Power supply driving method and system based on silicon carbide MOS tube

Abstract

The present invention relates to the field of power drive technology, and discloses a power drive method and system based on a silicon carbide MOSFET. The method comprises: extracting preset indicator features from historical drive data of the silicon carbide MOSFET and performing feature screening to obtain target indicator features; performing indicator data fusion on target indicator data corresponding to the target indicator features to obtain fused data for each target indicator feature; calculating the power supply power consumption and switch power consumption of the silicon carbide MOSFET based on power supply parameters, and constructing total power consumption based on the power supply power consumption and switch power consumption; generating a parameter particle population based on the fused data, performing differential optimization on the parameter particle population based on the total power consumption to obtain an optimized parameter population; determining indicator parameter data based on the optimized parameter population, and performing power drive on the power supply corresponding to the silicon carbide MOSFET based on the indicator parameter data. The present invention can improve the accuracy of power drive.

Description

Technical Field

[0001] The present invention relates to the field of power drive technology, and in particular to a power drive method and system based on a silicon carbide MOS tube. Background Art

[0002] Currently, a significant portion of global energy is consumed during energy conversion, with semiconductor power devices accounting for the majority of this consumption. Once-dominant silicon power devices are increasingly unable to meet the demands of high power, high temperature, high frequency, and miniaturized devices. Silicon carbide field-effect transistors (SiC MOS), a new wide-bandgap transistor, not only offer high breakdown voltage, strong current capability, and excellent thermal stability, but also boast high carrier saturation drift rates and excellent thermal conductivity, exceeding three times that of conventional silicon. As a third-generation semiconductor material, SiC is rapidly gaining popularity due to its superior properties.

[0003] Compared with traditional silicon MOS tubes, silicon carbide MOS tubes have the characteristics of fast turn-on and turn-off speeds, low switching losses, and low on-resistance. They are suitable for operating in fields with higher operating frequencies. At the same time, their high-temperature characteristics are better than those of traditional silicon devices, and they can operate in higher temperature environments. However, when the power supply is driven, if the positive disturbance of the power supply exceeds the threshold voltage Vgs(th) of the silicon carbide MOS tube, it will cause the silicon carbide MOS tube to be mis-turned on, causing a power short circuit. When the negative disturbance is too large and exceeds the maximum negative voltage allowed by the silicon carbide MOS tube, it may cause gate-source damage or even breakdown. Therefore, how to improve the accuracy of power supply driving based on the safe voltage of the silicon carbide MOS tube and ensure power safety has become an urgent problem to be solved. Summary of the Invention

[0004] The present invention provides a power driving method and system based on a silicon carbide MOS tube, the main purpose of which is to solve the problem of poor power driving accuracy.

[0005] To achieve the above objectives, the present invention provides a power driving method based on a silicon carbide MOS tube, comprising:

[0006] Obtain historical driving data and power supply parameters of the silicon carbide MOS tube, extract indicator feature data of preset indicator features from the historical driving data, perform feature screening on the indicator features according to the indicator feature data, and obtain target indicator features;

[0007] Extracting target indicator data corresponding to the target indicator feature from the historical driving data, performing indicator data fusion on the target indicator data, and obtaining fused data of each target indicator feature;

[0008] Calculating the power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube according to the power supply parameters, and constructing the total power consumption according to the power consumption and the switch power consumption;

[0009] The power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube are calculated using the following formula:

[0010]

[0011]

[0012] Among them, P1 represents power consumption, I a , I b Respectively represent the power supply working current, the power supply capacitor working current, R a 、R b Respectively represent the equivalent internal resistance of the power supply and the power supply capacitor in the power supply parameters, P2 represents the switching power consumption, H represents the switching frequency of the silicon carbide MOS tube in the power supply parameters, U represents the electrolytic capacitor voltage at the output end in the power supply parameters, I L Indicates the inductor current, t r , t f Respectively represent the switch rising transition time and the switch falling transition time in the power supply parameters;

[0013] generating a parameter particle population according to the fusion data, performing differential optimization on the parameter particle population according to the total power consumption, and obtaining an optimized parameter population;

[0014] The index parameter data of the silicon carbide MOS tube is determined according to the total group of optimization parameters, and the power supply corresponding to the silicon carbide MOS tube is driven according to the index parameter data.

[0015] Optionally, the performing feature screening on the indicator features according to the indicator feature data to obtain target indicator features includes:

[0016] Dividing the indicator feature data into a minority class data set and a majority class data set, and calculating the nearest neighbor data of each minority class data in the minority class data set;

[0017] Performing linear interpolation on the minority class data set according to the neighbor data and the majority class data set to obtain interpolated data corresponding to the indicator feature data;

[0018] generating interpolation feature data of the indicator feature data according to the interpolation data;

[0019] Calculating the indicator feature weight of each indicator feature data according to the interpolation feature data;

[0020] Target indicator features are selected from the indicator features according to the indicator feature weights.

