Fuzzy c-means-based method and device for identifying electric vehicle charging behavior in a transformer area

Abstract

The application discloses a kind of based on fuzzy C mean's district area electric vehicle charging behavior identification method and device, method includes the following steps: constructing the spatial data matrix and phase data matrix of electric vehicle charging;Set fuzzy value;With the initialization district area outgoing line load data fuzzy clustering center matrix and three-phase load data fuzzy clustering center matrix are constructed;Using fuzzy clustering analysis and through iterative calculation, determine optimal fuzzy clustering matrix and optimal fuzzy clustering center matrix;Obtain district area outgoing line load data fuzzy clustering center matrix optimal value and three-phase load data fuzzy clustering center matrix optimal value;Calculate the average value of district area outgoing line and three-phase load data power;According to power average value, identify three-phase electric vehicle charging power, confirm charging behavior.This method can identify district area electric vehicle charging behavior, determine the outgoing line and phase of electric vehicle in charging, provide theoretical guidance for district area operation management, charging station configuration, load prediction, control scheduling.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of electric vehicle charging behavior recognition, and particularly relates to a distribution area electric vehicle charging behavior recognition method and device based on fuzzy C-means. BACKGROUND

[0002] In a modern power system, a distribution area is a kind of power grid form in which distributed sources (small hydropower, small wind power, photovoltaic power generation) and loads (water, electricity, gas, cold, heat load) are integrated in a certain way within the distribution area. Electric vehicles and energy storage devices have dual attributes, which can be a kind of distributed power source and a kind of distributed load. The distribution area is connected to the main power grid at voltage levels of 380V, 10kV, 35kV, etc. Under normal operating conditions, the distribution area is connected to the main power grid in parallel, absorbs power from the main power grid under heavy load, and may inject power into the main power grid under light load. In the case of local faults in the main power grid or in the case of faults in adjacent distribution areas, the distribution area can operate in an isolated grid. On the premise of ensuring the quality of electric energy, the distribution area provides electric power to the load by the distributed power source within the distribution area, realizes the normal power supply state of the fault-free distribution area, reduces the power outage time, and improves the power supply reliability.

[0003] In a family, the charging and discharging power of electric vehicles and energy storage devices is small, and is generally connected in single phase at a voltage level of 220V. The charging power is not greater than 10kW. However, due to different phase sequence connections of different families and distribution areas, and the randomness of family charging, different charging characteristics are presented in different spatial areas, different phase sequences of distribution areas, and different time scales, such as: different charging powers of distribution areas facing different areas to form charging spatial characteristics; different charging powers of A, B and C three-phase to form charging phase characteristics; different charging powers of 1-24 time period to form charging time characteristics. Due to the charging and discharging of family electric vehicles and energy storage devices in single phase, combined with the significant differences in space, phase and time characteristics, the imbalance of three-phase power of the distribution area is often caused, and this situation is more serious when the management is not in place, which has an adverse effect on the operation of the distribution network and the distribution area. In a public charging station, the charging and discharging power of electric vehicles and energy storage devices is large, and is generally connected in three-phase at a voltage level of 380V, which will not cause the imbalance of three-phase power of the distribution area. However, due to the randomness of charging, the distribution area facing the public charging station will also form significant differences in charging space, phase and time characteristics.

[0004] The distribution area for distributed small hydropower is a distribution area mainly in the form of small hydropower supply. In the water, wind and light distribution area, most of the hydropower stations are runoff type, the dam generally has no water storage function, the reservoir has no water storage and water regulation capacity, the water energy utilization of the small hydropower station completely depends on the water value of the reservoir, and the generation state and output scale of the small hydropower unit also completely depend on the water value of the reservoir. In this case, in order to realize efficient utilization of water energy generation of small hydropower station, it is necessary to achieve the generation of water. The water value of the small hydropower station reservoir has randomness, and the water value in different hydrological periods is completely different. In the wet season, the water value is large, and in the dry season, the water value is small. Therefore, the river flow value of the small hydropower station basin often shows the minimum flow value, the maximum flow value, the average flow value, the multi-year average flow value, the calculated average flow value, the weighted average flow value, the mathematical average flow value and other table forms. Different flow value table forms are adopted, and different installed capacity values of the small hydropower station are obtained. Different installed capacity values often result in different power generation and power generation values of the small hydropower station in different hydrological periods, and the optimal leads to different water energy utilization rates, power generation equipment utilization rates and annual maximum utilization hours of power generation equipment of the small hydropower station.

[0005] The small hydropower-wind power station is a kind of station which integrates small hydropower and small wind power, certain capacity load and connects the distributed power and load in a certain way. In the small hydropower-wind power station, not only the reservoir inflow value, reservoir flow value and power generation flow value of the small hydropower station have uncertainty and randomness, but also the wind speed of the small wind power station has uncertainty and randomness. In the water, wind and light power station, most of the hydropower stations are runoff type, the dam generally has no water storage function, the reservoir has no water storage and water regulation capacity, and the water energy utilization of the small hydropower station completely depends on the reservoir inflow value. In this case, the small hydropower station must generate power according to the amount of water to realize efficient utilization of water energy. The reservoir inflow value of the small hydropower station has randomness, and the inflow value is completely different in different hydrological periods. The inflow value is large in the wet season and small in the dry season. Therefore, the river flow value of the small hydropower station often shows the minimum flow value, the maximum flow value, the average flow value, the multi-year average flow value, the calculated average flow value, the weighted average flow value, the mathematical average flow value and the like. The small hydropower station will obtain different installed capacity levels by using different flow value forms. Different installed capacity levels often lead to different power generation power and power generation value of the small hydropower station in different hydrological periods, and the optimal leads to different water energy utilization rate, power generation equipment utilization rate and annual maximum utilization hours of power generation equipment of the small hydropower station. When the wind speed is less than the cut-in wind speed or greater than the cut-out wind speed, the wind turbine has no output power. When the wind speed is less than the rated wind speed for the cut-in wind speed, the wind turbine output power is less than the rated power. When the wind speed is less than the cut-out wind speed for the rated wind speed, the wind turbine outputs the rated power. The wind speed is completely different in different seasons of a year and completely different in different time periods of a day, and the wind speed has randomness, volatility and intermittency. Therefore, the wind speed of the small wind power station often shows the minimum wind speed, the maximum wind speed, the average wind speed, the multi-year average wind speed, the calculated average wind speed, the weighted average wind speed, the mathematical average wind speed and the like. The small wind power station will obtain different installed capacity levels by using different wind speed forms. Different installed capacity levels often lead to different power generation power and power generation value of the small wind power station in different seasons, and the optimal leads to different wind energy utilization rate, power generation equipment utilization rate and annual maximum utilization hours of power generation equipment of the small wind power station.

