Boost converter model predictive control method and system based on disturbance observer

Abstract

The application belongs to the technical field of power electronic converter control, and particularly relates to a Boost converter model predictive control method and system based on a disturbance observer, which comprises the following steps: constructing a nonlinear disturbance observer (NDO) to estimate the lumped disturbance to which a Boost converter system model is subjected in real time, so as to obtain an input voltage estimation value and a load resistance estimation value; constructing a disturbance compensator to compensate the disturbance to the Boost converter system model by using the above-mentioned estimation values; constructing a model predictive controller (MPC) to construct a state prediction model according to the Boost converter system model subjected to the disturbance compensation, to construct a finite time domain optimization problem based on the state prediction model, and to generate an optimal control amount by solving the finite time domain optimization problem to realize tracking control. The application can guarantee the rapid recovery and small overshoot of the output voltage, and can keep the system output within a bounded range around the reference value when there is an estimation error, thereby embodying good dynamic performance.

Description

Technical Field

[0001] This invention belongs to the field of power electronic converter control technology, specifically relating to a Boost converter model predictive control method and system based on a disturbance observer. Background Technology

[0002] As the core device of a power conversion system, the accuracy and stability of the output voltage of the Boost converter are crucial to ensuring the good operating performance of downstream electrical equipment. However, the high-performance voltage control design of the Boost converter faces two major constraints. First, its inherent non-minimum phase characteristic and bilinear nonlinearity make it difficult for traditional linear controllers to balance stability and speed. Second, uncertainties such as sudden changes in load resistance, input voltage fluctuations, and parameter perturbations in actual operation severely affect control performance. Therefore, researching an efficient and feasible control strategy is of great practical significance for ensuring the stable and high-quality operation of the Boost converter.

[0003] Model Predictive Control (MPC) has shown great potential in power electronic converter control due to its unique advantages of explicitly handling system constraints and achieving multi-objective performance trade-offs through rolling optimization. However, on the one hand, when MPC is directly applied to the voltage loop of a Boost converter, its control performance heavily relies on the accuracy of the predictive model. Uncertainties such as sudden load resistance changes, input voltage fluctuations, and parameter perturbations in actual operation constitute lumped disturbances. This makes the composite MPC strategy, which requires online dynamic compensation based on the baseline model and has rigorous closed-loop stability proof, significantly mismatched with the predictive model based on nominal parameters. This leads to decreased MPC prediction accuracy, increased steady-state error, and even deterioration in stability. On the other hand, existing composite control schemes based on disturbance observers typically treat simultaneous load resistance changes and input voltage fluctuations as separate disturbances for estimation and compensation, rather than modeling them as a unified lumped disturbance. This not only limits the disturbance compensation efficiency but also increases the design complexity of the observer and controller. Furthermore, most existing studies have not conducted rigorous input-to-state stability analysis on the composite closed-loop system consisting of the observer and the model predictive controller, resulting in a lack of sufficient theoretical guarantees for the safety and stability of the system when estimation errors exist. Summary of the Invention

[0004] The purpose of this invention is to address the aforementioned problems in the prior art by providing a Boost converter model predictive control method and system based on a disturbance observer, which can uniformly estimate lumped disturbances and improve the voltage tracking control performance of the Boost converter predictive model under complex disturbances.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] In a first aspect, the present invention provides a Boost converter model predictive control method based on a disturbance observer, the model predictive control method comprising:

[0007] S1. Construct a nonlinear disturbance observer (NDO) to estimate the lumped disturbances experienced by the Boost converter system model in real time, thereby obtaining input voltage and load resistance estimates.

[0008] S2. Construct a disturbance compensator to compensate for disturbances in the Boost converter system model using the input voltage estimate and load resistance estimate.

[0009] S3. Construct a Model Predictive Controller (MPC). The MPC constructs a state prediction model based on the Boost converter system model after disturbance compensation. Based on the state prediction model, a finite-time optimization problem is constructed. The finite-time optimization problem is solved to generate the optimal control quantity. The optimal control quantity is applied to the PWM modulator of the Boost converter system to achieve tracking control.

[0010] In S1, the expression for the nonlinear perturbation observer NDO is:

[0011] ;

[0012] ;

[0013] In the above formula, , , , They are respectively , , , The estimated value, , These are the inductor current and the output voltage, respectively. , These are the input voltage disturbance estimate and the load resistance disturbance estimate, respectively. The lumped disturbance includes the input voltage disturbance estimate and the load resistance disturbance estimate. For energy storage inductors; This is the output filter capacitor; The duty cycle of the controlled switch; , These are the input voltage and the load resistance, respectively. , These are the nominal values ​​of the input voltage and the load resistance, respectively. , , , All of these are observer gain parameters;

[0014] The nonlinear disturbance observer (NDO) observes the error in the inductor current. and output voltage error Obtain the input voltage disturbance estimate and load resistance disturbance estimate Based on the input voltage disturbance estimate and load resistance disturbance estimate Perturbation compensation is performed on the input voltage and load resistance of the Boost converter to obtain the input voltage estimate. and estimated load resistance The disturbance compensation formula is as follows:

[0015] ;

[0016] The expression for the Boost converter system model after disturbance compensation in S3 is as follows:

[0017] ;

[0018] In the above formula, , These are the estimated input voltage and the estimated load resistance, respectively. For energy storage inductors; This is the output filter capacitor; , These are the inductor current and the output voltage, respectively. The duty cycle of the controlled switch;

[0019] The construction of the state prediction model based on the Boost converter system model after disturbance compensation includes the following steps:

[0020] Define system state ,in , Define the control quantities as inductor current and output voltage, respectively. The Boost converter system model after disturbance compensation is then converted into the following control function:

[0021] ;

[0022] In the above formula, System status The derivative;

[0023] Based on the control function, the following state-space equations are established:

[0024] ;

[0025] ; ; ;

[0026] In the above formula, for The system state at any given moment; for The derivative of the system state at time t. for Duty cycle at any given time; This refers to the output voltage. , , These are the state matrix, input matrix, and output matrix in the state-space equation, respectively.

[0027] System state deviation is constructed based on real-time sampled inductor current and output voltage. ,in , These are the inductor current deviation and the output voltage deviation, respectively. The duty cycle deviation is constructed based on the real-time sampled duty cycle. Based on system state deviation and duty cycle deviation The state-space equations are transformed into the following discrete deviation system equations:

[0028] ;

[0029] ; ;

[0030] In the above formula, , For respectively time, The system state deviation at time t, ; The time is the real-time sampling time; for Duty cycle deviation at any given time; The equivalent disturbance term caused by the disturbance estimation error; for The output of the discrete-time deviation system equations ; The sampling period; , These are the state matrix and input matrix in the discrete deviation system equations, respectively.

