KR101570302B1 - Method and system for predicting demand of recharging electric vehicles for electricity network management - Google Patents

Abstract

A method for predicting the charging demand of an electronic vehicle (EV) by using a computer includes the steps of: determining the number of all EVs and the number of operating EVs at a reference point and a prediction point; determining the respective values of the matrix elements in a charging amount distribution matrix and in a conversion matrix in order to convert the charging amount distribution matrix, which expresses the charging amount distribution of all EVs at the reference point as the number of EVs in multiple converted discharge time units, to a predicted charging amount distribution matrix, which expresses the charging amount distribution of all EVs at the prediction point as the number of EVs in multiple converted discharge time units, by the conversion matrix; and calculating the number of EVs having increased charging amounts at the prediction point with respect to the reference point as the number of charging prediction EVs based on the values of the matrix elements in the charging amount distribution matrix and the conversion matrix enabling the convergence of the number of operation prediction EVs at the prediction point, which is obtained by multiplying the charging amount distribution matrix at the reference point and the conversion matrix, to the number of operating EVs at the prediction point.

Description

[0001] METHOD AND SYSTEM FOR PREDICTING ELECTRIC VEHICLE FOR ELECTRICITY NETWORK MANAGEMENT [0002] FIELD OF THE INVENTION [

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a power system management technique, and more particularly, to a power system management technique that considers charging demand of an electric vehicle.

The plug-in hybrid electric vehicle, which includes only a battery without an engine, can charge the battery from the commercial system or can be charged from the internal combustion engine power, as well as a narrow meaning electric vehicle charging the battery from the commercial system.

Because electric vehicles have fewer than several tens to several hundreds of kWh of batteries, even if they are fully charged and operated only a few times, their electricity usage exceeds the normal monthly electricity consumption of a typical household.

As electric cars with such large capacity batteries are increasingly commercialized and sold, the influence of charging demand of electric vehicles on power demand management is becoming more and more important. At first glance, many electric cars that will be supplied to the market are expected to have a relatively low rate of charging during commute time, but it is expected that the rate of charging and charging will increase relatively during the day or night.

But simple and rough estimates of this level of operation and charging patterns for hundreds of thousands of electric cars are not enough to manage power demand.

In addition, as the electric car charging technology develops, even a large capacity battery can be charged in only one to two hours through the commercial power distribution system. Therefore, large-capacity batteries of electric vehicles that are waiting to be charged while parking are expected to be effective measures to distribute power demand in combination with smart grid technology. That is, the power once charged into the battery can be fed back into the commercial distribution system in accordance with a macroscopic power management policy. The arrival of the electric car era has complicated the management of the power system.

Generally, it is desirable that the power generation system is appropriately adjusted according to the load amount so that the balance of power supply and demand is maintained. If the supply and demand of the power system is unbalanced, the voltage and frequency quality of the power system may be deteriorated, causing large-scale power outages locally somewhere in the power distribution system or degrading the service life of the equipment.

Accordingly, there is a need for a methodology capable of accurately predicting the operation patterns and electric power demand of electric vehicles, and thereby providing predicted power demand information for reference in determining power generation in the power system.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and system for estimating the number of charged electric vehicles.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and system for predicting charging demand of an electric vehicle, which can consider collective operation, charging, discharging and standby patterns of electric vehicles irrespective of the performance of individual electric vehicles or battery capacity.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and system for predicting charging demand of an electric vehicle that can improve the performance of an electric vehicle and the trend of increasing battery capacity.

A problem to be solved by the present invention is to provide a method and system for predicting charging demand of an electric car, which can take into account an increase in the number of electric vehicles to be supplied.

The present invention provides a method and system for predicting charging demand of an electric vehicle, which can easily and flexibly adapt to various conditions that may vary depending on countries, regions and times.

The solution to the problem of the present invention is not limited to those mentioned above, and other solutions not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a computer-

The computer comprising:

(a) determining the total number of electric vehicles and the number of electric vehicles in operation at the reference time and the forecasting time;

(b) a charge amount distribution matrix expressing a charge amount distribution of all the electric vehicles at the reference time point as electric number logarithmic values for each of the plurality of converted discharge time units, wherein the charge amount distribution of all the electric vehicles at the predicted time point is divided into a plurality of converted discharge time units Determining values of matrix elements of the charge amount distribution matrix and the transform matrix so as to be transformed into a predicted charge amount distribution matrix expressed by electric vehicle logarithmic values;

(c) determining whether the driving predictive electric-power logarithm of the prediction time obtained by multiplying the conversion matrix by the charge-amount distribution matrix at the reference point converges to the electric-vehicle logarithm of the driving of the prediction time point;

(d) repeating steps (b) and (c) if it is not determined that the number of vehicles predicted to converge to the number of electric vehicles during operation; And

(e) If it is determined that the driving forecast number of electric vehicles is converged to the number of electric vehicles during operation, the number of electric vehicles whose charging amount is increased at the predicted time point in comparison with the reference time is charged based on the values of the matrix elements of the charging amount distribution matrix and the conversion matrix As an estimated number of electric vehicles,

The converted discharge time unit may be a value obtained by converting the charged amount into a time unit.

According to an embodiment, the total number of electric vehicles and the number of electric vehicles in operation can be determined in consideration of the number of electric vehicles compared to the total number of vehicles based on the traffic volume of all the vehicles and the ratio of the vehicles.

According to an embodiment, each of the charge amount distribution matrix and the predicted charge amount distribution matrix may be matrices obtained by arranging all the electric vehicles according to a charge amount unit converted into a plurality of time units with respect to each charge amount.

According to one embodiment, the relation between the charge amount distribution matrix, the predicted charge amount distribution matrix, and the transformation matrix is expressed by the following equation

(T +? T) is a predicted charge distribution distribution matrix of the reference time t, B (t) is a transformation matrix, A (t + Is a 1 × n matrix composed of n matrix elements, the transformation matrix is an n × n matrix, n is a maximum charge amount of the electric vehicle As a closest natural number which is larger or smaller than a value obtained by dividing the discharge time by a unit of discharge discharge time.

According to one embodiment, the transformation matrix is defined by the following equation

는 기준 시점(t)에서 (j-1) 환산 방전 시간 단위의 충전량을 소모한 전기차가 예측 시점(t+Δt)에 (i-1) 환산 방전 시간 단위의 충전량을 소모한 전기차로 되는 비율일 수 있다., And matrix elements (i, j) of each matrix position (i, j) of the transformation matrix B

Is the ratio of the electric vehicle consuming the charging amount of the (j-1) converted discharge time unit at the reference time t to the electric vehicle consuming the charging amount of the (i-1) converted discharge time unit at the prediction time t + .

According to one embodiment, step (b)

(T) and a transformation matrix B (t) at a reference time t satisfying the simultaneous equations of the matrix A (t) and the matrix F (t) according to the Monte Carlo method.

According to an exemplary embodiment, the step (b) may include calculating at least one of the values of some matrix elements of the conversion matrix as at least one of a conversion discharge time unit, a battery efficiency of an electric vehicle, a rated motor capacity, Or 0 based on the number of bits of the data.

According to one embodiment, some of the matrix elements of the transformation matrix may be represented by the following equations

Where? T is a time interval between the prediction time and the reference time, and? H may be a unit of the converted discharge time.

