Single-phase h-bridge inverter system control method based on improved fish eagle algorithm

Abstract

The application discloses a single-phase H-bridge inverter system control method based on an improved fish-eagle algorithm, which adopts a double-closed-loop control strategy of active damping, voltage outer loop PI control and current inner loop QPR control, outputs control signals of switching tubes, and enables the inverter to quickly respond to load changes and input fluctuations, reduces harmonic distortion and steady-state error of an output waveform. Then, proportional coefficients and integral coefficients of the voltage outer loop PI control are taken as two-dimensional positions of the fish-eagle, optimization is carried out through the fish-eagle algorithm, the calculation method of the fish-eagle feeding position in the development stage is improved, negative values are introduced in the molecular coefficient part, and when the position of the fish-eagle is in the upper boundary, the updated position is always in the upper boundary. The position corresponding to the optimal fitness value output by the improved fish-eagle algorithm is used for double-closed-loop control of the single-phase H-bridge inverter system, so that the system can maintain the best response speed and accuracy under various working conditions.

Description

Technical Field

[0001] This invention belongs to the field of electronic power control technology, and relates to intelligent power grid control, specifically to a control method for a single-phase H-bridge inverter system based on an improved Osprey algorithm. Background Technology

[0002] Single-phase inverters are widely used in home appliances, uninterruptible power supplies (UPS), photovoltaic power generation systems, and electric vehicle charging stations. To meet the load's demand for high-quality power, the inverter's output waveform must satisfy the requirements of low harmonic content and high stability. Faced with complex load changes and nonlinear interference, traditional control methods struggle to effectively suppress the resonance peaks of LC filters, leading to increased total harmonic distortion (THD) and decreased dynamic stability of the output waveform. PI control achieves voltage and current regulation of the inverter system through a dual-closed-loop structure, but it relies on passive damping when dealing with resonance peak issues, which can easily lead to additional losses. Therefore, designing an efficient inverter control strategy to ensure that the inverter can quickly and accurately track the input waveform under various operating conditions has become a key problem that needs to be solved.

[0003] Besides the control strategy, the selection of various controller parameters also directly affects the response speed, accuracy, and stability of the control system. Traditional control strategies often rely on experience to adjust parameters, which is not only time-consuming and labor-intensive, but also makes it difficult to ensure that the system achieves optimal performance under all operating conditions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a control method for a single-phase H-bridge inverter system based on an improved Osprey algorithm. By combining outer-loop PI control and inner-loop QPR control, the system transfer function is reconstructed using active damping modulation to suppress resonance peaks. Furthermore, the improved Osprey algorithm is used to optimize PI control parameters, thereby improving convergence efficiency and enabling the system to maintain high-precision tracking even under sudden load changes.

[0005] A control method for a single-phase H-bridge inverter system based on an improved Osprey algorithm, the specific steps of which are as follows:

[0006] Step 1: Active damping suppresses resonance peak control

[0007] Based on the resonant frequency requirements of the inverter system, determine the values ​​of inductor L and capacitor C in the LC filter structure.

[0008] Using an active damping control method, an equivalent control block diagram and system transfer function of a single-phase H-bridge inverter system are established. Bode plots of active damping control strategy and undamped control strategy under no-load conditions are compared. Based on the suppression of the resonance peak, the active damping parameter Ka is adjusted.

[0009] Step 2: Dual Closed-Loop Control Strategy

[0010] A dual closed-loop control strategy employing an outer-loop PI control and an inner-loop QPR control is adopted. First, the output voltage Vo of the single-phase H-bridge inverter system is sampled and compared with the reference voltage Vref to obtain the voltage error value. This voltage error value is then input to the PI control to obtain the reference current value. Next, the output current value is sampled and compared with the reference current value obtained from the PI control to obtain the current error value. This current error value is then input to the QPR control to obtain a reference output value y. This y is accumulated with the active damping parameter Ka obtained in step 1 and compared with the input voltage Vi to calculate the duty cycle (Duty) of the PWM input for the switching transistor control.

[0011] Duty=(y-iL*Ka) / Vi

[0012] Where iL represents the current of the LC filter structure.

[0013] Step 3: Optimization of PI Control Parameters

[0014] A simulation model of a single-phase H-bridge inverter system is built based on the transfer function. The output voltage Vo of the system is measured using a dual closed-loop control method.