[0031] Define the prediction time domain and control time domain The discrete deviation system equations are then transformed into the following state prediction model:

[0032] ;

[0033] ; ; ;

[0034] ; ; ;

[0035] In the above formula, Indicates in Time is crucial for controlling the time domain The sequence of control quantities within, In order to be in Time is crucial for controlling the time domain Inner Control variables for time steps In order to be in Time for prediction time domain Interference sequences within; for Time for prediction time domain The system state deviation sequence within, For state prediction models in Time for prediction time domain The output voltage deviation sequence within, In order to be in Time is crucial for controlling the time domain Inner Output voltage deviation at time step; The system state matrix of the state prediction model; This is the input matrix for the state prediction model; The process disturbance input matrix is ​​used for the state prediction model; This is the output matrix of the state prediction model; This is the direct transfer matrix of the state prediction model; This is the measurement noise input matrix for the state prediction model.

[0036] In S3, the steps for constructing the finite-time optimization problem include:

[0037] Defined in Time for prediction time domain Reference trajectory within Define the output voltage reference error , For reference trajectory Inner One element, This is a reference value for the output voltage;

[0038] Defined in Time is crucial for controlling the time domain Incremental sequence of control quantities within Among them Time is crucial for controlling the time domain Inner Increment of control quantity at time step , , They are respectively in Time is crucial for controlling the time domain Inner Time step, number Control parameters for time steps;

[0039] The following finite-time optimization problem is derived:

[0040] ;

[0041] In the above formula, , They are respectively with Item and The weighted matrix corresponding to each item.

[0042] The finite-time optimization problem is transformed into a standard quadratic programming (QP) form and solved according to the following steps:

[0043] First, let in Time is crucial for controlling the time domain Inner Control quantity of time step ,in , for The control quantity, In order to be in Time is crucial for controlling the time domain Inner Increment of control quantity at time step For steady-state duty cycle, it will be Time is crucial for controlling the time domain Internal control sequence Rewrite it in the following form: , The control quantity for the previous moment, , It is a lower triangular identity matrix;

[0044] Will Substituting into the following equation of the state prediction equation This yields the first transformation form of the state prediction equation: ;

[0045] Subsequently, the output free response equation is constructed. Based on the output free response equation, the first transformation form of the state prediction equation is transformed into the second transformation form: intermediate matrix ;

[0046] Then, construct the voltage tracking vector. The deviation between the output of the free response equation and the reference trajectory is obtained. The deviation between the output and the reference trajectory based on the free response equation The finite-time optimization problem is transformed into the following standard quadratic programming (QP) form:

[0047] ;

[0048] In the above formula, , All are intermediate quantities. , ;

[0049] Finally, solve the standard quadratic programming QP form to generate... Optimal control increment sequence at time 1 Based on the incremental sequence of optimal control quantity The first control increment Calculated Optimal control quantity at time 1 The obtained optimal control quantity A PWM modulator is applied to the Boost converter system to make the output voltage of the Boost converter system track the voltage reference value.

[0050] Secondly, the present invention provides a Boost converter model predictive control system based on a disturbance observer, the model predictive control system comprising:

[0051] The nonlinear disturbance observation module is used to construct a nonlinear disturbance observer (NDO). The NDO is used to estimate the lumped disturbances experienced by the Boost converter system model in real time, so as to obtain the input voltage estimate and the load resistance estimate.

[0052] The disturbance compensation module is used to construct a disturbance compensator, which uses the input voltage estimate and load resistance estimate to perform disturbance compensation on the Boost converter system model.

[0053] The Model Predictive Control module is used to construct a Model Predictive Controller (MPC). The MPC constructs a state prediction model based on the disturbance-compensated Boost converter system model, constructs a finite-time optimization problem based on the state prediction model, generates the optimal control quantity by solving the finite-time optimization problem, and applies the optimal control quantity to the PWM modulator of the Boost converter system to achieve tracking control.

[0054] The expression for the nonlinear perturbation observer (NDO) is:

[0055] ;

[0056] ;

[0057] In the above formula, , , , They are respectively , , , The estimated value, , These are the inductor current and the output voltage, respectively. , These are the input voltage disturbance estimate and the load resistance disturbance estimate, respectively. The lumped disturbance includes the input voltage disturbance estimate and the load resistance disturbance estimate. For energy storage inductors; This is the output filter capacitor; The duty cycle of the controlled switch; , These are the input voltage and the load resistance, respectively. , These are the nominal values ​​of the input voltage and the load resistance, respectively. , , , All of these are observer gain parameters;

[0058] The nonlinear disturbance observer (NDO) is used to observe the inductor current error. and output voltage error Obtain the input voltage disturbance estimate and load resistance disturbance estimate Based on the input voltage disturbance estimate and load resistance disturbance estimate Perturbation compensation is performed on the input voltage and load resistance of the Boost converter to obtain the input voltage estimate. and estimated load resistance The disturbance compensation formula is as follows:

[0059] ;

[0060] The expression for the Boost converter system model after disturbance compensation is as follows:

[0061] ;

[0062] In the above formula, , These are the estimated input voltage and the estimated load resistance, respectively. For energy storage inductors; This is the output filter capacitor; , These are the inductor current and the output voltage, respectively. The duty cycle of the controlled switch;

[0063] The Model Prediction Controller (MPC) is used to construct a state prediction model according to the following steps:

[0064] Define system state ,in , Define the control quantities as inductor current and output voltage, respectively. The Boost converter system model after disturbance compensation is then converted into the following control function:

[0065] ;

[0066] In the above formula, System status The derivative;

[0067] Based on the control function, the following state-space equations are established:

[0068] ;

[0069] ; ; ;

[0070] In the above formula, for The system state at any given moment; for The derivative of the system state at time t. for Duty cycle at any given time; This refers to the output voltage. , , These are the state matrix, input matrix, and output matrix in the state-space equation, respectively.