[0021] Optionally, performing linear interpolation on the minority class data set based on the neighbor data and the majority class data set to obtain interpolated data corresponding to the indicator feature data includes:

[0022] Determine the boundary data in the minority class data set according to the neighbor data and the majority class data set, and calculate the boundary neighbor data of each boundary data;

[0023] Performing data interpolation on the boundary neighbor data to obtain minority class samples, and adding the minority class samples to the indicator feature data to obtain minority class indicator data;

[0024] Data cleaning is performed on the minority class indicator data to obtain interpolation data corresponding to the indicator feature data.

[0025] Optionally, performing indicator data fusion on the target indicator data to obtain fused data of each target indicator feature includes:

[0026] Using a pre-built multi-layer perceptron to perform data mapping on the target indicator data to obtain a data vector corresponding to the target indicator data;

[0027] Constructing a data matrix of the target indicator data according to the data vector;

[0028] The multi-layer perceptron is used to perform data fusion according to the data matrix to obtain fused data of the target indicator characteristics.

[0029] Optionally, constructing the total power consumption according to the power consumption of the power supply and the power consumption of the switch includes:

[0030] Obtaining the power supply output current corresponding to the silicon carbide MOS tube, and calculating the power supply power consumption and the current output power consumption corresponding to the switch power consumption according to the power supply output current;

[0031] Calculating current change power consumption according to the power supply output current;

[0032] The current output power consumption and the current change power consumption are weighted and summed to obtain the total power consumption.

[0033] Optionally, calculating the current change power consumption according to the power supply output current includes:

[0034] Use the following formula to calculate the power consumption due to current change:

[0035] P3=(I k -I k-1 )2 , k=1,2,…,N

[0036] Among them, P3 represents the current change power consumption, I k , I k-1 They represent the power supply output current at time k and time k-1 respectively, and N represents a positive integer.

[0037] Optionally, generating a total swarm of parameter particles according to the fusion data includes:

[0038] Obtaining a data constraint range corresponding to each target indicator feature in the fused data;

[0039] Generate constraint data corresponding to each target indicator feature according to the data constraint range;

[0040] The constraint data are combined to obtain a total swarm of parameter particles.

[0041] Optionally, performing differential optimization on the parameter particle population according to the total power consumption to obtain an optimized parameter population includes:

[0042] Generating a fitness function according to the total power consumption, and calculating the fitness value of each parameter particle in the total parameter particle population;

[0043] updating the parameter particle swarm according to the fitness value to obtain an updated particle swarm;

[0044] Mutating and crossovering the updated particle population to obtain an initial optimized population;

[0045] The initial optimization total group is iterated until the number of iterations is greater than a preset iteration number threshold, thereby obtaining the optimization parameter total group.

[0046] Optionally, determining the index parameter data of the silicon carbide MOS tube according to the total group of optimization parameters includes:

[0047] Calculating the fitness value of each optimization parameter in the total group of optimization parameters;

[0048] The optimization parameter corresponding to the minimum fitness value is selected as the index parameter data of the silicon carbide MOS tube.

[0049] In order to solve the above problems, the present invention also provides a power drive system based on silicon carbide MOS tube, the system comprising:

[0050] An indicator feature screening module is used to obtain historical driving data and power supply parameters of the silicon carbide MOS tube, extract indicator feature data of preset indicator features from the historical driving data, and perform feature screening on the indicator features according to the indicator feature data to obtain target indicator features;

[0051] An indicator data fusion module is used to extract target indicator data corresponding to the target indicator feature from the historical driving data, perform indicator data fusion on the target indicator data, and obtain fused data of each target indicator feature;

[0052] a power consumption calculation module, configured to calculate the power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube according to the power supply parameters, and construct a total power consumption according to the power consumption and the switch power consumption;

[0053] The power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube are calculated using the following formula:

[0054]

[0055]

[0056] Among them, P1 represents power consumption, I a , I b Respectively represent the power supply working current, the power supply capacitor working current, R a 、R b Respectively represent the equivalent internal resistance of the power supply and the power supply capacitor in the power supply parameters, P2 represents the switching power consumption, H represents the switching frequency of the silicon carbide MOS tube in the power supply parameters, U represents the electrolytic capacitor voltage at the output end in the power supply parameters, I L Indicates the inductor current, t r , t f Respectively represent the switch rising transition time and the switch falling transition time in the power supply parameters;

[0057] a differential optimization module, configured to generate a parameter particle population according to the fusion data, and perform differential optimization on the parameter particle population according to the total power consumption to obtain an optimized parameter population;

[0058] A power driving module is used to determine the index parameter data of the silicon carbide MOS tube according to the total group of optimization parameters, and to drive the power supply corresponding to the silicon carbide MOS tube according to the index parameter data.

