GB2529294A

GB2529294A - Automatic train control apparatus and automatic train control method

Info

Publication number: GB2529294A
Application number: GB1510155.3A
Authority: GB
Inventors: Kentaroh Maki; Tsutomu Miyauchi; Ikuo Niimi; Takahiro Nakajima; Tetsuya Shimomura; Takeshi Narita
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2014-07-10
Filing date: 2015-06-11
Publication date: 2016-02-17
Anticipated expiration: 2035-06-11
Also published as: JP6357372B2; JP2016019411A; DE102015212847B4; DE102015212847A8; GB2529294B; CN105270445A; CN105270445B; GB201510155D0; DE102015212847A1

Abstract

An automatic train control system 501 has a vehicle characteristics learning unit 505 that learns vehicle characteristics based on the train speed and a braking command and estimated using a unit 507. An adoption judgment unit 508 determined whether the characteristics should be adopted in a braking command, and an adoption unit 509 adopts the characteristics in a braking command only if judgment unit 508 determines that the characteristics should be adopted. Judgment unit 508 may adjust the measured deceleration and shift it on a time axis to correct for vehicle characteristics, and may determine whether to adopt characteristics according to how often the measured or commanded deceleration is zero or negative, corresponding to wheel slide or rapid release and reapplication of the brake. The system may avoid learning characteristics during wheel slide or other abnormal situations that may not be obviously abnormal or impossible, avoiding capturing inaccurate performance data.

Description

[Tifle of Invention] AUTOMATIC TRAIN CONTROL APPARATUS AND

AUTOMATIC TRAIN CONTROL METHOD

[Technica Field]

[0001] The present invention relates to an automatic train control apparatus and an automatic train control method. Particulady, this invention is suited for use in an automatic train control apparatus and automatic train control method for controfling travefling of trains by utilizing vehicle characteristics learned from previous travelling resufts.

[Background Art]

[0002] In recent years, introduction ot automatic train control (ATO) apparatuses has been promoted for the purpose of reducing burdens on crews and reducing labor costs on the background of, for example, highdensity train operation diagrams and enhanced introduction of platform doors. Of the ATO apparatuses, particularly train automatic stop control (TASC) apparatuses that stop trains by accurately locating vehicle door positions at plafforrn door positions have been being actively introduced for many train lines along with advancement of introduct!on of platform doors at existing train stations.

[0003] PU 1 discloses a technique relating to an ATO apparatus. Specifically speaking, PTL I discloses an automatic train control apparatus that: executes online processing of data acquired while a train is traveihng; automatically learns, during travefling of the train, control parameters to be used at the time of train travelling, train characteristics, and train line characteristics based on the online processed data and previously obtained data; and automatically drives the train by using the automat!cally-learned train characteristics and train line characteristics.

[0004] Furthermore, there is a disclosed technique that corrects an estimation result within a limited characteristic value when the train characteristics are estimated by using driving test results before commercial driving and the estimation result is a characteristic value which is actually impossible to happen, or when the estimation result is not identical to the limited characteristic value which may actually happen.

[0005] This PIL 1 describes that the train characteristics and the train line characteristics can be automaticafly carried onhne while a train is travefling; and the train can be automaticafly driven by using the automaticafly learned results.

Furthermore, when the estimation result which can never happen is obtained, it is possible to avoid using this inappropriate estimation result for control in advance by correcting the estimation result of the train characteristics.

[Citation List] [Patent Literature] [0006] [PTL 1]Patent No. 3940649

[Summary of Invention]

[Problems to be Solved by the Invention] [0007] The technique described in PTL I estimates the train characteristics by using the driving test results before commercial driving; however, when the driving test before commercial driving (or driving after commerc!al driving) includes disturbance events such as slipping or skidding and the train characteristics are estniated by using travelling results ncluding these disturbance events, the estimation result may not be judged to be obviously abnormal in some cases.

[0008] If the estimation result cannot be judged to be obviously abnormal, the inappropriate estimation result may sometimes be used for control. In this case, problems of degradat!on of tollowab!lity to make an actual vehicle speed properly follow a predetermined target speed and degradation of stop position accuracy and comfortable ride may occur.

[0009] The present invention was devised in consideration of the above-described circumstances and suggests an automatic train control apparatus capable of preventing degradation of foflowability and enhancing stop posftion accuracy and comfortable hde.

