Model prediction control method of load commutation inverter driving synchronous motor

Abstract

According to the model prediction control method for the load commutation inverter driving synchronous motor, the dynamic response capacity of a unit to load changes can be remarkably improved, the influence of other system disturbance such as sudden load increasing and sudden load decreasing under different working conditions can be dealt with, and reliable and stable operation of the synchronous motor is achieved; according to the method, the measured quantity is a common quantity, and only the residual computing power under the load commutation low control frequency needs to be applied, so that the system can be controlled more efficiently and more precisely to cope with different conditions encountered by a unit; the method has different control effects on different prediction step lengths and control step lengths, and different control effects can also be generated on the system by setting different weights.

Description

Technical Field

[0001] The invention belongs to the field of synchronous motor speed regulation control and relates to a model prediction control method for a synchronous motor driven by a load commutation inverter. Background Art

[0002] Load commutated inverter (LCI) uses thyristors as its power devices, so it has the advantages of large capacity and high reliability. It is widely used in pumped storage units, synchronous phase regulators, large gas turbines, rolling mills and other main drive systems. With the increasing demand for power transmission control in various industries, in order to meet the application requirements of high control accuracy and dynamic response, many intelligent control algorithms with large computational complexity have gradually been applied to the field of motor control.

[0003] For transmission systems with high control accuracy such as finishing mills, loading and unloading are performed under step load disturbance conditions. The traditional dual closed-loop control proportional-integral (PI) control method has fixed controller parameters, no adaptive ability, poor stability and dynamic response, and has great limitations in dealing with external load disturbances. When the working conditions change, the system's poor dynamic response will inevitably affect the yield rate of steel products and lead to a decline in economic benefits. For the dual closed-loop feedforward control method, if the feedforward coefficient is too large or too small, it is easy to cause problems such as overshoot or insufficient adjustment of the system. At the same time, a large number of defective products will be generated during debugging, increasing the debugging cost of the feedforward system. And the adjustment of the feedforward coefficient of the LCI high-coupling system also increases its debugging difficulty.

[0004] Model predictive control is a hot topic in the research of motor control methods in recent years. The biggest features of this method are high accuracy, good control performance, strong real-time performance and strong fault tolerance. This method adopts online optimization parameters and solves them, replacing the offline design of the traditional double closed-loop PI controller, realizing independent and precise control of the system at all times, and improving the real-time performance of the drive system. The driving strategy of the finishing mill is to drive the torque requirements of the steel to the rollers. Due to frequent changes in working conditions and frequent loading and unloading, the frequent torque pulsation causes vibration and noise of the motor, reduces the reliability and control accuracy of the motor drive system, and increases the additional energy consumption of the motor, shortening the life of the motor. Therefore, for the load-commutated inverter drive system, it is of great research significance to improve the real-time tracking capability of the system speed and the dynamic response capability to cope with changes in working conditions. Summary of the invention

[0005] The present invention aims to improve the real-time tracking capability of the system speed and the dynamic response capability to changes in operating conditions for a load-commutated inverter drive system, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides a model predictive control method for a load commutation inverter driving a synchronous motor, comprising the following steps:

[0008] Step 1: Collect the DC bus current between the rectifier bridge and the inverter bridge of the LCI system, and construct the DC circuit equivalent voltage equation based on the LCI system:

[0009]

[0010] Step 2: Collect the voltage at the inverter bridge end, and use the inverter bridge voltage u2 and bus current i d And the motor speed ω to get the electromagnetic torque T e :

[0011]

[0012] Step 3: Collect the mechanical speed of the motor, construct the mechanical motion equation of the synchronous motor, and obtain the electromagnetic torque equation of the motor based on the collected motor speed and bus current:

[0013]

[0014] Step 4: construct a system state equation according to the DC circuit equivalent voltage equation described in step 1 and the motor torque equation described in step 3;

[0015] Step 5: To ensure the stable operation of the system, the input change of the control quantity should not change in a large range, and the system should adopt the discrete system incremental state equation;

[0016] Step 6: Based on the system measurement value at the current moment, predict the future system state from time k+1 to time k+p, and obtain the dynamic system equation for the future p period;

[0017] Step 7: Select J that minimizes the error between the actual speed, the speed prediction value, and the control amount expected value as the objective function;

[0018] Step 8: According to the optimal control algorithm, the first element Δu(k) of the optimal control sequence is selected and superimposed with the current control signal u(k), that is, the control signal u(k) = Δu(k) + u(k-1), and acts on the system. When the next moment comes, the calculation is repeated, and so on.

