodiment, the following contents are included:

[0007] 1) Define the improved sliding mode speed control reaching law according to the following equation:

[0008]

[0009] In the formula, s ω is the speed error sliding surface, e ω is the speed error, k lωis a parameter greater than 0, k tω is a parameter greater than 0, α ω is a parameter between 0 and 1, k fω is a parameter greater than 0, ξ ω is a parameter between 0 and 1, δ ω is a parameter greater than 0, σ ω is a parameter greater than 0;

[0010] 2) Define the improved sliding mode speed control saturation function y according to the following equation:

[0011]

[0012] In the formula, β ω is a saturated function boundary layer greater than 0;

[0013] 3) Based on the augmented state space equation that defines the speed loop disturbance as the speed loop disturbance torque, the speed loop disturbance torque observation equation is established, and the speed loop disturbance torque observer convergence rate is defined according to the following equation:

[0014]

[0015] In the formula, λ smo For observer switching gain greater than 1, J cp is the air compressor moment of inertia, e T is the disturbance observation error, s smo is the sliding surface of the speed observation error.

[0016] In a preferred embodiment, the target current corrector receives the current signal of the current sensor at the current moment, the target quadrature-axis current signal of the improved speed controller based on speed loop disturbance observation, and the target direct-axis current signal which is constantly 0, and outputs the target current correction signal through an improved sliding mode control algorithm.

[0017] In a preferred embodiment, the following contents are included:

[0018] 1) Define the direct axis current static difference sliding mode surface s according to the following equation: id , cross-axis current static difference sliding mode surface s iq :

[0019]

[0020] In the formula, e id is the difference between the direct axis target current value and the direct axis current sampling value at the current moment, e iq is the difference between the quadrature axis target current value and the quadrature axis current sampling value at the current moment, δ i is the integral sliding surface gain, g d is the direct axis sliding mode limit function, gq is the cross-axis sliding mode limit function;

[0021] 2) Define the orthogonal axis sliding mode limit function g according to the following equation:

[0022]

[0023] In the formula, α i is a parameter greater than 0;

[0024] 3) Define the current static difference compensation value differential equation according to the following equation:

[0025]

[0026] In the formula, i drc is the direct axis target current compensation value, i qrc is the quadrature axis target current compensation value, L d is the direct-axis inductance, L q is the quadrature-axis inductance, R s is the air compressor drive motor line resistance, k i is a parameter greater than 0, θ i is a parameter greater than 0, and ri is a parameter between 0 and 1.

[0027] In a preferred embodiment, the two-step ideal voltage calculator receives the orthogonal axis current corrected target current signal of the target current corrector and the orthogonal axis voltage signal of the motor voltage sensor at the current moment, and based on the extended model current prediction algorithm taking disturbance into consideration, reversely infers the ideal voltage vector that can reach the orthogonal axis current target correction signal value at time k+1 and k+2, and outputs it to the two-step current prediction controller based on current loop disturbance observation.

[0028] In a preferred embodiment, the following contents are included:

[0029] 1) Define the speed prediction equation as follows:

[0030]

[0031] In the formula, “~” represents prediction, T SG is the controller sampling time;

[0032] 2) According to the following equation, the ideal voltage vector U of the orthogonal axis that reaches the target current value at time k+1 is derived: id1 , U iq1 , the ideal voltage vector U of the orthogonal axis that reaches the target current value at time k+2 is derived id2 , U iq2 ;

[0033]

[0034] In the formula, A U , B U , D U is the extended current prediction state space matrix considering speed change and current loop disturbance, d U is the system disturbance vector.

[0035] In a preferred embodiment, the two-step current prediction controller based on current loop disturbance observation receives the corrected orthogonal axis target current signal of the target current corrector and the ideal voltage vector signal of the two-step ideal voltage calculator, generates a voltage pulse signal, outputs it to the inverter, and drives the air compressor to rotate.