[0006] The small hydropower-wind-solar power station is a kind of power station which integrates three kinds of distributed power sources, i.e. small hydropower, small wind power and solar power, and connects the distributed power sources and loads in a certain way. In the small hydropower-wind-solar power station, not only the water inflow value, reservoir flow value and power generation flow value of the small hydropower station, but also the wind speed of the small wind power station and the solar intensity have uncertainty and randomness. In the small hydropower-wind-solar power station, most of the hydropower stations are runoff type, the dam generally has no water storage function, and the reservoir has no water storage and water regulation capacity. The water energy utilization of the small hydropower station completely depends on the water inflow value of the reservoir, and the generation state and output scale of the small hydropower unit also completely depend on the water inflow value of the reservoir. In this case, in order to realize the efficient utilization of water energy, the small hydropower station must generate electricity according to the water inflow value. However, the water inflow value of the small hydropower station has randomness, and the water inflow value is completely different in different hydrological periods. The water inflow value is large in the wet season and small in the dry season. Therefore, the river flow value of the small hydropower station often shows the minimum flow value, maximum flow value, average flow value, multi-year average flow value, calculated average flow value, weighted average flow value, mathematical average flow value and other table forms. By using different flow value table forms, the small hydropower station will obtain different installed capacity levels. Different installed capacity levels often lead to different power generation power and power generation value of the small hydropower station in different hydrological periods, and the optimal leads to different water energy utilization rate, power generation equipment utilization rate and annual maximum utilization hours of power generation equipment of the small hydropower station. When the wind speed is less than the cut-in wind speed or greater than the cut-out wind speed, the wind turbine has no output power. When the wind speed is less than the rated wind speed for the cut-in wind speed, the wind turbine output power is less than the rated power. When the wind speed is less than the cut-out wind speed for the rated wind speed, the wind turbine outputs the rated power. The wind speed is completely different in different seasons of a year and completely different in different time periods of a day, and the wind speed has randomness, volatility and intermittency. Therefore, the wind speed of the small wind power station often shows the minimum wind speed, maximum wind speed, average wind speed, multi-year average wind speed, calculated average wind speed, weighted average wind speed, mathematical average wind speed and other table forms. By using different wind speed table forms, the small wind power station will obtain different installed capacity levels. Different installed capacity levels often lead to different power generation power and power generation value of the small wind power station in different seasons, and the optimal leads to different wind energy utilization rate, power generation equipment utilization rate and annual maximum utilization hours of power generation equipment of the small wind power station. The greater the solar intensity, the greater the output power of the photovoltaic power generation system. The solar intensity is completely different in different seasons of a year and completely different in different time periods of a day, and the solar intensity has randomness, volatility and intermittency. Therefore, the solar intensity of the photovoltaic power station often shows the minimum solar intensity, maximum solar intensity, average solar intensity, multi-year average solar intensity, calculated average solar intensity, weighted average solar intensity, mathematical average solar intensity and other table forms. By using different solar intensity table forms, the photovoltaic power station will obtain different installed capacity levels.The power generation and value of photovoltaic power station in different seasons are different at different installed capacity levels, and the optimal causes the wind energy utilization rate, power generation equipment utilization rate, and annual maximum utilization hours of power generation equipment to be different.

[0007] The integration of different load levels and distributed power capacity scales in the transformer area changes the structure and power flow characteristics of the transformer area. Due to the access of small hydropower, small wind power, photovoltaic power and other distributed power sources, different voltage levels will be adopted due to the different capacity scales of the accessed power sources. Due to the randomness of electricity consumption, the load power will change at different time and space scales, and has obvious time period characteristics. At the same time, the output of wind power, photovoltaic power and other distributed power sources has intermittency, randomness and time period characteristics, and the output of small hydropower units has seasonality. Therefore, the balance between the load power and the power of the transformer area is often difficult to maintain. When the load power is greater than the power of the transformer area, the transformer area needs to obtain supplementary power from the main power grid, and when the load power is less than the power of the transformer area, the remaining power of the transformer area needs to be injected into the main power grid, forming a random bidirectional power flow characteristic. The random bidirectional power flow characteristic has an impact on the voltage quality of the nodes in the transformer area. When the output of distributed power is large and the load is light, the node voltage in the local area of the transformer area will be high, and when the output of distributed power is small and the load is heavy, the node voltage in the local area of the transformer area will be low. Therefore, the limitation and requirement of the node voltage in the transformer area have an impact and restriction on the configuration, operation mode and voltage control strategy of the distributed power capacity in the transformer area, and the configuration, operation mode and voltage control strategy of the distributed power capacity in the transformer area need to consider the limitation and requirement of the node voltage in the transformer area. The node access of the transformer area in the distribution network of different voltage levels will cause the node voltage of the distribution network to be high or low due to the different sizes of the power absorbed or injected by the transformer area from the distribution network, and the configuration, operation mode and voltage control strategy of the distributed power capacity in the transformer area need to consider the limitation and requirement of the node voltage of the distribution network.

[0008] The distribution type power supply system of a transformer area is a system with complex relations and interaction of random and fuzzy uncertainty events or parameters. Under the influence of various random and fuzzy events or parameters, the power generation and generation value of the distribution type power supply system of a transformer area become more random and fuzzy, which greatly affects the capacity configuration of the distribution type power supply system of a transformer area. In the past, the power generation and generation value of the distribution type power supply system of a transformer area were usually calculated by a deterministic method, and some were calculated by a probabilistic analysis method. The deterministic method usually calculates the power generation, generation value and installed capacity of the distribution type power supply system of a transformer area under the assumption that the inflow value and flow value of a small hydropower station, the regional sunshine intensity and wind speed are determined, and the influence of factors such as transformer area and distribution network voltage regulation requirements and flexible control mode is not considered. The calculation result is unique and deterministic, and often cannot reflect the actual situation of the power generation, generation value and installed capacity of the distribution type power supply system of a transformer area. The probabilistic analysis method usually calculates the power generation, generation value and installed capacity of the distribution type power supply system of a transformer area under the assumption that only a single factor such as the inflow value and flow value of a small hydropower station, the regional sunshine intensity and wind speed is an uncertain factor. The calculation result is a probability value with a certain confidence level. In fact, the power generation, generation value and installed capacity of the distribution type power supply system of a transformer area are affected by multiple uncertain factors. Moreover, these factors usually have random uncertainty or fuzzy uncertainty, or both random and fuzzy uncertainty, and often exist as random and fuzzy uncertainty events or parameters. Therefore, the existing technology for calculating the power generation, generation value and installed capacity of the distribution type power supply system of a transformer area does not fully consider the uncertainty and randomness of the influencing factors, and the applicability, practicality and applicability of the calculation method are difficult to meet.

[0009] Under the interaction and common influence of distributed power supply, electric vehicles and energy storage devices, and the random and fuzzy uncertainty of load, the peak and valley load difference is often formed in the transformer area, and the peak and valley load space, phase and time characteristics are also significantly different. In order to identify these characteristics, the traditional method is to use artificial method to verify the user information of the transformer area through regular inspection. The traditional method lacks real-time performance and is time-consuming and laborious. For the transformer area facing old streets, the line is complex, and the information cannot be updated in time, which causes the transformer area to have the problems of incomplete, imperfect and missing information. From the perspective of transformer area management, technicians often change the user connection mode, or need to modify the line to balance the transformer area load, which easily leads to unclear user line relationship and inaccurate attribution relationship record. These factors increase the difficulty and complexity of identifying the interaction and common influence characteristics of the distribution type power supply, electric vehicles and energy storage devices of a transformer area.

[0010] With the increasing number of users in the transformer area, the traditional method of manual transcription has been difficult to adapt. Using special recognition equipment has become a common practice, such as carrier communication method and pulse current method. Carrier signal and pulse signal are easy to be disturbed, and transformer is a greater obstacle to communication. Therefore, the feature recognition technology based on carrier communication method and pulse current method can only be applied in the same phase line range of the same transformer area.