[Means for Solving the Problems] [0010] In order to solve the above-described problems, the present irwention provides an automatic train control apparatus equipped with a vehicle charactehstics earning unit that learns vehicle characteristics based on a speed of its own trail) and a braking command to control travefling of its own train, wherein the vehicle characteristics learning unit includes: a vehicle characteristics estimation unit that estimates the vehicle characteristics; a vehicle characteristics adoption possibihty judgment unit that judges whether the vehicle charactebstics should be adopted in a braking command or not; and a vehicle characteristics adoption unit that adopts the vehicle characteristics in a braking command to be newly calculated: and wherein the vehicle characteristics adoption unit adopts the vehicle characteristics, which are estimated by the vehicle characteristics estimation unit, in the braking command to be newly calculated only when the vehicle characteristics adoption possibility judgment unit determines that the vehicle characteristics can be adopted in the braking command [0011] Furthermore, in order to solve the above--described objects, the present invention provides an automatic train control method for an automatic train control apparatus equipped with a vehide characteristics learning unit that learns vehicle characteristics based on a speed of its own train and a braking command to control travelling of its own train; wherein the vehicle characteristics learning unit includes: a first step of estimating the vehicle characteristics; a second step of judging whether the vehcle characteristics should be adopted in a braking command or not; and a third step of reflecting the vehicle characteristics in a braking corn mand to be newly calculated; and wherein in the third step, the vehicle characteristics, which are esLimated in the first step, are adopted in the braking corn mand to be newly calculated only when it is determined in the second step that the vehicle characteristics can be adopted in the braking command.

[Advantageous Effects of Invention] [0012] According to the present invention, it is possible to prevent deterioration of followability and enhance stop position accuracy and comfortable ride.

[Brief Description of Drawings]

[0013] [Fig. 1] Fig. I is a functional configuration diagram of a train automatic stop control device; [Fig. 2] Fig. 2 is a relationship diagram of a vehicle speed, commanded deceleration and occurred deceleration, and elapsed time; [Fig. 3] Fig. 3 is a functional configuration diagram of a train automatic stop device equipped with a vehicle characteristics learning function; [Fig. 4A] Fig. 4A is a relationship diagram of the vehicle speed and the elapsed time when skidding and re-adhesion occur; [Fig. 4B] Fig. 4B is a relationship diagram of the commanded deceleration and the occurred deceleration, and the elapsed time; [Fig. 5] Fig. 5 is a functional configuration diagram of an automatic train control apparatus according to a first embodiment; [Fig. 6] Fig. 6 is an explanatory diagram of a residual deviation; [Fig. 7A] Fig. TA is a relationship diagram of the vehicle speed, the commanded deceleration and the occurred deceleration, and the elapsed time; [Fig. 7B] Fig. TB is a relationship diagram of the vehicle speed, the commanded deceleration and the occurred deceleration, and the elapsed time; [Fig. 8] Fig. 8 is a flowchart illustrating adjustment parameter update processing according to the first embodiment; [Fig. 9] Fig. 9 is a functional configuration diagram of an automatic train control apparatus according to a second embodiment; [Fig. IDA] Fig. IDA is a relationship diagram of the vehicle speed and the elapsed time when skidding occurs; [Fig. I OB] Fig. I OB is a diagram illustrating characteristics represented by an occurred deceleration waveform when skidding occurs; [Fig. 11] Fig. 11 is a diagram illustrating the deceleration waveform when a braking command is turned on and off repeatedly at short intervals; [Fig. 12] Fig. 12 is a flowchart illustrating adjustment parameter update processing according to the second embodiment; [Fig. 131 Fig. 13 is a functional configuration diagram of an automatic train control apparatus according to a third embodiment; and [Fig. 14] Fig. 14 is a flowchart illustrating adjustment parameter update processing according to the third embodiment.

[Mode for Carrying out the Inventioni

[0014] Eembodiments of the present invention will be explained below with reference to drawings. Incidentally, Fig. I to Fig. 4 are explanatory diagrams of the configuration and object of a train automatic stop control device (TASC device) included in a general automatic train control apparatus (ATO apparatus) and Fig. 5 to Fig. 14 are explanatory diagrams of the configuration of an automatic train control apparatus (particularly, a train automatic stop control device) according to the embodiments.

[0015] Furthermore, each unit illustrated in Fig. I to Fig. 14 is equipment configured by a processor, storage media, or programs, or any combination thereof. For example, the processor implements various functions by reading the programs stored in the storage media.

[0016] (1) Outlines of Train Automatic Stop Control Device Fig. I illustrates a functional structure of a train automatic stop control device (TASC device). The TASC device has a speed and position detection unit which detects a speed signal from a tacho-generator installed at a wheel shaft and detect positional information from a pickup which communicates with a beacon. Furthermore, the TASC device has a control command calculation unit which calculates a braking command based on the obtained speed signal and positional information and outputs the calculated braking command to a train management system.

[0017] The TASC device has two major functions as described above.

Specifically speaking, the TASC device has a speed and position detection function detecting the speed signal and the positional information, and a control command calculation function calculating the braking command. A control command calculation unit which has the control command calculation function of the above-mentioned functions is further composed of a planning unit equipped with a planning function and a follow-up unit equipped with a follow-up function.

[0018] The planning function is a function that calculates a target speed in light of a current vehicle position relative to a braking speed to reach a station stop position retained in advance. Furthermore, the follow-up function is a function that acquires a speed deviation between the target speed and a current vehicle speed and calculates a braking force to be output. The TASC device includes this calculated braking force in a braking command and outputs the braking command to a train management system.

