Soft switch of high-frequency switch transformer and distributed control method and system thereof

Abstract

The invention relates to the technical field of power electronics, and discloses a soft switch of a high-frequency switch transformer and a distributed control method and system thereof, and the method comprises the steps: achieving the precise synchronization of nodes through the construction of a ring topology network, combining the monitoring of multiple physical quantities with the collaborative optimization of parameters, dynamically adjusting the parameters of a resonant network and a driving time sequence, and achieving the precise synchronization of the nodes. A zero-voltage switching state is maintained, and electromagnetic interference is suppressed by adopting the composite optimization model; the system comprises a network initialization module, a multi-physical-quantity monitoring module, a space-time synchronization control module, a resonance parameter adjustment module, an electromagnetic optimization decision module, a dynamic adjustment module and a parameter evolution module. The stability of the soft switch is improved through a distributed network and time-space synchronization, and the loss is reduced by adopting a dynamic optimization algorithm; constructing an electromagnetic interference optimization model to enhance compatibility; performing multi-parameter fusion and weight distribution to optimize dynamic response; the reliability is improved through a high-precision protocol and coevolution; and the tensor product framework realizes multi-dimensional intelligent cooperative control.

Description

Technical Field

[0001] The present invention relates to the field of power electronics technology, and in particular to a soft switch of a high-frequency switching transformer and a distributed control method and system thereof. Background Art

[0002] As power electronics devices evolve toward higher frequencies and higher power densities, the performance of high-frequency switching transformers, core components of energy conversion, directly impacts system efficiency and reliability. Traditional hard-switching technologies struggle to meet the demands of modern, efficient power conversion due to high switching losses and severe electromagnetic interference. Existing soft-switching solutions often employ centralized control architectures, which suffer from limitations such as insufficient synchronization accuracy and lag in parameter adjustment when dealing with rapid load fluctuations and multiple modules in parallel.

[0003] Current technologies typically use fixed-value designs for resonant network parameters, making them unable to adapt to varying operating conditions across a wide load range. This easily violates zero-voltage switching conditions. Time synchronization mechanisms often rely on a single master clock, and the accumulation of phase deviations between nodes can easily lead to drive signal mismatches, resulting in uneven stress on the switch tubes. Electromagnetic interference suppression methods are often limited to passive filtering, lacking active optimization and control of interference sources, making it difficult to meet stringent electromagnetic compatibility standards.

[0004] Furthermore, existing control methods often employ fixed-weight strategies when addressing multi-objective optimization problems, failing to dynamically balance the conflicting demands of efficiency improvement and electromagnetic suppression. The independent control dimensions of the parameter adjustment process lack a collaborative optimization mechanism, limiting overall system performance. Existing solutions are particularly prone to thermal runaway and oscillatory instability under extreme operating conditions, such as high temperatures and high loads, hindering the long-term reliable operation of high-frequency switching transformers.

[0005] Therefore, the present invention proposes a soft switching of a high-frequency switching transformer and a distributed control method and system thereof to address the deficiencies of the prior art. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention provides a soft switch of a high-frequency switching transformer and its distributed control method and system, which solves the problems of soft switching state instability, insufficient electromagnetic interference suppression and low multi-node collaborative control accuracy of the high-frequency switching transformer under complex working conditions.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A soft switching and distributed control method for a high-frequency switching transformer, comprising the following steps:

[0008] S1. Build a distributed control network with a ring topology, load the improved precise time protocol, initialize the equivalent inductance and capacitance parameters of the resonant cavity, and establish a clock deviation compensation matrix between nodes;

[0009] S2. Use voltage sensors and current sensors to collect the drain-source voltage, drain current, and load current change rate of each node in real time, simultaneously monitor the junction temperature data of the power device, and calculate the core magnetic flux change based on the voltage integration method;

[0010] S3. Solve the second-order partial derivative of the phase angle between nodes based on the load current change rate and the magnetic flux calculation results to generate a synchronous drive pulse signal with time stamp calibration;

[0011] S4. Based on the timing characteristics of the synchronous drive pulse signal and the rate of change of the drain current, iteratively update the resonant network parameters using a quasi-Newton method to maintain a zero voltage switching state under different load conditions;

[0012] S5. Establish a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, and obtain the optimal switching trajectory command by solving the Hamiltonian equation;

[0013] S6. Based on the real-time monitored junction temperature data and load current change rate, dynamically allocate weight coefficients for efficiency optimization and electromagnetic suppression, suppress abnormal propagation through the error energy attenuation function, and perform tensor product co-evolution of resonance parameters and synchronization parameters.

[0014] Preferably, step S1 includes: constructing a distributed control network with a ring topology, loading an improved precise time protocol including a phase angle differential compensation mechanism, initializing a set of equivalent inductance and capacitance parameters of the resonant cavity, and establishing a clock deviation compensation matrix based on the second-order mixed partial derivatives of the phase angle between nodes, wherein the construction process of the compensation matrix includes a cross-differentiation operation on the phase change rates of adjacent nodes.