According to one embodiment, some of the matrix elements of the transformation matrix may be represented by the following equations

, Where e is the ratio of the available travel time to be charged when the electric vehicle is charged by the charge time, and [Delta] h may be the converted discharge time unit.

According to one embodiment, the charge amount distribution matrix at the reference time point is calculated by the following equation

The predicted charge amount distribution matrix of the electric vehicles predicted to be on standby at the predicted time is multiplied by the diagonal matrix of the conversion matrix,

Wherein A _REST (t + Δt) is the distribution predicted charge of the electric vehicle is expected to be waiting in the prediction time matrix, D (t) is a diagonal matrix of the transformation matrix, the prediction charge distribution matrix A _REST (t + Δt ) May be the logarithm of the electric vehicles predicted to be waiting at the predicted time.

According to one embodiment, the charge amount distribution matrix at the reference time point is calculated by the following equation

The predicted charge amount distribution matrix of the electric vehicle, which is predicted to be operated at the predicted point in time, is obtained by multiplying the lower triangular matrix of the conversion matrix,

Wherein A _DRIVING (t + Δt) is the distribution predicted charge of the electric vehicle that is expected while operating on the prediction time matrix, the L (t) is a lower triangular matrix of the transformation matrix, the prediction charge distribution matrix A _DRIVING (t + Δt) may be the number of electric vehicles estimated to be operating at the predicted time.

According to one embodiment, the charge amount distribution matrix at the reference time point is calculated by the following equation

The predicted charge amount distribution matrix of the electric vehicle, which is predicted to be charged at the predicted time point, is multiplied by the upper triangular matrix of the transformation matrix,

Wherein A _CHARGING (t + Δt) is the distribution predicted charge of the electric vehicle is predicted to being charged to the prediction point matrices, U (t) is a lower triangular matrix of the transformation matrix, the prediction charge distribution matrix A _CHARGING (t + Δt) may be the number of electric vehicles estimated to be charged at the predicted time.

According to one embodiment, the method for predicting charging demand for an electric vehicle

Repeating the procedures from step (a) to step (e) with respect to a new reference time point and a new prediction time point.

A computer program according to another aspect of the present invention is a computer program stored in a computer readable recording medium for implementing the steps of a method for predicting electric charge demand in a computer according to an exemplary embodiment of the present invention.

According to another aspect of the present invention,

An electric vehicle algebra number calculation unit for determining the total number of electric vehicles and the number of electric vehicles in operation based on the externally given ratio of electric vehicles, the traffic volume, and the driving vehicle to the entire vehicle;

Wherein the charge distribution matrix expressing the charge amount distribution of all of the electric vehicles at the reference time point as a plurality of electric charge logarithmic values by the unit of the discharge time unit is a predicted charge amount expressed by the electric charge logarithm values of the electric vehicles A transformation matrix optimizer for defining the transformation matrix to be transformed into a distribution matrix;

A matrix element operation unit for determining the values of the charge distribution matrix and the matrix elements of the transformation matrix;

For the values of the matrix elements of the charge accumulation matrix and the charge accumulation matrix at the reference point determined by the matrix element arithmetic unit, multiplying the charge accumulation matrix of the reference point by the transformation matrix to obtain the driving predicted electric power logarithm of the predicted time point, A solution verifying unit for determining whether or not the number of electric vehicles to be driven converges among the estimated number of electric vehicles to be operated and the number of electric vehicles to be operated at the predicted time provided by the electric vehicle number calculating unit;

And a charge amount distribution matrix which is determined by the solution verifying section to converge the logarithm of the driving predictive electric potential and the electric potential difference between the driving of the prediction point and the charging electric potential difference matrix, And a charge demand calculation unit for calculating the charge demand at the predicted time based on the number of charge predicted electric vehicles at the predicted time and the charger capacity.

According to an embodiment, each of the charge amount distribution matrix and the predicted charge amount distribution matrix may be matrices obtained by arranging all the electric vehicles according to a charge amount unit converted into a plurality of time units with respect to each charge amount.

According to an embodiment, the conversion matrix optimizer may calculate values of some matrix elements of the conversion matrix by at least one of a conversion discharge time unit, a battery efficiency of an electric vehicle, a rated motor capacity, a rated charging capacity, And can be operated to determine 0 or 1 on a basis.

According to one embodiment, the matrix element operation unit

The sum of the matrix elements of the charge amount distribution matrix is equal to the total number of the electric motors of the reference time point and the sum of the values of the matrix elements for each column of the conversion matrix is equal to 1, And to determine values of the charge distribution distribution matrix and the matrix elements of the transformation matrix, respectively.

According to one embodiment, the matrix element operation unit may operate to apply the Monte Carlo method to solve the solution of the simultaneous equations.

According to an embodiment, the charge demand calculator may calculate a sum of the matrix elements of the predicted charge amount distribution matrix at the prediction time obtained by multiplying the upper triangular matrix of the transformation matrix by the charge amount distribution matrix at the reference time, Which is the number of electric vehicles predicted to be charged in the electric power generating system.

According to the method and system for predicting charging demand of an electric vehicle of the present invention, collective operation, charging, discharging and waiting patterns of electric vehicles can be taken into consideration regardless of the performance of an individual electric vehicle or the battery capacity.

According to the method and system for predicting the charge demand of an electric vehicle of the present invention, it is possible to take into account the improvement in performance of an electric car and an increase in battery capacity.

According to the method and system for predicting charging demand of an electric vehicle of the present invention, an increase in the number of electric vehicles to be supplied can be considered.

According to the method and system for predicting charging demand of an electric vehicle according to the present invention, it is possible to easily and flexibly adapt to various conditions that may vary from country to country, region and time.

The effects of the present invention are not limited to those mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the following description.

FIG. 1 is a flowchart illustrating an electric vehicle charge demand forecasting method according to an embodiment of the present invention.
FIG. 2 is a table illustrating the ratio of vehicles traveling by hour provided from the traffic volume information providing system of the Ministry of Land, Transport and Traffic.
FIG. 3 is a diagram illustrating an outline of a conversion matrix according to an electric vehicle charging demand forecasting method according to an embodiment of the present invention.
FIG. 4 is a graph showing a change in the demand for charging an electric car in the day predicted according to an electric vehicle charging demand forecasting method according to an embodiment of the present invention.
5 is a block diagram illustrating an electric vehicle charging demand forecasting system according to an embodiment of the present invention.

For the embodiments of the invention disclosed herein, specific structural and functional descriptions are set forth for the purpose of describing an embodiment of the invention only, and it is to be understood that the embodiments of the invention may be practiced in various forms, The present invention should not be construed as limited to the embodiments described in Figs.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals are used for the same constituent elements in the drawings and redundant explanations for the same constituent elements are omitted.

FIG. 1 is a flowchart illustrating an electric vehicle charge demand forecasting method according to an embodiment of the present invention.

Referring to FIG. 1, an electric vehicle charge demand prediction method is basically performed in a computer that can process data and can be programmed to calculate a solution of a predetermined equation or formula.

The method for predicting the charging demand of the electric vehicle may start with the step of determining the total number of electric vehicles and the number of electric vehicles in operation at the reference time and the forecasting time, respectively, in step S11.