[0011] With the increasing maturity of big data and artificial intelligence technology, data mining and artificial intelligence technology are gradually applied in the interaction and common influence feature recognition of distributed power supply, electric vehicles and energy storage devices in the transformer area, such as: obtaining transformer area and user data through smart meters, identifying the transformer area and phase of the user by using data mining method; using transformer voltage data, using data mining method to identify the space-time correlation of equipment. However, due to the difference in gray correlation, the recognition accuracy is not high in some cases. In the case of very limited voltage time series data, and without considering the random and fuzzy uncertainty of distributed power generation, electric vehicles and energy storage devices charging and discharging and load, it is difficult to identify the dynamic state and dynamic characteristics of the transformer area based on big data and artificial intelligence feature recognition technology, and it is more difficult to identify the peak-valley load characteristics of the transformer area. SUMMARY

[0012] The main purpose of the present application is to overcome the shortcomings and deficiencies of the prior art, and provide a transformer area electric vehicle charging behavior recognition method and device based on fuzzy C mean. The method constructs a spatial data matrix and a phase data matrix of electric vehicle charging from historical and real-time data, and constructs and initializes a fuzzy clustering center matrix by using fuzzy C mean clustering analysis method. Then, the optimal value is obtained by iterative calculation, and the average value of the load power is calculated. Finally, the three-phase charging power is identified according to the average value, and the outgoing line and phase of the electric vehicle during charging are confirmed, which provides theoretical guidance for transformer area operation management, charging station configuration, load prediction and control scheduling.

[0013] In order to achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0014] The present application provides a transformer area electric vehicle charging behavior recognition method based on model C mean, which comprises the following steps:

[0015] According to historical and real-time data, a spatial data matrix and a phase data matrix of electric vehicle charging are constructed;

[0016] Set the fuzzy value; the fuzzy value c includes: minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value;

[0017] Fuzzy clustering analysis is used to construct and initialize the fuzzy clustering center matrix of the transformer area outgoing line load data and the fuzzy clustering center matrix of the three-phase load data;

[0018] The optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix are determined through iterative calculation;

[0019] The optimal value of the fuzzy clustering center matrix of the substation outlet load data is obtained;

[0020] The optimal value of the fuzzy clustering center matrix of the three-phase load data of the substation outlet is obtained;

[0021] The average value of the power of the substation outlet load data and the three-phase load data is calculated;

[0022] According to the average value of the three-phase load data power, the three-phase electric vehicle charging power is identified, and the charging behavior is confirmed.

[0023] As a preferred technical solution, the construction of the spatial data matrix and the phase data matrix of electric vehicle charging is specific to:

[0024] The substation outlet load active power data and the three-phase load active power data are obtained from historical and real-time data;

[0025] By processing, calculating and analyzing the substation outlet load active power data, the spatial data matrix X S of electric vehicle charging is constructed:

[0026]

[0027] Wherein, x it is the load active power of the substation outlet i at time period t, i=1,2,...,N L , t=1,2,...,T; N L is the number of substation outlets; T is the number of time periods representing the substation outlet load active power data;

[0028] By processing, calculating and analyzing the substation outlet three-phase load active power data, the phase data matrix of electric vehicle charging is constructed, respectively X A , X B , X C , which is represented as:

[0029]

[0030]

[0031]

[0032] Wherein, X A , X B , X C are the A, B, C phase data matrices of the electric vehicle charging of the substation outlet i, x Ait , xBit , x Cit are the active power of A, B, C phase of the feeder i at time period t, i = 1, 2, …, N L , t = 1, 2, …, T.

[0033] As a preferred technical solution, the construction and initialization of the fuzzy clustering center matrix of the feeder load data and the fuzzy clustering center matrix of the three-phase load data are specific to:

[0034] The fuzzy clustering center matrix of the active power of the feeder i load is constructed by fuzzy clustering analysis:

[0035] C Si = {C Si1 , C Si2 , C Si3 , …, C Sic}

[0036] Where C Si1 , C Si2 , C Si3 , C Si4 , C Si5 , C Si6 , C Si7 , C Si8 , C Si9 are the minimum, minimum, small, small, median, small, large, maximum, maximum fuzzy clustering center matrix of the active power of the feeder i load

[0037] Let C Sijt = [C Si1t , C Si2t , …, C Sict ] T , (j = 1, 2, …, c), T is a row-to-column operation; for indicating the minimum, minimum, small, small, median, small, large, maximum, maximum fuzzy clustering center matrix of the active power of the feeder i at time period t;

[0038] Let X MSit be the multi-year average of the active power of the feeder i at time period t, then the initial value of the minimum, minimum, small, small, median, small, large, maximum, maximum fuzzy clustering center matrix of the active power of the feeder i at time period t is:

[0039]

[0040] Where k j represents the coefficient corresponding to the fuzzy value;

[0041] Fuzzy clustering analysis was used to construct the fuzzy clustering center matrix of the active power of phases A, B, and C of the outgoing line i in the transformer substation:

[0042] C Ai ={C Ai1 C Ai2 C Ai3 …,C Aic}

[0043] C Bi ={C Bi1 C Bi2 C Bi3 …,C Bic}

[0044] C Ci ={C Ci1 C Ci2 C Ci3 …,C Cic}

[0045] Among them, C Ai1 C Ai2 C Ai3 C Ai4 C Ai5 C Ai6 C Ai7 C Ai8 C Ai9 C is the fuzzy clustering center matrix for the minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power of the A-phase load of transformer substation i; Bi1 C Bi2 C Bi3 C Bi4 C Bi5 C Bi6 C Bi7 C Bi8 C Bi9 The fuzzy clustering center matrix represents the minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power values ​​of the B-phase load of transformer substation i; C Ci1 C Ci2 C Ci3 C Ci4 C Ci5 C Ci6 C Ci7 C Ci8 C Ci9 The fuzzy clustering center matrix represents the minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power of the C-phase load of transformer area i.

[0046] Let C Aijt =[CAi1t , Ai2t , Aict ] T , (j = 1, 2,..., c), for representing the minimum, minimum, small, small, median, small, large, maximum, maximum fuzzy clustering center matrix of the A-phase load active power of the substation outgoing line i at the time period t;

[0047] Let C Bijt = [C Bi1t , C Bi2t ,..., C Bict ] T , (j = 1, 2,..., c), for representing the minimum, minimum, small, small, median, small, large, maximum, maximum fuzzy clustering center matrix of the B-phase load active power of the substation outgoing line i at the time period t;

[0048] Let C Cijt = [C Ci1t , C Ci2t ,..., C Cict ] T , (j = 1, 2,..., c), for representing the minimum, minimum, small, small, median, small, large, maximum, maximum fuzzy clustering center matrix of the C-phase load active power of the substation outgoing line i at the time period t;

[0049] Let the multi-year average of the three-phase load active power of the substation outgoing line i at the time period t be X MAit , X MBit , X MCit , respectively, then the initial values of the minimum, minimum, small, small, median, small, large, maximum, maximum fuzzy clustering center matrix of the three-phase load active power of the substation outgoing line i at the time period t are:

[0050]

[0051]

[0052]

[0053] Wherein, k j represents the coefficient corresponding to the fuzzy value.

[0054] As a preferred technical solution, the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix are determined by iterative calculation, specifically:

[0055] The iterative calculation is performed using the charging load active power data;

[0056] First set the fuzzy coefficient f, set the iterative calculation end error value ε, set the iteration number t;

[0057] The random number is randomly generated on [0, 1] by using a random function to initialize the fuzzy clustering matrix U (0) ;

[0058] Iterate and calculate the element value of the fuzzy clustering matrix, in the tth iteration, the element value of the fuzzy clustering matrix U (t) The update formula of the element value of the fuzzy clustering matrix is:

[0059]

[0060] Wherein, d() is the distance function;

[0061] The element value of the fuzzy clustering center matrix C (t) The minimum value, average value, maximum value of the element value of the fuzzy clustering center matrix The iterative update formula is:

[0062]

[0063] Wherein, i=1, 2,..., c, j=1, 2,..., m;

[0064] If The iteration calculation is ended, otherwise the next period t=t+1 is entered to continue the iteration calculation; after the iteration calculation is completed, the optimal value of the fuzzy clustering matrix is obtained as:

[0065]

[0066] The optimal value of the fuzzy clustering center matrix is obtained as:

[0067]

[0068] As a preferred technical scheme, the optimal value of the fuzzy clustering center matrix of the obtained transformer area outgoing line load data is specifically:

[0069] Using the load active power data, the optimal value of the fuzzy clustering center matrix of the minimum value, minimum value, small value, small value, median value, slightly large value, large value, maximum value, maximum value of the transformer area outgoing line i load active power is obtained:

[0070]

[0071] Wherein P Sitj (i=1, 2,..., N L ​The optimal value of the fuzzy clustering center matrix of the minimum value, the minimum value, the small value, the small value, the median value, the small value, the large value, the maximum value, the maximum value of the three-phase load data of the outgoing line of the transformer area is obtained.

