A high-frequency transformer energy scheduling and load balancing method and system

Abstract

The present application relates to the technical field of transformer control, in particular to a high-frequency transformer energy scheduling and load balancing method and system. The present application extracts input energy supply, bus voltage, load current and branch carrying state in the same control cycle, first identifies the limited working condition of the coincidence of input energy contraction and load demand increase, then calculates the upper limit of the total distributable current increment of the transformer, so that the current distribution is constrained by the overall schedulable energy boundary. Then, the current increment is distributed in combination with the branch temperature, equivalent impedance and carrying margin, and the distribution result is corrected according to the actual response slope deviation after the driving signal is issued, the abnormal branch is eliminated, and the compensation amount exceeding the current carrying capacity is delayed, thereby reducing the excessive carrying of local branches, improving the consistency of load sharing of parallel branches, and enhancing the stability of scheduling under limited working conditions.

Description

Technical Field

[0001] This invention relates to the field of transformer control technology, and in particular to a method and system for energy dispatching and load balancing of high-frequency transformers. Background Technology

[0002] The high-frequency transformer energy dispatch and load balancing method refers to the method for controlling and managing the energy distribution and load-bearing relationships among the input, output, and load sides of transformers operating in high-frequency power conversion scenarios.

[0003] Traditional high-frequency transformer energy dispatch and load balancing methods, while capable of conventionally controlling energy transfer based on input power, output power, or load current, are prone to problems when photovoltaic input power supply rapidly declines, DC bus voltage drops synchronously, and load-side current demand continues to rise. Relying solely on fixed allocation ratios or static rated current boundaries can easily lead to a disconnect between the dispatch criteria and actual operating conditions. Specifically, input-side energy contraction reduces the dispatchable power space, while increased load-side demand drives further current increases in parallel branches. Different parallel branches differ in temperature, equivalent resistance, equivalent leakage inductance, and current response speed. Traditional methods struggle to promptly distinguish which branches still have carrying capacity, which are nearing their carrying capacity limits, and which have failed to respond as expected after the drive signal is issued. Therefore, under constrained operating conditions, some branches may continue to receive large current increments, while the actual response deviations of other branches may be ignored, resulting in a discrepancy between the load balancing results and the real-time carrying capacity of each parallel branch. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a high-frequency transformer energy dispatching and load balancing method and system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a high-frequency transformer energy dispatching and load balancing method, comprising the following steps: Obtain the set of transformer operating parameters within the control cycle of transformer energy dispatch; Each transformer operating characteristic is determined based on the set of transformer operating parameters and integrated into a set of transformer operating characteristics. Based on the set of transformer operating characteristics, it is determined whether the transformer is in a restricted operating condition. When it is determined to be in a restricted operating condition, the upper limit of the total distributable current increment of the transformer is determined according to the set of transformer operating parameters. Based on the set of transformer operating parameters and the set of transformer operating characteristics, the dynamic carrying capacity of each parallel branch of the transformer is determined, and the current increment of each parallel branch is calculated in combination with the upper limit of the total distributable current increment of the transformer. The drive signal for each parallel branch is generated by referring to the current increment of each parallel branch and the transformer operating characteristic set. The actual current slope and current ramp-up slope of each parallel branch are calculated after the drive signal is issued. The actual response slope deviation of each parallel branch is calculated based on the actual current slope and current ramp-up slope of each parallel branch. The current increment of each parallel branch is adjusted in combination with the dynamic load margin of each parallel branch to obtain the load balancing scheduling result.

[0006] As a further aspect of the present invention, the set of transformer operating parameters within the control cycle of transformer energy dispatch includes: The transformer's electrical parameters are obtained by acquiring the output voltage and current of the photovoltaic array, the instantaneous voltage of the DC bus, the total load current, and the actual current and driving voltage of each parallel branch of the transformer during the control cycle of the transformer energy dispatch. The local temperature of each parallel branch of the transformer is collected during the control cycle to obtain the transformer temperature parameters; By integrating all the transformer electrical parameters and the transformer temperature parameters, a set of transformer operating parameters is obtained.

[0007] As a further aspect of the present invention, the step of determining each transformer operating characteristic based on the transformer operating parameter set and integrating them into a transformer operating characteristic set includes: The output power of the photovoltaic array is obtained by multiplying the output voltage and output current of the photovoltaic array in the transformer operating parameter set. The sampling time before the start of the control cycle is obtained as the data before the control cycle, and the sampling time after the end of the control cycle is obtained as the data after the control cycle. The power difference of the photovoltaic array output power before and after the control cycle is calculated, and the ratio of the power difference to the control cycle is calculated to obtain the photovoltaic array output power change rate. The product of the instantaneous voltage of the transformer DC bus and the total load current of the transformer is calculated to obtain the total active power. The voltage difference of the instantaneous voltage of the transformer DC bus before and after the control cycle is calculated, and the ratio of the voltage difference to the control cycle is calculated to obtain the instantaneous derivative of the DC bus voltage. The current difference of the total load current before and after the control cycle is calculated, and the ratio of the current difference to the control cycle is calculated to obtain the instantaneous rise slope of the total load current. The jump difference of the total active power before and after the control cycle is calculated to obtain the jump amplitude of the total active power. Referring to the set of transformer operating parameters, calculate the temperature difference between the preset core reference saturation temperature and the local temperature of each parallel branch, and calculate the ratio of the temperature difference to the preset reference saturation temperature difference to obtain the temperature correction coefficient of each parallel branch. Calculate the difference between the actual current flowing through each parallel branch before and after the control cycle, and calculate the ratio of the branch current difference to the control cycle to obtain the instantaneous rate of change of current for each parallel branch. An equivalent circuit model of a transformer branch is established with preset initial leakage inductance parameters and preset initial resistance parameters. The instantaneous rate of change of current, the actual current flowing through each parallel branch, and the driving voltage of each parallel branch are substituted into the equivalent circuit model of the transformer branch. The preset initial leakage inductance parameters and preset initial resistance parameters are combined to perform parameter fitting calculation on the equivalent circuit model of the transformer branch to obtain the basic value of leakage inductance and the basic value of resistance for each parallel branch. The product of the basic value of leakage inductance and the corresponding temperature correction coefficient is calculated to obtain the equivalent leakage inductance for each parallel branch. The product of the basic value of resistance and the corresponding temperature correction coefficient is calculated to obtain the equivalent resistance for each parallel branch. The output power of the photovoltaic array, the rate of change of the output power of the photovoltaic array, the instantaneous derivative of the DC bus voltage, the instantaneous rise slope of the total load current, the jump amplitude of the total active power, and the equivalent leakage inductance and equivalent resistance of each parallel branch are integrated as each transformer operating characteristic to construct a transformer operating characteristic set.

[0008] As a further aspect of the present invention, the step of determining whether the transformer is in a restricted operating condition based on the transformer operating characteristic set, and determining the upper limit of the total distributable current increment of the transformer based on the transformer operating parameter set when the condition is determined to be a restricted operating condition, includes: When the photovoltaic array output power change rate in the transformer operating characteristic set is less than the preset photovoltaic power contraction threshold, the instantaneous derivative of the DC bus voltage is less than the preset bus voltage drop threshold, the instantaneous rise slope of the total load current is greater than the preset load current rise threshold, and the jump amplitude of the total active power is greater than the preset power jump threshold, the transformer is determined to be in a restricted operating condition where the input energy contraction coincides with the increase in load demand. When the transformer is under the restricted operating conditions, the transformer bus capacitance is obtained, and the product of the photovoltaic array output power change rate and the control cycle in the transformer operating characteristic set is calculated to obtain the photovoltaic array output power change. The sum of the photovoltaic array output power change and the photovoltaic array output power is calculated to obtain the photovoltaic array predicted injection power. At the same time, the product of the square of the instantaneous voltage of the transformer DC bus, the transformer bus capacitance, and the preset energy coefficient in the transformer operating parameter set is calculated to obtain the bus instantaneous capacitance energy. The ratio of the bus instantaneous capacitance energy to the control cycle is calculated to obtain the bus instantaneous discharge power. The sum of the bus instantaneous discharge power and the photovoltaic array predicted injection power is calculated to obtain the total dispatchable predicted power. The ratio of the total dispatchable predicted power to the instantaneous voltage of the transformer DC bus is calculated to obtain the total dispatchable predicted current. The current difference between the total dispatchable predicted current and the total transformer load current is calculated as the upper limit of the total distributable current increment of the transformer.