Next, in step S12, the computer determines whether or not the charge amount distribution matrix expressed by the electric-charge logarithmic values of the plurality of converted discharge time units as the charge amount distribution of all of the electric vehicles at the reference time point, The values of the charge amount distribution matrix and the matrix elements of the transformation matrix can be determined so that the distribution is converted into the predicted charge amount distribution matrix expressed by the electric vehicle logarithm values by the plurality of converted discharge time units. Here, the converted discharge time unit is a value obtained by converting the charged amount into a time unit of running, specifically, the amount of consumed charge according to the operation is a value converted in units of time in accordance with the rated motor capacity, battery efficiency and average running time of the electric vehicle.

In step S13, the computer calculates the number of driving predicted electric vehicles at the prediction time obtained by multiplying the charge amount distribution matrix at the reference point determined at step S12 by the conversion matrix, converges to the number of electric vehicles during operation at the prediction point determined at step S11. Or not.

If it is not determined in step S13 that the number of driving predictive electric vehicles converges to the number of electric vehicles in operation, the procedure returns to step S12 to repeat step S12 and step S13.

If it is determined in step S13 that the number of driving predicted electric vehicles converges to the number of electric vehicles in operation, the process can proceed to step S14.

In step S14, based on the values of the matrix elements of the charge amount distribution matrix and the conversion matrix determined to converge in step S13, the computer calculates the number of electric vehicles whose charged amount is increased at the predicted time point, It is calculated as the number of electric cars.

By repeating the procedures from the step S11 to the step S14 with respect to the next prediction time in the step S15, it is also possible to predict a considerably long-term charging prediction electric power logarithm.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram for explaining a charge demand of an electric vehicle at a predicted time point; FIG. The solution of the simultaneous equations generated from the matrix elements of the charge amount distribution matrix with respect to the charge amount distribution of the conversion matrix and the reference point electric loads is obtained. Next, the present invention is able to know the distribution of charge amounts of electric vehicles at the reference time and the predicted time from the above-described solution, identify the number of charged electric vehicles at the reference time, and estimate the number of electric cars to be charged at the predicted time have.

The base time and the forecasting time may be current or future, provided that only basic data can be provided to determine the traffic volume and vehicle ratios at each time point. However, it is obvious that the prediction time point is a time point exceeding the reference time point by a predetermined prediction interval.

The charge amount distribution matrix and the predicted charge amount distribution matrix are matrices obtained by classifying all electric vehicles into a plurality of predetermined charge amount units according to the charge amount at each time point. At this time, it is important that the charged amount unit for classifying electric vehicles is not an electric energy unit such as kWh, but a converted discharge time unit in terms of the nominal time unit, that is, the difference of the present charged amount with respect to the maximum charged amount Do.

That is, the charge amount distribution matrix and the predicted charge amount distribution matrix are obtained by dividing all the electric vehicles by the number of the maximum charged electric cars, the number of electric cars charged in a state corresponding to the amount of power consumed by one conversion time unit compared to the maximum charge amount, The number of electric vehicles charged to the extent corresponding to the consumption of electric power by two conversion discharge time units as compared with the maximum electric energy consumption time of the electric energy consumed, It is a one-dimensional matrix arranged in the same way as the number of electric cars.

However, it is preferable that the maximum charge amount is determined by a representative value of the maximum charge amount values of registered individual electric vehicles, for example, an average value, a middle value, or a mode value so that the performance deviation of the individual electric vehicle can be diluted.

According to the embodiment, instead of the maximum charge amount, the charge amount distribution matrix and the predicted charge amount distribution matrix may be arranged in accordance with the converted discharge time unit in which the present charge amount based on the full discharge is converted into the travelable time.

In this case, the charge amount distribution matrix and the predicted charge amount distribution matrix are obtained by dividing all the electric vehicles by the number of fully discharged electric vehicles, the number of electric cars charged in a state corresponding to the electric power that can be operated by one conversion discharge time unit, May be a one-dimensional matrix arranged in the same manner as the number of fully charged electric vehicles to the extent that they can be operated by the number of electric cars charged in a state corresponding to the electric power by the time unit, .

The following description is based on the case where the charge amount distribution matrix and the predicted charge amount distribution matrix are arranged according to the conversion discharge time corresponding to the difference of the present charge amount with respect to the maximum charge amount.

In this regard, each matrix element in the charge amount distribution matrix means the logarithm of electric vehicles having a specific charge amount at the reference time point, and each matrix element in the predicted charge amount distribution matrix means the logarithm of the electric charge having a specific charge amount at the prediction time point .

In addition, the conversion matrix capable of converting the charge amount distribution matrix at the reference time point into the predicted charge amount distribution matrix at the prediction time is a conversion matrix in which the charge amount of each electric vehicle, that is, the calculated discharge time, , Which means that it does not change while waiting, or that it decreases through sales of electricity or commercial lines.

As will be described below, appropriate constraints can be defined to reflect various charging and discharging requirements such as battery performance, charging charge, operating distance, user's habits, etc., and conversion Some of the matrix elements of the matrix can be forcibly fixed.

Matrix elements of the predicted charge amount distribution matrix can be determined from the solutions of the charge amount distribution matrix and each matrix element of the transformation matrix by, for example, a conventional simultaneous equations solution method or a Monte Carlo method.

Accordingly, if the matrix elements of the predicted charge amount distribution matrix at the prediction time point and the matrix elements of the charge amount distribution matrix at the reference time point are compared with each other, then the logarithm of the electric vehicles whose charge amount has increased rather than time elapses, The number of electric cars that are expected to be charged while reaching the point of time, that is, the number of charge predicted electric cars.

In step S11, the computer determines the total number of electric vehicles and the number of electric vehicles in operation at the reference time t and the forecasting time t + Δt, respectively .

The interval between the prediction time point (t +? T) and the reference time point (t) is the prediction interval? T.

On the other hand, the total number of the electric vehicles and the number of the electric vehicles may be the data obtained by examining in real time whether the individual electric vehicles are operated individually or not, but in a more realistic sense, The number of electric vehicles compared to the total number of vehicles based on the traffic volume and the ratio of vehicles to vehicles. The traffic volume and the vehicle traffic ratio can be calculated from data obtained by measuring the number of vehicles traveling on the road. For example, data certified from the traffic volume information providing system of the Ministry of Land, Transport and Maritime Affairs can be provided.

For example, in the case of continuously estimating fluctuating charge ratios during a day, that is, from 0 to 24 hours, the amount of traffic of all vehicles acquired every hour from one hour to 24 hours And driving vehicle ratio data can be prepared in advance.

If you want to predict the percentages of charge that can vary by specific time or region, you can prepare data for the traffic volume and vehicle ratio of all vehicles separately by year, season, month, period, and region.

In this case, the traffic volume and vehicle ratio data of all the vehicles can be composed of data of the average annual average, average monthly average, period average or local average traffic volume, and running vehicle rate collected every hour from 0:00 to 24:00.

The ratio of vehicles that are running compared to the total number of vehicles at the base time can be calculated based on the traffic volume and the running vehicle ratio.

If the total number of registered cars is 1000, and the percentage of vehicles driving at a specific point in time is 20%, 200 cars are on the road.

In order to determine the total number of alleys and the number of trains at the base time, it is assumed that the proportion of the running electric vehicles among the vehicles operating at the reference time is equal to the proportion of the electric vehicles among all the vehicles.

If there are 200 registered electric vehicles among the total registered automobiles of 1000, and the percentage of vehicles driving at a specific point in time is 20%, it can be estimated that there are 20 electric cars among the 200 cars running on the road.

Thus, the total number of electric vehicles and the number of electric vehicles in operation at the reference time t and the forecasting time t + t can be determined based on the information on the traffic volume and the running vehicle collected in the past.

FIG. 2 is a table illustrating an Average Domestic Traffic Volume according to hour provided from the traffic volume information providing system of the Ministry of Land, Transport and Traffic. FIG.

Referring to FIG. 2, it is estimated that 2.78% of the entire fossil fuel vehicles are running on the road from 0:00 to 1:00 in some areas. At 8:00 am - 9:00 am, 6.03 percent of all cars are on the road.

The values of the total number of electric vehicles and the number of electric vehicles during operation can be obtained at the reference time and the forecasting time in consideration of the ratio of the traffic volume and the running vehicle ratio data to the registered number of electric vehicles registered for the total number of registered automobiles.

Then, returning back to Fig. 1, in step S12, the computer calculates a charge amount distribution matrix A (t (t)) expressing the distribution of charge amounts of all the electric vehicles at the reference time t as a plurality of electric discharge vehicle- (T +? T) expressing the charge amount distribution of all the electric vehicles at the prediction time point (t +? T) as the electric power logarithm values by the plurality of converted discharge time units? H by the transformation matrix B ) And the matrix elements of the charge distribution matrix A (t) and the transformation matrix B (t), respectively.

Here, the converted discharge time unit? H is a value obtained by converting the consumed amount of consumed amount according to the operation into the unit of time, for example, 30 minutes or 1 hour. In order to simplify the prediction process, the charge amounts of all the electric vehicles are clustered according to an integral multiple of the converted discharge time unit (? H) by raising or lowering the actual running time.

It is preferable that the predicted interval? T and the converted discharge time unit? H have the same value or an integral multiple of each other. For example, if an electric vehicle charge demand is predicted every hour of the day, the predicted interval? T may be one hour and the converted discharge time unit? H may be one hour or 30 minutes.

At the current technology level, the battery of the electric car consumes the charged electric power during operation, so the shorter the predicted interval (Δt) and the converted discharge time unit (Δh), the better predict the actual charge demand. However, in such a case, the calculation amount required for predicting the charging demand may become large and the calculation time may increase. Also, since the data on the traffic volume and the running vehicle ratio are provided by being measured in real time, there is no guarantee that the charging demand will be more accurately predicted even if the converted discharge time unit (Δh) becomes shorter. However, interpolation of traffic volume and traffic volume ratio data per hour, for example, every 30 minutes, or traffic volume and traffic volume data is obtained more frequently, for example every 30 minutes, DELTA t) and the converted discharge time unit DELTA h can be made equal to each other.

The relation between the charge amount distribution matrix A (t), the predicted charge amount distribution matrix A (t +? T), and the transformation matrix B (t)

Here, the charge amount distribution matrix A (t) is a 1 × (n) matrix consisting of n matrix elements: A ₁ , A ₂ ,, A _i , ..., A _n n matrix. At this time, A ₁ is the logarithm of the electric vehicles in the buffered state where the converted discharge time is 0 (never operated after fully charged) at time t, and A ₂ is the number of electric vehicles A _i is the number of electric vehicles having battery charge corresponding to the (i-1) converted discharge time unit (Δh) at time t.

Naturally, the _sum of the values A ₁ to A _n as in Equation (2) is equal to the total number of electric cars at the reference time (t).

The transformation matrix B (t) can be expressed by the following equation (3).

는 기준 시점(t)에서 (j-1) 환산 방전 시간 단위(Δh)의 충전량 (j-1)Δh을 소모한 전기차가 예측 시점(t+Δt)에 (i-1) 환산 방전 시간 단위(Δh)의 충전량 (i-1)Δh을 소모한 전기차로 되는 비율을 의미한다.The matrix elements of each (i, j) of the transformation matrix B (t)

(I-1) converted discharge time unit (i-1) at the predicted time (t + t) by the electric vehicle consuming the charged amount (j-1)? H of the (I-1) < / RTI >< RTI ID = 0.0 > h, < / RTI >

For example, when the converted discharge time unit (Δh) is 0.5 hour, B _{2 ← 4} indicates that an electric car having a charged amount of 1.5 hours running at the reference time (t) travels 0.5 hour at the forecasting time (t + Δt) It is the ratio that becomes electric car with charge amount. In this case, the charging amount is increased when the converted discharge time is reduced. This means that the electric vehicle charges between the reference time t and the forecasting time t + t.

Accordingly, the transformation matrix can be interpreted as shown in FIG. Referring to FIG. 3, FIG. 3 illustrates an outline of a conversion matrix according to an EV demand prediction method according to an embodiment of the present invention. Referring to FIG.

In Fig. 3, the matrix elements of the transformation matrix B (t) are divided into a diagonal area with i = j, a lower triangular area with i > j, and an upper triangular area with i < j.

The matrix elements of the diagonal area with i = j are matrix elements for predicting the electric motors, that is, the electric motors to wait between the reference time t and the prediction time t +? t.

The matrix elements of the lower triangular region i > j are matrix elements for predicting the electric vehicles to increase the converted discharge time between the reference time t and the prediction time t + t, that is, electric cars to be operated or discharged.

On the other hand, the matrix elements of the upper triangular region i <j are matrix elements for predicting the reduced electric discharge time between the reference time t and the prediction time t + Δt, that is, electric cars to be charged.

As a result, the result obtained by multiplying the charged amount distribution matrix A (t) of the reference time t by the diagonal matrix D (t) of the transformation matrix B (t) predictions that predicted air electric vehicle is predicted to be a charge distribution matrix a _REST (t + Δt). The value obtained by adding the matrix elements of the predicted charge amount distribution matrix A _REST (t +? T) in Equation (4) is the atmospheric predicted electric vehicle logarithm.

The result obtained by multiplying the charge amount distribution matrix A (t) of the reference time t by the lower triangular matrix L (t) of the transformation matrix B (t) as shown in equation (5) A _DRIVING (t + Δt) is a predicted charge distribution matrix of the predicted electric vehicles that are expected to operate. The value obtained by adding the matrix elements of the predicted charge amount distribution matrix A _DRIVING (t +? T) in Equation (5) is the logarithm of the driving predictive electric vehicle.

Similarly, as shown in Equation 6, the result obtained by multiplying the charge amount distribution matrix A (t) of the reference time t by the upper triangular matrix U (t) of the transformation matrix B (t) (T +? T) of the predicted charge _quantity distribution matrix A _CHARGING The value obtained by adding the matrix elements of the predicted charge amount distribution matrix A _CHARGING (t +? T) in Equation (6) is the number of charge predicted electric vehicles.

Of course, A (t +? T) = A _REST (t +? T) + A _DRIVING (t +? T) + A _CHARGING (t +?

