Digital power supply error feedback information generation method, digital power supply control method, system, device and computer equipment

Abstract

The application relates to a digital power error feedback information generation method, a digital power control method, a system, a device and computer equipment. The method comprises the following steps: acquiring each feedback value of a collected digital power; storing each feedback value in a two-dimensional space matrix in time sequence, wherein each feedback value between adjacent columns in the two-dimensional space matrix is continuous in time; performing first fusion on each feedback value in each row in the two-dimensional space matrix to obtain each fusion feature, the fusion feature is used for representing the instantaneous fluctuation of each feedback value of the digital power in the time domain, and each fusion feature is stored in the form of a two-dimensional matrix; and performing second fusion on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information, the target error feedback information is used for representing the change trend of each feedback value of the digital power. The method can improve the stability of the digital power loop.

Description

Technical Field

[0001] This application relates to the field of digital power supply technology, and in particular to a method for generating digital power supply error feedback information, a digital power supply control method, a system, an apparatus, and a computer device. Background Technology

[0002] In integrated circuit testing equipment, the VI source, as the excitation voltage source for the unit under test, needs to provide a controllable and measurable power supply. Currently, this technology is continuously developing towards multi-channel and high-precision directions to adapt to more chip testing application scenarios. However, due to factors such as line loss, electrical issues, and time in actual use, the voltage or current applied across the load by the actual VI source deviates from the expected value. Therefore, a stable negative feedback regulation loop is needed to adjust the voltage or current applied to the load in order to achieve the goal of accurate output voltage or current.

[0003] In traditional PID digital loop control methods that adjust the voltage or current applied to the load through negative feedback loops, the system output accuracy is low, and the output waveform of the power supply regulation circuit will oscillate for loads with different characteristics, such as capacitive or inductive loads. Summary of the Invention

[0004] Therefore, it is necessary to provide a digital power supply error feedback information generation method, digital power supply control method, system, device, and computer equipment that can improve the stability of digital power supply circuits, addressing the aforementioned technical problems.

[0005] In a first aspect, this application provides a method for generating digital power supply error feedback information, the method comprising:

[0006] Acquire the feedback values ​​of the collected digital power supply;

[0007] The feedback values ​​are stored in a two-dimensional spatial matrix in chronological order, and the feedback values ​​between adjacent columns in the two-dimensional spatial matrix are continuous in time.

[0008] The feedback values ​​in each row of the two-dimensional spatial matrix are first fused to obtain fused features. The fused features are used to characterize the instantaneous fluctuation of each feedback value of the digital power supply in the time domain, and each fused feature is stored in the form of a two-dimensional matrix.

[0009] A second fusion is performed on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information, which is used to characterize the changing trend of each feedback value of the digital power supply.

[0010] In one embodiment, acquiring the feedback values ​​of the collected digital power supply includes:

[0011] Each feedback value from the digital power supply is cached in a register;

[0012] The feedback values ​​of the collected digital power supply are obtained from the buffer.

[0013] In one embodiment, before storing the feedback values ​​in a two-dimensional spatial matrix in chronological order, the method further includes:

[0014] The target number of rows and columns of the two-dimensional spatial matrix are determined. When the product of the target number of rows and columns is fixed, increasing the target number of rows or decreasing the target number of columns enhances the real-time performance of the digital power supply but weakens its stability; or...

[0015] When the product of the target number of rows and the target number of columns is fixed, as the target number of rows decreases or the target number of columns increases, the real-time performance of the digital power supply weakens, while its stability is enhanced.

[0016] In one embodiment, determining the target number of rows and columns of the two-dimensional spatial matrix includes:

[0017] Determine the initial number of rows and columns of the two-dimensional spatial matrix, and store the collected feedback values ​​of the digital power supply into the initial two-dimensional spatial matrix, wherein the number of rows of the initial two-dimensional spatial matrix is ​​the initial number of rows and the number of columns is the initial number of columns;

[0018] Based on the initial two-dimensional spatial matrix and the feedback values ​​of the digital power supply, the initial number of rows and the initial number of columns are adjusted synchronously to obtain the reference number of rows and the reference number of columns;

[0019] Based on the reference row number and reference column number, as well as the feedback values ​​of the digital power supply, the reference row number and reference column number are adjusted respectively to obtain the target row number and target column number.

[0020] In one embodiment, adjusting the initial number of rows and columns based on the initial two-dimensional spatial matrix and the feedback values ​​of the digital power supply to obtain the reference number of rows and columns includes:

[0021] The feedback values ​​of the digital power supply are stored in the initial two-dimensional spatial matrix;

[0022] The feedback values ​​in the initial two-dimensional space matrix are fused to obtain the initial error feedback information;

[0023] The digital power supply is controlled based on the initial error feedback information to obtain the initial output information of the digital power supply.

[0024] If, based on the initial output information, it is determined that the stability of the digital power supply does not meet the preset stability condition, the initial row number and the initial column number are increased synchronously according to the first preset step size until the stability of the digital power supply meets the preset stability condition.

[0025] If, based on the initial output information, it is determined that the real-time performance of the digital power supply does not meet the preset real-time performance conditions, the initial number of rows and the initial number of columns are reduced synchronously according to the second preset step size until the real-time performance of the digital power supply meets the preset real-time performance conditions.

[0026] Based on the adjusted initial number of rows and columns, the reference number of rows and reference number of columns of the two-dimensional spatial matrix are obtained.

[0027] In one embodiment, adjusting the reference row number and reference column number based on the reference row number and reference column number and the feedback values ​​of the digital power supply to obtain the target row number and target column number includes:

[0028] The system receives control instructions, wherein if it is determined that real-time performance needs to be enhanced, the control instructions include at least one of increasing the number of reference rows or decreasing the number of reference columns; if it is determined that real-time performance needs to be weakened, the control instructions include at least one of decreasing the number of reference rows or increasing the number of reference columns; if it is determined that stability needs to be enhanced, the control instructions include at least one of decreasing the number of reference rows or increasing the number of reference columns; and if it is determined that stability needs to be weakened, the control instructions include at least one of increasing the number of reference rows or decreasing the number of reference columns.

[0029] Adjust at least one of the reference row number and the reference column number based on the control command;

[0030] The intermediate two-dimensional space matrix is ​​determined based on the adjusted number of reference rows and reference columns, and each feedback value of the digital power supply is stored in the intermediate two-dimensional space matrix.

[0031] The feedback values ​​in the intermediate two-dimensional space matrix are fused to obtain reference error feedback information, and the digital power supply is controlled based on the reference error feedback information to obtain reference output information of the digital power supply.

[0032] If the stability and real-time performance of the digital power system meet the corresponding first requirement based on the reference output information, the adjusted reference row number is taken as the target row number, and the adjusted reference column number is taken as the target column number.

[0033] If, based on the reference output information, it is determined that at least one of the stability and real-time performance of the digital power system does not meet the corresponding first requirement, the step of receiving control commands continues.

[0034] In one embodiment, the first fusion of the feedback values ​​in each row of the two-dimensional spatial matrix to obtain fused features includes:

[0035] Obtain the relationship between the first fusion coefficient and the first fusion function;

[0036] Based on the first fusion coefficient and the first fusion function relationship, at least two feedback values ​​located in the same row in the two-dimensional space matrix are first fused to obtain each fusion feature; wherein at least two feedback values ​​are located in adjacent columns or non-adjacent columns.

[0037] In one embodiment, the first fusion of at least two feedback values ​​located in the same row of the two-dimensional spatial matrix, based on the first fusion coefficient and the first fusion function relationship, to obtain various fusion features includes:

[0038] Based on the first fusion coefficient and the first fusion function relationship, each pair of feedback values ​​in the two-dimensional space matrix is ​​fused to obtain each fusion feature, wherein each pair of feedback values ​​is two feedback values ​​located in the same row and adjacent columns in the two-dimensional space matrix.

[0039] In one embodiment, the first fusion coefficient and the rate of change of the fusion feature show opposite trends.

[0040] In one embodiment, the second fusion of the fusion features in each column of the two-dimensional matrix to obtain the target error feedback information includes:

[0041] Obtain the relationship between the second fusion coefficient and the second fusion function;

[0042] Based on the second fusion coefficient and the second fusion function relationship, at least two of the fusion features located in the same column in the two-dimensional matrix are fused in the second fusion to obtain the target error feedback information; wherein at least two of the fusion features are located in adjacent rows or non-adjacent rows.

[0043] In one embodiment, the second fusion is performed on at least two fusion features located in the same column of the two-dimensional matrix based on the second fusion coefficient and the second fusion function relationship to obtain target error feedback information, including:

[0044] Based on the second fusion coefficient and the second fusion function relationship, a second fusion is performed on every two fusion features in the two-dimensional matrix to obtain the target error feedback information, wherein every two fusion features are two feedback values ​​located in the same column and adjacent rows in the two-dimensional matrix.

[0045] In one embodiment, the second fusion coefficient exhibits the same trend of change as the stability of the digital power supply.

[0046] In one embodiment, before performing a first fusion on the feedback values ​​in each row of the two-dimensional spatial matrix to obtain the fused features, the method further includes:

[0047] The first fusion coefficient is set to a fixed initial value, and the second fusion coefficient is adjusted based on at least one of the stability and real-time performance of the digital power supply until the stability and real-time performance of the digital power supply meet the corresponding second requirements, thus obtaining the second fusion coefficient.

[0048] Secondly, this application also provides a digital power supply control method, the method comprising:

[0049] The output of the digital power supply is sampled using a sampling device to obtain various feedback values;

[0050] Based on the above-described digital power supply error feedback information generation method, error feedback information for each target is obtained;

[0051] The control parameter values ​​are obtained based on the target error feedback information and the target PID control algorithm.

[0052] The digital power supply is controlled based on the control parameter values.

[0053] Thirdly, this application also provides a digital power supply system, the system comprising:

[0054] A sampling device is used to sample the output of the digital power supply to obtain various feedback values;

[0055] An FPGA is used to execute the above-described digital power supply error feedback information generation method to obtain each target error feedback information, and to obtain control parameter values ​​based on each target error feedback information and the target PID control algorithm.

[0056] A DAC device for controlling the digital power supply based on the control parameter values.

[0057] Fourthly, this application also provides a digital power supply error feedback information generation device, the device comprising:

[0058] The data storage module is used to acquire the feedback values ​​of the collected digital power supply.

[0059] The time-domain to spatial-domain conversion module is used to store each of the feedback values ​​in time order into a two-dimensional spatial matrix, wherein each of the feedback values ​​in the two-dimensional spatial matrix is ​​continuous in time between adjacent columns;

[0060] The data fusion operation module is used to perform a first fusion on each feedback value in each row of the two-dimensional spatial matrix to obtain each fusion feature. The fusion feature is used to characterize the instantaneous fluctuation of each feedback value of the digital power supply in the time domain, and each fusion feature is stored in the form of a two-dimensional matrix. The module also performs a second fusion on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information. The target error feedback information is used to characterize the changing trend of each feedback value of the digital power supply.

[0061] Fifthly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0062] The aforementioned digital power supply error feedback information generation method, digital power supply control method, system, device, and computer equipment acquire various feedback values ​​of the collected digital power supply; store each feedback value in a two-dimensional spatial matrix in chronological order, wherein the feedback values ​​in adjacent columns of the two-dimensional spatial matrix are continuous in time; perform a first fusion on each feedback value in each row of the two-dimensional spatial matrix to obtain various fusion features, wherein the fusion features are used to characterize the instantaneous fluctuation of each feedback value of the digital power supply in the time domain, and each fusion feature is stored in a two-dimensional matrix; perform a second fusion on each fusion feature in each column of the two-dimensional matrix to obtain various target error feedback information, wherein the target error feedback information is used to characterize the changing trend of each feedback value of the digital power supply. This can fuse instantaneous fluctuations and changing trends, thereby improving accuracy and enhancing the stability of the digital power supply circuit. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a schematic diagram of a digital loop PID control system in traditional technology;

[0065] Figure 2 This is a schematic diagram of a digital power supply system in one embodiment;

[0066] Figure 3 This is a flowchart illustrating a method for generating digital power supply error feedback information in one embodiment;

[0067] Figure 4This is a schematic diagram of the two-dimensional spatial matrix transformation steps in one embodiment;

[0068] Figure 5 This is a schematic diagram of the data fusion module processing steps in one embodiment;

[0069] Figure 6 The following is a simulation result diagram for an embodiment where k1 = 0.5, k2 = 0.5, and k3 = 0.9;

[0070] Figure 7 The following is a simulation result diagram for an embodiment where k1 = 0.5, k2 = 0.5, and k3 = 0.5;

[0071] Figure 8 The following is a simulation result diagram for an embodiment where k1 = 0.5, k2 = 0.5, and k3 = 0.1;

[0072] Figure 9 This is a flowchart illustrating a digital power control method in one embodiment;

[0073] Figure 10 This is an execution diagram of a digital power control method in one embodiment;

[0074] Figure 11 This is a flowchart illustrating a digital power control method in another embodiment;

[0075] Figure 12 This is a structural block diagram of a digital power supply error feedback information generation device in one embodiment. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0077] like Figure 1 As shown, Figure 1This diagram illustrates a traditional digital loop PID control system. In traditional technology, when a 5V voltage needs to be applied to the DUT (Device Under Test) load, line losses often result in an actual voltage below 5V. Therefore, a four-wire method is used for measurement. The actual voltage value on the DUT is acquired through the HS and LS channels, amplified, conditioned, and then sent to the ADC (Analog-to-Digital Converter). The FPGA acquires this value and performs PID calculations, sending the result to the DAC output. If the voltage applied to the DUT is less than 5V, the ADC acquires this signal and transmits it to the FPGA. The FPGA then increases the input to the DAC to further increase the voltage applied to the DUT. Similarly, if the voltage applied to the DUT is higher than 5V, the ADC acquires this signal and transmits it to the FPGA. The FPGA then decreases the input to the DAC to further decrease the voltage applied to the DUT, thus stabilizing the output voltage or current. However, using this traditional PID digital loop control method results in a long system output voltage stabilization time and low accuracy. Moreover, for loads with different characteristics, such as capacitive or inductive loads, it can cause the power supply regulation circuit output waveform to oscillate or the output settling time to vary.

[0078] To address the aforementioned technical problems, this application provides a method for generating digital power supply error feedback information that can effectively improve the stability and accuracy of digital power supply circuits. The method for generating digital power supply error feedback information provided in this embodiment can be applied to applications such as... Figure 2 The digital power supply system shown.

[0079] The digital power supply system includes a sampling device, an FPGA, and a DAC device. The sampling device is used to sample the output of the digital power supply to obtain various feedback values. The FPGA is used to execute the digital power supply error feedback information generation method proposed in this application to obtain various target error feedback information, and to obtain control parameter values ​​based on the target error feedback information and the target PID control algorithm. The DAC device is used to control the digital power supply based on the control parameter values.

[0080] The sampling device is Figure 1 The ADC and DAC devices in the middle are Figure 1 The DAC in this application only improves the method in the FPGA device. Therefore, the digital power supply error feedback information generation method proposed in this application can be encapsulated into an FPGA algorithm IP, with reserved external calling interface, and can be flexibly and seamlessly integrated into any FPGA-based digital power supply architecture.

