CI-SIDO Buck converter control method based on variable gain superhelix ESO

By adopting variable gain superhelical ESO and generalized superhelical sliding mode algorithm in single-inductor dual-output converter, the problems of coupling and cross-influence are solved, the system's decoupling and robustness are improved, and the anti-interference is significantly improved Performance and response speed.

CN118944428APending Publication Date: 2024-11-12SHAANXI SCI TECH UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410761987.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-13
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

When the existing single-inductor dual-output converter control method faces the cross-influence of coupling and output branches, especially for coupled single-inductor dual-output converters, there are problems of complex design and difficulty in decoupling. Traditional linear control The method does not apply.

Method used

Using a control method based on variable gain superhelical ESO, the hyperbolic function replaces the symbolic function in the superhelical sliding mode expansion state observer, designs a variable gain function that can change with observation error, and introduces a generalized superhelical sliding mode algorithm and integral The fast non-singular sliding mode surface is combined with adaptive law to improve the robustness and immunity of the system.

Benefits of technology

It effectively suppresses the cross-influence between output branches, improves the robustness and transient response speed of the system, realizes the decoupling of the system, and improves the immunity performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118944428A_ABST
    Figure CN118944428A_ABST
Patent Text Reader

Abstract

According to the CI-SIDO Buck converter control method based on the variable-gain superhelix ESO, improvement is carried out on the basis of superhelix ESO and superhelix sliding mode control, a hyperbolic function is adopted to replace a switching function of a superhelix expansion state observer, system buffeting is reduced, and in order to improve the observation capacity of the observer for disturbance, the CI-SIDO Buck converter control method based on the variable-gain superhelix ESO is improved. A variable gain function capable of adjusting the observation gain along with the observation error is designed; an integral type nonsingular fast superhelix surface is designed for a CI-SIDO Buck converter to be combined with a self-adaptive superhelix sliding mode controller to control the system, the robustness of the system is improved, the cross influence between output branches is restrained, system decoupling is achieved, and the anti-interference performance and the transient response speed of the system are further improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of power electronic control, and in particular relates to a CI-SIDO Buck converter control method based on a variable gain super-helical ESO. Background Art

[0002] Existing related literature on coupled single inductor dual output converter (CI-SIDO) has studied the converter's working mode, voltage and current ripple analysis, coupling coefficient selection, and cross-influence analysis. Among them, in the decoupling control of CI-SIDOBuck converter, the main method at home and abroad is to compensate the model based on the small signal mathematical model of the converter to achieve decoupling; on the other hand, the inverse dynamics system is constructed through precise feedback linearization for decoupling. In the fields with strong coupling effects such as SI-SIDOBuck converters, motors, and drones, the control strategy of decoupling the system by using an extended state observer combined with nonlinear control is widely used.

[0003] At present, there are two main strategies for the control methods of traditional single-inductor dual-output converters at home and abroad. One is a linear control method based on timing and transfer function, which realizes system decoupling and suppresses cross-influence, but this method is too dependent on the mathematical model of the converter, and has limitations in the treatment of unmodeled parts and internal disturbances such as parasitic resistance. The other is to use a nonlinear control scheme to design the system by taking advantage of the characteristics of not relying on mathematical models, which shows good results in decoupling and can improve system performance. The coupled single-inductor dual-output converter has similar coupling and cross-influence between output branches as the traditional single-inductor dual-output converter, but the coupled single-inductor dual-output converter uses a reverse coupled inductor, the system is a fourth-order system, and the energy supply mode of the inductor is very different from that of the SI-SIDO converter. The linear decoupling control method based on timing is not suitable for the coupled single-inductor dual-output converter.

[0004] Due to the modeling complexity and model coupling of CI-SIDO converters, the design of linear control in CI-SIDO converters is relatively complicated. Designing a control method with a simple control structure that does not require complex modeling methods and can decouple the model has become a research problem that needs to be solved. Summary of the invention

[0005] The purpose of the present invention is to provide a control method for a CI-SIDO Buck converter based on a variable gain super-helical ESO, which improves the robustness of the system and suppresses the cross-effect between output branches.

