CN110492744B - Constant power load control method and circuit applied to DC-DC converter - Google Patents

Abstract

The invention discloses a constant power load control method and circuit applied to a DC-DC converter, comprising the steps of: 1) respectively sampling the output voltage and the inductor current through a voltage sampling module and a current sampling module, and inputting the sampling signals ADC module; 2) The input sampling signal is converted into a digital signal through the ADC module, and input to the correction module; 3) The correction module compares the digital signal with the predicted value obtained by solving the prediction model, and the difference is input as a correction signal to MLD-MPC Control module; 4) The MLD-MPC control module solves the control model according to the correction signal, and realizes the constant power load control of the DC-DC converter according to the switching duty ratio obtained by the solution. The invention solves the problems of steady-state error and poor dynamic performance in the traditional constant power control method, so that the DC-DC converter can adjust quickly and stably in the face of disturbances such as sudden changes in input voltage and load, and maintain constant power output.

Description

Constant power load control method and circuit applied to DC-DC converter

technical field

The invention relates to the technical field of power electronic DC-DC converters, in particular to a constant power load control method and circuit applied to the DC-DC converters.

Background technique

Power electronic DC-DC converters are widely used in DC microgrids due to their small size, light weight and high efficiency. As the application scenarios of DC-DC converters become more and more diverse, non-linear loads such as constant power loads put forward higher requirements for the control method of DC-DC converters. Traditional linearization modeling methods are difficult to accurately describe the nonlinear characteristics of DC-DC converters with constant power loads, so the constant power control of DC-DC converters requires more accurate models.

At present, the constant power load control methods applied to DC-DC converters mainly include: 1. State feedback control: take the output capacitance and output resistance of the DC-DC converter as its feedback variables, list the system state equation and carry out Taylor expansion, The state equation is linearized by discarding high-order infinitesimal terms, and the approximate linear system is controlled. 2. Virtual impedance control: A current sampling module is added to the DC-DC converter, so that the sampling current flows through a virtual resistor and then added to the control link to improve system stability. Because the Thevenin equivalent circuit of the control link is equivalent to a resistor in series in the circuit, but in fact the virtual resistor does not consume power, it is also called virtual impedance control. Although the existing constant power load control method of DC-DC converter can achieve the control effect due to the linear approximation, it has problems such as steady-state error, overshoot, and long adjustment time in terms of accuracy and dynamic performance.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and proposes a constant power load control method and circuit applied to a DC-DC converter, so as to solve the problems such as steady-state error and poor dynamic performance in the traditional constant power control method. , so that the DC-DC converter can adjust quickly and stably in the face of disturbances such as sudden changes in input voltage and load, and maintain constant power output.

In order to achieve the above purpose, the technical solution provided by the present invention is as follows: a constant power load control method applied to a DC-DC converter needs to be configured with a voltage sampling module, a current sampling module, an ADC module, a calibration module and an MLD-MPC control module , wherein the voltage sampling module is used to sample the output voltage, the current sampling module is used to sample the inductor current, the ADC module is used to convert the sampling signal into a digital signal, and the correction module is used to correct control model error, the MLD-MPC control module is used to control the duty cycle of the DC-DC converter;

The constant power load control method includes the following steps:

1) Sample the output voltage and the inductor current through the voltage sampling module and the current sampling module respectively, and input the sampling signal into the ADC module;

2) Convert the input sampling signal into a digital signal through the ADC module, and input it into the correction module;

3) The correction module compares the predicted value obtained by solving the digital signal with the prediction model, and the difference is input to the MLD-MPC control module as a correction signal;

4) The MLD-MPC control module solves the control model according to the correction signal, and realizes the constant power load control of the DC-DC converter according to the obtained switching duty cycle.

Further, the correction module internally uses a mixed logic dynamic model, namely the MLD model, to predict the output voltage and inductor current. The basic features of the MLD model are: by defining 0-1 logic variables, the nonlinearity of the DC-DC converter is calculated. The system is divided into a set of M finite linear subsystems, and auxiliary variables are constructed to express it in the form of a linear system. The switching logic between subsystems is described by mixed integer inequalities. The MLD model has the following form:

Among them, M represents that the nonlinear system can be divided into M linear sub-models; k represents the time of the discrete model, and the corresponding continuous time is t=kTs, where Ts is the sampling time of the discrete model; x(k) represents the state variable of the system at time k; x(k+1) represents the state variable of the system at time k+1; u(k) represents the input variable of the system at time k; y(k) represents the output variable of the system at time k; δ _j is the system at time k 0-1 logical variables; A _j , B _j , a _j , C _j , D _j , b _j are the corresponding parameter matrices of the system state equation, respectively.

Further, the control process of the MLD-MPC control module is as follows:

S1. The MLD-MPC control module is initialized and begins to optimize the control of the system at the current moment;

S2. The MLD-MPC control module detects whether the system parameters change; if the system parameters change, rebuild the discrete state equation of the linear sub-model according to the new system parameters, update the MLD model according to the reconstructed discrete state equation, and put the new The MLD model is applied to the calibration module and the MLD-MPC control module, and step S3 is performed; if the system parameters do not change, step S3 is directly performed;

S3. After updating the MLD model, the MLD-MPC control model is reconstructed and solved using the updated MLD model as a constraint, and the process is completed in the MLD-MPC control module;

S4. Determine whether the MLD-MPC control model has a feasible solution;

S5: If there is no feasible solution, it is determined that the system is unstable, and the MLD-MPC control module reports an error and initializes it;

S6: If the control model has a feasible solution, take the first element of the system control sequence of the optimal solution as the system control strategy at this moment, and return to step S1 to start the control optimization at the next moment.

The constant power load control circuit applied to the DC-DC converter is composed of a voltage sampling module, a current sampling module, an ADC module, a correction module and an MLD-MPC control module, wherein the voltage sampling module and the DC-DC converter are connected. The two ends of the output capacitor of the main circuit are connected, the current sampling module is connected in series with the energy storage inductance of the main circuit of the DC-DC converter, the ADC module is respectively connected with the voltage sampling module, the current sampling module and the calibration module, and the calibration module is connected to the The MLD-MPC control module is connected, and the MLD-MPC control module is driven and connected to the main circuit switch tube of the DC-DC converter; during operation, the voltage sampling module and the current sampling module respectively sample the output voltage and the inductor current, and collect The sampled signal is input to the ADC module, the ADC module converts the input sampled signal into a digital signal, and the correction module is inputted. The MLD-MPC control module solves the control model according to the correction signal, and controls the conduction state of the switch tube of the main circuit of the DC-DC converter according to the obtained switch duty ratio, so as to realize the constant power load control of the DC-DC converter.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. The present invention has good starting performance, and the starting process of the converter can respond quickly without overshoot.

2. The present invention has strong stability, and can keep the output power constant when the input or the load end is disturbed greatly.

3. The present invention can be extended to constant power control in which multiple converter devices are cascaded or connected in parallel. For multiple converter devices connected in parallel or in parallel, as long as they can be divided into a combination of finite linear systems, the method of the present invention can be applied. .

4. The present invention is convenient for configuration and tuning, and the control model can be generated in the control module independently only by inputting the parameters of the converter elements.

Description of drawings

Fig. 1 is the control logic diagram of the MLD-MPC control module.

FIG. 2 is a circuit diagram of a constant power load control circuit applied to a DC-DC converter.

FIG. 3 shows the output power response of the constant power load control method applied to the DC-DC converter under the disturbance of the input voltage.

FIG. 4 shows the output power response of the constant power load control method applied to the DC-DC converter under the disturbance of the load resistance.

FIG. 5 shows the actual output power response of the constant power load control method applied to the DC-DC converter under the change of the output power setting value.

Detailed ways

The present invention will be described in further detail below with reference to the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

In the constant power load control method applied to the DC-DC converter provided by this embodiment, firstly, a voltage sampling module, a current sampling module, an ADC module, a correction module and an MLD-MPC control module need to be configured, wherein the voltage The sampling module is used to sample the output voltage, the current sampling module is used to sample the inductor current, the ADC module is used to convert the sampled signal into a digital signal, the correction module is used to correct the control model error, the The MLD-MPC control module is used to control the duty cycle of the DC-DC converter; the constant power load control method includes the following steps:

1) Sample the output voltage and the inductor current through the voltage sampling module and the current sampling module respectively, and input the sampling signal into the ADC module;

2) Convert the input sampling signal into a digital signal through the ADC module, and input it into the correction module;

3) The correction module compares the predicted value obtained by solving the digital signal with the prediction model, and the difference is input to the MLD-MPC control module as a correction signal;

4) The MLD-MPC control module solves the control model according to the correction signal, and realizes the constant power load control of the DC-DC converter according to the obtained switching duty cycle.

The mixed logic dynamic model, namely the MLD model, is used in the correction module to predict the output voltage and inductor current. The basic characteristics of the MLD model are: by defining 0-1 logic variables, the nonlinear system of the DC-DC converter is divided into M A set of finite linear subsystems, and auxiliary variables are constructed to express it in the form of a linear system. The switching logic between subsystems is described by mixed integer inequalities. The MLD model has the following form:

Among them, M represents that the nonlinear system can be divided into M linear sub-models; k represents the time of the discrete model, and the corresponding continuous time is t=kTs, where Ts is the sampling time of the discrete model; x(k) represents the state variable of the system at time k; x(k+1) represents the state variable of the system at time k+1; u(k) represents the input variable of the system at time k; y(k) represents the output variable of the system at time k; δ _j is the system at time k 0-1 logical variables; A _j , B _j , a _j , C _j , D _j , b _j are the corresponding parameter matrices of the system state equation, respectively.

As shown in Figure 1, the control process of the MLD-MPC control module is as follows:

S1. The MLD-MPC control module is initialized and begins to optimize the control of the system at the current moment;

S2. The MLD-MPC control module detects whether the system parameters change; if the system parameters change, rebuild the discrete state equation of the linear sub-model according to the new system parameters, update the MLD model according to the reconstructed discrete state equation, and put the new The MLD model is applied to the calibration module and the MLD-MPC control module, and then step S3 is performed; if the system parameters do not change, step S3 is directly performed;

S3. After updating the MLD model, the MLD-MPC control model is reconstructed and solved using the updated MLD model as a constraint, and the process is completed in the MLD-MPC control module;

S4. Determine whether the MLD-MPC control model has a feasible solution;

S5: If there is no feasible solution, it is determined that the system is unstable, and the MLD-MPC control module reports an error and initializes it;

S6: If the control model has a feasible solution, take the first element of the system control sequence of the optimal solution as the system control strategy at this moment, and return to step S1 to start the control optimization at the next moment.

As shown in Figure 2, a constant power load control circuit applied to a DC-DC converter is given. The DC-DC converter is specifically a Boost converter, which includes an input DC power supply, an energy storage inductor L, and an N channel. MOS tube, power diode D, output capacitor C, output load R, the constant power load control circuit includes voltage sampling module, current sampling module, ADC module, correction module and MLD-MPC control module; the positive pole of the input DC power supply is connected to the energy storage inductor One end of L; the other end of the energy storage inductor L is connected to the anode of the power diode D and the D pole of the N-channel MOS transistor respectively; the cathode of the power diode D is respectively connected to one end of the output capacitor C and one end of the output load R; the output The other end of the load R is connected to the other end of the output capacitor C, the S pole of the N-channel MOS tube, and the negative pole of the input DC power supply; the voltage sampling module is connected to both ends of the output capacitor C, and the current sampling module is connected to the energy storage inductor L. The inductor current is collected, the ADC module is respectively connected with the voltage sampling module, the current sampling module and the calibration module, the calibration module is connected with the MLD-MPC control module, and the MLD-MPC control module is connected with the N-channel MOS transistor driver. The output voltage V _out and the inductor current I _l are sampled and transmitted to the correction module. After comparing with the predicted signal of the MLD model inside the module, the error signal is input to the MLD-MPC control module as a correction variable. The MLD-MPC control module passes the correction variable and state variable. The optimization solution controls the on and off of the switch to realize the constant power control of the load.

The output voltage u _C and the inductor current i _L are used as its control variable and state variable, and the state variable is recorded as:

x ₌ [i Lu _C ] ^T

From the working logic of the Boost converter, the state equations of the discrete linear subsystems can be obtained as follows:

x(k+1)=A _d1 x(k)+B _d1 u(k)

x(k+1)=A _d2 x(k)+B _d2 u(k)

x(k+1)=A _d3 x(k)

Among them, A _d1 , A _d2 , A _d3 , B _d1 , B _d2 are the discrete linear subsystem state equation parameter matrix, k represents the discrete model time, and the corresponding continuous time is t=kT _s . According to the working logic of the converter, define the 0-1 discrete logic variable and establish its MLD model, as follows:

Among them, C and D are the corresponding parameter matrices of the calculated output variables, respectively, δ ₃ (k), δ ₄ (k), δ ₅ (k) represent the time k, respectively, when the system works in three discrete linear subsystems 0-1 logical variable.

The actual state variables of the boost converter are compared with the state variables calculated by the MLD model, and the difference is used as a correction variable, which is input to the MLD-MPC control module together with the state variables at the current moment.

The internal optimization control function of the MLD-MPC control module is as follows

y(k|t)=x(k|t)

Among them, x(k|t) represents the prediction of the system based on the state variable at time k, and the state variable at the future time t, and y(k|t) represents the prediction of the system based on the state variable at time k, and at the future time t The output variable of , K represents the prediction time domain length, y _set represents the preset value of the output variable, y _set = [I ₀ U ₀ ] ^T , I ₀ is the preset value of the inductor current, and U ₀ is the preset value of the capacitor voltage. Solving the equation according to the load power P and the resistance value R can get:

By solving the optimization function inside the MLD-MPC control module, the optimal switching logic at the next moment is obtained, and then the on-state of the switch tube at the next moment is determined to keep the control variable stable at the set value.

