Boost PFC converter dynamic boost control method based on PI-MPC control interleaved parallel connection

Abstract

The invention discloses a PI-MPC control-based interleaved Boost PFC converter dynamic boost control method, which specifically comprises the following steps: step 1, acquiring an uncontrolled rectified voltage Vrec, an output voltage Vo and an output current Io of an interleaved Boost PFC converter for a current inner ring to obtain a reference current Iref, and calculating ga when a switching tube is switched on and goff when the switching tube is switched off according to the reference current Iref; 2, acquiring phases of output voltage Vo, input voltage Vin and input current iin for a voltage outer ring, optimizing by using a particle swarm algorithm to obtain optimal kp and ki of the voltage ring, and obtaining high pulse vPH and low pulse vPL as driving pulse signals through operation; and step 3, constructing a cost function gopt to obtain an effective control pulse P. The problem that the Boost PFC dynamic response speed is low in existing ACM control is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of electronic power, and relates to a dynamic boost control method of a Boost PFC converter based on PI-MPC control interleaved parallel connection. Background Art

[0002] In modern power systems, with the widespread use of various electronic devices, the requirements for power quality are increasing. In order to improve the efficiency of power utilization and reduce grid pollution, power factor correction (PFC) technology is essential. Boost PFC converters have the advantages of increasing output voltage and reducing line current ripple due to the series connection of the input inductor. They are widely used in many AC-DC power conversion scenarios, such as switching power supplies and electronic rectifiers.

[0003] Average current mode control (ACM) is usually used in Boost PFC converters in continuous conduction mode (CCM). ACM controllers have two control loops - current control loop and voltage control loop. Its current control loop is used to adjust the average inductor current to follow the current reference, which is obtained by multiplying the rectified input voltage with the output of the compensator in the voltage control loop. Since the pulsating input power of the Boost PFC converter does not match the constant DC output power, a double frequency output voltage ripple will be generated. After this ripple enters the current control loop, it will cause serious distortion of the input current, thereby reducing the input power factor. In addition, in order to suppress the interference of the double frequency output voltage ripple on the current control, the voltage loop bandwidth is often designed to be much lower than the double frequency (usually less than 20Hz). This makes the voltage control loop compensator like a low-pass filter. Although it can attenuate the ripple, it leads to a slow dynamic response speed, which is easy to cause large output voltage overshoot and undershoot, affecting the stable operation of the entire system. Summary of the invention

[0004] The purpose of the present invention is to provide a dynamic boost control method for Boost PFC converters based on PI-MPC control interleaved parallel connection, which solves the problem of slow dynamic response speed of Boost PFC in existing ACM control.

[0005] The technical solution adopted by the present invention is a dynamic boost control method of a Boost PFC converter with staggered parallel control based on PI-MPC, which specifically includes the following steps:

[0006] Step 1: collect the uncontrolled rectifier voltage V of the interleaved parallel BoostPFC converter for the current inner loop rec , output voltage V o , output current I o, , and obtain the reference current I ref , according to the reference current I refCalculate the g when the switch is turned on on and g when the switch is turned off off ;

[0007] Step 2: Collect the output voltage V for the voltage outer loop o 、Input voltage V in and input current i in Phase, the particle swarm algorithm is used to find the optimal k for the voltage loop p , k i , and obtain the high pulse v through calculation PH and low pulse v PL As a driving pulse signal;

[0008] Step 3: Combine steps 1 and 2 to construct the cost function g opt , and obtain the effective control pulse P, which in turn controls the interleaved parallel Boost PFC converter.

[0009] The present invention is also characterized in that:

[0010] The specific process of step 1 is:

[0011] Step 1.1, collect the uncontrolled rectifier voltage V rec , output voltage V o , output current I o , Assuming the converter efficiency is 100%, when the CCM Boost PFC converter is unity power factor, the peak current I of the total inductor is obtained m :

[0012]

[0013] v rec =V m |sin(ωt)| (2)

[0014]

[0015] Among them, V in is the inductor input voltage, I in is the input current, V m is the peak voltage, I m is the peak current, w is the line frequency, and t is the time;

[0016] Step 1.2, let V ref =V o , the peak current I obtained from step 1.1 m , the reference current i ref , as shown in the following formula (4):

[0017]

[0018] Step 1.3, set the reference current i ref Discretize i ref (t n ), as shown in the following formula (5):

[0019]

[0020] Among them, t n is the nth moment;