[0019] Furthermore, the DC circuit equivalent voltage equation is:

[0020]

[0021] Among them, u1 is the voltage at the rectifier bridge terminal, u2 is the voltage at the inverter bridge terminal, L d is the smoothing reactance, R is the loop resistance, i d is the bus current.

[0022] Furthermore, the motor torque equation is:

[0023]

[0024] Where T e is the electromagnetic torque, T L is the load torque, J is the motor moment of inertia, B m is the viscosity coefficient and ω is the mechanical torque.

[0025] Furthermore, the system state equation selects the bus current i d and motor speed ω as the state quantity, that is, x=[i d (t)ω(t)] T , select the trigger angle of the rectifier bridge as the control quantity, that is, u=[u1(t)0] T , select the load torque of the system as the disturbance d=[u2(t)T L (t)] T , then the system state equation is:

[0026]

[0027] y(t)=Cx(t)

[0028] in C = [0 1],

[0029] Furthermore, the incremental state equation of the discretized system is:

[0030] Δx(k+1)=MΔx(k)+NΔu(k)+WΔd(k)

[0031] y(k)=CΔx(k)+y(k-1)

[0032] in

[0033] Furthermore, the dynamic system equation for the future p period is:

[0034] Y(k+p|k)=S x Δx(k)+Ey(k)+S d Δd(k)+S u ΔU(k)

[0035]

[0036] Furthermore, the objective function J is:

[0037]

[0038] Furthermore, the objective function is transformed into a quadratic function that is only related to the control output ΔU(k), and the terms that are not related to the control output can be ignored. The objective function is:

[0039] J c =ΔU(k) T HΔU(k)+LΔU(k)

[0040] in

[0041] Furthermore, solving the quadratic function J c The optimal control sequence ΔU(k) ​​selects its first sequence as its output to act on the system, and when the next moment comes, repeats the above steps, and so on.

[0042] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0043] 1. The model predictive control method of the load-commutated inverter driving the synchronous motor of the present invention can significantly improve the dynamic response capability of the unit to load changes, can cope with the influence of other system disturbances such as sudden increase and decrease of load under different working conditions, and realize the reliable and stable operation of the synchronous motor;

[0044] 2. The model predictive control method of the load-commutated inverter driving the synchronous motor of the present invention measures commonly used quantities and only requires the use of the remaining computing power of the processor, thereby achieving more efficient and higher-precision control of the system to cope with different situations encountered by the unit;

[0045] 3. The model predictive control method of the load commutation inverter driving the synchronous motor of the present invention has different control effects for different prediction step sizes and control step sizes, and different weight settings can also produce different control effects on the system.

[0046] The model predictive control method for a load-commutated inverter-driven synchronous motor proposed in the present invention can be used in applications where the system has strong dynamic response and high control accuracy requirements: steel finishing mills can reduce speed fluctuations and dynamic stabilization time under variable loads, thereby enhancing the economic benefits of the enterprise; pumped storage applications can better resist interference and have better robustness; large gas turbine applications can improve the unit startup success rate and system reliability; and fans and pumps, when the wind power generation system copes with various working conditions, can optimize the efficiency and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 is the LCI DC loop equivalent circuit model;

[0049] Figure 2 This is the schematic diagram of model predictive control;

[0050] Figure 3 This is the flow chart of the model predictive control algorithm;

[0051] Figure 4 PI and MPC control methods load and unload speed simulation waveforms;

[0052] Figure 5 Loading and unloading simulation waveform under PI control;

[0053] Figure 6 Loading and unloading simulation waveforms under model predictive control

[0054] Figure 7 Speed ​​difference at different step lengths;

[0055] Figure 8 Speed ​​difference under different weight coefficients. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Example 1

[0058] A model predictive control method for a load-commutated inverter-driven synchronous motor is implemented by the following steps:

[0059] Figure 3 For the online solution flow chart of model predictive control, set the initial parameter S x , S d , S u, Q1, Q2, and give the predetermined speed and the measured bus current and speed, solve its optimal value through the objective function J, and judge whether the value is under the constraint condition. If it is confirmed that there is a solution, output its first element to act on the system, otherwise continue to use the control quantity of the previous cycle to act on the system, and repeat this cycle.

