Fractional order finite control set model control method based on single-phase quasi-Z-source rectifier

Abstract

The invention discloses a fractional order finite control set model control method based on a single-phase quasi-Z-source rectifier, and belongs to the field of power electronics. Compared with traditional integer order finite control set model prediction control and fractional order finite set model prediction control without delay compensation, the method has the advantages that the input side alternating current distortion rate is reduced, the fluctuation of the output side direct current voltage is reduced, the dynamic response speed of the output side direct current voltage is increased, and higher control precision is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of power electronics technology, and in particular to a fractional-order finite control set model control method based on a single-phase quasi-Z-source rectifier. Background Art

[0002] With the development of power electronics technology, PWM rectifiers have been widely used in power systems, electric traction and other fields. Among them, the quasi-Z-source rectifier has broad application prospects due to its simple topology, ability to boost or buck, allow direct flow and strong reliability. It is currently used in vehicle integrated chargers. The traditional quasi-Z-source rectifier is controlled by a dual closed-loop control system. Both the current inner loop and the voltage outer loop use PI control, which has the disadvantages of slow dynamic response and complex control parameter debugging. In order to overcome these shortcomings, model predictive control has been widely used in PWM rectifiers. Model predictive control has the characteristics of good dynamic performance, simple design, and can achieve multi-objective control. However, model predictive control still has shortcomings in control accuracy.

[0003] With the development of fractional-order calculus theory, fractional-order modeling and control have entered people's field of vision. Compared with traditional integer-order modeling, fractional-order calculus equations can better describe the actual operating state of the rectifier. Compared with traditional control, fractional-order control adds an adjustable parameter, the fractional order, which makes the control effect more flexible and the control accuracy higher. The fractional-order calculus operator is introduced into the traditional model predictive control to obtain the fractional-order model predictive control. In addition, considering that sampling and calculation require a certain amount of time, the traditional model predictive control still has a control delay of one cycle, which affects the control accuracy of the algorithm. Summary of the invention

[0004] The present invention provides a fractional-order finite control set model control method based on a single-phase quasi-Z-source rectifier, which can achieve the control goals of operating the rectifier at a unity power factor, reducing the distortion rate of the input current on the AC side, improving the power quality, reducing the output DC voltage fluctuation, and improving the steady-state performance and dynamic performance.

[0005] An embodiment of the present invention provides a fractional-order finite control set model control method based on a single-phase quasi-Z-source rectifier, comprising the following steps:

[0006] The input current on the AC side, the inductor current and capacitor voltage of the quasi-Z source network on the DC side are selected as control targets, and the operating characteristics of the single-phase quasi-Z source rectifier are analyzed. Based on the analysis of the AC side circuit, the state equation corresponding to the input current on the AC side in different states of the rectifier bridge switch tube is obtained. The fractional-order calculus operator is introduced, and the integer-order state equation is replaced by the fractional-order state equation. Based on the modal analysis of the DC side equivalent circuit, the state equations of the inductor current and capacitor voltage of the quasi-Z source network in different circuit states are obtained. When the circuit is in a non-through state, the switch tube is turned on, and when the circuit is in a through state, the switch tube is turned off.

[0007] The forward Euler method is used to discretize the state equations of the selected inductor current and capacitor voltage to obtain the prediction model of the required current and voltage. The possible combinations of switch states of the rectifier bridge switch tube and the quasi-Z source network switch tube are listed, and the combinations are matched with the prediction model to construct the value function of the associated multi-objective control and compensate for the delay time.

[0008] The required voltage and current are sampled through voltage and current sensors. According to the sampled values ​​and calculated values, predictive control is performed through the improved fractional-order finite control set model. The switch state combination that minimizes the value function is obtained by traversal optimization and determined as the optimal switch state combination. According to the optimal switch state combination, the switch states of the rectifier bridge switch tube and the switch tube in the quasi-Z source network are determined.