[0021] Step 1.4, collect the total inductor current i L , the total inductor current i L Discretization gives L(n) , based on the volt-second balance principle, when the switch tubes S1 and S2 are turned on, formula (6) is obtained; when the switch tubes S1 and S2 are turned off, formula (7) is obtained; according to the forward Euler method, formula (8) is obtained; L in formula (6) and formula (7) is shown in the following formula (9):

[0022]

[0023] Step 1.5, substitute formula (8) into formula (6) and (7) to obtain the prediction model equation when the switch tubes S1 and S2 are turned on / off as follows:

[0024]

[0025] Among them, i L-on (n+1) is the inductor current value when the switch is turned on at time n+1, i L-off (n+1) is the inductor current value when the switch is turned off at time n+1;

[0028] Step 1.6, combining steps 1.3 and 1.5, calculate the difference g between the inductor current and the reference current at time n+1 when the switches S1 and S2 are turned on by the following formula (12): on ; The difference g between the inductor current and the reference current at time n+1 when the switches S1 and S2 are turned off is calculated by the following formula (13): off :

[0029] g on =|i L-on (n+1)-i L-ref (n+1)| (12)

[0030] g off =|i L-off (n+1)-i L-ref (n+1)| (13).

[0031] The specific process of step 2 is:

[0032] Step 2.1, collect the output voltage V o , and input voltage V in and input current i in The phase, overshoot time when switching loads, voltage error Vout_error and power factor obtained through the phase;

[0033] Step 2.2, initialize the data collected in step 2.1 and construct the initialization parameters of the particle swarm algorithm;

[0034] Step 2.3, based on the initialization conditions of step 2.2, extract the current particle parameters and start calling the staggered parallel Boost PFC simlink model for simulation;

[0035] Step 2.4, establish the objective function of the interleaved parallel boost PFC converter;

[0036] Step 2.5, determine whether the convergence condition is met through the objective function established in step 2.4. If not, update the data and iterate again, re-update the inertia weight w and fitness weight r, and adjust the particle position and speed of PSO; if it meets the convergence condition, convergence ends and the optimal k is output. p and k i ;

[0037] Step 2.6, get the optimal k from step 2.5 p and k i , find the high pulse drive v PH and low pulse drive v PL .

[0038] The specific process of step 2.2 is:

[0039] Initialize the data collected in step 2.1 and set the initial k p , k i , and construct the number of initialized particles, maximum number of iterations, PSO parameter settings, inertia weight w, fitness function weight setting r, objective function fitness of fitness evaluation and the optimal convergence condition fitness<1e-6 of the objective function.

[0040] In step 2.4, the objective function is shown in the following formula (14):

[0041]

[0042] In the formula, w_power_factor represents the power factor weight, PowerFactor represents the power factor, w_overshoot represents the overshoot weight, Overshoot represents the overshoot, w_adjust_time represents the adjustment time weight, AdjustTime represents the overshoot time, w_voltage represents the output voltage deviation weight, Vout_error represents the output voltage error, w_phase_diff represents the phase difference weight between the input voltage and the input current, and abs(Phase_diff) represents the phase, and i is the number of iterations.

[0043] In step 2.5, when the convergence condition fitness < 1e-6 is met, the convergence ends and the optimal k is output. p and k i , the process is shown in the following formula (15):

[0044]

[0045] Among them, Pos(:,fitMin_index) indicates the parameter position of the current optimal particle, fitMin: is the minimum fitness value in the current iteration, and updates the global optimal solution G best is the optimal k p and k i .

[0046] In step 2.6, the high pulse drive v is solved by the following formulas (16) and (17) respectively: PH and low pulse drive v PL :

[0047]

[0048] In the formula, v con is the voltage control signal output by the controller, v o (t n ) is t n The output voltage at the moment, v rec (t n ) is t n Peak voltage at the moment.

[0049] The specific process of step 3 is as follows:

[0050] The cost function g is constructed by the following formula (18): opt :

[0051]

[0052] when i L(n+1) Less than i ref(n+1) When g onLess than g off When the drive controller selects high power pulse v PH As the effective control pulse P of the next switching cycle; when i L(n+1) Greater than i ref(n+1) When g on Greater than g off The drive controller selects low power pulse v PL As the effective control pulse P of the next switching cycle, it further controls the interleaved parallel Boost PFC converter.

[0053] The beneficial effects of the present invention are as follows:

[0054] 1) Significantly improved dynamic performance: The present invention effectively reduces the switching time of the dynamic process through innovative control strategies, and improves the dynamic performance of the system. Experimental verification shows that the dynamic process can be reduced from 180ms to 30ms, thereby improving the dynamic response capability of the equipment.

