Flexible power distribution method in charging process of vehicle direct current charging equipment

Abstract

The invention belongs to the technical field of new energy vehicle charging equipment management and control, and relates to a flexible power distribution method in the charging process of vehicle direct current charging equipment, which comprises the following steps: S1, starting equipment flexible power unit distribution; s2, sampling a health degree vector sample of the idle charging power unit; s3, calculating an average vector and a covariance of the pattern class samples; s4, judging a health degree mode class of the idle charging power units by using Bayesian classification, starting scheduling distribution of the healthy idle charging power units when the number of the power units needing to be charged is smaller than that of the healthy power units, otherwise, waiting for starting of the healthy idle charging power units, and starting scheduling distribution of the sub-healthy power units; s5, scheduling and allocating charging power unit operation degree vector sampling; s6, calculating scheduling distribution response priorities of the scheduling distribution charging power units; and S7, starting the high-priority charging power unit according to requirements. According to the method, the charging power units are flexibly scheduled according to charging requirements, and the overall cooperative capability, flexibility and effective utilization rate of equipment are improved.

Description

Technical Field

[0001] This invention belongs to the field of new energy vehicle charging equipment management and control technology, and particularly relates to a flexible power distribution method during the charging process of vehicle DC charging equipment. Background Technology

[0002] Currently, megawatt-level charging piles (power ≥ 1MW) are becoming an important development direction for new energy vehicle energy replenishment technology. With the rapid development of new energy electric vehicle battery technology, there are many types of vehicle charging power. Charging equipment has problems such as insufficient effective utilization rate of power units, lack of flexibility and balance, resulting in wasted equipment capacity, weak coordination ability and shortened lifespan. Flexible power distribution technology is usually used for adjustment.

[0003] Common flexible power allocation methods for new energy vehicle charging equipment include ring / star-ring topology, matrix dynamic and fixed allocation, and parameter threshold allocation. Ring / star-ring topology achieves charging power unit scheduling through multi-path connections; however, due to limited scheduling paths, it suffers from poor scheduling flexibility and difficulty in improving equipment utilization. Matrix dynamic and fixed allocation combines fixed and dynamic charging power to achieve charging power unit scheduling; however, due to reliance on fixed units, it results in low overall equipment utilization and insufficient equipment balance. Parameter threshold allocation combines control unit performance parameters with thresholds for scheduling; however, it lacks real-time analysis and prediction of control unit performance, leading to weak equipment coordination capabilities. Summary of the Invention

[0004] The purpose of this invention is to provide a flexible power allocation method for vehicle DC charging equipment during the charging process. Based on the power unitization and full matrix distribution output design of ultra-high power DC group charging equipment for vehicles, this invention solves the problems that common flexible power allocation methods for new energy vehicle charging equipment cannot simultaneously overcome the issues of low effective utilization, poor flexibility, insufficient balance, and weak coordination of power units during the charging process.

[0005] A flexible power distribution method for vehicle DC charging equipment during the charging process, the specific steps of which are as follows:

[0006] S1: The charging equipment connects with the vehicle and starts charging. The vehicle-charging station interacts to enter the charging stage and obtains the vehicle's BCL (Battery Charging Request) message. Based on the vehicle's charging power requirement and the status of the equipment's charging power unit, the equipment's flexible power unit allocation is enabled.

[0007] S2: Based on the charging power unit status information database stored in real time by the charging equipment control unit, the status information database stores the operating status information of all charging power units in real time, deeply analyzes the overall performance of the charging power units of the equipment, and samples the health vector samples of idle charging power units.

[0008] S3: The charging power unit sample information database is stored in the charging equipment control unit. The sample information database is updated and downloaded based on information such as the type and flexibility of the charging power unit. The health samples in the sample information database are divided into three categories: healthy, sub-healthy, and faulty. The control unit obtains the type and flexibility of the charging power unit by interacting with the idle charging power unit.

[0009] S4: In the charging equipment control unit, the health vector x corresponding to the charging power unit in the idle state is respectively... i Substitute into the Bayesian classification decision function b tj (x i ), where when b t12 (x i ) > 0, and b t13 (x i ) > 0 indicates a health mode for the charging power unit module; when b t12 (x i )≤0, and b t23 (x i A value greater than 0 indicates a sub-health mode for the charging power unit module; the rest are identified as unit module failure modes.

[0010]

[0011] b tjk (x i )=b tj (x i )-b tk (x i j,k=1,2,3

[0012] In the formula: P(w tj ) is the pattern class w tj Probability of occurrence;

[0013] Fault-mode charging power units no longer participate in power scheduling and allocation. Based on the vehicle's charging power demand, healthy and sub-healthy charging power units are scheduled and allocated, and priority calculations for whether to participate in scheduling and allocation are determined.

[0014] S5: Based on the charging power unit status information database stored in real time by the charging equipment control unit, deeply analyze the overall operating degree of the charging power units of the equipment, and only sample the vector samples of the operating degree of the charging power units participating in power distribution scheduling.

[0015] S6: Based on the real-time storage of charging power unit status information database in the charging equipment control unit, deeply analyze the priority of the operation of charging power units in the same health mode of the charging equipment.

[0016] S7: Charging power unit scheduling and allocation decision: Based on the vehicle's charging power demand, subtract the total power of the charging power units in the ready-to-start state, and divide the remaining demanded charging power by the unit flexibility to determine the number of charging power units to be scheduled; charging power units of the same health status participating in power allocation scheduling are selected according to the scheduling and allocation response d. j (y i The smaller the value, the higher the priority. The charging power units with the same number of high priorities are selected and put into the ready-to-start state. Finally, the charging power units in the ready-to-start state are started to charge.

[0017] Preferably, the specific content of sampling the health vector of the idle charging power unit in step S2 is as follows:

[0018] Construct a system based on the total failure rate x of the charging power unit during the charging process. i1 Charging power unit failure rate x i2 Operating temperature of charging power unit x i3 Describe the health vector x of the charging power unit of the sub-unit i ;

[0019]

[0020] Where: N ei The cumulative number of faults for the charging power unit at address i, N es The total number of faults for all charging power units, N ri T is the cumulative number of operations for the charging power unit at address i. pi The temperature of the charging power unit at address i. The average temperature of the charging power unit in all standby states, T max This is the current maximum allowable deviation temperature.

[0021] Preferably, step S3 further includes selecting x-vector samples of the corresponding charging power unit type, unit flexibility, and average temperature from the sample information database, and extracting the average vector m of the pattern class samples respectively. tj The covariance matrix C tj The sample information database is updated via OTA or locally, and the average vector m of the pattern class samples is recalculated after each update. tj The covariance matrix C tj ;

[0022]

[0023] In the formula: t is the average temperature value, with an accuracy of 1℃, w j There are three pattern-based sample libraries, N sj It is class w j The number of sample vectors.

[0024] Preferably, in step S4, if the vehicle's charging power demand is greater than the total power of the healthy and sub-healthy charging power units, then all healthy and sub-healthy charging power units will start charging and will no longer participate in the scheduling priority calculation; if the vehicle's charging power demand is only greater than the total power of the healthy charging power units, then all healthy charging power units will be in a ready-to-start state and will no longer participate in the scheduling priority calculation, while the sub-healthy charging power units will participate in the scheduling priority calculation; if the vehicle's charging power demand is less than the total power of the healthy charging power units, then the healthy charging power units will participate in the scheduling priority calculation, while the sub-healthy charging power units will no longer participate in the scheduling priority calculation.

[0025] Preferably, the specific content of step S5, which involves sampling the operating degree vector of the charging power unit participating in power allocation scheduling, is as follows:

[0026] Construct a system based on the percentage of times the charging power unit operates (y). i1 The proportion of output power of the charging power unit y i2 and the percentage of charging power unit operating time y i3 To describe the charging power unit operating degree vector y of the sub-subject i ;

[0027]

[0028] Where: N ri The cumulative number of operations for the charging power unit at address i, N r P is the cumulative number of operations for all charging power units. oi For the cumulative output power of the charging power unit at address i, P o The output power is accumulated for all charging power units, T ri T is the cumulative operating time of the charging power unit at address i. r The total runtime of all charging power units.

[0029] Preferably, the specific content of step S6, which involves in-depth analysis of the operating priority of the health mode charging power unit, is as follows:

[0030] In the sample information database, select the module health mode class, charging power unit type, and charging power unit operation degree weight vector θ corresponding to the charging power unit participating in power allocation scheduling, based on the unit's flexibility. j The scheduling and allocation response is a weighted sum d of the operating degree vectors of the charging power units. j (y i );

[0031]

[0032] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0033] This method, during the startup and charging process of vehicle DC charging equipment, relies on a sample library of charging power units to conduct in-depth analysis of the overall performance of the charging power units. It uses Bayesian decision-making perception to classify idle charging power units into health modes, and combines the scheduling and allocation response of charging power units to determine the priority of charging power units with the same health mode. This enables flexible scheduling of charging power units according to charging demand, improving the overall coordination capability, flexibility, and effective utilization rate of the equipment. At the same time, it can predict fault mode modules in real time, solving the problems of ineffective idle capacity and insufficient balance of charging power units, and extending the service life of the equipment. Detailed Implementation

[0034] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments.

[0035] Example 1

[0036] This embodiment applies to a 1280kW group charging system supporting 16 charging terminals. The flexible charging power units are 40kW each, with a total of 32 units. The system employs a full matrix output distribution design, allowing each charging power unit to reach any charging terminal. Operating parameters are retrieved from the charging power unit status information database. The charging power unit type is UR100040-G6. The total number of failures for all charging power units in the system is 67, the total number of operations is 3612, the total output power is 157.472MW, and the total operating time is 4921 hours. The current charging vehicle requires a voltage of 545V and a current of 120A. The average temperature of all standby charging power units is 17.2℃, the maximum allowable temperature deviation is 7.1℃, and there are 21 standby charging power units. The operating parameters for 7 addresses are as follows:

[0037]

[0038] A flexible power distribution method for vehicle DC charging equipment during the charging process, the specific steps of which are as follows:

[0039] S1: The charging equipment connects with the vehicle and starts charging. The vehicle-charging station interacts to enter the charging stage and obtains the vehicle's BCL (Battery Charging Request) message. Based on the vehicle's charging power requirement and the status of the equipment's charging power unit, the equipment's flexible power unit allocation is enabled.

[0040] S2: Based on the real-time storage of charging power unit status information in the charging equipment control unit, conduct in-depth analysis of the overall performance of the charging power units, and sample the health vector of idle charging power units; construct a system based on the total failure rate of the charging power units during the charging process x i1 Charging power unit failure rate xi2 Operating temperature of charging power unit x i3 Describe the health vector x of the charging power unit of the sub-unit i ;

[0041]

[0042] Where: N ei The cumulative number of faults for the charging power unit at address i, N es The total number of faults for all charging power units, N ri T is the cumulative number of operations for the charging power unit at address i. pi The temperature of the charging power unit at address i. The average temperature of the charging power unit in all standby states, T max This is the current maximum allowable deviation temperature;

[0043] In this embodiment, the health vectors corresponding to the 7 address charging power units selected can be obtained as follows:

[0044] x1 = (0.0448, 0.0211, 0) T x3 = (0, 0, 0.0141) T x6 = (0, 0, 0.0282) T , x7=(0.0149,0.0072,0.0423) T x 14 =(0.0299,0.0163,0.0282) T x 16 =(0.0299,0.0147,0.0141) T x 30 =(0.3134,0.5833,0.0141) T ;