[0015] Preferably, step S2 includes: collecting the instantaneous value of the drain-source voltage of each node through a voltage sensor, obtaining the drain current and the load current change rate through a current sensor, synchronously monitoring the real-time junction temperature data of the power device, and calculating the core magnetic flux change based on the voltage integration method, wherein the calculation process of the voltage integration method includes a compensation term for the equivalent resistance of the magnetic core, and the calculation result of the magnetic flux change is in a nonlinear integral relationship with the drain-source voltage sampling value.

[0016] Preferably, step S2 also includes: establishing a dynamic coupling correction model of junction temperature monitoring data and magnetic flux estimation value, and using a sliding time window algorithm to perform time domain smoothing on the load current change rate, and the smoothing process includes a weighted average operation of the current sampling value and a mutation suppression mechanism.

[0017] Preferably, step S3 includes: solving the second-order mixed partial derivative of the phase angle between nodes based on the calculation results of the load current change rate and the magnetic flux, wherein the calculation process of the partial derivative introduces a dynamic weight factor of the load current change rate, and generates a synchronous drive pulse signal with timestamp calibration, and the timestamp calibration includes a phase advance correction based on a feedforward compensation mechanism, and the generation process of the drive pulse signal and the update of the resonant network parameters form a closed-loop feedback control.

[0018] Preferably, step S4 includes: based on the timing characteristics of the synchronous drive pulse signal and the drain current change rate, using the quasi-Newton method to iteratively update the resonant network parameters, the update process comprehensively calculates the efficiency gradient, electromagnetic interference gradient and time synchronization gradient, and performs parameter correction through the inverse operation of the Jacobian matrix, and dynamically adjusts the resonant cavity quality factor to maintain the zero voltage switching state, wherein the adjustment of the quality factor and the load current change rate form a closed-loop feedback control.

[0019] Preferably, step S5 includes: establishing a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, constructing a Hamiltonian equation containing the quadratic derivability constraint of the switching trajectory, and using the adjoint variable method to process the boundary conditions to solve the optimal switching trajectory instruction, wherein the calculation process of the adjoint variable is data-coupled with the update of the resonant network parameters, and the generation of the optimal switching trajectory instruction is synchronously fed back to the timing adjustment module of the synchronous drive pulse.

[0020] Preferably, step S6 includes: based on the real-time monitored junction temperature data and load current change rate, using a sigmoid function and a cosine modulation function to combine and calculate the weight coefficients of dynamic allocation efficiency optimization and electromagnetic suppression, constructing a composite error energy attenuation function including a Gaussian kernel function and a frequency domain band limiting characteristic, and suppressing abnormal propagation through convolution operation, wherein the bandwidth parameter of the Gaussian kernel function is positively correlated with the system switching frequency, and the frequency domain band limiting characteristic is determined by the maximum operating frequency of the system.

[0021] Preferably, step S6 also includes: performing tensor product co-evolution of resonance parameters and synchronization parameters, establishing a parameter space mapping relationship based on the node coupling matrix, and defining a symbiotic evolution rule including energy conservation constraints and phase continuity conditions, wherein the construction of the coupling matrix reflects the correlation strength of the electrical parameters of adjacent nodes, and the evolution rule ensures that the parameter update process maintains the system stability boundary.

[0022] The present invention also provides a soft switch of a high-frequency switching transformer and a distributed control system thereof, wherein the system includes the following modules:

[0023] A network initialization module, which configures a distributed control network for building a ring topology and loads the improved precision time protocol;

[0024] A multi-physical quantity monitoring module is configured to collect real-time data on drain-source voltage, drain current, and load current change rate, while simultaneously monitoring power device junction temperature data and calculating core magnetic flux changes.

[0025] a spatiotemporal synchronization control module configured to solve the second-order partial derivatives of the phase angle between nodes and generate a synchronous driving pulse signal for time stamp calibration;

[0026] a resonant parameter adjustment module configured to iteratively update the resonant network parameters based on a quasi-Newton method to maintain a zero voltage switching state;

[0027] An electromagnetic optimization decision module is configured to establish a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, and to generate an optimal switching trajectory instruction by solving the Hamiltonian equation;

[0028] A dynamic adjustment module configures weight coefficients for dynamically allocating efficiency optimization and electromagnetic suppression, and suppresses abnormal propagation through a composite error energy attenuation function;

[0029] The parameter evolution module is configured to perform tensor product co-evolution of resonance parameters and synchronization parameters to achieve multi-parameter space mapping and symbiotic evolution.

[0030] The present invention provides a soft switch of a high-frequency switching transformer and a distributed control method and system thereof. It has the following beneficial effects:

[0031] 1. This invention achieves precise coordination of multi-node drive signals through a distributed control network and spatiotemporal synchronization, effectively suppressing zero-voltage switching instability caused by sudden load changes. A dynamic resonant parameter optimization algorithm ensures soft switching characteristics over a wide load range, reducing switching losses.

[0032] 2. This invention significantly reduces high-frequency harmonic radiation by constructing an electromagnetic interference spectrum norm optimization model and combining it with frequency-domain band-limited suppression technology. By using a composite error energy attenuation function, it effectively suppresses electromagnetic interference propagation under abnormal operating conditions and improves the system's adaptability to electromagnetic environments.