[0019] Incidentally, types of the braking command include, for example, a brake notch command and a torque command. The train management system is a device for managing onboard information transmission; and after the braking command is input from the TASC device, the input braking command is output to a braking and driving control device. The braking and driving control device controls travelling of the relevant train based on the input braking command.

[0020] Incidentally, some difference usually occurs between time when the braking command is output from the TASC device and time when operation based on the braking command is actually adopted in the train. Furthermore, it is known that some difference occurs between the braking force included in the braking command from the TASC device and a braking force that actually acts on the train.

[0021] Fig. 2 illustrates the relationship between a vehicle speed, commanded deceleration and occurred deceleration, and elapsed time. The vehicle speed is represented by a dotted line, the commanded deceleration is represented by a solid line, and the occurred deceleration is represented by an alternate long and short dash line. Referring to Fig. 2, some difference occurs between time when a braking command, which orders deceleration, is output from the TASC device and time when the commanded deceleration included in this braking command is actuaRy adopLed in the train. This time difference is called dead time.

[0022] Furthermore, some difference also occurs between the commanded deceleration and the occurred deceleration actuaRy acting on the train. This dftference is called a deceleration deviation. As the dead time and the deceleration deviation increase, followabihty to follow the target speed degrades.

Therefore, when introducing a TASC device: operators need to adjust various parameters sufficiently and it is necessary to minimize adverse effects of the dead time and the deceleration deviation on the stop position accuracy.

[0023] Meanwhile, in recent years, development of a function that automaticafly learns the vehicle characteristics (the dead time and the deceleration deviation) based on travelling results, which are obtained when trains actually travel, is being promoted in order to reduce manhour for the adjustment work by the operators. This function is caRed a vehicle characteristics learning function.

[0024] Fig. 3 illustrates a functional structure of a TASC device equipped with a vehicle characteristics learning unit. The vehicle characteristics learning unit is a function that acquires a braking command and a current speed, estmates the dead time relating to deceleration and the deceleration deviation, and outputs adjustment parameters, including the vehicle characteristic value, to the control command calculation unit.

[0025] The control command calculation unit calculates a predicted position and predicted speed after the dead time with respect to the vehicle pos!tion and the speed and performs predictive control in consideration of any delay by the dead time. The control command calculation unit also corrects a commanded value of the braking force in consideration of the deceleration deviation.

[0026] Now, regarding estimation of the vehicle characteristics, values of the dead time and the deceleration deviation can be estimated properly, if a waveform of the commanded deceleraUon is almost similar to that of the occurred deceleration as illustrated in Fig. 2 and these waveforms can he overlaid with one another by paraflel translation in the time axis direction and in the deceleration axis direction.

[0027] However, if an attempt is made to estimate the vehicle characteristics based on the travefling results including a disturbance event such as slipping or skidding. the adjustment parameters used to control travelling of the train cannot be set properly. In this case, the followability worsens and the stop position accuracy and the comfortable ride degrade.

[0028] Fig. 4A illustrates the relationship between the vehcle speed and the &apsed time when skidding and re-adhesion occur. Furthermore, Fig. 4B illustrates the relationship between the commanded deceleration and the occurred deceleration, and the elapsed time when skidding and re-adhesion occur. The vehicle speed is represented by a dotted line, the commanded deceleration is represented by a solid line, and the occurred deceleration is represented by an alternate long and short dash line.

[0029] When skidding and re-adhesion occur while braking as illustrated in Fig. 4A and Fig. 4B, the vehicle speed becomes unstable due to repeated skidding and re-adhesion and the waveform of the occurred deceleration calculated from the vehicle speed shows wide swings although constant commanded deceleration is output. In this case, it is difficult to estimate the vehicle characteristics properly by comparing the commanded deceleration waveform with the occurred deceleration waveform.

[0030] (2) First Embodiment A first embodiment will be explained about an automatic train control apparatus (ATO apparatus) that calculates a residual deviation by comparing the commanded deceleration waveform with the occurred deceleration waveform; does not update adjustment parameter values used to control travelling of the train when a disturbance event occurs and the residual deviation becomes equal to or more than a specified threshold value; and updates the adjustment parameter values only when the residual deviation is less than the specified threshold value.

[0031] (2-1) Configuration of Automatic Train Control Apparatus Fig. 5 illustrates a functional structure of an automatic train control apparatus 501 according to the first embodiment. The automatic train control apparatus 501 controls travelling of the train by calculating the braking command and outputting the calculated braking command via a braking and driving control device 502 to an actuator 503.

[0032] The automatic train control apparatus 501 acquires the position and speed of its own train from a speed and position detection unit 504 and acquires a vehicle characteristic value from a vehicle characteristics earning unit 505.

Then, the automatic train control apparatus 501 has a control command calculation unit 506 which acquires the position and speed from the speed and position detection unit 504 and the vehicle characteristic value from the vehicle characteristics learning unit 505 and, thereby calculates the braking command.