[0009] As a further aspect of the present invention, the step of determining the dynamic carrying capacity margin of each parallel branch of the transformer based on the transformer operating parameter set and the transformer operating characteristic set, and calculating the current increment of each parallel branch in conjunction with the upper limit of the total distributable current increment of the transformer, includes: Obtain the upper limit of the rated current of each parallel branch of the transformer, calculate the ratio of the equivalent resistance to the equivalent leakage inductance of each parallel branch in the transformer operating characteristic set, obtain the transient current increase sensitivity of each parallel branch, and calculate the current difference between the upper limit of the rated current of each parallel branch and the actual current flowing through each parallel branch in the transformer operating parameter set, as the basic bearing margin of each parallel branch; The ratio of the preset derating reference temperature to the local temperature of each parallel branch in the transformer operating parameter set is calculated as the temperature derating compensation coefficient for each parallel branch. The product of the basic bearing margin of each parallel branch and the temperature derating compensation coefficient is calculated to obtain the dynamic bearing margin of each parallel branch. The transient current increase sensitivity and dynamic bearing margin of each parallel branch are normalized to obtain the normalized transient current increase sensitivity and dynamic bearing margin of each parallel branch. Calculate the product of the normalized transient current surge sensitivity and the normalized dynamic carrying capacity of each parallel branch to obtain the comprehensive evaluation value of each parallel branch; calculate the sum of the comprehensive evaluation values ​​of all parallel branches to obtain the total comprehensive evaluation value of the branches; calculate the ratio of the comprehensive evaluation value of each parallel branch to the total comprehensive evaluation value of the branches, and use it as the comprehensive allocation weight coefficient of each parallel branch; calculate the product of the comprehensive allocation weight coefficient of each parallel branch and the upper limit of the total distributable current increment of the transformer to obtain the current increment of each parallel branch.

[0010] As a further aspect of the present invention, the step of generating a drive signal for each parallel branch by referring to the current increment of each parallel branch and the transformer operating characteristic set, and calculating the actual current slope and current ramp-up slope of each parallel branch after issuing the drive signal includes: Based on the current increment of each parallel branch and the equivalent leakage inductance and equivalent resistance of each parallel branch in the transformer operating characteristic set, a drive signal for each parallel branch is generated. After the monitoring transformer sends the drive signal for each parallel branch, the actual current flowing through each parallel branch of the transformer within the control cycle is calculated. The difference in the actual current flow before and after the control cycle is calculated. The ratio of the difference in the current flow to the control cycle is calculated to obtain the actual current slope of each parallel branch. Calculate the ratio of the current increment to the control cycle for each parallel branch to obtain the current ramp-up slope for each parallel branch.

[0011] As a further aspect of the present invention, the step of calculating the actual response slope deviation of each parallel branch based on the actual current slope and current ramp-up slope of each parallel branch, and adjusting the current increment of each parallel branch in conjunction with the dynamic load margin of each parallel branch to obtain the load balancing scheduling result includes: Calculate the slope difference between the actual current slope of each parallel branch and the current ramp-up slope of each parallel branch to obtain the actual response slope deviation of each parallel branch. Calculate the product of the actual response slope deviation and the control period to obtain the branch deviation current of each parallel branch. When the actual response slope deviation is less than the preset health response tolerance threshold, each parallel branch is judged to have a normal response and is marked as a healthy branch. When the actual response slope deviation is greater than or equal to the health response tolerance threshold, each parallel branch is judged to have an abnormal response and is marked as an abnormal branch for removal. For the selected healthy branches, the normalized dynamic carrying margin of each parallel branch is extracted as the dynamic carrying margin of the healthy branch. The maximum current increment step of the healthy branch within the control cycle is calculated in combination with the dynamic carrying margin of the healthy branch. The sum of the dynamic carrying margins of all healthy branches is calculated to obtain the total healthy dynamic carrying margin. The margin ratio of the dynamic carrying margin of the healthy branches to the total healthy dynamic carrying margin is calculated as the theoretical compensation share coefficient. The product of the branch deviation current of each parallel branch corresponding to the healthy branch and the theoretical compensation share coefficient is calculated to obtain the theoretical compensation current. When the theoretical compensation current is greater than the maximum current increment step of the healthy branch, the compensation difference between the theoretical compensation current and the maximum current increment step of the healthy branch is calculated to obtain the remaining compensation current, which is then stored in the transformer's remaining compensation delay queue. When the theoretical compensation current is less than or equal to the maximum current increment step of the healthy branch, it is determined that there is no remaining compensation current, and the operation of storing data in the transformer's remaining compensation delay queue is stopped. Based on the actual response slope deviation of all the parallel branches and the transformer's remaining compensation delay queue, the current increment of each parallel branch is uniformly adjusted to obtain the load balancing scheduling result. This result is used to dynamically allocate the carrying current boundary of each parallel branch when the high-frequency transformer faces the condition of input energy contraction and load demand increase.

[0012] A high-frequency transformer energy dispatch and load balancing system, the system comprising: The parameter acquisition module obtains the set of transformer operating parameters within the control cycle of transformer energy dispatch; The feature extraction module determines each transformer operation feature based on the set of transformer operation parameters and integrates them into a set of transformer operation features. The incremental determination module determines whether the transformer is in a restricted operating condition based on the transformer operating characteristic set. When it is determined to be in a restricted operating condition, it determines the upper limit of the total distributable current increment of the transformer based on the transformer operating parameter set. The branch dispatch module determines the dynamic carrying capacity of each parallel branch of the transformer based on the transformer operating parameter set and transformer operating characteristic set, and calculates the current increment of each parallel branch in combination with the upper limit of the total distributable current increment of the transformer. The slope calculation module generates a drive signal for each parallel branch by referring to the current increment of each parallel branch and the transformer operating characteristic set, and calculates the actual current slope and current ramp-up slope of each parallel branch after the drive signal is issued. The load balancing scheduling module calculates the actual response slope deviation of each parallel branch based on the actual current slope and current ramp-up slope of each parallel branch, and adjusts the current increment of each parallel branch in combination with the dynamic load margin of each parallel branch to obtain the load balancing scheduling result.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are as follows: This invention extracts changes in input power supply, bus voltage, load current, and branch load status simultaneously within the same control cycle. It first determines whether the system has entered a constrained operating condition where input energy contraction coincides with increased load demand. Then, under this condition, it calculates the upper limit of the total distributable current increment of the transformer, ensuring that subsequent current allocation is constrained by the overall dispatchable energy boundary. Subsequently, this invention combines the temperature correction, equivalent impedance characteristics, and remaining load capacity of each parallel branch to obtain a dynamic load margin that more closely reflects the actual state of the branch, and allocates the current increment of each parallel branch accordingly. After the drive signal is issued, it identifies branch response deviations by comparing the actual current slope with the theoretical current ramp-up slope, eliminates branches with abnormal responses, and continues to compensate healthy branches according to their load margin. Compensation exceeding the current step size is stored in a delay queue. Therefore, this invention can dynamically constrain the total current increase when both insufficient input energy and increased load demand occur simultaneously, and adjust the current carrying capacity of each parallel branch according to changes in temperature, impedance, and response state, thereby reducing excessive load on local branches, improving the consistency of load sharing among parallel branches, and enhancing scheduling stability under constrained operating conditions. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the steps of the method of the present invention; Figure 2 This is a flowchart illustrating the steps and principles of the method of the present invention. Figure 3 This is a block diagram of the system of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] Please see Figure 1-2 This invention provides a technical solution, a method for energy dispatching and load balancing of high-frequency transformers, comprising the following steps: S1: Obtain the set of transformer operating parameters within the control cycle of transformer energy dispatch, including: The transformer's electrical parameters are obtained by acquiring the output voltage and current of the photovoltaic array, the instantaneous voltage of the DC bus, the total load current, and the actual current and driving voltage of each parallel branch of the transformer during the control cycle of the transformer energy dispatch. First, the output voltage of the photovoltaic array is collected by the voltage sampling device at the output end of the transformer photovoltaic array. The output current of the photovoltaic array is collected by the current sampling device at the output end of the transformer photovoltaic array. The instantaneous voltage of the transformer DC bus is collected by the transformer DC bus voltage sampling device. The total load current of the transformer is collected by the transformer load total current sampling device. The actual current flowing through each parallel branch of the transformer is collected by the current sampling device of each parallel branch according to the branch number. The drive voltage of each parallel branch of the transformer is collected by the drive voltage sampling device of each parallel branch according to the branch number. The control cycle consists of one sampling, calculation, and drive issuance time slice. The high-frequency transformer is a transformer that completes the conversion of input energy and load energy under high-frequency switching drive. The photovoltaic array of the transformer is a photovoltaic power generation branch that provides input power to the transformer DC bus. The transformer DC bus is a DC voltage node connecting the input side and the load side. Each parallel branch of the transformer is a conduction path that shares the total load current of the transformer. Each sampling device outputs sampled values ​​at the start and end of the control cycle, and at the sampling time preceding the start of the control cycle. These sampled values ​​are written into the sampling record according to the control cycle number, sampling time, sampled object name, branch number, sampled value, and unit of the physical quantity. Sampling records with the same control cycle number and sampling time are grouped into the same timing record. Similarly, the actual current flowing through each parallel branch of the transformer with the same branch number is grouped into the same branch record. After completing the aggregation of timing and branch records, the transformer's electrical parameters are obtained.