[0015] Given an osprey population size N, a maximum number of iterations T, and an optimization parameter dimension D, the scaling factor Kp and integral factor Ki are used as the two-dimensional positions of the ospreys. Upper and lower boundaries of Kp and Ki are set to generate N initial values, which serve as the initial position matrix for the osprey population. The absolute error integral criterion ITAE is used as the fitness function.

[0016]

[0017] e(m) = Vref - Vo

[0018] Where m is the operating time of the single-phase H-bridge inverter system, and the output voltage Vo is obtained from the simulation model.

[0019] The exploration and development phases are conducted sequentially, updating the osprey's position. The feeding position X of the i-th osprey is determined during the design and development phase. i,j 2(t), based on the original osprey algorithm's method for calculating feeding positions, uses random number R... i,j The range of values ​​for 2 has been changed to [-0.6, 0.6], and the denominator has been changed to t. 2 To accelerate the convergence speed:

[0020] X i,j 2(t)=X i,j 1(t)+(lb j +R i,j 2(ub j -lb j )) / t2

[0021] Among them, X i,j 1(t) represents the updated position of the Osprey during the exploration phase, where t represents the current iteration number, and lb represents the position of the Osprey. j ub j Let represent the lower and upper limits of the optimization parameter j, respectively, where j = 1, 2, 3, ..., D.

[0022] After each iteration, the simulation model is called to calculate the fitness value ITAE, with a smaller value indicating better fitness. The fitness function is calculated based on the updated position of each osprey, and the position corresponding to the optimal fitness value fbest is recorded as the optimal position Xbest. When the number of iterations reaches T, the optimal position Xbest is output as the proportional coefficient Kp and integral coefficient Ki for PI control, controlling the single-phase H-bridge inverter system.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1. Using active damping control reduces the impact of the resonance peak of the LC filter under no-load conditions and avoids the additional power loss to the system caused by the passive damping control strategy.

[0025] 2. A dual closed-loop control strategy is adopted, in which the outer voltage loop PI control effectively stabilizes the output voltage, while the inner current loop QPR control ensures accurate tracking of the output current, enabling the inverter to respond quickly to load changes and input fluctuations, and reducing harmonic distortion and steady-state error of the output waveform.

[0026] 3. The position calculation method in the development stage of the Osprey algorithm has been improved, resulting in faster convergence. The optimized PI control parameters enable the system to maintain optimal response speed and accuracy under various operating conditions, thereby improving the stability and robustness of the inverter and reducing the harmonic content of the output voltage and current. Attached Figure Description

[0027] Figure 1 This is the circuit diagram of a single-phase H-bridge inverter system.

[0028] Figure 2 This is a block diagram of the LC filter active damping control for a single-phase H-bridge inverter system.

[0029] Figure 3 This is a control block diagram of a single-phase H-bridge inverter system based on a dual closed-loop control strategy.

[0030] Figure 4 Flowchart of the improved Osprey algorithm.

[0031] Figure 5 This is a comparison chart showing the changes in fitness values ​​of different optimization algorithms in the embodiments.

[0032] Figure 6 The output current, output voltage, and reference output voltage waveforms of the single-phase H-bridge inverter system in the embodiment are shown.

[0033] Figure 7 The diagram shows the harmonic content of the output voltage controlled by the single-phase H-bridge inverter system in the embodiment. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described clearly and completely below in conjunction with the accompanying drawings and embodiments.

[0035] like Figure 1 As shown, the single-phase H-bridge inverter system includes an H-bridge inverter structure composed of four switching transistors Q1, Q2, Q3, and Q4, and an LC filter structure composed of inductor L and capacitor C. The input DC voltage Vi is converted into an output AC voltage Vo after passing through the H-bridge inverter and LC filter. Ci is the input capacitor, and R... L R is the resistance coefficient of the inductor itself. load For output load.

[0036] This invention provides a control method for a single-phase H-bridge inverter system based on an improved Osprey algorithm. It utilizes active damping to suppress resonance peaks and combines an outer-loop PI control with an inner-loop QPR control dual-closed-loop control strategy to control the single-phase H-bridge inverter system. The specific steps are as follows:

[0037] Step 1: Active damping suppresses resonance peak control

[0038] Based on the resonant frequency requirements of the inverter system, determine the values ​​of inductor L and capacitor C in the LC filter structure.

[0039] To suppress the resonance peak of an LC filter structure, common control strategies include passive damping and active damping. Passive damping attenuates the resonance peak under no-load conditions by connecting resistors in series or parallel with the filter capacitor or inductor; however, the added components result in power loss, affecting system efficiency. Therefore, this embodiment employs active damping control.