[0045] S3: The charging power unit sample information database is stored in the charging equipment control unit. The health status samples in the database are divided into three categories: healthy, sub-healthy, and faulty. The control unit obtains the type and flexibility of the charging power unit by interacting with idle charging power units. From the sample information database, an x-vector sample is selected with the charging power unit type UR100040-G6, a unit flexibility of 40KW, and an average temperature of 17℃. The temperature is stored separately for each mode type sample with a resolution of 1℃. Each sample information contains 2000 sample values, and the average vector m of each mode type sample is extracted. tj The covariance matrix C tj The sample information database can be updated periodically via OTA on the operation and maintenance platform or locally. Only the average vector m of the pattern class samples needs to be recalculated after each update. tjThe covariance matrix C tj ;

[0046]

[0047] In the formula: t is the average temperature value, with an accuracy of 1℃, w j There are three pattern-based sample libraries, N sj It is class w j The number of sample vectors; for this type of sample, we can obtain:

[0048] m1 = (0.0216, 0.0103, 0.0486) T m2 = (0.091, 0.0423, 0.0711) T m3 = (0.2007, 0.0949, 0.081) T

[0049]

[0050] S4: In the charging equipment control unit, the health vector x corresponding to the 21 charging power units in the idle state is respectively... i Substitute into the Bayesian classification decision function b tj (x i ), where when b t12 (x i ) > 0, and b t13 (x i If b > 0, it is identified as a health mode class for the charging power unit module. t12 (x i )≤0, and b t23 (x i If the value is greater than 0, it is identified as a sub-health mode of the charging power unit module; otherwise, it is identified as a fault mode of the unit module.

[0051]

[0052] b tjk (x i )=b tj (x i )-b tk (x i j,k=1,2,3

[0053] Where P(w) tj The corresponding probabilities in the sample information database are (0.428, 0.314, 0.258). T The Bayesian classification decision function identified the 21 standby charging power units as follows: 14 healthy, 5 sub-healthy, and 2 faulty.

[0054] The Bayesian classification decision function values ​​corresponding to the seven address charging power units selected in this embodiment are shown in the table below:

[0055]

[0056]

[0057] Based on the above data, the following conclusions were drawn: charging module units with addresses 1, 3, 6, 14, and 16 were identified as being in healthy mode; charging module unit with address 7 was identified as being in sub-health mode; and charging module unit with address 30 was identified as being in fault mode. Analysis of the control unit's operating logs revealed that charging module unit with address 7 had a loose communication interface, causing recent communication interruptions; charging module unit with address 30 had an output undervoltage fault and has been replaced and repaired.

[0058] In this embodiment, the current charging vehicle requires a voltage of 545V, a current of 120A, and a charging power of 65.4KW. The total power that can be allocated in the healthy mode is 640KW. Therefore, only the healthy charging power units participate in the scheduling and allocation priority calculation, while the sub-healthy charging power units no longer participate in the scheduling and allocation priority calculation.

[0059] S5: Based on the real-time storage of charging power unit status information in the charging equipment control unit, deeply analyze the overall operating degree of the charging power units in the equipment, and only sample the operating degree vector of the charging power units participating in power allocation and scheduling; construct a system based on the proportion of charging power unit operation times y i1 The proportion of output power of the charging power unit y i2 and the percentage of charging power unit operating time y i3 To describe the charging power unit operating degree vector y of the sub-subject i In this embodiment, only a health mode type charging power unit is constructed;

[0060]

[0061] Where: N ri The cumulative number of operations for the charging power unit at address i, N r P is the cumulative number of operations for all charging power units. oi For the cumulative output power of the charging power unit at address i, P o The output power is accumulated for all charging power units, T ri T is the cumulative operating time of the charging power unit at address i. r Accumulate the running time for all charging power units;

[0062] The health vector of the charging power unit in the health mode type of charging power unit in this embodiment can be obtained as follows:

[0063] y1=(0.0393, 0.0434, 0.0382), y3= (0.0379, 0.0325, 0.0415), y6= (0.0418, 0.0424, 0.0471), y 14 = (0.0341, 0.0315, 0.0348), y 16 = (0.0377, 0.0514, 0.0496)

[0064] S6: Based on the real-time storage of charging power unit status information in the charging equipment control unit, deeply analyze the priority of the operation of charging power units in the same health mode. In the sample information database, select the weight vector θ1 = (0.28, 0.39, 0.33) of the operation of charging power units of type UR100040-G6, with a unit flexibility of 40KW and belonging to the health mode category. Calculate the scheduling allocation response d1(y) of the charging power units in the health mode category. i );

[0065]

[0066] The scheduling and allocation response of the health mode charging power unit in this embodiment can be obtained as follows:

[0067] Where: d1(y1)=0.0406, d1(y3)=0.037, d1(y6)=0.0438, d1(y 14 ) = 0.0333, d1(y 16 S7: In this embodiment, the vehicle's charging power demand is 65.4KW, and the charging power flexibility is 40KW. Therefore, the number of charging power units that need to be scheduled is 2. The scheduling and allocation response d of 14 healthy mode charging power units is as follows: j (y i The ordering shows that addresses 14 and 3 have higher priority, and the corresponding charging power units are activated.

[0068] Example 2

[0069] This embodiment applies to a 640kW group charging system supporting 10 charging terminals. The flexible charging power units are 40kW each, with a total of 16 units. The system employs a full matrix output distribution design, allowing each charging power unit to reach any charging terminal. Operating parameters are retrieved from the charging power unit status information database. The charging power unit type is EVR1000-40000D. The total number of failures for all charging power units in the system is 329, the total number of operations is 10268, the total output power is 598.829MW, and the total operating time is 17743 hours. The current charging vehicle requires a voltage of 680V and a current of 400A. The average temperature of all standby charging power units is 5.2℃, the maximum allowable temperature deviation is 4.1℃, and there are 9 standby charging power units with the following operating parameters:

[0070]

[0071] A flexible power distribution method for vehicle DC charging equipment during the charging process, the specific steps of which are as follows:

[0072] S1: The charging equipment connects with the vehicle and starts charging. The vehicle-charging station interacts to enter the charging stage and obtains the vehicle's BCL (Battery Charging Request) message. Based on the vehicle's charging power requirement and the status of the equipment's charging power unit, the equipment's flexible power unit allocation is enabled.

[0073] S2: Based on the real-time storage of charging power unit status information in the charging equipment control unit, conduct in-depth analysis of the overall performance of the charging power units, and sample the health vector of idle charging power units; construct a system based on the total failure rate of the charging power units during the charging process x i1 Charging power unit failure rate x i2 Operating temperature of charging power unit x i3 Describe the health vector x of the charging power unit of the sub-unit i ;

[0074]

[0075] Where: N ei The cumulative number of faults for the charging power unit at address i, N es The total number of faults for all charging power units, N ri T is the cumulative number of operations for the charging power unit at address i. pi The temperature of the charging power unit at address i. The average temperature of the charging power unit in all standby states, T max This is the current maximum allowable deviation temperature;

[0076] In this embodiment, the health vector corresponding to the idle charging power unit can be obtained as follows:

[0077] x1 = (0.0061, 0.0029, 0.0244) T x2 = (0.0091, 0.0045, 0) T , x3=(0.0426,0.0237,0.0488) T , x6=(0.0334,0.0196,0.0488) T x8 = (0, 0, 0.0244) T x 10 =(0.0486,0.0266,0.0488) T x 12 =(0,0,0.0244) T x 13 =(0.0213,0.0110,0) T x 15 =(0.0547,0.0308,0.0244) T , ;