[0009] Optionally, in one embodiment of the present invention, the AC side circuit of the single-phase quasi-Z-source rectifier is analyzed, and a fractional-order calculus operator is introduced to obtain a state equation of the input side AC current. The fractional-order state equation is discretized using the forward Euler method to obtain a fractional-order prediction expression of the input side AC current:

[0010]

[0011] Among them, i g (k+1) and i g (k) represents the AC input current value at the k+1th and kth sampling moments respectively; T s is the control period; L g is the inductance value of the AC side; u g (k) and u pn (k) represents the AC input voltage and the rectifier bridge output voltage at the kth sampling moment; α is the order of the introduced fractional calculus operator; S1 is the operation mode of the rectifier, which takes the value of -1, 0 or 1;

[0012] The DC side circuit of the single-phase quasi-Z-source rectifier is analyzed to obtain the state equations of the capacitor voltage and inductor current of the quasi-Z-source network. The state equations are discretized using the forward Euler method to obtain the prediction expression of the inductor current:

[0013]

[0014] Among them, i L2 (k+1) and i L2 (k) represents the inductor current value of the quasi-Z source network at the k+1th and kth sampling moments respectively; u C2 (k) and u C1 (k) and u Co (k) respectively represent the capacitor voltage value and output DC voltage value of the quasi-Z source network at the kth sampling moment; S2 is the switch state of the quasi-Z source network switch tube, which takes a value of 0 or 1. When the rectifier works in the direct-flow state, S2 takes a value of 0, and when the rectifier works in the non-direct-flow state, S2 takes a value of 1;

[0015] The predicted expression of capacitor voltage is obtained as:

[0016]

[0017] Among them, u C2 (k+1) and u C2 (k) represents the capacitor voltage value of the quasi-Z source network at the k+1th and kth sampling moments respectively; i pn (k), i L2 (k) and i L1 (k) respectively represent the output current value of the rectifier bridge and the inductor current value of the quasi-Z source network at the kth sampling moment.

[0018] Optionally, in one embodiment of the present invention, the value function for constructing the associated multi-objective control is:

[0019]

[0020] Among them, i g (k+1), i L2 (k+1) and u C2 (k+1) respectively represent the input current value of the AC side and the inductor current and capacitor voltage value of the quasi-Z source network at the k+1th sampling moment; λ is the weight factor; i gref 、i L2ref and u C2ref They represent the reference values ​​of the AC input current and the inductor current and capacitor voltage of the quasi-Z source network respectively; i gref It is obtained by multiplying the output of the capacitor voltage outer loop of the quasi-Z source network controlled by PI with the phase of the AC side input voltage to achieve unity power factor operation; i L2ref Obtained by the output of the DC output voltage outer loop controlled by PI; u C2ref is a fixed constant.

[0021] Optionally, in one embodiment of the present invention, compensating for the delay time includes:

[0022] A two-step prediction method of one-step prediction compensation delay plus fractional-order model prediction is used to compensate for control delay, that is, the optimal switch state acting at this moment is obtained in the previous cycle.

[0023] The fractional-order finite control set model control method based on the single-phase quasi-Z-source rectifier of the embodiment of the present invention can achieve the control goal of the rectifier operating at a power factor and outputting a stable DC voltage. Compared with the traditional integer-order model predictive control and the model predictive control without delay compensation strategy, the control performance disclosed by the present invention is more flexible and the control accuracy is higher. Under the same conditions, the present invention reduces the distortion rate of the input current on the AC side of the single-phase quasi-Z-source rectifier, improves the power quality, makes the output DC voltage more stable, and reduces the voltage fluctuation. The dynamic performance and robustness of the rectifier are improved. When the system is disturbed by the outside world, the rectifier can be adjusted quickly to achieve the stability of the output DC voltage.

[0024] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0026] Figure 1 Schematic diagram of the topological structure of a single-phase quasi-Z-source rectifier in an embodiment of the present invention;

[0027] Figure 2 is a control block diagram of a single-phase quasi-Z-source rectifier in an embodiment of the present invention;

[0028] Figure 3 It is a flow chart of the main steps of the improved fractional-order finite control set model predictive control method for a single-phase quasi-Z-source rectifier in an embodiment of the present invention;

[0029] Figure 4 It is a schematic diagram of the principle of the delay compensation strategy of the fractional-order finite control set model predictive control in an embodiment of the present invention;