[0033] 3. This invention integrates multi-physical quantity monitoring data to achieve real-time compensation and correction of key parameters such as junction temperature and magnetic flux. It also uses a dynamic weight allocation mechanism to achieve an adaptive balance between efficiency optimization and electromagnetic suppression, quickly responding to load fluctuations and changes in operating conditions.

[0034] 4. This invention improves inter-node synchronization accuracy to nanoseconds by introducing an improved precise time protocol and calculating the second-order partial derivative of the phase angle. Through a parameter co-evolution mechanism, a multi-parameter spatial mapping relationship is established, enhancing the system's parameter fault tolerance and long-term operational stability.

[0035] 5. This invention builds a tensor product co-evolution framework, breaking through the limitations of traditional single-parameter optimization and achieving multi-dimensional joint optimization of resonance and synchronization parameters. It also employs a distributed decision-making architecture, enabling closed-loop data interaction between functional modules and enhancing the overall intelligent control level of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flow chart of the method of the present invention;

[0037] Figure 2 This is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0039] See also Figure 1 The embodiment of the present invention provides a soft switching of a high-frequency switching transformer and a distributed control method thereof, comprising the following steps:

[0040] S1. Build a distributed control network with a ring topology, load the improved precise time protocol, initialize the equivalent inductance and capacitance parameters of the resonant cavity, and establish a clock deviation compensation matrix between nodes;

[0041] In this embodiment, a distributed control network with a ring topology is constructed using at least three control nodes to form a closed-loop communication architecture, with information exchange between nodes achieved via a bidirectional data channel. The improved precision time protocol, based on the standard IEEE-1588 protocol, adds a phase angle differential compensation mechanism. Specifically, by embedding the phase change rate data field of adjacent nodes in the protocol frame structure, this mechanism enables real-time correction of dynamic phase deviations between nodes.

[0042] When initializing the resonant cavity parameters, the equivalent inductance parameter set includes the equivalent inductance value L at the fundamental operating frequency. eq and its third harmonic component L 3rd , where L eq Determined by the product of the core air gap length and the square of the number of winding turns, L 3rd The capacitance parameter set is obtained by fitting the nonlinear BH curve of the core material. r and transformer parasitic capacitance C p The parallel equivalent value of C total =C r +Cp and calibrated by actual measurement using an impedance analyzer.

[0043] When establishing the clock deviation compensation matrix, the second-order mixed partial derivative calculation based on the phase angle between nodes is as follows: let the phase angle of the i-th node be θ i , the phase angle of adjacent nodes is θ i-1 and θ i+1 , then the compensation matrix element M i,j Determined by the following formula:

[0044]

[0045] Where α is the correlation factor of the phase change rate of adjacent nodes, obtained by offline calibration; θ i-1 ,θ i+1 is the phase angle of adjacent nodes; θ i is the phase angle of the i-th node; θ j is the phase angle of the jth node; t is the time variable. The cross differential operation is implemented by the hardware acceleration unit of the digital signal processor, and the calculation cycle is synchronized with the switching frequency.

[0046] Preferably, the phase angle differential compensation mechanism is implemented in the synchronization message processing layer of the protocol stack, using a two-stage correction strategy of pre-compensation and post-compensation: before sending the synchronization message, the compensation amount is predicted based on the historical phase change trend; after receiving the message, a secondary fine-tuning is performed based on the measured deviation. During the resonant cavity parameter initialization process, the resonant points of the inductor and capacitor are dynamically matched by the frequency sweep method to ensure that the following are met at the rated operating frequency:

[0047]

[0048] Where f0 is the rated operating frequency of the resonant cavity; L eq is the equivalent inductance value, including the fundamental and third harmonic components; C total is the total equivalent capacitance, including resonant capacitance and parasitic capacitance.

[0049] Furthermore, the ring topology network utilizes redundant links. In the event of a single node failure, bypass relays automatically reconfigure the communication path. Clock synchronization accuracy between nodes is maintained by dynamically updating the compensation matrix, with an update period of 1 / Nth of the switching period (N being the total number of nodes), ensuring that the phase deviation across the entire network is controlled within ±5 nanoseconds.

[0050] S2. Use voltage sensors and current sensors to collect the drain-source voltage, drain current, and load current change rate of each node in real time, simultaneously monitor the junction temperature data of the power device, and calculate the core magnetic flux change based on the voltage integration method;

[0051] In this embodiment, the instantaneous value of the drain-source voltage V is collected by a voltage sensor configured at the source of the power device. ds (t), the voltage sensor adopts a differential sampling architecture with a common mode rejection ratio of not less than 80dB to eliminate switching noise interference. d The monitoring of (t) is achieved through a closed-loop Hall effect sensor, and the output of the sensor is connected to a differential circuit to obtain the load current change rate. The time constant of the differential circuit is inversely proportional to the switching period.

[0052] When monitoring the junction temperature of power devices simultaneously, a temperature-sensitive diode is embedded inside the device package, and the junction temperature T is inferred by detecting the temperature drift characteristics of the diode forward voltage drop. j (t), the sampled data is converted into a digital signal by a Σ-Δ modulator. Preferably, the junction temperature monitoring system adopts a time-interleaved sampling technique to complete at least three independent measurements within a single switching cycle to eliminate transient thermal noise.