[0071] System state deviation is constructed based on real-time sampled inductor current and output voltage. ,in , These are the inductor current deviation and the output voltage deviation, respectively. The duty cycle deviation is constructed based on the real-time sampled duty cycle. Based on system state deviation and duty cycle deviation The state-space equations are transformed into the following discrete deviation system equations:

[0072] ;

[0073] ; ;

[0074] In the above formula, , For respectively time, The system state deviation at time t, ; The time is the real-time sampling time; for Duty cycle deviation at any given time; The equivalent disturbance term caused by the disturbance estimation error; for The output of the discrete-time deviation system equations ; The sampling period; , These are the state matrix and input matrix in the discrete deviation system equations, respectively.

[0075] Define the prediction time domain and control time domain The discrete deviation system equations are then transformed into the following state prediction model:

[0076] ;

[0077] ; ; ;

[0078] ; ; ;

[0079] In the above formula, Indicates in Time is crucial for controlling the time domain The sequence of control quantities within, In order to be in Time is crucial for controlling the time domain Inner Control variables for time steps In order to be in Time for prediction time domain Interference sequences within; for Time for prediction time domain The system state deviation sequence within, For state prediction models in Time for prediction time domain The output voltage deviation sequence within, In order to be in Time is crucial for controlling the time domain Inner Output voltage deviation at time step; The system state matrix of the state prediction model; This is the input matrix for the state prediction model; The process disturbance input matrix is ​​used for the state prediction model; This is the output matrix of the state prediction model; This is the direct transfer matrix of the state prediction model; This is the measurement noise input matrix for the state prediction model.

[0080] The Model Predictive Controller (MPC) is used to construct a finite-time optimization problem according to the following steps:

[0081] Defined in Time for prediction time domain Reference trajectory within Define the output voltage reference error , For reference trajectory Inner One element, This is a reference value for the output voltage;

[0082] Defined in Time is crucial for controlling the time domain Incremental sequence of control quantities within Among them Time is crucial for controlling the time domain Inner Increment of control quantity at time step , , They are respectively in Time is crucial for controlling the time domain Inner Time step, number Control variables for time steps;

[0083] The following finite-time optimization problem is derived:

[0084] ;

[0085] In the above formula, , They are respectively with Item and The weighted matrix corresponding to each item.

[0086] The Model Predictive Controller (MPC) is used to transform the finite-time optimization problem into a standard quadratic programming (QP) form for solution according to the following steps:

[0087] First, let in Time is crucial for controlling the time domain Inner Control quantity of time step ,in , for The control quantity, In order to be in Time is crucial for controlling the time domain Inner Increment of control quantity at time step For steady-state duty cycle, it will be Time is crucial for controlling the time domain Internal control sequence Rewrite it in the following form: , The control quantity for the previous moment, , It is a lower triangular identity matrix;

[0088] Will Substituting into the following equation of the state prediction equation This yields the first transformation form of the state prediction equation: ;

[0089] Subsequently, the output free response equation is constructed. Based on the output free response equation, the first transformation form of the state prediction equation is transformed into the second transformation form: intermediate matrix ;

[0090] Then, construct the voltage tracking vector. The deviation between the output of the free response equation and the reference trajectory is obtained. The deviation between the output and the reference trajectory based on the free response equation The finite-time optimization problem is transformed into the following standard quadratic programming (QP) form:

[0091] ;

[0092] In the above formula, , All are intermediate quantities. , ;

[0093] Finally, solve the standard quadratic programming QP form to generate Optimal control increment sequence at time 1 Based on the incremental sequence of optimal control quantity The first control increment Calculated Optimal control quantity at time 1 The obtained optimal control quantity A PWM modulator is applied to the Boost converter system to make the output voltage of the Boost converter system track the voltage reference value.

[0094] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0095] 1. The method described in this invention models load mutations and input voltage fluctuations as lumped disturbances. Based on the extended state observer theory, a nonlinear disturbance observer (NDO) for the Boost converter is constructed for real-time estimation of the lumped disturbance. The disturbance estimate is then used to perform online dynamic compensation for the state prediction model. Compared with existing technologies that separate the disturbance, this significantly improves the matching accuracy between the prediction model and the actual system, effectively suppresses prediction bias and steady-state error caused by model mismatch, and enables the Boost converter to achieve fast and accurate tracking of the output voltage through rolling optimization under complex disturbances, thus balancing dynamic response speed and steady-state control accuracy.

[0096] 2. The method described in this invention designs a finite-time optimization problem based on a compensated state prediction model, and transforms the finite-time optimization problem into a standard quadratic programming problem for online rolling solution. The standard quadratic programming form can explicitly handle physical constraints such as duty cycle, inductor current, and control increment. The online rolling solution can resolve the optimal control sequence in the future finite steps based on the current state in each sampling period, which meets the real-time control requirements of the Boost converter under high switching frequency. This allows the system to quickly obtain the deterministic optimal duty cycle command when the load changes suddenly and the input voltage fluctuates, taking into account both dynamic response speed and control accuracy.

[0097] 3. The method described in this invention ensures rapid recovery of the output voltage and minimal overshoot, while maintaining the system output within a bounded range near the reference value even when estimation errors exist, demonstrating good dynamic performance. Attached Figure Description

[0098] Figure 1 This is a flowchart of the method described in this invention.

[0099] Figure 2 The simulation results show the response waveforms of the inductor current and output voltage when the reference voltage is stepped down from 24V to 20V.

[0100] Figure 3 The simulation results show the inductor current and output voltage response waveforms when the load resistance changes abruptly from 45Ω to 95Ω.

[0101] Figure 4 The simulation results show the inductor current and output voltage response waveforms when the load resistance changes abruptly from 95Ω to 45Ω.

[0102] Figure 5 The simulation results show the inductor current and output voltage response waveforms when the input voltage changes abruptly from 12V to 15V.

[0103] Figure 6 The simulation results show the inductor current and output voltage response waveforms when the input voltage changes abruptly from 12V to 15V.

[0104] Figure 7 To add an amplitude of 2V and an angular frequency of [value] to the nominal input voltage. Simulation results of the output voltage response waveform under sinusoidal interference signal conditions.

[0105] Figure 8 This is a structural block diagram of the system described in this invention. Detailed Implementation

[0106] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings.

[0107] Example 1:

[0108] See Figure 1 A model predictive control method for a Boost converter based on a disturbance observer is proposed, which proceeds in the following steps:

[0109] S1. Construct a nonlinear disturbance observer (NDO) to estimate the lumped disturbances experienced by the Boost converter system model in real time, thereby obtaining input voltage and load resistance estimates.

[0110] Specifically, the following state equations are established for the Boost converter system model:

[0111] ;

[0112] In the above formula, , These are the inductor current and the output voltage, respectively. Input voltage; For energy storage inductors; This is the output filter capacitor; The duty cycle of the controlled switch; , These are the input voltage and the load resistance, respectively.

[0113] Due to the input voltage in practical applications and load resistance It will be affected by external uncertainties, and take , These are the nominal values ​​of the input voltage and the load resistance, respectively. The disturbance caused by the change in load resistance is... The disturbance caused by input voltage fluctuation is Then we can obtain the following perturbation model:

[0114] ;

[0115] Substituting the disturbance model into the state equation, we define the lumped disturbance. The state equation under the disturbance is obtained as follows:

[0116] ; .

[0117] Specifically, the aggregated disturbance The vector is considered as the extended state of the system, and its rate of change is bounded. The expression for the nonlinear perturbation observer (NDO) is designed as follows:

[0118] ;

[0119] In the above formula, , , , They are respectively , , , The estimated value, , These are the inductor current and the output voltage, respectively. , These are the input voltage disturbance estimate and the load resistance disturbance estimate, respectively. The lumped disturbance includes the input voltage disturbance estimate and the load resistance disturbance estimate. For energy storage inductors; This is the output filter capacitor; The duty cycle of the controlled switch; , These are the input voltage and the load resistance, respectively. , , , All of these are observer gain parameters.

[0120] The nonlinear disturbance observer (NDO) observes the error in the inductor current. and output voltage error Obtain the input voltage disturbance estimate and load resistance disturbance estimate .

[0121] S2. Construct a disturbance compensator to compensate for disturbances in the Boost converter system model using the input voltage estimate and load resistance estimate.

[0122] Specifically, the disturbance compensation module is based on the input voltage disturbance estimate. and load resistance disturbance estimate Perturbation compensation is performed on the input voltage and load resistance of the Boost converter to obtain the input voltage estimate. and estimated load resistance The disturbance compensation formula is as follows:

[0123] ;

[0124] Using input voltage estimates and estimated load resistance Compensating the state equations of the Boost converter system model, the state equation expression of the Boost converter system model after disturbance compensation is as follows:

[0125] ;

[0126] S3. Construct a Model Predictive Controller (MPC). The MPC constructs a state prediction model based on the Boost converter system model after disturbance compensation. Based on the state prediction model, a finite-time optimization problem is constructed. The optimal control quantity is generated by solving the finite-time optimization problem. The optimal control quantity is applied to the PWM modulator of the Boost converter system to achieve tracking control.

[0127] The construction of the state prediction model based on the Boost converter system model after disturbance compensation includes the following steps:

[0128] Define system state ,in , Define the control quantities as inductor current and output voltage, respectively. The Boost converter system model after disturbance compensation is then converted into the following control function:

[0129] ;

[0130] In the above formula, System status The derivative;

[0131] Based on the control function, the following state-space equations are established:

[0132] ;

[0133] ; ; ;

[0134] In the above formula, for The system state at any given moment; for The derivative of the system state at time t. for Duty cycle at any given time; This refers to the output voltage. , , These are the state matrix, input matrix, and output matrix in the state-space equation, respectively.

[0135] System state deviation is constructed based on real-time sampled inductor current and output voltage. ,in , These are the inductor current deviation and the output voltage deviation, respectively. The duty cycle deviation is constructed based on the real-time sampled duty cycle. Based on system state deviation and duty cycle deviation The state-space equations are transformed into the following discrete deviation system equations:

[0136] ;

[0137] ; ;

[0138] In the above formula, , For respectively time, The system state deviation at time t, ; The time is the real-time sampling time; for Duty cycle deviation at any given time; The equivalent disturbance term caused by the disturbance estimation error is assumed in this embodiment to be bounded, and the disturbance at all future times is equal to the disturbance at the current time. for The output of the discrete-time deviation system equations is the system output voltage deviation. ; The sampling period; , These are the state matrix and input matrix in the discrete deviation system equations, respectively.

[0139] Define the prediction time domain and control time domain The discrete deviation system equations are then transformed into the following state prediction model:

[0140] ;

[0141] ; ; ;

[0142] ; ; ;

[0143] In the above formula, Indicates in Time is crucial for controlling the time domain The sequence of control quantities within, In order to be in Time is crucial for controlling the time domain Inner Control variables for time steps In order to be in Time for prediction time domain Interference sequences within, for Time for prediction time domain The system state deviation sequence within, For state prediction models in Time for prediction time domain The output voltage deviation sequence within, In order to be in Time is crucial for controlling the time domain Inner Output voltage deviation at time step; The system state matrix of the state prediction model; This is the input matrix for the state prediction model; The process disturbance input matrix is ​​used for the state prediction model; This is the output matrix of the state prediction model; This is the direct transfer matrix of the state prediction model; This is the measurement noise input matrix for the state prediction model.

[0144] Since the nonlinear disturbance observer (NDO) can only achieve bounded estimation of lumped disturbances, the state prediction model still contains an equivalent disturbance term caused by the disturbance estimation error. Therefore, the model predictive controller needs to construct a finite-time optimization problem and continuously correct the control quantity within the finite-time domain through rolling optimization to suppress the impact of the disturbance estimation error on the prediction performance, thereby improving the robustness of the system. The steps for constructing the finite-time optimization problem include:

[0145] Defined in Time for prediction time domain Reference trajectory of internal output voltage Define the output voltage reference error , For reference trajectory Inner One element, This is a reference value for the output voltage, which is the nominal value designed to withstand any interference or error.

[0146] Defined in Time is crucial for controlling the time domain Incremental sequence of control quantities within Among them Time is crucial for controlling the time domain Inner Increment of control quantity at time step , , They are respectively in Time is crucial for controlling the time domain Inner Time step, first Control parameters for time steps;

[0147] To make the output voltage deviation Tracking output voltage reference error With smooth control, the following finite-time optimization problem is constructed:

[0148] ;

[0149] In the above formula, , They are respectively with Item and The weighted matrix corresponding to each term;

[0150] The finite-time optimization problem is transformed into a standard quadratic programming (QP) form and solved according to the following steps:

[0151] First, let in Time is crucial for controlling the time domain Inner Control quantity of time step ,in , for The control quantity, In order to be in Time is crucial for controlling the time domain Inner Increment of control quantity at time step For steady-state duty cycle, it will be Time is crucial for controlling the time domain Internal control sequence Rewrite it in the following form: , The control quantity for the previous moment, , It is a lower triangular identity matrix. ;

[0152] Will Substituting into the following equation of the state prediction equation This yields the first transformation form of the state prediction equation: ;

[0153] Subsequently, the output free response equation is constructed. Based on the output free response equation, the first transformation form of the state prediction equation is transformed into the second transformation form: intermediate matrix Due to disturbance Since the predicted time is unavailable, this embodiment does not explicitly consider this term when solving the optimization problem, but treats it as an equivalent bounded perturbation for subsequent closed-loop stability analysis.

[0154] Then, construct the voltage tracking vector. The deviation between the output of the free response equation and the reference trajectory is obtained. The deviation between the output and the reference trajectory based on the free response equation The finite-time optimization problem is transformed into the following standard quadratic programming (QP) form, ignoring constant terms irrelevant to the optimization variables during the transformation:

[0155] ;

[0156] In the above formula, , All are intermediate quantities. , ;

[0157] Finally, a commercial solver (such as MATLAB) can be used to solve the above standard quadratic programming QP form to generate... Optimal control increment sequence at time 1 Based on the incremental sequence of optimal control quantity The first control increment Calculated Optimal control quantity at time 1 The obtained optimal control quantity A PWM modulator is applied to the Boost converter system to make the output voltage of the Boost converter system track the voltage reference value.

[0158] To ensure physical feasibility and safe operation, constraints need to be imposed on control variables and system states. Specifically, this includes control increment constraints. Duty cycle constraint Inductor current constraint ,in A column vector consisting entirely of 1s, with free response. .

[0159] The above constraints are finally transformed into the following form:

[0160] ;

[0161] ; .

[0162] Since the model parameters are updated and compensated in real time through a disturbance observer, the state prediction model changes with time at each sampling moment; therefore, this model predictive control belongs to time-varying model predictive control. Under the condition that the disturbance estimation error is bounded, it can be modeled as a system driven by a bounded disturbance. This invention performs a stability analysis on the system model. Based on discrete Lyapunov stability theory, the system's stability with respect to disturbances can be obtained. It is the input that makes the state stable.

[0163] Performance verification:

[0164] To verify the effectiveness of the proposed method (MPC), its performance was compared with that of traditional sliding mode control (SMC). A system was built in the Simulink Specialized Power Systems environment as follows: Figure 1 The simulation model of the Boost converter is shown below. The system parameters of the Boost converter are shown in Table 1. Controller parameters are selected as follows: , , , , , , Select the observer gain parameter: , , , .

[0165] Table 1 System parameters of the Boost converter

[0166] ;

[0167] The simulation test consisted of four sets of operating conditions, which were used to verify the robustness and tracking performance of the method described in this invention under reference voltage step, load resistance sudden change, input voltage step, and input voltage superimposed high-frequency disturbance. The specific four simulation conditions and their results are analyzed below:

[0168] (1) The initial reference voltage of the system is 24V, which decreases to 20V in a stepwise manner over 0.25s. The simulation results of the inductor current and output voltage response waveforms under this condition are as follows: Figure 2 As shown in the figure, the steady-state error of the output voltage proposed in this invention is approximately 0.16V, which is better than the 0.29V of traditional sliding diaphragm control. At the same time, the inductor current responds to changes in the reference voltage much faster, the output can follow the changes in the reference voltage very well, and the adjustment time is significantly shorter than that of traditional sliding diaphragm control.

[0169] (2) Consider the load resistance changing abruptly from 45Ω to 95Ω and then back to 45Ω to test the system's robustness to load resistance disturbances. The simulation results of the inductor current and output voltage response waveforms when the load resistance changes abruptly from 45Ω to 95Ω are as follows: Figure 3 As shown, Figure 3 This indicates that the inductor current response speed of the proposed method is faster when the load resistance changes. Regarding voltage output, the traditional sliding diaphragm control has a maximum overshoot of approximately 0.73V, a settling time of approximately 4.4ms, and a steady-state error of approximately 0.31V, while the proposed method has a maximum overshoot of only 0.38V, a settling time of 4.3ms, and a steady-state error of approximately 0.17V. The simulation results of the inductor current and output voltage response waveforms when the load resistance changes abruptly from 95Ω to 45Ω are as follows: Figure 4 As shown, Figure 4 This also demonstrates that the method proposed in this invention outperforms traditional sliding diaphragm control in terms of inductor current response speed and output voltage performance when faced with load resistance disturbances.

[0170] (3) Consider the operating conditions where the input voltage changes abruptly from 12V to 15V and back to 12V to test the system's robustness to input voltage disturbances. The simulation results of the inductor current and output voltage response waveforms when the input voltage changes abruptly from 12V to 15V are as follows: Figure 5 As shown, Figure 5This indicates that both the proposed method and traditional sliding diaphragm control can rapidly respond to changes in input voltage. However, in terms of output voltage, the traditional sliding diaphragm control has a maximum overshoot of approximately 1.27V, a settling time of approximately 3.8ms, and a steady-state error of approximately 0.27V. In contrast, the proposed method has a maximum overshoot of only approximately 0.42V, a settling time of approximately 4.2ms, and a steady-state error of 0.15V. Simulation results of the inductor current and output voltage response waveforms when the input voltage changes abruptly from 12V to 15V are shown below. Figure 6 As shown, Figure 6 The results show that, since the nominal input voltage is 12V, both methods exhibit significant overshoot during the startup phase when the initial input voltage is set to 15V. However, the traditional sliding plate control requires approximately 20ms of adjustment and recovery time. Similar conclusions are drawn when the input voltage experiences a sudden change at 0.25s. All these results demonstrate that the method proposed in this invention outperforms traditional sliding plate control in terms of inductor current response speed and output voltage performance when facing input voltage disturbances.

[0171] (4) Consider the input voltage parameter perturbation, that is, add an amplitude of 2V and an angular frequency of to the nominal input voltage. The sinusoidal interference signal was used. The simulation results of the output voltage response waveform under this operating condition are as follows: Figure 7 As shown in the figure, the nominal input voltage is affected by high-frequency disturbances at 0.25s. At this time, the output voltage fluctuation amplitude of the method proposed in this invention increases from the original 0.065V to 0.32V, and the average steady-state error is approximately 0.15V. The output voltage fluctuation amplitude of the traditional sliding diaphragm control increases from the original 0.13V to 0.55V, and the average steady-state error is approximately 0.4V. The above results show that the method proposed in this invention has a stronger ability to suppress high-frequency disturbances in the input voltage.

[0172] Example 2:

[0173] See Figure 8 A model predictive control system for a Boost converter based on a disturbance observer includes a nonlinear disturbance observation module, a disturbance compensation module, and a model predictive control module. The nonlinear disturbance observation module constructs a nonlinear disturbance observer (NDO) to estimate the lumped disturbance experienced by the Boost converter system model in real time, thereby obtaining input voltage and load resistance estimates. Specifically, the nonlinear disturbance observation module is used to observe inductor current errors through the NDO. and output voltage error To obtain the input voltage disturbance estimate and load resistance disturbance estimate The expression for the nonlinear perturbation observer (NDO) is:

[0174] ;

[0175] ;

[0176] In the above formula, , , , They are respectively , , , The estimated value, , These are the inductor current and the output voltage, respectively. , These are the input voltage disturbance estimate and the load resistance disturbance estimate, respectively. The lumped disturbance includes the input voltage disturbance estimate and the load resistance disturbance estimate. For energy storage inductors; This is the output filter capacitor; The duty cycle of the controlled switch; , These are the input voltage and the load resistance, respectively. , These are the nominal values ​​of the input voltage and the load resistance, respectively. , , , All are observer gain parameters;

[0177] The disturbance compensation module is used to construct a disturbance compensator, which performs disturbance compensation on the Boost converter system model using the input voltage estimate and the load resistance estimate. Specifically, the disturbance compensator is based on the input voltage disturbance estimate. and load resistance disturbance estimate Perturbation compensation is performed on the input voltage and load resistance of the Boost converter to obtain the input voltage estimate. and estimated load resistance The disturbance compensation formula is as follows:

[0178] ;

[0179] The expression for the Boost converter system model after disturbance compensation is as follows:

[0180] ;

[0181] In the above formula, , These are the estimated input voltage and the estimated load resistance, respectively. For energy storage inductors; This is the output filter capacitor; , These are the inductor current and the output voltage, respectively. The duty cycle of the controlled switch;

[0182] The model predictive control module is used to construct a model predictive controller (MPC). The MPC constructs a state prediction model based on the perturbation-compensated Boost converter system model. Based on the state prediction model, a finite-time optimization problem is constructed. Solving the finite-time optimization problem generates the optimal control quantity, which is then applied to the PWM modulator of the Boost converter system to achieve tracking control. Specifically, the MPC constructs the state prediction model according to the following steps:

[0183] Define system state ,in , Define the control quantities as inductor current and output voltage, respectively. The Boost converter system model after disturbance compensation is then converted into the following control function:

[0184] ;

[0185] In the above formula, System status The derivative;

[0186] Based on the control function, the following state-space equations are established:

[0187] ;

[0188] ; ; ;

[0189] In the above formula, for The system state at any given moment; for The derivative of the system state at time t. for Duty cycle at any given time; This refers to the output voltage. , , These are the state matrix, input matrix, and output matrix in the state-space equation, respectively.

[0190] System state deviation is constructed based on real-time sampled inductor current and output voltage. ,in , These are the inductor current deviation and the output voltage deviation, respectively. The duty cycle deviation is constructed based on the real-time sampled duty cycle. Based on system state deviation and duty cycle deviation The state-space equations are transformed into the following discrete deviation system equations:

[0191] ;

[0192] ; ;

[0193] In the above formula, , For respectively time, The system state deviation at time t, ; The time is the real-time sampling time; for Duty cycle deviation at any given time; The equivalent disturbance term caused by the disturbance estimation error; for The output of the discrete-time deviation system equations ; The sampling period; , These are the state matrix and input matrix in the discrete deviation system equations, respectively.

[0194] Define the prediction time domain and control time domain The discrete deviation system equations are then transformed into the following state prediction model:

[0195] ;

[0196] ; ; ;

[0197] ; ; ;

[0198] In the above formula, Indicates in Time is crucial for controlling the time domain The sequence of control quantities within, In order to be in Time is crucial for controlling the time domain Inner Control variables for time steps In order to be in Time for prediction time domain Interference sequences within; for Time for prediction time domain The system state deviation sequence within, For state prediction models in Time for prediction time domain The output voltage deviation sequence within, In order to be in Time is crucial for controlling the time domain Inner Output voltage deviation at time step; The system state matrix of the state prediction model; This is the input matrix for the state prediction model; The process disturbance input matrix is ​​used for the state prediction model; This is the output matrix of the state prediction model; This is the direct transfer matrix of the state prediction model; The measurement noise input matrix is ​​used for the state prediction model;

[0199] The Model Predictive Controller (MPC) is used to construct a finite-time optimization problem according to the following steps:

[0200] Defined in Time for prediction time domain Reference trajectory within Define the output voltage reference error , For reference trajectory Inner One element, This is a reference value for the output voltage;

[0201] Defined in Time is crucial for controlling the time domain Incremental sequence of control quantities within Among them Time is crucial for controlling the time domain Inner Increment of control quantity at time step , , They are respectively in Time is crucial for controlling the time domain Inner Time step, first Control parameters for time steps;

[0202] The following finite-time optimization problem is derived:

[0203] ;

[0204] In the above formula, , They are respectively with Item and The weighted matrix corresponding to each item.

[0205] The Model Predictive Controller (MPC) is used to transform the finite-time optimization problem into a standard quadratic programming (QP) form for solution according to the following steps:

[0206] First, let in Time is crucial for controlling the time domain Inner Control quantity of time step ,in , for The control quantity, In order to be in Time is crucial for controlling the time domain Inner Increment of control quantity at time step For steady-state duty cycle, it will be Time is crucial for controlling the time domain Internal control sequence Rewrite it in the following form: , For the previous moment Timing control quantity sequence , It is a lower triangular identity matrix;

[0207] Will Substituting into the following equation of the state prediction equation This yields the first transformation form of the state prediction equation: ;

[0208] Subsequently, the output free response equation is constructed. Based on the output free response equation, the first transformation form of the state prediction equation is transformed into the second transformation form: intermediate matrix ;

[0209] Then, construct the voltage tracking vector. The deviation between the output of the free response equation and the reference trajectory is obtained. The deviation between the output and the reference trajectory based on the free response equation The finite-time optimization problem is transformed into the following standard quadratic programming (QP) form:

[0210] ;

[0211] In the above formula, , All are intermediate quantities. , ;

[0212] Finally, solve the standard quadratic programming QP form to generate... Optimal control increment sequence at time 1 Based on the incremental sequence of optimal control quantity The first control increment Calculated Optimal control quantity at time 1 The obtained optimal control quantity A PWM modulator is applied to the Boost converter system to make the output voltage of the Boost converter system track the voltage reference value.