[0017] The local temperature of each parallel branch of the transformer is collected during the control cycle to obtain the transformer temperature parameters; Temperature sampling devices installed on each parallel branch of the transformer collect the local temperature of each parallel branch. The local temperature of each parallel branch is the temperature value of the sampled location within the control cycle. The sampling location corresponds to the area adjacent to the conductor or winding of each parallel branch. The temperature sampling devices output temperature values ​​according to the control cycle number, sampling time, and branch number. The temperature values ​​are mapped to the actual current and drive voltage of each parallel branch according to the branch number. When the control cycle number and sampling time are the same, the temperature value is assigned to the corresponding branch number. When the control cycle number is the same but the sampling time is different, the temperature value closest to the electrical sampling time within the same control cycle is assigned to the corresponding branch number. After completing the time-series mapping of temperature values ​​under all branch numbers, the transformer temperature parameters are obtained.

[0018] By integrating all transformer electrical parameters and transformer temperature parameters, a set of transformer operating parameters is obtained; Using the control cycle number as the primary index and the sampling time and branch number as corresponding indices, transformer electrical parameters and transformer temperature parameters are correlated within the same cycle. The output voltage and current of the transformer photovoltaic array, the instantaneous voltage of the transformer DC bus, and the total load current of the transformer are included in the common operation record. The actual current flowing through each parallel branch of the transformer, the driving voltage, and the local temperature are included in the branch operation record according to the branch number. The common operation record and the branch operation records together form the electrical and temperature quantities available for subsequent calculations within the same control cycle. When the control cycle number of the common operation record matches the control cycle number of the branch operation record, a corresponding cycle is established between the common operation record and the branch operation record; when the branch numbers in the branch operation records match, a corresponding branch is established for the actual current flowing through each parallel branch of the transformer. After completing the corresponding cycle between the common operation record and all branch operation records, the set of transformer operating parameters is obtained.

[0019] S2: Determine each transformer operating characteristic based on the transformer operating parameter set, and integrate them into a transformer operating characteristic set, including: The output power of the photovoltaic array is obtained by multiplying the output voltage and output current of the photovoltaic array in the transformer operating parameter set. The sampling time before the start of the control cycle is used as the data before the control cycle, and the sampling time after the end of the control cycle is used as the data after the control cycle. The power difference of the photovoltaic array output power before and after the control cycle is calculated, and the ratio of the power difference to the control cycle is calculated to obtain the photovoltaic array output power change rate. The product of the instantaneous voltage of the transformer DC bus and the total load current of the transformer is calculated to obtain the total active power. The voltage difference of the instantaneous voltage of the transformer DC bus before and after the control cycle is calculated, and the ratio of the voltage difference to the control cycle is calculated to obtain the instantaneous derivative of the DC bus voltage. The current difference of the total load current before and after the control cycle is calculated, and the ratio of the current difference to the control cycle is calculated to obtain the instantaneous rise slope of the total load current. The jump difference of the total active power before and after the control cycle is calculated to obtain the jump amplitude of the total active power. After the transformer operating parameter set is formed, the output voltage and current of the transformer photovoltaic array, the instantaneous voltage of the transformer DC bus, and the total load current of the transformer are read from the common operating record before and after the control cycle. The data before the control cycle is taken from the data corresponding to the previous sampling time at the start sampling time of the control cycle, and the data after the control cycle is taken from the data corresponding to the sampling time at the end sampling time of the control cycle. The output power of the photovoltaic array is the power output by the transformer photovoltaic array at the sampling time. The total active power is the power corresponding to the instantaneous voltage of the transformer DC bus and the total load current of the transformer. The instantaneous derivative of the DC bus voltage is the rate of change of the instantaneous voltage of the transformer DC bus within the control cycle. The instantaneous rise slope of the total load current is the rate of change of the total load current within the control cycle. The jump amplitude of the total active power is the change in the total active power before and after the control cycle. The output power of the photovoltaic array is calculated as follows: output voltage of the photovoltaic array × output current of the photovoltaic array = output power of the photovoltaic array; output power of the photovoltaic array after the control cycle - output power of the photovoltaic array before the control cycle = power difference; power difference / control cycle = rate of change of output power of the photovoltaic array; instantaneous voltage of the transformer DC bus × total load current of the transformer = total active power; instantaneous voltage of the transformer DC bus after the control cycle - instantaneous voltage of the transformer DC bus before the control cycle = voltage difference; voltage difference / control cycle = instantaneous derivative of DC bus voltage; total load current of the transformer after the control cycle - total load current of the transformer before the control cycle = current difference; current difference / control cycle = instantaneous rise slope of total load current; total active power after the control cycle - total active power before the control cycle = total active power jump amplitude. After completing the data calculations at the corresponding sampling times in the common operation record, the output power of the photovoltaic array, the rate of change of output power of the photovoltaic array, the instantaneous derivative of DC bus voltage, the instantaneous rise slope of total load current, and the jump amplitude of total active power are obtained.

[0020] Referring to the transformer operating parameter set, calculate the temperature difference between the preset core reference saturation temperature and the local temperature of each parallel branch, and calculate the ratio of the temperature difference to the preset reference saturation temperature difference to obtain the temperature correction coefficient of each parallel branch. Calculate the difference between the actual current flowing through each parallel branch before and after the control cycle, and calculate the ratio of the branch current difference to the control cycle to obtain the instantaneous rate of change of current for each parallel branch. The local temperature and actual current flowing through each parallel branch are read from the branch operation record. The preset core reference saturation temperature and preset reference saturation temperature difference are read from the controller parameter table. The preset core reference saturation temperature is the temperature reference value used for core temperature determination. The preset reference saturation temperature difference is the difference between the preset core reference saturation temperature and the core reference operating temperature. The temperature correction factor is the proportional value used to correct the parameters of subsequent branches based on the local temperature of each parallel branch. The process for setting the preset core reference saturation temperature is as follows: read the nominal core saturation temperature value from the core material specification and write it into the controller parameter table. The process for setting the preset reference saturation temperature difference is as follows: read the preset core reference saturation temperature and the core reference operating temperature from the controller parameter table. Preset core reference saturation temperature - core reference operating temperature = preset reference saturation temperature difference. Preset core reference saturation temperature - local temperature of each parallel branch = temperature difference; temperature difference / preset reference saturation temperature difference = temperature correction factor for each parallel branch. The closer the local temperature of each parallel branch is to the preset core reference saturation temperature, the smaller the temperature difference, and the smaller the temperature correction coefficient for each parallel branch. Conversely, the greater the difference between the local temperature of each parallel branch and the preset core reference saturation temperature, the larger the temperature difference, and the larger the temperature correction coefficient for each parallel branch. The actual current flowing through each parallel branch after the control cycle minus the actual current flowing through each parallel branch before the control cycle equals the branch current difference. The branch current difference / control cycle equals the instantaneous rate of change of current for each parallel branch. After completing the temperature correction and current change calculations corresponding to each branch number, the temperature correction coefficient and the instantaneous rate of change of current for each parallel branch are obtained.