[0053] When calculating the change of magnetic core magnetic flux based on the voltage integration method, the core equivalent resistance R is introduced core The specific calculation formula for the compensation item is:

[0054]

[0055] Where, Φ(t) is the core magnetic flux at time t; Φ residual is the residual magnetic flux of the core, measured by the initial demagnetization process; τ is the integral time variable. The equivalent resistance R core Established with junction temperature T j The association model:

[0056] R core (T j )=R0·[1+β·(T j -T0)];

[0057] Where R0 is the equivalent resistance value at the reference temperature T0; β is the temperature coefficient, which is obtained by testing the frequency domain impedance characteristics of the magnetic core material; T j is the junction temperature of the power device; T0 is the resistor calibration reference temperature.

[0058] Preferably, the nonlinear integral relationship is realized by inserting a variable gain amplifier in the integrator feedback loop, whose gain coefficient K(t) is related to the magnetic flux operating point and satisfies:

[0059]

[0060] Among them, γ is the core saturation characteristic parameter, which is determined according to the second-order derivative of the BH curve of the magnetic material.

[0061] When establishing a dynamic coupling correction model for junction temperature monitoring data and magnetic flux estimation, a transfer function is constructed:

[0062] Φ corr (t) = Φ(t)·[1-η·(T j (t)-T ref )];

[0063] Where η is the junction temperature-magnetic flux coupling coefficient; T ref The correction model is updated in real time through the recursive operation of the digital filter, and its update cycle is synchronized with the magnetic flux calculation cycle.

[0064] When using the sliding time window algorithm to perform time domain smoothing on the load current change rate, the window length W is defined as an integer multiple of the switching period, and the sampling values ​​within the window are distributed according to the exponential weight:

[0065]

[0066] Weight coefficient w k =e -λk The attenuation factor λ is adaptively adjusted according to the system noise level. The mutation suppression mechanism is implemented by comparing the variance change rate of adjacent windows. When a mutation event is detected, it automatically switches to the weighting mode of the previous window historical data.

[0067] Preferably, the smoothing process is implemented in a convolution acceleration unit of a digital signal processor, using a circular buffer to store window data. For high-frequency interference components, a cascaded digital low-pass filter is applied before the sliding window operation, with its cutoff frequency maintaining a fixed proportional relationship with the system switching frequency.

[0068] S3. Solve the second-order partial derivative of the phase angle between nodes based on the load current change rate and the magnetic flux calculation results to generate a synchronous drive pulse signal with time stamp calibration;

[0069] In this embodiment, when solving the second-order mixed partial derivative of the phase angle between nodes, based on the phase angle data θ of the adjacent nodes, i (t) and θ j (t), combined with the load current change rate And the calculation results of magnetic flux Φ(t), construct the following partial differential equation:

[0070]

[0071] Wherein, w(t) is a dynamic weight factor, which ranges from [0, 1] and is determined by normalizing the absolute value of the load current change rate. The specific expression is:

[0072]

[0073] Among them, I load (t) is the load current at time t; ∈ is a small positive number to prevent the denominator from being zero. The calculation of the second-order mixed partial derivative is implemented by the hardware differential unit of the digital signal processor, and the calculation cycle is synchronized with the switching frequency.

[0074] When generating synchronous drive pulse signals, the timestamp calibration mechanism is used to correct the phase advance of the pulse leading edge. The feedforward compensation mechanism calculates the correction amount Δt by predicting the phase change trend of the next N switching cycles. adv :

[0075]

[0076] Among them, k p 、k d is the feedforward compensation coefficient, obtained through offline system identification; is the phase angle acceleration of the i-th node; Preferably, the prediction algorithm adopts a third-order Taylor expansion model and uses historical phase angle data to fit the instantaneous angular acceleration.

[0077] The closed-loop feedback control is realized by setting the rising edge time t actual Compared with the theoretical calculated value t theory Deviation Δt=t actual -t theory The input is a proportional-integral regulator, and the output is used to dynamically adjust the resonant network parameter update rate. The regulator transfer function is:

[0078]

[0079] Among them, K p , K i is the regulator parameter; τ is the system response delay time, which is determined by loop stability analysis.

[0080] Preferably, the timestamp calibration process is implemented in the timing control unit of the FPGA, and a high-precision digital delay line is used to adjust the pulse signal to the sub-nanosecond level. The duty cycle of the driving pulse is based on the rate of change of the magnetic flux. Dynamic adjustment to meet:

[0081]

[0082] Where D0 is the reference duty cycle; β is the flux coupling coefficient, which is calibrated through the core saturation characteristic experiment.

[0083] Furthermore, the data interaction between the closed-loop feedback control and the resonant network parameter update is realized through a shared memory pool, and a double buffer mechanism is used to avoid read and write conflicts. After each parameter update, the feedforward compensation coefficient k is dynamically optimized by comparing the phase consistency errors of the adjacent node drive pulses. p 、k d , ensuring that the synchronization accuracy of the entire network remains within the preset threshold.