[0006] The technical solution adopted by the present invention is a CI-SIDO Buck converter control method based on a variable gain superhelical ESO. First, a hyperbolic function is used to replace the sign function in a superhelical sliding mode extended state observer, and a variable gain function that can change in real time with the observation error is designed. For the superhelical sliding mode controller, a generalized superhelical sliding mode algorithm with linear terms is introduced as the reaching law of the system to smooth the system control law, and an adaptive law is designed to weaken the dependence of the generalized superhelical sliding mode algorithm on the disturbance boundary. An integral fast non-singular sliding mode surface is designed, and the convergence speed of the system is further improved by the linear terms in the integral fast non-singular sliding mode surface.

[0007] The present invention is also characterized in that:

[0008] Follow the steps below to implement it:

[0009] Step 1: Improved supercoil ESO design

[0010] Step 1.1: Design of inner ring improved supercoil ESO;

[0011] Step 1.2: Design of outer ring improved supercoil ESO;

[0012] Step 2: Design of an integral-type fast non-singular adaptive super-helical sliding mode controller

[0013] Step 2.1: Design of improved super-helical sliding mode controller for the current inner loop;

[0014] Step 2.2: Design of improved super-helical sliding mode controller for voltage outer loop.

[0015] Step 1.1 is as follows:

[0016] The state space average model of SIDO Buck converter is:

[0017]

[0018] In formula (3.6), v in is the input voltage, L1 and L2 are self-inductance, M is mutual inductance, the two branches of SIDO Buck converter are branch a and branch b, C a , C b is the output filter capacitor of branch a and branch b, i L1 、i L2 is the inductor current of branch a and branch b, i oa 、i ob is the output current of branch a and branch b, d a d b is the duty cycle of branch a and branch b, v a 、v bis the output voltage of branch a and branch b;

[0019] The unmodeled part of the current inner loop and other uncertain disturbances are regarded as the internal disturbances f of the output branches a and b. ca 、f cb , then according to formula (3.6), the inductor current state equation of branch a and branch b is restated as:

[0020]

[0021] In formula (3.7), x c1 、x c2 is the state quantity of output branch a and branch b;

[0022] The inner loop coupling term and the unmodeled part as well as the internal and external disturbances of the inner loop are regarded as the total disturbance of the inner loop, and f c1 , f c2 It means that, combining equation (3.6) and equation (3.7), we get:

[0023]

[0024] In formula (3.8), f ca 、f cb is the total disturbance of the unmodeled part of the inner loop of branch a and branch b as well as the internal and external parts;

[0025] The error term in the improved superhelical ESO adopts a smooth hyperbolic function, the hyperbolic function F(e i ) function is expressed as:

[0026]

[0027] In formula (4.2), e i is the observation error; m is greater than 0, which is the adjustable hyperbola slope value;

[0028] The improved supercoil ESO expression is:

[0029]

[0030] In formula (4.3), β i is the variable gain function, e i is the observer error, θ 1i for y i The observed value, θ 2i is the observed value of the total disturbance;

[0031]

[0032] In formula (4.4), β 1i , β 2iis the observer variable gain coefficient;

[0033] The expression of variable gain L(t) is:

[0034]

[0035] In formula (4.5), ω i , All are greater than 0 and are adjustable positive numbers;

[0036] According to equation (3.8) and combined with equation (4.3), the variable gain ST-ESO of the converter current inner loop is established as:

[0037]

[0038] In formula (4.6), e c1 and e c1 are the errors between the observed values ​​and the actual values ​​of the inner loop observers of branches a and b respectively; x c1 and x c2 is the actual value of the inner loop observer of branch a and branch b, β1 and β2 are the variable gain functions of the observer of branch a; β3 and β4 are the variable gain functions of the observer of branch b.

[0039] Step 1.2 is as follows:

[0040] The system outer loop branches a and b that are susceptible to unmodeled partial disturbances and other uncertain disturbances are regarded as outer loop disturbances f v1 、f v2 , then according to formula (3.6), the voltage outer loop is re-expressed as:

[0041]

[0042] In formula (3.17), x v1 、x v2 is the voltage state quantity of output branch a and branch b.