[0021] An equivalent circuit model of the transformer branch is established with preset initial leakage inductance and initial resistance parameters. The instantaneous rate of change of current, the actual current flowing through each parallel branch, and the driving voltage of each parallel branch are substituted into the equivalent circuit model. The model is then fitted with the preset initial leakage inductance and initial resistance parameters to obtain the basic values ​​of leakage inductance and resistance for each parallel branch. The product of the basic leakage inductance value and the corresponding temperature correction coefficient is calculated to obtain the equivalent leakage inductance for each parallel branch, and the product of the basic resistance value and the corresponding temperature correction coefficient is calculated to obtain the equivalent resistance for each parallel branch. Specifically, the process of fitting the transformer branch equivalent circuit model to obtain the basic values ​​of leakage inductance and resistance for each parallel branch involves constructing a model containing the parameters of the leakage inductance and resistance to be optimized. The set to be optimized is determined, and the preset initial leakage inductance parameters and preset initial resistance parameters are assigned to the leakage inductance parameters and resistance parameters to be optimized, respectively. The product of the leakage inductance parameters to be optimized and the instantaneous rate of change of current in each parallel branch is calculated to obtain the branch inductive theoretical voltage. The product of the resistance parameters to be optimized and the actual current flowing through each parallel branch is calculated to obtain the branch resistive theoretical voltage. The sum of the branch inductive theoretical voltage and the branch resistive theoretical voltage is calculated to obtain the branch comprehensive estimated voltage. The difference between the branch comprehensive estimated voltage and the driving voltage of each parallel branch is calculated, and the leakage inductance parameters and resistance parameters to be optimized are updated synchronously according to the difference and the preset adjustment step size. When the difference is less than the preset fitting convergence threshold, the parameter update is stopped and the current leakage inductance parameters to be optimized are extracted as the basic value of leakage inductance for each parallel branch, and the current resistance parameters to be optimized are extracted as the basic value of resistance. In MATLAB / Simulink, an equivalent circuit model of the transformer branch is constructed. The equivalent circuit model of the transformer branch is a circuit expression that describes the correspondence between the driving voltage of each parallel branch, the instantaneous rate of change of current of each parallel branch, and the actual current flowing through each parallel branch. The set to be optimized is a set of parameters that store the leakage inductance parameter and the resistance parameter to be optimized. The parameter fitting operation is to successively correct the leakage inductance parameter and the resistance parameter to be optimized by estimating the voltage of the branch and the driving voltage of each parallel branch. During construction, the instantaneous rate of change of current, the actual current flowing through each parallel branch, and the driving voltage of each parallel branch are first set as input terminals. Then, the leakage inductance parameter to be optimized and the resistance parameter to be optimized are set as updateable parameters. Subsequently, an inductive voltage calculation node is set to receive the leakage inductance parameter to be optimized and the instantaneous rate of change of current of each parallel branch. A resistive voltage calculation node is set to receive the resistance parameter to be optimized and the actual current flowing through each parallel branch. A voltage sum calculation node is set to receive the theoretical inductive voltage and the theoretical resistive voltage of the branch. A difference calculation node is set to receive the comprehensive estimated voltage of the branch and the driving voltage of each parallel branch. A parameter update node is set to receive the difference and the preset adjustment step size. The preset initial leakage inductance parameter is the initial leakage inductance value pre-written in the controller parameter table, which is taken from the design value of the winding leakage inductance of each parallel branch of the transformer. The preset initial resistance parameter is the initial resistance value pre-written in the controller parameter table, which is taken from the design value of the conductor resistance of each parallel branch of the transformer. The preset adjustment step size is the amplitude of each synchronous update of the leakage inductance parameter and the resistance parameter to be optimized. When setting, the minimum resolvable voltage of the driving voltage sampling device of each parallel branch is read first, and then the update ratio in the controller parameter table is read. Minimum resolvable voltage × update ratio = preset adjustment step size. The preset fitting convergence threshold is the allowable voltage difference boundary between the branch comprehensive estimated voltage and the driving voltage of each parallel branch. When setting, the minimum resolvable voltage of the driving voltage sampling device of each parallel branch is written into the controller parameter table. After the set to be optimized is established, the preset initial leakage inductance parameter is written into the leakage inductance parameter to be optimized, and the preset initial resistance parameter is written into the resistance parameter to be optimized. The following parameters are used to calculate the optimal leakage inductance parameter multiplied by the instantaneous rate of change of current in each parallel branch: The theoretical inductive voltage of the branch is calculated as follows: The theoretical resistive voltage of the branch is calculated as follows: The theoretical inductive voltage of the branch plus the theoretical resistive voltage of the branch equals the comprehensive estimated voltage of the branch. The comprehensive estimated voltage of the branch minus the driving voltage of each parallel branch equals the difference. This difference, along with the preset adjustment step size, is updated synchronously. When the difference is positive, the parameters to be optimized (leakage inductance and resistance) are updated in the direction that decreases the comprehensive estimated voltage of the branch. When the difference is negative, the parameters to be optimized (leakage inductance and resistance) are updated in the direction that increases the comprehensive estimated voltage of the branch. When the difference is zero, the parameters to be optimized (leakage inductance and resistance) remain at their current values.When the difference is less than the preset fitting convergence threshold, the current leakage inductance parameter to be optimized is used as the base value of the leakage inductance for each parallel branch, and the current resistance parameter to be optimized is used as the base value of the resistance. The base value of leakage inductance × the corresponding temperature correction factor = the equivalent leakage inductance for each parallel branch; the base value of resistance × the corresponding temperature correction factor = the equivalent resistance for each parallel branch. After completing the construction of the equivalent circuit model of the transformer branch, parameter fitting calculation, and temperature correction, the equivalent leakage inductance and equivalent resistance of each parallel branch are obtained.

[0022] The output power of the photovoltaic array, the rate of change of the output power of the photovoltaic array, the instantaneous derivative of the DC bus voltage, the instantaneous rise slope of the total load current, the jump amplitude of the total active power, the equivalent leakage inductance and equivalent resistance of each parallel branch are integrated as each transformer operating characteristic to construct a transformer operating characteristic set. Common characteristic records and branch characteristic records are generated according to the control cycle number. The photovoltaic array output power, photovoltaic array output power change rate, instantaneous derivative of DC bus voltage, instantaneous rise slope of total load current, and total active power jump amplitude are written into the common characteristic record according to the control cycle number. The equivalent leakage inductance and equivalent resistance of each parallel branch are written into the branch characteristic record according to the branch number. The common characteristic record and the branch characteristic record together serve as characteristic data for subsequent operating condition determination and branch allocation. When the control cycle number of the common characteristic record matches the control cycle number of the branch characteristic record, a same-cycle correspondence is established between the common characteristic record and the branch characteristic record; when the branch numbers of the equivalent leakage inductance and equivalent resistance of each parallel branch match, a same-branch correspondence is established for each parallel branch. After completing the same-cycle correspondence between the common characteristic record and the branch characteristic record, the transformer operating characteristic set is obtained.

[0023] S3: Determine whether the transformer is in a restricted operating condition based on the transformer operating characteristic set. If it is determined to be in a restricted operating condition, determine the upper limit of the total distributable current increment of the transformer based on the transformer operating parameter set, including: When the rate of change of photovoltaic array output power in the transformer operating characteristic set is less than the preset photovoltaic power contraction threshold, the instantaneous derivative of DC bus voltage is less than the preset bus voltage drop threshold, the instantaneous rise slope of total load current is greater than the preset load current rise threshold, and the jump amplitude of total active power is greater than the preset power jump threshold, the transformer is determined to be in a restricted operating condition where the input energy contraction coincides with the increase in load demand. After the transformer operating characteristic set is formed, the photovoltaic array output power change rate, the instantaneous derivative of the DC bus voltage, the instantaneous rise slope of the total load current, and the jump amplitude of the total active power are compared with the corresponding thresholds. The restricted operating condition is an operating state where input energy contraction and load demand increase occur simultaneously within the same control cycle. The preset photovoltaic power contraction threshold is the judgment boundary for the photovoltaic array output power change rate. When setting it, the rated output power of the transformer photovoltaic array and the allowable power decrease ratio during the control cycle are read. The photovoltaic array rated output power × allowable power decrease ratio during the control cycle / control cycle = photovoltaic power contraction speed reference value. The negative value of the photovoltaic power contraction speed reference value is written into the controller parameter table. The preset bus voltage drop threshold is the judgment boundary for the instantaneous derivative of the DC bus voltage. When setting it, the rated DC bus voltage of the transformer and the allowable voltage drop ratio during the control cycle are read. The rated DC bus voltage × allowable power drop ratio during the control cycle = photovoltaic power contraction speed reference value. The allowable voltage drop ratio during the control cycle / control cycle = bus voltage drop rate reference value. Write the negative value of the bus voltage drop rate reference value into the controller parameter table. The preset load current rise threshold is the boundary for judging the instantaneous rise slope of the total load current. When setting, read the rated value of the total transformer load current and the allowable current rise ratio during the control cycle. The rated value of the total transformer load current × the allowable current rise ratio during the control cycle / control cycle = preset load current rise threshold. The preset power jump threshold is the boundary for judging the jump amplitude of the total active power. When setting, read the rated value of the total active power and the allowable power jump ratio during the control cycle. The rated value of the total active power × the allowable power jump ratio during the control cycle = preset power jump threshold. The rate of change of photovoltaic array output power is compared with a preset photovoltaic power contraction threshold; if it is less than the threshold, it is considered an input energy contraction. The instantaneous derivative of DC bus voltage is compared with a preset bus voltage drop threshold; if it is less than the threshold, it is considered a bus voltage drop. The instantaneous rise slope of total load current is compared with a preset load current rise threshold; if it is greater than the threshold, it is considered an increase in load demand. The jump amplitude of total active power is compared with a preset power jump threshold; if it is greater than the threshold, it is considered a power jump. When the rate of change of photovoltaic array output power is less than the preset photovoltaic power contraction threshold, the instantaneous derivative of DC bus voltage is less than the preset bus voltage drop threshold, the instantaneous rise slope of total load current is greater than the preset load current rise threshold, and the jump amplitude of total active power is greater than the preset power jump threshold, the transformer is determined to be in a restricted operating condition where input energy contraction and load demand increase coincide.