[0084] S4. Based on the timing characteristics of the synchronous drive pulse signal and the rate of change of the drain current, iteratively update the resonant network parameters using a quasi-Newton method to maintain a zero voltage switching state under different load conditions;

[0085] In this embodiment, when the quasi-Newton method is used to iteratively update the resonant network parameters, the optimization objective function F(x) is defined as the weighted sum of the efficiency index, the electromagnetic interference index and the timing synchronization index, where the parameter vector x={L eq ,C total ,Q} includes equivalent inductance, total capacitance and quality factor. The specific expression of the objective function is:

[0086] F(x)=α·η(x)+β·||EMI(x)||2+γ·Δt sync (x);

[0087] Where α, β, and γ are weight coefficients; η is the calculated conversion efficiency; ||EMI||2 represents the L2 norm of the electromagnetic interference spectrum; Δt sync is the synchronization time deviation between nodes.

[0088] efficiency gradient The calculation is achieved through the perturbation observation method. While keeping the switching frequency unchanged, a small perturbation δx is applied to the resonant cavity parameters and the efficiency change Δη is measured:

[0089]

[0090] Electromagnetic interference gradient The acquisition uses a near-field probe array to sample the spatial radiation field distribution, and extracts the energy change rate of the characteristic frequency band through fast Fourier transform. It is calculated by the rising edge time difference of the driving pulses of adjacent nodes.

[0091] During the inverse operation of the Jacobian matrix, the Hessian matrix approximation matrix H is constructed k The BFGS update formula is used when:

[0092]

[0093] Among them, H k is the Hessian matrix approximation of the kth iteration; s k =xk+1 -x k is the parameter update vector, is the gradient change vector. The matrix inversion operation is performed through Cholesky decomposition to ensure numerical stability.

[0094] When dynamically adjusting the resonant cavity quality factor Q, establish a constant with the load current change rate. Feedback relationship:

[0095]

[0096] Where Q0 is the reference quality factor; κ is the adjustment amplitude coefficient; ξ is the slope control factor, which is calibrated by the load transient response test. The closed-loop feedback control is implemented by adjusting the resonant network damping resistance value through a digital potentiometer to ensure that the zero voltage switching condition V is maintained when the load suddenly changes. ds (t switch )=0.

[0097] Preferably, the iterative step size of the quasi-Newton method is determined using the Armijo line search strategy, satisfying the sufficient descent condition:

[0098]

[0099] in, is the search direction; c∈(0,1) is the attenuation coefficient; λ k The iteration termination condition is set as the rate of change of the objective function for three consecutive iterations is less than the preset threshold.

[0100] Furthermore, the parameter update process forms a data exchange channel with the spatiotemporal synchronization control module, triggering an emergency correction mechanism for the synchronization pulse phase angle when the quality factor adjustment is detected to exceed a safety threshold. The safety threshold is determined based on the maximum allowable current stress of the resonant cavity and the thermal design margin.

[0101] S5. Establish a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, and obtain the optimal switching trajectory command by solving the Hamiltonian equation;

[0102] In this embodiment, when establishing the control model, the optimization objective function J is defined as a weighted combination of the time integral of the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm:

[0103]

[0104] Among them, t0 is the optimization start time; is the drain-source voltage change rate; λ is the weight coefficient; represents Fourier transform; ||·|| ∞The objective function constrains the switching transient characteristics in both the time and frequency domains to ensure soft switching conditions and electromagnetic compatibility requirements.

[0105] When constructing the Hamiltonian equation, the state variable x=[V ds ,I d ,Φ] T And the control variable u(t) (drive pulse duty cycle), and additional switching trajectory quadratic diffusibility constraints:

[0106]

[0107] Among them, K max The upper limit of the acceleration of the control variable is determined by the safe working area of ​​the power device. Hamiltonian function Defined as:

[0108]

[0109] Among them, p is the adjoint variable vector; f(x,u) is the system state equation.

[0110] When solving with the adjoint variable method, first establish the adjoint equation:

[0111]

[0112] The boundary condition is processed by the virtual terminal method, at the end point of the time domain t f The extended virtual state satisfies p(t f )=0. In the solution process, the update of the accompanying variables and the iteration of the resonant network parameters form data coupling, specifically by exchanging the quality factor Q and equivalent inductance L in the shared storage area eq Real-time value of .

[0113] Preferably, the generation of the optimal switching trajectory instruction uses the spectral method to discretize the Hamiltonian equation and transform the continuous time problem into a nonlinear programming problem. k =kΔt, the Legendre polynomial is used to approximate the control variable u(t), and the KKT condition is solved by the interior point method to obtain the optimal solution.

[0114] The specific implementation of the data coupling mechanism is: after each resonance parameter update, the Jacobian matrix of the system state equation f(x,u) is recalculated And update the coefficient matrix of the adjoint equation. At the same time, the optimal switching trajectory instruction u * The derivative information of (t) is fed back to the spatiotemporal synchronization control module to adjust the rising edge slope of the driving pulse.

[0115] Furthermore, the electromagnetic interference spectrum norm is calculated using a windowed FFT technique, with a Blackman-Harris window selected as the window function to suppress spectrum leakage. The frequency band is primarily monitored within the odd harmonics of the switching frequency, and the infinite norm is calculated after extracting characteristic frequency components through bandpass filtering.

