JP3350228B2 - Railway load prediction device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electric railway load predicting apparatus which is used as a power load managing apparatus for a railway substation and which predicts the electric power consumption of a substation after T minutes.

[0002]

2. Description of the Related Art In conventional railways, the load of a substation exceeds the contracted power by an operator, judging from the number of trains on a feeder section, for the purpose of efficient use of electric power for train operation and reduction of electric charges. When it is predicted that the train operation is notched, an instruction is issued to limit the train operation notch, thereby reducing the power for train operation.

[0003] However, when the operator predicts the load power of the substation from only the number of trains on the feeder section in this way, there is a problem in terms of prediction accuracy, the burden on the operator is large, and the prediction is incorrect. If the power load exceeds the contracted power with the power company, the surplus contracted power will be penalized.

[0004] Therefore, there has been proposed a railway load predicting apparatus for automatically predicting a railway load performed by such an operator. This uses a power consumption prediction method based on vehicle kilometers (the sum of the traveling distances of each vehicle that has traveled in a feeder section within a certain time) that has a strong correlation with the power consumption during train travel.

[0005]

However, the following problem arises in the power consumption prediction method in that "vehicle kilometers" is used as a parameter for power consumption prediction. That is, the value of the vehicle kilometer includes both the distance that the train travels while outputting the traction force and the distance that the train travels without outputting the traction force. In the former, the power used by the train to travel is large, but in the latter it is small. Therefore, even under the same vehicle kilometers, the traction force output by the train is different between when traveling on a track condition with a large running resistance, for example, traveling on an uphill, and when traveling on a track condition with a small traveling resistance, for example, on a downhill. The power will be different. Therefore, this prediction method has a problem in prediction accuracy.

[0006] Therefore, an object of the present invention is to perform more accurate power consumption prediction, and based on the prediction result, to appropriately limit and cancel the train operation notch and maintain the power consumption below the contract power. An object of the present invention is to provide a simple railway load prediction device.

[0007]

In order to achieve the above-mentioned object, the invention according to claim 1 is based on the present invention, wherein the present train position and the planned speed pattern suitable for each train type train prepared in advance are provided for each train. , The predicted speed pattern of the target feeding section for the next T minutes is determined, the power running distance is determined from the speed change and the running resistance, and the sum of the power running travel distances for all vehicles, ie, the power running kilometer, is calculated. From the train position T minutes ago, the current train position, and the actual train speed of each point in this section from the train speed T minutes before the present to the present time. A powering / km actual value calculating means for calculating a powering / running distance obtained by calculating a pattern, calculating a distance of the powering / running based on the speed change and the running resistance, and calculating a total of the powering / running distance of all the vehicles. , Based on the use the actual power value and the calculated power running kilometers actual value at this power running kilometer actual value calculating means,
The power running km-power characteristic statistical processing means for making a database of the power running km-power characteristic for a certain period of time, and the power running km-power characteristic database obtained by the power running km-power characteristic statistical processing means, and the normal An electric railway load prediction device comprising: a predicted powering kilometer predicted by the hourly powering kilometer predicted value calculating means; and a predicted power consumption value calculating means for predicting power consumption for the next T minutes.

In order to achieve the above object, the invention according to claim 2 is characterized in that in claim 1, the predicted power consumption value calculated by the predicted power consumption value calculation means is compared with the contract power with a power company. This is an electric railway load prediction device to which a notch restriction command determination unit that outputs a train notch restriction command and a notch restriction release command is added.

In order to achieve the above object, the invention according to claim 3 is based on claim 1 in which the notch is used for an abnormal time when the train is running at a low speed due to the notch restriction in the target feeding section. For the next T minutes when the restriction is lifted,
By performing a running simulation using the current train position and running speed as initial values, a predicted speed pattern of a target feeding section for the next T minutes of each train at the time of releasing the notch restriction is obtained, and a powering km is calculated from the predicted speed pattern. This is an electric railway load prediction device to which an extraordinary power running kilometer prediction value calculation means for predicting is added.

In order to achieve the above object, the invention according to claim 4 is characterized in that, in claim 3, the predicted power consumption value calculated by the predicted power consumption value calculation means is compared with the contract power with the power company. This is an electric railway load prediction device to which a notch restriction command determination unit that outputs a train notch restriction command and a notch restriction release command is added.