[0033] Incidentally, examples of a method for having the speed and position detection unit 504 detect the speed of its own train include a detection method by multiplying a rotational speed of an axle by a circumferential length of a wheel by using a speed signal from a tacho-generator installed at the axle and a method of measuring a speed difference from the ground by using a Doppler radar.

[0034] Next, the vehicle characteristics learning unit 505 will be explained. The vehicle characteristics learning unit 505 is composed of a vehicle characterist!cs estimation unit 507, a vehicle characteristics adoption possibility judgment unit 508, and a vehicle characteristics adoption unit 509.

[0035] The vehicle characteristics estimation unit 507 acquires the position and speed of its own train from the speed and position detection unit 504 and a braking command of its own train from the control command calculation unit 506 and estimates the vehicle characteristics of its own train.

[0036] The vehicle characteristics adoption possihihty judgment unit 508 judges whether or not the vehicle characteristics estimated by the vehicle characteristics estimation unit 507 should be adopted in the adjustment parameters used to control travelhng of the relevant train, based on the vehicle characteristics estimated by the vehicle characteristics estimation unit 507 and the deceleration waveform used in the estimation process; and if it is determined to reflect the vehicle characteristics in the adjustment parameters, the vehicle characteristics adoption possibdity judgment unit 508 outputs the vehicle characteristics to the vehicle characteristics adoption unit 509.

[0037] The vehicle characteristics adoption unit 509: accumulates, and executes statistical processing on, vehicle characteristic values regarding which t has been determined by the vehicle characteristics adoption possibility judgment unit 508 that they can be adopted in the adjustment parameters; and then outputs the vehicle characteristic values to the control command calculation unit 506.

[0038] Next, the details of the vehicle characteristics estimation unit 507 will be explained. The vehicle characteristics estimaton unit 507 is composed of a travehing result data accumulation unit 510, a commanded deceleration waveform calculation unit 511 an occurred deceleration waveform calculation unit 512. and a waveform shift amount calculation unit 513.

[0039] The travelling result data accumulation unit 510 accumulates positions and speeds of its own tra!n from the speed and posit!on detection unit 504 and braking commands of its own train from the control command calculation unit 506 as data indicative of travelling results.

[0040] The commanded deceleration waveform calculation unit 511 calculates time-series data of deceleration commanded when braking (commanded deceleration waveform) based on the braking commands accumulated by the travelling result data accumulation unit 510. Furthermore, the occurred deceleration waveform calculation unit 512 calculates time-series data of deceleration occurred when braking (occurred deceleration waveform) based on the positions and speeds accumulated by the travelling result data accumulation unit 510.

[0041] Incidentally, in the process of calculating the deceleration waveform which has occurred when braking, it is possible to eliminate influence of grade resistance on the occurred deceleration and accurately calculate the deceleration which has actually occurred on the train by referring to train line gradient information, which is not shown in the drawing, using positional data of its own train.

[0042] The waveform shift amount calculation unit 513 acquires the commanded deceleration waveform calculated by the commanded deceleration waveform calculation unit 511 and the occurred deceleration waveform calculated by the occurred deceleration waveform calculation unit 512.

[0043] Then, the waveform shift amount calculation unit 513 calculates a waveform shift amount indicative of how much these waveforms are shifted (differences) in the time axis direction and the deceleration axis direction. This waveform shift amount is the vehicle characteristic value estimated by the vehicle characteristics estimation unit 507.

[0044] Next, the details of the vehicle characteristics adoption possibility judgment unit 506 will be explained. The vehicle characteristics adoption possibility judgment unit 508 is composed of a residual deviation calculation unit 514 and a residual deviation threshold judgment unit 515.

[0045] The residual deviation calculation unit 514 calculates a residual deviation by acquiring the commanded deceleration waveform, the occurred deceleration waveform, and the waveform shift amount calculated by the waveform shift amount calculation unit 513.

[00461 Fig. 6 is an explanatory diagram of the residual deviation. The residual deviation is a quantitative value that is a time-based average of absolute values of deceleration-direction deviations, which have remained between the commanded deceleration waveform and a waveform obtained after shifting the occurred deceleration waveform by the waveform shift amount (corrected occurred deceleration waveform), and represents a shape difference between the commanded deceleration waveform and the occurred deceleration waveform.

[00471 Fig. 7A illustrates the relationship between the vehicle speed, the commanded deceleration and the occurred deceleration, and the elapsed time when the residual deviation is small. The shape of the commanded deceleration waveform and the shape of the occurred deceleration waveform during speed reduction mostly correspond with each other. In this case, the vehicle characteristics can be estimated properly based on the waveform shift amount for overlaying these waveforms with one another.

[0048] Fig. 7B illustrates the relationship between the vehicle speed, the commanded deceleration and the occurred deceleration, and the elapsed time when the residual deviation is large. The shape of the commanded deceleration waveform is significantly different from that of the occurred deceleration waveform. In this case, the vehicle characteristics cannot be estimated properly based on the waveform shift amount for overlaying these waveforms with one another.