[0030] Figure 5 (a) and Figure 5 (b) are respectively comparison diagrams of voltage and current simulation waveforms of the single-phase quasi-Z-source rectifier before and after adding the delay compensation strategy in the embodiment of the present invention, Figure 5 (c) and Figure 5(d) are comparison diagrams of the percentage of the current harmonic content to the fundamental current content (THD) of the AC side current of the single-phase quasi-Z-source rectifier before and after adding the delay compensation strategy in the embodiment of the present invention;

[0031] Figure 6 The present invention is a comparison diagram of the percentage of the harmonic content of the AC side current of the single-phase quasi-Z-source rectifier to the fundamental current content (THD) under different input voltages and output powers under the improved fractional-order finite control set model predictive control, integer-order finite control set model predictive control and fractional-order finite control set model predictive control in an embodiment of the present invention, wherein (a) is a comparison diagram of the THD under three types of control when the input voltage is different, and (b) is a comparison diagram of the THD under three types of control when the output power is different;

[0032] Figure 7 It is a simulation waveform comparison diagram of the output voltage change of a single-phase quasi-Z-source rectifier when a given output voltage changes under the improved fractional-order finite control set model predictive control and the traditional PI double closed-loop control and the fractional-order finite control set model predictive control in an embodiment of the present invention, wherein (a), (b) and (c) respectively represent the comparison diagrams of the voltage waveforms under the traditional PI double closed-loop control, the fractional-order finite control set model predictive control and the improved fractional-order finite control set model predictive control;

[0033] Description of the symbols in the figure: V g is the AC input voltage, I g is the input current on the AC side, V Co is the DC side output voltage. IOMPC is integer order model predictive control, FOMPC is fractional order model predictive control, and TDC-FOMPC is improved fractional order model predictive control. DETAILED DESCRIPTION

[0034] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0035] The present invention proposes an improved fractional-order finite control set model predictive control method for a single-phase quasi-Z-source rectifier, which enables the rectifier to operate at unity power factor and improves the steady-state and dynamic performance of the rectifier.

[0036] The fractional-order finite control set model control method based on a single-phase quasi-Z-source rectifier comprises the following steps:

[0037] In step S101, the input current on the AC side, the inductor current and capacitor voltage of the quasi-Z source network on the DC side are selected as control targets, and the operating characteristics of the single-phase quasi-Z source rectifier are analyzed. Based on the analysis of the AC side circuit, the state equation corresponding to the input current on the AC side in different states of the rectifier bridge switch tube is obtained. At the same time, in order to reduce the distortion rate of the input current on the AC side and improve the power quality, fractional-order calculus operators are introduced to replace integer-order state equations with fractional-order state equations. Based on the modal analysis of the DC side equivalent circuit, the state equations of the inductor current and capacitor voltage of the quasi-Z source network in different circuit states are obtained. When the circuit is in a non-through state, the switch tube is turned on, and when the circuit is in a through state, the switch tube is turned off.

[0038] In step S102, the state equations of the selected inductor current and capacitor voltage are discretized by using the forward Euler method to obtain the prediction model of the required current and voltage, and the possible combinations of switch states of the rectifier bridge switch tube and the quasi-Z source network switch tube are listed, and the combinations are matched with the prediction model to construct the value function of the associated multi-objective control, wherein the reference value of the input side AC current is obtained by multiplying the phase of the input AC voltage by the outer loop output of the quasi-Z source network capacitor voltage through the PI control, and the reference value of the quasi-Z source network inductor current is obtained by the output DC voltage through the outer loop output of the PI control. Considering the delay in the sampling and calculation process, the control method is improved by the two-step prediction method, and the delay time compensation is added.

[0039] In step S103, the required voltage and current are sampled by voltage and current sensors, and predictive control is performed through the improved fractional-order finite control set model based on the sampled values ​​and calculated values. The switch state combination that minimizes the value function is obtained by traversal and optimization, and is determined as the optimal switch state combination. Based on the optimal switch state combination, the switch states of the rectifier bridge switch tube and the switch tube in the quasi-Z source network are determined.

[0040] By analyzing the AC side circuit based on the topological structure of the single-phase quasi-Z-source rectifier, the state equation of the AC side input current can be obtained as follows:

[0041]

[0042] Among them, L g is the AC side filter inductor; i g ,u g ,u pn They represent the input current on the AC side, the input voltage on the AC side and the output voltage of the rectifier bridge respectively; S1 represents different operating modes of the rectifier, which is related to the different switching states of the rectifier bridge switch tube, and its value is -1, 0 or 1.

[0043] In order to improve the power quality, reduce the distortion rate of the input current on the AC side, and increase the flexibility of control, the fractional-order calculus operator α is introduced, and the above state equation becomes:

[0044]

[0045] Discretizing the above equations using the forward Euler method, we can obtain:

[0046]

[0047] Among them, i g (k+1) and i g (k) represents the AC input current value at the k+1th and kth sampling moments respectively; T s To control the cycle.

[0048] Combining the above two equations, the fractional-order prediction expression of the input current on the AC side can be obtained as follows:

[0049]

[0050] Among them, u g (k) and u pn (k) represent the AC side input voltage and the rectifier bridge output voltage value at the kth sampling moment respectively.

[0051] By analyzing the DC side of the single-phase quasi-Z-source rectifier, the mathematical model of the equivalent circuit can be obtained as follows:

[0052]

[0053] Among them, the left equation is the state equation of the circuit in the direct state, and the right equation is the state equation of the circuit in the non-direct state. L1 and L2 are the inductors in the quasi-Z source network, C1 and C2 are the capacitors in the quasi-Z source network, and C o is the output side capacitance, and R is the output side load.

[0054] Based on the mathematical model of the DC side of the above single-phase quasi-Z-source rectifier, the state equation of the inductor L2 current is:

[0055]

[0056] Among them, i L2 ,u C1 ,u C2 and u Co They represent the current of the quasi-Z source network inductor L2, the voltage of the quasi-Z source network capacitors C1 and C2, and the output side capacitor C ovoltage; S2 is the switching state of the quasi-Z source network switch tube, which takes a value of 0 or 1. When the rectifier works in the through state, S2 takes a value of 0, and when the rectifier works in the non-through state, S2 takes a value of 1.

[0057] Similarly, based on the state equation of the current, the forward Euler method is used to obtain the prediction expression of the current of the quasi-Z source network inductor L2:

[0058]

[0059] Among them, i L2 (k+1) and i L2 (k) represents the inductor current value of the quasi-Z source network at the k+1th and kth sampling moments respectively; u C2 (k) and u C1 (k) and u Co (k) respectively represent the capacitor voltage value and output DC voltage value of the quasi-Z source network at the kth sampling moment; S2 is the same as above.

[0060] The state equation for the capacitor C2 voltage is:

[0061]

[0062] Among them, i L1 ,i L2 and i pn They respectively represent the current of the quasi-Z source network inductor L1, the current of the quasi-Z source network inductor L1 and the output current of the rectifier bridge; S2 is the same as above.

[0063] Similarly, based on the voltage state equation, the forward Euler method is used to obtain the predicted expression of the voltage of the quasi-Z source network capacitor C2:

[0064]

[0065] Among them, u C2 (k+1) and u C2 (k) represents the capacitor voltage value of the quasi-Z source network at the k+1th and kth sampling moments respectively; i pn (k), i L2 (k) and i L1 (k) respectively represent the output current value of the rectifier bridge and the inductor current value of the quasi-Z source network at the kth sampling moment; S2 is the same as above.

[0066] The single-phase quasi-Z-source rectifier has five switch tubes. The combination of the switch states of different switch tubes corresponds to different current and voltage prediction models through corresponding different values ​​of S1 and S2. The corresponding relationship is shown in Table 1:

[0067] Table 1 Correspondence table

[0068] Serial number n 1 2 3 4 5 6 7 8 9 Switching tube status 01101 10011 01011 10101 01110 10110 11010 11100 11110 S1 -1 1 0 0 0 0 0 0 0 S2 1 1 1 0 0 0 0 0 0

[0069] In order to evaluate the predicted values ​​of current and voltage under different combinations of switch states, the cost function of multi-objective control associated with the controlled voltage and current is constructed as:

[0070]