[0043] In order to improve the anti-disturbance ability of the system, the i in the outer loop that is susceptible to external disturbances is oa 、i ob The unmodeled part and the inner and outer disturbances of the outer loop are regarded as the total disturbance of the outer loop, and f v1 、f v2 Expressed, combining equation (3.6) and equation (3.17) we get:

[0044]

[0045] In formula (3.18), f va 、f vb The unmodeled parts of branch a and branch b as well as the internal and external disturbances;

[0046] According to equation (3.18) and equation (4.3), the variable gain ST-ESO of the converter voltage outer loop, i.e. the outer loop improved super helical ESO, is established as:

[0047]

[0048] In formula (4.7), e v1 、e v1 are the errors between the observed values ​​and actual values ​​of the voltage outer loop observer of branch a and branch b respectively; v1 、x v2 is the actual value of the voltage outer loop observer of branch a and branch b, β5 and β6 are the variable gain functions of the observer of branch a; β7 and β8 are the variable gains of the observer of branch b.

[0049] Step 2.1 is as follows:

[0050] Select the non-singular fast terminal sliding surface S cvi for:

[0051]

[0052] In formula (4.21), λ1, λ2>0,

[0053] When the system enters the sliding mode, Right now:

[0054]

[0055] Derivative and simplified formula (4.22) yields:

[0056]

[0057] From the analysis of formula (4.23), we know that the linear terms and play different leading roles in different motion stages of the system, achieving global rapid convergence, and introducing integral terms when constructing the sliding surface, without differential terms, avoiding singular phenomena;

[0058] A parameter adaptive law is introduced to further improve the generalized super-helical sliding mode controller. The adaptive super-helical sliding mode control law is designed as follows:

[0059]

[0060] In formula (4.24), m a is the integral of the switching term. k2=2εk1+β+4ε 2 ,in γ1,μ,ε,β,k mAll are positive numbers, k1 and k2 are adjusted adaptively along with the sliding surface;

[0061] According to equations (3.8) and (4.6), the VGST-ESO error system of the inner current loop of the CI-SIDOBuck converter is constructed as follows:

[0062]

[0063] The fast non-singular sliding mode surface of the converter inner loop integral type designed by equations (4.20) and (4.25) is:

[0064]

[0065] Derivative of equation (4.26) yields:

[0066]

[0067] Substituting equation (4.25) into equation (4.27), we get:

[0068]

[0069] make The adaptive super-helical sliding mode control law designed by combining formula (4.24) is used to obtain the current inner loop control law:

[0070]

[0071] Step 2.2 is as follows:

[0072] Combining equations (3.18) and (4.7), we get the error system of the CI-SIDOBuck converter VGST-ESO as:

[0073]

[0074] The converter voltage outer loop integral type fast non-singular terminal sliding mode surface constructed by combining equations (4.21) and (4.30) is:

[0075]

[0076] Derivative of equation (4.31) yields:

[0077]

[0078] Substituting equation (4.30) into equation (4.32), we get:

[0079]

[0080] make Combined with equation (4.33), the voltage outer loop control law is obtained as:

[0081]

[0082] The beneficial effects of the present invention are:

[0083] The invention discloses a control method for a CI-SIDO Buck converter based on a variable gain superhelical ESO, which is improved on the basis of superhelical ESO and superhelical sliding mode control, and adopts a hyperbolic function to replace the switching function of a superhelical extended state observer to reduce system jitter. In order to improve the observation capability of the observer to disturbances, a variable gain function that can adjust the observation gain with the observation error is designed. An integral non-singular fast superhelical surface is designed for the CI-SIDO Buck converter, and an adaptive superhelical sliding mode controller is combined with the system to control the system, thereby improving the robustness of the system, suppressing the cross-influence between output branches, realizing system decoupling, and further improving the anti-disturbance performance and transient response speed of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is the block diagram of the improved sliding mode control of the CI-SIDOBuck converter;

[0085] Figure 2 This is the block diagram of adaptive sliding mode decoupling control based on variable gain super-helical ESO;

[0086] Figure 3 is the hyperbola F(e i )The trajectory diagram of the function;