On the other hand, the matrix elements B _{i ← j} (1 _{≤ i ≤} n) belonging to each column in the transformation matrix B (t) of Equation (3) represent the charging amounts of j-converted discharge time units at the reference time t (T +? T) to the electric charges in the range from the charging amount of 0 conversion discharge time unit to the charging amount of n-1 conversion discharge time unit, the matrix elements B _{i The} value added by the column by _j should be 1 as shown in the following Equation 7. &quot; (7) &quot;

However, what these matrix elements actually mean can be changed depending on whether the calculated discharge time is defined as a time or a time when the operation can be performed. In the description of the present invention, It is defined as one hour.

The converted discharge time units Δh and n for determining the sizes of the charge amount distribution matrix A (t) and the conversion matrix B (t) can be determined according to the performance indices of the electric vehicle.

For example, if the rated battery capacity of an electric car is 24.4 kWh and the rated power consumption of the motor is 6.2 kWh, the maximum serviceable time of a fully charged electric car is about 4 hours. Therefore, if the converted discharge time unit DELTA h is 30 minutes, the charged amount of the electric vehicles can be divided from the 0 converted discharge time unit (0 DELTA h) to the 8 converted discharge time unit (DELTA h) The size n of the transformation matrix B (t) can be determined as (maximum travel time / unit of converted discharge time) + 1 = 9.

On the other hand, the product of the charge distribution matrix A (t) and the transformation matrix B (t) actually constitutes a simultaneous equations in which n (n + 1) unknowns produce n equations The solution of this simultaneous equations can not be obtained by the ordinary equation solution, and the solution of the simultaneous equations can be found by using the Monte-Carlo method.

The Monte Carlo method is a technique for finding the solution of the simultaneous equations by finding a random number satisfying a predetermined condition by repeatedly substituting a random number into the unknown. The algorithm of the Monte Carlo method is relatively simple, but if the number of unknowns is large as described above, it may take a long time to obtain the solution.

Accordingly, in order to reduce the number of unknowns in the simultaneous equations, embodiments of the present invention introduce a plurality of constraints related to the characteristics of the electric car, the electric bill, and the driving situation as follows.

Illustratively, the limiting conditions include battery efficiency (fuel economy), rated motor capacity, rated charger capacity, and may, according to embodiments, further include a utility bill.

For example, according to the traffic statistics of the National Statistical Office, the average daily mileage of a conventional car is about 40 km, and if an electric car is used for the same purpose as an existing car, the average daily mileage of the electric car is about 40 km.

At present, the efficiency of an electric car is known to be about 6.5 km / kWh, so the average daily electric power consumption of an electric car is about 40 kW / 6.5 km / kWh = about 6.2 kWh. If the charged charger capacity is 3.3 kWh, it is expected that the average used electric car will charge 1 hour per 0.5 hour operation.

The charging efficiency e = the additional driving time / charging time ≒ the rated charging capacity / the rated motor capacity can be regarded as the charging efficiency e when the ratio of the charging time to the charging time of the electric vehicle is defined as the charging efficiency e. In the above example, the charging efficiency e is about 0.5.

If the power demand is predicted on an hourly basis, the converted discharge time unit is preferably 0.5 hour, or 30 minutes. As the charging technology develops, if it can be operated for one hour by charging for one hour or for one hour by charging for 0.5 hours, the conversion discharge time unit can be appropriately changed accordingly.

It is preferable that the performance index of the electric vehicle is determined by a representative value of the individual performance index values of registered electric vehicles, for example, an average value, an intermediate value, or a mode value so that the performance deviation of the individual electric vehicle can be diluted.

On the other hand, considering each of the matrix elements B _{i ← j} (1≤i, _j ) of the transformation matrix B (t) in the equation (3), taking into account the rated motor capacity, the rated capacity of the charger, If we look more closely at j ≤ n, we can find matrix elements that have a value of zero if the average electric car is used normally.

For example, in the above example, since the predicted interval? T is one hour and the converted discharge time unit? H is 0.5 hour, the electric vehicle traveling logically within the one hour predicted interval? (2? H). That is to say, the average electric vehicle that has been consumed only by one conversion discharge time unit (1? H) at the reference time (t) can be converted into the electric vehicle consumed by the three conversion discharge time unit (3? H) at the prediction time (t + And can not be converted into an electric vehicle consumed by a 2-converted discharge time unit (2Δh) or a 4-converted discharge time unit (4Δh). Therefore, matrix elements such as B _{3 ← 1} and B _{4 ← 2} may have non-zero values, but matrix elements such as B _{2 ← 1} , B _{4 ← 1} , B _{3 ← 2} , B _{5 ← 2} , .

Accordingly, some matrix elements satisfying the condition of Equation (8) can be regarded as 0.

Further, in the above example, since the average electric vehicle can be operated for 0.5 hour (here, it is equal to one converted discharge time) at the time of charging with the average charger for one hour, the average electric vehicle Quot; only replenishes the charged amount by one converted discharge time unit (DELTA h). That is to say, the average electric vehicle consumed by the four-point discharge time unit 4? H at the reference time t is only converted to the electric car consumed by the three-time discharge time unit 3? H at the prediction time t + 1 after one hour , It can never be converted to a consumed or buffered electric vehicle by 2 converted discharge time units (2? H). Therefore, matrix elements such as B _{4 ← 5} and B _{3 ← 4} can have non-zero values, but matrix elements such as B _{3 ← 5} and B _{2 ← 4} can all be considered as zero.

Accordingly, some matrix elements satisfying the condition of Equation (9) can be regarded as zero.

Accordingly, according to the above example, the transformation matrix B (t) of Equation (3) can be simply summarized as Equation (10).

Using the fact that the sum of the matrix elements of each column is 1 according to the relation of Equation (7), the transformation matrix B (t) of Equation (10) can be derived the following simultaneous equations.

According to this description, the matrix elements of the charge amount distribution matrix A (t) and the transformation matrix B (t) are respectively determined in step S12 by obtaining solutions of the simultaneous equations of the equations (2) and (11) according to the Monte Carlo method It can be said that.

Further, according to an embodiment, the conversion matrix B (t) of Equation (10) can be made simpler if the limiting conditions further include the charging tendency condition of the driver according to the electric bill.

For example, as shown in Table 1, KEPCO charges electric power charges for electric vehicles differently depending on the load of the whole system, the supply voltage, or the season (season), as shown in Table 1.

division
Basic charge
(Won / kWh) Electricity charge (won / kWh) Load Summer
(July to August) Spring fall
(3 ~ 6, 9 ~ October) Winter
(November to February) Low pressure 2,320 Light load 55.80 56.90 78.20 Intermediate load 140.80 68.30 124.20 Maximum load 255.30 73.10 184.90 High pressure 2,500 Light load 50.90 51.80 67.70 Intermediate load 107.30 62.30 97.90 Maximum load 158.60 66.10 134.50

Table 1 below shows the light load, intermediate load, and peak load times set by KEPCO.

division Summer
(July to August) Spring fall
(3 ~ 6, 9 ~ October) Winter
(November to February) Light load 23:00 - 09:00 23:00 - 09:00 23:00 - 09:00 Intermediate load 09:00 - 11:00
12:00 - 13:00
17:00 - 23:00 09:00 - 11:00
12:00 - 13:00
17:00 - 23:00 09:00 - 10:00
12:00 - 17:00
20:00 - 22:00 Maximum load 11:00 - 12:00
13:00 - 17:00 11:00 - 12:00
13:00 - 17:00 10:00 - 12:00
22:00 - 23:00

According to Tables 1 and 2, the electricity rate varies by up to about five times, depending on the season, supply voltage and load time. Therefore, the user of the electric vehicle can not ignore the charging time zone when charging the electric vehicle.