[0049] Incidentally, the commanded deceleration waveform illustrated in Fig. 7B is a waveform to which a step-shaped large notch change command is added manually during the speed reduction by the automatic train control apparatus 501.

Regarding the occurred deceleration waveform, it is indicated that elements of temporary delay and oscillations are formed in response to the commanded deceleration. Another example of a case where the residual deviation becomes large is a case where regeneration cancellation occurs during the speed reduction using an electric brake.

[0050] When the regeneration cancellation occurs, a burden ratio of an air brake which is more effective than the electric brake increases. So, even if a constant braking command is retained, that is, even if the commanded deceleration is constant, the occurred deceleration will increase in the middle of travelling. As a result, the shape difference between the commanded deceleration waveform and the occurred deceleration waveform occurs and the residual deviation becomes large.

[00511 The residual deviation calculated by the residual deviation calculation unit 514 is output to the residual deviation threshold judgment unit 515. The residual deviation threshold judgment unit 515 judges whether the input residual deviation is less than a specified residual deviation threshold value or not. If the residual deviation is less than the residual deviation threshold value, the residual deviation threshold judgment unit 515 outputs the waveform shift amount calculated by the waveform shift amount calculation unit 513 to the vehicle characteristics adoption unit 509.

[0052] Next, the details of the vehicle characteristics adoption unit 509 will be explained. The vehicle characteristics adoption unit 509 is composed of a vehicle characteristic data accumulation unit 516 and a vehicle characteristic statistical processing unit 517. The vehicle characteristic data accumulation unit 516 accumulates the waveform shift amount from the vehicle characteristics adoption possibility judgment unit 508 as vehicle characteristic values in a database.

[0053] The vehicle characteristic statistical processing unit 517 refers to a set of vehicle characteristic values accumulated in the vehicle characteristic data accumulation unit 516, decides a representative value from among the set of vehicle characteristic values by executing statistical processing, and updates the adjustment parameters retained by the control command calculation unit 506 with the representative value.

[0054] The control command calculation unft 506 calculates the braking command by using the updated adjustment parameters and outputs the calculated braking command via the braking and driving contro device 502 to the actuator 503, thereby controlling travelling of the train.

[0055] (2-2) Flowchart Fig. 8 illustrates a processing sequence of adjustment parameter update processing executed by the vehicle characteristics learning unit 505. Firstly. the vehicle characteristics estimation unit 507 acquires the position of its own train, which is detected by the speed and position detection unit 504, and judges whether its own train has reached a TASC control starting point (a specified distance before the relevant dest!nation) or not (SF1).

[0056] When the vehicle characteristics estimation unit 507 obtains a negative result in the judgment of step SF1 it continues to obtain the position of its train detected by the speed and position detection unit 504 at each constant period and repeats the judgment in step SP1.

[0057] On the other hand. when the vehicle characteristics estmation unit 507 obtains an affirmative result in the judgment of step SP1, it accumulates subsequent positions and speeds of its own train and braking commands as travehing result data in the traveWng result data accumulation unit 510 (SP2).

[0058] Then, the vehicle characteristics estimation unit 507 judges whether TASC control has terminated or not (SF3). When the vehicle characteristics estimation unit 507 obtains a negative result in the judgment of step SP3, it continues to accumulate the travelling result data in the travelling result data accumulation unit 510 at each constant period.

[0059] On the other hand, when the vehicle characteristics estimation unit 507 obtains an affirmative result in the judgment of step SF3, it has the commanded deceleration waveform calculation unit 511 calculate the commanded deceleration waveform and has the occurred deceleration waveform calculation unft 512 calculate the occurred deceleration waveform based on the travefling resuft data accumLilated by the travefling result data accumulation unft 510 (SP4).

[0060] Subsequenfly, the vehicle characterisflcs estimaflon unft 507 has the waveform shift amount calculation unit 513 correct the occurred deceleration waveform by shifting the occurred deceleration waveform in the time axis direction and the deceleration axis direction by a specified amount to make it doser to the corrimanded deceleration waveform. Then, the vehicle characteristics estimation unit 507 calculates the then-shifted amount as a waveform shift amount (SP5) and outputs the calculated waveform shift amount to the vehicle characteristics adoption possibility judgment unit 508.

[0061] After acquiring the waveform shift amount from the vehicle characteristics estimation unit 507, the vehicle characteristics adoption possibility judgment unit 508 has the residual deviation calculation unit 514 calculate the residual deviation based on the commanded deceleration waveform: the corrected occurred deceleration waveform, and the waveform shift amount (SF6).

[0062] Next, the vehicle characterist!cs adoption possibility judgment unt 508 has the residual deviation threshold judgment unit 515 judge whether the residual deviation calculated in step SF6 is less than a specified threshold value or not (SP7).