[0213] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program goods. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0214] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0215] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0216] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0217] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A Boost converter model predictive control method based on a disturbance observer, characterized in that: The model predictive control method includes: S1. Construct a nonlinear disturbance observer (NDO) to estimate the lumped disturbances experienced by the Boost converter system model in real time, thereby obtaining input voltage and load resistance estimates. S2. Construct a disturbance compensator to compensate for disturbances in the Boost converter system model using the input voltage estimate and load resistance estimate. S3. Construct a Model Predictive Controller (MPC). The MPC constructs a state prediction model based on the Boost converter system model after disturbance compensation. Based on the state prediction model, a finite-time optimization problem is constructed. The finite-time optimization problem is solved to generate the optimal control quantity. The optimal control quantity is applied to the PWM modulator of the Boost converter system to achieve tracking control.

2. The Boost converter model predictive control method based on a disturbance observer according to claim 1, characterized in that: In S1, the expression for the nonlinear perturbation observer NDO is: ； ； In the above formula, , , , They are respectively , , , The estimated value, , These are the inductor current and the output voltage, respectively. , These are the input voltage disturbance estimate and the load resistance disturbance estimate, respectively. The lumped disturbance includes the input voltage disturbance estimate and the load resistance disturbance estimate. For energy storage inductors; This is the output filter capacitor; The duty cycle of the controlled switch; , These are the input voltage and the load resistance, respectively. , These are the nominal values ​​of the input voltage and the load resistance, respectively. , , , All are observer gain parameters; The nonlinear disturbance observer (NDO) observes the error in the inductor current. and output voltage error Obtain the input voltage disturbance estimate and load resistance disturbance estimate Based on the input voltage disturbance estimate and load resistance disturbance estimate Perturbation compensation is performed on the input voltage and load resistance of the Boost converter to obtain the input voltage estimate. and estimated load resistance The disturbance compensation formula is as follows: 。 3. The Boost converter model predictive control method based on a disturbance observer according to claim 1 or 2, characterized in that: The expression for the Boost converter system model after disturbance compensation in S3 is as follows: ； In the above formula, , These are the estimated input voltage and the estimated load resistance, respectively. For energy storage inductors; This is the output filter capacitor; , These are the inductor current and the output voltage, respectively. The duty cycle of the controlled switch; The construction of the state prediction model based on the Boost converter system model after disturbance compensation includes the following steps: Define system state ,in , Define the control quantities as inductor current and output voltage, respectively. The Boost converter system model after disturbance compensation is then converted into the following control function: ； In the above formula, System status The derivative; Based on the control function, the following state-space equations are established: ； ； ； ； In the above formula, for The system state at any given moment; for The derivative of the system state at time t. for Duty cycle at any given time; This refers to the output voltage. , , These are the state matrix, input matrix, and output matrix in the state-space equation, respectively. System state deviation is constructed based on real-time sampled inductor current and output voltage. ,in , These are the inductor current deviation and the output voltage deviation, respectively. The duty cycle deviation is constructed based on the real-time sampled duty cycle. Based on system state deviation and duty cycle deviation The state-space equations are transformed into the following discrete deviation system equations: ； ； ； In the above formula, , For respectively time, The system state deviation at time t, ; The time is the real-time sampling time; for Duty cycle deviation at any given time; The equivalent disturbance term caused by the disturbance estimation error; for The output of the discrete-time deviation system equations ; The sampling period; , These are the state matrix and input matrix in the discrete deviation system equations, respectively. Define the prediction time domain and control time domain The discrete deviation system equations are then transformed into the following state prediction model: ； ； ； ； ； ； ； In the above formula, Indicates in Time is crucial for controlling the time domain The sequence of control quantities within, In order to be in Time is crucial for controlling the time domain Inner Control variables for time steps In order to be in Time for prediction time domain Interference sequences within; for Time for prediction time domain The system state deviation sequence within, For state prediction models in Time for prediction time domain The output voltage deviation sequence within, In order to be in Time is crucial for controlling the time domain Inner Output voltage deviation at time step; The system state matrix of the state prediction model; This is the input matrix for the state prediction model; The process disturbance input matrix is ​​used for the state prediction model; This is the output matrix of the state prediction model; This is the direct transfer matrix of the state prediction model; This is the measurement noise input matrix for the state prediction model.

4. The Boost converter model predictive control method based on a disturbance observer according to claim 3, characterized in that: In S3, the steps for constructing the finite-time optimization problem include: Defined in Time for prediction time domain Reference trajectory within Define the output voltage reference error , For reference trajectory Inner One element, This is a reference value for the output voltage; Defined in Time is crucial for controlling the time domain Incremental sequence of control quantities within Among them Time is crucial for controlling the time domain Inner Increment of control quantity at time step , , They are respectively in Time is crucial for controlling the time domain Inner Time step, first Control parameters for time steps; The following finite-time optimization problem is derived: ； In the above formula, , They are respectively with Item and The weighted matrix corresponding to each item.

5. The Boost converter model predictive control method based on a disturbance observer according to claim 4, characterized in that: The finite-time optimization problem is transformed into a standard quadratic programming (QP) form and solved according to the following steps: First, let in Time is crucial for controlling the time domain Inner Control quantity of time step ,in , for The control quantity, In order to be in Time is crucial for controlling the time domain Inner Increment of control quantity at time step For steady-state duty cycle, it will be Time is crucial for controlling the time domain Internal control sequence Rewrite it in the following form: , The control quantity for the previous moment, , It is a lower triangular identity matrix; Will Substituting into the following equation of the state prediction equation This yields the first transformation form of the state prediction equation: ; Subsequently, the output free response equation is constructed. Based on the output free response equation, the first transformation form of the state prediction equation is transformed into the second transformation form: intermediate matrix ; Then, construct the voltage tracking vector. The deviation between the output of the free response equation and the reference trajectory is obtained. The deviation between the output and the reference trajectory based on the free response equation The finite-time optimization problem is transformed into the following standard quadratic programming (QP) form: ； In the above formula, , All are intermediate quantities. , ; Finally, solve the standard quadratic programming QP form to generate... Optimal control increment sequence at time 1 Based on the incremental sequence of optimal control quantity The first control increment Calculated Optimal control quantity at time 1 The obtained optimal control quantity A PWM modulator is applied to the Boost converter system to make the output voltage of the Boost converter system track the voltage reference value.