[0071] Among them, i g (k+1), i L2 (k+1) and u C2 (k+1) respectively represent the input current value of the AC side and the inductor current and capacitor voltage values ​​of the quasi-Z source network at the k+1th sampling moment; λ is the weight factor, which is a fixed constant, and its size can highlight the focus of multi-objective control; i gref 、i L2ref and u C2ref Respectively represent the reference values ​​of the input current on the AC side and the inductor current and capacitor voltage of the quasi-Z source network. gref It is obtained by multiplying the phase of the outer loop output of the voltage of the quasi-Z source network capacitor C2 controlled by PI and the input voltage on the AC side to achieve unity power factor operation; i L2ref Obtained by the outer loop output of the DC output voltage controlled by PI; u C2ref is a fixed constant.

[0072] By minimizing the value function, the optimal state combination of the five switching tubes can be obtained, and at each moment, a rolling optimization can be performed through the fractional-order prediction model to control each switching tube in real time, thereby achieving the goal of the rectifier operating at unity power factor and outputting a stable DC voltage.

[0073] Considering that the fractional-order model predictive control has the characteristic of one-step delay, that is, the optimal switch state combination of the switch tube obtained in this cycle will act on the switch tube at the beginning of the next cycle, and the existence of control delay will affect the control accuracy and control effect of the algorithm, a two-step prediction method of one-step prediction compensation delay plus fractional-order model prediction is used to compensate for the control delay commonly existing in this method, that is, the optimal switch state acting at this moment is obtained in the previous cycle. Taking the input current on the AC side as an example, at t k At this moment, the optimal switch state obtained in the previous cycle is applied to sample the AC input current value i at the moment tk. g (k), based on the sampled value and the switch state, the prediction model is used to calculate the AC input current value i at time t(k+1) g (k+1), in i g(k+1), and then make a prediction based on the prediction value i at time t(k+2) under different switch states using the value function of multi-objective control. g (k+2) is used for evaluation, and the optimal switch state is obtained and acts on the system at the beginning of the next sampling period, that is, at time t(k+1). The fractional-order model predictive control algorithm with time delay compensation has better real-time control performance.

[0074] It is understandable that, ideally, the sampling of the required voltage and current values, the calculation of the prediction model, and the optimization process of the value function of the multi-objective control are all performed at t k The optimal switch state combination occurs at the sampling moment and acts on the switch tube, thereby achieving k+1 At the sampling moment, the error between the actual value of the control target and the given reference value is minimized. However, in actual digital signal processing systems, sampling and calculation require a certain amount of time, and the optimal switch state obtained in this cycle will not be applied to the switch until the next sampling cycle arrives. Therefore, there is a one-cycle delay in model predictive control, and the control accuracy will be affected.

[0075] In order to solve the above delay problem, a two-step prediction method is used to compensate for the delay, and an improved fractional-order model predictive control based on the time delay compensation strategy is adopted to control the single-phase quasi-Z source rectifier. The specific process is as follows: at t k At the moment, the optimal switch state combination obtained in the k-1th sampling cycle acts on the switch tube, and the sampling t k The required current and voltage values ​​at time t are obtained by predicting the control target at time t based on the voltage value and the optimal switch state. k+1 The predicted value at time t k+2 The predicted values ​​of the controlled current and voltage under different switch states at the time and the corresponding value function are selected, and the switch state that minimizes the value function is selected as the optimal switch state. k+1 Acting on each switch tube in the circuit at all times.

[0076] According to the above principle, the controlled current and voltage are k+2 The prediction model at time t can be respectively k+1 The prediction model for the moment is launched.

[0077] The input current of AC side is at t k+2 The prediction model at this moment is:

[0078]

[0079] The current of the quasi-Z source network inductor L2 is k+2 The prediction model at this moment is:

[0080]

[0081] The voltage of the capacitor C2 of the quasi-Z source network is k+2 The prediction model at this moment is:

[0082]

[0083] Use the following value function to select in advance k+1 The optimal switching state at the moment:

[0084]

[0085] It can be seen from the defined value function that the smaller the value function is, the smaller the error between the sampled values ​​of the controlled current and voltage and the given reference current and voltage values ​​is. Therefore, the switch state combination that minimizes the value function should be selected as the optimal switch state combination to control the five switch tubes in the circuit.

[0086] like Figure 1 As shown, it is a schematic diagram of the topological structure of a single-phase quasi-Z source rectifier in an embodiment of the present invention, including an input side AC power supply and a filter inductor, a rectifier bridge, a quasi-Z source network, and an output side filter capacitor and a load. The rectifier allows the bridge arm to be directly connected in a short time, the circuit reliability is high, and the step-up and step-down conversion can be realized. When the circuit is in the direct connection state, the switch tube S5 in the quasi-Z source network is turned off, and when the circuit is in the non-direct connection state, the switch tube S5 is turned on.

[0087] like Figure 2 FIG. 1 is a control block diagram of an improved fractional-order model predictive control for a single-phase quasi-Z-source rectifier in an embodiment of the present invention. C2ref 、u Coref 、i gref and i L2ref are the reference values ​​of the voltage of the quasi-Z source network capacitor C2, the output voltage, the AC side input current and the current of the quasi-Z source network inductor L2 respectively; i g (k), i L2 (k) and u C2 (k) are t k The AC input current, the current of the quasi-Z source network inductor L2, and the voltage value of the quasi-Z source network capacitor C2 at the moment i g (k+1), i L2 (k+1) and u C2 (k+1) are t k+1 The AC input current, the current of the quasi-Z source network inductor L2, and the voltage value of the quasi-Z source network capacitor C2 at the moment i g (k+2), i L2 (k+2) and u C2(k+2) are t k+2 The AC input current, the current of the quasi-Z source network inductor L2, and the voltage value of the quasi-Z source network capacitor C2 at the moment g n (x) is the value of the value function under different switch state combinations. As can be seen from the figure, the reference value of the input current on the AC side is obtained by multiplying the output of the quasi-Z source network capacitor C2 voltage outer loop under PI control and the phase of the input voltage on the AC side, and the reference value of the current of the quasi-Z source network inductor L2 is obtained by the output of the output side DC voltage outer loop under PI control. According to each sampled value and calculated value, the optimal switch state of each sampling cycle is obtained through the proposed fractional-order model predictive control based on two-step prediction improvement, and the opening and closing of the five switch tubes in the control circuit are controlled.

[0088] like Figure 3 As shown, it is a schematic diagram of the main flow of the improved fractional-order model predictive control for a single-phase quasi-Z-source rectifier in an embodiment of the present invention, including the following steps: defining the switch state combination and other parameters of the five switch tubes in the circuit; sampling the required current and voltage in the kth sampling period to obtain the discretized current and voltage; obtaining the prediction model by adding a prediction model with delay compensation; evaluating the associated value function composed of the reference value and the predicted value of the current and voltage under different switch states, and obtaining the switch state combination corresponding to the minimum value function; and applying the optimal switch state combination to the rectifier.

[0089] like Figure 4 As shown in FIG. 1 , it is a schematic diagram of the principle of the delay compensation strategy of the fractional-order finite control set model predictive control in an embodiment of the present invention. In an ideal state, the sampling and calculation of the model predictive control method and the optimization of the switch state are completed in an instant, achieving the control goal of a small error between the actual value of the controlled current and voltage and the reference value. However, the sampling and calculation of current and voltage take time, such as Figure 4 As shown in (a), at t k Sampling is performed, and the optimal switching state C3 can be applied to the circuit only after a time delay of sampling and calculation, resulting in a deviation between the predicted value and the actual value of the current and voltage at the beginning of the next switching cycle, and the control accuracy of the system is reduced. In addition, the optimal switching state obtained will actually be applied to the system when the next sampling cycle arrives, such as Figure 4 As shown in (b), S3 at t k+1 The optimal switching state obtained in the previous sampling cycle will only take effect at the beginning of the next sampling cycle. Therefore, the traditional model predictive control algorithm has a time delay of one cycle, which will affect the control accuracy of the algorithm. Therefore, an improved fractional-order model predictive control based on the two-step prediction method is adopted, such as Figure 4 As shown in (c), at t kAt the moment, the optimal switch state combination obtained in the k-1th sampling cycle acts on the switch tube, and the sampling t k The required current and voltage values ​​at time t are obtained by predicting the control target at time t based on the voltage value and the optimal switch state. k+1 The predicted value at time t k+2 The predicted values ​​of the controlled current and voltage under different switch states at the time and the corresponding value function are selected, and the switch state that minimizes the value function is selected as the optimal switch state. k+1 It acts on each switch in the circuit at all times, which compensates for the control delay and improves the control accuracy of the algorithm.