[0078] S3: The charging power unit sample information database is stored in the charging equipment control unit. The health status samples in the database are divided into three categories: healthy, sub-healthy, and faulty. The control unit obtains the type and flexibility of the charging power unit by interacting with idle charging power units. It selects x-vector samples from the sample information database that have a charging power unit type of EVR1000-40000D, a unit flexibility of 40KW, and an average temperature of 5.2℃. Temperature information is stored separately for each mode type sample with a resolution of 1℃. Each sample information contains 4200 sample values. The average vector m of each mode type sample is then extracted. tj The covariance matrix C tj The sample information database can be updated periodically via OTA on the operation and maintenance platform or locally. Only the average vector m of the pattern class samples needs to be recalculated after each update. tj The covariance matrix C tj ;

[0079]

[0080]

[0081] In the formula: t is the average temperature value, with an accuracy of 1℃, w j There are three pattern-based sample libraries, N sj It is class w j The number of sample vectors; for this type of sample, we can obtain:

[0082] m1 = (0.0114, 0.0037, 0.0841) T, m2=(0.0391,0.0106,0.1768) T , m3=(0.0967,0.0239,0.2146) T

[0083]

[0084] S4: In the charging equipment control unit, the health vector x corresponding to the 9 charging power units in the idle state is respectively... i Substitute into the Bayesian classification decision function b tj (x i ), where when b t12 (x i ) > 0, and b t13 (x i If b > 0, it is identified as a health mode class for the charging power unit module. t12 (x i )≤0, and b t23 (x i If the value is greater than 0, it is identified as a sub-health mode of the charging power unit module; otherwise, it is identified as a fault mode of the unit module.

[0085]

[0086] b tjk (x i )=b tj (x i )-b tk (x i j,k=1,2,3

[0087] Where P(w) tj The corresponding probabilities in the sample information database are (0.389, 0.315, 0.296). T The Bayesian classification decision function identified the 21 standby charging power units as follows: 5 healthy, 3 sub-healthy, and 1 faulty.

[0088] The Bayesian classification decision function values ​​corresponding to the idle charging power units in this embodiment are shown in the table below:

[0089]

[0090] Based on the above data, the following conclusions were drawn: charging module units with addresses 1, 2, 8, 12, and 13 were identified as being in healthy mode; charging module units with addresses 3, 6, and 10 were identified as being in sub-health mode; and charging module unit with address 15 was identified as being in fault mode. Analysis of the control unit's operating log revealed that charging module unit with address 15 had a high-voltage output protection above 500V, and it has been replaced and repaired.

[0091] In this embodiment, the current charging vehicle requires a voltage of 630V, a current of 400A, and a charging power of 272KW. The total allocable power for the healthy mode is 200KW, and the total allocable power for the sub-healthy mode is 120KW. Since the allocable power for the healthy mode is less than the required power, all charging power units in the healthy mode are in a ready-to-start state and no longer participate in the scheduling and allocation priority calculation. Charging power units in the sub-healthy mode participate in the scheduling and allocation priority calculation.

[0092] S5: Based on the real-time storage of charging power unit status information in the charging equipment control unit, deeply analyze the overall operating degree of the charging power units in the equipment, and only sample the operating degree vector of the charging power units participating in power allocation and scheduling; construct a system based on the proportion of charging power unit operation times y i1 The proportion of output power of the charging power unit y i2 and the percentage of charging power unit operating time y i3 To describe the charging power unit operating degree vector y of the sub-subject i In this embodiment, only a health mode type charging power unit is constructed;

[0093]

[0094] Where: N ri The cumulative number of operations for the charging power unit at address i, N r P is the cumulative number of operations for all charging power units. oi For the cumulative output power of the charging power unit at address i, P o T is the cumulative output power of all charging power units. ri T is the cumulative operating time of the charging power unit at address i. r Accumulate the running time for all charging power units;

[0095] In this embodiment, the health vector of the charging power unit in the sub-health mode can be obtained as follows:

[0096] y3=(0.0576, 0.0417, 0.0400), y6= (0.0547, 0.0593, 0.0538), y 10 = (0.0586, 0.0585, 0.0543)

[0097] S6: Based on the real-time storage of charging power unit status information in the charging equipment control unit, deeply analyze the priority of the operation of charging power units in the same health mode. In the sample information database, select the weight vector θ1 = (0.31, 0.39, 0.30) of the operation of charging power units of type EVR1000-40000D, with a unit flexibility of 40KW and in the health mode category, and calculate the scheduling allocation response d1(y) of the charging power units in the health mode category. i );

[0098]

[0099] The scheduling and allocation response of the health mode charging power unit in this embodiment can be obtained as follows:

[0100] Where: d2(y3)=0.0461, d2(y6)=0.0563, d2(y 10 ) = 0.0573;

[0101] S7: In this embodiment, the vehicle's charging power demand is 272KW, and the charging power flexibility is 40KW. Therefore, the number of charging power units that need to be scheduled is 7, and the scheduling and allocation response d for 3 sub-optimal charging power units is as follows: j (y i The sorting results show that addresses 3 and 6 have higher priority, so all healthy mode charging power units and sub-health mode charging power units with addresses 3 and 6 are activated.

[0102] Through actual verification with the Fenxi Electronic Cluster Charging Equipment, the effective utilization rate of charging power capacity under full load is ≥97.1%, and the accuracy rate of charging power unit fault warning is ≥92.1%. The algorithm is designed in the embedded microprocessor chip STM32F413, and the algorithm lag time is ≤110ms, which meets the real-time decision requirements of the charging equipment. At the same time, it overcomes the problems of low effective utilization rate, poor flexibility, insufficient balance and weak coordination of power units in the charging process.

[0103] The above embodiments are merely preferred technical solutions of the present invention. Any simple substitutions made by those skilled in the art based on the content of the present invention specification should fall within the patent protection scope of the present invention.

Claims

1. A flexible power distribution method for vehicle DC charging equipment during the charging process, characterized in that, The specific steps are as follows: S1: The charging equipment connects with the vehicle and starts charging. The vehicle-charging station interacts to enter the charging stage and obtains the vehicle's BCL (Battery Charging Request) message. Based on the vehicle's charging power requirement and the status of the equipment's charging power unit, the equipment's flexible power unit allocation is enabled. S2: Based on the real-time storage of charging power unit status information database in the charging equipment control unit, deeply analyze the overall performance of the charging power units of the equipment and sample the health vector of idle charging power units. S3: The charging power unit sample information database is stored in the charging equipment control unit. The sample information database is updated and downloaded based on information such as the type and flexibility of the charging power unit. The health samples in the sample information database are divided into three categories: healthy, sub-healthy, and faulty. The control unit obtains the type and flexibility of the charging power unit by interacting with the idle charging power unit. S4: In the charging equipment control unit, the charging power unit corresponding health degree vector x i is substituted into the Bayesian classification decision function b tj (x i ) t12 (x i ) > 0, and b t13 (x i ) > 0 are identified as the charging power unit module health mode class; when b t12 (x i ) ≤ 0, and b t23 (x i ) > 0 are identified as the charging power unit module sub-health mode class; The rest are identified as unit module failure mode classes; b tjk (x i )=b tj (x i )-b tk (x i )j ,k=1,2,3, In the formula: P(w tj ) is the pattern class w tj Probability of occurrence; Fault-mode charging power units no longer participate in power scheduling and allocation. Based on the vehicle's charging power demand, healthy and sub-healthy charging power units are scheduled and allocated, and priority calculations for whether to participate in scheduling and allocation are determined. S5: Based on the charging power unit status information database stored in real time by the charging equipment control unit, deeply analyze the overall operating degree of the charging power units of the equipment, and only sample the vector samples of the operating degree of the charging power units participating in power distribution scheduling. S6: Based on the real-time storage of charging power unit status information in the charging equipment control unit, deeply analyze the priority of the operation of charging power units in the same health mode of the charging equipment. S7: Charging power unit scheduling and allocation decision: Based on the vehicle's charging power demand, subtract the total power of the charging power units in the ready-to-start state, and divide the remaining demanded charging power by the unit flexibility to determine the number of charging power units to be scheduled; charging power units of the same health status participating in power allocation scheduling are selected according to the scheduling and allocation response d. j (y i The smaller the value, the higher the priority. The charging power units with the same number of high priorities are selected and put into the ready-to-start state. Finally, the charging power units in the ready-to-start state are started to charge.

2. The flexible power distribution method for vehicle DC charging equipment during charging according to claim 1, characterized in that, The specific details of sampling the health vector of the idle charging power unit in step S2 are as follows: Construct a system based on the total failure rate x of the charging power unit during the charging process. i1 Charging power unit failure rate x i2 Operating temperature of charging power unit x i3 Describe the health vector x of the charging power unit of the sub-unit i ; Where: N ei The cumulative number of faults for the charging power unit at address i, N es The total number of faults for all charging power units, N ri T is the cumulative number of operations for the charging power unit at address i. pi The temperature of the charging power unit at address i. The average temperature of the charging power unit in all standby states, T max This is the current maximum allowable deviation temperature.

3. The flexible power distribution method for vehicle DC charging equipment during charging according to claim 1, characterized in that, Step S3 further includes selecting x-vector samples of the corresponding charging power unit type, unit flexibility, and average temperature from the sample information database, and extracting the average vector m of the pattern class samples respectively. tj The covariance matrix C tj ; The sample information database is updated via OTA or locally, and the average vector m of the pattern class samples is recalculated after each update. tj The covariance matrix C tj ; In the formula: t is the average temperature value, with an accuracy of 1℃, w j There are three pattern-based sample libraries, N sj It is class w j The number of sample vectors.

4. The flexible power distribution method for vehicle DC charging equipment during charging according to claim 1, characterized in that, In step S4, if the vehicle's charging power demand is greater than the total power of the healthy and sub-healthy charging power units, all healthy and sub-healthy charging power units will start charging and will no longer participate in the scheduling priority calculation; if the vehicle's charging power demand is only greater than the total power of the healthy charging power units, all healthy charging power units will be in a ready-to-start state and will no longer participate in the scheduling priority calculation, while the sub-healthy charging power units will participate in the scheduling priority calculation; if the vehicle's charging power demand is less than the total power of the healthy charging power units, the healthy charging power units will participate in the scheduling priority calculation, while the sub-healthy charging power units will no longer participate in the scheduling priority calculation.

5. A flexible power distribution method for vehicle DC charging equipment during charging according to claim 1, characterized in that, The specific details of step S5, which involves sampling the operating degree vector of the charging power unit participating in power allocation scheduling, are as follows: Construct a system based on the percentage of times the charging power unit operates (y). i1 The proportion of output power of the charging power unit y i2 and the percentage of charging power unit runtime y i3 To describe the charging power unit operating degree vector y of the sub-subject i ; Where: N ri The cumulative number of operations for the charging power unit at address i, N r P is the cumulative number of operations for all charging power units. oi For the cumulative output power of the charging power unit at address i, P o The output power is accumulated for all charging power units, T ri T is the cumulative operating time of the charging power unit at address i. r The total runtime of all charging power units.

6. The flexible power distribution method for vehicle DC charging equipment during charging according to claim 1, characterized in that, The specific details of step S6, which involves in-depth analysis of the operational priority of the health mode charging power unit, are as follows: In the sample information database, select the module health mode class, charging power unit type, and charging power unit operation degree weight vector θ corresponding to the charging power unit participating in power allocation scheduling, based on the unit's flexibility. j The scheduling and allocation response is a weighted sum d of the operating degree vectors of the charging power units. j (y i );

Images