[0116] Preferably, the quadratic differentiability constraint is implemented by introducing a slack variable ξ(t) to transform the original constraint into:

[0117]

[0118] Among them, u(t) is the control variable.

[0119] The constraint conditions are then integrated into the objective function through the barrier function method to form an unconstrained optimization problem. During the iterative solution process, an adaptive adjustment strategy of the barrier parameters is adopted to balance the convergence speed and calculation accuracy.

[0120] The time-domain waveform of the optimal switching trajectory command is generated through a combination of a lookup table and real-time interpolation, with waveform storage and DAC output control implemented in the FPGA. The command update cycle and the resonant parameter adjustment cycle are kept as integer multiples, ensuring consistency in the system's dynamic response.

[0121] S6. Based on the real-time monitored junction temperature data and load current change rate, dynamically allocate weight coefficients for efficiency optimization and electromagnetic suppression, suppress abnormal propagation through an error energy attenuation function, and perform tensor product co-evolution of resonance parameters and synchronization parameters;

[0122] In this embodiment, when dynamically allocating efficiency optimization and electromagnetic suppression weight coefficients, a combination function is constructed:

[0123] w eff (t)=σ(α·T j (t))·cos(2πf m t);

[0124] w emi (t)=1-w eff (t);

[0125] Among them, w eff (t) is the efficiency optimization weight coefficient; σ(·) is the sigmoid function; α is the junction temperature sensitivity coefficient; f m is the modulation frequency, which is the same as the system switching frequency f sw Satisfy f m =f sw / N (N is an integer); T j (t) is the real-time junction temperature. The junction temperature data T j(t) From the online monitoring system of the power device, it is mapped to the weight domain through the temperature-efficiency transfer function.

[0126] When constructing the composite error energy attenuation function, the time-frequency domain joint processing function is defined:

[0127]

[0128] Where E(t) is the error energy at time t; ∈(τ) is the time domain error energy; G(t;σ) is the Gaussian kernel function, and its bandwidth parameter is (k is the proportional coefficient); σ is the Gaussian kernel bandwidth parameter; B(f) is the frequency domain band-limiting function:

[0129]

[0130] Among them, f max The convolution operation is implemented by the cooperation of analog integrator circuit and digital FIR filter to suppress the propagation of abnormal oscillation caused by sudden load change.

[0131] When performing tensor product coevolution, define the resonance parameter vector R = [L eq ,C total ,Q] and synchronization parameter vector S=[θ,Δt adv ,k p ]’s tensor product space:

[0132]

[0133] Where R is the resonance parameter vector; S is the synchronization parameter vector; is the tensor product operator, generating a 9×1 parameter space vector.

[0134] Construct the node coupling matrix M to reflect the parameter correlation strength, and the matrix element M ij Determined by the electrical parameter correlation coefficient of adjacent nodes:

[0135]

[0136] Where Cov represents covariance and σ represents standard deviation. The symbiotic evolution rules include:

[0137] Energy conservation constraints:

[0138]

[0139] Phase continuity condition:

[0140] |θ i (t+Δt)-θ i (t)|≤Δθ max ;

[0141] Among them, δP max is the allowable power deviation threshold; Δθ max is the maximum phase jump angle.

[0142] Preferably, the co-evolution process adopts a genetic algorithm framework and defines a fitness function:

[0143]

[0144] Among them, T opt is the historical optimal parameter tensor. The mutation operation is performed under the coupling matrix constraint to ensure that the parameter update direction descends along the energy loss gradient.

[0145] The frequency domain band-limiting characteristic is achieved by cascading multiple elliptic filters to form a filter bank, and the stopband attenuation of each filter is not less than 60dB. Preferably, the bandwidth parameter σ of the Gaussian kernel function is dynamically adjusted according to the real-time switching frequency to meet σ·f sw =C (C is a constant), ensuring that the time-frequency resolution product is constant.

[0146] Furthermore, parameter space mapping is implemented through a self-organizing feature mapping neural network, projecting the high-dimensional tensor space onto a two-dimensional grid to visually monitor parameter evolution. When a parameter combination approaches a stability boundary, a parameter rollback mechanism is triggered, loading the historical safety configuration.

[0147] The symbiotic evolution rules are implemented using a constrained optimization algorithm, verifying energy conservation and phase continuity after each parameter update. Candidate solutions that violate the constraints are projected onto the boundaries of the feasible region, while the number of violations is recorded to adjust the learning rate of the evolution strategy.

[0148] See also Figure 2 The present invention also provides a soft switch of a high-frequency switching transformer and a distributed control system thereof, wherein the system includes the following modules:

[0149] A network initialization module, which configures a distributed control network for building a ring topology and loads the improved precision time protocol;

[0150] Distributed control nodes are deployed in a ring topology, with redundant communication links between nodes forming bidirectional data channels. An improved time synchronization mechanism is integrated into the underlying protocol stack to support dynamic compensation for phase deviation and clock drift correction. Upon node startup, a resonant cavity parameter scan is automatically performed to establish an initial inductor-capacitor matching table, and master and slave node roles are determined through a negotiation algorithm.