6. A Boost converter model predictive control system based on a disturbance observer, characterized in that: The model prediction control system includes: The nonlinear disturbance observation module is used to construct a nonlinear disturbance observer (NDO). The NDO is used to estimate the lumped disturbances experienced by the Boost converter system model in real time, so as to obtain the input voltage estimate and the load resistance estimate. The disturbance compensation module is used to construct a disturbance compensator, which uses the input voltage estimate and load resistance estimate to perform disturbance compensation on the Boost converter system model. The Model Predictive Control module is used to construct a Model Predictive Controller (MPC). The MPC constructs a state prediction model based on the disturbance-compensated Boost converter system model, constructs a finite-time optimization problem based on the state prediction model, generates the optimal control quantity by solving the finite-time optimization problem, and applies the optimal control quantity to the PWM modulator of the Boost converter system to achieve tracking control.

7. The Boost converter model predictive control system based on a disturbance observer according to claim 6, characterized in that: The expression for the nonlinear perturbation observer (NDO) is: ； ； In the above formula, , , , They are respectively , , , The estimated value, , These are the inductor current and the output voltage, respectively. , These are the input voltage disturbance estimate and the load resistance disturbance estimate, respectively. The lumped disturbance includes the input voltage disturbance estimate and the load resistance disturbance estimate. For energy storage inductors; This is the output filter capacitor; The duty cycle of the controlled switch; , These are the input voltage and the load resistance, respectively. , These are the nominal values ​​of the input voltage and the load resistance, respectively. , , , All are observer gain parameters; The nonlinear disturbance observation module is used to observe the inductor current error through a nonlinear disturbance observer (NDO). and output voltage error Obtain the input voltage disturbance estimate and load resistance disturbance estimate Based on the input voltage disturbance estimate and load resistance disturbance estimate Perturbation compensation is performed on the input voltage and load resistance of the Boost converter to obtain the input voltage estimate. and estimated load resistance The disturbance compensation formula is as follows: 。 8. The Boost converter model predictive control system based on a disturbance observer according to claim 6 or 7, characterized in that: The expression for the Boost converter system model after disturbance compensation is as follows: ； In the above formula, , These are the estimated input voltage and the estimated load resistance, respectively. For energy storage inductors; This is the output filter capacitor; , These are the inductor current and the output voltage, respectively. The duty cycle of the controlled switch; The Model Prediction Controller (MPC) is used to construct a state prediction model according to the following steps: Define system state ,in , Define the control quantities as inductor current and output voltage, respectively. The Boost converter system model after disturbance compensation is then converted into the following control function: ； In the above formula, System status The derivative; Based on the control function, the following state-space equations are established: ； ； ； ； In the above formula, for The system state at any given moment; for The derivative of the system state at time t. for Duty cycle at any given time; This refers to the output voltage. , , These are the state matrix, input matrix, and output matrix in the state-space equation, respectively. System state deviation is constructed based on real-time sampled inductor current and output voltage. ,in , These are the inductor current deviation and the output voltage deviation, respectively. The duty cycle deviation is constructed based on the real-time sampled duty cycle. Based on system state deviation and duty cycle deviation The state-space equations are transformed into the following discrete deviation system equations: ； ； ； In the above formula, , For respectively time, The system state deviation at time t, ; The time is the real-time sampling time; for Duty cycle deviation at any given time; The equivalent disturbance term caused by the disturbance estimation error; for The output of the discrete-time deviation system equations ; The sampling period; , These are the state matrix and input matrix in the discrete deviation system equations, respectively. Define the prediction time domain and control time domain The discrete deviation system equations are then transformed into the following state prediction model: ； ； ； ； ； ； ； In the above formula, Indicates in Time is crucial for controlling the time domain The sequence of control quantities within, In order to be in Time is crucial for controlling the time domain Inner Control variables for time steps In order to be in Time for prediction time domain Interference sequences within; for Time for prediction time domain The system state deviation sequence within, For state prediction models in Time for prediction time domain The output voltage deviation sequence within, In order to be in Time is crucial for controlling the time domain Inner Output voltage deviation at time step; The system state matrix of the state prediction model; This is the input matrix for the state prediction model; The process disturbance input matrix is ​​used for the state prediction model; This is the output matrix of the state prediction model; This is the direct transfer matrix of the state prediction model; This is the measurement noise input matrix for the state prediction model.

9. The Boost converter model predictive control system based on a disturbance observer according to claim 8, characterized in that: The Model Predictive Controller (MPC) is used to construct a finite-time optimization problem according to the following steps: Defined in Time for prediction time domain Reference trajectory within Define the output voltage reference error , For reference trajectory Inner One element, This is a reference value for the output voltage; Defined in Time is crucial for controlling the time domain Incremental sequence of control quantities within Among them Time is crucial for controlling the time domain Inner Increment of control quantity at time step , , They are respectively in Time is crucial for controlling the time domain Inner Time step, first Control parameters for time steps; The following finite-time optimization problem is derived: ； In the above formula, , They are respectively with Item and The weighted matrix corresponding to each item.

10. The Boost converter model predictive control method based on a disturbance observer according to claim 4, characterized in that: The Model Predictive Controller (MPC) is used to transform the finite-time optimization problem into a standard quadratic programming (QP) form for solution according to the following steps: First, let in Time is crucial for controlling the time domain Inner Control quantity of time step ,in , for The control quantity, In order to be in Time is crucial for controlling the time domain Inner Increment of control quantity at time step For steady-state duty cycle, it will be Time is crucial for controlling the time domain Internal control sequence Rewrite it in the following form: , The control quantity for the previous moment, , It is a lower triangular identity matrix; Will Substituting into the following equation of the state prediction equation This yields the first transformation form of the state prediction equation: ; Subsequently, the output free response equation is constructed. Based on the output free response equation, the first transformation form of the state prediction equation is transformed into the second transformation form: intermediate matrix ; Then, construct the voltage tracking vector. The deviation between the output of the free response equation and the reference trajectory is obtained. The deviation between the output and the reference trajectory based on the free response equation The finite-time optimization problem is transformed into the following standard quadratic programming (QP) form: ； In the above formula, , All are intermediate quantities. , ; Finally, solve the standard quadratic programming QP form to generate... Optimal control increment sequence at time 1 Based on the incremental sequence of optimal control quantity The first control increment Calculated Optimal control quantity at time 1 The obtained optimal control quantity A PWM modulator is applied to the Boost converter system to make the output voltage of the Boost converter system track the voltage reference value.

Images