In order to achieve the above object, a fifth aspect of the present invention provides a power consumption predicted value calculating means according to any one of the first to fourth aspects, wherein the auxiliary power consumption predicted value is added. It is a railway load prediction device.

[0012]

According to the invention corresponding to claim 1, unlike the prediction method of the railway load prediction device which has been proposed in the past, a power running kilometer is used instead of a vehicle kilometer as a parameter of power consumption prediction. A database of the characteristics of the kilometer and the actual power consumption is used, and by referring to this database, the power consumption for the next T minutes is predicted from the powering / km prediction value. The load power can be predicted with higher accuracy, so that the operator can accurately perform the train operation notch limitation and the release of the limit based on the result of the load power prediction value, and reduce the power consumption to the contract power or less. Can be maintained.

[0013] According to the invention corresponding to claim 2, according to the invention corresponding to claim 1, the predicted power consumption value and the contract power with the power company are compared, and the train operation notch restriction and restriction release commands are output. Since the function is added, the burden on the operator is reduced as compared with the invention according to claim 1, and the train operation notch control is performed more accurately and accurately.

According to a third aspect of the present invention, there is provided the invention according to the first aspect, wherein the notch restriction is canceled for an abnormal time when the train is running at a low speed due to the notch restriction in the target power section. Since the power running km prediction function for predicting the power running km in the case of the addition is added, the power prediction at the time of the abnormality becomes more accurate, and the timing of releasing the notch restriction becomes more accurate.

According to the invention corresponding to claim 4, in the invention corresponding to claim 3, the predicted power consumption value and the contract power with the power company are compared, and the train operation notch restriction and restriction release commands are output. Since the function is added, the burden on the operator is reduced as compared with the invention according to claim 3, and the train operation notch control is performed more accurately and accurately.
According to the invention corresponding to claim 5, the accuracy of predicting the power consumption is further improved as compared with the invention corresponding to claims 1 to 4.

The use of the railway load predicting apparatus according to the first to fifth aspects of the present invention greatly reduces the contracted power surcharge which is taken as a penalty when the used power exceeds the contracted power with the power company. It has the effect of reducing.

[0017]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a functional block diagram showing a first embodiment of the present invention. In this embodiment, a normal running power / km prediction value calculating means 1 and an abnormal running power / km prediction value calculating means 2 are shown.
Powering kilometer actual value calculating means 3, powering kilometer-power characteristic statistical processing means 4, used power predicted value calculating means 5, and notch limit command determining means 6.

The normal-time power running kilometer predicted value calculating means 1 calculates a current train position and a speed pattern (hereinafter referred to as a planned speed pattern) suitable for each train type prepared in advance for each train. Next, the predicted speed pattern of the target feeding section for the next T minutes is determined, the power running distance is determined from the speed change and the running resistance, and the sum of the power running distances for all vehicles (hereinafter referred to as power running distance) is calculated. (Kilometers).

The abnormal-time power running kilometer predicted value calculation means 2 is used for the next T minutes in the case where the notch restriction is released for an abnormal time when the train is running at a low speed under the notch restriction in the target power section. By performing a running simulation with the current train position and running speed as initial values, a predicted speed pattern of a target feeding section for the next T minutes of each train at the time of releasing the notch restriction is obtained, and from the predicted speed pattern, The powering kilometer is predicted in the same manner as the hourly powering kilometer predicted value calculation means 1.

The powering-km actual value calculating means 3 calculates a speed pattern (T minutes before the present time to the present time) for each train from the train position T minutes ago, the current train position, and the actual train speed at each point in this section. Hereinafter, this is referred to as an actual speed pattern), the distance of the power running is calculated from the speed change and the running resistance, and the sum of the power running distance is calculated for all the vehicles.

The powering kilometer-power characteristic statistical processing means 4 performs a powering kilometer for a certain period of time based on the actual power consumption value and the powering kilometer actual value calculated by the powering kilometer actual value calculating means.
Create a database of power characteristics.

The power consumption prediction value calculating means 5 calculates the power running kilo-
Using the powering kilometer-power characteristic database obtained by the power characteristic statistical processing means 4, the power consumption for the next T minutes is predicted from the predicted powering kilometers predicted by the powering kilometer predicted value calculation means 1 and 2.