[0072] As a preferred technical solution, the optimal value of the fuzzy clustering center matrix of the three-phase load data of the outgoing line of the transformer area is obtained, specifically:

[0073] The optimal value of the fuzzy clustering center matrix of the minimum value, the minimum value, the small value, the small value, the median value, the small value, the large value, the maximum value, the maximum value of the three-phase load data of the outgoing line of the transformer area is obtained.

[0074]

[0075]

[0076]

[0077] Where P Aitj , P Bitj , P Citj (i=1,2,...,N L , t=1,2,...,T, j=1,2,...,c) are the optimal values of the minimum value, the minimum value, the small value, the small value, the median value, the small value, the large value, the maximum value, the maximum value of the fuzzy clustering center matrix of the A, B, C phase load active power of the outgoing line i at time period t.

[0078] As a preferred technical solution, the average value of the load data and the three-phase load data power of the outgoing line of the transformer area is calculated, specifically:

[0079] The average value of the load power of the outgoing line i at time period t is calculated using the charging load active power data:

[0080]

[0081] The average value of the A, B, C phase load power of the outgoing line i at time period t is calculated:

[0082] The average value of the A phase load power is:

[0083]

[0084] The average value of the B phase load power is:

[0085]

[0086] The average value of the C phase load power is:

[0087]

[0088] As a preferred technical solution, the step of identifying the charging power of the electric vehicle phase based on the average power of the three-phase load data and confirming the charging behavior specifically includes:

[0089] The charging behavior of electric vehicles is determined by the difference between two adjacent time periods.

[0090] When confirming the charging behavior of electric vehicle phase A, the charging power of electric vehicle phase A must meet three conditions:

[0091] 1) The difference in the average active power of phase A load on transformer substation i during time periods t and t-1 is greater than P. EV k EV Times:

[0092] ΔP AAit =P AAit -P AAit-1 ≥k EV P EV

[0093] Among them, P EV For the power of slow charging of electric vehicles, k EV ≥1 is a positive coefficient;

[0094] 2) The difference between the average active power of phase A load and the average active power of phase B or phase C load of transformer substation i during time period t is greater than P. EV k EV times: P AAit -P ABit ≥k EV P EV or P AAit -P ACit ≥k EV P EV ;

[0095] 3) The difference between the average active power of phase A load of transformer substation line i in time period t and the average active power of load of transformer substation line i in time period t satisfies: and( or );

[0096] When confirming the charging behavior of the B-phase electric vehicle, the charging power of the B-phase electric vehicle must meet three conditions:

[0097] 1) The difference between the average active power of load B on transformer substation i during time periods t and t-1 is greater than P. EV k EV Times:

[0098] ΔP ABit =PABit -P ABit-1 ≥k EV P EV

[0099] wherein, P EV is the power of the electric vehicle slow charging, k EV ≥1 is a positive coefficient;

[0100] 2) the difference between the average value of the B-phase active power of the load of the transformer outlet i in the time period t and the average value of the active power of the load of the A-phase or C-phase is greater than k EV times of P EV : P ABit -P AAit ≥k EV P EV or P ABit -P ACit ≥k EV P EV ;

[0101] 3) the difference between the average value of the B-phase active power of the load of the transformer outlet i in the time period t and the average value of the active power of the load of the transformer outlet i in the time period t satisfies: and or ;

[0102] When confirming the charging behavior of the C-phase electric vehicle, the charging power of the C-phase electric vehicle satisfies three conditions:

[0103] 1) the difference between the average value of the C-phase active power of the load of the transformer outlet i in the time period t and the average value of the C-phase active power of the load of the transformer outlet i in the time period t-1 is greater than k EV times of P EV :

[0104] ΔP ACit =P ACit -P ACit-1 ≥k EV P EV ;

[0105] 2) the difference between the average value of the C-phase active power of the load of the transformer outlet i in the time period t and the average value of the active power of the load of the A-phase or B-phase is greater than k EV times of P EV : P ACit -P AAit ≥k EV P EV or P ACit -P ABit ≥k EV P EV ;

[0106] 3), the difference value of the C-phase active power average value of the transformer area outgoing line i in the time period t and the active power average value of the load in the time period t of the transformer area outgoing line i satisfies: and ( or ).

[0107] Another aspect of the present application provides a fuzzy C-means based transformer area electric vehicle charging behavior identification system, applied to the fuzzy C-means based transformer area electric vehicle charging behavior identification method described above, comprising a data matrix construction module, a cluster center matrix construction module, an iterative calculation module, an optimal value acquisition module, an average value calculation module and a charging behavior identification module.

[0108] The data matrix construction module constructs the spatial data matrix and the phase data matrix of electric vehicle charging according to historical and real-time data.

[0109] The cluster center matrix construction module sets a fuzzy value, adopts fuzzy clustering analysis, and constructs the fuzzy clustering center matrix of the transformer area outgoing line load data and the fuzzy clustering center matrix of the three-phase load data.

[0110] The iterative calculation module determines the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix through iterative calculation.

[0111] The optimal value acquisition module is used to acquire the optimal value of the fuzzy clustering center matrix of the transformer area outgoing line and the three-phase load data.

[0112] The average value calculation module is used to calculate the average value of the power of the transformer area outgoing line load data and the three-phase load data.

[0113] The charging behavior identification module identifies the three-phase electric vehicle charging power according to the average value of the three-phase load data power and confirms the charging behavior.

[0114] The present application also provides a computer readable storage medium storing a program, characterized in that the program, when executed, implements the fuzzy C-means based transformer area electric vehicle charging behavior identification method described above.

[0115] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0116] The fuzzy C-means based transformer area electric vehicle charging behavior identification method can identify the transformer area electric vehicle charging behavior, determine the outgoing line and phase of the electric vehicle during charging, and reflect the randomness of the transformer area electric vehicle charging for many years, thereby providing theoretical guidance for transformer area operation management, charging station configuration, load prediction, control scheduling, and necessary technical support for distributed new energy power generation and intelligent power grid dispatching operation. BRIEF DESCRIPTION OF DRAWINGS

[0117] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0118] Figure 1 This is a flowchart of the method for identifying the charging behavior of electric vehicles in a transformer substation based on fuzzy C-means according to an embodiment of the present invention;

[0119] Figure 2 This is a structural diagram of the electric vehicle charging behavior identification system based on fuzzy C-means in an embodiment of the present invention.

[0120] Figure 3 This is a structural diagram of a computer-readable storage medium according to an embodiment of the present invention. Detailed Implementation

[0121] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort are within the scope of protection of the present application.

[0122] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application can be combined with other embodiments.

[0123] like Figure 1 As shown, this embodiment of the electric vehicle charging behavior identification method based on fuzzy C-means includes the following steps:

[0124] S1. Based on historical and real-time data, construct the spatial data matrix and phase-specific data matrix for electric vehicle charging, specifically as follows:

[0125] Active power data of outgoing loads and active power data of three-phase loads in the transformer substation are obtained from historical and real-time data over many years.