[0063] When the vehicle characteristics adoption possibility judgment unit 508 obtains a negative result in the judgment of step SPY, it terminates this processing. On the other hand, when the vehicle characteristics adoption poss!bility judgment unit 508 obtains an affirmative result in the judgment of step SF7, it outputs the waveform shift amount calculated by the vehicle characteristics estimation unit 507 to the vehicle characteristics adoption unit 509.

[0064] After acquiring the waveform shift amount from the vehicle characteristics adoption possibility judgment unit 508, the vehicle characteristics adoption unit 509 accumulates the waveform shift amount as vehicle characteristic data in the vehicle characteristic data accumulation unit 516 (SP8).

[0065] Subsequently, the vehicle characteristics adoption unit 509 has the vehicle characteristic statistical processing unit 517 execute vehicle characteristic data statistical processing on a set of vehicle characteristic data accumulated in the vehicle characteristic data accumulation unit 516 as population (SP9).

Incidentally, an example of the vehicle characteristic data statistical processing can be mode processing; however, any processing method may be executed as long as the method is to decide a representative value.

[0066] Lasfly, the veh!cle characteristics adoption unit 509 updates the adjustment parameter values in the control command calculation unit 506 by reflecting the representative value decided by the vehicle characteristic data.

statistical processing in the control command calculation unit 506 (SPIO), thereby terminating this processing.

[0067] (2-3) Advantageous Effects of First Embodiment Accord!ng to the first embodiment as described above, the residual deviation is calculated by comparing the commanded deceleration waveform with the occurred deceleration waveform; the adjustment parameter values are rot updated when a disturbance event occurs and the residual deviation becomes equal to or more than a specified threshold value; and the adjustment parameter values are updated only when the residual deviation is less than the specified threshold value. So, it is possible to control travelling of the train based on appropriate adjustment parameters. Therefore, it is possible to prevent degradation of followab!lity with respect to the target speed and enhance the stop position accuracy and the comfortable ride.

[0068] (3) Second Embodiment In a second embodiment, an explanation will be given about another automatic train control apparatus (ATO apparatus) that: calculates a characteristic amount existing in the occurred deceleration waveform; does not update the adjustment parameter values when a disturbance event occurs and the characteristic amount is equal to or more than the specified threshold value; and updates the adjustment parameter values only when the characteristic amount is less than the specified threshold value. Incidentally, the same reference numerals as those in the first embodiment are assigned to the same components as those of the first embodiment and an explanation about them has been omitted.

[0069] (3-1) Configuration of Automatic Train Control Apparatus Fig. 9 illustrates a functional structure of an automatic train control apparatus 501A according to the second embodiment. A vehicle characteristics learning unit 505A is composed of the vehicle characteristics estimation unit 507, a vehicle characteristics adoption possibility judgment unit 508A, and the vehicle characteristics adoption unit 509. In this example, an explanation about the vehicle characteristics estimation unit 507 and the vehicle characteristics adoption unit 509 has been omitted and the vehicle characteristics adoption possibility judgment unit 508A will be explained.

[0070] The vehicle characteristics adoption possibility judgment unit 508A judges whether the vehicle characteristics estimated by the vehicle characteristics estimation unit 507 should be adopted in the control command calculation unit 506 or not, based on the occurred deceleration waveform calculated by the vehicle characteristics estimation unit 507; and if it is determined that the vehicle characteristics should be adopted in the control command calculation unit 506, the vehicle characteristics adoption possibility judgment unit 506A outputs the vehicle characteristics to the vehicle characteristics adoption unit 509.

[0071] Next, the details of the vehicle characteristics adoption possibility judgment unit 508A will be explained. The vehicle characteristics adoption possibility judgment unit 508A is composed of an occurred deceleration waveform characteristic amount calculation unit 514A and a waveform characteristic amount threshold value judgment unit 51 5A.

[0072] The occurred deceleration waveform characteristic amount calculation unit 514A acquires the occurred deceleration waveform calculated by the occurred deceleration waveform c&culation unit 512 and calculates the characteristic amount existing in the occurred deceleration waveform.

[0073] Fig. 10 and Fig. 11 are explanatory diagrams of the characteristic amount.

Specificafly speaking, Fig. 1OA is a diagram iflustrating the relationship between the vehicle speed and the elapsed time when skidding occurs; and Fig. lOB is a diagram illustrating the relationship between the occurred deceleration waveform and the elapsed time when skidding occurs. Furthermore, Fig. 11 is a diagram illustrating the deceleration waveform when the braking command is turned on and off repeatedly at short intervals.

[0074] An example of the characteristic amount existing in the occurred deceleration waveform can be the number of moments when the occurred deceleration is located between positive sections or the occurred deceleration becomes zero or a negative value." Such a phenomenon occurs when skidding occurs during braking or when a brake notch is turned on and off minutely.

[0075] When skidding occurs during braking or when the brake notch is turned on and off minutely, there is a gap between the shape of the commanded deceleration waveform and the shape of the occurred deceleration waveform; and a large residual deviation remains even if the occurred deceleration waveform is shifted in the time axis direction and the deceleration axis direction.