[0087] Figure 4 is i oa The experimental waveform of the mutation, where Figure 4(a) is i oa The experimental waveform of 1A→2A, Figure 4(b) is i oa Experimental waveform of 2A→1A;

[0088] Figure 5 is i oa The experimental waveform of the mutation, where Figure 5(a) is i ob The experimental waveform of 1A→2A, Figure 5(b) is i ob Experimental waveform of 2A→1A;

[0089] Figure 6 is v in The experimental results of the sudden change are shown in Figure 6(a) , where the input voltage is 45V→35V, and Figure 6(b) , where the input voltage is 35V→55V. DETAILED DESCRIPTION

[0090] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0091] Example 1

[0092] This embodiment provides a control method for a CI-SIDO Buck converter based on a variable gain super-twisting expanded state observer (VGST-ESO). Figure 1 As shown in the figure, firstly, a hyperbolic function is used to replace the sign function in the super-helical sliding mode extended state observer to reduce the system chattering, and a variable gain function that can change in real time with the observation error is designed to improve the observation ability of disturbances. For the super-helical sliding mode controller, a generalized super-helical sliding mode algorithm with linear terms is introduced as the reaching law of the system to smooth the system control law, and an adaptive law is designed to weaken the dependence of the generalized super-helical sliding mode algorithm on the disturbance boundary, thereby improving the anti-disturbance ability of the system. In order to further improve the robustness of the system, an integral fast non-singular sliding surface is designed, and the linear terms in the integral fast non-singular sliding surface are used to further improve the convergence speed of the system, improve the overall performance of the system, and ensure the stability and anti-disturbance performance of the control.

[0093] The block diagram of adaptive sliding mode decoupling control based on variable gain super-helical ESO is as follows: Figure 2 shown.

[0094] Example 2

[0095] This embodiment provides a control method for a CI-SIDO Buck converter based on a variable gain super-helical ESO. Based on Embodiment 1, the method is specifically implemented according to the following steps:

[0096] Step 1: Improved supercoil ESO design

[0097] Compared with the linear extended state observer (ESO), the super-helical sliding mode extended state observer (ST-ESO) has higher observation accuracy and better robustness, but the error term in the ST-ESO uses the switching function sign, which makes the system have certain chattering problems. In order to reduce the chattering problem, a smooth hyperbolic function is used instead of the discontinuous switching function sign.

[0098] Step 1.1: Design of inner ring improved supercoil ESO;

[0099] Step 1.1 is as follows:

[0100] The state space average model of SIDO Buck converter is:

[0101]

[0102] In formula (3.6), v inis the input voltage, L1 and L2 are self-inductance, M is mutual inductance, the two branches of SIDO Buck converter are branch a and branch b, C a , C b is the output filter capacitor of branch a and branch b, i L1 、i L2 is the inductor current of branch a and branch b, i oa 、i ob is the output current of branch a and branch b, d a d b is the duty cycle of branch a and branch b, v a 、v b is the output voltage of branch a and branch b;

[0103] The unmodeled part of the current inner loop and other uncertain disturbances are regarded as internal disturbances of output branches a and b. ca 、f cb , then according to formula (3.6), the inductor current state equation of branch a and branch b is restated as:

[0104]

[0105] In formula (3.7), x c1 、x c2 is the state quantity of output branch a and branch b;

[0106] The inner loop coupling term and the unmodeled part as well as the internal and external disturbances of the inner loop are regarded as the total disturbance of the inner loop, and f c1 , f c2 It means that, combining equation (3.6) and equation (3.7), we get:

[0107]

[0108] In formula (3.8), f ca 、f cb is the total disturbance of the unmodeled part of the inner loop of branch a and branch b as well as the internal and external parts;

[0109] The error term in the improved superhelical ESO adopts a smooth hyperbolic function, the hyperbolic function F(e i ) function is expressed as:

[0110]

[0111] In formula (4.2), e i is the observation error, m is greater than 0, and is the adjustable hyperbola slope value;

[0112] Hyperbolic function F(e i ) is trending as Figure 3As shown, by formula (4.2) and Figure 3 It can be seen that the switching function F(e i ) is a continuous and smooth function. Unlike the symbolic function sign, it does not have discontinuities. In theory, it can weaken the chattering problem and improve the observation capability of the generalized superhelical ESO to disturbances.