[0036] In a preferred embodiment, the following contents are included:

[0037] 1) Based on the deadbeat current control principle, taking into account the current disturbance caused by motor parameter mismatch, an extended model current prediction algorithm considering disturbance is designed according to the following equation:

[0038]

[0039] Where U d is the direct axis voltage value of the motor voltage sensor, U q is the quadrature axis voltage value of the motor voltage sensor, f d is the observed value of direct axis current disturbance, f q is the observed value of the quadrature-axis current disturbance, γ d is the observer parameter greater than 0, γ q is an observer parameter greater than 0, F d is the direct axis sliding mode control function, F q is the cross-axis sliding mode control function;

[0040] 2) Define the current loop perturbation sliding mode observation function according to the following equation:

[0041]

[0042] In the formula, s dsmo is the sliding surface of the direct axis observation error, s qsmo is the sliding surface of the cross-axis observation error, M d is the adaptive saturation function of the direct axis boundary layer, M q is the cross-axis boundary layer adaptive saturation function, R d is the adaptive reaching law of the direct axis boundary layer, R q is the adaptive reaching law of the cross-axis boundary layer;

[0043] 3) Define the boundary layer adaptive reaching law R according to the following equation:

[0044]

[0045] In the formula, ε c is the approach rate gain greater than 0, e smo is the observation error, s smo is the observed error sliding surface, β c is the boundary layer gain greater than 0, η c is a parameter greater than 0, λ c is a parameter greater than 0;

[0046] 4) Define the boundary layer adaptive saturation function M according to the following equation:

[0047]

[0048] The boundary layer adaptive saturation function uses an inverse tangent function to ensure that the boundary layer thickness does not increase infinitely. When the sliding mode observation error approaches 0, the boundary layer also approaches 0.

[0049] 5) According to the extended model current prediction algorithm considering disturbance, U d (k) U q (k) is defined as the ideal voltage vector from the two-step ideal voltage calculator, and current predictions for steps k+1 and k+2 are performed according to the following evaluation indicators:

[0050]

[0051] Find g i The ideal voltage vector corresponding to the smallest one is the first optimal voltage vector;

[0052] 6) Based on the introduction of duty cycle calculation, the duty cycle is calculated first to obtain the first optimal vector action time, and the zero voltage vector corresponding to the remaining voltage vector action time is replaced by 7 basic voltage vectors, and the combined voltage vector is calculated. The orthogonal axis predicted current corresponding to the combined voltage vector of k+1 steps and k+2 steps is calculated by the extended model current prediction algorithm considering disturbances, and the evaluation index shown is substituted to obtain the second optimal voltage vector.

[0053] The present invention also provides a model predictive control system for a permanent magnet synchronous motor for an air compressor, which runs a model predictive control method for a permanent magnet synchronous motor for an air compressor, comprising an improved speed controller (1) based on speed loop disturbance observation, a target current corrector (2), a two-step ideal voltage calculator (3), and a two-step current predictive controller (4) based on current loop disturbance observation.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The present application provides a model predictive control method and system for a permanent magnet synchronous motor for an air compressor. Compared with a traditional PI speed controller, an improved speed controller based on speed loop disturbance observation is adopted, which reduces overshoot and increases response speed, overcomes the problem of large overshoot of the traditional PI speed controller when facing an ultra-high-speed permanent magnet synchronous motor, ensures that the ultra-high-speed permanent magnet synchronous motor components are not damaged due to excessive overshoot, and simultaneously observes in real time the speed loop disturbance caused by parameter mismatch caused by operating condition changes and fatigue aging during the operation of the air compressor, thereby overcoming the influence of parameter mismatch on control accuracy and speed fluctuation.

[0056] The present application provides a model predictive control method and system for a permanent magnet synchronous motor for an air compressor. The method adopts a target current corrector. Compared with the traditional target current correction algorithm with direct feedback of a low-pass filter, a nonlinear integral sliding surface is used. This method can avoid competing for the control dominance of the speed loop and the current loop over the system before the system is stable, thereby ensuring stable operation of the system. At the same time, without affecting the speed fluctuation and current fluctuation, the current control error caused by the model prediction algorithm is smoothed, thereby further improving the control accuracy.

[0057] The present application provides a model predictive control method and system for a permanent magnet synchronous motor for an air compressor, which adopts a two-step ideal voltage calculator and derives an ideal voltage vector based on an extended current prediction equation that takes into account current loop disturbances and speed changes within a sampling period, thereby reducing current fluctuations while saving controller calculation costs.

[0058] The present application provides a model predictive control method and system for a permanent magnet synchronous motor for an air compressor. The method adopts a two-step current predictive controller based on current loop disturbance observation. Compared with the traditional current disturbance observation algorithm, the method effectively reduces the system jitter problem of the current loop caused by disturbance observation. At the same time, based on the extended model current prediction algorithm that considers disturbance, the predicted current of the alternative voltage vector in k+1 step and k+2 step is derived, and a new evaluation index is substituted to finally select the first optimal voltage vector and the second optimal voltage vector, so as to better adapt to the actual situation that there is a control delay of two beats from the controller to the inverter and from the inverter to the motor. At the same time, the influence of current fluctuation caused by parameter mismatch is overcome, the robustness of the system is improved, the system fatigue caused by current fluctuation is reduced, and the life of the system is extended. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a schematic diagram of a model predictive control system for a permanent magnet synchronous motor for an air compressor according to a preferred embodiment of the present invention.

[0060] Figure 2 It is a schematic diagram of the interior of an improved speed controller based on speed loop disturbance observation according to a preferred embodiment of the present invention.

[0061] Figure 3 Schematic diagram of the interior of the target current corrector according to a preferred embodiment of the present invention.

[0062] Figure 4 It is an internal schematic diagram of a two-step current prediction controller based on current loop disturbance observation according to a preferred embodiment of the present invention.

[0063] Figure 5 This is a control effect diagram of the preferred embodiment of the present invention in response to speed step response and load step disturbance under the condition of parameter mismatch.

[0064] Figure 6 This is a control effect diagram of the preferred embodiment of the present invention in response to speed cosine response and load sinusoidal disturbance under the condition of parameter mismatch.

[0065] In the figure: 1-improved speed controller based on speed loop disturbance observation, 2-target current corrector, 3-two-step ideal voltage calculator, 4-two-step current prediction controller based on current loop disturbance observation. DETAILED DESCRIPTION

[0066] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0067] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0068] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or their combinations.

[0069] See also Figure 1 The present invention provides a model predictive control method and system for a permanent magnet synchronous motor for an air compressor, comprising an improved speed controller 1 based on speed loop disturbance observation, a target current corrector 2, a two-step ideal voltage calculator 3, and a two-step current predictive controller 4 based on current loop disturbance observation.

[0070] The improved speed controller 1 based on speed loop disturbance observation receives the quadrature-axis current signal and speed error signal of the motor current sensor at the current moment, adds the load disturbance to the improved sliding mode control equation based on the improved sliding mode observation algorithm, and outputs the target quadrature-axis current value; the target current corrector 2 receives the current signal from the current sensor at the current moment, the target quadrature-axis current signal of the improved speed controller 1 based on speed loop disturbance observation, and the target direct-axis current signal which is always 0, and outputs the target current correction signal through the improved sliding mode control algorithm; the two-step ideal voltage calculator 3 receives the target The orthogonal axis current corrected target current signal of the target current corrector 2 and the orthogonal axis voltage signal of the motor voltage sensor at the current moment are used to reversely infer the ideal voltage vector that can reach the orthogonal axis current target correction signal value at time k+1 and k+2, and output it to the two-step current prediction controller 4 based on current loop disturbance observation; the two-step current prediction controller 4 based on current loop disturbance observation receives the corrected orthogonal axis target current signal from the target current corrector 2 and the ideal voltage vector signal from the two-step ideal voltage calculator 3, generates a voltage pulse signal, and outputs it to the inverter to drive the air compressor to rotate.

[0071] See also Figure 2 The working principle and design method of the improved speed controller 1 based on speed loop disturbance observation are as follows:

[0072] 1) Design the disturbance observation equation according to the following equation:

[0073]

[0074] In the formula, ω m is the mechanical angular velocity of the air compressor, T d is the speed ring disturbance torque, B is the viscous friction coefficient of the air compressor drive motor, J cp is the air compressor rotation inertia, P n is the number of pole pairs of the air compressor drive motor, The nameplate magnetic flux of the air compressor drive motor, i q (k) is the quadrature axis current sampling value of the motor current sensor at the current moment.

[0075] 2) Design the disturbance observation error equation according to the following equation:

[0076]

[0077] In the formula, “^” represents observation, l d is the observer gain greater than 0, and Z is the observer control rate.

[0078] 3) Design the observer sliding surface s according to the following equation smo :

[0079]

[0080] In the formula, c ωsmo Integration parameters.

[0081] 4) Define the sliding mode approach rate according to the following equation:

[0082]

[0083] In the formula, ξ smo is the controllable parameter of the observer.

[0084] 5) Define the observer control rate Z according to the following equation:

[0085]

[0086] In the formula, μ is a parameter greater than 1.

[0087] 6) Establish the control-oriented air compressor rotation differential equation:

[0088]

[0089] Where, T e is the electromagnetic torque of the air compressor drive motor, T cp is the compression torque at the compression end of the air compressor, T exp It is the expansion torque of the expansion end of the air compressor.

[0090] 7) Establish the air compressor speed error sliding surface according to the following equation:

[0091]

[0092] In the formula, e ω =ω ref -ω m ,ω ref is the target angular velocity, c ω is the speed error integral coefficient.

[0093] 8) Define the speed equation considering the speed ring disturbance as follows:

[0094]

[0095] 9) The improved sliding mode speed control reaching law is defined as follows:

[0096]

[0097] In the formula, k lω is a parameter greater than 0, k tω is a parameter greater than 0, α ω is a parameter between 0 and 2, kfω is a parameter greater than 0, ξ ω is a parameter between 0 and 1, δ ω is a parameter greater than 0, σ ω A parameter greater than 0.

[0098] 10) Establish the improved sliding mode speed control saturation function y according to the following equation:

[0099]

[0100] In the formula, β ω A parameter greater than 0.

[0101] 11) Output the target quadrature axis current signal i according to the following equation: qr :

[0102]

[0103] See also Figure 3 , the working principle and design method of the target current corrector 2 are as follows:

[0104] 1) Define the direct axis current error e according to the following equation: id , quadrature axis current error e iq :

[0105]

[0106] 2) Define the direct axis current static difference sliding mode surface s according to the following equation: id , the sliding surface of the cross-axis current error s iq :

[0107]

[0108] In the formula, δ i is the sliding mode gain, g d is the direct axis sliding mode control rate, g q is the cross-axis sliding mode control rate.

[0109] 3) Define the orthogonal axis sliding mode control rate g according to the following equation:

[0110]

[0111] 4) Define the sliding mode approach rate according to the following equation:

[0112]

[0113] In the formula, k i is a parameter greater than 0, θ i is a parameter greater than 0, and ri is a parameter between 0 and 1.

[0114] 5) Output the direct axis target current compensation value i according to the following equation: drc , quadrature axis target current compensation value i qrc :

[0115]

[0116] Where, L d is the direct-axis inductance, L q is the quadrature-axis inductance, R s It is the line resistance of the air compressor drive motor.

[0117] 6) Output the direct axis target current correction value i according to the following equation: drr , quadrature axis target current correction value i qrr :

[0118]

[0119] See also Figure 4 The working principle and design method of the two-step current prediction controller 4 based on current loop disturbance observation are as follows:

[0120] 1) According to the basic voltage-current equation of permanent magnet synchronous motor, the sliding mode disturbance observation differential equation is defined as follows:

[0121]

[0122] In the formula, γ d is a parameter greater than 0, γ q is a parameter greater than 0, F d is the direct axis sliding mode control function, F q is the cross-axis sliding mode control function.

[0123] 2) Define the direct axis sliding mode observation error e according to the following equation: dsmo , cross-axis sliding mode observation error e qsmo :

[0124]

[0125] 3) Define the direct axis integral sliding surface s according to the following equation: dsmo , cross-axis integral sliding surface s qsmo :

[0126]

[0127] In the formula, c smo is the observer integral gain parameter.

[0128] 4) Define the sliding mode observer approach rate according to the following equation:

[0129]

[0130] Where M d is the adaptive saturation function of the direct axis boundary layer, M q is the cross-axis boundary layer adaptive saturation function, R d is the adaptive reaching law of the direct axis boundary layer, R q is the adaptive reaching law for the cross-axis boundary layer.

[0131] 5) Define the boundary layer adaptive reaching law R according to the following equation:

[0132]

[0133] In the formula, ε c is the approach rate gain greater than 0, e smo is the observation error, s smo is the observed error sliding surface, β c is the boundary layer gain greater than 0, η c is a parameter greater than 0, λ c A parameter greater than 0.