[0055] 2) The implementation method is simple and easy: The implementation process of the present invention is more concise and clear, and is easy to apply in engineering, which not only reduces the complexity of technical implementation, but also shortens the product development cycle, which is conducive to the rapid promotion and industrialization of technology;

[0056] 3) The present invention significantly improves the dynamic performance of the system through the newly established PI-MPC control strategy, thereby improving the efficiency and energy efficiency of the converter, dynamic performance and promoting technological industrialization. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a structural block diagram of a Boost PFC converter used in a dynamic boost control method of Boost PFC converters with staggered parallel connection based on PI-MPC control of the present invention;

[0058] Figure 2 It is a control block diagram of a dynamic boost control method of a Boost PFC converter based on PI-MPC control interleaved parallel connection of the present invention;

[0059] Figure 3 It is a specific process of combining the PSO algorithm and the converter of the dynamic improvement control method of the Boost PFC converter based on PI-MPC control staggered parallel connection of the present invention;

[0060] Figure 4(a) to Figure 4(c) It is the simulation transient response diagram of the three traditional methods;

[0061] Figure 5It is a transient response diagram of the Boost PFC converter working at 600W to 1000W under the PI-MPC control method in the dynamic improvement control method of the Boost PFC converter with staggered parallel connection based on PI-MPC control of the present invention. DETAILED DESCRIPTION

[0062] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0063] Example 1

[0064] The present invention is based on a PI-MPC (a hybrid control strategy of proportional integral (PI) control and model predictive control (MPC)) control method for staggered parallel Boost PFC converters. The control block diagram of the Boost PFC converter used is as follows: Figure 1 As shown in the figure, the interleaved parallel Boost PFC converter circuit has v in and V o are the input voltage and output voltage of the interleaved parallel Boost PFC, i in is the input current, v in (V) The value range is 110-265V, V o The value range of (V) is 380-410V, V rec is the peak voltage after rectification, S1 and S2 are switches, D1 and D2 are diodes, L1 and L2 are inductors, i C is the output capacitor current, I o is the output current V o is the output current.

[0065] Example 2

[0066] The dynamic improvement control method of Boost PFC converter based on PI-MPC control interleaved parallel connection specifically includes the following steps:

[0067] Step 1: collect the uncontrolled rectifier voltage V of the interleaved parallel BoostPFC converter for the current inner loop rec , output voltage V o , output current I o, , the reference current I can be obtained through calculation ref . Sample the total inductor current i L The inductor current at time n is discretized, and the inductor current at time n+1 is calculated. The reference current is subtracted from the current of the sampled inductor to obtain the g when the switch is turned on. on and g when the switch is turned off off .

[0068] Step 2: Collect the output voltage V for the voltage outer loop o、Input voltage V in and input current i in Phase, the particle swarm optimization (PSO) algorithm is used to find the optimal k for the voltage loop. p , k i . And through calculation, we get the high pulse v PH and low pulse v PL as a driving pulse signal.

[0069] Step 3: Combine steps 1 and 2 to construct the cost function g opt , and obtain the effective control pulse P, which in turn controls the interleaved parallel Boost PFC converter.

[0070] Example 3

[0071] The specific process of step 1 is as follows:

[0072] Step 1.1, collect the uncontrolled rectifier voltage V rec , output voltage V o , output current I o , Assuming the converter efficiency is 100%, when the CCM Boost PFC converter is unity power factor, the peak current I of the total inductor (L1+L2) can be obtained. m :

[0073]

[0074] v rec =V m |sin(ωt)| (2)

[0075]

[0076] Among them, V in is the inductor input voltage, I in is the input current, V m is the peak voltage, I m is the peak current, w is the line frequency, and t is the time.

[0077] Step 1.2, the peak current I obtained from step 1.1 m , and let V ref =V o , then the reference current i can be derived ref express:

[0078]

[0079] Step 1.3, from step 1.2, we can get the reference current i ref , discretizing the reference current yields i ref (t n ), tn For n moments:

[0080]

[0081] Step 1.4, collect the total inductor current i L (the sum of the current of L1 and the current of L2), discretize the inductor current to get i L(n) Based on the volt-second balance principle, the inductor voltage is obtained. When the switch tubes S1 and S2 are turned on, the formula (6) is obtained, and when the switch tubes S1 and S2 are turned off, the formula (7) is obtained. And the formula (8) is obtained according to the forward Euler method. L in formula (6) and (7) is L1 and L2 in parallel to obtain formula (9)

[0082]

[0083] In the formula, V L (n) is the inductor voltage at time n, v rec (n) is the uncontrolled rectifier voltage V at time n rec ,i L (n) is the inductor current at time n, i L (n+1) is the inductor current at time n+1, T S for one switching cycle.