[0113] The improved supercoil ESO expression is:

[0114]

[0115] In formula (4.3), β i is the variable gain function, e i is the observer error, θ 1i for y i The observed value, θ 2i is the observed value of the total disturbance;

[0116]

[0117] In formula (4.4), β 1i , β 2i is the observer variable gain coefficient;

[0118] The expression of variable gain L(t) is:

[0119]

[0120] In formula (4.5), ω i , All are greater than 0 and are adjustable positive numbers;

[0121] According to equation (3.8) and combined with equation (4.3), the variable gain ST-ESO of the converter current inner loop is established as:

[0122]

[0123] In formula (4.6), e c1 and e c1 are the errors between the observed values ​​and the actual values ​​of the inner loop observers of branches a and b respectively; x c1 and x c2 is the actual value of the inner loop observer of branch a and branch b, β1 and β2 are the variable gain functions of the observer of branch a; β3 and β4 are the variable gain functions of the observer of branch b.

[0124] Step 1.2: Design of outer ring improved supercoil ESO;

[0125] Step 1.2 is as follows:

[0126] The system outer loop branches a and b that are susceptible to unmodeled partial disturbances and other uncertain disturbances are regarded as outer loop disturbances f v1 、f v2 , then according to formula (3.6), the voltage outer loop can be re-expressed as:

[0127]

[0128] In formula (3.17), x v1 、x v2 is the voltage state quantity of output branch a and branch b.

[0129] In order to improve the anti-disturbance ability of the system, the i in the outer loop that is susceptible to external disturbances is oa,ob The unmodeled part and the inner and outer disturbances of the outer loop are regarded as the total disturbance of the outer loop, and f v1 , f v2 Expressed, combining equation (3.6) and equation (3.17) we get:

[0130]

[0131] In formula (3.18), f va 、f vb The unmodeled parts of branch a and branch b as well as the internal and external disturbances;

[0132] According to equation (3.18) and equation (4.3), the variable gain ST-ESO of the converter voltage outer loop, i.e. the outer loop improved super helical ESO, is established as:

[0133]

[0134] In formula (4.7), e v1 、e v1 are the errors between the observed values ​​and actual values ​​of the voltage outer loop observer of branch a and branch b respectively; v1 and x v2 is the actual value of the voltage outer loop observer of branch a and branch b, β5 and β6 are the variable gain functions of the observer of branch a; β7 and β8 are the variable gains of the observer of branch b;

[0135] Step 2: Design of an integral-type fast non-singular adaptive super-helical sliding mode controller

[0136] Step 2.1: Design of improved super-helical sliding mode controller for the current inner loop;

[0137] Step 2.2: Design of improved super-helical sliding mode controller for voltage outer loop.

[0138] Example 3

[0139] This embodiment provides a control method for a CI-SIDO Buck converter based on a variable gain super-helical ESO. Based on Embodiment 1-2, step 2.1 is specifically as follows:

[0140] Select the non-singular fast terminal sliding surface S cvi for:

[0141]

[0142] In formula (4.21), λ1, λ2>0,

[0143] When the system enters the sliding mode, Right now:

[0144]

[0145] Derivative and simplified formula (4.22) yields:

[0146]

[0147] From the analysis of formula (4.23), we know that the linear terms and play different leading roles in different motion stages of the system, achieving global rapid convergence, and introducing integral terms when constructing the sliding surface, without differential terms, avoiding singular phenomena;

[0148] A parameter adaptive law is introduced to further improve the generalized super-helical sliding mode controller. The adaptive super-helical sliding mode control law is designed as follows:

[0149]

[0150] In formula (4.24), m a is the integral of the switching term. k2=2εk1+β+4ε 2 ,in γ1,μ,ε,β,k m They are all positive numbers, k1 and k2 can be adjusted adaptively along with the sliding surface;

[0151] According to equations (3.8) and (4.6), the VGST-ESO error system of the inner current loop of the CI-SIDOBuck converter is constructed as follows:

[0152]

[0153] The fast non-singular sliding mode surface of the converter inner loop integral type designed by equations (4.20) and (4.25) is:

[0154]

[0155] Derivative of equation (4.26) yields:

[0156]

[0157] Substituting equation (4.25) into equation (4.27), we get:

[0158]

[0159] make The adaptive super-helical sliding mode control law designed by combining formula (4.24) is used to obtain the current inner loop control law:

[0160]

[0161] Step 2.2 is as follows:

[0162] Combining equations (3.18) and (4.7), we get the error system of the CI-SIDOBuck converter VGST-ESO as:

[0163]

[0164] The converter voltage outer loop integral type fast non-singular terminal sliding mode surface constructed by combining equations (4.21) and (4.30) is:

[0165]

[0166] Derivative of equation (4.31) yields:

[0167]

[0168] Substituting equation (4.30) into equation (4.32), we get:

[0169]

[0170] make Combined with equation (4.33), the voltage outer loop control law is obtained as:

[0171]

[0172] Experimental analysis

[0173] In order to verify the effectiveness of the control method of the present invention, a simulation circuit was built in MATLAB / Simulink simulation software, and a comparative analysis was performed with the sliding mode decoupling control strategy based on super-bolt ESO. The controller parameters of the control method of the present invention are shown in Table 4-1.

[0174] Table 4-1 CI-SIDO Buck converter control parameters

[0175]

[0176]

[0177] (I) Anti-load disturbance experimental analysis

[0178] Figure 4 shows the experimental results of the control method of the present invention against branch a load disturbance. Analysis of Figure 4(a) shows that when the load of branch a suddenly changes from 1A to 2A, the voltage overshoot of branch a caused by the load disturbance is 0.5V, and the system recovers to a steady state after 2.5ms; the voltage overshoot of branch b caused by the cross-effect of branch a is 0.2V, and the system recovers to a steady state after 1.5ms.

[0179] From the analysis of Figure 4(b), we can see that when the load of branch a suddenly changes from 2A to 1A, the voltage overshoot caused by the load disturbance on branch a is 0.4V, and the system recovers to a steady state after 2ms; the voltage on branch b caused by the cross-shadow of branch a is 0.2V, and the system recovers to a steady state after 1.5ms.

[0180] Table 4-2 compares the experimental results of the sliding mode decoupling control based on the super bolt ESO and the decoupling control of the control method of the present invention.

[0181] Table 4-2 Comparison of experimental results of different decoupling controls

[0182]

[0183] By analyzing Figure 4 and combining with Table 4-2, it can be seen that the control method of the present invention realizes the decoupling between converters, suppresses the cross-influence of branch a on branch b, and further improves the resistance to branch a load disturbance and dynamic response based on the decoupling control based on superhelical ESO.

[0184] Figure 5 is the experimental result of resisting branch b load disturbance. Analysis of Figure 5(a) shows that when the branch b load suddenly changes from 1A to 2A, the voltage overshoot caused by the load disturbance on branch b is 0.2V, and the system recovers to steady state after 1.2ms; the voltage overshoot caused by the cross-effect of branch b on branch a is 0.3V, and the system recovers to steady state after 1.2ms.

[0185] From the analysis of Figure 5(b), we can see that when the load of branch b suddenly changes from 2A to 1A, the voltage overshoot caused by the load disturbance on branch b is 0.2V, and the system recovers to a steady state after 1.5ms; the voltage on branch a caused by the cross-shadow of branch b is 0.3V, and the system recovers to a steady state after 1.5ms.

[0186] Table 4-3 is a comparison of the experimental results of the sliding mode decoupling control based on the super bolt ESO and the decoupling control of the control method of the present invention to resist the b branch load disturbance.

[0187] Table 4-3 Comparison of experimental results of different decoupling controls

[0188]

[0189] By analyzing Figure 5 and combining with Table 4-3, it can be seen that the control method of the present invention realizes the decoupling between converters, suppresses the cross-influence of branch b on branch a, and further improves the resistance to branch b load disturbances and dynamic response based on the decoupling control in Chapter 3.