[0134] 6) Define the boundary layer adaptive saturation function M according to the following equation:

[0135]

[0136] 7) According to the following equation, the sliding mode control function is derived:

[0137]

[0138] 8) Based on the deadbeat current control principle and taking into account the motor parameter mismatch, the extended model current prediction algorithm is designed according to the following equation:

[0139]

[0140] In the formula, “~” represents prediction, U d is the direct axis voltage value of the motor voltage sensor, U q is the quadrature-axis voltage value of the motor voltage sensor.

[0141] 9) According to the following process, retrieve the first optimal voltage vector U pot1 :

[0142] It is known that the voltage component U of the ideal voltage vector of the two-step ideal voltage calculator 3 under the orthogonal axis is id1 , U iq1 , U id2 , U iq2 , and obtain U by inverse Park transform alpha1 , U beta1 , U alpha2, U beta2 , and then get the theoretical optimal voltage vector sector N i1 、N i2 .

[0143] According to N i1 、N i2 , get the first optimal voltage vector U A1 , U A2 , U A3 , U A4 , calculate the current component U under the orthogonal axis respectively Ad1 , U Aq1 , U Ad2 , U Aq2 , U Ad3 , U Aq3 , U Ad4 , U Aq4 , and then respectively:

[0144]

[0145] Substitute them into the predicted current state space equation respectively to obtain the current prediction values ​​at time k+1 and time k+2 respectively, and substitute the obtained current prediction values ​​into the current prediction evaluation index shown below:

[0146]

[0147] Let g i The candidate first optimal voltage vector corresponding to the smallest one is the first optimal voltage vector.

[0148] 10) According to the following equation, calculate the zero voltage vector action i q The slope S0 is:

[0149]

[0150] 11) According to the basic principle of the two-level three-bridge inverter, there are 8 basic spatial voltage vectors. After removing a repeated zero voltage vector, the remaining 7 basic spatial voltage vectors are made the second best candidate voltage vectors.

[0151] 12) According to the following equation, the seven second optimal voltage vectors are calculated respectively when i q The slope S opt2 :

[0152]

[0153] Where U qq are the quadrature-axis voltage components corresponding to the second optimal candidate voltage vectors respectively.

[0154] 13) Calculate the first optimal voltage vector action time T according to the following equation: opt1 :

[0155]

[0156] 14) According to the following equation, the direct-axis voltage component U corresponding to the 7 synthetic voltage vectors synthesized by the 7 second candidate optimal voltage vectors and the first optimal voltage vector is obtained: mixd , quadrature axis voltage component U mixq :

[0157]

[0158] Where U opt1d For U opt1 The direct axis voltage component, U dd are the direct-axis voltage components corresponding to the second optimal candidate voltage vectors respectively.

[0159] 15) Substituting the 7 synthetic voltage vectors into the predicted current state space equation respectively, obtaining 14 two-step current prediction values, substituting them into the current prediction evaluation index, g i The composite voltage vector corresponding to the smallest one is the optimal composite voltage vector, and the second optimal candidate voltage vector corresponding to the optimal composite voltage vector is the second optimal voltage vector.

[0160] 16) The first optimal voltage vector and the second optimal voltage vector signals are output to the inverter, and finally drive the air compressor to rotate.

[0161] As a preferred method, an extended Kalman filter speed observer is designed to receive the current signal of the motor current sensor and the voltage signal of the motor voltage sensor at the current moment, output the air compressor angular velocity observation value, realize the vector control of the air compressor without position sensor, save the manufacturing cost of the air compressor drive motor position sensor, and reduce the overall size of the air compressor. The specific design method includes the following steps:

[0162] 1) According to the voltage and current equation of the permanent magnet synchronous motor, the current equation of the air compressor drive motor in the stationary coordinate system is established:

[0163]

[0164] In the formula, i α is the current component of the Alpha axis, i β is the current component of the Beta axis, u α is the voltage component of the Alpha axis, u β is the voltage component of the Beta axis, L s is the motor inductance, θ e is the motor electrical angle.

[0165] 2) Based on the basic principles of model prediction, establish the prediction equation:

[0166]

[0167] In the formula, “~” represents prediction and “~” represents state observation.

[0168] 3) Calculate the output prediction value:

[0169]

[0170] 4) Calculate the error covariance matrix:

[0171]

[0172] 5) Define Q and R matrices:

[0173] Q=cos(V)

[0174] R=cos(W)

[0175] Where V is the system noise with a mathematical expectation of 0, and W is the measurement noise with a mathematical expectation of 0.