The notch limit command judging means 6 compares the predicted power value calculated by the predicted power value calculation means 5 with the contract power with the power company, and outputs the output of the train notch restriction command and the notch restriction release command. Do.

The operation of the first embodiment of the railway load predicting apparatus having the above-described configuration will be described below in order. FIG. 2 is a flowchart for explaining this operation.

The flow of the overall operation of the present embodiment is simply represented by the following S1 to S6. These series of operations are started at a certain time interval Δt, and the power consumption is estimated for the next T minutes. I do.

S1: With respect to the target feeding section, the actual running power km value in the past T minutes is calculated, the actual power consumption in the past T minutes is read, and the power running km-power characteristic database is updated.

S2: It is determined whether the notch is currently being restricted. If the notch restriction is not issued, the process of S4 is performed as normal power consumption prediction. If the notch restriction is being performed, the vehicle is running under the notch restriction due to an accident or the like, and the processing of S5 is performed as abnormal power consumption prediction. Perform processing.

S3: It is determined whether or not the target feeding section is in a traffic jam. The determination as to whether or not there is traffic congestion is made when the speed of the train on the line between the stations is close to zero. If the traffic congestion occurs in S3, a notch restriction command is output, and the processes in S4 and S6 are not performed.

S4: Predict normal power consumption. S5: Predict power consumption at the time of abnormality. S6: Compare the predicted value of the used power in S4 or S5 with the contract power with the power company, determine whether to output the notch restriction command or the notch restriction release command, and output the notch restriction command as in S7. Or a notch limit release command is output as in S8.

The processing of each of S1, S3, S4, S5, and S6 will be specifically described below. [Powering kilometer-power characteristic database statistical processing in S1] This function will be described with reference to the flowchart in FIG.

A1: A train that has traveled in the target power section is extracted from the train position T minutes before each train and the current position.
This is to extract a train if the power travel section is included at all in the travel of the train for the past T minutes. For example, FIG.
When the trains A, B, C, and D run as in the above, the trains A, B, and C are extracted.

A2: For each target train, the actual speed pattern of the target T section during the past T minutes is obtained. This actual speed pattern gives the train speed to the position, and stores the passing time of the track circuit boundary point for the past T minutes, and calculates the average running speed of each track circuit in the target feeding section from this, Create using this.

A3: Based on the actual speed pattern of each train, the actual running distance of each train is obtained. The powering travel distance is determined, for example, as follows. The target speed pattern in FIG. 8 is finely divided with respect to the position, and it is determined from the speed change and the running resistance in each divided section whether or not the train has run in power, and these are summed.

Then, the tractive force output by the train is obtained by the equation (1). Fp = F + R (1) where Fp is the traction force output by the train on the average of the divided sections.
[N] F: Force in the traveling direction applied to the train on the average of the divided sections [N] F = ΔV × W (ΔV is the speed change at the division point, W is the train weight) R: Running resistance force on the average of the divided sections [N] Here, if the traction force F output by the train is positive, it is determined that the vehicle has run power, and the sum of the distances of the divided sections in which the train has run is defined as the running distance for each train.

A4: The powering travel distance obtained in a3 is multiplied by the number of vehicles of each train, and the total is calculated for all vehicles, and this is set as the actual powering kilometer value. b1: Read the actual value of substation use power for the past T minutes.

C1: The powering kilometer-power characteristic database is updated from the actual powering kilometer value of a4 and the actual power consumption value of c1. [Judgment of traffic congestion in S3] When the train is congested due to an accident or the like and running at a low speed, when the accident is resolved, the train accelerates all at once, and the power consumption increases suddenly and the contract power increases. It may be over.

In order to prevent this, when the train runs at low speed in the middle of the station, it is determined whether or not the target power section is congested in an accident, and a notch limit command is output. The power consumption when the restriction is released is predicted, and the timing of the notch restriction release command is measured.

On the other hand, when there is no traffic congestion in the accident, the normal-time power consumption is predicted in S4. The judgment of whether or not there is traffic congestion is smaller than the threshold that the train speed is low,
When the train position is in the middle of the station, it is assumed that the traffic is congested in the accident.

[Calculation of predicted value of power consumption during normal use in S4] The power consumption prediction in the present invention predicts the speed pattern of each train from now on, and predicts the power running km based on this. It is based on the prediction of electric power by pulling.

The normal use power estimation value calculation function performs a normal power estimation. The normal state is a case where it is not an “abnormal state” described later. In normal times, trains have speed patterns that match the schedule (planned speed patterns).
, And the power consumption for the next T minutes is predicted.

In addition, the prediction is performed including the case where the temporary speed limit section is set between the stations during the normal time. The operation of this function will be described with reference to the flowchart of FIG.

E1: Selection of a train that is likely to travel in the feeder section in the next T minutes. The criteria of the train to be selected are based on the following conditions.・ A train whose current train position is in the target feeding section ・ A current train position is in the upstream section of the target feeding section,
A train e2 that is within a distance of reaching the target feeding section in the next T minutes e2: It is determined whether or not the temporary speed control section is set in the target feeding section.

E3: Predict the speed pattern for the next T minutes when no temporary speed limit section is set. For this speed pattern, a planned speed pattern, that is, a speed pattern when the vehicle travels at the arrival time between stations substantially on a train schedule is prepared for each train type in advance, and is used for the next T minutes of each train. Predict speed patterns. The planned speed pattern gives the train speed to the position.

For example, as shown in FIG. 9, a point where the current train position matches the position of the planned speed pattern is set as time on the planned speed pattern = current time, and the planned speed pattern is applied as a speed pattern for the next T minutes.

E4: Predict the speed pattern for the next T minutes when there is a temporary speed limit section. In the temporary speed limit section, a running simulation is performed only for a section that runs below the speed limit and a section that accelerates through the section and reaches the planned speed pattern. For the other sections, the planned speed pattern is applied, and the next Predict speed patterns.

For example, as shown in FIG. 10, when the temporary speed limit is set between P1 and P2, the running simulation is started with the train speed and position of the planned speed pattern at P1 as initial values, and the speed limit is set. Regarding decelerating running, running at a constant speed below the speed limit, and then accelerating from the point P2 that has left the speed limit section to the point P3 where the vehicle returns to the planned speed pattern, from the running resistance determined by the traction force characteristics of the vehicle and the track conditions, Calculate train speed and position. The specific flow of the traveling simulation is the same as that of the traveling simulation performed by the function for calculating a predicted power consumption value at abnormal time described later.

E5: The running distance of each train is calculated based on the speed pattern of each train for the next T minutes. This is a
It is calculated in the same manner as in 3.

E6: The powering travel distance obtained in e5 is multiplied by the number of vehicles of each train, and the total is calculated for all vehicles, and this is set as a powering kilometer predicted value. f1: From the predicted powering kilometer, a predicted power consumption value for the next T minutes is obtained based on the powering kilometer-power characteristic database.

[Calculation of Predicted Power Consumption for Abnormal Operation in S5] The function for calculating the predicted power consumption for abnormal use performs prediction of power consumption for an abnormal situation.

The abnormal state is a state in which the train is running at a low speed in response to the notch restriction command and the timing of the release command is applied. Therefore, in this function, the power consumption when the current notch restriction is released is predicted. The operation of this function will be described with reference to the flowchart of FIG.

S51: A train that is likely to travel in the feeder section during the next T minutes is selected. This is performed in the same manner as e1 in the case of the normal use power calculation. S50: Predict the speed pattern of each train for the next T minutes via S52 to S59.

In S2, when the notch restriction is released, each train is assumed to run so as to return to the planned speed pattern from the current train speed, and a running simulation S50 for the next T minutes is performed on each train. For example,
As shown in FIG. 11, it is assumed that the vehicle is accelerated toward the planned speed pattern using the current train position and the train speed as initial values, and travels at the planned speed pattern from a point P1 where the planned speed pattern is restored.

The running simulation 50 gives the current train speed and train position as initial values for each train,
The speed and the position are calculated based on the vehicle performance and the running resistance at every certain interval, and a speed pattern for T minutes is created.

S60: The power running distance of each train is obtained based on the speed pattern of each train for the next T minutes. this is,
It is calculated in the same manner as in a3. S61: The running distance calculated in S60 is multiplied by the number of vehicles in each train, and the total is calculated for all the trains. Then, a predicted running value is calculated in S62.

S63: From the powering kilometer predicted value calculated in S62, a predicted power consumption value for the next T minutes is obtained based on the powering kilometer-power characteristic database. [Notch Limit On / Off Determination in S6] This function will be described with reference to FIG.

S65: Here, it is determined whether or not the notch is being limited (notch restriction command ON / OFF). S66: If notch restriction is not being performed, it is determined whether or not the predicted power consumption value obtained in S4 of FIG. 2 exceeds the contract power with the power company.

S68: If the predicted value of the used power exceeds the contracted power with the power company in S66, a notch restriction command is output. S67: The power threshold value at the time of releasing the notch restriction is set to be lower than the contract power with the power company by ΔP, and the set value is set as shown in FIG.
It is determined whether or not it is larger than the predicted value of the used power obtained in S4. In this case, the set value is set to be lower than the contract power with the power company by ΔP in order to prevent the contract power from exceeding the contract power again in the near future. S69: S
If the predicted power consumption value is larger than the set value in 67, a notch restriction release command is output. In S66, if the predicted value of the used power is smaller than the contract power with the power company, a notch restriction release command is output.

FIG. 12 is a block diagram showing a second embodiment of the present invention, which is configured by adding an auxiliary motor use electric power predicted value calculating means 8 to the embodiment of FIG. This calculation means 8
Is used to calculate a predicted value of electric power used by an auxiliary motor for air conditioning and lighting of a train running in a feeder section. The operation of the second embodiment thus configured will be described. The electric power used by the auxiliary motor does not depend on the running state of the train, that is, whether the train is running, coasting or decelerating, but depends on the number of vehicles that have traveled in the target power section, the running time, and the performance of the auxiliary motor. The power used by the air conditioner in the vehicle also depends on the seasonal factor and the time factor. Therefore, the power consumption of the auxiliary motor is model-calculated and predicted from the number of vehicles traveling in the feeding section of the target train, the running time, the auxiliary motor performance parameters, and the parameters representing the seasonal suitable element and the time appropriate element.

The predicted value of the used electric power of the auxiliary motor calculated by this additional function is added to the predicted value of the used electric power in the powering kilo-power characteristic database of the first embodiment, and this is used as the predicted value of the whole used electric power.

However, at the time of the statistical processing of the powering kilometer-power characteristic database at this time, the actual power value includes both the power used by the main motor and the power used by the auxiliary motor for running the train. The second powering killer power characteristic statistical processing means 4 uses a value obtained by subtracting the auxiliary motor use power actual value from the actual power value.

According to the second embodiment described here, the first
In the embodiment, an auxiliary motor use electric power predicted value calculating means 8 for calculating an auxiliary electric motor use electric power predicted value for air conditioning and lighting of a train which fluctuates depending on a season and a time zone is added. This has the effect of increasing the prediction accuracy of the power consumption.

The present invention is not limited to the above-described embodiments, but can be modified as follows. That is, as an invention corresponding to claim 1, FIG.
In the embodiment of the present invention, the normal power running kilometer predicted value calculating means 1
It is also possible to adopt a configuration in which only the powering kilometer actual value calculating means 3, the powering killer power characteristic statistical processing means 4, and the used power predicted value calculating means 5 are combined. Even in this case, a power running kilometer is used as a parameter of the power usage prediction, and the characteristics of the power running kilometer and the actual power usage are stored in a database. Since the power consumption is predicted, the load power for train operation at the railway substation can be predicted with higher accuracy. Notch restriction and restriction release can be accurately performed, and the power consumption can be maintained at or below the contract power.

Further, as an invention corresponding to claim 2, in the embodiment shown in FIG. 1 or FIG. 12, the normal running power / km predicted value calculating means 1, the running power / km actual value calculating means 3, and the running power / kilow power characteristic statistical processing. It is also possible to adopt a configuration in which only the means 4, the predicted power value calculation means 5, and the notch restriction command determination means 6 are combined. In this case, the invention according to claim 1 has a function of comparing the predicted power consumption value and the contract power with the power company and outputting a train operation notch restriction and restriction release command. As compared with the invention corresponding to (1), the burden on the operator is reduced, and the train operation notch control is performed more accurately and accurately.

Further, as an invention corresponding to claim 3,
Normal power running kilometer predicted value calculating means 1, abnormal power running kilometer predicted value calculating means 2, power running kilometer actual value calculating means 3, power running killer power characteristic statistical processing means 4, and used power predicted value calculating means 5 only May be combined. In this case, in the case of an abnormality which predicts a power running kilometer when the notch restriction is released in a case where the notch restriction is canceled in a case where the train is running at a low speed due to the notch restriction in the power supply section according to the first aspect of the present invention. The addition of the power running kilometer prediction function makes the power prediction at the time of abnormality more accurate, and the timing of releasing the notch restriction becomes more accurate.

[0065]

According to the railway load prediction apparatus of the present invention, the train operation load power applied to the railway substation is automatically predicted with high accuracy, and the train operation notch restriction and restriction are performed based on the predicted value. Release can be performed accurately, and the used power can be maintained at or below the contracted power.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a functional configuration of a first embodiment of the present invention.

FIG. 2 is a flowchart for explaining the operation of FIG. 1;

FIG. 3 is a flowchart showing in detail a powering killer power characteristic statistical processing function of FIG. 2;

FIG. 4 is a flowchart showing details of a normal use power prediction function of FIG. 2;

FIG. 5 is a flowchart showing in detail an abnormal use power prediction function of FIG. 2;

FIG. 6 is a flowchart showing the notch restriction command determining function of FIG. 2 in detail;

FIG. 7 is a view for explaining a method of extracting a target train in FIG. 3;

FIG. 8 is a view for explaining how to obtain a powering traveling distance from the speed pattern of FIG. 3;

FIG. 9 is a view for explaining a method of creating a predicted speed pattern using the planned speed pattern of FIG. 4;

FIG. 10 is a view for explaining a method of creating a predicted speed pattern when setting a temporary speed limit section in FIG. 4;

FIG. 11 is a diagram for explaining a method of creating the predicted speed pattern at the time of abnormality in FIG. 5 by running simulation.

FIG. 12 is a block diagram showing a functional configuration of a second embodiment of the present invention.

[Explanation of symbols]

1: Normal power running kilometer predicted value calculation means, 2: Abnormal power running kilometer predicted value calculation means, 3: Power running kilometer actual value calculation means, 4 ...
Powering killer power characteristic statistical processing means, 5 ... used power predicted value calculating means, 6 ... notch limit command determining means, 8 ... auxiliary motor used power predicted value calculating means.

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Tadashi Okada 2-4-2, Shibata, Kita-ku, Osaka-shi, Osaka Inside West Japan Railway Company (72) Inventor Hiroyuki Hirayama 2-chome, Shibata, Kita-ku, Osaka, Osaka No. 4-24 Inside West Japan Railway Company (72) Inventor Tetsuro Sakamoto 2-4-2 Shibata Kita-ku, Osaka City, Osaka Prefecture Inside West Japan Railway Company (72) Inventor Mieko Hayashi 1 Toshiba-cho, Fuchu-shi, Tokyo Address Toshiba Corporation Fuchu Plant (56) References JP-A-5-252563 (JP, A) JP-A-7-304353 (JP, A) JP-A-5-176457 (JP, A) JP-A-5 -16808 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) B60M 3/00 B61L 27/00 H02J 3/00

Claims (5)

