Traffic energy intelligent fusion management system and method based on artificial intelligence

Abstract

The invention relates to the technical field of traffic energy management, in particular to a traffic energy intelligent fusion management system and method based on artificial intelligence. According to the calculated coincidence degree and the function value of each charging area, the alternative charging areas selected by recommendation are output, allocation can be carried out according to the mobility of the electric vehicles, and the peak load of the power distribution network is reduced; furthermore, the standby power utilization demand of the target vehicle is analyzed and calculated, the recommended charging duration of the target vehicle is calculated according to the standby power utilization demand and is pushed to the user side, the charging duration of the vehicle can be planned, the coincidence duration between the charging duration of the vehicle and the power utilization peak time period is shortened, and the charging efficiency of the vehicle is improved. Therefore, the load scheduling of the power distribution network can be realized, the charging cost of a user can be reduced, the charging duration can be reduced in a peak period, the charging duration can be increased in a low peak period, and the charging cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic energy management, and in particular to an artificial intelligence-based traffic energy intelligent fusion management system and method. Background Art

[0002] With the popularity of electric vehicles, their charging needs will significantly increase the load on the distribution network. In order to alleviate the impact of electric vehicles on the distribution network and give full play to the potential of electric vehicles, the industry has proposed an interactive model between electric vehicles and the distribution network. For example, the "car-pile-network" interactive model can achieve peak shaving and valley filling, and absorb new energy through the mobile energy storage characteristics of electric vehicles, thereby reducing the peak-to-valley difference of the distribution network load, improving the new energy absorption capacity and reducing the demand for capacity expansion and transformation of the distribution network.

[0003] However, electric vehicles may be charged in a disorderly manner. For example, an electric vehicle may suddenly need to be charged during use, and the charging location may be randomly selected. This will cause the peak load of the distribution network to increase significantly. The disorderly charging behavior of electric vehicles may further aggravate the peak-to-valley difference in the power grid load, making the dispatching and operation management of the distribution network more complicated and difficult. Summary of the invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based transportation energy intelligent fusion management system and method to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: an intelligent fusion management method for traffic energy based on artificial intelligence, the management method comprising the following steps: Step S1, obtaining the target charging area marked by the target vehicle, predicting the arrival time and charging time period of the target vehicle, obtaining the peak power consumption time period of the target charging area, determining whether the charging time of the target vehicle overlaps with the peak power consumption time period of the target charging area, and calculating the degree of overlap; Step S2: if the overlap is greater than the overlap threshold, obtain the passing area of ​​the target vehicle's route, extract the charging area with idle charging piles in the passing area as the candidate charging area, calculate the function value of the candidate charging area, and output the recommended replacement candidate charging area according to the function value; Step S3: After the target vehicle is connected to the charging pile, the charging area where the connected charging pile is located is obtained, the peak power consumption time period of the power distribution area where the charging area is located is determined, and the coverage value between the current vehicle's charging time period and the peak power consumption time period of the power distribution area is calculated; Step S4: If the coverage value is greater than the coverage value threshold, the historical charging data of the target vehicle is obtained, the historical charging record of the target vehicle is extracted, the charging frequency of the target vehicle is calculated, the standby power demand of the target vehicle is analyzed, and the recommended charging time of the target vehicle is calculated based on the standby power demand, and pushed to the user end.

[0006] Furthermore, the step S1 includes: Step S101, select a target charging area, obtain historical power consumption data of the target charging area, analyze the peak power consumption time period of the target charging area based on the historical power consumption data, and record the peak power consumption time period as [t1, t2], where t1 represents the start time point of the peak power consumption time period, and t2 represents the end time point of the peak power consumption time period; Step S102: record the time point when the target vehicle arrives at the target charging area as t3, obtain the charging record of the target vehicle, extract the record of the charging amount reaching full charge in the charging record as the target charging record data, obtain the charging record duration corresponding to each target charging record data, input the charging amount and charging record duration in each charging record data into the training model, obtain the charging duration prediction model, input the target charging amount of the target vehicle, and output the predicted charging duration TD of the target vehicle; Step S103, taking the time point when the target vehicle arrives at the target charging area as the charging start time, and obtaining the charging time period of the target vehicle as [t3, t4] according to the predicted charging time TD, wherein t3 represents the starting time point of the charging time period, and t4 represents the ending time point of the charging time period, and establishing a time axis, marking the peak power consumption time period and the charging time period on the time axis, marking the overlapping time period of the two as [t5, t6], and calculating the overlap degree C1, the calculation formula of which is C1=t6-t5 / t3-t4, wherein t3-t4 represents the charging time period of the charging time period [t3, t4], and t6-t5 represents the overlapping time period of the overlapping time period [t5, t6].

[0007] Furthermore, the step S2 includes: Step S201: Set the overlap threshold to C 阈 , if C1>C 阈When , the route of the target vehicle to the target charging area is obtained, the area passed by the route is marked as the passing area, all the charging areas in the passing area are extracted, the peak power consumption time period of each charging area is obtained and marked on the time axis, and the time point t9 when the target vehicle arrives at the designated charging area is predicted, and the time point when the target vehicle arrives at the designated charging area is used as the charging start time. The charging time period of the target vehicle is predicted to be [t9, t10], and the time period when the charging time period overlaps with the peak power consumption time period of the designated charging area is marked as [t7, t8]. The overlap degree C2 between the target vehicle and the corresponding charging area is calculated, and the calculation formula is C2=t8-t7 / t10-t9, where t10-t9 represents the charging time length of the charging time period [t9, t10], and t8-t7 represents the overlap time length of the overlap time period [t7, t8]. The charging area with C2<C1 is used as the alternative charging area; Step S202: Eliminate all the charging areas in the candidate charging area where the charging piles are fully occupied, obtain all the types of charging piles in the charging area with idle charging piles, divide the charging piles into fast charging type and slow charging type, obtain the number of idle charging piles and the types of idle charging piles in each candidate charging area, record the number of fast charging type charging piles as D1, record the number of slow charging type charging piles as D2, and calculate the slow charging proportion q=D2 / D1+D2; Step S203, obtain the charging record of each idle charging pile in the candidate charging area, mark the record of unfinished charging order in the charging record as abnormal charging record, obtain the end time of the charging order corresponding to all abnormal charging records, extract the start time of the next charging record recorded by the current charging pile after the abnormal charging record, calculate the time difference T1 between the two records, if T1<T1 阈值 , then the abnormal charging record is judged as the target abnormal data, all target abnormal data are extracted, and the abnormal charging rate Q of the charging pile is calculated. i =n / m, where n is the number of all target abnormal data of a single charging pile, m is the number of all abnormal charging records in the candidate charging area, Q i is the abnormal charging rate of the i-th charging pile; Step S204: Calculate the function value K of each candidate charging area, and the calculation formula is K=α1*∑Q i +α2*q, where i=1, 2, 3, ..., i, α1 and α2 represent weight values, * represents the product of two values, and the charging area with the smallest function value is output as the recommended candidate charging area.

[0008] Furthermore, the step S3 includes: Step S301, obtain the divided distribution network area, mark the charging pile to which the target vehicle is connected, obtain the distribution network area where the charging pile is located, extract the peak power consumption time period [tR1, tR2] of the distribution network area, where tR1 represents the starting time point of the peak power consumption time period of the distribution network area, and tR2 represents the ending time point of the peak power consumption time period of the distribution network area, obtain the current estimated charging time period [tR3, tR4] of the target vehicle, where tR3 represents the starting time point of the estimated charging time period, and tR4 represents the ending time point of the estimated charging time period, obtain the overlapping time period of the estimated charging time period and the peak power consumption time period and mark it as [tR5, tR6], calculate the coverage value S, and the calculation formula is S=tR6-tR5 / tR4-tR3, where tR6-tR5 represents the overlapping duration of the overlapping time period [tR5, tR6], and tR4-tR3 represents the charging duration of the estimated charging time period [tR3, tR4].

[0009] Further, the step S4 includes: Step S401: Set the coverage threshold to S 阈 , if S>S 阈值 , then the historical charging data of the target vehicle is obtained, the historical charging record of the target vehicle is extracted, and the record of the target vehicle completing a charging order is marked as a charging number; Step S402, calculate the charging frequency F of the target vehicle, the calculation formula is F=the number of charging times of the target vehicle / the total usage time of the target vehicle, obtain the daily usage of the target vehicle, extract the urban area where the target vehicle travels every day, mark the urban area where the target vehicle appears most times as the frequently used urban area, obtain the daily power consumption of the target vehicle in the frequently used urban area to form a line graph, calculate the daily average power consumption P1 of the target vehicle in the frequently used urban area, and obtain the maximum value P of the daily power consumption man ; Step S403, extracting the target vehicle's single-day driving route in the commonly used urban area, recording the target vehicle's one-day driving route in the commonly used urban area as a target event, obtaining the driving routes in two adjacent target events, eliminating the same sections in the two driving routes corresponding to the two target events, retaining different sections as target sections, calculating the power consumption used by the target vehicle on the target section, marking it as reserved power consumption, obtaining all reserved power consumption, generating a line graph in units of days, and calculating the reserved power consumption average value P2 of the reserved power consumption; Step S404: Calculate the standby power demand based on the daily average power consumption and the average reserved power consumption. The calculation formula is: W = P1 + P2 * β1 + P man*β2, where β1 and β2 are weight values. According to the current remaining power of the target vehicle, the charging time when the standby power demand W is reached is predicted, and the standby charging time period is obtained as [tR7, tR8]. The end time point of the standby charging time period is obtained as tR8, and the end time point of the peak power consumption time period is tR2. According to the time sequence, if tR8 < tR2, the time between tR8 and tR2 is marked as T2. According to the formula, TP is calculated as (T2 / tR8-tR7) + b, where TP represents the waiting time ratio, tR8-tR7 represents the charging time of the standby charging time period, and b is a fixed value. If TP> T2 阈 , then the recommended charging time is [tR7, tR8]; if tR8 ≥ tR2 or TP ≤ T2 阈 , the recommended output power is full charge.

[0010] A traffic energy intelligent fusion management system based on artificial intelligence is applied to a traffic energy intelligent fusion management method based on artificial intelligence. The management system includes a power peak prediction module, a charging area selection module and a power demand analysis module; the power peak prediction module is used to predict the power peak time period of each charging area and the power peak time period of each distribution network area, and calculate the overlap between the charging time period of the target vehicle and the power peak time period; the charging area selection module is used to output the recommended alternative charging area according to the calculated overlap and the function value of each charging area; the power demand analysis module is used to analyze and calculate the standby power demand of the target vehicle, calculate the recommended charging time of the target vehicle according to the standby power demand, and push it to the user end.

[0011] The power consumption peak prediction module includes a charging area unit, a distribution area unit and a target vehicle unit. The charging area unit is used to obtain the power consumption peak time period of the target charging area, the distribution area unit is used to obtain the power consumption peak time period of the divided distribution network area; and the target vehicle unit is used to obtain the charging time period of the target vehicle.

[0012] The charging area selection module includes a charging pile analysis unit and an area selection unit. The charging pile analysis unit is used to calculate the abnormal probability of each charging pile and the fast charging proportion of the charging area corresponding to the charging pile; the area selection unit calculates the function value of the area according to the abnormal charging rate and the fast charging proportion of the charging area, and outputs the recommended alternative charging area according to the function value.

[0013] The power demand analysis module includes a power statistics unit, a redundant unit and an output unit. The power statistics unit is used to obtain the daily power consumption of the target vehicle and calculate the average daily power consumption of the target vehicle; the redundant unit is used to calculate the average reserved power consumption of the target vehicle; the output unit is used to calculate the standby power demand based on the daily power consumption and the average reserved power consumption, and output the recommended charging time based on the standby power demand.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention analyzes the charging time period of the target vehicle with the peak power consumption time period of each charging area and the peak power consumption time period of the distribution network area, calculates the overlap between the charging time period of the target vehicle and the peak power consumption time period, and the function value of each charging area, and outputs the recommended alternative charging area according to the calculated overlap and the function value of each charging area, which can be deployed according to the mobility of the electric vehicle to reduce the peak load of the distribution network; and further analyzes and calculates the standby power demand of the target vehicle, calculates the recommended charging time of the target vehicle according to the standby power demand, and pushes it to the user end, so that the charging time of the vehicle can be planned, and the overlap time between the charging time of the vehicle and the peak power consumption time period can be shortened, thereby realizing the load scheduling of the distribution network, and also reducing the charging cost of the user, and can reduce the charging time during peak hours and increase the charging time during low peak hours to reduce the charging cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a flow chart of an intelligent fusion management method for transportation energy based on artificial intelligence of the present invention; Figure 2 This is a structural schematic diagram of an artificial intelligence-based transportation energy intelligent fusion management system of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1: Figure 1 As shown, the present invention provides a transportation energy intelligent fusion management method based on artificial intelligence, and the management method comprises the following steps: Step S1, obtaining the target charging area marked by the target vehicle, predicting the arrival time and charging time period of the target vehicle, obtaining the peak power consumption time period of the target charging area, determining whether the charging time of the target vehicle overlaps with the peak power consumption time period of the target charging area, and calculating the degree of overlap; The step S1 comprises: Step S101, select a target charging area, obtain historical power consumption data of the target charging area, analyze the peak power consumption time period of the target charging area based on the historical power consumption data, and record the peak power consumption time period as [t1, t2], where t1 represents the start time point of the peak power consumption time period, and t2 represents the end time point of the peak power consumption time period; Step S102: record the time point when the target vehicle arrives at the target charging area as t3, obtain the charging record of the target vehicle, extract the record of the charging amount reaching full charge in the charging record as the target charging record data, obtain the charging record duration corresponding to each target charging record data, input the charging amount and charging record duration in each charging record data into the training model, obtain the charging duration prediction model, input the target charging amount of the target vehicle, and output the predicted charging duration TD of the target vehicle; Step S103, taking the time point when the target vehicle arrives at the target charging area as the charging start time, and obtaining the charging time period of the target vehicle as [t3, t4] according to the predicted charging time TD, wherein t3 represents the starting time point of the charging time period, and t4 represents the ending time point of the charging time period, and establishing a time axis, marking the peak power consumption time period and the charging time period on the time axis, marking the overlapping time period of the two as [t5, t6], and calculating the overlap degree C1, the calculation formula of which is C1=t6-t5 / t3-t4, wherein t3-t4 represents the charging time period of the charging time period [t3, t4], and t6-t5 represents the overlapping time period of the overlapping time period [t5, t6].

[0018] For example, the target vehicle arrives at the target charging area at 19:00, the peak power consumption period in the target charging area is [17:00, 21:00], and the predicted duration of the target vehicle is 1 hour. The charging period of the target vehicle is [19:00, 20:00]. According to the time axis, the overlapping period of the peak power consumption period and the charging period is [19:00, 20:00], where the overlap degree C1=1h / 1h=1.

[0019] Step S2: if the overlap is greater than the overlap threshold, obtain the passing area of ​​the target vehicle's route, extract the charging area with idle charging piles in the passing area as the candidate charging area, calculate the function value of the candidate charging area, and output the recommended replacement candidate charging area according to the function value; The step S2 comprises: Step S201: Set the overlap threshold to C 阈 , if C1>C 阈 When , the route of the target vehicle to the target charging area is obtained, the area passed by the route is marked as the passing area, all the charging areas in the passing area are extracted, the peak power consumption time period of each charging area is obtained and marked on the time axis, and the time point t9 when the target vehicle arrives at the designated charging area is predicted, and the time point when the target vehicle arrives at the designated charging area is used as the charging start time. The charging time period of the target vehicle is predicted to be [t9, t10], and the time period when the charging time period overlaps with the peak power consumption time period of the designated charging area is marked as [t7, t8]. The overlap degree C2 between the target vehicle and the corresponding charging area is calculated, and the calculation formula is C2=t8-t7 / t10-t9, where t10-t9 represents the charging time length of the charging time period [t9, t10], and t8-t7 represents the overlap time length of the overlap time period [t7, t8]. The charging area with C2<C1 is used as the alternative charging area; Step S202: Eliminate all the charging areas in the candidate charging area where the charging piles are fully occupied, obtain all the types of charging piles in the charging area with idle charging piles, divide the charging piles into fast charging type and slow charging type, obtain the number of idle charging piles and the types of idle charging piles in each candidate charging area, record the number of fast charging type charging piles as D1, record the number of slow charging type charging piles as D2, and calculate the slow charging proportion q=D2 / D1+D2; Step S203, obtain the charging record of each idle charging pile in the candidate charging area, mark the record of unfinished charging order in the charging record as abnormal charging record, obtain the end time of the charging order corresponding to all abnormal charging records, extract the start time of the next charging record recorded by the current charging pile after the abnormal charging record, calculate the time difference T1 between the two records, if T1<T1 阈值 , then the abnormal charging record is judged as the target abnormal data, all target abnormal data are extracted, and the abnormal charging rate Q of the charging pile is calculated. i =n / m, where n is the number of all target abnormal data of a single charging pile, m is the number of all abnormal charging records in the candidate charging area, Q i is the abnormal charging rate of the i-th charging pile; Step S204: Calculate the function value K of each candidate charging area, and the calculation formula is K=α1*∑Q i +α2*q, where i=1, 2, 3, ..., i, α1 and α2 represent weight values, * represents the product of two values, and the charging area with the smallest function value is output as the recommended candidate charging area.

[0020] For example, setting C 阈=0.2, since C1=1 is greater than C2=0.2, the method in step S1 is used to calculate the value of C2, the process is: obtain the charging time period to reach the designated charging area [19:30, 20:30], the peak power consumption time period of the designated charging area is [17:00, 20.00], the charging time period and the time period of the designated charging area overlap is [19:30, 20:00], then C2=0.5h / 1h=0.5, since C2 is less than C1, C2 is used as the alternative charging area; The number of available fast-charging charging piles in the candidate charging area is 5, and the number of available slow-charging charging piles is 3. The fast-charging ratio is q=3 / 5+3=0.375. Extract the abnormal charging records of the charging piles in the candidate charging area, and record the time difference between a certain abnormal charging record and the next charging record as T1. If T1 is less than T1 阈值 , it means that the target vehicle is recharged again in a very short period of time, which means that the abnormal record is determined to be caused by charging abnormality, not the completion of the charging order or the charging vehicle actively stopping charging; If the number of abnormal charging data of the charging pile is 20 and the total number of abnormal charging data of the charging area where the charging pile is located is 110, then the abnormal charging rate of the charging pile is calculated as Q = 20 / 110 = 0.18. Add the abnormal charging rates of all idle charging piles to get ∑Q i =1.2; Set α1 and α2 to fixed values, the sum of α1 and α2 is 1, generally α1 is 0.6, α2 is 0.4, calculate K=0.6*1.2+0.4*0.375=0.87, the larger the calculated K value, the more likely the charging area is to have abnormal charging or it is not easy to use fast charging type charging piles for charging in the charging area. Both of the above situations will affect the charging efficiency. Therefore, the larger the K value, the less recommended it is to select the charging area. Finally, the charging area corresponding to the smallest K value is output as the recommended alternative charging area.

[0021] Step S3: After the target vehicle is connected to the charging pile, the charging area where the connected charging pile is located is obtained, the peak power consumption time period of the power distribution area where the charging area is located is determined, and the coverage value between the current vehicle's charging time period and the peak power consumption time period of the power distribution area is calculated; The step S3 comprises: Step S301, obtain the divided distribution network area, mark the charging pile to which the target vehicle is connected, obtain the distribution network area where the charging pile is located, extract the peak power consumption time period [tR1, tR2] of the distribution network area, where tR1 represents the starting time point of the peak power consumption time period of the distribution network area, and tR2 represents the ending time point of the peak power consumption time period of the distribution network area, obtain the current estimated charging time period [tR3, tR4] of the target vehicle, where tR3 represents the starting time point of the estimated charging time period, and tR4 represents the ending time point of the estimated charging time period, obtain the overlapping time period of the estimated charging time period and the peak power consumption time period and mark it as [tR5, tR6], calculate the coverage value S, and the calculation formula is S=tR6-tR5 / tR4-tR3, where tR6-tR5 represents the overlapping duration of the overlapping time period [tR5, tR6], and tR4-tR3 represents the charging duration of the estimated charging time period [tR3, tR4].

[0022] For example, the peak power consumption time period in the distribution network area is [15:30, 21:30], and the expected charging time period is [19:50, 21.20]. The expected charging time period overlaps with the peak power consumption time period at [19:50, 21.20], and S = 1.5h / 1.5 = 1 Step S4: If the coverage value is greater than the coverage value threshold, the historical charging data of the target vehicle is obtained, the historical charging record of the target vehicle is extracted, the charging frequency of the target vehicle is calculated, the standby power demand of the target vehicle is analyzed, and the recommended charging time of the target vehicle is calculated based on the standby power demand, and pushed to the user end.

[0023] The step S4 comprises: Step S401: Set the coverage threshold to S 阈 , if S>S 阈值 , then the historical charging data of the target vehicle is obtained, the historical charging record of the target vehicle is extracted, and the record of the target vehicle completing a charging order is marked as a charging number; Step S402, calculate the charging frequency F of the target vehicle, the calculation formula is F=the number of charging times of the target vehicle / the total usage time of the target vehicle, obtain the daily usage of the target vehicle, extract the urban area where the target vehicle travels every day, mark the urban area where the target vehicle appears most times as the frequently used urban area, obtain the daily power consumption of the target vehicle in the frequently used urban area to form a line graph, calculate the daily average power consumption P1 of the target vehicle in the frequently used urban area, and obtain the maximum value P of the daily power consumption man ; Step S403, extracting the target vehicle's single-day driving route in a commonly used urban area, recording the target vehicle's one-day driving route in a commonly used urban area as a target event, obtaining the driving routes in two adjacent target events, eliminating the same sections in the two driving routes corresponding to the two target events, retaining different sections as target sections, calculating the power consumption used by the target vehicle on the target section, marking it as reserved power consumption, obtaining all reserved power consumption, generating a line graph in units of days, and calculating the reserved power consumption average value P2 of the reserved power consumption; wherein, calculating the power consumption of different sections as the reserved power consumption is to avoid insufficient power consumption caused by unexpected situations when the user uses the vehicle.

[0024] Step S404: Calculate the standby power demand based on the daily average power consumption and the average reserved power consumption. The calculation formula is: W = P1 + P2 * β1 + P man *β2, where β1 and β2 are weight values. According to the current remaining power of the target vehicle, the charging time when the standby power demand W is reached is predicted, and the standby charging time period is obtained as [tR7, tR8]. The end time point of the standby charging time period is obtained as tR8, and the end time point of the peak power consumption time period is tR2. According to the time sequence, if tR8 < tR2, the time between tR8 and tR2 is marked as T2. According to the formula, TP is calculated as (T2 / tR8-tR7) + b, where TP represents the waiting time ratio, tR8-tR7 represents the charging time of the standby charging time period, and b is a fixed value. If TP> T2 阈 , then the recommended charging time is [tR7, tR8]; if tR8 ≥ tR2 or TP ≤ T2 阈 , the recommended output power is full charge.

[0025] Among them, if the end time of the backup charging time period is less than the end time of the peak power consumption time period, it means that after the backup charging is completed, it is still in the peak power consumption time period of the area. At this time, the end time of the backup charging time period and the end time of the peak power consumption time period are obtained to obtain the remaining time for the peak to end, and calculate TP=0.5h*2h+3=4. If TP>T2 阈 , it means that there is still a long time before the end of the peak period of electricity consumption. At this time, it is not recommended to continue charging. Instead, it is recommended to stop charging after charging to the standby power demand; if TP≤T2 阈 , the manual indicates that the end time of the peak electricity consumption period is very close, and you can continue to wait for the end of the peak electricity consumption period, so you can output the recommended full charge result. If the end time of the standby charging period is greater than the peak electricity consumption period, that is, the end time of charging is after the peak electricity consumption period, it means that there will be no peak and valley load here, so you can directly charge it fully, and the recommended power output is full charge.

[0026] Embodiment 2: Figure 2 As shown, an artificial intelligence-based intelligent fusion management system for transportation energy is applied to an artificial intelligence-based intelligent fusion management method for transportation energy, the management system includes a peak power consumption prediction module, a charging area selection module and a power demand analysis module; the peak power consumption prediction module is used to predict the peak power consumption time period of each charging area and the peak power consumption time period of each distribution network area, and calculate the overlap between the charging time period of the target vehicle and the peak power consumption time period; the charging area selection module is used to output recommended alternative charging areas based on the calculated overlap and the functional value of each charging area; the power demand analysis module is used to analyze and calculate the backup power demand of the target vehicle, calculate the recommended charging time of the target vehicle based on the backup power demand, and push it to the user end.

[0027] The power consumption peak prediction module includes a charging area unit, a distribution area unit and a target vehicle unit. The charging area unit is used to obtain the power consumption peak time period of the target charging area, the distribution area unit is used to obtain the power consumption peak time period of the divided distribution network area; and the target vehicle unit is used to obtain the charging time period of the target vehicle.

[0028] The charging area selection module includes a charging pile analysis unit and an area selection unit. The charging pile analysis unit is used to calculate the abnormal probability of each charging pile and the fast charging proportion of the charging area corresponding to the charging pile; the area selection unit calculates the function value of the area according to the abnormal charging rate and the fast charging proportion of the charging area, and outputs the recommended alternative charging area according to the function value.

[0029] The power demand analysis module includes a power statistics unit, a redundant unit and an output unit. The power statistics unit is used to obtain the daily power consumption of the target vehicle and calculate the average daily power consumption of the target vehicle; the redundant unit is used to calculate the average reserved power consumption of the target vehicle; the output unit is used to calculate the standby power demand based on the daily power consumption and the average reserved power consumption, and output the recommended charging time based on the standby power demand.

[0030] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An intelligent fusion management method for transportation energy based on artificial intelligence, characterized by: The management method comprises the following steps: Step S1, obtaining the target charging area marked by the target vehicle, predicting the arrival time and charging time period of the target vehicle, obtaining the peak power consumption time period of the target charging area, determining whether the charging time of the target vehicle overlaps with the peak power consumption time period of the target charging area, and calculating the degree of overlap; Step S2: if the overlap is greater than the overlap threshold, obtain the passing area of ​​the target vehicle's route, extract the charging area with idle charging piles in the passing area as the candidate charging area, calculate the function value of the candidate charging area, and output the recommended replacement candidate charging area according to the function value; Step S3: After the target vehicle is connected to the charging pile, the charging area where the connected charging pile is located is obtained, the peak power consumption time period of the power distribution area where the charging area is located is determined, and the coverage value between the current vehicle's charging time period and the peak power consumption time period of the power distribution area is calculated; Step S4: If the coverage value is greater than the coverage value threshold, the historical charging data of the target vehicle is obtained, the historical charging record of the target vehicle is extracted, the charging frequency of the target vehicle is calculated, the standby power demand of the target vehicle is analyzed, and the recommended charging time of the target vehicle is calculated based on the standby power demand, and pushed to the user end.