[0059] The embodiment of the present invention extracts the index feature data of the preset index features from the historical driving data of the silicon carbide MOSFET. According to the index feature data, the index features can be feature screened, the interference index features can be removed, and the target index data corresponding to the more accurate target index features can be obtained; data fusion of the target index data can be performed to integrate the target index data, further reduce data redundancy, improve calculation efficiency, and at the same time, the correlation between the data can be mined; the total power consumption can be constructed according to the power supply parameters. The power consumption during power drive and the switching loss of the silicon carbide MOSFET can be comprehensively considered, and the subsequent index parameter data can be optimized by the loss model to obtain the total optimized parameter group; the index parameter data is determined according to the total optimized parameter group, and the power supply corresponding to the silicon carbide MOSFET is driven according to the index parameter data. The power consumption can be minimized in a safe working environment based on the silicon carbide MOSFET, and the power drive can be performed more accurately. Therefore, the power drive method and system based on the silicon carbide MOSFET proposed in the present invention can solve the problem of poor accuracy of power drive. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic flow chart of a power supply driving method based on a silicon carbide MOS tube provided in one embodiment of the present invention;

[0061] Figure 2 A schematic diagram of a process for performing indicator data fusion on target indicator data provided by one embodiment of the present invention;

[0062] Figure 3 A schematic diagram of a process for generating a parameter particle swarm based on fusion data according to an embodiment of the present invention;

[0063] Figure 4 This is a functional module diagram of a power drive system based on silicon carbide MOS tubes provided in one embodiment of the present invention.

[0064] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0065] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] The embodiment of the present application provides a power driving method based on silicon carbide MOS tube. The execution subject of the power driving method based on silicon carbide MOS tube includes but is not limited to at least one of the electronic devices such as the server, the terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the power driving method based on silicon carbide MOS tube can be executed by software or hardware installed on the terminal device or the server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0067] Reference Figure 1 FIG2 is a flow chart of a power supply driving method based on a silicon carbide MOS tube according to an embodiment of the present invention. In this embodiment, the power supply driving method based on a silicon carbide MOS tube includes:

[0068] S1. Obtain historical driving data and power supply parameters of the silicon carbide MOS tube, extract indicator feature data of preset indicator features in the historical driving data, and perform feature screening on the indicator features according to the indicator feature data to obtain target indicator features.

[0069] In the embodiment of the present invention, the silicon carbide MOS tube is a metal oxide semiconductor field effect tube based on silicon carbide semiconductor material, which is generally used in high-voltage, high-power power supplies, and the power circuit is driven by the switch of the driving circuit composed of the silicon carbide MOS tube.

[0070] Specifically, the historical driving data includes the data of the silicon carbide MOSFET when it is driven by the power supply within a preset time period, including driving data such as circuit voltage, output current, power, and load of each node of the power supply when the power supply is working; the power supply parameters include but are not limited to the driving data of the silicon carbide MOSFET at the current moment, the switching frequency of the silicon carbide MOSFET, the electrolytic capacitor voltage at the output end, the switching transition time, the threshold voltage of the silicon carbide MOSFET, the threshold capacitance, the working current of the capacitor in the power supply, the resistance, and the working current, output current, and equivalent internal resistance of the power supply when the power supply is working.

[0071] In an embodiment of the present invention, the preset indicator feature is a data indicator representing power driving, and the historical driving data is summarized and collected through the indicator feature. For example, data whose indicator feature is circuit voltage is extracted from the historical driving data to obtain characteristic indicator data corresponding to the circuit voltage, and the characteristic indicator data is sorted according to time. By extracting the indicator characteristic data, the indicator features can be feature screened, interference indicator features can be removed, and more accurate target indicator features can be obtained.

[0072] In the embodiment of the present invention, the step of screening the indicator features according to the indicator feature data to obtain target indicator features includes:

[0073] Dividing the indicator feature data into a minority class data set and a majority class data set, and calculating the nearest neighbor data of each minority class data in the minority class data set;

[0074] Performing linear interpolation on the minority class data set according to the neighbor data and the majority class data set to obtain interpolated data corresponding to the indicator feature data;

[0075] generating interpolation feature data of the indicator feature data according to the interpolation data;

[0076] Calculating the indicator feature weight of each indicator feature data according to the interpolation feature data;

[0077] Target indicator features are selected from the indicator features according to the indicator feature weights.

[0078] In an embodiment of the present invention, the indicator feature data corresponding to each indicator feature can be divided into minority class data and majority class data according to a preset data ratio. For example, data division is performed according to a 3:1 data ratio, and the data distance between each minority class data in the minority class data and other minority class data or majority class data is calculated. According to the data distance, a preset number k of neighboring data are selected from the indicator feature data.

[0079] Specifically, the feature weight of each indicator feature can be calculated using the filter feature selection method Relief, so that a preset number of target indicator features can be selected according to the size of the weight.

[0080] Among them, in the Relief method, it assigns weights to indicator features according to the strength of the correlation between the indicator feature and another indicator feature (binary classification), and deletes indicator features with weights below a specific threshold. It defines correlation as the ability of the indicator feature to distinguish neighboring observation points. Specifically, the Relief method randomly selects an observation point S from the interpolated feature data corresponding to the indicator feature, and then calculates the nearest neighbor observation point of S from the interpolated feature data of the same indicator feature, called NearHit, and also calculates the nearest neighbor observation point of S from the interpolated feature data corresponding to different indicator features, called NearMiss.

[0081] The weight of each indicator feature is updated according to the following rules: 1. If the distance between observation point S and NearHit on a certain indicator feature is greater than the distance between it and NearMiss, the weight of this indicator feature will increase, because the indicator feature helps distinguish the indicator feature data in the nearest neighbor situation; 2. Conversely, if the distance between observation point S and NearHit on a certain indicator feature is less than the distance between it and NearMiss, the weight of this indicator feature will decrease. The above process is repeated m times to obtain the average weight of each indicator feature. The larger the average weight of the indicator feature, the stronger the classification ability of the indicator feature. Therefore, a preset number of target indicator features can be selected from the indicator features based on the numerical value of the average weight of each indicator feature.

[0082] In an embodiment of the present invention, performing linear interpolation on the minority class data set based on the nearest neighbor data and the majority class data set to obtain interpolated data corresponding to the indicator feature data includes:

[0083] Determine the boundary data in the minority class data set according to the neighbor data and the majority class data set, and calculate the boundary neighbor data of each boundary data;

[0084] Performing data interpolation on the boundary neighbor data to obtain minority class samples, and adding the minority class samples to the indicator feature data to obtain minority class indicator data;

[0085] Data cleaning is performed on the minority class indicator data to obtain interpolation data corresponding to the indicator feature data.

[0086] In an embodiment of the present invention, the neighbor data is the neighbor data of each minority class data in the minority class data set, and the number of neighbor data is k. The neighbor data may include minority class data and majority class data. If there are k / 2 to k neighbor data in the neighbor data that belong to the majority class data, then the minority class data corresponding to the neighbor data belongs to the boundary data. The data distance between the boundary data and the minority class data set is calculated, and a preset number of neighbor data for each boundary data is determined to obtain the boundary neighbor data. Data interpolation is performed on the boundary neighbor data to obtain minority class samples. Specifically, the SMOTE algorithm can be used for data interpolation to limit the boundaries of the data interpolation, so that the data distribution of the minority class samples is more accurate.

[0087] In an optional embodiment of the present invention, data cleaning of minority class indicator data is a process of deleting duplicate data in the minority class indicator data and performing data preprocessing on missing values, outliers, etc. Data cleaning can remove noise data in the minority class indicator data, avoid interference from noise data, and obtain more accurate interpolation data.

[0088] In the embodiment of the present invention, linear interpolation can be used to balance the data imbalance problem between different indicator feature data. At the same time, data cleaning can reduce the influence of noise data and improve the accuracy of subsequent feature selection.

[0089] S2. Extract target indicator data corresponding to the target indicator feature from the historical driving data, perform indicator data fusion on the target indicator data, and obtain fused data of each target indicator feature.

[0090] In one embodiment, target indicator data corresponding to each target indicator feature is extracted from the historical driving data, for example, the historical driving data corresponding to the target indicator feature when the target indicator feature is the circuit voltage, or the historical driving data corresponding to the target indicator feature when the target indicator feature is the load of each node of the power supply, etc., to obtain the target indicator data corresponding to each target indicator feature.

[0091] In the embodiment of the present invention, see Figure 2 As shown, the target indicator data is fused to obtain fused data of each target indicator feature, including:

[0092] S21. Using a pre-built multi-layer perceptron to perform data mapping on the target indicator data, to obtain a data vector corresponding to the target indicator data;

[0093] S22. Constructing a data matrix of the target indicator data according to the data vector;

[0094] S23. Perform data fusion using the multi-layer perceptron according to the data matrix to obtain fused data of the target indicator features.

[0095] In an embodiment of the present invention, a multilayer perceptron (MLP) is an artificial neural network with a forward structure, comprising an input layer, an output layer and multiple hidden layers composed of neurons. The target indicator data can be mapped to corresponding data vectors through the corresponding mapping function in the multilayer perceptron. Each data vector is used as a row vector to construct a data matrix of the target indicator data. The data matrix is ​​input into the multilayer perceptron to perform vector mapping to obtain a mapping vector. The mapping vector is activated to obtain fused data of the target indicator features.