[0190] (II) Input voltage disturbance experiment analysis

[0191] From the analysis of FIG6(a), it can be seen that when the input voltage suddenly drops from 45V to 35V, the voltage overshoot caused by the input voltage disturbance in branch a is 0.3V, and the system recovers to a steady state after 1.8ms; the voltage overshoot caused by the input voltage disturbance in branch b is 0.2V, and the system recovers to a steady state after 1.8ms.

[0192] From the analysis of Figure 6(b), we can see that when the input voltage suddenly increases from 35V to 55V, the voltage overshoot caused by the input voltage disturbance in branch a is 0.5V, and the system recovers to a steady state after 1.5ms; the voltage overshoot caused by the input voltage disturbance in branch b is 0.2V, and the system recovers to a steady state after 2ms.

[0193] Table 4-3 is a comparison of the experimental results of the sliding mode decoupling control based on the super bolt ESO and the decoupling control of the control method of the present invention in terms of resistance to input voltage disturbances.

[0194] Table 4-4 Comparison of experimental results of different decoupling controls

[0195]

[0196] By analyzing FIG. 6 and combining with Table 4-4, it can be seen that in the case of resisting input voltage disturbance, the control method of the present invention further improves the ability to resist input voltage disturbance based on the control proposed based on superhelical ESO, and improves the robustness and dynamic performance of the system.

Claims

1. Control method of CI-SIDO Buck converter based on variable gain super-helical ESO, It is characterized in that Firstly, a hyperbolic function is used to replace the sign function in the superhelical sliding mode extended state observer, and a variable gain function that changes in real time with the observation error is designed. For the superhelical sliding mode controller, a generalized superhelical sliding mode algorithm with linear terms is introduced as the reaching law of the system to smooth the system control law. An adaptive law is designed to weaken the dependence of the generalized superhelical sliding mode algorithm on the disturbance boundary, and an integral fast non-singular sliding surface is designed. The convergence speed of the system is further improved by the linear terms in the integral fast non-singular sliding surface.

2. The control method of the CI-SIDO Buck converter based on variable gain super-helical ESO according to claim 1, It is characterized in that Follow the steps below to implement it: Step 1: Improved supercoil ESO design Step 1.1: Design of inner ring improved supercoil ESO; Step 1.2: Design of outer ring improved supercoil ESO; Step 2: Design of an integral-type fast non-singular adaptive super-helical sliding mode controller Step 2.1: Design of improved super-helical sliding mode controller for the current inner loop; Step 2.2: Design of improved super-helical sliding mode controller for voltage outer loop.

3. The control method of the CI-SIDO Buck converter based on variable gain super-helical ESO according to claim 2, It is characterized in that Step 1.1 is as follows: The state space average model of SIDO Buck converter is: In formula (3.6), v in is the input voltage, L 1 , L 2 is the self-inductance, M is the mutual inductance, the two branches of the SIDO Buck converter are branch a and branch b, C a , C b is the output filter capacitor of branch a and branch b, i L1 、i L2 is the inductor current of branch a and branch b, i oa 、i ob is the output current of branch a and branch b, d a d b is the duty cycle of branch a and branch b, v a 、v b is the output voltage of branch a and branch b; The unmodeled part of the current inner loop and other uncertain disturbances are regarded as the internal disturbances of output branch a and branch b. ca 、f cb , then according to formula (3.6), the inductor current state equation of branch a and branch b is restated as: In formula (3.7), x c1 、x c2 is the state quantity of output branch a and branch b; The inner loop coupling term and the unmodeled part as well as the internal and external disturbances of the inner loop are regarded as the total disturbance of the inner loop, and f c1 , f c2 It means that, combining equation (3.6) and equation (3.7), we get: In formula (3.8), f ca 、f cb is the total disturbance of the unmodeled part of the inner loop of branch a and branch b as well as the internal and external parts; The error term in the improved superhelical ESO adopts a smooth hyperbolic function, the hyperbolic function F(e i ) function is expressed as: In formula (4.2), e i is the observation error; m is greater than 0, which is the value of the hyperbola slope; The improved supercoil ESO expression is: In formula (4.3), β i is the variable gain function, e i is the observer error, θ 1i for y i The observed value, θ 2i is the observed value of the total disturbance; In formula (4.4), β 1i , β 2i is the observer variable gain coefficient; The expression of variable gain L(t) is: In formula (4.5), ω i , All are greater than 0 and are adjustable positive numbers; According to equation (3.8) and combined with equation (4.3), the variable gain ST-ESO of the converter current inner loop is established as: In formula (4.6), e c1 and e c1 are the errors between the observed values ​​and the actual values ​​of the inner loop observers of branches a and b respectively; x c1 and x c2 is the actual value of the inner loop observer of branch a and branch b, β 1 , β 2 is the variable gain function of the observer of branch a; β 3 , β 4 is the variable gain function of the observer of branch b.