2. According to the method of intelligent fusion management of traffic energy based on artificial intelligence in claim 1, it is characterized by: The step S1 comprises: Step S101, select a target charging area, obtain historical power consumption data of the target charging area, analyze the peak power consumption time period of the target charging area based on the historical power consumption data, and record the peak power consumption time period as [t1, t2], where t1 represents the start time point of the peak power consumption time period, and t2 represents the end time point of the peak power consumption time period; Step S102: record the time point when the target vehicle arrives at the target charging area as t3, obtain the charging record of the target vehicle, extract the record of the charging amount reaching full charge in the charging record as the target charging record data, obtain the charging record duration corresponding to each target charging record data, input the charging amount and charging record duration in each charging record data into the training model, obtain the charging duration prediction model, input the target charging amount of the target vehicle, and output the predicted charging duration TD of the target vehicle; Step S103, taking the time point when the target vehicle arrives at the target charging area as the charging start time, and obtaining the charging time period of the target vehicle as [t3, t4] according to the predicted charging time TD, wherein t3 represents the starting time point of the charging time period, and t4 represents the ending time point of the charging time period, and establishing a time axis, marking the peak power consumption time period and the charging time period on the time axis, marking the overlapping time period of the two as [t5, t6], and calculating the overlap degree C1, the calculation formula of which is C1=t6-t5 / t3-t4, wherein t3-t4 represents the charging time period of the charging time period [t3, t4], and t6-t5 represents the overlapping time period of the overlapping time period [t5, t6].

3. According to claim 2, a method for intelligent fusion management of traffic energy based on artificial intelligence is characterized by: The step S2 comprises: Step S201: Set the overlap threshold to C 阈 , if C1>C 阈 When , the route of the target vehicle to the target charging area is obtained, the area passed by the route is marked as the passing area, all the charging areas in the passing area are extracted, the peak power consumption time period of each charging area is obtained and marked on the time axis, and the time point t9 when the target vehicle arrives at the designated charging area is predicted, and the time point when the target vehicle arrives at the designated charging area is used as the charging start time. The charging time period of the target vehicle is predicted to be [t9, t10], and the time period when the charging time period overlaps with the peak power consumption time period of the designated charging area is marked as [t7, t8]. The overlap degree C2 between the target vehicle and the corresponding charging area is calculated, and the calculation formula is C2=t8-t7 / t10-t9, where t10-t9 represents the charging time length of the charging time period [t9, t10], and t8-t7 represents the overlap time length of the overlap time period [t7, t8]. The charging area with C2<C1 is used as the alternative charging area; Step S202: Eliminate all the charging areas in the candidate charging area where the charging piles are fully occupied, obtain all the types of charging piles in the charging area with idle charging piles, divide the charging piles into fast charging type and slow charging type, obtain the number of idle charging piles and the types of idle charging piles in each candidate charging area, record the number of fast charging type charging piles as D1, record the number of slow charging type charging piles as D2, and calculate the slow charging proportion q=D2 / D1+D2; Step S203, obtain the charging record of each idle charging pile in the candidate charging area, mark the record of unfinished charging order in the charging record as abnormal charging record, obtain the end time of the charging order corresponding to all abnormal charging records, extract the start time of the next charging record recorded by the current charging pile after the abnormal charging record, calculate the time difference T1 between the two records, if T1<T1 阈值 , then the abnormal charging record is judged as the target abnormal data, all target abnormal data are extracted, and the abnormal charging rate Q of the charging pile is calculated. i =n / m, where n is the number of all target abnormal data of a single charging pile, m is the number of all abnormal charging records in the candidate charging area, Q i is the abnormal charging rate of the i-th charging pile; Step S204: Calculate the function value K of each candidate charging area, and the calculation formula is K=α1*∑Q i +α2*q, where i=1, 2, 3, ..., i, α1 and α2 represent weight values, * represents the product of two values, and the charging area with the smallest function value is output as the recommended candidate charging area.

4. The method for intelligent fusion management of transportation energy based on artificial intelligence according to claim 1 is characterized in that: The step S3 comprises: Step S301, obtain the divided distribution network area, mark the charging pile to which the target vehicle is connected, obtain the distribution network area where the charging pile is located, extract the peak power consumption time period [tR1, tR2] of the distribution network area, where tR1 represents the starting time point of the peak power consumption time period of the distribution network area, and tR2 represents the ending time point of the peak power consumption time period of the distribution network area, obtain the current estimated charging time period [tR3, tR4] of the target vehicle, where tR3 represents the starting time point of the estimated charging time period, and tR4 represents the ending time point of the estimated charging time period, obtain the overlapping time period of the estimated charging time period and the peak power consumption time period and mark it as [tR5, tR6], calculate the coverage value S, and the calculation formula is S=tR6-tR5 / tR4-tR3, where tR6-tR5 represents the overlapping duration of the overlapping time period [tR5, tR6], and tR4-tR3 represents the charging duration of the estimated charging time period [tR3, tR4].

5. The method for intelligent fusion management of transportation energy based on artificial intelligence according to claim 4 is characterized in that: The step S4 comprises: Step S401: Set the coverage threshold to S 阈 , if S>S 阈值 , then the historical charging data of the target vehicle is obtained, the historical charging record of the target vehicle is extracted, and the record of the target vehicle completing a charging order is marked as a charging number; Step S402, calculate the charging frequency F of the target vehicle, the calculation formula is F=the number of charging times of the target vehicle / the total usage time of the target vehicle, obtain the daily usage of the target vehicle, extract the urban area where the target vehicle travels every day, mark the urban area where the target vehicle appears most times as the frequently used urban area, obtain the daily power consumption of the target vehicle in the frequently used urban area to form a line graph, calculate the daily average power consumption P1 of the target vehicle in the frequently used urban area, and obtain the maximum value P of the daily power consumption man ; Step S403, extracting the target vehicle's single-day driving route in the commonly used urban area, recording the target vehicle's one-day driving route in the commonly used urban area as a target event, obtaining the driving routes in two adjacent target events, eliminating the same sections in the two driving routes corresponding to the two target events, retaining different sections as target sections, calculating the power consumption used by the target vehicle on the target section, marking it as reserved power consumption, obtaining all reserved power consumption, generating a line graph in units of days, and calculating the reserved power consumption average value P2 of the reserved power consumption; Step S404: Calculate the standby power demand based on the daily average power consumption and the average reserved power consumption. The calculation formula is: W = P1 + P2 * β1 + P man *β2, where β1 and β2 are weight values. According to the current remaining power of the target vehicle, the charging time when the standby power demand W is reached is predicted, and the standby charging time period is obtained as [tR7, tR8]. The end time point of the standby charging time period is obtained as tR8, and the end time point of the peak power consumption time period is tR2. According to the time sequence, if tR8 < tR2, the time between tR8 and tR2 is marked as T2. According to the formula, TP is calculated as (T2 / tR8-tR7) + b, where TP represents the waiting time ratio, tR8-tR7 represents the charging time of the standby charging time period, and b is a fixed value. If TP> T2 阈 , then the recommended charging time is [tR7, tR8]; if tR8 ≥ tR2 or TP ≤ T2 阈 , the recommended output power is full charge.

6. An artificial intelligence-based intelligent fusion management system for transportation energy, which is applied to an artificial intelligence-based intelligent fusion management method for transportation energy according to any one of claims 1 to 5, characterized in that: The management system includes a peak power consumption prediction module, a charging area selection module and a power demand analysis module; the peak power consumption prediction module is used to predict the peak power consumption time period of each charging area and the peak power consumption time period of each distribution network area, and calculate the overlap between the charging time period and the peak power consumption time period of the target vehicle; The charging area selection module is used to output the recommended alternative charging area based on the calculated overlap and the functional value of each charging area; the power demand analysis module is used to analyze and calculate the backup power demand of the target vehicle, calculate the recommended charging time of the target vehicle based on the backup power demand, and push it to the user end.

7. The traffic energy intelligent fusion management system based on artificial intelligence according to claim 6 is characterized by: The power consumption peak prediction module includes a charging area unit, a distribution area unit and a target vehicle unit. The charging area unit is used to obtain the power consumption peak time period of the target charging area, the distribution area unit is used to obtain the power consumption peak time period of the divided distribution network area; and the target vehicle unit is used to obtain the charging time period of the target vehicle.

8. The traffic energy intelligent fusion management system based on artificial intelligence according to claim 6 is characterized by: The charging area selection module includes a charging pile analysis unit and an area selection unit. The charging pile analysis unit is used to calculate the abnormal probability of each charging pile and the fast charging proportion of the charging area corresponding to the charging pile; the area selection unit calculates the function value of the area according to the abnormal charging rate and the fast charging proportion of the charging area, and outputs the recommended alternative charging area according to the function value.

9. The traffic energy intelligent fusion management system based on artificial intelligence according to claim 6 is characterized by: The power demand analysis module includes a power statistics unit, a redundant unit and an output unit. The power statistics unit is used to obtain the daily power consumption of the target vehicle and calculate the average daily power consumption of the target vehicle; the redundant unit is used to calculate the average reserved power consumption of the target vehicle; the output unit is used to calculate the standby power demand based on the daily power consumption and the average reserved power consumption, and output the recommended charging time based on the standby power demand.

Images