[0090] like Figure 5 As shown in (a)-(d), both the fractional-order model predictive control and the improved fractional-order model predictive control based on the delay compensation strategy can achieve the same phase of the AC side voltage and current, the rectifier operates at unity power factor, and the DC side output voltage is stable at 320V. Before the delay compensation strategy was adopted, the current harmonic content of the AC side current as a percentage of the fundamental current content (THD) was 2.51%, and the fluctuation of the DC side output voltage was 0.35V. After the delay compensation strategy was adopted, the THD of the AC side current was 2.10%, and the fluctuation of the DC side output voltage was 0.175V. It can be seen that after the delay compensation strategy is adopted, the THD of the AC side current is reduced, the fluctuation of the DC side output voltage is reduced, and the steady-state performance of the system is improved.

[0091] like Figure 6 As shown in (a)-(b), the THD of the AC side current increases with the increase of input voltage and decreases with the increase of output power. In addition, it can be seen that the fractional-order model predictive control is better than the integer-order model predictive control, and the improved fractional-order model predictive control is better than the fractional-order model predictive control. It can be seen that the improved fractional-order model predictive control based on the delay compensation strategy is more effective in harmonic compensation, reduces more harmonic pollution on the AC side, and can obtain better power quality.

[0092] like Figure 7As shown in (a)-(c), when the given value of the DC side output voltage suddenly increases, the amplitude of the AC side input current increases, and when the given value of the DC side output voltage suddenly decreases, the amplitude of the AC side input current decreases. During the period when the DC voltage changes, the phase of the AC side input current is still the same as the phase of the input voltage, the output voltage of the rectifier is adjustable, and has good dynamic performance. It can be seen from the figure that under the double closed-loop control, the DC side output voltage takes 0.4s to achieve stable tracking of the reference value; under the fractional-order model predictive control, the DC side output voltage takes 0.14s to achieve stable tracking of the reference value; under the improved fractional-order model predictive control based on the delay compensation strategy, the DC side output voltage takes 0.07s to achieve stable tracking of the reference value. Therefore, using the improved fractional-order model predictive control, the system has the fastest dynamic response and can reach steady state in the shortest time.

[0093] According to the fractional-order finite control set model control method based on a single-phase quasi-Z source rectifier proposed in an embodiment of the present invention, PI control is performed on the capacitor voltage of the quasi-Z source network and the DC voltage on the output side to obtain reference values ​​of the input side AC current and the inductor current of the quasi-Z source network; in order to reduce the distortion rate of the input current on the AC side, a fractional-order calculus operator is introduced into the prediction model of the input side inductor current, and the state equation is discretized by using the forward Euler method to obtain the prediction models of the input side AC current, the capacitor voltage of the quasi-Z source network and the inductor current, and an associated multi-control objective value function is established; based on the influence of sampling and calculation delay time on control accuracy, a two-step prediction method is used to compensate for time delay of the control strategy; the required voltage and current are sampled, and based on all pre-defined switch states, the value functions under different switch states are solved, and the switch state that minimizes the value function is determined as the optimal switch state by traversal optimization; the optimal switch state is used to control the rectifier bridge and the switch tube of the quasi-Z source network to achieve the control target. The implementation of the present invention can have the following effects: compared with the traditional integer-order finite control set model predictive control and the fractional-order finite set model predictive control without delay compensation, it reduces the input side AC current distortion rate, reduces the fluctuation of the output side DC voltage, improves the dynamic response speed of the output side DC voltage, and has higher control accuracy.