[0151] A multi-physical quantity monitoring module is configured to collect real-time data on drain-source voltage, drain current, and load current change rate, while simultaneously monitoring power device junction temperature data and calculating core magnetic flux changes.

[0152] An integrated high-precision sensor array, including isolated voltage probes, zero-flux current sensors, and infrared thermal imaging units, enables simultaneous acquisition of electrical and thermal parameters. A built-in magnetic flux calculation engine uses an improved integration algorithm to eliminate measurement deviations caused by core loss and outputs a real-time magnetic flux rate-of-change data stream. An adaptive filtering unit dynamically adjusts the sampling frequency and signal bandwidth based on operating conditions.

[0153] a spatiotemporal synchronization control module configured to solve the second-order partial derivatives of the phase angle between nodes and generate a synchronous driving pulse signal for time stamp calibration;

[0154] A phase angle processing unit is deployed based on a distributed computing architecture, building a global phase relationship matrix through data exchange between adjacent nodes. The drive pulse generation unit integrates a programmable delay line, supports sub-nanosecond timestamp calibration, and features dual-mode operation with feedforward compensation and feedback regulation. An impedance matching network is configured at the synchronization signal output to ensure signal integrity throughout the power stage.

[0155] a resonant parameter adjustment module configured to iteratively update the resonant network parameters based on a quasi-Newton method to maintain a zero voltage switching state;

[0156] A built-in parameter optimization engine uses a hybrid optimization algorithm to adjust the equivalent values ​​of resonant network components in real time. A parameter safety monitoring unit triggers a rapid correction mechanism when it detects that the resonance point offset exceeds the threshold. This system works in conjunction with the thermal management module to dynamically adjust the upper quality factor limit based on the heat sink temperature to prevent component overstress damage.

[0157] An electromagnetic optimization decision module is configured to establish a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, and to generate an optimal switching trajectory instruction by solving the Hamiltonian equation;

[0158] An integrated multi-objective optimization solver builds a joint time-frequency domain analysis model. The switch trajectory planning unit employs a model predictive control framework, combining historical data with real-time state predictions to generate the optimal actuation sequence. A spectrum analysis front-end is configured to capture near-field radiation signatures using a multi-channel data acquisition card, providing feedback input for the optimization model.

[0159] A dynamic adjustment module configures weight coefficients for dynamically allocating efficiency optimization and electromagnetic suppression, and suppresses abnormal propagation through a composite error energy attenuation function;

[0160] The weight allocation unit utilizes a nonlinear function generator to dynamically adjust optimization target priorities based on operational characteristics. The anomaly suppression channel utilizes a multi-stage filtering network, including a time-domain convolution kernel and a frequency-domain limiter, to achieve broadband interference suppression. A built-in self-diagnostic function identifies oscillation patterns and automatically switches suppression strategies.

[0161] A parameter evolution module, configured to perform tensor product coevolution of resonance parameters and synchronization parameters, achieving multi-parameter space mapping and symbiotic evolution;

[0162] Deploy a co-evolutionary algorithm framework and establish a multi-dimensional parameter space mapping relationship library. The evolutionary process is supervised by a stability constraint controller to ensure that parameter updates conform to the system dynamics. Configure an evolution history database to support rapid retrieval and loading of optimal parameter combinations, enhancing the system's anti-disturbance capabilities.

[0163] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A soft switch of a high-frequency switching transformer and a distributed control method thereof, characterized in that: The following steps are involved: S1. Build a distributed control network with a ring topology, load the improved precise time protocol, initialize the equivalent inductance and capacitance parameters of the resonant cavity, and establish a clock deviation compensation matrix between nodes; S2. Use voltage sensors and current sensors to collect the drain-source voltage, drain current, and load current change rate of each node in real time, simultaneously monitor the junction temperature data of the power device, and calculate the core magnetic flux change based on the voltage integration method; S3. Solve the second-order partial derivative of the phase angle between nodes based on the load current change rate and the magnetic flux calculation results, and generate a synchronous drive pulse signal with time stamp calibration; S4. Based on the timing characteristics of the synchronous drive pulse signal and the rate of change of the drain current, iteratively update the resonant network parameters using a quasi-Newton method to maintain a zero voltage switching state under different load conditions; S5. Establish a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, and obtain the optimal switching trajectory command by solving the Hamiltonian equation; S6. Based on the real-time monitored junction temperature data and load current change rate, dynamically allocate weight coefficients for efficiency optimization and electromagnetic suppression, suppress abnormal propagation through the error energy attenuation function, and perform tensor product co-evolution of resonance parameters and synchronization parameters.

2. The soft switching and distributed control method of the high-frequency switching transformer according to claim 1, characterized in that: The step S1 includes: constructing a distributed control network with a ring topology, loading an improved precise time protocol including a phase angle differential compensation mechanism, initializing a set of equivalent inductance and capacitance parameters of the resonant cavity, and establishing a clock deviation compensation matrix based on the second-order mixed partial derivatives of the phase angle between nodes, wherein the construction process of the compensation matrix includes a cross-differentiation operation on the phase change rates of adjacent nodes.

3. The soft switching and distributed control method of the high-frequency switching transformer according to claim 1, characterized in that: The step S2 includes: collecting the instantaneous value of the drain-source voltage of each node through a voltage sensor, obtaining the drain current and the load current change rate through a current sensor, synchronously monitoring the real-time junction temperature data of the power device, and calculating the core magnetic flux change based on the voltage integration method, wherein the calculation process of the voltage integration method includes a compensation term for the equivalent resistance of the magnetic core, and the calculation result of the magnetic flux change is in a nonlinear integral relationship with the drain-source voltage sampling value.

4. The soft switching and distributed control method of the high-frequency switching transformer according to claim 3, characterized in that: The step S2 also includes: establishing a dynamic coupling correction model of junction temperature monitoring data and magnetic flux estimation value, and using a sliding time window algorithm to perform time domain smoothing on the load current change rate, wherein the smoothing process includes a weighted average operation of the current sampling value and a mutation suppression mechanism.

5. The soft switching and distributed control method of the high-frequency switching transformer according to claim 1, characterized in that: The step S3 includes: solving the second-order mixed partial derivative of the phase angle between nodes based on the load current change rate and the magnetic flux calculation results, wherein the calculation process of the partial derivative introduces a dynamic weight factor of the load current change rate, and generates a synchronous drive pulse signal with time stamp calibration, and the time stamp calibration includes a phase advance correction based on a feedforward compensation mechanism. The generation process of the drive pulse signal and the update of the resonant network parameters form a closed-loop feedback control.

6. The soft switching and distributed control method of the high-frequency switching transformer according to claim 1, characterized in that: The step S4 includes: based on the timing characteristics of the synchronous drive pulse signal and the rate of change of the drain current, using the quasi-Newton method to iteratively update the parameters of the resonant network, the update process comprehensively calculates the efficiency gradient, the electromagnetic interference gradient and the time synchronization gradient, and performs parameter correction through the inverse operation of the Jacobian matrix, and dynamically adjusts the quality factor of the resonant cavity to maintain the zero voltage switching state, wherein the adjustment of the quality factor and the load current change rate form a closed-loop feedback control.

7. The soft switching and distributed control method of the high-frequency switching transformer according to claim 1, characterized in that: The step S5 includes: establishing a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, constructing a Hamiltonian equation containing the quadratic differentiability constraint of the switching trajectory, and using the adjoint variable method to process the boundary conditions to solve the optimal switching trajectory instruction, wherein the calculation process of the adjoint variable is data-coupled with the update of the resonant network parameters, and the generation of the optimal switching trajectory instruction is synchronously fed back to the timing adjustment module of the synchronous drive pulse.

8. The soft switching and distributed control method of the high-frequency switching transformer according to claim 1, characterized in that: The step S6 includes: based on the real-time monitored junction temperature data and the load current change rate, using a sigmoid function and a cosine modulation function to combine and calculate the weight coefficients of dynamically allocating efficiency optimization and electromagnetic suppression, constructing a composite error energy attenuation function including a Gaussian kernel function and a frequency domain band limiting characteristic, and suppressing abnormal propagation through a convolution operation, wherein the bandwidth parameter of the Gaussian kernel function is positively correlated with the system switching frequency, and the frequency domain band limiting characteristic is determined by the maximum operating frequency of the system.

9. The soft switching and distributed control method of the high-frequency switching transformer according to claim 1, characterized in that: The step S6 also includes: performing tensor product co-evolution of resonance parameters and synchronization parameters, establishing a parameter space mapping relationship based on the node coupling matrix, and defining a symbiotic evolution rule including energy conservation constraints and phase continuity conditions, wherein the construction of the coupling matrix reflects the correlation strength of the electrical parameters of adjacent nodes, and the evolution rule ensures that the parameter update process maintains the system stability boundary.

10. A soft switch of a high-frequency switching transformer and a distributed control system thereof, applied to the method according to any one of claims 1 to 9, characterized in that: The system includes the following modules: A network initialization module, which configures a distributed control network for building a ring topology and loads the improved precision time protocol; A multi-physical quantity monitoring module is configured to collect real-time data on drain-source voltage, drain current, and load current change rate, while simultaneously monitoring power device junction temperature data and calculating core magnetic flux changes. a spatiotemporal synchronization control module configured to solve the second-order partial derivatives of the phase angle between nodes and generate a synchronous driving pulse signal for time stamp calibration; a resonant parameter adjustment module configured to iteratively update the resonant network parameters based on a quasi-Newton method to maintain a zero voltage switching state; An electromagnetic optimization decision module is configured to establish a control model with the square of the drain-source voltage change rate and the electromagnetic interference spectrum norm as optimization targets, and to generate an optimal switching trajectory instruction by solving the Hamiltonian equation; A dynamic adjustment module configures weight coefficients for dynamically allocating efficiency optimization and electromagnetic suppression, and suppresses abnormal propagation through a composite error energy attenuation function; The parameter evolution module is configured to perform tensor product co-evolution of resonance parameters and synchronization parameters to achieve multi-parameter space mapping and symbiotic evolution.

Images