4. The control method of the CI-SIDO Buck converter based on variable gain super-helical ESO according to claim 3, It is characterized in that Step 1.2 is as follows: The system outer loop branches a and b that are susceptible to unmodeled partial disturbances and other uncertain disturbances are regarded as outer loop disturbances f v1 、f v2 , then according to formula (3.6), the voltage outer loop is re-expressed as: In formula (3.17), x v1 、x v2 is the voltage state quantity of output branch a and branch b; In order to improve the anti-disturbance ability of the system, the i in the outer loop that is susceptible to external disturbances is oa 、i ob The unmodeled part and the inner and outer disturbances of the outer loop are regarded as the total disturbance of the outer loop, and f v1 、f v2 Expressed, combining equation (3.6) and equation (3.17) we get: In formula (3.18), f va 、f vb The unmodeled parts of branch a and branch b as well as the internal and external disturbances; According to equation (3.18) and equation (4.3), the variable gain ST-ESO of the converter voltage outer loop, i.e. the outer loop improved super-helical ESO, is established as: In formula (4.7), e v1 、e v1 are the errors between the observed values ​​and actual values ​​of the voltage outer loop observer of branch a and branch b respectively; v1 、x v2 is the actual value of the voltage outer loop observer of branch a and branch b, β 5 , β 6 is the variable gain function of the observer of branch a; β 7 , β 8 is the variable gain of the observer of branch b.

5. The control method of the CI-SIDO Buck converter based on variable gain super-helical ESO according to claim 4, It is characterized in that Step 2.1 is as follows: Select the non-singular fast terminal sliding surface S cvi for: In formula (4.21), λ 1 ,λ 2 >0, When the system enters the sliding mode, Right now: Derivative and simplified formula (4.22) yields: From the analysis of formula (4.23), we know that the linear terms and play different leading roles in different motion stages of the system, achieving global rapid convergence, and introducing integral terms when constructing the sliding surface, without differential terms, avoiding singular phenomena; A parameter adaptive law is introduced to further improve the generalized super-helical sliding mode controller. The adaptive super-helical sliding mode control law is designed as follows: In formula (4.24), m a is the integral of the switching term, k 2 =2εk 1 +β+4ε 2 ,in γ 1 , μ, ε, β, k m are all positive numbers, k 1 , k 2 With the sliding surface adaptive adjustment; According to equations (3.8) and (4.6), the VGST-ESO error system of the inner current loop of the CI-SIDOBuck converter is constructed as follows: The fast non-singular sliding mode surface of the converter inner loop integral type designed by equations (4.20) and (4.25) is: Derivative of equation (4.26) yields: Substituting equation (4.25) into equation (4.27), we get: make The adaptive super-helical sliding mode control law designed by combining equation (4.24) is used to obtain the current inner loop control law:

6. The control method of the CI-SIDO Buck converter based on variable gain super-helical ESO according to claim 5, It is characterized in that Step 2.2 is as follows: Combining equations (3.18) and (4.7), we get the error system of the CI-SIDOBuck converter VGST-ESO as: The converter voltage outer loop integral type fast non-singular terminal sliding mode surface constructed by combining equations (4.21) and (4.30) is: Derivative of equation (4.31) yields: Substituting equation (4.30) into equation (4.32), we get: make Combined with equation (4.33), the voltage outer loop control law is obtained as: