A dynamic wireless charging output power control method based on sliding mode predictive control

Abstract

The present invention discloses a dynamic wireless charging output power control method based on sliding mode predictive control, which belongs to the field of wireless power transmission technology. The method is applied to an LCC-S type resonant circuit, which can keep the output voltage stable during the dynamic process, maximize the output power of the magnetic coupling resonant wireless power transmission system, and can be controlled by sampling the load voltage and the inductor current. The sliding mode predictive controller based on the method of the present invention can realize feedback self-correction of the sliding mode control. Compared with conventional controllers, the controller implemented by the method of the present invention shows more accurate and faster performance in both transient and steady-state environments. The controller is a new predictive control algorithm based on certain conditions, which improves the available operating range of the system and ensures the reliability of the wireless charging system under dynamic changes in the coupling coefficient over a large range.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless power transmission, and in particular to a dynamic wireless charging output power control method based on sliding mode predictive control. Background Art

[0002] Electric vehicles and AGV mobile robots face problems with battery life and charging during use. The use of wireless power transmission can solve problems such as interface limitations and cumbersome procedures, and effectively improve charging reliability. At present, there are two main wireless power transmission methods: static and dynamic. However, static wireless charging has problems such as fixed charging positions and low output power. Dynamic wireless charging technology (dynamic wireless power transmission, DWPT) provides energy supply to objects in motion in a non-contact manner, effectively improving battery life. Considering that high-speed driving brings insecurity to charging stability and system design. For this reason, some scholars have proposed that fast dynamic wireless charging during the low-speed driving stage can effectively reduce the difficulty and cost of system design.

[0003] However, during dynamic charging, the receiving coil keeps moving in real time, and the transmission distance between the coils inevitably changes. This requires the system to have a wide offset range, that is, when the coupling coefficient between the coils changes within a certain range, the transmission power remains stable. Some scholars have shown that the essential reason for the power fluctuation caused by the change in the relative position between the coils is the change in mutual inductance. From the perspective of the coupling structure, changes in the coupling coefficient under different compensation topologies will affect the power and efficiency, and changes in the coupling coefficient need to be compensated from a structural or control perspective. Since wireless charging is in a dynamic state, the system is complex and nonlinear, and is easily affected by internal and external parameter disturbances, a controller with good robustness is required.

[0004] Sliding mode control is a discontinuous nonlinear control that has good control performance for nonlinear systems. However, when there is external noise interference and parameter perturbation, the system's interlacing on the sliding surface will become violent, causing the controller to vibrate. Predictive control is a special type of control that solves an optimal control at each sampling moment. In the DWPT system, it is necessary to consider the impact of the wide range and rapid changes of the coupling coefficient on the reliability of the controlled system. In the case of motion, the coupling coefficient varies greatly. Therefore, there are high requirements for the adjustment speed of the control algorithm. Traditional control algorithms cannot adjust the system quickly, which means that the control voltage and output current of the system will fluctuate during the dynamic process.

[0005] Sliding mode predictive control is an optimal control strategy that combines the ideas of variable structure control and predictive optimization. This method has a fast adjustment speed and can ensure good tracking performance when the coupling coefficient changes rapidly over a large range. By using sliding mode predictive control to adjust the DC-input voltage, the DWPT system can operate reliably. When the coupling coefficient changes rapidly over a large range, the traditional proportional-integral (PI) control algorithm is difficult to adjust quickly, resulting in system oscillation. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a dynamic wireless charging output power control method based on sliding mode predictive control. A sliding mode predictive control algorithm is proposed for a dynamically coupled discrete nonlinear system to improve the output stability problem of the object to be charged during dynamic charging at low speed.

[0007] In order to solve the above-mentioned technical problems, the present invention is achieved as follows:

[0008] A dynamic wireless charging output power control method based on sliding mode predictive control, the method is applied to an LCC-S type resonant circuit, comprising:

[0009] Step 10, constructing a system topology model based on an LCC-S type resonant circuit, selecting the error between the output voltage and the control voltage as the state variable of the control, the state variable including the tracking deviation, the tracking deviation change rate and the tracking deviation integral, and then derivatizing the state variable to obtain the system state space equation, the full-bridge inverter of the system adopts a phase shift modulation method;

[0010] Step 20, construct a sliding surface. When the system is close to the expected output power, the sliding mode control predicts the next output. If overshoot occurs, the current predicted value minus the predicted value at the previous moment is used as feedback correction to give the current sliding mode predictive control step. After the receiving end reaches the expected power, the output voltage remains stable. The sliding surface is derived with respect to time to obtain the control algorithm of the sliding mode controller.

[0011] Step 30, collecting the filter capacitor current and load voltage at the receiving end, discretizing through the Euler formula, predicting the filter capacitor current and load voltage at the next moment, substituting the prediction result into the control algorithm of the sliding mode controller, and calculating the sliding mode control output value at the next moment;

[0012] Step 40: Feedback correction is performed on the sliding mode prediction output value at time k+1 by using the difference between the sliding mode control output value at the current time and the sliding mode prediction value at time k-1, and the corrected prediction output value satisfies the cost function to gradually approach 0;

[0013] Step 50: Substitute the corrected predicted output value into the phase shift control equation, calculate the phase shift angle, control the phase shift of the full-bridge inverter, and realize the control of the output power.

[0014] Furthermore, in step 10, the state variable is specifically:

[0015]

[0016] Among them, x 1 (k) is the tracking deviation, x 2 (k) is the rate of change of tracking deviation, x 3 (k) is the tracking deviation integral, V ref is the reference voltage, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, u 0 (k) is the load voltage value that changes with time, and u(k) is the switching control law, which is equivalent to the effective value of the secondary side rectifier bridge output voltage u out (k), k is the current time divided by the control period.

[0017] Furthermore, the system state space equation is:

[0018]

[0019] in, is the derivative of the tracking deviation, is the derivative of the rate of change of the tracking deviation, is the derivative of the tracking deviation integral, X(k) is the system state variable at time k, u(k) is the system control variable at time k, and u 0 (k) is the load voltage value at time k.

[0020] Furthermore, in step 20, the sliding surface is specifically:

[0021] s(k)=σ T x(k)-α k s(0)

[0022] Among them, σ T =[σ 1 ,…,σ n ]; α k s(0) is the prediction equation, which means T The prediction of x(k) changes is converged by an exponential function, 0<α<1; k is the current time divided by the control period;

[0023] The control algorithm of the sliding mode controller is specifically:

[0024]

[0025] Among them, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, k 1 , k 2 , k 3 is the sliding mode control coefficient, i c (k) is the filter capacitor current value at time k, u 0 (k) is the load voltage value at time k, u eq (k) is the sliding mode control output value at time k, V ref is the reference voltage.

[0026] Furthermore, in step 30, the formula for predicting the filter capacitor current and load voltage at the next moment is specifically:

[0027]

[0028] Among them, i c (k) is the filter capacitor current value at time k, T s is the control period, u 0 (k) is the load voltage value at time k, u(k) is the switching control law at time k, which is equivalent to the effective value u of the secondary side rectifier bridge output voltage at time k. out (k).

[0029] Furthermore, the calculation formula of the sliding mode control output value at the next moment is:

[0030]

[0031] Among them, i c (k+1) is the current value of the filter capacitor at time k+1, u 0 (k+1) is the load voltage value at time k+1, 0<α<1, and k is the current time divided by the control period.

[0032] Furthermore, in step 40, the feedback correction specifically adopts the following formula:

[0033]

[0034] in,

[0035]

[0036]

[0037]

[0038] u eq(k) is the sliding mode control output value at the current moment, u eq (k-1) is the sliding mode prediction value at time k-1, u eq (k+1) is the sliding mode prediction output value at time k+1, ζ p is the feedback correction coefficient, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, k 1 , k 2 , k 3 is the sliding mode control coefficient, i c (k) is the filter capacitor current value at time k, u 0 (k) is the load voltage value at time k, V ref is the reference voltage, 0<α<1, and k is the current time divided by the control period.

[0039] Furthermore, in step 40, the cost function is specifically:

[0040]

[0041] Among them, V ref is the reference voltage, u 0 (k) is the load voltage value that changes with time, is the output value of the sliding mode predictive controller at time k+1.

[0042] Furthermore, in step 50, the phase shift control equation is:

[0043]

[0044] Where θ is the phase shift angle, U out is the effective value of the rectifier bridge output voltage, E is the inverter bridge input DC voltage, M is the mutual inductance, L is the r Compensate the topological inductance at the transmitter end.

[0045] The technical solution provided in the embodiments of the present invention has the following technical effects or advantages:

[0046] The algorithm proposed in the present invention can keep the output voltage stable during the dynamic process, maximize the output power of the magnetic coupling resonant wireless power transmission system, and can be controlled by only sampling the load voltage and inductor current.

[0047] The sliding mode predictive controller of the present invention can realize feedback self-correction of sliding mode control. Compared with conventional controllers, the controller proposed by the present invention shows more accurate and faster performance in both transient and steady-state environments. The controller is based on a new predictive control algorithm under certain conditions, which improves the available working range of the system and ensures the reliability of the wireless charging system under dynamic changes in the coupling coefficient over a large range.

[0048] The controller of the present invention is suitable for wireless charging of electric vehicles, AGVs, etc. at low speeds. The DWPT system under sliding mode predictive control has good output stability and dynamic tracking performance, while the output power and efficiency are improved. This method can be used to solve the power loss and load voltage oscillation problems generated in dynamic wireless charging.

[0049] The system designed by the present invention has made great progress compared to the PI control system, showing more accurate and faster performance in both dynamic and steady-state environments, can feedback and correct the algorithm model according to changes in the environment, has strong anti-disturbance ability, and can run well in practical applications. Compared with other traditional control systems in wireless power transmission technology, the control system designed by the method of the present invention is easier to implement in hardware.

[0050] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The present invention will be further described below in conjunction with embodiments with reference to the accompanying drawings.

[0052] Figure 1 The topological structure diagram of the wireless power transmission system LCC-S based on sliding mode predictive control;

[0053] Figure 2 A logical schematic diagram for implementing an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of a control flow of an embodiment of the present invention;

[0055] Figure 4 Simulation of dynamic displacement of moving coil at 15km / h;

[0056] Figure 5 Simulation of dynamic displacement of moving coil at 30km / h;

[0057] Figure 6 is the switching control quantity u eq Schematic diagram;

[0058] Figure 7 This is the experimental diagram of the dynamic offset of the system in the open loop state;

[0059] Figure 8 This is the experimental diagram of the dynamic offset of the system under PI control;

[0060] Fig. 9This is the experimental diagram of the dynamic deviation of the system under sliding mode predictive control;

[0061] Fig.10 It is an experiment of load resistance mutation under sliding mode predictive control;

[0062] Fig.11 This is the load resistance mutation experiment of the system under PI control;

[0063] Fig.12 Schematic diagram of the process of the embodiment of the present invention. DETAILED DESCRIPTION

[0064] The embodiment of the present application provides a dynamic wireless charging output power control method based on sliding mode predictive control, and proposes a sliding mode predictive control algorithm for a dynamically coupled discrete nonlinear system to improve the output stability problem of a charged object during dynamic charging at low speed.

[0065] The technical solution in the embodiment of the present application has the following general idea:

[0066] In the field of research and engineering application of wireless power transmission systems, four basic resonant compensation structures and composite resonant compensation structures such as LCC-S (inductor-capacitor-capacitor-series) and LCC-LCC (inductor-capacitor-capacitor-inductor-capacitor-capacitor) are commonly used. In a fixed circuit with parameters ignoring the internal resistance of each inductor coil, the track current in the LCC-S compensation topology is only related to the parameters of the transmitter, has a constant current output characteristic, and the receiving end can operate in an open circuit. In order to consider the stable operation of the system, the present invention selects the LCC-S resonant compensation structure.

[0067] like Figure 1 As shown, this embodiment adopts a DWPT system topology of an LCC-S type resonant circuit. The system as a whole consists of a DC power supply, a resonant mechanism, a controller, an inverter, a rectifier filter module and a load. The system includes a DC voltage source E, a phase-shifted bridge inverter, a primary resonant cavity, a secondary resonant cavity, a rectifier bridge, and a primary compensation capacitor C. p , parallel compensation capacitor C 1 , compensation inductance L 1 , secondary capacitor C s 、Filter capacitor C f , filter inductor L f and load resistance R L ; L p and L s are the self-inductance of the transmitting coil and the receiving coil respectively, M is the mutual inductance between the coils; U inis the fundamental effective value of the high frequency inverter output voltage, U out is the effective value of the output voltage of the rectifier bridge. The DC power output by the regulated DC power supply will be converted into high-frequency AC power with a frequency consistent with the resonant frequency through the bridge inverter, and then input into the resonant mechanism. According to the working principle of magnetic coupling resonance, high-frequency AC power will be converted into electromagnetic waves, transmitted from the transmitting end to the receiving end through the resonator, and generate induced AC power of the same frequency at the receiving end. The bridge rectifier rectifies the induced AC power into DC power, and outputs it to the load through capacitor filtering. The resonant circuit is composed of an inductor and a capacitor in series, and its resonant frequency can be defined as:

[0068]

[0069] The full-bridge inverter adopts phase shift modulation, and the phase shift angle θ is proportional to the fundamental effective value of the high-frequency inverter output voltage U in The relationship is:

[0070]

[0071] When the system is in a resonant state and the resonant frequency is equal to the inverter operating frequency, the impedance of the system can be simplified. At this time, the effective value of the rectifier bridge output voltage U out It can be expressed as:

[0072]

[0073] When the relative distance of the resonant coils changes, it will cause changes in the electromagnetic field of the electromagnetic wave transmission medium, that is, the air gap. When the transmitting coil and the receiving coil are coaxially symmetrical spiral coils with the same structure, the transmission distance and mutual inductance have the following relationship:

[0074]

[0075] The present invention selects phase shift modulation technology to generate a driving signal for an inverter power tube.

[0076] The sliding mode predictive controller proposed in the present invention can analyze the sampled system output voltage and current according to the prediction model of the DWPT system, and calculate the value of the phase shift angle θ transmitted to the phase shift modulator at the next moment. When the system is close to the expected output power, the sliding mode control predicts the next output. If overshoot occurs, the current prediction value minus the prediction value at the previous moment is used as feedback correction to give the current sliding mode predictive control step.

[0077] like Figure 2 As shown, the dynamic wireless charging output power control method based on sliding mode predictive control proposed in the present invention has the following implementation logic:

[0078] A mathematical model based on the LCC-S topology is established, the error between the output voltage and the control voltage is selected as the state variable of the control, and the tracking deviation x is calculated based on the forward Euler method discretization method. 1 , Tracking deviation change rate x 2 , tracking deviation integral x 3 It can be expressed as:

[0079]

[0080] Among them, x 1 (k) is the tracking deviation, x 2 (k) is the rate of change of tracking deviation, x 3 (k) is the tracking deviation integral, V ref is the reference voltage, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, u 0 (k) is the load voltage value that changes with time, and u(k) is the switching control law, which is equivalent to the effective value of the secondary side rectifier bridge output voltage u out (k), k is the current time divided by the control period.

[0081] Considering that discrete integrals are not easy to express, the state variable value is differentiated. The system state space equation required by the controller can be obtained by differentiating the equation with respect to differentials, namely:

[0082]

[0083] Collect filter capacitor current and load voltage as control requirements. The control process is as follows: Figure 3 shown.

[0084] Among them, the filter current and load voltage values ​​are discretized using the Euler formula to obtain the filter capacitor current and load voltage at the next moment:

[0085]

[0086] When the object is charged dynamically at a low speed, the axial distance between the coils does not change, but the radial distance is offset. When the offset is less than ±4cm, the change rate of the mutual inductance is less than 7.8%; when the offset is greater than ±4cm, the mutual inductance begins to decrease in an approximately linear proportion, at which point the overall transmission power and efficiency of the system decrease rapidly. When the coil offset is close to ±4cm, the mutual inductance coupling meets the transmission requirements, and the impact of the mutual inductance change on the system within this range can be compensated by the controller.

[0087] The system in the present invention has a constant output power, that is, a constant voltage output. Therefore, when the system meets the power that can be transmitted, the mutual inductance value is still increasing, and the output power needs to be controlled to reduce the fluctuation of the receiving end voltage. At this time, the system state is approaching the sliding surface, and the maximum power output of the phase shift angle continues to be maintained. After the output power is met, the system state jumps over the sliding surface, and the system state is interspersed back and forth on the sliding surface through the action of the constant speed convergence law. The waveform of the output voltage at this time will show an overshoot and then tend to be stable.

[0088] The sliding surface is defined as:

[0089] s(k)=σ T x(k)-α k s(0) (8)

[0090] Among them, σ T =[σ 1 ,…,σ n ], take n = 3; choose a suitable σ i To ensure the stability and dynamic performance of the sliding mode. k s(0) is the prediction equation, which means T For the prediction of x(k) changes, we use an exponential function with convergence of 0<α<1. According to the system conditions, we select α=0.99. k=T / T s , k is the current time divided by the control period. The sliding surface is represented by S = {x(k)|s(x(k)} = 0}.

[0091] When k = 0, s(0) = σ T x(0)-s(0)=0, that is, under any conditions, the initial state of the system is on the sliding surface. When k tends to infinity, when 0<α<1, the sliding surface formula is simplified to s(k=σ T s(k)-α ∞ s(0)=s(k), the sliding surface tends to be stable.

[0092] Define the Lyapunov function as:

[0093]

[0094] To ensure When s approaches 0+, there are:

[0095]

[0096] When s approaches 0, there is:

[0097]

[0098] Taking the derivative of the sliding mode surface with time, we get:

[0099]

[0100] Substituting formula (6) and formula (12) into formula (10) and formula (11), it can be obtained that the necessary and sufficient condition for the existence of a sliding mode zone on the switching surface is that the equivalent control satisfies:

[0101]

[0102] According to the nominal error system and sliding mode prediction model, it can be derived that:

[0103] s(k+p)=σ T x(k+p)-α k+p s(0) (14)

[0104] Taking p = 1, we get s(k+1) = σ T x(k+1)-α k+1 s(0), where:

[0105]

[0106] When kp is the starting time, the output value of the sliding mode prediction model at the current time It is expressed as:

[0107]

[0108] Then the output value at the previous moment is expressed as:

[0109]

[0110] According to formula (15) and formula (17), it can be deduced that:

[0111]

[0112]

[0113] In actual operation, due to the existence of time-varying, nonlinear and external interference in the system, the output value of the sliding mode prediction model will inevitably differ from the output value of the actual switching function. eq (k) and the advance prediction value u at time k-1 eq The difference between (k-1) is used to output the sliding mode prediction model u eq (k+1) performs feedback correction. Therefore, the output of the sliding mode predictive control is It can be expressed as:

[0114]

[0115] Among them, p∈R is the feedback correction coefficient. The effect of feedback correction will change with ζ p decreased due to the reduction of .

[0116] In view of the system requirement to achieve the goal of output voltage control, the cost function of the sliding mode prediction algorithm can be defined as:

[0117]

[0118] The value of It is irrelevant, because the system satisfies the sliding surface The quasi-sliding mode arrival condition in the x neighborhood of . Therefore, the actual sliding mode motion of the closed-loop system must converge to the x neighborhood of the sliding surface and remain within it. Since the stability and dynamic performance of the sliding mode have been guaranteed in advance, the closed-loop control system composed of is robust and stable.

[0119] In order to verify the reliability of the controller proposed in the present invention on the DWPT system, simulation and experimental verification were carried out. The system parameters of the simulation and experiment are shown in Table 1.

[0120] Table 1

[0121]

[0122] Figure 4 and Figure 5 The simulation results of coil movement at 15km / h and 30km / h are given respectively, where S1 represents the simulation result of sliding mode predictive control and S2 represents the simulation result of PI control. The figure shows the system output voltage u 0 , output current i 0 , secondary current i s When the mutual inductance coefficient between the receiving coil and the transmitting coil begins to rise rapidly, the output voltage can be stabilized to 10V.

[0123] The system is under sliding mode predictive control, and within 30% error of the optimal mutual inductance, u 0 It is able to return the reference value; and the waveform is consistent at different speeds. When the coupling coefficient changes rapidly over a wide range, the sliding mode predictive controller has good tracking performance, so the waveform changes consistently at different speeds. The simulation results show that the system with the sliding mode predictive control algorithm has a short settling time and fast dynamic response, which shows that the algorithm is suitable for the DWPT system.

[0124] Figure 6 To simulate the coil movement at a speed of 15 km / h, the control quantity u is switched between feedback correction and non-feedback correction. eq Feedback correction makes the control amount converge to 0 faster, avoids steady-state error, and the phase angle is denser.

[0125] In order to verify the reliability of the proposed controller in practical applications, a DWPT experimental platform device was established based on TMS320F28027DSP as the control core. The system consists of a DC power supply, an inverter, an LCC-S type resonator, a rectifier and filter module, an electronic load and a controller.

[0126] In the experiment, the resonator is an equilateral and symmetrical circular coil wound with Litz wire, and a magnetic isolation sheet is attached to the back of the coil. An oscilloscope is used to record the data.

[0127] In order to verify the performance of the algorithm, a comparative experiment between the algorithm proposed in the present invention and PI was carried out on the experimental platform under different conditions. The control coefficients of the simulation and the experiment are shown in Table 2.

[0128] Table 2

[0129]

[0130] In the experiment, the axial distance between the receiving coil and the transmitting coil is 15 cm, and the receiving coil moves at a constant speed of 1 km / h. The load size is 12Ω. The experimental results under open loop and different control algorithms are as follows: Figure 7-9 As shown, the system output voltage u 0 , output current i 0 , secondary current i s .

[0131] From the experimental results Figures 7 to 9 It can be seen that when the system is in an open loop state, u 0 As the coupling coefficient decreases, the output power also decreases. When the sliding mode predictive control is adopted, the system output is relatively stable within a certain working range, and the output power and efficiency are improved, which effectively improves the offset adaptability of the system.

[0132] Regardless of whether the system adopts sliding mode predictive control or PI control, it has a smaller overshoot; however, the system stable energy transmission time under sliding mode predictive control is 2.46s, which is greater than PI's 2.14s, that is, the sliding mode prediction algorithm has a wider offset range. At the same time, within the offset stable transmission range, the sliding mode predictive control output ripple is smaller, the voltage impact on the load is smaller, and it has a better control effect. When the coupling coefficient varies within a large range, the control response time of the DWPT system is very important to the safety and reliability of the system. The sliding mode prediction algorithm proposed in the present invention is suitable for the DWPT system under low-speed driving.

[0133] Fig.10 and Fig.11 They represent the system output voltage u under sliding mode predictive control and PI control when the load changes between 12Ω and 22Ω respectively. 0 , output current i0 , secondary current i s The coupling coefficient of the system is constant, u 0 Keep at 10V. When the load changes from 12Ω to 22Ω, the adjustment time of the system under sliding mode predictive control and PI control is 2ms and 12ms respectively; when the load changes from 22Ω to 12Ω, the adjustment time of the system is 5ms and 32ms respectively. In the DWPT system, the settling time is significant.

[0134] A specific embodiment is as follows:

[0135] This embodiment provides a dynamic wireless charging output power control method based on sliding mode predictive control. Fig.12 As shown, the method is applied to an LCC-S type resonant circuit, comprising:

[0136] Step 10, constructing a system topology model based on an LCC-S type resonant circuit, selecting the error between the output voltage and the control voltage as the state variable of the control, the state variable including the tracking deviation, the tracking deviation change rate and the tracking deviation integral, and then derivatizing the state variable to obtain the system state space equation, the full-bridge inverter of the system adopts a phase shift modulation method;

[0137] Step 20, construct a sliding surface. When the system is close to the expected output power, the sliding mode control predicts the next output. If overshoot occurs, the current predicted value minus the predicted value at the previous moment is used as feedback correction to give the current sliding mode predictive control step. After the receiving end reaches the expected power, the output voltage remains stable. The sliding surface is derived with respect to time to obtain the control algorithm of the sliding mode controller.

[0138] Step 30, collecting the filter capacitor current and load voltage at the receiving end, discretizing through the Euler formula, predicting the filter capacitor current and load voltage at the next moment, substituting the prediction result into the control algorithm of the sliding mode controller, and calculating the sliding mode control output value at the next moment;

[0139] Step 40: Feedback correction is performed on the sliding mode prediction output value at time k+1 by using the difference between the sliding mode control output value at the current time and the sliding mode prediction value at time k-1, and the corrected prediction output value satisfies the cost function to gradually approach 0;

[0140] Step 50: Substitute the corrected predicted output value into the phase shift control equation, calculate the phase shift angle, control the phase shift of the full-bridge inverter, and realize the control of the output power.

[0141] In a possible implementation, in step 10, the state variable is specifically:

[0142]

[0143] Among them, x 1 (k) is the tracking deviation, x 2 (k) is the rate of change of tracking deviation, x 3 (k) is the tracking deviation integral, V ref is the reference voltage, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, u 0 (k) is the load voltage value that changes with time, and u(k) is the switching control law, which is equivalent to the effective value of the secondary side rectifier bridge output voltage u out (k), k is the current time divided by the control period.

[0144] The system state space equation is:

[0145]

[0146] in, is the derivative of the tracking deviation, is the derivative of the rate of change of the tracking deviation, is the derivative of the tracking deviation integral, X(k) is the system state variable at time k, u(k) is the system control variable at time k, and u 0 (k) is the load voltage value at time k.

[0147] In step 20, the sliding surface is specifically:

[0148] s(k)=σ T x(k)-α k s(0)

[0149] Among them, σ T =[σ 1 ,…,σ n ]; α k s(0) is the prediction equation, which means T The prediction of x(k) changes is converged by an exponential function, 0<α<1; k is the current time divided by the control period;

[0150] The control algorithm of the sliding mode controller is specifically:

[0151]

[0152] Among them, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, k 1 , k 2 , k 3 is the sliding mode control coefficient, i c(k) is the filter capacitor current value at time k, u 0 (k) is the load voltage value at time k, u eq (k) is the sliding mode control output value at time k, V ref is the reference voltage.

[0153] In step 30, the formula for predicting the filter capacitor current and load voltage at the next moment is specifically:

[0154]

[0155] Among them, i c (k) is the filter capacitor current value at time k, T s is the control period, u 0 (k) is the load voltage value at time k, u(k) is the switching control law at time k, which is equivalent to the effective value u of the secondary side rectifier bridge output voltage at time k. out (k).

[0156] The calculation formula of the sliding mode control output value at the next moment is:

[0157]

[0158] Among them, i c (k+1) is the current value of the filter capacitor at time k+1, u 0 (k+1) is the load voltage value at time k+1, 0<α<1, and k is the current time divided by the control period.

[0159] In step 40, the feedback correction specifically adopts the following formula:

[0160]

[0161] in,

[0162]

[0163]

[0164]

[0165] u eq (k) is the sliding mode control output value at the current moment, u eq (k-1) is the sliding mode prediction value at time k-1, u eq (k+1) is the sliding mode prediction output value at time k+1, ζ p is the feedback correction coefficient, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, k 1 , k2 , k 3 is the sliding mode control coefficient, i c (k) is the filter capacitor current value at time k, u 0 (k) is the load voltage value at time k, V ref is the reference voltage, 0<α<1, and k is the current time divided by the control period.

[0166] In step 40, the cost function is specifically:

[0167]

[0168] Among them, V ref is the reference voltage, u 0 (k) is the load voltage value that changes with time, is the output value of the sliding mode predictive controller at time k+1.

[0169] In step 50, the phase shift control equation is:

[0170]

[0171] Where θ is the phase shift angle, U out is the effective value of the rectifier bridge output voltage, E is the inverter bridge input DC voltage, M is the mutual inductance, L is the r Compensate the topological inductance at the transmitter end.

[0172] The algorithm proposed in the present invention can keep the output voltage stable in the dynamic process, maximize the output power of the magnetic coupling resonant wireless power transmission system, and can be controlled by sampling the load voltage and inductor current. The sliding mode predictive controller of the present invention can realize feedback self-correction of the sliding mode control. Compared with the conventional controller, the controller proposed in the present invention shows more accurate and faster performance in both transient and steady-state environments. The controller is a new predictive control algorithm based on certain conditions, which improves the available working range of the system and ensures the reliability of the wireless charging system under dynamic changes in the coupling coefficient over a large range. The controller of the present invention is suitable for wireless charging of electric vehicles, AGV cars, etc. at low speeds. The DWPT system under sliding mode predictive control has good output stability and dynamic tracking performance, while the output power and efficiency are improved. This method can be used to solve the power loss and load voltage oscillation problems generated in dynamic wireless charging. The system designed by the present invention has made great progress compared to the PI control system, and shows more accurate and faster performance in both dynamic and steady-state environments. The algorithm model can be feedback-corrected according to environmental changes, has strong anti-disturbance ability, and can run well in practical applications. Compared with other traditional control systems in wireless power transmission technology, the control system designed by the method of the present invention is easier to implement in hardware.

[0173] Although the specific implementation modes of the present invention are described above, those skilled in the art should understand that the specific implementation modes described are only illustrative and are not intended to limit the scope of the present invention. Equivalent modifications and changes made by those skilled in the art in accordance with the spirit of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A dynamic wireless charging output power control method based on sliding mode predictive control, characterized in that: The method is applied to an LCC-S type resonant circuit, comprising: Step 10, constructing a system topology model based on an LCC-S type resonant circuit, selecting the error between the output voltage and the control voltage as the state variable of the control, the state variable including the tracking deviation, the tracking deviation change rate and the tracking deviation integral, and then derivatizing the state variable to obtain the system state space equation, the full-bridge inverter of the system adopts a phase shift modulation method; Step 20, construct a sliding surface. When the system is close to the expected output power, the sliding mode control predicts the next output. If overshoot occurs, the current predicted value minus the predicted value at the previous moment is used as feedback correction to give the current sliding mode predictive control step. After the receiving end reaches the expected power, the output voltage remains stable. The sliding surface is derived with respect to time to obtain the control algorithm of the sliding mode controller. Step 30, collecting the filter capacitor current and load voltage at the receiving end, discretizing through the Euler formula, predicting the filter capacitor current and load voltage at the next moment, substituting the prediction result into the control algorithm of the sliding mode controller, and calculating the sliding mode control output value at the next moment; Step 40: The current sliding mode control output value is compared with the k -1 moment to calculate the difference between the sliding mode prediction values k +1 time, the sliding mode prediction output value is feedback corrected, and the corrected prediction output value satisfies the cost function to gradually approach 0; Step 50, substituting the corrected predicted output value into the phase shift control equation, calculating the phase shift angle, controlling the phase shift of the full-bridge inverter, and realizing the control of the output power; In step 10, the state variable is specifically: in, x 1( k ) is the tracking deviation, x 2( k ) is the tracking deviation change rate, x 3( k ) is the tracking deviation integral, V ref is the reference voltage, R L is the load resistance, L f is the filter inductor, C f is the filter capacitor, u ( k ) is equivalent to the effective value of the secondary side rectifier bridge output voltage u out ( k ), k The current time is divided by the control period; The system state space equation is: in, is the derivative of the tracking deviation, is the derivative of the rate of change of the tracking deviation, is the derivative of the tracking deviation integral, X ( k )for k The system state variables at time for k The system control quantity at the moment, u 0( k ) is the load voltage value at time k; In step 20, the sliding surface is specifically: in, ; is the prediction equation, which means The prediction of changes, using exponential function convergence, ; k The current time is divided by the control period; The control algorithm of the sliding mode controller is specifically: in, k 1. k 2. k 3 is the sliding mode control coefficient, i c ( k )for k The filter capacitor current value at the moment, u eq ( k )for k The sliding mode control output value at time , V ref is the reference voltage.

2. The method according to claim 1, characterized in that: In step 30, the formula for predicting the filter capacitor current and load voltage at the next moment is specifically: in, i c ( k )for k The current value of the filter capacitor at the moment, T s To control the cycle.

3. The method according to claim 2, characterized in that: The calculation formula of the sliding mode control output value at the next moment is: in, i c ( k +1) k +1 moment filter capacitor current value, u 0( k+1 )for k The load voltage value at time +1, , k It is the current time divided by the control period.

4. The method according to claim 1, characterized in that: In step 40, the feedback correction specifically adopts the following formula: in, u eq ( k -1) k The sliding mode prediction value at time -1, u eq ( k +1) k +1 moment sliding mode prediction output value, is the feedback correction coefficient, , Sliding mode predictive controller k +1 is the output value at the moment.

5. The method according to claim 1, characterized in that: In step 40, the cost function is specifically: in, Sliding mode predictive controller k +1 is the output value at the moment.

6. The method according to claim 1, characterized in that: In step 50, the phase shift control equation is: in, θ is the phase shift angle, U out is the effective value of the rectifier bridge output voltage, E is the inverter bridge input DC voltage, M is the mutual inductance, L r Compensate the topological inductance at the transmitter end.

Images