[0094] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0095] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "N" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0096] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

Claims

1. A fractional-order finite control set model control method based on a single-phase quasi-Z-source rectifier, characterized in that: The following steps are involved: The input current on the AC side, the inductor current and capacitor voltage of the quasi-Z source network on the DC side are selected as control targets, and the operating characteristics of the single-phase quasi-Z source rectifier are analyzed. Based on the analysis of the AC side circuit, the state equation corresponding to the input current on the AC side in different states of the rectifier bridge switch tube is obtained. The fractional-order calculus operator is introduced, and the integer-order state equation is replaced by the fractional-order state equation. Based on the modal analysis of the DC side equivalent circuit, the state equations of the inductor current and capacitor voltage of the quasi-Z source network in different circuit states are obtained. When the circuit is in a non-through state, the switch tube is turned on, and when the circuit is in a through state, the switch tube is turned off. The forward Euler method is used to discretize the state equations of the selected inductor current and capacitor voltage to obtain the prediction model of the required current and voltage. The possible combinations of switch states of the rectifier bridge switch tube and the quasi-Z source network switch tube are listed, and the combinations are matched with the prediction model to construct the value function of the associated multi-objective control and compensate for the delay time. The required voltage and current are sampled through voltage and current sensors. According to the sampled values ​​and calculated values, predictive control is performed through the improved fractional-order finite control set model. The switch state combination that minimizes the value function is obtained by traversal optimization and determined as the optimal switch state combination. According to the optimal switch state combination, the switch states of the rectifier bridge switch tube and the switch tube in the quasi-Z source network are determined.

2. The method according to claim 1, characterized in that The AC side circuit of the single-phase quasi-Z-source rectifier is analyzed, and the fractional-order calculus operator is introduced to obtain the state equation of the input side AC current. The fractional-order state equation is discretized using the forward Euler method, and the fractional-order prediction expression of the input side AC current is obtained as follows: Among them, i g (k+1) and i g (k) represents the AC input current value at the k+1th and kth sampling moments respectively; T s is the control period; L g is the inductance value of the AC side; u g (k) and u pn (k) represents the AC input voltage and the rectifier bridge output voltage at the kth sampling moment; α is the order of the introduced fractional calculus operator; S1 is the operation mode of the rectifier, which takes the value of -1, 0 or 1; The DC side circuit of the single-phase quasi-Z-source rectifier is analyzed to obtain the state equations of the capacitor voltage and inductor current of the quasi-Z-source network. The state equations are discretized using the forward Euler method to obtain the prediction expression of the inductor current: Among them, i L2 (k+1) and i L2 (k) represents the inductor current value of the quasi-Z source network at the k+1th and kth sampling moments respectively; u C2 (k) and u C1 (k) and u Co (k) respectively represent the capacitor voltage value and output DC voltage value of the quasi-Z source network at the kth sampling moment; S2 is the switch state of the quasi-Z source network switch tube, which takes a value of 0 or 1. When the rectifier works in the direct-flow state, S2 takes a value of 0, and when the rectifier works in the non-direct-flow state, S2 takes a value of 1; The predicted expression of capacitor voltage is obtained as: Among them, u C2 (k+1) and u C2 (k) represents the capacitor voltage value of the quasi-Z source network at the k+1th and kth sampling moments respectively; i pn (k), i L2 (k) and i L1 (k) respectively represent the output current value of the rectifier bridge and the inductor current value of the quasi-Z source network at the kth sampling moment.

3. The method according to claim 1, characterized in that The value function of building associated multi-objective control is: Among them, i g (k+1), i L2 (k+1) and u C2 (k+1) respectively represent the input current value of the AC side and the inductor current and capacitor voltage value of the quasi-Z source network at the k+1th sampling moment; λ is the weight factor; i gref 、i L2ref and u C2ref They represent the reference values ​​of the AC input current and the inductor current and capacitor voltage of the quasi-Z source network respectively; i gref It is obtained by multiplying the output of the capacitor voltage outer loop of the quasi-Z source network controlled by PI with the phase of the AC side input voltage to achieve unity power factor operation; i L2ref Obtained by the output of the DC output voltage outer loop controlled by PI; u C2ref is a fixed constant.

4. The method according to claim 1, characterized in that Compensation for delay time, including: A two-step prediction method of one-step prediction compensation delay plus fractional-order model prediction is used to compensate for control delay, that is, the optimal switch state acting at this moment is obtained in the previous cycle.

Images