[0081] Continue to combine Figure 2As shown, the FPGA includes a data access module, a time-domain to spatial domain conversion module, a digital fusion operation module, and an algorithm execution module. The data access module primarily receives voltage and current signals acquired by the acquisition device and then buffers the acquired data in a FIFO to match the processing speed of subsequent algorithms. The time-domain to spatial domain conversion module converts the time-dimensional voltage and current data sent from the data access module into an M*N two-dimensional spatial domain data table (also called a two-dimensional spatial matrix) to match the operation rules of subsequent algorithms. The data fusion operation module 1203 employs a clever operation approach, fusing the acquired voltage and current signals in a two-dimensional spatial dimension. Through algorithmic operations, it can predict and infer the voltage and current fluctuations over time, anticipating the output of the digital power supply loop and controlling the digital loop's adjustment in a timely manner, thereby improving system stability. Finally, the algorithm execution module injects the operation results from the data fusion operation module into the original digital power supply loop, i.e., sends it to the DAC device to achieve the final effect.

[0082] In one exemplary embodiment, such as Figure 3 As shown, a method for generating digital power supply error feedback information is provided, which can be applied to... Figure 1 The following steps, 302 to 308, are used as an example to illustrate the process.

[0083] S302: Acquire the feedback values ​​of the collected digital power supply.

[0084] The feedback values ​​of the digital power supply can be at least one of voltage or current. In this application, the feedback values ​​of the digital power supply are acquired by an acquisition device, for example, by acquiring the feedback values ​​of the digital power supply through an ADC.

[0085] In some optional embodiments, acquiring the feedback values ​​of the collected digital power supply includes: caching the feedback values ​​of the digital power supply in a buffer; and acquiring the feedback values ​​of the collected digital power supply from the buffer.

[0086] The buffer can be an asynchronous FIFO. The method in this application needs to adapt to many complex and variable digital power supply loops, which may lead to a mismatch between the feedback value acquisition speed and the algorithm operation speed. Therefore, adding a data access module can effectively solve the problem of direct speed mismatch between system feedback and algorithm operation. Specifically, an asynchronous FIFO is used, controlling different clocks and storage depths at the read and write ends to match the data acquisition speed of the digital loop and the operation speed of the data fusion operation module. Taking a 1MHz ADC data sampling speed and a 100kHz data processing speed as an example, the FIFO write clock frequency is 1MHz, the data bit width is 9 bits (for ease of explanation, this invention processes 9 data points at a time), and the storage depth is 4096; the FIFO read clock frequency is 100kHz, the data bit width is 9 bits, and the storage depth is 512. After FIFO buffering, the data is sent to the time-domain to spatial-domain conversion module for further data conversion.

[0087] S304: Store each feedback value in a two-dimensional spatial matrix in chronological order, where the feedback values ​​between adjacent columns in the two-dimensional spatial matrix are continuous in time.

[0088] Here, a two-dimensional spatial matrix is ​​a two-dimensional matrix that includes rows and columns, which can be specifically combined with... Figure 4 As shown, the feedback values ​​are stored in a two-dimensional spatial matrix in chronological order as t1, t2, ... up to t... n The n feedback values ​​at time points are stored in an M*N two-dimensional data table. The feedback values ​​in adjacent columns of the two-dimensional spatial matrix are continuous in time, for example... Figure 4 In the first row, feedback values ​​3.1 and 3.3 are sequential in time, and feedback values ​​3.3 and 3 are sequential in time. This means that storing the feedback values ​​in the two-dimensional spatial matrix in chronological order involves: filling each row of the two-dimensional spatial matrix with the feedback values ​​sequentially in chronological order, continuing to fill the next row after the previous row is filled, until all feedback values ​​are stored in the two-dimensional spatial matrix.

[0089] In one optional embodiment, before storing each feedback value into the two-dimensional spatial matrix in chronological order, the method further includes: determining the target number of rows and the target number of columns of the two-dimensional spatial matrix, wherein when the product of the target number of rows and the target number of columns is fixed, increasing the target number of rows or decreasing the target number of columns enhances the real-time performance of the digital power supply and weakens its stability; or, when the product of the target number of rows and the target number of columns is fixed, decreasing the target number of rows or increasing the target number of columns weakens the real-time performance of the digital power supply and enhances its stability.

[0090] Among them, stability is obtained from the waveform output by the digital power supply system, and stability can be represented by the difference between the maximum peak and the trough of the output waveform; real-time performance is represented by the time corresponding to the first peak of the waveform output by the power supply system.

[0091] In this embodiment, when the product of the target number of rows M and the target number of columns N is fixed, the larger the target number of rows M and the smaller the target number of columns N of the two-dimensional spatial matrix, the shorter the time base of the advance prediction system, the worse the effect of predicting changes (i.e., the reduced stability), but the stronger the real-time performance of the system (i.e., the enhanced real-time performance). Conversely, the smaller the target number of rows M and the larger the target number of columns N, the longer the time base of the advance prediction system, the better the effect of predicting changes (i.e., the enhanced stability), but the worse the real-time performance of the system (i.e., the reduced real-time performance).

[0092] S306: Perform a first fusion on each feedback value in each row of the two-dimensional spatial matrix to obtain each fusion feature. The fusion feature is used to characterize the instantaneous fluctuation of each feedback value of the digital power supply in the time domain, and each fusion feature is stored in the form of a two-dimensional matrix.

[0093] S308: Perform a second fusion on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information. The target error feedback information is used to characterize the changing trend of each feedback value of the digital power supply.

[0094] The core idea of ​​the data fusion and calculation module is to combine four adjacent data points (or four spatially close data points, where every two data points are in the same row and every two data points are in the same column) from a data table into a single data point through formula calculation. The fused data point physically represents the fluctuation of the data over time, which is reflected in the digital power supply loop as the fluctuation pattern of the output voltage over time. Therefore, after processing by this algorithm, the control quantity of the digital loop has predictive power for the output voltage, and the weight of real-time performance and predictive power can be adjusted according to actual needs to adapt to different circuit systems, thereby effectively improving the stability of the digital power supply loop.

[0095] Because digital loops have many unstable factors, the collected data fluctuates with the system. The feedback values ​​collected continuously exhibit a relationship between adjacent data points in a two-dimensional space. Data processed according to certain rules can still be used as system feedback values. The basic principle of this processing rule is to balance system stability and response speed, and the weights of stability and response speed in the system can be adjusted using parameters. Following this principle, the feedback values ​​in each row of the two-dimensional space matrix are first fused to obtain fusion characteristics. Data satisfying these characteristics have short fluctuation periods in the time domain, insufficient to predict the overall trend of system feedback value changes, but can effectively suppress instantaneous change rates.

[0096] Secondly, based on the fusion characteristics, a second fusion is performed on each fusion characteristic in each column to obtain the target error feedback information. Data that satisfies the spatial domain secondary fusion characteristics has a long system fluctuation period in the time domain, which can be used to predict the changing trend of the entire system feedback value. This allows for proactive control of the digital power supply loop regulation, effectively suppressing system output oscillations and improving output stability.

[0097] The first fusion involves fusing each row of the two-dimensional spatial matrix. This fusion process can involve fusing at least two feedback values ​​from each row, such as two or three feedback values, without specific limitations. Furthermore, the feedback values ​​participating in the first fusion can be temporally adjacent or non-adjacent. In some optional embodiments, the fusion function of the first fusion can be a first-order function or other functions, without specific limitations. The first fusion in this application can be the fusion of two temporally and spatially adjacent feedback values ​​in the two-dimensional spatial matrix.

[0098] The second fusion involves fusing each column of the two-dimensional matrix separately. This fusion process can involve fusing at least two fusion features from each column, such as two or three fusion features, without specific limitations. Furthermore, the fusion features participating in the second fusion can be temporally adjacent or non-adjacent. In some optional embodiments, the fusion function relationship of the second fusion can be first-order or other functional relationships, without specific limitations. The second fusion in this application can be the fusion of two spatially adjacent fusion features in the two-dimensional matrix.

[0099] In some optional embodiments, this application first fuses two adjacent data in each row to obtain each fused feature, and then fuses two adjacent fused features in each column to obtain target error feedback information.

[0100] The algorithm execution module's role is to apply the computation results of the data fusion algorithm module to the digital power supply loop. Specifically, it uses the data fusion algorithm's computation results—that is, the target error feedback information—as the error feedback value input to the PID controller of the digital loop algorithm execution module. The algorithm execution module still uses the traditional PID controller of the digital power supply control loop, and the controller parameter adjustments also follow the previous PID controller parameters. Finally, the output of the algorithm execution module is applied to the DAC as the output of the digital loop, ultimately achieving closed-loop control of the digital loop.

[0101] The aforementioned method for generating digital power supply error feedback information involves acquiring various feedback values ​​from the collected digital power supply; storing these feedback values ​​in a two-dimensional spatial matrix in chronological order, where the feedback values ​​in adjacent columns of the two-dimensional spatial matrix are continuous in time; performing a first fusion on the feedback values ​​in each row of the two-dimensional spatial matrix to obtain fused features, which characterize the instantaneous fluctuations of the digital power supply's feedback values ​​in the time domain, and storing these fused features in a two-dimensional matrix; and performing a second fusion on the fused features in each column of the two-dimensional matrix to obtain target error feedback information, which characterizes the changing trends of the digital power supply's feedback values. This method integrates instantaneous fluctuations and changing trends, improving accuracy and enhancing the stability of the digital power supply circuit.

[0102] In one optional embodiment, determining the target number of rows and the target number of columns of the two-dimensional spatial matrix includes: determining the initial number of rows and the initial number of columns of the two-dimensional spatial matrix; storing the collected feedback values ​​of the digital power supply into the initial two-dimensional spatial matrix, wherein the number of rows in the initial two-dimensional spatial matrix is ​​the initial number of rows and the number of columns is the initial number of columns; adjusting the initial number of rows and the initial number of columns synchronously based on the initial two-dimensional spatial matrix and the feedback values ​​of the digital power supply to obtain the reference number of rows and the reference number of columns; and adjusting the reference number of rows and the reference number of columns respectively based on the reference number of rows and the reference number of columns and the feedback values ​​of the digital power supply to obtain the target number of rows and the target number of columns.

[0103] In this application, the target number of rows and the target number of columns are predetermined. For an unknown new system, an initial two-dimensional spatial matrix can be initialized. For example, the initial number of rows and the initial number of columns of an initial two-dimensional spatial matrix can be set. In some optional embodiments, the initial number of rows is equal to the initial number of columns. Optionally, M = N = 10. In other embodiments, the initial number of rows and the initial number of columns can be selected with other values. Then, the characteristics of the system are verified to adjust the initial number of rows and the initial number of columns. For example, the feedback values ​​of the collected digital power supply are stored in the initial two-dimensional spatial matrix, and the initial number of rows and the initial number of columns are subsequently adjusted based on the feedback values ​​to obtain the target number of rows and the target number of columns.

[0104] In this application, the adjustment of the initial number of rows and columns includes a coarse adjustment process and a fine adjustment process. The coarse adjustment process is to determine the approximate values ​​of the number of rows and columns, that is, the reference number of rows and reference number of columns mentioned above. The fine adjustment process is to determine the precise values ​​of the number of rows and columns, that is, the target number of rows and target number of columns.

[0105] In some optional embodiments, the coarse adjustment process is a synchronous adjustment process, which adjusts the initial number of rows and columns synchronously based on the real-time performance and stability of the digital power system output, for example, by increasing or decreasing them simultaneously. The fine adjustment process is a separate adjustment process, which adjusts the reference number of rows or reference number of columns separately based on the real-time performance and stability of the digital power system output to obtain the target number of rows and columns.

[0106] In one optional embodiment, the coarse adjustment process, namely adjusting the initial number of rows and columns based on the initial two-dimensional spatial matrix and the feedback values ​​of the digital power supply to obtain the reference number of rows and columns, includes: storing the feedback values ​​of the digital power supply into the initial two-dimensional spatial matrix; fusing the feedback values ​​in the initial two-dimensional spatial matrix to obtain initial error feedback information; controlling the digital power supply based on the initial error feedback information to obtain the initial output information of the digital power supply; if the stability of the digital power supply does not meet the preset stability condition based on the initial output information, synchronously increasing the initial number of rows and columns by a first preset step size until the stability of the digital power supply meets the preset stability condition; if the real-time performance of the digital power supply does not meet the preset real-time performance condition based on the initial output information, synchronously decreasing the initial number of rows and columns by a second preset step size until the real-time performance of the digital power supply meets the preset real-time performance condition; and obtaining the reference number of rows and columns of the two-dimensional spatial matrix based on the adjusted initial number of rows and columns.

[0107] In this application, the feedback values ​​of the digital power supply are first stored in an initial two-dimensional spatial matrix; the feedback values ​​in the initial two-dimensional spatial matrix are fused to obtain initial error feedback information; the digital power supply is controlled based on the initial error feedback information to obtain the initial output information of the digital power supply, which is a waveform, such as a voltage waveform or a current waveform.

[0108] Then, based on the initial output information, the initial stability and initial real-time performance are obtained. The calculation of initial real-time performance and initial stability can be found in the definitions of real-time performance and stability above. If the initial stability or initial real-time performance does not meet the preset stability or real-time performance conditions, the number of rows and columns is adjusted synchronously.

[0109] Specifically, if the initial real-time performance is too slow, that is, if the initial real-time performance is less than the standard real-time performance in the preset real-time performance conditions, the initial number of rows and columns will be reduced synchronously until the real-time performance of the digital power supply meets the preset real-time performance conditions.

[0110] If the initial stability is poor, for example, if the fluctuation is large, that is, if the initial stability is greater than the standard stability in the preset stability condition, then the initial number of rows and columns will be increased simultaneously until the stability of the digital power supply meets the preset stability condition.

[0111] The preset real-time conditions and preset stability conditions mentioned above can be set based on system type or based on needs, and no specific restrictions are made here.

[0112] Furthermore, the first preset step size is a step size determined for stability adjustment, and the second preset step size is a step size determined for real-time adjustment. The first preset step size can be equal to the second preset step size, or it can be different from the second preset step size. No specific limitations are made on the first preset step size and the second preset step size here.

[0113] In the above embodiments, if the system output response speed is found to be too slow, the initial number of rows and columns is gradually reduced; conversely, if the system output is found to have large fluctuations and poor stability, the initial number of rows and columns is gradually increased.

[0114] In one optional embodiment, adjusting the number of reference rows and reference columns based on the number of reference rows and reference columns, and the feedback values ​​of the digital power supply, to obtain the target number of rows and target number of columns includes: receiving a control command, wherein if it is determined that real-time performance needs to be enhanced, the control command includes at least one of increasing the number of reference rows or decreasing the number of reference columns; if it is determined that real-time performance needs to be weakened, the control command includes at least one of decreasing the number of reference rows or increasing the number of reference columns; if it is determined that stability needs to be enhanced, the control command includes at least one of decreasing the number of reference rows or increasing the number of reference columns; if it is determined that stability needs to be weakened, the control command includes at least one of increasing the number of reference rows or decreasing the number of reference columns; and adjusting the number of reference rows and reference columns based on the control command. At least one of the following steps is performed: A two-dimensional space matrix is ​​determined based on the adjusted number of reference rows and reference columns, and each feedback value of the digital power supply is stored in the intermediate two-dimensional space matrix; the feedback values ​​in the intermediate two-dimensional space matrix are fused to obtain reference error feedback information, and the digital power supply is controlled based on the reference error feedback information to obtain reference output information of the digital power supply; if the stability and real-time performance of the digital power supply system meet the corresponding first requirements based on the reference output information, the adjusted number of reference rows is used as the target number of rows, and the adjusted number of reference columns is used as the target number of columns; if at least one of the stability and real-time performance of the digital power supply system does not meet the corresponding first requirements based on the reference output information, the step of receiving control commands continues.

[0115] The control command is input by the user and is used to adjust at least one of the reference row number and reference column number. The user can determine the stability and real-time performance by observing the waveform, and then determine the adjustment strategy for stability or real-time performance, such as at least one of real-time performance enhancement, real-time performance reduction, stability enhancement, and stability reduction. In this way, the corresponding control command can be determined based on the adjustment strategy.

[0116] Specifically, when it is determined that real-time performance needs to be enhanced, the control instruction includes at least one of increasing the number of reference rows or decreasing the number of reference columns; when it is determined that real-time performance needs to be weakened, the control instruction includes at least one of decreasing the number of reference rows or increasing the number of reference columns; when it is determined that stability needs to be enhanced, the control instruction includes at least one of decreasing the number of reference rows or increasing the number of reference columns; when it is determined that stability needs to be weakened, the control instruction includes at least one of increasing the number of reference rows or decreasing the number of reference columns; and at least one of the number of reference rows and the number of reference columns is adjusted based on the control instruction.

[0117] The system adjusts at least one of the reference row and column numbers based on control commands. Then, an intermediate two-dimensional space matrix is ​​determined based on the adjusted reference row and column numbers. The feedback values ​​of the digital power supply are stored in this intermediate two-dimensional space matrix. Reference feedback values ​​are obtained based on the two-dimensional space matrix and input into the digital power supply system. Reference output information is then acquired, and stability and real-time performance are recalculated based on this information until both stability and real-time performance meet the requirements. The target row and column numbers are then obtained. If at least one of the stability or real-time performance requirements does not meet the corresponding first requirement, the user can continue to adjust it by inputting control commands.

[0118] The first requirement is the requirement for real-time performance and the requirement for stability. This first requirement can be determined by the user based on their needs, and no specific limitations are made here.

[0119] In the above embodiments, after coarsely determining the approximate values ​​of M and N, that is, the number of reference rows and the number of reference columns, the specific values ​​of M and N are further refined and adjusted. If a faster system output response speed is desired, M is increased and N is decreased; conversely, if a smaller system output fluctuation and better stability are desired, N is increased and M is decreased. M and N are both ≥2 and are positive integers. In this way, M and N that meet the corresponding first requirement are obtained as the target number of rows and the target number of columns.

[0120] In one optional embodiment, a first fusion is performed on each feedback value in each row of the two-dimensional spatial matrix to obtain each fusion feature, including: obtaining a first fusion coefficient and a first fusion function relationship; based on the first fusion coefficient and the first fusion function relationship, a first fusion is performed on at least two feedback values ​​located in the same row of the two-dimensional spatial matrix to obtain each fusion feature; wherein at least two feedback values ​​are located in adjacent columns or non-adjacent columns.

[0121] This application primarily describes the specific process of the first fusion, wherein the first fusion function can be a first-order function or another, and the first fusion coefficient can be preset. In one optional embodiment, the first fusion coefficient and the rate of change of the fusion feature exhibit opposite trends. For example, the larger the first fusion coefficient, the smaller the rate of change of the fusion feature; conversely, the smaller the first fusion coefficient, the larger the rate of change of the fusion feature.

[0122] The first fusion involves fusing at least two feedback values ​​from the same row. This fusion can involve two, three, or four feedback values ​​from the same row, and there are no specific limitations on this. Furthermore, the at least two feedback values ​​can be adjacent or non-adjacent.

[0123] In one optional embodiment, based on a first fusion coefficient and a first fusion function relationship, at least two feedback values ​​located in the same row in a two-dimensional spatial matrix are first fused to obtain each fusion feature, including: based on the first fusion coefficient and the first fusion function relationship, each pair of feedback values ​​is two feedback values ​​located in the same row and adjacent columns in a two-dimensional spatial matrix to obtain each fusion feature.

[0124] To reduce computational load, this application performs a first fusion on every two adjacent feedback values ​​in the two-dimensional spatial matrix to obtain each fusion feature. Each pair of feedback values ​​consists of two feedback values ​​located in the same row and adjacent columns of the two-dimensional spatial matrix.

[0125] For ease of understanding, combined with Figure 5 As shown, taking a 3*3 two-dimensional spatial matrix as an example, a, b…i represent the voltage values ​​of ADC2 corresponding to 9 consecutive sampling times from t1 to t9, and A, B, C, and D represent the values ​​after data fusion processing. 11 t 12 t 13 t 14 The voltage values ​​at four time points will be sent to the algorithm execution module as input to the PID controller, i.e., target error feedback information for subsequent adjustment.

[0126] For example, consider the feedback values ​​a and b at times t1 and t2. Assume the fusion coefficient between these two feedback values ​​is k1, and the range of k1 is 0-1. The first fusion relationship satisfies a first-order function relationship, i.e., ab = a*k1 + b*(1-k1) ①. The rate of change of two adjacent feedback values ​​in the time domain is controlled by adjusting the value of k1. The specific analysis is as follows: Figure 5 Taking a = 3.1 and b = 3.3 as an example,

[0127] With k1 = 0.1, the value of a and b after merging is ab = 3.1 * 0.1 + 3.3 * (1 - 0.1) = 3.28;

[0128] Taking k1 = 0.9, the value after merging a and b is ab = 3.1 * 0.9 + 3.3 * (1 - 0.9) = 3.12

[0129] Clearly, the larger k1 is, the smaller the rate of change of the feedback value, and vice versa.

[0130] The feedback values ​​b and c at times t2 and t3 are processed in the same way. Assume the fusion coefficient between the two feedback values ​​b and c is k2, and the range of k2 is set to 0-1. The first fusion relationship satisfies a first-order function relationship, i.e., bc = b*k2 + b*(1-k2)②. The rate of change of two adjacent feedback values ​​in the time domain is controlled by adjusting the value of k2. The specific analysis is as follows: Figure 5 Taking b=3.3 and c=3.0 as an example,

[0131] With k2 = 0.1, the value of b and c after merging is bc = 3.3 * 0.1 + 3.0 * (1 - 0.1) = 3.03;

[0132] Taking k2 = 0.9, the value after merging b and c is bc = 3.3 * 0.9 + 3.0 * (1 - 0.9) = 3.27

[0133] Clearly, the larger k2 is, the smaller the rate of change of the feedback value, and vice versa.

[0134] Similarly, we can conclude that:

[0135] The fused value of the feedback values ​​d and e at times t4 and t5 is de=d*k1+e*(1-k1)③;

[0136] The fused value of the feedback values ​​e and f at times t5 and t6 is ef=a*k2+b*(1-k2)④.

[0137] In one optional embodiment, a second fusion is performed on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information, including: obtaining a second fusion coefficient and a second fusion function relationship; based on the second fusion coefficient and the second fusion function relationship, a second fusion is performed on at least two fusion features located in the same column of the two-dimensional matrix to obtain each target error feedback information; wherein at least two fusion features are located in adjacent rows or non-adjacent rows.

[0138] This application primarily describes the specific process of the second fusion, where the second fusion function can be a first-order function or another, and the second fusion coefficient can be preset. In one optional embodiment, the second fusion coefficient exhibits the same trend as the stability of the digital power supply. For example, a larger second fusion coefficient results in higher stability of the digital power supply, while a smaller second fusion coefficient results in lower stability of the digital power supply.

[0139] The second fusion involves fusing at least two features from the same column. This fusion can be two features from the same column, three features from the same column, or four features from the same column, etc., without specific limitations. Furthermore, these at least two features can be adjacent or non-adjacent.

[0140] In one optional embodiment, based on the second fusion coefficient and the second fusion function relationship, a second fusion is performed on at least two fusion features located in the same column of the two-dimensional matrix to obtain the target error feedback information, including: based on the second fusion coefficient and the second fusion function relationship, a second fusion is performed on every two fusion features in the two-dimensional matrix to obtain the target error feedback information, wherein every two fusion features are two feedback values ​​located in the same column and adjacent rows of the two-dimensional matrix.

[0141] To reduce computational load, this application performs a second fusion on every two adjacent fusion features in the two-dimensional matrix to obtain the target error feedback information. Each pair of fusion features consists of two fusion features located in the same column and adjacent rows of the two-dimensional matrix.