[0096] In an embodiment of the present invention, the target indicator data can be integrated through indicator data fusion, further reducing data redundancy and improving the efficiency of subsequent calculations. At the same time, the multi-layer perceptron can be used to fully explore the correlation between the data in the target indicator data, thereby improving the accuracy of subsequent power driving.

[0097] S3. Calculate the power consumption of the silicon carbide MOSFET and the switch power consumption of the silicon carbide MOSFET according to the power supply parameters, and construct a total power consumption according to the power consumption and the switch power consumption.

[0098] In the embodiment of the present invention, the power consumption of the power supply represents the power consumption when the power supply is driven, and the switch power consumption of the silicon carbide MOS tube reflects the power consumption when the silicon carbide MOS tube is working. The power consumption and the switch power consumption can reflect the energy utilization rate of the power supply, thereby improving the accuracy of the power supply driving.

[0099] In the embodiment of the present invention, the power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube are calculated using the following formula:

[0100]

[0101]

[0102] Among them, P1 represents power consumption, I a , I b Respectively represent the power supply working current, the power supply capacitor working current, R a 、R b Respectively represent the equivalent internal resistance of the power supply and the power supply capacitor in the power supply parameters, P2 represents the switching power consumption, H represents the switching frequency of the silicon carbide MOS tube in the power supply parameters, U represents the electrolytic capacitor voltage at the output end in the power supply parameters, I L Indicates the inductor current, t r , t f They respectively represent the switch rising transition time and the switch falling transition time in the power supply parameters.

[0103] In the embodiment of the present invention, the power consumption of the power supply and the silicon carbide MOSFET is constructed by the power consumption of the power supply and the switch power consumption, and the energy utilization of the power supply can be allocated according to the total power consumption to improve the accuracy of the power supply driving.

[0104] In an embodiment of the present invention, constructing the total power consumption according to the power consumption of the power supply and the power consumption of the switch includes:

[0105] Obtaining the power supply output current corresponding to the silicon carbide MOS tube, and calculating the power supply power consumption and the current output power consumption corresponding to the switch power consumption according to the power supply output current;

[0106] Calculating current change power consumption according to the power supply output current;

[0107] The current output power consumption and the current change power consumption are weighted and summed to obtain the total power consumption.

[0108] In the embodiment of the present invention, the power supply output current is the output current value of the power supply corresponding to the silicon carbide MOS tube at different times. The current output power consumption represents the sum of the power supply power consumption caused by the output current of the power supply at a certain moment and the switch power consumption. The current change power consumption represents the power consumption caused by the change of the output current of the power supply between time intervals. The total power consumption can be used to calculate the power consumption when the power supply is driven as a whole, so as to drive the power supply more accurately.

[0109] Specifically, the current output power consumption can be calculated as the sum of the power consumption of the power supply and the switch power consumption, and the sum of the power consumption is multiplied by the power supply output current at a certain moment to obtain the current output power consumption at the corresponding moment.

[0110] Specifically, the current change power consumption is calculated using the following formula:

[0111] P3=(I k -I k-1 ) 2 , k=1,2,…,N

[0112] Among them, P3 represents the current change power consumption, I k , I k-1 They represent the power supply output current at time k and time k-1 respectively, and N represents a positive integer.

[0113] In the embodiment of the present invention, by constructing the total power consumption, the power consumption during power driving and the switching loss of the silicon carbide MOS tube can be comprehensively considered, and then the subsequent indicator parameter data can be optimized through the loss model to perform power driving more accurately.

[0114] S4. Generate a parameter particle population according to the fusion data, perform differential optimization on the parameter particle population according to the total power consumption, and obtain an optimized parameter population.

[0115] In one embodiment, the fused data can be used as a group of parameter particles when the power is driven, and a total group of parameter particles consisting of multiple groups of parameter data is generated by fusing the data.

[0116] Specifically, see Figure 3 As shown, the generating of the parameter particle swarm according to the fusion data includes:

[0117] S31, obtaining the data constraint range corresponding to each target indicator feature in the fused data;

[0118] S32. Generate constraint data corresponding to each target indicator feature according to the data constraint range;

[0119] S33. Combining the constraint data to obtain a total swarm of parameter particles.

[0120] In an embodiment of the present invention, the data constraint range is a constraint condition for each target indicator feature. Constraint data is randomly generated within the data constraint range. The constrained data ensures that the obtained parameter particle population meets the safety requirements during power drive. For example, the threshold voltage Vgs(th) of the silicon carbide MOS tube in the parameter particle population is within the data constraint range, thereby ensuring that the silicon carbide will not be mis-turned on during power drive, thereby improving power safety.

[0121] In the embodiment of the present invention, performing differential optimization on the parameter particle population according to the total power consumption to obtain the optimized parameter population includes:

[0122] Generating a fitness function according to the total power consumption, and calculating the fitness value of each parameter particle in the total parameter particle population;

[0123] updating the parameter particle swarm according to the fitness value to obtain an updated particle swarm;

[0124] Mutating and crossovering the updated particle population to obtain an initial optimized population;

[0125] The initial optimization total group is iterated until the number of iterations is greater than a preset iteration number threshold, thereby obtaining the optimization parameter total group.

[0126] In the embodiment of the present invention, the minimum value of total power consumption is used as the fitness function, and parameter particles with fitness values ​​less than a preset fitness threshold in the parameter particle population are selected as the update particle population to update the parameter particle population.

[0127] It should be noted that the data in the power supply parameters will be affected by the driving of the silicon carbide MOS tube, resulting in different parameter data resulting in different power supply power consumption and switch power consumption. Therefore, the power supply power consumption and switch power consumption are calculated in the form of functions. Therefore, the total power consumption is also calculated in the form of a function. The fitness function can be further constructed through the total power consumption to perform differential optimization on the total number of parameter particles to obtain the optimized parameter total number.

[0128] Among them, mutation is to select three other parameter particles for a parameter particle A first, and then generate the mutated parameter particle corresponding to parameter particle A through the other three parameter particles. Crossover is to cross each parameter particle with the offspring mutated parameter particles it generates. Specifically, for each parameter particle, the offspring mutated parameter particles (otherwise the original parameter particles) are selected according to a certain probability to generate the parameter particles after crossover.

[0129] In the embodiment of the present invention, differential optimization can be used to generate parameter particles with low power consumption, optimal power output current, and optimal power output power change rate, thereby improving the accuracy of power drive under the safe voltage of the silicon carbide MOS tube.

[0130] S5. Determine index parameter data of the silicon carbide MOSFET according to the total group of optimized parameters, and drive a power supply corresponding to the silicon carbide MOSFET according to the index parameter data.

[0131] In the embodiment of the present invention, the index parameter data is the index characteristic data corresponding to the target index characteristic corresponding to the optimal parameter particle selected from the total group of optimized parameters, which can further improve the accuracy of power driving.

[0132] In an embodiment of the present invention, determining the index parameter data of the silicon carbide MOS tube according to the total group of optimization parameters includes:

[0133] Calculating the fitness value of each optimization parameter in the total group of optimization parameters;

[0134] The optimization parameter corresponding to the minimum fitness value is selected as the index parameter data of the silicon carbide MOS tube.

[0135] In an embodiment of the present invention, the driving parameters of the target indicator characteristics when the silicon carbide MOSFET is driven by a power supply are determined by the indicator parameter data, and the corresponding power supply is controlled by the indicator parameter data to drive, for example, the line voltage, output current, power of the power supply and the switching frequency of the silicon carbide MOSFET.

[0136] In the embodiment of the present invention, power driving is performed using indicator parameter data to minimize power consumption in a safe working environment based on silicon carbide MOS tubes, thereby performing power driving more accurately.

[0137] like Figure 4 , which is a functional module diagram of a power drive system based on silicon carbide MOS tubes provided in one embodiment of the present invention.

[0138] The power drive system 400 based on a silicon carbide MOSFET described in the present invention can be installed in an electronic device. Depending on the functionality implemented, the power drive system 400 based on a silicon carbide MOSFET can include an indicator feature screening module 401, an indicator data fusion module 402, a power consumption calculation module 403, a differential optimization module 404, and a power drive module 405. The modules described in the present invention, also referred to as units, refer to a series of computer program segments that can be executed by an electronic device processor and can perform fixed functions, and are stored in the electronic device's memory.

[0139] In this embodiment, the functions of each module / unit are as follows:

[0140] The indicator feature screening module 401 is used to obtain historical driving data and power supply parameters of the silicon carbide MOS tube, extract indicator feature data of preset indicator features from the historical driving data, and perform feature screening on the indicator features according to the indicator feature data to obtain target indicator features;

[0141] The indicator data fusion module 402 is used to extract the target indicator data corresponding to the target indicator feature from the historical driving data, perform indicator data fusion on the target indicator data, and obtain fused data of each target indicator feature;

[0142] The power consumption calculation module 403 is configured to calculate the power consumption of the silicon carbide MOSFET and the switch power consumption of the silicon carbide MOSFET according to the power supply parameters, and construct a total power consumption according to the power consumption and the switch power consumption;

[0143] The power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube are calculated using the following formula:

[0144]

[0145]

[0146] Among them, P1 represents power consumption, I a , I b Respectively represent the power supply working current, the power supply capacitor working current, R a 、R bRespectively represent the equivalent internal resistance of the power supply and the power supply capacitor in the power supply parameters, P2 represents the switching power consumption, H represents the switching frequency of the silicon carbide MOS tube in the power supply parameters, U represents the electrolytic capacitor voltage at the output end in the power supply parameters, I L Indicates the inductor current, t r , t f Respectively represent the switch rising transition time and the switch falling transition time in the power supply parameters;

[0147] The differential optimization module 404 is configured to generate a parameter particle population based on the fusion data, and perform differential optimization on the parameter particle population based on the total power consumption to obtain an optimized parameter population;

[0148] The power driving module 405 is used to determine the index parameter data of the silicon carbide MOS tube according to the total group of optimization parameters, and to drive the power supply corresponding to the silicon carbide MOS tube according to the index parameter data.