[0126] By processing, calculating, and analyzing the active power data of the outgoing loads in the transformer substation, a spatial data matrix X for electric vehicle charging is constructed. S :

[0127]

[0128] wherein x it is the active power of the load of the outgoing line i of the transformer area at the time period t, i = 1, 2, …, N L , t = 1, 2, …, T; N L is the number of outgoing lines of the transformer area; and T is the number of time periods representing the active power data of the load of the outgoing line of the transformer area;

[0129] By processing, calculating and analyzing the three-phase active power data of the load of the outgoing line of the transformer area, the phase data matrix of the electric vehicle charging is constructed, respectively X A , X B , X C , which is represented as:

[0130]

[0131]

[0132]

[0133] wherein X A , X B , X C are the A, B, C phase data matrices of the electric vehicle charging of the outgoing line i of the transformer area, x Ait , x Bit , x Cit are the A, B, C phase active power of the load of the outgoing line i of the transformer area at the time period t, i = 1, 2, …, N L , t = 1, 2, …, T.

[0134] S2, 9 fuzzy values c are set, including: minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value;

[0135] S3, the fuzzy clustering analysis method is adopted to construct and initialize the fuzzy clustering center matrix of the load data of the outgoing line of the transformer area and the fuzzy clustering center matrix of the three-phase load data, specifically:

[0136] By adopting the fuzzy clustering analysis, c = 9 is selected to construct the fuzzy clustering center matrix of the active power of the load of the outgoing line i of the transformer area:

[0137] C Si = {C Si1 , C Si2 , C Si3 …, C Sic}

[0138] wherein C Si1 , C Si2 , CSi3 C Si4 C Si5 C Si6 C Si7 C Si8 C Si9 The fuzzy clustering center matrix represents the minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power values ​​of the i-th load in the transformer substation.

[0139] Let C Sijt =[C Si1t C Si2t ,...,C Sict ] T ,(j=1,2,...,c),T is a row-to-column operation; used to represent the minimum, minimum, small, minute, median, micro-large, large, maximum, and maximum fuzzy clustering center matrix of the load active power of the outgoing line i in the time period t;

[0140] Let X be the multi-year average value of the active power of the load on the outgoing line i of the transformer substation during time period t. MSit Then, the initial value of the fuzzy clustering center matrix for the minimum, minimum, small, small, median, micro-large, large, maximum, and maximum active power of the load on line i in time period t is:

[0141]

[0142] Where, k j This represents the coefficient corresponding to the fuzzy value.

[0143] In this embodiment, the fuzzy value coefficient k1 corresponding to the minimum value is 0.1, the fuzzy value coefficient k2 corresponding to the minimum value is 0.3, the fuzzy value coefficient k3 corresponding to the small value is 0.5, the fuzzy value coefficient k4 corresponding to the minute value is 0.8, the fuzzy value coefficient k5 corresponding to the median value is 1, the fuzzy value coefficient k6 corresponding to the slightly large value is 0.1, the fuzzy value coefficient k7 corresponding to the large value is 0.1, the fuzzy value k8 corresponding to the maximum value is 0.1, and the fuzzy value coefficient k9 corresponding to the maximum value is 0.1.

[0144] Using fuzzy clustering analysis, with c=9 selected, a fuzzy clustering center matrix of the active power of the A, B, and C phase loads of transformer substation i is constructed:

[0145] C Ai ={C Ai1 C Ai2 C Ai3 …,C Aic}

[0146] C Bi ={C Bi1 CBi2 ,C Bi3 …,C Bic}

[0147] C Ci ={C Ci1 ,C Ci2 ,C Ci3 …,C Cic}

[0148] wherein C Ai1 , C Ai2 , C Ai3 , C Ai4 , C Ai5 , C Ai6 , C Ai7 , C Ai8 , C Ai9 are the minimum value, the minimum value, the minimum value, the minimum value, the median value, the minimum value, the maximum value, the maximum value, the maximum value fuzzy clustering center matrix of the A-phase active power of the load of the outgoing line i of the transformer area; C Bi1 , C Bi2 , C Bi3 , C Bi4 , C Bi5 , C Bi6 , C Bi7 , C Bi8 , C Bi9 are the minimum value, the minimum value, the minimum value, the minimum value, the median value, the minimum value, the maximum value, the maximum value, the maximum value fuzzy clustering center matrix of the B-phase active power of the load of the outgoing line i of the transformer area; C Ci1 , C Ci2 , C Ci3 , C Ci4 , C Ci5 , C Ci6 , C Ci7 , C Ci8 , C Ci9 are the minimum value, the minimum value, the minimum value, the minimum value, the median value, the minimum value, the maximum value, the maximum value, the maximum value fuzzy clustering center matrix of the C-phase active power of the load of the outgoing line i of the transformer area;

[0149] Let C Aijt = [C Ai1t , C Ai2t ,..., C Aict ] T , (j = 1, 2,..., c) be used to represent the minimum value, the minimum value, the minimum value, the minimum value, the median value, the minimum value, the maximum value, the maximum value, the maximum value fuzzy clustering center matrix of the A-phase active power of the load of the outgoing line i in the time period t;

[0150] Let C Bijt = [C Bi1t , CBi2t ,...,C Bict ] T , (j = 1, 2,..., c), for representing the minimum value, the minimum value, the small value, the small value, the median value, the small value, the large value, the maximum value, the maximum value fuzzy clustering center matrix of the B-phase load active power of the substation outgoing line i at the time period t;

[0151] Let C Cijt = [C Ci1t ,C Ci2t ,...,C Cict ] T , (j = 1, 2,..., c), for representing the minimum value, the minimum value, the small value, the small value, the median value, the small value, the large value, the maximum value, the maximum value fuzzy clustering center matrix of the C-phase load active power of the substation outgoing line i at the time period t;

[0152] Let the multi-year average of the three-phase load active power of the substation outgoing line i at the time period t be X MAit , X MBit , X MCit , then the initial value of the minimum value, the minimum value, the small value, the small value, the median value, the small value, the large value, the maximum value, the maximum value fuzzy clustering center matrix of the three-phase load active power of the substation outgoing line i at the time period t is respectively:

[0153]

[0154]

[0155]

[0156] Wherein, k j represents the coefficient corresponding to the fuzzy value.

[0157] S4, determine the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix by iterative calculation, specifically:

[0158] The iterative calculation is performed by using the charging load active power data to determine the optimal value of the fuzzy clustering matrix, and the optimal value of the minimum value, the minimum value, the small value, the small value, the median value, the small value, the large value, the maximum value, the maximum value fuzzy clustering center matrix;

[0159] S41, set the fuzzy coefficient f first, set the iterative calculation end error value ε, and set the iteration number t;

[0160] S42, a random number is randomly generated on [0, 1] by using a random function to initialize the fuzzy clustering matrix U (0) ;

[0161] S43, iteration is performed and element values of the fuzzy clustering matrix are calculated, in the tth iteration, the update formula of the element values of the fuzzy clustering matrix U (t) The update formula of the element values of the fuzzy clustering matrix C

[0162]

[0163] Wherein, d() is a distance function;

[0164] S44, element values of the fuzzy clustering center matrix C (t) The minimum value, average value, maximum value fuzzy clustering center matrix element value The update formula is:

[0165]

[0166] Wherein, i = 1, 2,..., c, j = 1, 2,..., m;

[0167] S45, if The iteration calculation is ended, otherwise, go to step S33 to enter the next period t = t + 1 to continue the iteration calculation;

[0168] S46, after the iteration calculation is completed, the optimal value of the fuzzy clustering matrix is obtained:

[0169]

[0170] The optimal value of the fuzzy clustering center matrix is obtained:

[0171]

[0172] In the embodiment, the fuzzy coefficient The iteration calculation ending judgment error value ε = 0.001, the iteration number t = 1.