As shown in Figure 3, the output power response under input voltage disturbance is given. Figure 3 (a) is the input voltage disturbance curve, and Figure 3 (b) is the load power response curve, which can be seen intuitively from Figure 3. Out, when the input voltage has a sudden change in the range of 50%-200%, the output power can be quickly adjusted and maintained at the set value.

As shown in Figure 4, the output power response under load resistance disturbance is given. Figure 4(a) is the load resistance disturbance curve, and Figure 4(b) is the load power response curve. It can be seen intuitively from Figure 4 Out, when the input voltage has a sudden change in the range of 25%-200%, the output power can be quickly adjusted and maintained at the set value.

As shown in Figure 5, the actual output power response under the change of the output power setting value is given. Figure 5 (a) is the load power setting value change curve, Figure 5 (b) is the output power response curve, It can be seen intuitively from Figure 5 that when the load power setting value changes abruptly in the range of 50%-150%, the output power can be quickly adjusted and maintained at the new setting value and remains stable.

The above-mentioned embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited by the above-mentioned embodiments, and any other changes, modifications, substitutions, combinations, The simplification should be equivalent replacement manners, which are all included in the protection scope of the present invention.

Claims (2)

1. be applied to the constant power load control method of DC-DC converter, it is characterized in that: need to be equipped with voltage sampling module, current sampling module, ADC module, correction module and MLD-MPC control module, wherein, described voltage sampling module used to sample the output voltage, the current sampling module is used to sample the energy storage inductor current of the main circuit, the ADC module is used to convert the sampled signal into a digital signal, the correction module is used to correct the control model error, The MLD-MPC control module is used to control the duty cycle of the DC-DC converter; The constant power load control method includes the following steps: 1) Sample the output voltage and the inductor current through the voltage sampling module and the current sampling module respectively, and input the sampling signal into the ADC module; 2) Convert the input sampling signal into a digital signal through the ADC module, and input it into the correction module; 3) The correction module compares the predicted value obtained by solving the digital signal with the prediction model, and the difference is input to the MLD-MPC control module as a correction signal; 4) The MLD-MPC control module solves the control model according to the correction signal, and realizes the constant power load control of the DC-DC converter according to the obtained switching duty cycle; The correction module uses a mixed logic dynamic model, namely the MLD model, to predict the output voltage and inductor current. The basic characteristics of the MLD model are: by defining 0-1 logic variables, the nonlinear system of the DC-DC converter is divided For a set of M finite linear subsystems, construct auxiliary variables to express it in the form of a linear system, and describe the switching logic between subsystems through mixed integer inequalities. The MLD model has the following form:

Among them, M represents that the nonlinear system can be divided into M linear sub-models; k represents the time of the discrete model, and the corresponding continuous time is t=kTs, where Ts is the sampling time of the discrete model; x(k) represents the state variable of the system at time k; x(k+1) represents the state variable of the system at time k+1; u(k) represents the input variable of the system at time k; y(k) represents the output variable of the system at time k; δ _j is the system at time k 0-1 logical variables; A _j , B _j , a _j , C _j , D _j , b _j are the corresponding parameter matrices of the system state equation, respectively. 2. the constant power load control method applied to DC-DC converter according to claim 1, is characterized in that: the control process of described MLD-MPC control module is as follows: S1. The MLD-MPC control module is initialized and begins to optimize the control of the system at the current moment; S2. The MLD-MPC control module detects whether the system parameters change; if the system parameters change, rebuild the discrete state equation of the linear sub-model according to the new system parameters, update the MLD model according to the reconstructed discrete state equation, and put the new The MLD model is applied to the correction module and the MLD-MPC control module, and then step S3 is performed; if the system parameters do not change, step S3 is directly performed; S3. If the system parameters change and the MLD model is updated, the new MLD model is used as the constraint condition. If the system parameters do not change, the MLD-MPC control model is constructed and solved by using the original MLD model as the constraint condition. The process is completed in the MLD-MPC control module; S4. Determine whether the MLD-MPC control model has a feasible solution; S5: If there is no feasible solution, it is determined that the system is unstable, and the MLD-MPC control module reports an error and initializes it; S6: If the control model has a feasible solution, take the first element of the system control sequence of the optimal solution as the system control strategy at this moment, and return to step S1 to start the control optimization at the next moment.

Images