[0176] 6) Calculate the gain matrix K:

[0177]

[0178] 7) Calculate the optimized state estimation vector:

[0179]

[0180] 8) Calculate the error covariance matrix of the next prediction step, iterate sequentially, and finally output the air compressor angular velocity observation value:

[0181]

[0182] Preferably, the model predictive control method and system of a permanent magnet synchronous motor for an air compressor provided by the present invention require that the permanent magnet synchronous motor control integrator adopts a DSP multi-core processor with a program memory of 1MB, a data memory of 256KB, a stack memory of 8KB, a main frequency of 200MHz, a bit width of 32 bits, and support for single-precision floating-point operations; a high-precision ADC with a 32-bit resolution, a sampling rate of 200kHz, and a signal bandwidth of 50kHz; a MOSFET power device with a switching frequency of 100kHz, a power voltage of 600V, and SiC material; a filter and a protection circuit are used to reduce the influence of noise and vibration; the EtherCAT communication protocol is used, and the transmission rate is required to be up to 50Mbps; current and voltage sensors with a frequency of more than 50kHz are used to provide a solid hardware foundation for the model predictive control system of a permanent magnet synchronous motor for an air compressor provided by the present invention, so as to ensure that the system can operate efficiently, stably and safely.

[0183] Carry out the parameter mismatch situation of the driving motor flux being 0.5 times of the nominal value, the inductance being 0.5 times of the nominal value, and the resistance being 1.5 times of the nominal value; the simulation time is 2s; the simulation of the step signal with the target speed being 80000rpm. Starting from the traditional control system, the improved speed controller 1 based on speed loop disturbance observation, the target current corrector 2, the two-step ideal voltage calculator 3, and the two-step current prediction controller 4 based on current loop disturbance observation described in the present invention are added in sequence, and the absolute value of the speed error is less than 100rpm and no longer exceeds 100rpm with time, which is defined as the system entering a stable state. The speed error after the system stabilizes, the quadrature axis current fluctuation after the system stabilizes, the maximum speed overshoot, and the stabilization time data and their analysis are as follows:

[0184] As shown in Table 1, system number (1) represents an air compressor system that adopts traditional PI speed control and traditional model predictive current control, system number (2) represents an air compressor system that adopts an improved speed controller 1 based on speed loop disturbance observation and traditional model predictive current control, system number (3) represents an air compressor system that adopts an improved speed controller 1 based on speed loop disturbance observation and a two-step current predictive controller 4 based on current loop disturbance observation, system number (4) represents an air compressor system that adopts an improved speed controller 1 based on speed loop disturbance observation, a two-step current predictive controller 4 based on current loop disturbance observation and a two-step ideal voltage calculator 3, and system number (5) represents the air compressor system of the present invention.

[0185] As shown in Table 1: starting from the traditional control system, with the sequential addition of the improved speed controller 1 based on speed loop disturbance observation, the target current corrector 2, the two-step ideal voltage calculator 3, and the two-step current prediction controller 4 based on current loop disturbance observation described in the present invention, the system stabilization time, the average absolute error (MAE) of the speed after stabilization, the average relative error (MRE) of the speed after stabilization, the root mean square error (RMSE) of the speed after stabilization, the maximum overshoot, the MAE of the cross-axis current after stabilization, the MRE of the cross-axis current after stabilization, and the RMSE of the cross-axis current after stabilization all show a decreasing trend; among them, from system number (three) to system number (four), all the data have increased, and the reason is that the two-step current prediction controller 4 is added. After the ideal voltage calculator 3, the current fluctuation is reduced, resulting in an increase in the interval between the target current and the actual current, showing an increase in the error; the stabilization time of the system of the present invention from system number (one) to system number (five) is reduced by 66.30%, the maximum overshoot is reduced by 99.9909%, the MAE of the speed after stabilization is reduced by 95.72%, the MRE of the speed after stabilization is reduced by 96.07%, the RMSE of the speed after stabilization is reduced by 90.27%, the MAE of the cross-axis current after stabilization is reduced by 72.32%, the MRE of the cross-axis current after stabilization is reduced by 64.32%, and the RMSE of the cross-axis current after stabilization is reduced by 72.49%. The present invention embodies good adaptability to working conditions in dealing with parameter mismatch problems.