[0149] In detail, each module described in the power drive system 400 based on silicon carbide MOS tube in the embodiment of the present invention adopts the same method as above when in use. Figures 1 to 3 The power driving method based on silicon carbide MOS tube is the same technical means as described in the previous section and can produce the same technical effects, so I will not go into details here.

[0150] The present invention also provides an electronic device, which may include a processor, a memory, a communication bus and a communication interface, and may also include a computer program stored in the memory and run on the processor, such as a power driving method program based on a silicon carbide MOS tube.

[0151] In some embodiments, the processor may be composed of an integrated circuit, for example, it may be composed of a single packaged integrated circuit, or it may be composed of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips.

[0152] The memory includes at least one type of readable storage medium, including a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory may be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device.

[0153] The communication bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable connection and communication between the memory and at least one processor, etc.

[0154] The communication interface is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface may also be a standard wired interface or a wireless interface.

[0155] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0156] Specifically, the specific implementation method of the processor for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0157] In the several embodiments provided herein, it should be understood that the disclosed devices, systems, and methods may be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the module division is merely a logical functional division, and actual implementation may employ other division methods.

[0158] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0159] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0160] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0161] Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the claims are intended to be embraced therein. Any reference to a figure in a claim should not be construed as limiting the claim to which it relates.

[0162] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0163] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems recited in a system claim may also be implemented by a single unit or system through software or hardware. Terms such as "first" and "second" are used to indicate names and do not imply any particular order.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A power driving method based on silicon carbide MOS tube, characterized in that: The method comprises: Obtain historical driving data and power supply parameters of the silicon carbide MOS tube, extract indicator feature data of preset indicator features in the historical driving data, and perform feature screening on the indicator features according to the indicator feature data to obtain target indicator features; wherein, the feature screening on the indicator features according to the indicator feature data to obtain target indicator features includes: dividing the indicator feature data into a minority class data set and a majority class data set, calculating the neighbor data of each minority class data in the minority class data set; performing linear interpolation on the minority class data set according to the neighbor data and the majority class data set to obtain interpolation data corresponding to the indicator feature data; generating interpolation feature data of the indicator feature data according to the interpolation data; calculating the indicator feature weight of each indicator feature data according to the interpolation feature data; and screening the target indicator feature from the indicator features according to the indicator feature weight; Extracting target indicator data corresponding to the target indicator feature from the historical driving data, performing indicator data fusion on the target indicator data, and obtaining fused data of each target indicator feature; Calculating the power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube according to the power supply parameters, and constructing the total power consumption according to the power consumption and the switch power consumption; The power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube are calculated using the following formula: Among them, P1 represents power consumption, I a , I b Respectively represent the power supply working current, the power supply capacitor working current, R a 、R b Respectively represent the equivalent internal resistance of the power supply and the power supply capacitor in the power supply parameters, P2 represents the switching power consumption, H represents the switching frequency of the silicon carbide MOS tube in the power supply parameters, U represents the electrolytic capacitor voltage at the output end in the power supply parameters, I L Indicates the inductor current, t r , t f Respectively represent the switch rising transition time and the switch falling transition time in the power supply parameters; generating a parameter particle population according to the fusion data, performing differential optimization on the parameter particle population according to the total power consumption, and obtaining an optimized parameter population; The index parameter data of the silicon carbide MOS tube is determined according to the total group of optimization parameters, and the power supply corresponding to the silicon carbide MOS tube is driven according to the index parameter data.

2. The power driving method based on silicon carbide MOS tube according to claim 1, characterized in that: The linear interpolation of the minority class data set according to the neighbor data and the majority class data set to obtain interpolation data corresponding to the indicator feature data includes: Determine the boundary data in the minority class data set according to the neighbor data and the majority class data set, and calculate the boundary neighbor data of each boundary data; Performing data interpolation on the boundary neighbor data to obtain minority class samples, and adding the minority class samples to the indicator feature data to obtain minority class indicator data; Data cleaning is performed on the minority class indicator data to obtain interpolation data corresponding to the indicator feature data.