[0173] S5, the optimal value of the fuzzy clustering center matrix of the substation area outgoing line load data is obtained, specifically:

[0174] Using the load active power data, the optimal value of the fuzzy clustering center matrix of the minimum value, minimum value, small value, small value, median value, slightly large value, large value, maximum value, maximum value of the active power of the substation area outgoing line i load is obtained:

[0175]

[0176] Wherein P Sitj (i = 1, 2,..., N L ​The optimal value of the fuzzy clustering center matrix of the minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value of the three-phase load data of the outgoing line of the transformer area is obtained, and the optimal value of the fuzzy clustering center matrix of the minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value of the three-phase load data of the outgoing line of the transformer area is obtained.

[0177] S6, the optimal value of the fuzzy clustering center matrix of the three-phase load data of the outgoing line of the transformer area is obtained, specifically:

[0178] The optimal value of the fuzzy clustering center matrix of the minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value of the three-phase load data of the outgoing line of the transformer area is obtained, and the optimal value of the fuzzy clustering center matrix of the minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value of the three-phase load data of the outgoing line of the transformer area is obtained.

[0179]

[0180]

[0181]

[0182] Wherein P Aitj , P Bitj , P Citj (i=1,2,...,N L , t=1,2,...,T, j=1,2,...,c) are the optimal values of the fuzzy clustering center matrix of the minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value of the three-phase load data of the outgoing line of the transformer area.

[0183] S7, the average value of the load data and the three-phase load data power of the outgoing line of the transformer area is calculated, specifically:

[0184] The average value of the load power of the outgoing line i of the transformer area at time period t is calculated by using the charging load active power data:

[0185]

[0186] The average value of the A, B, C phase load power of the outgoing line i of the transformer area at time period t is calculated:

[0187] The average value of the A phase load power is:

[0188]

[0189] The average value of the B phase load power is:

[0190]

[0191] The average value of the C phase load power is:

[0192]

[0193] S8, according to the three-phase load data power average, identify phase electric vehicle charging power, confirm charging behavior, specifically:

[0194] According to the three-phase load data power average, identify phase electric vehicle charging power, confirm charging behavior, specifically:

[0195] According to the difference between the adjacent two periods, to confirm the charging behavior of electric vehicles;

[0196] When confirming the charging behavior of A-phase electric vehicles, the charging power of A-phase electric vehicles meets 3 conditions:

[0197] 1), the difference between the A-phase load active power average of the transformer outlet i at time period t and t-1 is greater than k times of P: EV EV ΔP AAit = P AAit -P AAit-1 ≥ k EV P EV

[0199] Where P is the power of slow charging of electric vehicles, and k ≥1 is a positive coefficient;

[0200] 2), the difference between the A-phase load active power average of the transformer outlet i at time period t and the B-phase or C-phase load active power average is greater than k times of P: EV EV P AAit -P ABit ≥ k EV P EV or P AAit -P ACit ≥ k EV P EV ;

[0201] 3), the difference between the A-phase load active power average of the transformer outlet i at time period t and the load active power average of the transformer outlet i at time period t satisfies: and ( or );

[0202] When confirming the charging behavior of B-phase electric vehicles, the charging power of B-phase electric vehicles meets 3 conditions:

[0203] 1), the difference between the B-phase load active power average of the transformer outlet i at time period t and t-1 is greater than k times of P: EV EV ΔP ABit = P ABit -P ABit-1 ≥ k EV P EV ;​

[0204] ΔP ABit = P ABit -P ABit-1 ≥ k EV P EV

[0205] wherein, P EV is the power of the electric vehicle slow charging, k EV ≥ 1 is a positive coefficient;

[0206] 2) The difference between the average value of the B-phase active power of the distribution area outgoing line i at the time period t and the average value of the active power of the A-phase or C-phase load is greater than k EV times of P EV : P ABit -P AAit ≥ k EV P EV or P ABit -P ACit ≥ k EV P EV ;

[0207] 3) The difference between the average value of the B-phase active power of the distribution area outgoing line i at the time period t and the average value of the active power of the load of the distribution area outgoing line i at the time period t satisfies: and ( or );

[0208] When confirming the charging behavior of the C-phase electric vehicle, the charging power of the C-phase electric vehicle satisfies 3 conditions:

[0209] 1) The difference between the average value of the C-phase active power of the distribution area outgoing line i at the time period t and the average value of the active power of the load at the time period t-1 is greater than k EV times of P EV :

[0210] ΔP ACit = P ACit -P ACit-1 ≥ k EV P EV ;

[0211] 2) The difference between the average value of the C-phase active power of the distribution area outgoing line i at the time period t and the average value of the active power of the A-phase or B-phase load is greater than k EV times of P EV : P ACit -P AAit ≥ k EV P EV or P ACit -P ABit ≥ k EV P EV ;

[0212] 3) the difference between the average value of the C-phase active power of the transformer outlet i in the time period t and the average value of the active power of the load of the transformer outlet i in the time period t satisfies: and ( or ).

[0213] In the embodiment, P EV is 5kW.

[0214] It should be noted that, for the foregoing method embodiments, in order to facilitate description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the action sequence described, because according to the present application, certain steps can be performed in other sequences or simultaneously.

[0215] Based on the same idea as the fuzzy C-means-based transformer area electric vehicle charging behavior recognition method in the above embodiment, the present application also provides a fuzzy C-means-based transformer area electric vehicle charging behavior recognition system, which can be used to execute the above fuzzy C-means-based transformer area electric vehicle charging behavior recognition method. For the convenience of description, in the structural schematic diagram of the fuzzy C-means-based transformer area electric vehicle charging behavior recognition system embodiment, only the parts related to the embodiments of the present application are shown, and those skilled in the art can understand that the illustrated structure does not constitute a limitation on the device, and can include more or fewer components than the illustrated, or combine certain components, or different component arrangements.

[0216] As Figure 2 shown, another embodiment of the present application provides a fuzzy C-means-based transformer area electric vehicle charging behavior recognition system, which includes the following modules:

[0217] The data matrix construction module constructs the spatial data matrix and the phase data matrix of electric vehicle charging according to historical and real-time data;

[0218] The cluster center matrix construction module sets the fuzzy value, adopts fuzzy clustering analysis, and constructs the fuzzy clustering center matrix of the transformer outlet load data and the fuzzy clustering center matrix of the three-phase load data;

[0219] The iterative calculation module determines the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix through iterative calculation;

[0220] The optimal value acquisition module is used to acquire the optimal value of the fuzzy clustering center matrix of the transformer outlet and three-phase load data;

[0221] The average value calculation module is used to calculate the average value of the power of the transformer outlet load data and the three-phase load data;

[0222] The charging behavior recognition module recognizes the three-phase electric vehicle charging power according to the average value of the three-phase load data power, and confirms the charging behavior.

[0223] It should be noted that the fuzzy C-means based transformer area electric vehicle charging behavior recognition system of the present application corresponds to the fuzzy C-means based transformer area electric vehicle charging behavior recognition method of the present application. The technical features and advantages described in the embodiment of the fuzzy C-means based transformer area electric vehicle charging behavior recognition method are applicable to the embodiment of the fuzzy C-means based transformer area electric vehicle charging behavior recognition system. For specific content, please refer to the description in the method embodiment of the present application. Here, no further description is given, and this is hereby declared.