[0084] Step 1.5: From step 1.4, we can obtain that by substituting formula (8) into formulas (6) and (7), we can obtain the following prediction model equations when the switches S1 and S2 are turned on / off:

[0085]

[0086] Among them, i L-on (n+1) is the inductor current value when the switch is turned on at time n+1, i L-off (n+1) is the inductor current value when the switch tube is turned off at time n+1.

[0089] Step 1.6, combine step 1.3 and step 1.5, g on It represents the difference between the inductor current and the reference current at time n+1 when the switches S1 and S2 are turned on, g off It represents the difference between the inductor current and the reference current at time n+1 when the switches S1 and S2 are turned off.

[0090] g on =|i L-on (n+1)-i L-ref (n+1)| (12)

[0091] g off =|i L-off (n+1)-i L-ref(n+1)| (13).

[0092] Where i L_ref (n+1) is the reference current inductance value at time n+1.

[0093] Example 4

[0094] Draw the specific process of combining the interleaved parallel Boost PFC converter and the PSO algorithm, such as Figure 3 As shown, the algorithm is initialized first. In each iteration, the algorithm simulates and calculates the system performance under the controller parameters corresponding to each particle, including multiple indicators such as power factor, overshoot, voltage deviation, etc., and calculates the fitness function by combining these indicators. The particle swarm moves towards the individual optimal solution and the global optimal solution by updating the position and speed, and dynamically adjusts the inertia weight and fitness function weight to improve the search efficiency and finally obtain the optimal solution.

[0095] The specific operation process of step 2 is as follows

[0096] Step 2.1 Collect the output voltage V o , and input voltage V in and input current i in The phase of the load, the overshoot time when switching the load, the voltage error (output voltage minus the reference voltage) Vout_error, and the power factor obtained by the phase.

[0097] Step 2.2: Initialize the data collected in step 2.1 (set the initial k p , k i ), and construct the initialization particle number of the particle swarm algorithm (PSO), the maximum number of iterations, the PSO parameter settings, the inertia weight w, the fitness function weight setting r, the objective function fitness of the fitness evaluation, and the optimal convergence condition of the objective function (fitness<1e-6);

[0098] Step 2.3, extract the current particle parameter k from step 2.2 p and k i Start calling the interleaved parallel Boost PFCsimlink model for simulation.

[0099] Step 2.4, based on step 2.3, the fitness evaluation setting objective function for the interleaved parallel boost PFC converter can be obtained:

[0100] fitness(i)=w_power_factor*(1-PowerFactor)

[0101] +w_overshoot*Overshoot

[0102] +w_adjust_time*AdjustTime

[0103] +w_voltage*Vout_error

[0104] +w_phase_diff*abs(Phase_diff)(14)

[0105] In the formula, w_power_factor represents the power factor weight, PowerFactor represents the power factor, w_overshoot represents the overshoot weight, Overshoot represents the overshoot, w_adjust_time represents the adjustment time weight, AdjustTime represents the overshoot time, w_voltage represents the output voltage deviation weight, Vout_error represents the output voltage error, w_phase_diff represents the phase difference weight between the input voltage and the input current, and abs(Phase_diff) represents the phase, and i is the number of iterations.

[0106] Step 2.5, judge by the objective function obtained in step 2.4. If it does not meet the convergence condition (fitness ≥ 1e-6), update the data and iterate again, re-update the inertia weight w and fitness weight r, and adjust the particle position and speed of PSO.

[0107] Step 2.6: If the convergence condition (fitness < 1e-6) is met according to step 2.5, the convergence is complete. Output the optimal k p and k i :

[0108]

[0109] Pos(:,fitMin_index) indicates the parameter position of the current optimal particle. fitMin: This is the minimum fitness value in the current iteration, that is, the fitness value corresponding to the optimal particle. Update the global optimal solution G best is the optimal k p and k i .

[0110] Step 2.7, get the optimal k from step 2.6 p and k i , then multiply the duty cycle by K H and K L Get high and low duty cycle:

[0111]

[0112] In the formula, v PH For high pulse, vPL For low pulse drive, v con is the voltage control signal output by the controller, v o (t n ) is t n The output voltage at the moment, v rec (t n ) is t n Peak voltage at the moment.