Therefore:, the vehicle characteristic value cannot be estimated properly.

[0076] The characteristic amount calculated by the occurred deceleration waveform characteristic amount calculation unit 514A is output to the waveform characteristic amount threshold value judgment unit 515k The waveform characteristic amount threshold value judgment unit 51 5A judges whether the input characteristic amount is less than a specified threshold value or not.

[0077] When the characteristic amount is less than the specified threshold value, the waveform characteristic amount threshold value judgment unit 51 5A outputs the waveform shift amount calculated by the waveform shift amount calculation unit 513 to the vehide characteristics adoption unit 509 [0078] (3-2) Flowchart Fig. 12 Hustrates a processing sequence of adjustment parameter update processing executed by the vehicle characteristics earning unit 505A. Since processing from step SF11 to SF15 is the same as the adjustment parameter update processing (Fig. 8: from SRi to SF5) according to the first embodiment, an explanation about it has been omitted here.

[0079] In step SPI6, the vehicle characteristics adoption possibility judgment unit SOSA acquires the occurred deceleration waveform calculated by the vehicle characteristics estimation unit 507 and has the occurred deceleration waveform characteristic amount calculation unit 514A calculate the characteristic amount existing in the input occurred deceleration waveform (SF16).

[0080] Subsequently, the vehicle characterist!cs adoption possibility judgment unit 508A has the waveform characteristic amount threshold value judgment unit 51 5A judge whether the characteristic amount calculated in step SF16 is less than a specified threshold value or not (SF17).

[0081] When the vehicle characteristics adoption possibility judgment unit 508A obtains a negative result in the judgment of step SF17, the vehicle characteristics adoption possibility judgment unit SOBA terminates this processing. On the other hand, when the vehicle characteristics adoption possibility judgment un!t 508A obtains an affirmative result in the judgment of step SF7, t outputs the waveform shift amount calculated by the vehicle characteristics estimation unit 507 to the vehicle characteristics adoption unit 509.

[0082] Since processing of subsequent steps SPI8 to SF20 is the same as the adjustment parameter update processing (Fig. 8: from SPB to SP1O) according to the first embodiment, an explanation aboLit it has been omitted here.

[0083] (3-3) Advantageous Effects of Second Embodiment According to the second embodiment as described above, the characteristic amount existing in the occurred deceleration waveform is calculated; the adjustment parameter values are not updated when a disturbance event occurs and the characteristic amount is equal to or more than the specified threshold value; and the adjustment parameter values are updated only when the characteristic amount is less than the specified threshold value. So, it is possible to control travelling of the train based on the appropriate adjustment parameters.

Therefore, it is possible to prevent degradation of followability with respect to the target speed and enhance the stop position accuracy and the comfortable ride.

[0084] (4) Third Embodiment In a third embodiment, an explanation will be given about another automatic train control apparatus (ATO apparatus) that: calculates a characteristic amount existing n the commanded deceleration waveform: does not update the adjustment parameter values when a disturbance event occurs and the characteristic amount 5 eqUal to or more than the specified threshold value; and updates the adjustment parameter values only when the characteristic amount is less than the specified threshold value. Incidentally, the same reference numerals as those in the first embodiment are assigned to the same components as those of the first embodiment and an explanation about them has been omitted.

[0085] (4-1) Configuration of Automatic Train Control Apparatus Fig. 13 illustrates a functional structure of an automatic train control apparatus 5016 according to the third embodiment. A vehicle characteristics learning unit 5056 is composed of the vehicle characteristics estimation unit 507, a vehicle characteristics adoption possibility judgment unit 508B, and the vehicle characteristics adoption unit 509. In this example, an explanation about the vehicle characteristics estimation unit 507 and the vehicle characteristics adoption unit 509 has been omitted and the vehicle characteristics adoption possihihty judgment unft 5083 wifi be explained.

[0086] The vehicle characteristics adoption possibihty judgment unft 508B Judges whether the vehicle characteristics estimated by the vehicle characteristics estimation unit 507 should be adopted in the control command calculation unit 506 or not, based on the commanded deceleration waveform calculated by the vehicle characteristics estimation unit 507; and if it is determined that the vehicle characteristics should be adopted in the control command calculation unit 506, the vehicle characteristics adoption possibility judgment unit 508B outputs the vehicle characteristics to the vehicle characteristics adoption unit 509.

[0087] Next, the detaUs of the vehicle characteristics adoption possibility judgment unit 5083 will be explained The vehicle characteristics adoption possibility judgment unit 5083 is composed of a commanded deceleration waveform characteristic amount calculation unit 5143 and a waveform characteristic amount threshold value judgment unit 51 5B.

[0088] The commanded deceleration waveform characteristic amount calculation unit 5143 acquires the commanded deceleration waveform calculated by the commanded deceleration waveform calculation unit 511 and calculates the characteristic amount exiting in the commanded deceleration waveform.