[0142] The second fusion is a spatial secondary fusion of the results of data fusion that are adjacent in both time and space. Specifically, it involves a spatial secondary fusion of the fusion value ab at times t1 and t2 with the fusion value de at times t4 and t5. Assume the fusion coefficient between the two fusion features ab and de is k3, and set the range of k3 to 0-1. The fusion relationship satisfies a first-order function relationship, i.e., A = ab * k3 + de * (1-k3)⑤. This yields the spatial domain fusion values ​​at times t1, t2, t4, and t5, which is the target error feedback information A. The target error feedback information A will be used as the feedback value fed back to the PID controller of the algorithm execution module at time t11. Data A, which satisfies the spatial domain secondary fusion characteristic, has a long system fluctuation period in the time domain. This can be used to predict the changing trend of the entire system feedback value, thereby enabling proactive control of the digital power supply loop regulation, effectively suppressing system output oscillations, and improving output stability.

[0143] The trend of the feedback value of the entire system is controlled by adjusting the value of k3. The specific analysis is as follows: Figure 5Taking a = 3.1, b = 3.3, d = 2.9, and e = 2.8 as an example, and substituting k1 = 0.1 and k2 = 0.1 into equations ① and ③, we get:

[0144] ab=3.1*0.1+3.3*(1-0.1)=3.28;

[0145] de=2.9*0.1+2.8*(1-0.1)=2.81;

[0146] Substituting the above result into equation ⑤, we get:

[0147] When k3 = 0.1, t 11 The spatial fusion value at time A = 3.28 * 0.1 + 2.81 * (1 - 0.1) = 2.857;

[0148] When k3 = 0.9, t 11 The spatial fusion value at time A = 3.28 * 0.9 + 2.81 * (1 - 0.9) = 3.233;

[0149] Clearly, the larger k3 is, the stronger the suppression effect on the changing trend of the entire system feedback value, and the better the system stability; conversely, the smaller k3 is, the weaker the suppression effect on the changing trend of the entire system feedback value, and the faster the system response speed.

[0150] Substituting equations ① and ③ into equation ⑤ yields:

[0151] A=[a*k1+b*(1-k1)]*k3+[d*k2+e*(1-k2)]*(1-k3) (1)

[0152] The rest t 12 t 13 t 14 The spatial domain fusion values ​​B, C, and D at the three time points can be obtained similarly:

[0153] B=[b*k1+c*(1-k1)]*k3+[e*k2+f*(1-k2)]*(1-k3) (2)

[0154] C=[d*k1+e*(1-k1)]*k3+[g*k2+h*(1-k2)]*(1-k3) (3)

[0155] D=[e*k1+f*(1-k1)]*k3+[h*k2+i*(1-k2)]*(1-k3) (4)

[0156] The continuous t can be calculated using the four formulas above. 11 t 12 t 13 t 14The system feedback values ​​are input to the algorithm execution module at four time points.

[0157] Regarding the four formulas mentioned above, only a combination of... Figure 5 The calculation and derivation of the data fusion values ​​at four consecutive time points of the system are presented, and the analogy and generalization are applied to the actual application scenario of digital power supply loops as follows:

[0158] Assume the digital loop continuously samples t1, t2...t at equal intervals over a period of time. n There are a total of n feedback values ​​a1, a2...a at time points. n These data are used to construct an M*N two-dimensional data table, so M*N = n. The specific rules for the values ​​of M and N are detailed in the description of the time-domain to spatial-domain conversion module 1202 above. The output after data fusion is denoted as A1, A2…A… (M-1)*(N-1) Analogous to equation (1) above, we have:

[0159] A1=[a1*k1+a2*(1-k1)]*k3+[a N+1 *k2+a N+2 *(1-k2)]*(1-k3);

[0160] A2=[a2*k1+a3*(1-k1)]*k3+[a N+2 *k2+a N+3 *(1-k2)]*(1-k3);

[0161] A (M-1)*(N-1) =[a (M-1)*N-1 *k1+a (M-1)*N *(1-k1)]*k3+[a M*N-1 *k2+a M*N *(1-k2)]*(1-k3).

[0162] The actual simulation results of the above algorithm are as follows: Figures 6 to 8 As shown, the original data on the left side of the simulation experiment represents the direct conversion of t1, t2...t... n The feedback values ​​at each time point are fed into the PID control digital loop to achieve the final output effect; the right side represents t1, t2...t n Each feedback value at any time Figure 2 The data fusion and computation module in the middle processes and calculates the t after the operation. 11 t 12 ...t 1n The fused value at each moment is fed into the PID control digital loop to achieve the final output effect.

[0163] In one optional embodiment, before performing a first fusion on each feedback value in each row of the two-dimensional spatial matrix to obtain each fused feature, the method further includes: setting the first fusion coefficient to a fixed initial value, and adjusting the second fusion coefficient based on at least one of the stability and real-time performance of the digital power supply, until the stability and real-time performance of the digital power supply meet the corresponding second requirements, thereby obtaining the second fusion coefficient.

[0164] In this application, the first fusion coefficients k1 and k2 are generally taken as fixed initial values, which can be 0.5, or other values ​​in other embodiments.

[0165] The value of the second fusion coefficient k3 needs to be adjusted according to user needs. If better system stability is desired, a larger value of k3 is better; if faster system response is desired, a smaller value of k3 is better. The values ​​of the three parameters k1, k2, and k3 mentioned above are all between 0 and 1.

[0166] In one exemplary embodiment, such as Figure 9 As shown, a digital power supply control method is provided, including steps 902 to 908. Wherein:

[0167] S902: The output of the digital power supply is sampled by a sampling device to obtain various feedback values.

[0168] For convenience, combined Figure 10 As shown, Figure 10 The diagram shows the execution of a digital power supply control method in one embodiment, where the sampling devices are ADC1 and ACD2, which are used to sample the output of the digital power supply to obtain various feedback values, including current feedback values ​​and voltage feedback values.

[0169] S904: The error feedback information of each target is obtained based on the digital power supply error feedback information generation method of any one of the above.

[0170] The FPGA's data access module is used to cache these feedback values, the time-domain to spatial domain conversion module is used to read these cached feedback values ​​and convert them into the spatial domain, and the subsequent data fusion module performs data fusion in the manner described above to obtain the error feedback information of each target.

[0171] S906: The control parameter values ​​are obtained based on the error feedback information of each target and the target PID control algorithm.

[0172] S908: Controls digital power supplies based on control parameter values.

[0173] The control parameter values ​​are obtained by processing the error feedback information from each target and the desired voltage or current. Finally, the digital power supply is controlled based on the control parameter values. This approach only requires adding logic code to the FPGA, without affecting the original digital loop architecture or requiring readjustment of PID parameters. This greatly improves the portability and operational flexibility of the solution, and shortens the development cycle and reduces costs.

[0174] For ease of understanding, combine Figure 11 As shown, Figure 11 The flowchart below illustrates a digital power supply control method in another embodiment. In this embodiment, the digital power supply is powered on, the user sets the target voltage or current value for the digital power supply output, the PID controller calculates the output, the DAC converts it into an analog output, the subsequent power amplifier circuit processes it, and finally the output is sent to the DUT load. The acquisition device can collect the voltage or current feedback values ​​of the DUT load, sample and store them in a FIFO data buffer. The time-domain to spatial domain conversion module reads these buffered feedback values ​​and converts them to the spatial domain. The subsequent data fusion module performs data fusion as described above to obtain the target error feedback information. The target feedback information is then applied to the PID controller to adjust its output. By sending the voltage and current feedback values ​​acquired from the high-speed ADC to the algorithm module for processing, the algorithm-processed result is used as the new feedback quantity and re-inputted to the PID controller. Compared to the original system output, which had significant fluctuations, the algorithm-processed system output fluctuations are greatly reduced, significantly improving system stability.