[0186] Table 1. Data table of simulation results of various systems under parameter mismatch conditions

[0187]

[0188] See also Figure 5 The present invention is simulated under the condition that the flux of the driving motor is 0.5 times of the nominal value, the inductance is 0.5 times of the nominal value, the resistance is 1.5 times of the nominal value, the target speed is a step signal starting from 0.4s, from 60000rpm, increasing by 10000rpm every 0.2s, to 100000rpm, and then decreasing by 10000rpm every 0.2s; and a cyclic step load disturbance is applied, in which a sudden load torque mutation of 3Nm is applied from 0.3s, and a sudden load torque mutation of 3Nm is reduced every 0.2s, and a sudden load torque mutation of 3Nm is added again every 0.2s. The results are as follows. Figure 5 As shown, the present invention can always maintain high-precision tracking of the target speed, and when faced with a sudden load change, the speed error returns to zero in a very short time.

[0189] See also Figure 6The present invention is simulated under the conditions that the flux of the driving motor is 0.5 times the nominal value, the inductance is 0.5 times the nominal value, the resistance is 1.5 times the nominal value, the target speed is a cosine signal with an amplitude of 25000rpm, a bias of 55000rpm, and a frequency of 7.85Hz, and the load disturbance is a sinusoidal signal with an amplitude of 3Nm, a bias of 1.5Nm, and a frequency of 7.85Hz. The results are as follows Figure 6 As shown, the present invention can always maintain high-precision tracking of the target rotation speed and the target current curve, reflecting good adaptability to working conditions.

[0190] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A model predictive control method for a permanent magnet synchronous motor for an air compressor, characterized in that: The improved speed controller based on speed loop disturbance observation receives the quadrature-axis current signal and speed error signal of the current sensor at the current moment, and based on the improved sliding mode observation algorithm, simulates the parameter mismatch problem generated by the air compressor drive motor as a load torque, adds it to the improved sliding mode control equation, and outputs the target quadrature-axis current value.

2. A model predictive control method for a permanent magnet synchronous motor for an air compressor according to claim 1, characterized in that: Includes the following: 1) Define the improved sliding mode speed control reaching law according to the following equation: In the formula, s ω is the speed error sliding surface, e ω is the speed error, k lω is a controller parameter greater than 0, k tω is a parameter greater than 0, α ω is a parameter between 0 and 1, k fω is a parameter greater than 0, ξ ω is a parameter between 0 and 1, δ ω is a parameter greater than 0, σ ω is a parameter greater than 0; 2) Define the improved sliding mode speed control saturation function y according to the following equation: In the formula, β ω is a saturated function boundary layer greater than 0; 3) Based on the augmented state space equation that defines the speed loop disturbance as the speed loop disturbance torque, the speed loop disturbance torque observation equation is established, and the speed loop disturbance torque observer convergence rate is defined according to the following equation: In the formula, λ smo is the observer gain greater than 1, J cp is the air compressor moment of inertia, e T is the disturbance observation error, s smo is the sliding surface of the speed observation error.

3. The model predictive control method for a permanent magnet synchronous motor for an air compressor according to claim 1, characterized in that: The target current corrector receives the current signal of the current sensor at the current moment, the target quadrature-axis current signal of the improved speed controller based on speed loop disturbance observation, and the target direct-axis current signal which is always 0, and outputs the target current correction signal based on the improved sliding mode control algorithm.

4. A model predictive control method for a permanent magnet synchronous motor for an air compressor according to claim 3, characterized in that: Includes the following: 1) Define the direct axis current static difference sliding mode surface s according to the following equation: id , cross-axis current static difference sliding surface s iq : In the formula, e id is the difference between the direct axis target current value and the direct axis current sampling value at the current moment, e iq is the difference between the quadrature axis target current value and the quadrature axis current sampling value at the current moment, δ i is the integral sliding surface gain, g d is the direct axis sliding mode limit function, g q is the cross-axis sliding mode limit function; 2) Define the orthogonal axis sliding mode limit function g according to the following equation: In the formula, α i is a parameter greater than 0; 3) Define the current static difference compensation value differential equation according to the following equation: In the formula, i drc is the direct axis target current compensation value, i qrc is the quadrature axis target current compensation value, L d is the direct-axis inductance, L q is the quadrature-axis inductance, R s is the air compressor drive motor line resistance, k i is a parameter greater than 0, θ i is a parameter greater than 0, and ri is a parameter between 0 and 1.