3. The power driving method based on silicon carbide MOS tube according to claim 1, characterized in that: The step of performing indicator data fusion on the target indicator data to obtain fused data of each target indicator feature includes: Using a pre-built multi-layer perceptron to perform data mapping on the target indicator data to obtain a data vector corresponding to the target indicator data; Constructing a data matrix of the target indicator data according to the data vector; The multi-layer perceptron is used to perform data fusion according to the data matrix to obtain fused data of the target indicator characteristics.

4. The power driving method based on silicon carbide MOS tube according to claim 1, characterized in that: The constructing the total power consumption according to the power consumption of the power supply and the power consumption of the switch includes: Obtaining the power supply output current corresponding to the silicon carbide MOS tube, and calculating the power supply power consumption and the current output power consumption corresponding to the switch power consumption according to the power supply output current; Calculating current change power consumption according to the power supply output current; The current output power consumption and the current change power consumption are weighted and summed to obtain the total power consumption.

5. The power driving method based on silicon carbide MOS tube according to claim 4, characterized in that: The calculating the current change power consumption according to the power supply output current includes: Use the following formula to calculate the power consumption due to current change: P3=(I k -I k-1 ) 2 ,k=1,2,…,N Among them, P3 represents the current change power consumption, I k , I k-1 They represent the power supply output current at time k and time k-1 respectively, and N represents a positive integer.

6. The power driving method based on silicon carbide MOS tube according to claim 1, characterized in that: Generating a total swarm of parameter particles according to the fusion data includes: Obtaining a data constraint range corresponding to each target indicator feature in the fused data; Generate constraint data corresponding to each target indicator feature according to the data constraint range; The constraint data are combined to obtain a total swarm of parameter particles.

7. The power driving method based on silicon carbide MOS tube according to claim 1, characterized in that: The differential optimization of the parameter particle population according to the total power consumption to obtain the optimized parameter population includes: Generating a fitness function according to the total power consumption, and calculating the fitness value of each parameter particle in the total parameter particle population; updating the parameter particle swarm according to the fitness value to obtain an updated particle swarm; Mutating and crossovering the updated particle population to obtain an initial optimized population; The initial optimization total group is iterated until the number of iterations is greater than a preset iteration number threshold, thereby obtaining the optimization parameter total group.

8. The power driving method based on silicon carbide MOS tube according to claim 1, characterized in that: The determining of the index parameter data of the silicon carbide MOS tube according to the total group of optimization parameters includes: Calculating the fitness value of each optimization parameter in the total group of optimization parameters; The optimization parameter corresponding to the minimum fitness value is selected as the index parameter data of the silicon carbide MOS tube.

9. A power drive system based on silicon carbide MOS tube, characterized in that: The system comprises: An indicator feature screening module is used to obtain historical driving data and power supply parameters of silicon carbide MOS tubes, extract indicator feature data of preset indicator features in the historical driving data, and perform feature screening on the indicator features according to the indicator feature data to obtain target indicator features; wherein, the feature screening on the indicator features according to the indicator feature data to obtain target indicator features includes: dividing the indicator feature data into a minority class data set and a majority class data set, calculating the neighbor data of each minority class data in the minority class data set; performing linear interpolation on the minority class data set according to the neighbor data and the majority class data set to obtain interpolation data corresponding to the indicator feature data; generating interpolation feature data of the indicator feature data according to the interpolation data; calculating the indicator feature weight of each indicator feature data according to the interpolation feature data; and screening the target indicator feature from the indicator features according to the indicator feature weight; An indicator data fusion module is used to extract target indicator data corresponding to the target indicator feature from the historical driving data, perform indicator data fusion on the target indicator data, and obtain fused data of each target indicator feature; a power consumption calculation module, configured to calculate the power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube according to the power supply parameters, and construct a total power consumption according to the power consumption and the switch power consumption; The power consumption of the silicon carbide MOS tube and the switch power consumption of the silicon carbide MOS tube are calculated using the following formula: Among them, P1 represents power consumption, I a , I b Respectively represent the power supply working current, the power supply capacitor working current, R a 、R b Respectively represent the equivalent internal resistance of the power supply and the power supply capacitor in the power supply parameters, P2 represents the switching power consumption, H represents the switching frequency of the silicon carbide MOS tube in the power supply parameters, U represents the electrolytic capacitor voltage at the output end in the power supply parameters, I L Indicates the inductor current, t r , t f Respectively represent the switch rising transition time and the switch falling transition time in the power supply parameters; a differential optimization module, configured to generate a parameter particle population according to the fusion data, and perform differential optimization on the parameter particle population according to the total power consumption to obtain an optimized parameter population; A power driving module is used to determine the index parameter data of the silicon carbide MOS tube according to the total group of optimization parameters, and to drive the power supply corresponding to the silicon carbide MOS tube according to the index parameter data.

Images