[0175] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0176] Based on the same inventive concept, this application also provides a digital power supply error feedback information generation device for implementing the aforementioned digital power supply error feedback information generation method. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the digital power supply error feedback information generation device provided below can be found in the limitations of the digital power supply error feedback information generation method described above, and will not be repeated here.

[0177] In one exemplary embodiment, such as Figure 12 As shown, a digital power supply error feedback information generation device is provided, including: a data storage module 1201, a time-domain to spatial-domain conversion module 1202, and a data fusion and calculation module 1203, wherein:

[0178] The data storage module 1201 is used to acquire the feedback values ​​of the collected digital power supply;

[0179] The time-domain to spatial domain conversion module 1202 is used to store each feedback value in time order into a two-dimensional spatial matrix, where each feedback value in the two-dimensional spatial matrix is ​​continuous in time between adjacent columns;

[0180] The data fusion operation module 1203 is used to perform a first fusion on each feedback value in each row of the two-dimensional spatial matrix to obtain each fusion feature. The fusion feature is used to characterize the instantaneous fluctuation of each feedback value of the digital power supply in the time domain, and each fusion feature is stored in the form of a two-dimensional matrix. The module also performs a second fusion on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information. The target error feedback information is used to characterize the changing trend of each feedback value of the digital power supply.

[0181] In one optional embodiment, the data access module 1201 is specifically used to cache each feedback value of the digital power supply feedback to a buffer; and to obtain each feedback value of the collected digital power supply from the buffer.

[0182] In one optional embodiment, the aforementioned time-domain to spatial domain module 1202 is specifically used to determine the target number of rows and the target number of columns of the two-dimensional spatial matrix. When the product of the target number of rows and the target number of columns is fixed, as the target number of rows increases or the target number of columns decreases, the real-time performance of the digital power supply is enhanced, while its stability is weakened. Alternatively, when the product of the target number of rows and the target number of columns is fixed, as the target number of rows decreases or the target number of columns increases, the real-time performance of the digital power supply is weakened, while its stability is enhanced.

[0183] In one optional embodiment, the aforementioned time-domain to spatial domain module 1202 is specifically used to determine the initial number of rows and the initial number of columns of the two-dimensional spatial matrix, store the collected feedback values ​​of the digital power supply into the initial two-dimensional spatial matrix, the number of rows of the initial two-dimensional spatial matrix being the initial number of rows, and the number of columns being the initial number of columns; based on the initial two-dimensional spatial matrix and the feedback values ​​of the digital power supply, the initial number of rows and the initial number of columns are synchronously adjusted to obtain the reference number of rows and the reference number of columns; based on the reference number of rows and the reference number of columns and the feedback values ​​of the digital power supply, the reference number of rows and the reference number of columns are adjusted respectively to obtain the target number of rows and the target number of columns.

[0184] In one optional embodiment, the aforementioned time-domain to spatial domain conversion module 1202 is specifically used to store each feedback value of the digital power supply into an initial two-dimensional spatial matrix; fuse each feedback value in the initial two-dimensional spatial matrix to obtain initial error feedback information; control the digital power supply based on the initial error feedback information to obtain initial output information of the digital power supply; if it is determined based on the initial output information that the stability of the digital power supply does not meet the preset stability condition, synchronously increase the initial number of rows and columns by a first preset step size until the stability of the digital power supply meets the preset stability condition; if it is determined based on the initial output information that the real-time performance of the digital power supply does not meet the preset real-time performance condition, synchronously decrease the initial number of rows and columns by a second preset step size until the real-time performance of the digital power supply meets the preset real-time performance condition; and obtain the reference number of rows and reference number of columns of the two-dimensional spatial matrix based on the adjusted initial number of rows and initial number of columns.

[0185] In one optional embodiment, the aforementioned time-domain to spatial-domain conversion module 1202 is specifically configured to receive control commands, wherein, if it is determined that real-time performance needs to be enhanced, the control commands include at least one of increasing the number of reference rows or decreasing the number of reference columns; if it is determined that real-time performance needs to be weakened, the control commands include at least one of decreasing the number of reference rows or increasing the number of reference columns; if it is determined that stability needs to be enhanced, the control commands include at least one of decreasing the number of reference rows or increasing the number of reference columns; if it is determined that stability needs to be weakened, the control commands include at least one of increasing the number of reference rows or decreasing the number of reference columns; adjust at least one of the number of reference rows and the number of reference columns based on the control commands; and adjust based on the adjusted number of reference rows. The intermediate two-dimensional space matrix is ​​determined by the reference column number, and the feedback values ​​of the digital power supply are stored in the intermediate two-dimensional space matrix. The feedback values ​​in the intermediate two-dimensional space matrix are fused to obtain reference error feedback information, and the digital power supply is controlled based on the reference error feedback information to obtain the reference output information of the digital power supply. If the stability and real-time performance of the digital power supply system meet the corresponding first requirements based on the reference output information, the adjusted reference row number is taken as the target row number, and the adjusted reference column number is taken as the target column number. If at least one of the stability and real-time performance of the digital power supply system does not meet the corresponding first requirements based on the reference output information, the step of receiving control commands continues.

[0186] In one optional embodiment, the data fusion module is specifically used to obtain a first fusion coefficient and a first fusion function relationship; based on the first fusion coefficient and the first fusion function relationship, a first fusion is performed on at least two feedback values ​​located in the same row in the two-dimensional space matrix to obtain each fusion feature; wherein at least two feedback values ​​are located in adjacent columns or non-adjacent columns.

[0187] In one optional embodiment, the data fusion module is specifically used to perform a first fusion on every two feedback values ​​in the two-dimensional spatial matrix based on a first fusion coefficient and a first fusion function relationship to obtain each fusion feature, wherein every two feedback values ​​are two feedback values ​​located in the same row and adjacent columns in the two-dimensional spatial matrix.

[0188] In one alternative embodiment, the first fusion coefficient exhibits an opposite trend to the rate of change of the fusion feature.

[0189] In one optional embodiment, the data fusion module is specifically used to obtain a second fusion coefficient and a second fusion function relationship; based on the second fusion coefficient and the second fusion function relationship, a second fusion is performed on at least two fusion features located in the same column in the two-dimensional matrix to obtain the error feedback information of each target; wherein at least two fusion features are located in adjacent rows or non-adjacent rows.

[0190] In one optional embodiment, the data fusion module is specifically used to perform a second fusion on every two fusion features in the two-dimensional matrix based on the second fusion coefficient and the second fusion function relationship to obtain the target error feedback information, wherein every two fusion features are two feedback values ​​located in the same column and adjacent rows in the two-dimensional matrix.

[0191] In one alternative embodiment, the second fusion coefficient exhibits the same trend as the stability of the digital power supply.

[0192] In one optional embodiment, the data fusion module is specifically used to set the first fusion coefficient to a fixed initial value, and adjust the second fusion coefficient based on at least one of the stability and real-time performance of the digital power supply, until the stability and real-time performance of the digital power supply meet the corresponding second requirements, thereby obtaining the second fusion coefficient.

[0193] Each module in the aforementioned digital power supply error feedback information generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0194] In an exemplary embodiment, a computer device, which may be an FPGA, is provided. The computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a method for generating digital power supply error feedback information. The display unit is used to form a visually visible image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0195] Those skilled in the art will understand that the limitations in the above paragraphs are only partial structures related to the present application and do not constitute a limitation on the computer device on which the present application is applied. Specific computer devices may include more or fewer components than those mentioned in the above paragraphs, or combine certain components, or have different component arrangements.