[0089] An example of the characteristic amount existing in the commanded deceleration waveform can be the number of moments when the commanded deceleration is located between positive sections or the occurred deceleration becomes zero." Such a phenomenon occurs when the brake notch is turned on and off minutely (Fig. 11).

[0090] When the brake notch is turned on and off minutely, there is a gap between the shape of the commanded deceleration waveform and the shape of the occurred deceleration waveform; and a large residual deviation remains even if the waveform is shifted in the time axis direction and the deceleration axis direction. Therefore, the vehicle characteristic value cannot be estimated properly.

[0091] The characteristic amount calculated by the commanded deceleration waveform characteristic amount calculation unt 514B is output to the waveform characteristic amount threshold value judgment unit 51 SB. The waveform characteristic amount threshold value judgment unit 51 SB judges whether the input characteristic amount is less than a specified threshold value or not.

[0092] If the characteristic amount is less than the specified threshold value: the waveform charactensflc amount threshold value judgment unt 51 SB outputs the waveform shift amount calculated by the waveform shift amount calculation unit 513 to the vehicle characteristics adoption unit 509.

[0093] (4-2) Flowchart Fig. 14 iflustrates adjustment parameter update processing executed by the vehicle characteristics learning unit 505B. Since processing from step SP2I to SP2S is the same as the adjustment parameter update processing (Fig. 8: from SP1 to SP5) according to the first embodiment, an explanation about it has been orn itted here.

[0094] In step SP26, the vehicle characteristics adoption possibility judgment unit 5083 acquires the commanded deceleration waveform calculated by the vehicle characteristics estimation unit 507 and has the commanded deceleration waveform characteristic amount calculation unit 5143 calculate a characteristic amount existing n the input commanded deceleration waveform (SP26).

[0095] Next, the vehicle characteristics adoption possibility judgment unit 5083 has the waveform characteristic amount threshold value judgment unit 5153 judge whether the characteristic amount calculated in step SP26 is less than a specified threshold value or not (SP27).

[0096] If the vehicle characteristics adoption possibility judgment unit 508B obtains a negative result in the judgment of step SP27, it terminates this processing. On the other hand, if the vehicle characteristics adoption possibility judgment unit 508B obtains an affirmative result in the judgment of step SP27, it outputs a waveform shift amount calculated by the vehicle characteristics estimation unit 507 to the vehicle characteristics adoption unit 509.

[00971 Since processing of subsequent steps from step SP28 to SP3O is the same as the adjustment parameter update processing (Fig. 8: from SP8 to SPIO) according to the first embodiment, an explanation about it has been omitted here.

[0098) (4-3) Advantageous Effects of Third Embodiment According to the third embodiment as described above, the characteristic amount existing in the commanded deceleration waveform is calculated; the adjustment parameter values are not updated when a disturbance event occurs and the characteristic amount is equal to or more than the specified threshold value; and the adjustment parameter values are updated only when the characteristic amount is less than the specified threshold value. So, it is possible to control travelling of the train based on the appropriate adjustment parameters.

[0099) (5) Other Embodiments The first to third embodiments have described the automatic train control apparatus that properly adopts the estimated vehicle characteristics in control based on the estimated accuracy of the vehicle characteristics; however, an application target is not limited to the automatic train control apparatus and, for example, the invention may be applied to, for example, a vehicle status diagnosis device that notifies a driver or an operating company of the vehicle status.

[0100) In this case, the vehicle characteristic value output from the vehicle characteristic statistical processing unit 517 (Fig. 5, Fig. 9, and Fig. 13) is reported to the driver and the operating company regardless of onboard or on the ground, or in operation or after operation. The report is made by means of, for example, screen displays or sounds.

[01011 The driver can realize more stable transportation by changing driving operation based on the reported vehicle status. For example, when the vehicle status is reported to indicate that the deceleration deviation is less than usual, that is, it is more difficult than usual to implement deceleration, it is possible to reduce the possibility of overrun when stopping at a station by using a larger brake notch than usual or putting on the brake earlier than usual.

[0102] Furthermore, the operating company can realize more stable transportation by developing a maintenance plan based on the reported vehicle status. For example, when the vehicle status is reported to indicate that the deceleration deviation has been gradually decreasing, that is, the occurred deceleration for the same command has been decreasing, the possibility to cause an unsafe event of a disabled brake can be reduced by suspecting degradation of brake blocks and performing maintenance of the vehicles earlier.

[0103] Furthermore, when the vehicle status is reported to indicate that the deceleration deviation has reduced only between specified stations along the relevant train line, the possibility to cause an unsafe event can be reduced by suspecting a failure of ground equipment at the relevant location and optimizing a maintenance or inspection plan of the ground equipment.

[Reference Signs List] [0104] 501, 501A, 501 B automatic train control apparatus 502 braking and driving control device 503 actuator 504 speed and position detection unit 505, 505A, 505B vehicle characteristics learning unit 506 control command calculation unit 507 vehicle characteristics estimation unit 508, 508A, 508B vehicle characteristics adoption possibility judgment unit 509 vehce characteristics adoption unit