[0224] In addition, in the embodiment of the fuzzy C-means based transformer area electric vehicle charging behavior recognition system described above, the logical division of each program module is only an example. In actual application, the above-mentioned function allocation can be completed by different program modules according to the needs, for example, considering the configuration requirements of the corresponding hardware or the convenience of software implementation. That is, the internal structure of the fuzzy C-means based transformer area electric vehicle charging behavior recognition system is divided into different program modules to complete all or part of the functions described above.

[0225] As shown in FIG. Figure 3 In one embodiment, a computer readable storage medium is provided, which stores a program in the memory. When the program is executed by a processor, the fuzzy C-means based transformer area electric vehicle charging behavior recognition method is realized. Specifically:

[0226] According to historical and real-time data, a spatial data matrix and a phase data matrix of electric vehicle charging are constructed;

[0227] The fuzzy value c is set, including: minimum value, minimum value, small value, small value, median value, small value, large value, large value, maximum value, maximum value;

[0228] Fuzzy clustering analysis is used to construct and initialize the fuzzy clustering center matrix of the transformer area outgoing line load data and the fuzzy clustering center matrix of the three-phase load data;

[0229] Through iterative calculation, the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix are determined;

[0230] The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line load data is obtained;

[0231] The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line three-phase load data is obtained;

[0232] The average value of the transformer area outgoing line load data and the three-phase load data power is calculated;

[0233] According to three-phase load data power average value, three-phase electric vehicle charging power is identified, and charging behavior is confirmed.

[0234] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing related hardware, and the program can be stored in a non-volatile computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments of each method. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory.

[0235] Each technical feature of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of each technical feature in the above embodiments are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0236] The above embodiments are the preferred embodiments of the present application, but the embodiments of the present application are not limited to the above embodiments, and any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the present application are equivalent replacement methods and are included in the protection scope of the present application.

Claims

1. A method for identifying electric vehicle charging behavior in a transformer area based on fuzzy C-means, characterized in that, It comprises the following steps: According to historical and real-time data, a spatial data matrix and a phase data matrix of electric vehicle charging are constructed, specifically: From historical and real-time data, the active power data of the transformer area outgoing line load and the three-phase load active power data are obtained; By processing, calculating and analyzing the active power data of the outgoing line load of the transformer area, a spatial data matrix of electric vehicle charging is constructed X s : ， wherein, x it are the number of out-of-line loads of the transformer substation, i the active power of the load in the time interval t , i = 1, 2,..., N L , t = 1, 2,..., T ; N L are the number of out-of-line loads of the transformer substation; T are the time interval values representing the active power data of the out-of-line loads of the transformer substation. By processing, calculating and analyzing the three-phase active power data of the transformer area outlet, the phase data matrix of electric vehicle charging is constructed, respectively X A , X B , X C , which is expressed as: ， ， ， in, X A , X B , X C Each will qualify from the Taiwan area. i A, B, and C phase data matrix for electric vehicle charging. x Ait , x Bit , x Cit Each will qualify from the Taiwan area. i Phases A, B, and C in the time period t The active power of the load, i =1,2,…, N L , t =1,2,…, T ; Set fuzzy values; the fuzzy values include: minimum value, minimum value, small value, small value, median value, small value, large value, maximum value, maximum value; Using fuzzy clustering analysis, the fuzzy clustering center matrix of the transformer area outgoing line load data and the fuzzy clustering center matrix of the three-phase load data are constructed and initialized; Through iterative calculation, the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix are determined; The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line load data is obtained; The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line three-phase load data is obtained; The average value of the transformer area outgoing line load data and the three-phase load data power is calculated; According to the average value of the three-phase load data power, the three-phase electric vehicle charging power is identified, and the charging behavior is confirmed.

2. The fuzzy C-means based method for identifying the behavior of electric vehicle charging in a transformer area according to claim 1, characterized in that, The construction and initialization of the fuzzy clustering center matrix of the transformer area outgoing line load data and the fuzzy clustering center matrix of the three-phase load data are specifically: The fuzzy clustering analysis is adopted to construct the transformer area outgoing line i The fuzzy clustering center matrix of the load active power: C si = { C si1 , C si2 , C si3 ,…, C sic}， wherein C si1 , C si2 , C si3 , C si4 , C si5 , C si6 , C si7 , C si8 , C si9 for the district out of line i active power minimum, minimum, small, small, median, slightly large, large, maximum, maximum fuzzy clustering center matrix; Let C sijt = [ C si1t , C si2t ,…, C sict ] T , j =1,2,…, c , T be the row-to-column operation; For indicating the transformer area outgoing line i The minimum, minimum, small, small, median, slightly large, large, maximum, maximum fuzzy clustering center matrix of the load active power in the time period t The minimum, minimum, small, small, median, slightly large, large, maximum, maximum fuzzy clustering center matrix of the load active power in the time period Set the area out line i In the time period t The multi-year average of the load active power is X MSit Then the out line i In the time period t The minimum, minimum, small, small, median, slightly large, large, maximum, maximum fuzzy clustering center matrix of the minimum value of the load active power in the time period ， wherein k j denotes the coefficient corresponding to the blur value; Fuzzy clustering analysis is used to construct the transformer area outgoing line. i Fuzzy clustering center matrix of active power of phases A, B, and C loads: C Ai = { C Ai1 , C Ai2 , C Ai3 ,…, C Aic}， C Bi = { C Bi1 , C Bi2 , C Bi3 ,…, C Bic}， C Ci = { C Ci1 , C Ci2 , C Ci3 ,…, C Cic}， in, C Ai1 , C Ai2 , C Ai3 , C Ai4 , C Ai5 , C Ai6 , C Ai7 , C Ai8 , C Ai9 Qualifying for the Taiwan region i The fuzzy clustering center matrix of the minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power of the A-phase load; C Bi1 , C Bi2 , C Bi3 , C Bi4 , C Bi5 , C Bi6 , C Bi7 , C Bi8 , C Bi9 Qualifying for Taiwan i The fuzzy clustering center matrix of the minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power of the B-phase load; C Ci1 , C Ci2 , C Ci3 , C Ci4 , C Ci5 , C Ci6 , C Ci7 , C Ci8 , C Ci9 Qualifying for the Taiwan region i The fuzzy clustering center matrix of the minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power of the C-phase load; make C Aijt = [ C Ai1t , C Ai2t ,…, C Aict ] T , j =1,2,…, c Used to indicate the outgoing line of the transformer area. i During the period t The minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power of the A-phase load are represented by the fuzzy clustering center matrix. make C Bijt = [ C Bi1t , C Bi2t ,…, C Bict ] T , j =1,2,…, c Used to indicate the outgoing line of the transformer area. i During the period t The minimum, minimum, small, minute, median, micro-maximum, large, maximum, and maximum active power of the B-phase load are represented by the fuzzy clustering center matrix. Let C Cijt = [ C Ci1t , C Ci2t ,…, C Cict ] T , j =1,2,…, c , represent the minimum, minimum, minimum, minimum, median, minimum, maximum, maximum, maximum fuzzy clustering center matrix of the C-phase active power of the load in the time period i t .​ Set up the outgoing line of the station area i During the period t The multi-year average values ​​of the three-phase load active power are respectively X MAit , X MBit , X MCit Then the Taiwan district will qualify. i During the period t The initial values ​​of the fuzzy clustering center matrix for the minimum, minimum, small, small, median, micro-large, large, maximum, and maximum active power of the three-phase load are as follows: ， ， ， wherein k j denotes the coefficient corresponding to the blur value.