[0196] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0197] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0198] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0199] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0200] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0201] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for generating digital power supply error feedback information, characterized in that, The method includes: Acquire the feedback values ​​of the collected digital power supply; The feedback values ​​are stored in a two-dimensional spatial matrix in chronological order, and the feedback values ​​between adjacent columns in the two-dimensional spatial matrix are continuous in time. The first fusion is performed on each feedback value in each row of the two-dimensional spatial matrix to obtain each fusion feature, including: obtaining a first fusion coefficient and a first fusion function relationship; based on the first fusion coefficient and the first fusion function relationship, the first fusion is performed on every two feedback values ​​in the two-dimensional spatial matrix to obtain each fusion feature, wherein every two feedback values ​​are two feedback values ​​located in the same row and adjacent columns in the two-dimensional spatial matrix; the fusion features are used to characterize the instantaneous fluctuation of each feedback value of the digital power supply in the time domain, and each fusion feature is stored in the form of a two-dimensional matrix; The second fusion is performed on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information, including: obtaining a second fusion coefficient and a second fusion function relationship; based on the second fusion coefficient and the second fusion function relationship, the second fusion is performed on every two fusion features in the two-dimensional matrix to obtain each target error feedback information, wherein every two fusion features are two feedback values ​​located in the same column and adjacent rows in the two-dimensional matrix; the target error feedback information is used to characterize the changing trend of each feedback value of the digital power supply.

2. The method according to claim 1, characterized in that, The acquired feedback values ​​of the digital power supply include: Each feedback value from the digital power supply is cached in a register; The feedback values ​​of the collected digital power supply are obtained from the buffer.

3. The method according to claim 1, characterized in that, Before storing the feedback values ​​into a two-dimensional spatial matrix in chronological order, the method further includes: The target number of rows and columns of the two-dimensional spatial matrix are determined. When the product of the target number of rows and columns is fixed, increasing the target number of rows or decreasing the target number of columns enhances the real-time performance of the digital power supply but weakens its stability; or... When the product of the target number of rows and the target number of columns is fixed, as the target number of rows decreases or the target number of columns increases, the real-time performance of the digital power supply weakens, while its stability is enhanced.

4. The method according to claim 3, characterized in that, Determining the target number of rows and columns of the two-dimensional spatial matrix includes: Determine the initial number of rows and columns of the two-dimensional spatial matrix, and store the collected feedback values ​​of the digital power supply into the initial two-dimensional spatial matrix, wherein the number of rows of the initial two-dimensional spatial matrix is ​​the initial number of rows and the number of columns is the initial number of columns; Based on the initial two-dimensional spatial matrix and the feedback values ​​of the digital power supply, the initial number of rows and the initial number of columns are adjusted synchronously to obtain the reference number of rows and the reference number of columns; Based on the reference row number and reference column number, as well as the feedback values ​​of the digital power supply, the reference row number and reference column number are adjusted respectively to obtain the target row number and target column number.

5. The method according to claim 4, characterized in that, The adjustment of the initial row number and initial column number based on the initial two-dimensional spatial matrix and the feedback values ​​of the digital power supply to obtain the reference row number and reference column number includes: The feedback values ​​of the digital power supply are stored in the initial two-dimensional spatial matrix; The feedback values ​​in the initial two-dimensional space matrix are fused to obtain the initial error feedback information; The digital power supply is controlled based on the initial error feedback information to obtain the initial output information of the digital power supply. If, based on the initial output information, it is determined that the stability of the digital power supply does not meet the preset stability condition, the initial row number and the initial column number are increased synchronously according to the first preset step size until the stability of the digital power supply meets the preset stability condition. If, based on the initial output information, it is determined that the real-time performance of the digital power supply does not meet the preset real-time performance conditions, the initial number of rows and the initial number of columns are reduced synchronously according to the second preset step size until the real-time performance of the digital power supply meets the preset real-time performance conditions. Based on the adjusted initial number of rows and columns, the reference number of rows and reference number of columns of the two-dimensional spatial matrix are obtained.

6. The method according to claim 4, characterized in that, The step of adjusting the reference row number and reference column number based on the reference row number and reference column number and the feedback values ​​of the digital power supply to obtain the target row number and target column number includes: The system receives control instructions, wherein if it is determined that real-time performance needs to be enhanced, the control instructions include at least one of increasing the number of reference rows or decreasing the number of reference columns; if it is determined that real-time performance needs to be weakened, the control instructions include at least one of decreasing the number of reference rows or increasing the number of reference columns; if it is determined that stability needs to be enhanced, the control instructions include at least one of decreasing the number of reference rows or increasing the number of reference columns; and if it is determined that stability needs to be weakened, the control instructions include at least one of increasing the number of reference rows or decreasing the number of reference columns. Adjust at least one of the reference row number and the reference column number based on the control command; The intermediate two-dimensional space matrix is ​​determined based on the adjusted number of reference rows and reference columns, and each feedback value of the digital power supply is stored in the intermediate two-dimensional space matrix. The feedback values ​​in the intermediate two-dimensional space matrix are fused to obtain reference error feedback information, and the digital power supply is controlled based on the reference error feedback information to obtain reference output information of the digital power supply. If the stability and real-time performance of the digital power system meet the corresponding first requirement based on the reference output information, the adjusted reference row number is taken as the target row number, and the adjusted reference column number is taken as the target column number. If, based on the reference output information, it is determined that at least one of the stability and real-time performance of the digital power system does not meet the corresponding first requirement, the step of receiving control commands continues.

7. The method according to claim 1, characterized in that, The first fusion coefficient and the rate of change of the fusion feature show opposite trends.

8. The method according to claim 1, characterized in that, The second fusion coefficient exhibits the same trend as the stability of the digital power supply.

9. The method according to claim 1, characterized in that, Before performing a first fusion of the feedback values ​​in each row of the two-dimensional spatial matrix to obtain the fused features, the method further includes: The first fusion coefficient is set to a fixed initial value, and the second fusion coefficient is adjusted based on at least one of the stability and real-time performance of the digital power supply until the stability and real-time performance of the digital power supply meet the corresponding second requirements, thus obtaining the second fusion coefficient.

10. A digital power supply control method, characterized in that, The method includes: The output of the digital power supply is sampled using a sampling device to obtain various feedback values; The error feedback information of each target is obtained based on the digital power supply error feedback information generation method according to any one of claims 1 to 9; The control parameter values ​​are obtained based on the target error feedback information and the target PID control algorithm. The digital power supply is controlled based on the control parameter values.

11. A digital power supply system, characterized in that, The system includes: A sampling device is used to sample the output of the digital power supply to obtain various feedback values; An FPGA is used to execute the digital power supply error feedback information generation method according to any one of claims 1 to 9 to obtain target error feedback information, and to obtain control parameter values ​​based on the target error feedback information and the target PID control algorithm. A DAC device for controlling the digital power supply based on the control parameter values.

12. A digital power supply error feedback information generation device, characterized in that, The device includes: The data storage module is used to acquire the feedback values ​​of the collected digital power supply. The time-domain to spatial-domain conversion module is used to store each of the feedback values ​​in time order into a two-dimensional spatial matrix, wherein each of the feedback values ​​in the two-dimensional spatial matrix is ​​continuous in time between adjacent columns; The data fusion operation module is used to perform a first fusion on each feedback value in each row of the two-dimensional spatial matrix to obtain each fusion feature, including: obtaining a first fusion coefficient and a first fusion function relationship; based on the first fusion coefficient and the first fusion function relationship, performing a first fusion on every two feedback values ​​in the two-dimensional spatial matrix to obtain each fusion feature, wherein every two feedback values ​​are two feedback values ​​located in the same row and adjacent columns of the two-dimensional spatial matrix; the fusion feature is used to characterize the instantaneous fluctuation of each feedback value of the digital power supply in the time domain, and each fusion feature is stored in a two-dimensional matrix; and performing a second fusion on each fusion feature in each column of the two-dimensional matrix to obtain each target error feedback information, including: obtaining a second fusion coefficient and a second fusion function relationship; based on the second fusion coefficient and the second fusion function relationship, performing a second fusion on every two fusion features in the two-dimensional matrix to obtain each target error feedback information, wherein every two fusion features are two feedback values ​​located in the same column and adjacent rows of the two-dimensional matrix; the target error feedback information is used to characterize the changing trend of each feedback value of the digital power supply.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 9.

Images