[0113] Example 5

[0114] The specific process of step 3 is as follows

[0115] Combining steps 1 and 2, we can construct the cost function g opt , when i L(n+1) Less than i ref(n+1) When g on Less than g off When the drive controller selects high power pulse v PH As the effective control pulse P of the next switching cycle; when i L(n+1) Greater than i ref(n+1) When g on Greater than g off The drive controller selects low power pulse v PL As the effective control pulse P of the next switching cycle, it further controls the interleaved parallel Boost PFC converter.

[0116]

[0117] Example 6

[0118] FIG4 is a simulation of transient response diagrams of three conventional methods for controlling 600W to 1000W;

[0119] Figure 4(a) is the transient response diagram under PI (proportional integral) control from 600W to 1000W. It can be seen that the load switching from 600W to 1000W under PI control is 180ms;

[0120] Figure 4(b) is the transient response diagram under PCMC (predictive current control) control from 600W to 1000W. It can be seen that the load switching from 600W to 1000W under PCMC control is 110ms;

[0121] As shown in Figure 4(c), it is the transient response diagram of MPCC (model predictive control) control from 600W to 1000W. It can be seen that the load switching from 600W to 1000W under MPCC control is 90ms.

[0122] The transient response diagram of the interleaved parallel Boost PFC converter of the present invention working under the PI-MPC control method at 600W to 1000W is drawn, as shown in FIG. Figure 5 As shown, the output voltage V O It only takes 30ms to reach a stable state. Experimental verification shows that the dynamic process can be reduced from 180ms to 30ms, which improves the dynamic response capability of the equipment, thereby improving the efficiency and energy efficiency of the converter, dynamic performance and promoting the industrialization of technology.

[0123] The present invention establishes the mathematical relationship between the output voltage, output current, reference voltage, input peak voltage and reference current according to the principle of staggered parallel Boost PFC converter, and takes the current value when the switch is turned on at the next moment and the error between the current value when the switch is turned off at the next moment and the reference current value as the target to construct the function. The voltage loop uses the classic PSO algorithm to solve the optimal k p and k i。 , and set the high and low pulses through the high power pulse v PH With low power pulse v PL The drive controller can effectively make the average value of the inductor current i L Always follow the current reference i ref The transient effect of the interleaved parallel Boost PFC converter is further improved, and the practical applicability of the Boost converter is enhanced.

Claims

1. A dynamic boost control method for Boost PFC converters with staggered parallel connection based on PI-MPC control, characterized in that: The specific steps include: Step 1: collect the uncontrolled rectifier voltage V of the interleaved parallel BoostPFC converter for the current inner loop rec , output voltage V o , output current I o , and obtain the reference current I ref , according to the reference current I ref Calculate the g when the switch is turned on on and g when the switch is turned off off ; Step 2: Collect the output voltage V for the voltage outer loop o 、Input voltage V in and input current i in Phase, the particle swarm algorithm is used to find the optimal k for the voltage loop p , k i , and obtain the high pulse v through calculation PH and low pulse v PL As a driving pulse signal; Step 3: Combine steps 1 and 2 to construct the cost function g opt , and obtain the effective control pulse P, which in turn controls the interleaved parallel Boost PFC converter.