[0040] The H-bridge inverter structure in a single-phase H-bridge inverter system is equivalent to a square wave with positive and negative values. Kirchhoff's theorem is used to establish the state equation of the LC filter circuit. The superposition theorem is used to obtain the input and output relationship of the single-phase H-bridge inverter system. Combined with an active control strategy, the following is obtained: Figure 2 The system equivalent control block diagram shown is given, where iL is the LC filter loop current, Ka is the active damping parameter, S is the complex frequency domain variable, Zo is the equivalent impedance, and the corresponding no-load system transfer function G is given. a (S) is:

[0041] G a (S)=1 / (LCS 2 +KrCS+1)

[0042] By using the system transfer function, Bode plots comparing the active damping control strategy and the undamped control strategy under no-load conditions are drawn in SIMULINK to observe the suppression of the resonance peak and select an appropriate active damping parameter Ka.

[0043] Step 2: Dual Closed-Loop Control Strategy

[0044] A dual closed-loop control strategy, employing an outer voltage loop PI control and an inner current loop QPR control, is used to generate the input PWM for the switching transistor, such as... Figure 3 As shown, firstly, the output voltage value Vo of the single-phase H-bridge inverter system is sampled and compared with the reference voltage value Vref to obtain the voltage error value. This voltage error value is then input to the PI control to obtain the reference current value. Next, the output current value is sampled and compared with the reference current value obtained from the PI control to obtain the current error value. This current error value is then input to the QPR control to obtain a reference output value y. This y is accumulated with the active damping parameter Ka obtained in step 1 and compared with the input voltage Vi to calculate the duty cycle (Duty) of the PWM input for the switching transistor control.

[0045] Duty=(y-iL*Ka) / Vi

[0046] Where iL represents the current of the LC filter structure.

[0047] PI control is a control strategy that adjusts the inverter output voltage using a proportional-integral controller. The outer-loop PI control monitors the error between the output voltage and the setpoint, and adjusts the inverter's modulation signal using the PI controller parameters. It is primarily responsible for system stability and load regulation, possessing strong anti-interference capabilities and effectively handling load changes and input voltage fluctuations to ensure stable inverter output voltage. Its transfer function G... PI (S) is:

[0048] G PI (S) = Kp + (Ki / S)

[0049] Among them, K p K represents the proportionality coefficient. i This represents the control coefficient.

[0050] The inner current loop QPR control focuses on instantaneous current control. By precisely adjusting the current, it reduces harmonics generated by the inverter bridge and output current hysteresis, more effectively handling the tracking requirements of sinusoidal current waveforms. This ensures that the inverter output current accurately follows the set value. Its transfer function G... QPR (S) is:

[0051] G QPR (S)=Kp+Kr(2KrW c S) / (S 2 +2W c S+W 2 o )

[0052] Where Kr represents the resonance coefficient, W c W represents the cutoff bandwidth. o This indicates the resonant frequency.

[0053] Step 3: Optimization of PI Control Parameters

[0054] A simulation model of a single-phase H-bridge inverter system is built based on the transfer function. A dual-closed-loop control method is used to measure the system's output voltage Vo. Figure 4 As shown, the improved Osprey algorithm is used to optimize the proportional coefficient Kp and the control coefficient Ki. The specific steps are as follows:

[0055] s3.1. Set the osprey population size N=30, the maximum number of iterations T=30, and the dimension D=2. Use the scaling factor Kp and the integral factor Ki as the two-dimensional position of the osprey. Set the upper boundary of Kp to ub1=0.8 and the lower boundary to lb1=0.2, and the upper boundary of Ki to ub2=8000 and the lower boundary to lb2=2000. Generate N initial values ​​X. i,j (0) is used as the initial position matrix for the osprey population. The absolute error integral criterion ITAE is used as the fitness function:

[0056]

[0057] e(m) = Vref - Vo

[0058] Where m is the operating time of the single-phase H-bridge inverter system, and the output voltage Vo is obtained from the simulation model.

[0059] s3.2, Exploration Phase: The osprey explores and catches fish; the osprey's location is X. i,j (t) updated to X i,j 1(t):

[0060] For each osprey, the new position X is obtained according to the osprey exploration phase formula. i,j 1(t):

[0061] X i,j 1(t)=X i,j (t)+R i,j 1(SF i,j -I i,j X i,j (t))

[0062] Where t represents the current iteration number, R i,j 1 is a random number between [0, 1], SF i,j For N positions X i,j I is a randomly selected position in (t). i,j It is a random number, and I i,j =1 or 2.

[0063] s3.3 Calculate the fitness value of the updated Osprey position and compare it with X i,j (t) and X i,j The fitness value corresponding to 1(t) is such that if the new position X i,j If the fitness value of 1(t) is better, then the new position X is used. i,j 1(t) replaces the old position X i,j (t), otherwise keep the old position unchanged, let X i,j 1(t) = X i,j (t).

[0064] Then, based on the set upper and lower boundaries, update the position X. i,j 1(t) is used to constrain the position beyond the boundary by using the set boundary value.

[0065] s3.4, During the development phase, after the cormorant catches fish, it will move the fish from X. i,j 1(t) is brought to a new position X i,j 2(t) Feeding, the feeding position X of the i-th osprey i,j 2(t) is:

[0066] X i,j 2(t)=X i,j 1(t)+(lb j +R i,j 1(ub j -lb j )) / t

[0067] This embodiment will specify the feeding location X. i,j 2(t) is changed to:

[0068] X i,j 2(t)=X i,j 1(t)+(lb j +R i,j 2(ub j -lb j )) / t 2

[0069] Among them, R i,j 2 is a random number ranging from [-0.6, 0.6]. Compared to the original Osprey algorithm's calculation of the feeding position, the random number R in the numerator is...i,j The value of 2 can be negative, preventing the updated position from always being at the upper boundary when the osprey's position is already at the upper boundary. Furthermore, the t in the denominator is changed to t. 2 This allows the convergence range to gradually decrease as the number of iterations increases, thereby optimizing the local location search, accelerating the convergence speed, reducing the convergence time, and improving the convergence efficiency.

[0070] s3.5 Calculate the fitness value of the updated Osprey position and compare X again. i,j 2(t) and X i,j Fitness between 1(t), if X i,j If 2(t) has better fitness, then X should be used. i,j 2(t) replaces X i,j 1(t), otherwise keep the old position and let X i,j 2(t)=X i,j 1(t).

[0071] Then, based on the set upper and lower boundaries, update the position X. i,j 2(t) is used to constrain the position beyond the boundary by using the set boundary value.

[0072] s3.6, Update the position X of each osprey i,j 2(t) Calculate the fitness value and denote the position corresponding to the optimal fitness value fbest as the optimal position Xbest.

[0073] s3.7 Repeat s3.2~s3.6 until the number of iterations t=T, and output the optimal position Xbest as the proportional coefficient Kp and integral coefficient Ki of the PI control to control the single-phase H-bridge inverter system.

[0074] To illustrate the advantages of the improved Osprey algorithm in PI control parameter optimization, this embodiment uses the Osprey algorithm, Sparrow algorithm, and Particle Swarm Optimization algorithm to optimize the PI control parameters, and compares the changes in fitness values ​​of different optimization methods with the number of iterations. Figure 5 As shown, the improved Osprey algorithm requires fewer iterations to reach convergence, converges faster, and has higher convergence efficiency. Furthermore, its fitness value after convergence is the lowest among the four methods.

[0075] Figure 6 This is a simulation waveform diagram of the output current, output voltage, and reference output voltage of a single-phase H-bridge inverter system under the above control method, facing a sudden load change. The reference output voltage is a standard 220V RMS value, a 50Hz sine wave 311sin(2*pi*50). The single-phase H-bridge inverter system initially operates in full-load mode, switches to half-load mode at 0.1s, and switches to no-load mode at 0.2s. Figure 6 As can be seen, the waveform of the output voltage almost overlaps with that of the reference output voltage. When load changes occur at 0.1s and 0.2s, the output voltage and output current can still respond quickly and remain stable, indicating that the single-phase H-bridge inverter system has high stability and fast response capability under the control of this method.

[0076] Figure 7 The diagram shows the harmonic content of the output voltage of a single-phase H-bridge inverter system under the above control method. It can be seen that the harmonic content of the output voltage is low, indicating that this method can effectively suppress harmonics.