5. The model predictive control method for a permanent magnet synchronous motor for an air compressor according to claim 1, characterized in that: The two-step ideal voltage calculator receives the orthogonal axis current corrected target current signal of the target current corrector and the orthogonal axis voltage signal of the voltage sensor at the current moment, and based on the extended model current prediction algorithm taking disturbance into consideration, reversely infers the ideal voltage vector that can reach the orthogonal axis current target correction signal value at time k+1 and k+2, and outputs it to the two-step current prediction controller based on current loop disturbance observation.

6. A model predictive control method for a permanent magnet synchronous motor for an air compressor according to claim 5, characterized in that: Includes the following: 1) Define the speed prediction equation as follows: In the formula, "~" represents prediction, T SG is the controller sampling time; 2) According to the following equation, the ideal voltage vector U of the orthogonal axis that reaches the target current value at time k+1 is derived: id1 , U iq1 , the ideal voltage vector U of the orthogonal axis that reaches the target current value at time k+2 is derived id2 , U iq2 ; In the formula, A U , B U , D U is the extended current prediction state space matrix considering speed change and current loop disturbance, d U is the system disturbance vector.

7. The model predictive control method for a permanent magnet synchronous motor for an air compressor according to claim 1, characterized in that: The two-step current prediction controller based on current loop disturbance observation receives the corrected orthogonal axis target current signal of the target current corrector and the ideal voltage vector signal of the two-step ideal voltage calculator, generates a voltage pulse signal, outputs it to the inverter, and drives the air compressor to rotate.

8. A model predictive control method for a permanent magnet synchronous motor for an air compressor according to claim 7, characterized in that: Includes the following: 1) Based on the deadbeat current control principle, taking into account the current disturbance caused by motor parameter mismatch, an extended model current prediction algorithm considering disturbance is designed according to the following equation: Where U d is the direct axis voltage value of the motor voltage sensor, U q is the quadrature axis voltage value of the motor voltage sensor, f d is the observed value of direct axis current disturbance, f q is the observed value of the quadrature-axis current disturbance, γ d is a parameter greater than 0, γ q is a parameter greater than 0, F d is the direct axis sliding mode control function, F q is the cross-axis sliding mode control function; 2) Define the current loop perturbation sliding mode observation function according to the following equation: In the formula, s dsmo is the sliding surface of the direct axis observation error, s qsmo is the sliding surface of the cross-axis observation error, M d is the adaptive saturation function of the direct axis boundary layer, M q is the cross-axis boundary layer adaptive saturation function, R d is the adaptive reaching law of the direct axis boundary layer, R q is the adaptive reaching law of the cross-axis boundary layer; 3) Define the boundary layer adaptive reaching law R according to the following equation: In the formula, ε c is the approach rate gain greater than 0, e smo is the observation error, s smo is the observed error sliding surface, β c is the boundary layer gain greater than 0, η c is a parameter greater than 0, λ c is a parameter greater than 0; 4) Define the boundary layer adaptive saturation function M according to the following equation: 5) According to the extended model current prediction algorithm considering disturbance, U d (k) U q (k) is defined as the ideal voltage vector from the two-step ideal voltage calculator, and current predictions for steps k+1 and k+2 are performed according to the following evaluation indicators: Find g i The ideal voltage vector corresponding to the smallest one is the first optimal voltage vector; 6) Based on the introduction of duty cycle calculation, the duty cycle is calculated first to obtain the first optimal vector action time, and the zero voltage vector corresponding to the remaining voltage vector action time is replaced by 7 basic voltage vectors, and the combined voltage vector is calculated. The orthogonal axis predicted current corresponding to the combined voltage vector of k+1 steps and k+2 steps is calculated by the extended model current prediction algorithm considering disturbances, and the evaluation index shown is substituted to obtain the second optimal voltage vector.

9. A model predictive control system for a permanent magnet synchronous motor for an air compressor, characterized in that: A model predictive control method for a permanent magnet synchronous motor for an air compressor as described in any one of claims 1 to 8 above comprises an improved speed controller (1) based on speed loop disturbance observation, a target current corrector (2), a two-step ideal voltage calculator (3) and a two-step current predictive controller (4) based on current loop disturbance observation.

Images