[0024] When the transformer is under restricted operating conditions, the transformer bus capacitance is obtained, the product of the photovoltaic array output power change rate and the control cycle in the transformer operating characteristic set is calculated to obtain the photovoltaic array output power change, and the sum of the photovoltaic array output power change and the photovoltaic array output power is calculated to obtain the photovoltaic array predicted injection power; at the same time, the product of the square of the instantaneous voltage of the transformer DC bus, the transformer bus capacitance and the preset energy coefficient in the transformer operating parameter set is calculated to obtain the bus instantaneous capacitance energy, the ratio of the bus instantaneous capacitance energy to the control cycle is calculated to obtain the bus instantaneous discharge power, the sum of the bus instantaneous discharge power and the photovoltaic array predicted injection power is calculated to obtain the total dispatchable predicted power, the ratio of the total dispatchable predicted power to the instantaneous voltage of the transformer DC bus is calculated to obtain the total dispatchable predicted current, and the current difference between the total dispatchable predicted current and the total transformer load current is calculated as the upper limit of the total distributable current increment of the transformer; After determining that the operating condition is restricted, the following parameters are read: transformer bus capacitance, photovoltaic array output power change rate, photovoltaic array output power, instantaneous voltage of the transformer DC bus, and total transformer load current. The transformer bus capacitance is the capacitance value of the capacitor connected to the transformer DC bus, and this value is written into the controller parameter table from the nominal capacitance. The preset energy coefficient is the conversion factor used when calculating the instantaneous capacitance energy of the bus, and it is set to 0.5 according to the capacitor energy storage relationship. Photovoltaic array output power change rate × control cycle = photovoltaic array output power change; Photovoltaic array output power change + photovoltaic array output power = photovoltaic array predicted injection power; Transformer DC bus instantaneous voltage × transformer DC bus instantaneous voltage = transformer DC bus instantaneous voltage squared; Transformer DC bus instantaneous voltage squared × transformer bus capacitance × preset energy coefficient = bus instantaneous capacitance energy; Bus instantaneous capacitance energy / control cycle = bus instantaneous discharge power; Bus instantaneous discharge power + photovoltaic array predicted injection power = total dispatchable predicted power; Total dispatchable predicted power / transformer DC bus instantaneous voltage = total dispatchable predicted current; Total dispatchable predicted current - transformer load total current = upper limit of transformer total distributable current increment. After completing the power and current conversion under restricted operating conditions, the upper limit of transformer total distributable current increment is obtained.

[0025] S4: Based on the transformer operating parameter set and transformer operating characteristic set, determine the dynamic carrying capacity margin of each parallel branch of the transformer. Combined with the upper limit of the total distributable current increment of the transformer, calculate the current increment of each parallel branch, including: Obtain the upper limit of the rated current of each parallel branch of the transformer, calculate the ratio of the equivalent resistance to the equivalent leakage inductance of each parallel branch in the transformer operating characteristic set, obtain the transient current increase sensitivity of each parallel branch, and calculate the current difference between the upper limit of the rated current of each parallel branch and the actual current flowing through each parallel branch in the transformer operating parameter set, as the basic bearing margin of each parallel branch; The rated current limit of each parallel branch of the transformer is read from the controller parameter table by branch number. The equivalent resistance and equivalent leakage inductance of each parallel branch are read from the branch characteristic record in the transformer operating characteristic set. The actual current flowing through each parallel branch is read from the branch operating record in the transformer operating parameter set. The rated current limit of each parallel branch is the current boundary that each parallel branch can carry. The transient current surge sensitivity is the degree of response of each parallel branch to the current increment under the combined effect of equivalent resistance and equivalent leakage inductance. The basic carrying margin is the difference between the rated current limit of each parallel branch and the current actual current flowing through it. The process of setting the rated current limit of each parallel branch is as follows: read the allowable current of the conductor and the allowable current of the power device in each parallel branch, and write the smaller of the two values ​​into the controller parameter table. The equivalent resistance of each parallel branch / the equivalent leakage inductance of each parallel branch = the transient current surge sensitivity of each parallel branch; the rated current limit of each parallel branch - the actual current flowing through each parallel branch = the basic load-bearing margin of each parallel branch. After completing the branch-by-branch calculation, the transient current surge sensitivity and the basic load-bearing margin of each parallel branch are obtained.

[0026] The ratio of the preset derating reference temperature to the local temperature of each parallel branch in the transformer operating parameter set is calculated as the temperature derating compensation coefficient for each parallel branch. The product of the basic bearing margin and the temperature derating compensation coefficient for each parallel branch is calculated to obtain the dynamic bearing margin for each parallel branch. The transient current increase sensitivity and dynamic bearing margin of each parallel branch are normalized to obtain the normalized transient current increase sensitivity and dynamic bearing margin for each parallel branch. After setting the basic load-bearing margin, the preset derating reference temperature is read from the controller parameter table, the local temperature of each parallel branch is read from the branch operation record in the transformer operating parameter set, and the basic load-bearing margin of each parallel branch obtained from the previous step is also read. The preset derating reference temperature is the temperature reference value used when calculating the temperature derating compensation coefficient. The temperature derating compensation coefficient is the correction ratio of the local temperature of each parallel branch to the basic load-bearing margin, and the dynamic load-bearing margin is the incremental space of the current that can be carried after temperature correction. The process of setting the preset derating reference temperature is as follows: read the allowable temperature of the insulation material and the allowable temperature of the power devices of each parallel branch of the transformer, and write the smaller of the two values ​​into the controller parameter table. The local temperature of each parallel branch is read from the branch operation record in the transformer operating parameter set, and the basic load-bearing margin of each parallel branch is read from the result obtained from the previous step. Preset derating reference temperature / local temperature of each parallel branch = temperature derating compensation coefficient of each parallel branch; basic load-bearing margin of each parallel branch × temperature derating compensation coefficient = dynamic load-bearing margin of each parallel branch. Normalization is performed to map the values ​​under different branch numbers to a unified comparison scale. In MATLAB, the transient current increase sensitivity and dynamic load margin of each parallel branch are arranged by branch number. The maximum value, minimum value, maximum value and minimum value of the transient current increase sensitivity, dynamic load margin of all parallel branches are read respectively. When the maximum transient current surge sensitivity of all parallel branches is not equal to the minimum transient current surge sensitivity of all parallel branches, the normalized transient current surge sensitivity of each parallel branch is calculated as follows: (Transient current surge sensitivity of each parallel branch - Minimum transient current surge sensitivity of all parallel branches) / (Maximum transient current surge sensitivity of all parallel branches - Minimum transient current surge sensitivity of all parallel branches) = Normalized transient current surge sensitivity of each parallel branch. Similarly, when the maximum dynamic load margin of all parallel branches is not equal to the minimum dynamic load margin of all parallel branches, the normalized dynamic load margin of each parallel branch is calculated as follows: (Dynamic load margin of each parallel branch - Minimum dynamic load margin of all parallel branches) / (Maximum dynamic load margin of all parallel branches - Minimum dynamic load margin of all parallel branches) = Normalized dynamic load margin of each parallel branch. After temperature derating and normalization, the normalized transient current surge sensitivity and dynamic load margin of each parallel branch are obtained.

[0027] Calculate the product of the normalized transient current surge sensitivity of each parallel branch and the normalized dynamic carrying capacity of each parallel branch to obtain the comprehensive evaluation value of each parallel branch; calculate the sum of the comprehensive evaluation values ​​of all parallel branches to obtain the total comprehensive evaluation value of the branches; calculate the ratio of the comprehensive evaluation value of each parallel branch to the total comprehensive evaluation value of the branches, and use it as the comprehensive allocation weight coefficient of each parallel branch; calculate the product of the comprehensive allocation weight coefficient of each parallel branch and the upper limit of the total distributable current increment of the transformer to obtain the current increment of each parallel branch. The transient current surge sensitivity and dynamic carrying capacity of each parallel branch are read according to the branch number. The comprehensive evaluation value of each parallel branch is the evaluation value of each parallel branch participating in the current increment allocation, and the comprehensive allocation weight coefficient of each parallel branch is the proportion of the total allocable current increment limit of the transformer for each parallel branch. The larger the transient current surge sensitivity and the larger the dynamic carrying capacity of each parallel branch after normalization, the larger the corresponding comprehensive evaluation value of the branch. The normalized transient current surge sensitivity of each parallel branch × the normalized dynamic carrying capacity margin of each parallel branch = the comprehensive evaluation value of each parallel branch; the sum of the comprehensive evaluation values ​​of all parallel branches = the total comprehensive evaluation value of the branches; the comprehensive evaluation value of each parallel branch / the total comprehensive evaluation value of the branches = the comprehensive allocation weight coefficient of each parallel branch; the comprehensive allocation weight coefficient of each parallel branch × the upper limit of the total distributable current increment of the transformer = the current increment of each parallel branch. After completing the weight allocation of all parallel branches, the current increment of each parallel branch is obtained.