Claims

1. A single-phase H-bridge inverter system control method based on an improved fish eagle algorithm, characterized in that: The control signal of the output switch tube is output by using a double closed-loop control strategy of active damping and voltage outer loop PI control and current inner loop QPR control; The population size N of the fish eagle and the maximum iteration number T are set, the proportional coefficient Kp and the integral coefficient Ki of the voltage outer loop PI control are taken as the two-dimensional position of the fish eagle, and the initial positions of the N fish eagles are generated; From the initial position, the exploration phase and the development phase are carried out in turn, and the position of the fish eagle is updated; the feeding position X of the i-th fish eagle in the development phase i,j 2(t) is: X i,j 2(t) = X i,j 1(t) + (lb j + R i,j 2(ub j -lb j )) / t 2 where X i,j 1(t) denotes the updated position of the fish eagle in the exploration phase, lb j , ub j denote the lower and upper limits of the optimization parameter j, respectively, j = 1, 2, 3,..., D. After each iteration, the output voltage Vo of the single-phase H-bridge inverter system is calculated according to the updated position, the fitness value ITAE is calculated according to the absolute error integral value between the output voltage Vo and the set reference voltage Vref, and the position corresponding to the optimal fitness value fbest is recorded as the optimal position Xbest; when the iteration number reaches T, the optimal position Xbest is output as the proportional coefficient Kp and the integral coefficient Ki of the PI control, and the single-phase H-bridge inverter system is controlled.

2. The single-phase H-bridge inverter system control method based on the improved fish eagle algorithm according to claim 1, characterized in that: The values of the inductance L and the capacitance C in the single-phase H-bridge inverter system are determined according to the requirement of the resonant frequency.

3. The single-phase H-bridge inverter system control method based on the improved fish eagle algorithm according to claim 1, characterized in that: The equivalent control block diagram and the system transfer function of the single-phase H-bridge inverter system are established by using the active damping control method, the comparative Bode diagram of the active damping control strategy and the undamped control strategy under no-load condition is compared, and the active damping parameter Ka is adjusted according to the suppression of the resonant peak.

4. The single-phase H-bridge inverter system control method based on the improved fish eagle algorithm according to claim 1, characterized in that: The double closed-loop control strategy of the voltage outer loop PI control and the current inner loop QPR control is as follows: The output voltage value Vo of the single-phase H-bridge inverter system is sampled and compared with the reference voltage value Vref to obtain the voltage error value, and the voltage error value is input into the PI control to obtain the reference current value; Then the output current value is sampled and compared with the reference current value obtained by the PI control to obtain the current error value, the current error value is input into the QPR control to obtain a reference output value y, the reference output value y is accumulated with the active damping parameter Ka, and compared with the input voltage Vi to calculate the duty cycle of the switch tube control input PWM.

5. The single-phase H-bridge inverter system control method based on the improved fish eagle algorithm according to claim 1, characterized in that: The calculation method of the fitness value is as follows: ； e(m)= Vref-Vo; Wherein, m is the running time of the single-phase H-bridge inverter system, Vref and Vo represent the reference voltage and the output voltage of the single-phase H-bridge inverter system respectively.

6. The single-phase H-bridge inverter system control method based on the improved fish eagle algorithm according to claim 1, characterized in that: In the exploration phase, the fish eagle explores and catches fish, and the position of the fish eagle is updated as X i,j 1(t), the fitness value of the new position X i,j 1(t) is better, the new position X i,j 1(t) replaces the old position X i,j (t), otherwise the old position remains unchanged, and X i,j 1(t)=X i,j (t); then according to the set upper and lower boundaries, the updated position X i,j 1(t) is constrained, and the position exceeding the boundary is replaced by the set boundary value; In the development stage, the fish eagle brings fish from X i,j 1(t) to a new position X i,j 2(t) to eat, compare the fitness value before and after the position update, if the fitness value of the new position X i,j 2(t) is better, then replace the old position X i,j 1(t) with the new position X i,j 2(t), otherwise keep the old position unchanged, let X i,j 2(t)=X i,j 1(t); then according to the set upper and lower boundaries, constrain the updated position X i,j 2(t), use the set boundary value to replace the position beyond the boundary.

7. The single-phase H-bridge inverter system control method based on the improved fish eagle algorithm according to claim 1, characterized in that: The population size N of the fish eagle is set to 30, the maximum iteration number T is set to 30, the upper boundary ub1 of the proportional coefficient Kp is set to 0.8, the lower boundary lb1 is set to 0.2, the upper boundary ub2 of the integral coefficient Ki is set to 8000, and the lower boundary lb2 is set to 2000. 8.A computer readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method of any one of claims 1-7.

Images