3. The fuzzy C-means based method for identifying the behavior of electric vehicle charging in a transformer area according to claim 2, characterized in that, The iterative calculation module determines the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix through iterative calculation; Iterative calculation is performed using the charging load active power data; Setting the blur coefficient f Setting the iteration computation end determination error value ε Setting the iteration number t ; The random numbers are generated randomly on [0, 1] using a random function to initialize the fuzzy clustering matrix U (0) ; The iteration is performed and the element values of the fuzzy clustering matrix are calculated, and in the first iteration, the update formula of the element values of the fuzzy clustering matrix is: t U (t) The iteration is performed and the element values of the fuzzy clustering matrix are calculated, and in the first iteration, the update formula of the element values of the fuzzy clustering matrix is:​​ ， Wherein, d() is the distance function; fuzzy clustering center matrix C (t) the minimum value, the average value, the maximum value of the element values of the fuzzy clustering center matrix The iterative updating formula is: ， wherein i = 1, 2,..., c , j = 1, 2,..., m ; If then the iterative calculation is finished, otherwise the next time period is entered t = 0 t + 1 the iterative calculation is continued After the iterative calculation is completed, the optimal value of the fuzzy clustering matrix is: ， The optimal value of the fuzzy clustering center matrix is: 。 4. The fuzzy C-means based method for identifying the behavior of electric vehicle charging in a transformer area according to claim 3, characterized in that, The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line load data is obtained, specifically: The load active power data is used to obtain the outgoing line of the transformer area i Optimal value of fuzzy clustering center matrix of load active power minimum value, minimum value, small value, small value, median value, slightly large value, large value, maximum value, maximum value ， wherein P Sitj ( i = 1, 2, …, N L , t = 1, 2, …, T , j = 1, 2, …, c ) is the area of the line i The minimum, minimum, minimum, minimum, median, minimum, maximum, maximum, maximum fuzzy clustering center matrix of the optimal value of the active power of the load in the time period t .

5. The fuzzy C-means based method for identifying the behavior of electric vehicle charging in a transformer area according to claim 4, characterized in that, The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line three-phase load data is obtained, specifically: Using load active power data, obtain the outgoing lines of the transformer area. i The optimal values ​​of the fuzzy clustering center matrix for the minimum, minimum, small, minute, median, minute-large, large, maximum, and maximum active power of phases A, B, and C loads: ， ， ， in P Aitj , P Bitj , P Citj ( i =1,2,…, N L , t =1,2,…, T , j =1,2,…, c () are the teams that will advance from the Taiwan district. i During the period t The optimal value of the fuzzy clustering center matrix for the minimum, minimum, small, minute, median, micro-large, large, maximum, and maximum active power of the A, B, and C phase loads.

6. The fuzzy C-means based method for identifying the behavior of electric vehicle charging in a transformer area according to claim 5, characterized in that, The average value of the transformer area outgoing line load data and the three-phase load data power is calculated, specifically: The charging load active power data is used to calculate the feeder outgoing line i The load power average value in the time period t is calculated. ， Computing the distribution of the power of the outgoing line of the table area i The average value of the load power of A, B, C phase in the time period t : The average value of A-phase load power: ， The average value of B-phase load power: ， The average value of C-phase load power: 。 7. The fuzzy C-means based method for identifying electric vehicle charging behavior of a transformer area according to claim 6, characterized in that, According to the average value of the three-phase load data power, the three-phase electric vehicle charging power is identified, and the charging behavior is confirmed, specifically: According to the difference between the adjacent two time periods, the charging behavior of the electric vehicle is confirmed; When the charging behavior of the A-phase electric vehicle is confirmed, the charging power of the A-phase electric vehicle meets 3 conditions: 1) The transformer area outgoing line i In the time period t And t -1, the difference between the average active power of the A-phase load is greater than P EV k EV times:​ ， wherein P EV Pslow is the power for slow charging of the electric vehicle, k EV ≥ 1 is a positive coefficient; 2) the outgoing line of the transformer substation i the difference between the average value of the active power of the phase A load and the average value of the active power of the phase B or C load in the time period t is greater than P EV k EV times: ;​ 3) Qualifying from the Taiwan area i During the period t The average active power of phase A load and the output power of the transformer substation i During the period t The difference between the average active power of the load and the average active power satisfies: and( ); When the charging behavior of the B-phase electric vehicle is confirmed, the charging power of the B-phase electric vehicle meets 3 conditions: 1) Area of the line i In the time period t And t -1 B load active power average difference greater than P EV k EV Times:​ ， wherein P EV Pslow is the power for slow charging of the electric vehicle, k EV ≥1 is a positive coefficient; 2) the outgoing line of the transformer area i the difference between the average value of the active power of the phase B load and the average value of the active power of the phase A or C load in the time period t is greater than P EV k EV times: ;​ 3) Qualifying from the Taiwan area i During the period t The average active power of phase B load and the output power of the transformer substation i During the period t The difference between the average active power of the load and the average active power satisfies: and( ); When the charging behavior of the C-phase electric vehicle is confirmed, the charging power of the C-phase electric vehicle meets 3 conditions: 1) The transformer area outgoing line i In the time period t And t The active power average value difference of C load at -1 is greater than P EV k EV Times:​ ； 2) the difference between the average value of the active power of the C-phase load of the outgoing line of the transformer area in the time period and the average value of the active power of the A-phase or B-phase load is greater than P EV of k EV times: ; 3), the transformer area outgoing line i the C phase active power average value of the load in the time period t and the difference value of the load active power average value of the transformer area outgoing line i in the time period t satisfies: and ( ).

8. A fuzzy C-means based system for identifying electric vehicle charging behavior in a transformer area, characterized by, The fuzzy C-means based transformer area electric vehicle charging behavior identification method of any one of claims 1-7 comprises a data matrix construction module, a clustering center matrix construction module, an iterative calculation module, an optimal value acquisition module, an average value calculation module and a charging behavior identification module; The data matrix construction module constructs a spatial data matrix and a phase data matrix of electric vehicle charging according to historical and real-time data; The clustering center matrix construction module sets fuzzy values, uses fuzzy clustering analysis, and constructs and initializes the fuzzy clustering center matrix of the transformer area outgoing line load data and the fuzzy clustering center matrix of the three-phase load data; The iterative calculation module determines the optimal fuzzy clustering matrix and the optimal fuzzy clustering center matrix through iterative calculation; The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line load data is obtained, specifically: The optimal value of the fuzzy clustering center matrix of the transformer area outgoing line three-phase load data is obtained, specifically: The average value of the transformer area outgoing line load data and the three-phase load data power is calculated, specifically: The average value of A-phase load power: The average value of B-phase load power: The average value of C-phase load power: According to the average value of the three-phase load data power, the three-phase electric vehicle charging power is identified, and the charging behavior is confirmed, specifically: According to the difference between the adjacent two time periods, the charging behavior of the electric vehicle is confirmed; When the charging behavior of the A-phase electric vehicle is confirmed, the charging power of the A-phase electric vehicle meets 3 conditions: When the charging behavior of the B-phase electric vehicle is confirmed, the charging power of the B-phase electric vehicle meets 3 conditions: When the charging behavior of the C-phase electric vehicle is confirmed, the charging power of the C-phase electric vehicle meets 3 conditions: The optimal value acquisition module is configured to acquire optimal values of the fuzzy clustering center matrix of the feeder outlet and three-phase load data; The average value calculation module is configured to calculate average values of the feeder outlet load data and three-phase load data power; The charging behavior recognition module is configured to recognize three-phase electric vehicle charging power according to the average values of the three-phase load data power, and confirm the charging behavior.

9. A computer-readable storage medium storing a program, the program comprising instructions which, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. When the program is executed, the fuzzy C-means based feeder electric vehicle charging behavior recognition method according to any one of claims 1-7 is implemented.

Images