[0028] S5: Generate a drive signal for each parallel branch based on the current increment of each parallel branch and the transformer operating characteristic set. Calculate the actual current slope and current ramp-up slope of each parallel branch after the drive signal is issued, including: The drive signal for each parallel branch is generated based on the current increment of each parallel branch and the equivalent leakage inductance and equivalent resistance of each parallel branch in the transformer operating characteristic set. After determining the current increment of each parallel branch, the current increment of each parallel branch is read according to the branch number, and the equivalent leakage inductance and equivalent resistance of each parallel branch are read from the branch characteristic record in the transformer operating characteristic set. The drive signal is the drive value sent to the drive terminal of each parallel branch. The inductive drive component is the drive component formed by the equivalent leakage inductance and the rate of change of the current increment of each parallel branch, and the resistive drive component is the drive component formed by the equivalent resistance and the current increment of each parallel branch. Current increment / control cycle of each parallel branch = rate of change of the increment of each parallel branch; equivalent leakage inductance of each parallel branch × rate of change of the increment of each parallel branch = inductive drive component of each parallel branch; equivalent resistance of each parallel branch × current increment of each parallel branch = resistive drive component of each parallel branch; inductive drive component + resistive drive component of each parallel branch = drive signal value of each parallel branch. After completing the numerical calculation of the drive signal, the drive signal for each parallel branch is generated.

[0029] After the monitoring transformer sends the drive signal for each parallel branch, the actual current flowing through each parallel branch of the transformer within the control cycle is calculated. The difference in the actual current flow before and after the control cycle is calculated. The ratio of the difference in the current flow to the control cycle is calculated to obtain the actual current slope of each parallel branch. After the drive signal is issued, the current sampling device for each parallel branch of the transformer continues to collect the actual current flowing through each parallel branch before and after the control cycle, and matches it with the drive signal issued for each parallel branch according to the branch number. The actual current slope of each parallel branch is the rate of change of the actual current flowing through each parallel branch within the control cycle after the drive signal is issued, and the creepage difference is the difference between the actual current flowing through each parallel branch before and after the control cycle. Actual current flowing through each parallel branch after the control cycle - Actual current flowing through each parallel branch before the control cycle = Creepage difference; Creepage difference / control cycle = Actual current slope of each parallel branch. After completing the branch-by-branch sampling and slope calculation after the drive signal is issued, the actual current slope of each parallel branch is obtained.

[0030] Calculate the ratio of the current increment to the control cycle for each parallel branch to obtain the current ramp-up slope for each parallel branch. Corresponding to the actual current slope, the current increment of each parallel branch is read according to the branch number, and the control cycle is read according to the current control cycle number. The current ramp-up slope of each parallel branch is the theoretical current change rate corresponding to the current increment of each parallel branch. Current increment of each parallel branch / control cycle = current ramp-up slope of each parallel branch. After completing the branch-by-branch conversion, the current ramp-up slope of each parallel branch is obtained.

[0031] S6: Calculate the actual response slope deviation of each parallel branch based on the actual current slope and current ramp-up slope of each parallel branch, and adjust the current increment of each parallel branch in conjunction with the dynamic load margin of each parallel branch to obtain the load balancing scheduling results, including: Calculate the slope difference between the actual current slope of each parallel branch and the current ramp-up slope of each parallel branch to obtain the actual response slope deviation of each parallel branch. Calculate the product of the actual response slope deviation and the control period to obtain the branch deviation current of each parallel branch. Read the actual current slope and current ramp-up slope of each parallel branch according to the same branch number. The actual response slope deviation of each parallel branch is the difference between the actual current slope and the current ramp-up slope of each parallel branch. The branch deviation current of each parallel branch is the current deviation converted from the actual response slope deviation within the control cycle. Actual current slope of each parallel branch - current ramp-up slope of each parallel branch = actual response slope deviation of each parallel branch; actual response slope deviation × control cycle = branch deviation current of each parallel branch. After completing the slope deviation and current deviation conversion, the actual response slope deviation and branch deviation current of each parallel branch are obtained.

[0032] When the actual response slope deviation is less than the preset health response tolerance threshold, each parallel branch is judged to have a normal response and is marked as a healthy branch. When the actual response slope deviation is greater than or equal to the health response tolerance threshold, each parallel branch is judged to have an abnormal response and is marked as an abnormal branch for removal. For the selected healthy branches, the normalized dynamic carrying margin of each parallel branch is extracted as the dynamic carrying margin of the healthy branch. The maximum current increment step of the healthy branch within the control cycle is calculated in combination with the dynamic carrying margin of the healthy branch. The sum of the dynamic carrying margins of all healthy branches is calculated to obtain the total healthy dynamic carrying margin. The margin ratio of the dynamic carrying margin of the healthy branches to the total healthy dynamic carrying margin is calculated as the theoretical compensation share coefficient. The product of the branch deviation current of each parallel branch corresponding to the healthy branch and the theoretical compensation share coefficient is calculated to obtain the theoretical compensation current. The preset health response tolerance threshold is read from the controller parameter table, and the actual response slope deviation of each parallel branch is read according to the branch number. The preset health response tolerance threshold is the slope deviation boundary used to determine the response status of each parallel branch. A healthy branch is a parallel branch whose actual response slope deviation meets the preset health response tolerance threshold requirement, and an abnormal branch is a parallel branch whose actual response slope deviation does not meet the preset health response tolerance threshold requirement. The theoretical compensation share coefficient is the proportion of compensation current shared by the healthy branch. The setting process of the preset health response tolerance threshold is as follows: the minimum resolvable current and control cycle of the current sampling device of each parallel branch are read, and the minimum resolvable current / control cycle of the current sampling device of each parallel branch = the preset health response tolerance threshold. When the actual response slope deviation of each parallel branch is less than the preset health response tolerance threshold, each parallel branch is judged to have a normal response and is marked as a healthy branch; when the actual response slope deviation of each parallel branch is greater than or equal to the health response tolerance threshold, each parallel branch is judged to have an abnormal response and is marked as an abnormal branch for removal. For the selected healthy branches, the normalized dynamic carrying capacity margin of each parallel branch is used as the corresponding dynamic carrying capacity margin of the healthy branch; the maximum current increment step of the healthy branch is the upper limit of current compensation allowed to be carried by the healthy branch within the control cycle, and the maximum current increment step of the healthy branch is calculated from the dynamic carrying capacity margin of the healthy branch. The sum of the dynamic carrying capacity margins of all healthy branches = the total healthy dynamic carrying capacity margin; the ratio of the dynamic carrying capacity margin of healthy branches to the total healthy dynamic carrying capacity margin = the theoretical compensation share coefficient; the branch deviation current of each parallel branch corresponding to the healthy branch × the theoretical compensation share coefficient = the theoretical compensation current. After completing the healthy branch screening and compensation share calculation, the theoretical compensation current is obtained.