(57) [Claims] 1. For each train, a predicted speed pattern of a target power supply section for the next T minutes is obtained based on a current train position and a planned speed pattern prepared in advance for each train type schedule. From the speed change and the running resistance, the power running distance is determined, and the total power running travel distance obtained by summing the calculated power running distances for all the vehicles, that is, the normal power running km predicted value calculating means for predicting the power running km, From the train position T minutes ago, the current train position, and the actual train speed at each point in this section, the actual speed pattern from T minutes ago to the present is obtained, and from this speed change and running resistance, power running is performed. Powering / km actual value calculating means for calculating the powering / km distance, which is the sum of the powering / traveling distances obtained by summing the distances for all vehicles, and the powering / km actual value calculated by the powering / km actual value calculating means. Powering kilometer-power characteristic statistical processing means for making a database of powering kilometer-power characteristics for a certain period of time based on the value and the actual power consumption value; powering kilometer-power obtained by the powering kilometer-power characteristic statistical processing means An electric train load estimating device comprising: a predicted powering km calculated by the normal powering kilometer predicted value calculating means using a characteristic database; 2. For each train, a predicted speed pattern of a target power supply section for the next T minutes is calculated based on a current train position and a planned speed pattern prepared for each train type prepared in advance. From the speed change and the running resistance, the power running distance is determined, and the total power running travel distance obtained by summing the calculated power running distances for all the vehicles, that is, the normal power running km predicted value calculating means for predicting the power running km, From the train position T minutes ago, the current train position, and the actual train speed at each point in this section, the actual speed pattern from T minutes ago to the present is obtained, and from this speed change and running resistance, power running is performed. Powering / km actual value calculating means for calculating the powering / km distance, which is the sum of the powering / running distances calculated for all the vehicles, that is, the powering / km actual value calculated by the powering / km actual value calculating means. Powering kilometer-power characteristic statistical processing means for creating a database of powering kilometer-power characteristics for a certain period of time based on the value and the actual power consumption value; powering kilometer-power obtained by the powering kilometer-power characteristic statistical processing means A power consumption predicted value calculating means for predicting power consumption for the next T minutes from the predicted power running km predicted by the normal power running km predicted value calculating means using a characteristic database; A notch limit command determining unit that compares the calculated predicted power consumption value with the contract power with the power company and outputs a train notch limit command and a notch limit release command. 3. For each train, a predicted speed pattern of a target power supply section for the next T minutes is obtained based on a current train position and a planned speed pattern prepared for each train type prepared in advance. From the speed change and the running resistance, a power running distance is obtained, and a total of the power running travel distance obtained by summing the distances for all vehicles, that is, a normal power running km prediction value calculating means for predicting the power running km, and a target power supply section. By performing a running simulation using the current train position and running speed as initial values for the next T minutes when the notch limit is released, targeting an abnormal time when the train is running at low speed due to the notch limit. , Obtains a predicted speed pattern of the target feeding section for the next T minutes of each train when the notch restriction is released, and predicts a power running km from the predicted speed pattern. From the measured value calculating means, for each train, the actual speed pattern from T minutes before to the present time is obtained from the train position before T minutes, the current train position, and the actual train speed at each point in this section. From the speed change and the running resistance, the power running distance is obtained, and the sum of the power running travel distances of all the vehicles is calculated, that is, the power running km actual value calculating means for calculating the power running km, and the power running km actual value calculating means. Powering kilometer-power characteristic statistical processing means for making a database of powering kilometer-power characteristics for a certain period of time based on the calculated powering kilometer actual value and used power actual value, and powering kilometer-power characteristic statistical processing means. Using the database of the powering km-power characteristics obtained, the power consumption for the next T minutes is predicted from the predicted powering km predicted by the normal or abnormal powering km predicted value calculating means. And a power consumption prediction value calculating means for measuring the power consumption. 4. For each train, a predicted speed pattern of a target power supply section for the next T minutes is obtained based on a current train position and a planned speed pattern prepared for each train type for each train type prepared in advance. From the speed change and the running resistance, a power running distance is obtained, and a total of the power running travel distance obtained by summing the distances for all vehicles, that is, a normal power running km prediction value calculating means for predicting the power running km, and a target power supply section. By performing a running simulation using the current train position and running speed as initial values for the next T minutes when the notch limit is released, targeting an abnormal time when the train is running at low speed due to the notch limit. , Obtains a predicted speed pattern of the target feeding section for the next T minutes of each train when the notch restriction is released, and predicts a power running km from the predicted speed pattern. From the measured value calculating means, for each train, the actual speed pattern from T minutes before to the present time is obtained from the train position before T minutes, the current train position, and the actual train speed at each point in this section. From the speed change and the running resistance, the power running distance is obtained, and the sum of the power running travel distances of all the vehicles is calculated, that is, the power running km actual value calculating means for calculating the power running km, and the power running km actual value calculating means. Powering kilometer-power characteristic statistical processing means for making a database of powering kilometer-power characteristics for a certain period of time based on the calculated powering kilometer actual value and used power actual value, and powering kilometer-power characteristic statistical processing means. Using the database of the powering km-power characteristics obtained, the power consumption for the next T minutes is predicted from the predicted powering km predicted by the normal or abnormal powering km predicted value calculating means. Means for calculating a predicted power consumption value to be measured, and comparing the predicted power consumption value calculated by the predicted power consumption value calculation means with the contract power with the power company to output a train notch limit command and a notch limit release command. A notch limit command determining means; 5. An electric railway load predicting device, wherein an auxiliary electric motor use electric power predicted value is added to the used electric power predicted value calculation means according to any one of claims 1 to 4.