[0060] Step 1: Collect the voltage u1, u2 and DC bus current i between the rectifier bridge and inverter bridge of the LCI system d , and construct the DC circuit equivalent voltage equation based on the LCI system.

[0061]

[0062] Step 2: Collect the inverter bridge voltage u2, and use the inverter bridge voltage u2 and bus current i d The electromagnetic torque T is obtained by multiplying the product by the motor speed ω e .

[0063] Step 3: Collect the mechanical speed ω of the motor and convert the load torque T L Add the product of the moment of inertia J and the speed differential plus the product of the viscosity coefficient and the speed to get the electromagnetic torque T e . Construct the mechanical motion equations of the synchronous motor.

[0064] Step 4: Based on the above DC circuit equivalent voltage equation and motor motion equation, construct the system state equation. Select the bus current i d and motor speed ω as the state quantity, that is, x=[i d (t)ω(t)] T , select the trigger angle of the rectifier bridge as the control quantity, that is, u=[u1(t)0] T , select the load torque of the system as the disturbance d=[u2(t)T L (t)] T , then the system state equation is:

[0065]

[0066] y(t)=Cx(t)

[0067] in C = [0 1],

[0068] Step 5: To ensure the stable operation of the system, the input change of the control quantity should not change in a large range, and the system adopts the discrete system incremental state equation:

[0069] Δx(k+1)=MΔx(k)+NΔu(k)+WΔd(k)

[0070] y(k)=CΔx(k)+y(k-1)

[0071] in

[0072] Step 6: Based on the system measurement values ​​at the current moment, predict the future system states from time k+1 to time k+p, and obtain the dynamic system equation for the future p period of time.

[0073] Y(k+p|k)=S x Δx(k)+Ey(k)+S d Δd(k)+S u ΔU(k)

[0074]

[0075] Step 7: Select J that minimizes the error between the actual speed, the predicted speed, and the expected value of the control amount as the objective function:

[0076]

[0077] Step 8: According to the optimal control algorithm, the first element Δu(k) of the optimal control sequence is selected and superimposed with the current control signal u(k), that is, the control signal u(k) = Δu(k) + u(k-1), and acts on the system. When the next moment comes, the calculation is repeated, and so on.

[0078] Figure 1 For the LCI DC loop equivalent circuit model, the voltage equation is established according to the topological structure; Figure 2 This is the model prediction block diagram of the LCI driven synchronous motor. Through the speed setting, constant excitation setting, and the isomorphic model prediction control method, the rectifier and inverter control trigger angles are output; Figure 3 For the online solution flow chart of model predictive control, set the initial parameter S x , S d , S u , Q1, Q2, and give the predetermined speed and the measured bus current and speed, solve its optimal value through the objective function J, and judge whether the value is under the constraint condition. If it is confirmed that there is a solution, output its first element to act on the system, otherwise continue to use the control quantity of the previous cycle to act on the system, and repeat this cycle.

[0079] Figure 4 The speed fluctuations under different loading and unloading conditions for traditional double closed-loop PI control and model predictive control. The loading and unloading action lasts for 1 second in each case. These are the dynamic waveforms under 33%, 66%, and 100% rated load. The blue one is the speed waveform under PI control, and the red one is the speed waveform under model predictive control. Figure 5The waveforms of the motor speed, excitation current, commutation angle and bus current for PI control loading and unloading. Figure 6 The waveform diagram of the motor speed, excitation current, commutation angle and bus current under loading and unloading under model predictive control. It can be seen from the figure that compared with the traditional double closed-loop PI control, model predictive control has better control effects in terms of motor anti-interference, speed fluctuation and dynamic response time.

[0080] Figure 7 The fluctuation of motor speed under different values ​​of prediction step size p and control step size m for model predictive control. Figure 8 The speed weight Q1 and output change weight Q2 of the objective function show the fluctuation of the motor speed under different values.

[0081] So far, the embodiments of the present invention have been described in detail with reference to the drawings. It should be noted that the implementation methods not shown or described in the drawings or the body of the specification are all forms known to ordinary technicians in the relevant technical field and are not described in detail.

[0082] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A model predictive control method for a load-commutated inverter-driven synchronous motor, characterized in that: A model predictive control method for a load-commutated inverter-driven synchronous motor comprises the following steps: Step 1: Collect the DC bus current between the rectifier bridge and the inverter bridge of the LCI system, and construct the DC circuit equivalent voltage equation based on the LCI system: Step 2: Collect the voltage u1 at the rectifier bridge end, and convert the voltage u2 and bus current i d And the motor speed ω to get the electromagnetic torque T e : Step 3: Collect the mechanical speed of the motor, construct the mechanical motion equation of the synchronous motor, and obtain the electromagnetic torque equation of the motor based on the collected motor speed and bus current; Step 4: construct a system state equation according to the DC circuit equivalent voltage equation described in step 1 and the motor torque equation described in step 3; Step 5: To ensure the stable operation of the system, the input change of the control quantity should not change in a large range, and the system should adopt the discrete system incremental state equation; Step 6: Based on the system measurement value at the current moment, predict the future system state from time k+1 to time k+p, and obtain the dynamic system equation for the future p period; Step 7: Select J that minimizes the error between the actual speed, the speed prediction value, and the control amount expected value as the objective function; Step 8: According to the optimal control algorithm, the first element Δu(k) of the optimal control sequence is selected and superimposed with the current control signal u(k-1), that is, the control signal u(k)=Δu(k)+u(k-1), and acts on the system. When the next moment comes, the calculation is repeated, and so on.

2. A model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 1, characterized in that: The DC circuit equivalent voltage equation in step 1 is: Among them, u1 is the voltage at the rectifier bridge terminal, u2 is the voltage at the inverter bridge terminal, L d is the smoothing reactance, R is the loop resistance, i d is the bus current.

3. The model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 1, characterized in that: The electromagnetic torque in step 3 is: Where T e is the electromagnetic torque, T L is the load torque, J is the motor moment of inertia, B m is the viscosity coefficient and ω is the mechanical torque.

4. The model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 1, characterized in that: In step 4, the system state equation selects the bus current i d and motor speed ω as the state quantity, that is, x=[i d (t)ω(t)] T , select the trigger angle of the rectifier bridge as the control quantity, that is, u=[u1(t)0] T , select the load torque of the system as the disturbance d=[u2(t)T L (t)] T , the system state equation is: y(t)=Cx(t) in C=[01], 5. The model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 1, characterized in that: The incremental state equation of the discretized system is: Δx(k+1)=MΔx(k)+NΔu(k)+WΔd(k) y(k)=CΔx(k)+y(k-1) in 6. The model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 1, characterized in that: With the prediction step length of p and the control step length of m, the incremental state equation of the discretized system is transformed into the dynamic system state equation of the p period: Y(k+p|k)=S x Δx(k)+Ey(k)+S d Δd(k)+S u ΔU(k) 7. The model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 1, characterized in that: The speed error and output variation are set as the objective function, and different weights Q1 and Q2 are set respectively. The objective function J is:

8. A model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 7, characterized in that: The objective function is transformed into a quadratic function that is only related to the control output ΔU(k). The terms that are not related to the control output can be ignored. It is expressed as follows: J c =ΔU(k) T HΔU(k)+LΔU(k) in 9. A model predictive control method for a load-commutated inverter-driven synchronous motor according to claim 8, characterized in that: Solve the quadratic function J c The optimal control sequence ΔU(k) ​​selects the sum of its first sequence and the current output as its output to act on the system. When the next moment comes, repeat the above steps, and so on.

10. Application of a model predictive control method for a load-commutated inverter-driven synchronous motor in the control of a synchronous motor with strong system dynamic response and high control accuracy.

Images