[0033] When the theoretical compensation current is greater than the maximum current increment step of the healthy branch, the compensation difference between the theoretical compensation current and the maximum current increment step of the healthy branch is calculated to obtain the remaining compensation current, which is then stored in the transformer's remaining compensation delay queue. When the theoretical compensation current is less than or equal to the maximum current increment step of the healthy branch, it is determined that there is no remaining compensation current, and the operation of storing data in the transformer's remaining compensation delay queue is stopped. Based on the actual response slope deviation of all parallel branches and the transformer's remaining compensation delay queue, the current increment of each parallel branch is uniformly adjusted to obtain the load balancing scheduling result, which is used to address the transmission challenges faced by high-frequency transformers. When the operating conditions of energy contraction and increased load demand coincide, the carrying current boundary of each parallel branch is dynamically adjusted; the transformer residual compensation delay queue is specifically configured as a queue with a fixed depth and multiple initial null value points; the process of storing the residual compensation current into the transformer residual compensation delay queue includes: encapsulating the current timestamp of the control cycle with the residual compensation current to obtain the node to be stored; when the used depth of the transformer residual compensation delay queue is less than the fixed depth, extracting the target null value from multiple initial null value points; writing the node to be stored into the target null value and configuring a preset effective retention period; clearing the target null value when the effective retention period decreases to zero; The theoretical compensation current and the maximum current increment step of the healthy branch are read according to the branch number corresponding to the healthy branch. The fixed depth, multiple initial empty value positions, used depth, and preset effective retention period of the transformer remaining compensation delay queue are also read. The transformer remaining compensation delay queue is a queue configured with a fixed depth and multiple initial empty value positions. The fixed depth is the upper limit of the number of nodes to be stored that the transformer remaining compensation delay queue can hold. The initial empty value positions are the storage locations of nodes not yet written to. The preset effective retention period is the number of control cycles that the node to be stored will retain in the target empty space. The node to be stored is a queue write object formed by encapsulating the current timestamp of the control cycle with the remaining compensation current. The fixed depth and preset effective retention period are read from the controller parameter table. The fixed depth is set according to the number of control cycles that the transformer remaining compensation delay queue can retain, and the preset effective retention period is set according to the number of control cycles that the remaining compensation current allows to be delayed and retained. When the theoretical compensation current is greater than the maximum current increment step of the healthy branch, the theoretical compensation current minus the maximum current increment step of the healthy branch equals the remaining compensation current, and the remaining compensation current is stored in the transformer's remaining compensation delay queue. When the theoretical compensation current is less than or equal to the maximum current increment step of the healthy branch, it is determined that there is no remaining compensation current, and the operation of storing data in the transformer's remaining compensation delay queue is stopped. When storing the remaining compensation current in the transformer's remaining compensation delay queue, the current timestamp of the control cycle and the remaining compensation current are encapsulated to obtain the node to be stored. When the used depth of the transformer's remaining compensation delay queue is less than the fixed depth, the target empty slot is extracted according to the arrangement order of multiple initial empty slots. After the node to be stored is written into the target empty slot, a preset effective retention period is configured. Each time a control cycle passes, the preset effective retention period is decremented once. When the preset effective retention period decreases to zero, the target empty slot is cleared. Subsequently, based on the actual response slope deviation of all parallel branches and the transformer's remaining compensation delay queue, the current increment of each parallel branch is uniformly adjusted to obtain the load balancing scheduling result.

[0034] Please see Figure 3 A high-frequency transformer energy dispatch and load balancing system, the system comprising: The parameter acquisition module obtains the set of transformer operating parameters within the control cycle of transformer energy dispatch; The feature extraction module determines each transformer operation feature based on the set of transformer operation parameters and integrates them into a set of transformer operation features. The incremental determination module determines whether the transformer is in a restricted operating condition based on the transformer operating characteristic set. When it is determined to be in a restricted operating condition, it determines the upper limit of the total distributable current increment of the transformer based on the transformer operating parameter set. The branch dispatch module determines the dynamic carrying capacity of each parallel branch of the transformer based on the transformer operating parameter set and transformer operating characteristic set, and calculates the current increment of each parallel branch in combination with the upper limit of the total distributable current increment of the transformer. The slope calculation module generates a drive signal for each parallel branch by referring to the current increment of each parallel branch and the transformer operating characteristic set, and calculates the actual current slope and current ramp-up slope of each parallel branch after the drive signal is issued. The load balancing scheduling module calculates the actual response slope deviation of each parallel branch based on the actual current slope and current ramp-up slope of each parallel branch, and adjusts the current increment of each parallel branch in combination with the dynamic load margin of each parallel branch to obtain the load balancing scheduling result.

[0035] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for energy dispatching and load balancing of high-frequency transformers, characterized in that, Includes the following steps: Obtain the set of transformer operating parameters within the control cycle of transformer energy dispatch; Each transformer operating characteristic is determined based on the set of transformer operating parameters and integrated into a set of transformer operating characteristics. Based on the set of transformer operating characteristics, it is determined whether the transformer is in a restricted operating condition. When it is determined to be in a restricted operating condition, the upper limit of the total distributable current increment of the transformer is determined according to the set of transformer operating parameters. Based on the set of transformer operating parameters and the set of transformer operating characteristics, the dynamic carrying capacity of each parallel branch of the transformer is determined, and the current increment of each parallel branch is calculated in combination with the upper limit of the total distributable current increment of the transformer. The drive signal for each parallel branch is generated by referring to the current increment of each parallel branch and the transformer operating characteristic set. The actual current slope and current ramp-up slope of each parallel branch are calculated after the drive signal is issued. The actual response slope deviation of each parallel branch is calculated based on the actual current slope and current ramp-up slope of each parallel branch. The current increment of each parallel branch is adjusted in combination with the dynamic load margin of each parallel branch to obtain the load balancing scheduling result.

2. The high-frequency transformer energy dispatch and load balancing method according to claim 1, characterized in that, The set of transformer operating parameters within the control cycle for obtaining transformer energy dispatch includes: The transformer's electrical parameters are obtained by acquiring the output voltage and current of the photovoltaic array, the instantaneous voltage of the DC bus, the total load current, and the actual current and driving voltage of each parallel branch of the transformer during the control cycle of the transformer energy dispatch. The local temperature of each parallel branch of the transformer is collected during the control cycle to obtain the transformer temperature parameters; By integrating all the transformer electrical parameters and the transformer temperature parameters, a set of transformer operating parameters is obtained.

3. The high-frequency transformer energy dispatch and load balancing method according to claim 2, characterized in that, The process of determining each transformer operating characteristic based on the set of transformer operating parameters and integrating them into a set of transformer operating characteristics includes: The output power of the photovoltaic array is obtained by multiplying the output voltage and output current of the photovoltaic array in the transformer operating parameter set. The sampling time before the start of the control cycle is obtained as the data before the control cycle, and the sampling time after the end of the control cycle is obtained as the data after the control cycle. The power difference of the photovoltaic array output power before and after the control cycle is calculated, and the ratio of the power difference to the control cycle is calculated to obtain the photovoltaic array output power change rate. The product of the instantaneous voltage of the transformer DC bus and the total load current of the transformer is calculated to obtain the total active power. The voltage difference of the instantaneous voltage of the transformer DC bus before and after the control cycle is calculated, and the ratio of the voltage difference to the control cycle is calculated to obtain the instantaneous derivative of the DC bus voltage. The current difference of the total load current before and after the control cycle is calculated, and the ratio of the current difference to the control cycle is calculated to obtain the instantaneous rise slope of the total load current. The jump difference of the total active power before and after the control cycle is calculated to obtain the jump amplitude of the total active power. Referring to the set of transformer operating parameters, calculate the temperature difference between the preset core reference saturation temperature and the local temperature of each parallel branch, and calculate the ratio of the temperature difference to the preset reference saturation temperature difference to obtain the temperature correction coefficient of each parallel branch. Calculate the difference between the actual current flowing through each parallel branch before and after the control cycle, and calculate the ratio of the branch current difference to the control cycle to obtain the instantaneous rate of change of current for each parallel branch. An equivalent circuit model of a transformer branch is established with preset initial leakage inductance parameters and preset initial resistance parameters. The instantaneous rate of change of current, the actual current flowing through each parallel branch, and the driving voltage of each parallel branch are substituted into the equivalent circuit model of the transformer branch. The preset initial leakage inductance parameters and preset initial resistance parameters are combined to perform parameter fitting calculation on the equivalent circuit model of the transformer branch to obtain the basic value of leakage inductance and the basic value of resistance for each parallel branch. The product of the basic value of leakage inductance and the corresponding temperature correction coefficient is calculated to obtain the equivalent leakage inductance for each parallel branch. The product of the basic value of resistance and the corresponding temperature correction coefficient is calculated to obtain the equivalent resistance for each parallel branch. The output power of the photovoltaic array, the rate of change of the output power of the photovoltaic array, the instantaneous derivative of the DC bus voltage, the instantaneous rise slope of the total load current, the jump amplitude of the total active power, and the equivalent leakage inductance and equivalent resistance of each parallel branch are integrated as each transformer operating characteristic to construct a transformer operating characteristic set.

4. The high-frequency transformer energy dispatch and load balancing method according to claim 3, characterized in that, The method of determining whether a transformer is in a restricted operating condition based on a set of transformer operating characteristics, and determining the upper limit of the total distributable current increment of the transformer based on a set of transformer operating parameters when the transformer is determined to be in a restricted operating condition, includes: When the photovoltaic array output power change rate in the transformer operating characteristic set is less than the preset photovoltaic power contraction threshold, the instantaneous derivative of the DC bus voltage is less than the preset bus voltage drop threshold, the instantaneous rise slope of the total load current is greater than the preset load current rise threshold, and the jump amplitude of the total active power is greater than the preset power jump threshold, the transformer is determined to be in a restricted operating condition where the input energy contraction coincides with the increase in load demand. When the transformer is under the restricted operating conditions, the transformer bus capacitance is obtained, and the product of the photovoltaic array output power change rate and the control cycle in the transformer operating characteristic set is calculated to obtain the photovoltaic array output power change. The sum of the photovoltaic array output power change and the photovoltaic array output power is calculated to obtain the photovoltaic array predicted injection power. At the same time, the product of the square of the instantaneous voltage of the transformer DC bus, the transformer bus capacitance, and the preset energy coefficient in the transformer operating parameter set is calculated to obtain the bus instantaneous capacitance energy. The ratio of the bus instantaneous capacitance energy to the control cycle is calculated to obtain the bus instantaneous discharge power. The sum of the bus instantaneous discharge power and the photovoltaic array predicted injection power is calculated to obtain the total dispatchable predicted power. The ratio of the total dispatchable predicted power to the instantaneous voltage of the transformer DC bus is calculated to obtain the total dispatchable predicted current. The current difference between the total dispatchable predicted current and the total transformer load current is calculated as the upper limit of the total distributable current increment of the transformer.