Accordingly, it is possible to set exemplary charging tendency conditions as follows.

(1) If the predicted time is within a relatively high time zone, charge only the electric car when the charge amount is very low.

(2) If the forecasting time is relatively middle time, all waiting electric vehicles with a charge of less than mid-charge will be charged.

(3) If the predicted time is in a relatively low-cost time zone, all waiting electric vehicles except the very charged ones will be charged.

(4) Charged electric vehicles with minimal charge or fully discharged charge even if the charge rate is disadvantageous.

For example, the all of the A _n-1 (t) or A _n (t) in regardless of the plan charging inclination condition (4), the reference electric Among charge distribution of the matrix A (t) at the time (t) in accordance with Electric vehicles always charge at the predicted time (t + Δt).

In the above example in which n = 9, the electric vehicle consumed by the seven converted discharge time units 7? h or 8 converted discharge time units 8? h nearly discharged at the reference time t is the predicted time t +? t (7Δh) or 8-converted discharge time unit (8Δh) in the discharge time unit (6Δh) or 7-converted discharge time unit (7 [Delta] h).

B _{n ← n} and B _{n-1 ← n-1} among the matrix elements can be regarded as 0, and B _{n-2 ← n-1} and B _{n-1 ← n} are 1 Can be considered.

Reflecting this charge inclinational condition (4), the transformation matrix B (t) of Equation (10) can be further simplified as shown in Equation (12) and the simultaneous equations can be derived as Equation (13).

According to these explanations, the matrix elements of the charge amount distribution matrix A (t) and the transformation matrix B (t) are respectively determined in step S12 by obtaining the solution of the simultaneous equations of Equations 2 and 13 according to the Monte Carlo method It can be said that.

For example, according to the charge incentive condition (1), an electric car consumed by one conversion time unit (1? H) to four conversion time unit (4? H) at the reference time (t) , It can be estimated that the charged amount will not become higher at the predicted time (t + t).

In addition to these estimates, B _{1 ← 2} , B _{2 ← 3 and} so on can be regarded as 0, as well as B _{n ← n} and B _{n-1 ← n-1} , among the matrix elements.

Reflecting these charge bias conditions (1) and (4), the transformation matrix B (t) in equation (12) can be further simplified as shown in equation (14) and the derivation equation can be derived as in equation .

In addition, when the charging tendency condition (2) for predicting the charging demand in the time zone corresponding to the intermediate charging rate is applied, for example, if the charging amount of the 4-conversion discharge time unit (4Δh) ), Electric vehicles that were in a standby state with less than the middle charge amount are all charged, so that they can not have the same charge amount even at the predicted time (t + Δt), which is the time of charging for one hour logically. Thus, all matrix coefficients B _{i ← j} with i = j ≥ 5 can also be considered to be zero.

Similarly, when the charging tendency condition (3) for predicting the charging demand in the time zone corresponding to the low fare is applied, for example, when the charged amount of the 2-converted discharge time unit (4Δh) , The electric vehicles which were in a standby state with a charge amount lower than the 3-converted discharge time unit (3? H) are all charged, so that they can not still have the same charge amount at the predicted time (t +? T) . Thus, all matrix coefficients B _{i ← j} with i = j ≥ 3 can also be considered to be zero.

In this way, the simultaneous equations of Equations (11), (13) or (15) are reduced according to various exemplary charging tendency conditions and the number of unknowns is significantly reduced compared to Equation (7) There is little difficulty in seeking the sun.

In this way, taking into account the charging tendency conditions depending on the battery capacity, charging efficiency, charger performance, and charging plan for each reference time, a part of the matrix elements of the conversion matrix B (t) in Equation 3 is regarded as 0 or 1, You can solve the equation by applying the Monte Carlo method.

Determining the matrix elements of the charge amount distribution matrix A (t) and the transformation matrix B (t), respectively, in step S12 as described above, can be accomplished by solving the simultaneous equations of equations 3 and 11, 13 and 15 Can be said to be obtained by the Monte Carlo method.

If the matrix elements of the charge amount distribution matrix A (t) and the transformation matrix B (t) are determined in accordance with the Monte Carlo method respectively, then in step S13 the computer calculates the charge amount of the reference time point t determined in step S12 (T + t) determined at step S11 is calculated by subtracting the estimated driving electric-power logarithm A _DRIVING (t +? T) of the predicted moment t +? T obtained by multiplying the distribution matrix A (t) It is determined whether or not converging to the number of electric vehicles during operation.

In this step S13, the sum of the matrix elements of the driving predictive electric power logarithm A _DRIVING (t +? T) according to the equation (5) is calculated as the prediction time t +? T determined in step S11, Of the total number of electric vehicles in operation.

Operation predicting electric logarithmic A _DRIVING (t + Δt) is, like the above-described equation (5), the reference point in time (t) charge distribution matrix A (t) a lower triangular matrix L (t) of the transformation matrix B (t) of the .

If the transformation matrix B (t) of Equation 10 is used, step S13 can be expressed as a procedure for verifying the solution using Equation (17).

Since the simultaneous equations of Equation (11), Equation (13) or Equation (15) and Equation (2) are much greater than the number of equations, the elements of the solution set according to the Monte Carlo method can be plural. Among these plurality of solutions, a solution that does not converge to the number of electric vehicles during operation in which the number of driving predicted electric vehicles according to Equation (5) is determined according to the actual traffic volume and the vehicle driving ratio can be filtered out by Expression (17).

The error coefficient delta can be given, for example, to about 0.03.

Steps S12 and S13 may be repeated until a solution to be verified by equation (17) is derived.

In step S13, the sum of the matrix elements of A _DRIVING (t +? T), which is the driving predictive electric power number of the predicted time (t +? T), is made substantially equal to the electric vehicle number Is determined by the values of matrix elements of the charge distribution matrix A (t) and the transformation matrix B (t) of the electric vehicles at the reference time t.

In step S14, the computer compares the charging amount of the reference time point t with the charge amount distribution matrix A (t) based on the values of the matrix elements of the conversion matrix B (t) And calculates the number of electric vehicles in which the charge amount at the predicted time point (t +? T) increases as the number of charge predicted electric vehicles.

In other words, in step S14, as shown in equation (6), the charge-accumulation matrix A (t) of the reference time t is multiplied by the upper triangular matrix U (t) of the transformation matrix B The sum of the matrix elements of the predicted charge amount distribution matrix A _CHARGING (t + DELTA t) can be calculated as the number of charge predicted electric vehicles.

By repeating the procedures from the step S11 to the step S14 with respect to the next prediction time in the step S15, it is also possible to predict a considerably long-term charging prediction electric power logarithm.