2. The method for dynamically improving the Boost PFC converter based on PI-MPC control interleaved parallel connection according to claim 1, characterized in that: The specific process of step 1 is as follows: Step 1.1, collect the uncontrolled rectifier voltage V rec , output voltage V o , output current I o , assuming that the converter efficiency is 100%, when the CCM Boost PFC converter is a unity power factor, the peak current I of the total inductor is obtained m : v rec =V m |sin(ωt)| (2) Among them, V in is the inductor input voltage, I in is the input current, V m is the peak voltage, I m is the peak current, w is the line frequency, and t is the time; Step 1.2, let V ref =V o , the peak current I obtained from step 1.1 m , the reference current i ref , as shown in the following formula (4): Step 1.3, set the reference current i ref Discretize i ref (t n ), as shown in the following formula (5): Among them, t n is the n moment; Step 1.4, collect the total inductor current i L , the total inductor current i L Discretization gives L(n) , based on the volt-second balance principle, when the switch tubes S1 and S2 are turned on, formula (6) is obtained; when the switch tubes S1 and S2 are turned off, formula (7) is obtained; according to the forward Euler method, formula (8) is obtained; L in formula (6) and formula (7) is shown in the following formula (9): Step 1.5, substitute formula (8) into formula (6) and (7) to obtain the prediction model equation when the switch tubes S1 and S2 are turned on / off as follows: Among them, i Lon (n+1) is the inductor current value when the switch is turned on at time n+1, i Loff (n+1) is n+1-- The inductor current value when the switch tube is turned off at the moment; Step 1.6, combining steps 1.3 and 1.5, calculate the difference g between the inductor current and the reference current at time n+1 when the switches S1 and S2 are turned on by the following formula (12): on ; The difference g between the inductor current and the reference current at time n+1 when the switches S1 and S2 are turned off is calculated by the following formula (13): off : g on =|i L-on (n+1)-i L-ref (n+1)| (12) g off =|i L-off (n+1)-i L-ref (n+1)| (13)。 3. The method for dynamically improving the Boost PFC converter based on PI-MPC control interleaved parallel connection according to claim 2, characterized in that: The specific process of step 2 is: Step 2.1, collect the output voltage V o , and input voltage V in and input current i in The phase, overshoot time when switching loads, voltage error Vout_error and power factor obtained through the phase; Step 2.2, initialize the data collected in step 2.1 and construct the initialization parameters of the particle swarm algorithm; Step 2.3, based on the initialization conditions of step 2.2, extract the current particle parameters and start calling the staggered parallel BoostPFC simlink model for simulation; Step 2.4, establish the objective function of the interleaved parallel boost PFC converter; Step 2.5, determine whether the convergence condition is met through the objective function established in step 2.

4. If not, update the data and iterate again, re-update the inertia weight w and fitness weight r, and adjust the particle position and speed of PSO; if it meets the convergence condition, convergence ends and the optimal k is output. p and k i ; Step 2.6, get the optimal k from step 2.5 p and k i , find the high pulse drive v PH and low pulse drive v PL .

4. The method for dynamically improving the control of Boost PFC converters based on PI-MPC control interleaved parallel connection according to claim 3, characterized in that: The specific process of step 2.2 is as follows: Initialize the data collected in step 2.1 and set the initial k p , k i , and construct the number of initialized particles, maximum number of iterations, PSO parameter settings, inertia weight w, fitness function weight setting r, objective function fitness of fitness evaluation and the optimal convergence condition fitness<1e-6 of the objective function.

5. The method for dynamically improving the control of Boost PFC converters based on PI-MPC control interleaved parallel connection according to claim 4, characterized in that: In step 2.4, the objective function is as shown in the following formula (14): In the formula, w_power_factor represents the power factor weight, PowerFactor represents the power factor, w_overshoot represents the overshoot weight, Overshoot represents the overshoot, w_adjust_time represents the adjustment time weight, AdjustTime represents the overshoot time, w_voltage represents the output voltage deviation weight, Vout_error represents the output voltage error, w_phase_diff represents the phase difference weight between the input voltage and the input current, and abs(Phase_diff) represents the phase, and i is the number of iterations.

6. The method for dynamically improving the control of Boost PFC converters based on PI-MPC control in parallel according to claim 5, characterized in that: In step 2.5, when the convergence condition fitness < 1e-6 is met, the convergence ends and the optimal k is output. p and k i , the process is shown in the following formula (15): Among them, Pos(:,fitMin_index) indicates the parameter position of the current optimal particle, fitMin: is the minimum fitness value in the current iteration, and updates the global optimal solution G best is the optimal k p and k i .

7. The method for dynamically improving the control of Boost PFC converters based on PI-MPC control interleaved parallel connection according to claim 6, characterized in that: In step 2.6, the high pulse drive v is solved by the following formulas (16) and (17) respectively: PH and low pulse drive v PL : In the formula, v con is the voltage control signal output by the controller, v o (t n ) is t n The output voltage at the moment, v rec (t n ) is t n Peak voltage at the moment.

8. The method for dynamically improving the control of Boost PFC converters based on PI-MPC control in staggered parallel connection according to claim 7, characterized in that: The specific process of step 3 is as follows: The cost function g is constructed by the following formula (18): opt : when i L(n+1) Less than i ref(n+1) When g on Less than g off When the drive controller selects high power pulse v PH As the effective control pulse P of the next switching cycle; when i L(n+1) Greater than i ref(n+1) When g on Greater than g off The drive controller selects low power pulse v PL As the effective control pulse P of the next switching cycle, it further controls the interleaved parallel Boost PFC converter.

Images