5. The high-frequency transformer energy dispatch and load balancing method according to claim 4, characterized in that, The process of determining the dynamic carrying capacity margin of each parallel branch of the transformer based on the transformer operating parameter set and transformer operating characteristic set, and calculating the current increment of each parallel branch in conjunction with the upper limit of the total distributable current increment of the transformer, includes: Obtain the upper limit of the rated current of each parallel branch of the transformer, calculate the ratio of the equivalent resistance to the equivalent leakage inductance of each parallel branch in the transformer operating characteristic set, obtain the transient current increase sensitivity of each parallel branch, and calculate the current difference between the upper limit of the rated current of each parallel branch and the actual current flowing through each parallel branch in the transformer operating parameter set, as the basic bearing margin of each parallel branch; The ratio of the preset derating reference temperature to the local temperature of each parallel branch in the transformer operating parameter set is calculated as the temperature derating compensation coefficient for each parallel branch. The product of the basic bearing margin of each parallel branch and the temperature derating compensation coefficient is calculated to obtain the dynamic bearing margin of each parallel branch. The transient current increase sensitivity and dynamic bearing margin of each parallel branch are normalized to obtain the normalized transient current increase sensitivity and dynamic bearing margin of each parallel branch. Calculate the product of the normalized transient current surge sensitivity and the normalized dynamic carrying capacity of each parallel branch to obtain the comprehensive evaluation value of each parallel branch; calculate the sum of the comprehensive evaluation values ​​of all parallel branches to obtain the total comprehensive evaluation value of the branches; calculate the ratio of the comprehensive evaluation value of each parallel branch to the total comprehensive evaluation value of the branches, and use it as the comprehensive allocation weight coefficient of each parallel branch; calculate the product of the comprehensive allocation weight coefficient of each parallel branch and the upper limit of the total distributable current increment of the transformer to obtain the current increment of each parallel branch.

6. The high-frequency transformer energy dispatch and load balancing method according to claim 5, characterized in that, The process of generating a drive signal for each parallel branch by referencing the current increment of each parallel branch and the transformer operating characteristic set, and calculating the actual current slope and current ramp-up slope of each parallel branch after issuing the drive signal, includes: Based on the current increment of each parallel branch and the equivalent leakage inductance and equivalent resistance of each parallel branch in the transformer operating characteristic set, a drive signal for each parallel branch is generated. After the monitoring transformer sends the drive signal for each parallel branch, the actual current flowing through each parallel branch of the transformer within the control cycle is calculated. The difference in the actual current flow before and after the control cycle is calculated. The ratio of the difference in the current flow to the control cycle is calculated to obtain the actual current slope of each parallel branch. Calculate the ratio of the current increment to the control cycle for each parallel branch to obtain the current ramp-up slope for each parallel branch.

7. The high-frequency transformer energy dispatch and load balancing method according to claim 6, characterized in that, The process of calculating the actual response slope deviation of each parallel branch based on its actual current slope and current ramp-up slope, and adjusting the current increment of each parallel branch in conjunction with its dynamic load margin, to obtain the load balancing scheduling result includes: Calculate the slope difference between the actual current slope of each parallel branch and the current ramp-up slope of each parallel branch to obtain the actual response slope deviation of each parallel branch. Calculate the product of the actual response slope deviation and the control period to obtain the branch deviation current of each parallel branch. When the actual response slope deviation is less than the preset health response tolerance threshold, each parallel branch is judged to have a normal response and is marked as a healthy branch. When the actual response slope deviation is greater than or equal to the health response tolerance threshold, each parallel branch is judged to have an abnormal response and is marked as an abnormal branch for removal. For the selected healthy branches, the normalized dynamic carrying margin of each parallel branch is extracted as the dynamic carrying margin of the healthy branch. The maximum current increment step of the healthy branch within the control cycle is calculated in combination with the dynamic carrying margin of the healthy branch. The sum of the dynamic carrying margins of all healthy branches is calculated to obtain the total healthy dynamic carrying margin. The margin ratio of the dynamic carrying margin of the healthy branches to the total healthy dynamic carrying margin is calculated as the theoretical compensation share coefficient. The product of the branch deviation current of each parallel branch corresponding to the healthy branch and the theoretical compensation share coefficient is calculated to obtain the theoretical compensation current. When the theoretical compensation current is greater than the maximum current increment step of the healthy branch, the compensation difference between the theoretical compensation current and the maximum current increment step of the healthy branch is calculated to obtain the remaining compensation current, which is then stored in the transformer's remaining compensation delay queue. When the theoretical compensation current is less than or equal to the maximum current increment step of the healthy branch, it is determined that there is no remaining compensation current, and the operation of storing data in the transformer's remaining compensation delay queue is stopped. Based on the actual response slope deviation of all the parallel branches and the transformer's remaining compensation delay queue, the current increment of each parallel branch is uniformly adjusted to obtain the load balancing scheduling result. This result is used to dynamically allocate the carrying current boundary of each parallel branch when the high-frequency transformer faces the condition of input energy contraction and load demand increase.

8. The high-frequency transformer energy dispatch and load balancing method according to claim 3, characterized in that, The process of performing parameter fitting calculations on the equivalent circuit model of the transformer branch to obtain the basic values ​​of leakage inductance and resistance for each parallel branch is as follows: Construct a set of parameters to be optimized, including the leakage inductance parameter and the resistance parameter to be optimized, and assign the preset initial leakage inductance parameter and the preset initial resistance parameter to the leakage inductance parameter and the resistance parameter to be optimized, respectively; The theoretical inductive voltage of the branch is obtained by multiplying the leakage inductance parameter to be optimized with the instantaneous rate of change of the current in each parallel branch. The theoretical resistive voltage of the branch is obtained by multiplying the resistance parameter to be optimized with the actual current flowing through each parallel branch. Calculate the sum of the inductive theoretical voltage and the resistive theoretical voltage of the branch to obtain the comprehensive estimated voltage of the branch. Calculate the difference between the estimated voltage of the branch and the driving voltage of each parallel branch, and update the leakage inductance parameter and the resistance parameter to be optimized synchronously according to the difference and the preset adjustment step size. When the difference is less than the preset fitting convergence threshold, stop updating the parameters and extract the current leakage inductance parameter to be optimized as the basic value of leakage inductance for each parallel branch, and extract the current resistance parameter to be optimized as the basic value of resistance.

9. The high-frequency transformer energy dispatch and load balancing method according to claim 7, characterized in that, The transformer remaining compensation delay queue is specifically configured as a queue with a fixed depth and multiple initial null value positions; The process of storing the remaining compensation current into the transformer's remaining compensation delay queue includes: encapsulating the current timestamp of the control cycle with the remaining compensation current to obtain the node to be stored; When the used depth of the transformer's remaining compensation delay queue is less than the fixed depth, the target empty slot is extracted from multiple initial empty slots; the node to be stored is written into the target empty slot and a preset effective retention period is configured; the target empty slot is cleared when the effective retention period decreases to zero.

10. A high-frequency transformer energy dispatching and load balancing system, characterized in that, The system includes: The parameter acquisition module obtains the set of transformer operating parameters within the control cycle of transformer energy dispatch; The feature extraction module determines each transformer operation feature based on the set of transformer operation parameters and integrates them into a set of transformer operation features. The incremental determination module determines whether the transformer is in a restricted operating condition based on the transformer operating characteristic set. When it is determined to be in a restricted operating condition, it determines the upper limit of the total distributable current increment of the transformer based on the transformer operating parameter set. The branch dispatch module determines the dynamic carrying capacity of each parallel branch of the transformer based on the transformer operating parameter set and transformer operating characteristic set, and calculates the current increment of each parallel branch in combination with the upper limit of the total distributable current increment of the transformer. The slope calculation module generates a drive signal for each parallel branch by referring to the current increment of each parallel branch and the transformer operating characteristic set, and calculates the actual current slope and current ramp-up slope of each parallel branch after the drive signal is issued. The load balancing scheduling module calculates the actual response slope deviation of each parallel branch based on the actual current slope and current ramp-up slope of each parallel branch, and adjusts the current increment of each parallel branch in combination with the dynamic load margin of each parallel branch to obtain the load balancing scheduling result.

Images