FIG. 4 is a graph showing a change in the hourly charging ratio of the EV according to an embodiment of the present invention; FIG.

Referring to FIG. 4, even in the early morning hours where the electric vehicle is low in charge and the charging is completed during the night, the charge demand is low even though the charge is low, Is low. However, as the rate of operation is high in the morning and the discharged electric car is increased in the morning hours, the charging demand gradually increases even though the charge is expensive. This trend also appears in the afternoon. Demand for recharge is further increased as the rate rises to peak again in the evening hours and the charge is moderately low. At night time, charging demand rapidly increases at 23 o'clock, which is the cheapest rate during the day. However, as the number of electric cars that have been fully charged increases with the time of day, the demand for charging decreases gradually.

According to the graph of FIG. 4, it can be seen that there is a slight difference in the interval from afternoon to night due to the difference of the peak load time according to the season.

5 is a block diagram illustrating an electric vehicle charging demand forecasting system according to an embodiment of the present invention.

5, the electric vehicle charging demand forecasting system 50 includes an electric vehicle logarithmic calculating unit 51, a conversion matrix optimizing unit 52, a matrix element calculating unit 53, a solution verifying unit 54, 55).

More specifically, the electric vehicle logarithmic calculation unit 51 can determine the total number of electric vehicles and the number of electric vehicles during operation based on the externally given ratio of electric vehicles, the amount of traffic, and the ratio of vehicles to vehicles, .

The conversion matrix optimizing unit 52 compares the charge amount distribution matrix representing the charge amount distribution of all the electric vehicles at the reference point with the electric number logarithm values of the plurality of converted discharge time units into the electric charge distribution distributions of all the electric vehicles at the prediction time point, It is possible to define a transformation matrix to be transformed into a predicted charge distribution matrix expressed by logarithmic values. Here, the converted discharge time unit is a value obtained by converting the consumption amount of charge according to the operation into the time unit in accordance with the rated motor capacity, the battery efficiency and the average running time of the electric vehicle. The size of the conversion matrix can be determined among natural numbers close to the value obtained by dividing the maximum operating time determined based on the rated battery capacity of the electric car and the rated motor capacity divided by the converted discharge time unit.

According to the embodiment, the conversion matrix optimizing unit 52 converts the values of some matrix elements of the transformation matrix into the conversion discharge time unit, the battery efficiency of the electric vehicle, the rated motor capacity, the rated charger The charging condition may be determined to be 0 or 1 based on at least any one of the charging condition, the charge condition,

The matrix element operation unit 53 can determine the values of the matrix elements of the charge distribution distribution matrix and the optimized transformation matrix, respectively.

Specifically, the matrix element calculator 53 can determine the values of the matrix elements of the charge amount distribution matrix and the transform matrix according to Equations (2) and (7).

At this time, the matrix element computing unit 53 may apply the Monte Carlo method so that the number of simultaneous equations can be less than the number of unknowns.

The values of matrix elements of the charge amount distribution matrix and the transform matrix that are derived so as to satisfy the equations (2) and (7) may be a plurality of values.

Accordingly, the solution verifying unit 54 multiplies the charge accumulation matrix at the reference time with the transformation matrix, for the solution derived by the matrix element computing unit 53, that is, the values of the matrix elements of the charge accumulation matrix and the transformation matrix at the reference time The matrix element operating section 53 obtains the driving predicted number of electric vehicles at the predicted time point and determines whether or not the number of electric vehicles to be driven converges among the driving predicted electric vehicle number at the predicted point in time and the predicted point provided by the electric vehicle logarithm calculation section 51, And the number of years.

Specifically, the solution verifying unit 54 calculates a matrix of the predicted charge amount distribution matrix of the prediction point obtained by multiplying the lower triangular matrix of the conversion matrix by the charge amount distribution matrix of the reference point according to Equation (5) The values can be summed.

More specifically, the solution verifying unit 54 obtains a driving predictive electric power (hereinafter, referred to as &quot; driving predictive electric power &quot;) obtained by adding the values of the matrix elements of the predicted charge amount distribution matrix at the prediction time obtained by multiplying the lower triangular matrix of the conversion matrix by the charge- It is possible to determine whether or not convergence is based on whether the difference between the logarithm and the predicted point of time is smaller than a predetermined error.

The conversion matrix and the charge amount distribution matrix according to the solution proved by the solution verifying unit 54 that the number of electric vehicles to be operated is almost the same among the driving predicted electric vehicle number and the operation time at the prediction point is provided to the charge demand calculation unit 55.

The charging demand calculation unit 55 calculates the number of electric vehicles whose charging amount has increased at the prediction time with respect to the reference time as the number of charging predicted electric vehicles based on the values of the matrix elements of the charging amount distribution matrix and the conversion matrix, The charge demand at the predicted time is calculated based on the predicted number of electric vehicles and the capacity of the charger.

Specifically, the charge demand calculation unit 55 sums the values of the matrix elements of the predicted charge amount distribution matrix at the prediction time obtained by multiplying the upper triangular matrix of the transformation matrix by the charge amount distribution matrix at the reference time, according to Equation (6) The number of electric cars can be calculated.

It is possible to obtain future charging demand by electric vehicles by repeating such demand forecasting for a plurality of time zones in which traffic volume and running vehicle ratio data are provided.

Even if the performance of the electric cars and the performance of the charger are improved, if the conversion matrix is optimized by reflecting the values of the conversion discharge time unit, the battery efficiency, the rated motor capacity and the rated capacity of the charger, have.

It is to be understood that both the foregoing general description and the following detailed description of the present invention are exemplary and explanatory and are intended to provide further explanation of the invention as claimed. It will be understood that variations and specific embodiments which may occur to those skilled in the art are included within the scope of the present invention.

Further, the apparatus according to the present invention can be implemented as a computer-readable code on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the recording medium include ROM, RAM, optical disk, magnetic tape, floppy disk, hard disk, nonvolatile memory and the like. The computer-readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner.

50 Electric Vehicle Charging Demand Forecasting System
51 Electric Vehicle Algebra Calculator
52 transformation matrix optimization section
53 matrix element operating section
However,
55 Charging Demand Calculator

Claims (20)

As a method for predicting charging demand of an electric vehicle using a computer,
The computer comprising:
(a) determining the total number of electric vehicles and the number of electric vehicles in operation at the reference time and the forecasting time;
(b) a charge amount distribution matrix expressing a charge amount distribution of all the electric vehicles at the reference time point as electric number logarithmic values for each of the plurality of converted discharge time units, wherein the charge amount distribution of all the electric vehicles at the predicted time point is divided into a plurality of converted discharge time units Determining values of matrix elements of the charge amount distribution matrix and the transform matrix so as to be transformed into a predicted charge amount distribution matrix expressed by electric vehicle logarithmic values;
(c) determining whether the driving predictive electric-power logarithm of the prediction time obtained by multiplying the conversion matrix by the charge-amount distribution matrix at the reference point converges to the electric-vehicle logarithm of the driving of the prediction time point;
(d) repeating steps (b) and (c) if it is not determined that the number of vehicles predicted to converge to the number of electric vehicles during operation; And
(e) If it is determined that the driving forecast number of electric vehicles is converged to the number of electric vehicles during operation, the number of electric vehicles whose charging amount is increased at the predicted time point in comparison with the reference time is charged based on the values of the matrix elements of the charging amount distribution matrix and the conversion matrix As an estimated number of electric vehicles,
Wherein the converted discharge time unit is a value obtained by converting the charged amount into a time unit. The method according to claim 1, wherein the total number of electric vehicles and the number of electric vehicles during operation are determined in consideration of the number of electric vehicles compared to the total number of vehicles based on the traffic volume of all vehicles and the ratio of vehicles. 2. The electric vehicle according to claim 1, wherein each of the charge amount distribution matrix and the predicted charge amount distribution matrix is matrices obtained by arranging all the electric vehicles according to a charge amount unit converted into a plurality of time units with respect to each charge amount Demand forecasting method. 2. The method of claim 1, wherein the relationship between the charge amount distribution matrix, the predicted charge amount distribution matrix,

(T +? T) is a predicted charge distribution distribution matrix of the reference time t, B (t) is a transformation matrix, A (t + Is a 1 × n matrix composed of n matrix elements, the transformation matrix is an n × n matrix, n is a maximum charge amount of the electric vehicle Is determined as the nearest natural number which is larger or smaller than a value obtained by dividing the electric charge by the discharge discharge time unit. 5. The method of claim 4,

, And matrix elements (i, j) of each matrix position (i, j) of the transformation matrix B

Is a ratio of the electric vehicle consuming the charging amount of the (j-1) converted discharge time unit at the reference time t to the electric vehicle consuming the charging amount of the (i-1) converted discharge time unit at the prediction time t + And estimating the charging demand of the electric vehicle. 6. The method of claim 5, wherein step (b)

(T) and the matrix elements of the transformation matrix B (t) at a reference time point (t) satisfying a simultaneous equations consisting of a charge amount distribution matrix A Prediction method. [6] The method of claim 6, wherein the step (b) comprises the steps of: calculating values of some matrix elements of the transformation matrix by at least one of a conversion discharge time unit, a battery efficiency of an electric vehicle, a rated motor capacity, Further comprising the step of determining whether the charge demand is 0 or 1 on the basis of the estimated value. The method of claim 7, wherein some of the matrix elements of the transformation matrix

, Where DELTA t is a time interval between the prediction time and the reference time, and DELTA h is a unit of the converted discharge time. The method of claim 7, wherein some of the matrix elements of the transformation matrix

, Where e is the ratio of the available travel time added when charging the electric vehicle to the charge time, and [Delta] h is the converted discharge time unit. 5. The method according to claim 4, further comprising:

The predicted charge amount distribution matrix of the electric vehicles predicted to be on standby at the predicted time is multiplied by the diagonal matrix of the conversion matrix,
Wherein A _REST (t + Δt) is the distribution predicted charge of the electric vehicle is expected to be waiting in the prediction time matrix, D (t) is a diagonal matrix of the transformation matrix, the prediction charge distribution matrix A _REST (t + Δt ) Is a logarithm of electric vehicles predicted to be waiting at the predicted time point. 5. The method according to claim 4, further comprising:

The predicted charge amount distribution matrix of the electric vehicle, which is predicted to be operated at the predicted point in time, is obtained by multiplying the lower triangular matrix of the conversion matrix,
Wherein A _DRIVING (t + Δt) is the distribution predicted charge of the electric vehicle that is expected while operating on the prediction time matrix, the L (t) is a lower triangular matrix of the transformation matrix, the prediction charge distribution matrix A _DRIVING (t + Δt) is the logarithm of the electric vehicles estimated to be in operation at the predicted time point. 5. The method according to claim 4, further comprising:

The predicted charge amount distribution matrix of the electric vehicle, which is predicted to be charged at the predicted time point, is multiplied by the upper triangular matrix of the transformation matrix,
Wherein A _CHARGING (t + Δt) is the distribution predicted charge of the electric vehicle is predicted to being charged to the prediction point matrices, U (t) is a lower triangular matrix of the transformation matrix, the prediction charge distribution matrix A _CHARGING (t + DELTA t) is a logarithm of electric vehicles predicted to be charged at the predicted time point. The method of claim 1, further comprising repeating steps (a) through (e) with respect to a new reference time point and a new prediction time point. A computer program stored in a computer-readable recording medium for implementing the steps of a method for predicting charging demand of an electric vehicle according to any one of claims 1 to 13 in a computer. An electric vehicle algebra number calculation unit for determining the total number of electric vehicles and the number of electric vehicles in operation based on the externally given ratio of electric vehicles, the traffic volume, and the driving vehicle to the entire vehicle;
Wherein the charge distribution matrix expressing the charge amount distribution of all of the electric vehicles at the reference time point as a plurality of electric charge logarithmic values by the unit of the discharge time unit is a predicted charge amount expressed by the electric charge logarithm values of the electric vehicles A transformation matrix optimizer for defining the transformation matrix to be transformed into a distribution matrix;
A matrix element operation unit for determining the values of the charge distribution matrix and the matrix elements of the transformation matrix;
For the values of the matrix elements of the charge accumulation matrix and the charge accumulation matrix at the reference point determined by the matrix element arithmetic unit, multiplying the charge accumulation matrix of the reference point by the transformation matrix to obtain the driving predicted electric power logarithm of the predicted time point, A solution verifying unit for determining whether or not the number of electric vehicles to be driven converges among the estimated number of electric vehicles to be operated and the number of electric vehicles to be operated at the predicted time provided by the electric vehicle number calculating unit;
And a charge amount distribution matrix which is determined by the solution verifying section to converge the logarithm of the driving predictive electric potential and the electric potential difference between the driving of the prediction point and the charging electric potential difference matrix, And a charge demand calculation unit for calculating the number of chargeable electric vehicles as the number of charge predicted electric car numbers and calculating the charge demand at the predicted time based on the number of charge predicted electric vehicles at the predicted time and the charger capacity. 16. The electric vehicle according to claim 15, wherein each of the charge amount distribution matrix and the predicted charge amount distribution matrix is matrices obtained by arranging all the electric vehicles according to a charge amount unit converted into a plurality of time units with respect to each charge amount, Demand forecasting system. 16. The system of claim 15, wherein the conversion matrix optimizer is configured to calculate values of some matrix elements of the transformation matrix based on at least any one of a conversion discharge time unit, a battery efficiency of an electric car, a rated motor capacity, a rated charger capacity, So as to be 0 or 1, respectively. 16. The apparatus of claim 15,
The sum of the matrix elements of the charge amount distribution matrix is equal to the total number of the electric motors of the reference time point and the sum of the values of the matrix elements for each column of the conversion matrix is equal to 1, And to determine values of the charge distribution matrix and the matrix elements of the transformation matrix, respectively. 19. The system of claim 18, wherein the matrix element operating unit operates to apply the Monte Carlo method to solve the solution of the simultaneous equations. 16. The apparatus of claim 15, wherein the charge demand calculator calculates a sum of the matrix elements of the predicted charge amount distribution matrix at the prediction time obtained by multiplying the upper triangular matrix of the transformation matrix by the charge amount distribution matrix of the reference point, And calculates the number of charging predicted electric vehicles, which is the number of electric vehicles estimated to be charged.

Images