Title:
ELECTRONIC DEVICE FOR THE AUTOMATIC CONTROL OF A RAILWAY TRAIN
United States Patent 3604905


Abstract:
A system for the automatic control of a railway train wherein the itinerary is divided into several zones, in each of which the feed current of the motors can be cut off before the end of the zone is reached by the train if a time interval which is a determined function of the distance counted from the beginning of the zone becomes greater than the real time.



Inventors:
RIONDEL PIERRE
Application Number:
04/744553
Publication Date:
09/14/1971
Filing Date:
07/12/1968
Assignee:
SA. DES ATELIERS DE SECHERON
Primary Class:
Other Classes:
701/19
International Classes:
B61L27/04; B60L15/20; (IPC1-7): G06F15/48
Field of Search:
235/150
View Patent Images:
US Patent References:



Foreign References:
GB923915A1963-04-18
GB980687A
FR1382146A1964-12-18
Other References:

Railway Signaling & Communications, "Controls Vital to Mass Transit," Sept. 1963, pp. 13-18 (246-187B)..
Primary Examiner:
Morrison, Malcolm A.
Assistant Examiner:
Gruber, Felix D.
Claims:
I claim

1. An electronic device for the automatic control of an electrically operated railway engine along a railway track said track being divided into several zones between geographical points along said track, in each of which the feed current of the motors can be cut off automatically before the engine reaches the end of each selected zone if a time interval which is a predetermined function of the distance counted from the beginning of said selected zone becomes greater than the real time, said device comprising: first means for registering the distance of the engine from the beginning of each selected zone; first means for counting the distance run from the beginning of the first zone of said track; means for comparing the distance run with the distance registered by said first registering means; means operatively associated with said comparing means for generating a pulse when said train reaches said distance counted from the beginning of each zone; a first counting means operatively connected to the output of said pulse generating means for setting said first time counting means to zero and for indicating a time which is a predetermined function of the distance from the beginning of a corresponding zone; second means for registering in said first counting means the time representing the initial value of the function for the distance from the beginning of a zone; means for feeding during the run from one distance to another, said first counting means with pulses representing the values of said predetermined function for generating a time signal means for counting pulses and generating a signal corresponding to the real time; and further comprising means for comparing the real time signal with said time signal and means generating a pulse to cut off the current when the signals indicated by said first and second counting means become equal.

2. A device according to claim 1, wherein said registering means comprise resistances.

3. A device according to claim 1, wherein said function is represented by a curve comprising several segments of a straight line; and said feeding means comprises third means for registering the length of said segments and fourth means for registering the time corresponding to the length of said segments.

4. A device according to claim 1, wherein said function is represented by a continuous curve; and said feeding means comprises third means for registering a coefficient and means multiplying said coefficient by a parameter which caries with said distance run to modify the slope of said curve.

Description:
It is known that in certain cases the automatic control of a railway train is effected by cutting off the current to the motors, according to a well-defined criterium.

When the train effects a service in which the stops are close together, such a criterium may be the minimum energy consumption, a criterium which was defined by Mr. Bernard, during the congress of cybernetics of the UIC in Paris, in the autumn of 1963.

According to this criterium, the feed current of the motors is cut off, in a zone delimited between two consecutive stations, after the train has run for a distance calculated in advance and registered on board the motor carriage, in such a fashion that the time taken for the run between one station and the next is a predetermined function of the time remaining to reach the terminus.

This criterium of minimum consumption is moreover also applicable to direct trains effecting a long run without intermediate stops where this run represents different profiles. This run may be divided into several successive zones in each of which the current to the motors may be cut off before the train has reached the end of a zone.

Each zone requires the registering of the parameters representing the said function.

To accomplish the above requires the registration of as many sets of parameters as there are zones for each running direction. In addition, if the motor carriage or van is required to haul trains running on different lines, or else if the train effects services of different nature on a same line (local and semidirect services for example), the total number of parameters becomes very large.

Logic electronic devices have already been conceived which register the parameters by means of diode matrixes. The disadvantage of these devices resides in the large number of diodes required which may reach several tens of thousands. The reliability, the cost and the large size of these devices are not compatible with the exigencies of railway service.

The purpose of the invention is to provide a device which avoids these disadvantages.

According to the invention, the electronic device for the automatic control of a railway train along an itinerary comprising several zones delimited between geographical points along the itinerary, the positions of which depend on the nature of the train (local or direct train, etc.), in each of which the feed current of the motors can be cut off automatically before each zone end, if a time interval which is a determined function of the distance counted from the beginning of the zone becomes greater than the real time, is characterized in that it comprises a generator for generating the functions corresponding to the different zones, sets of electronic components for the registration of the parameters of these functions, these sets of components each comprising as many components as there are zones, means being provided for the commutation of the components of each of these sets when passing from one zone to the next.

The drawing illustrates, by way of examples, two embodiments of the device.

FIG. 1 is a diagram illustrating how the current cutoff control is effected when a curve Tr, representing the real time in function of the distance run, intersects a curve Tcc, representing the time at which the current must be cut off as a function of the distance run.

FIG. 2 is a diagram illustrating the approximation of the curve Tcc by means of a polygonal curve, for a single zone such as is used in the first embodiment.

FIG. 3 is a block diagram illustrating the first embodiment of the device.

FIG. 4 is a view corresponding to that of FIG. 2 and illustrating the approximation of the curve Tcc by means of an increasing monotonic curve, which is used in the second embodiment.

FIG. 5 is a block diagram illustrating the second embodiment.

FIG. 1 illustrates a railway itinerary comprising a departure station G1, two intermediate stations G2 and G3, and a terminus station G4. The abscissi X represent the lengths of run along the itinerary, and the ordinates Y, represent both the real time Tr in any particular case of running of a train as well as the time Tcc indicating the moments when the feed current of the motors must be cut off after the train has runup speed. The itinerary illustrated thus comprises three zones Z1, Z2, Z3 between the stations, as well as four station zones ZG1, ZG2, ZG3, ZG4. The curve Tcc represents for each section the instant when the current must be cut off in order that on the one hand the train may reach the terminus station G4 at the time fixed in advance, and on the other that the running of the train is effected with a minimum consumption of energy.

Thus, if at a distance from the departure station, the real time Tr, counting from the time of departure of the itinerary, is greater than the time Tcc, the current supply must be maintained, while if it is less, the current must be cut off. In other words, each time the curve Tr representing the real time intersects the curve Tcc, the current must be cut off. The curve Tcc is interrupted in the station zones in which braking and then starting intervene.

The automatic control device illustrated in FIG. 3 enables the current to be cut off automatically, this device generating successively the three sections of the curve Tcc in the form of an approximation, i.e., by assimilating each one of them with a polygonal curve composed of three segments of straight line, as will be seen later on, with reference to FIG. 2.

The device illustrated in FIG. 3 comprises a selector switch having a common shaft 1 which is actuated step by step by an electromagnetic control mechanism 2, one step being effected every time that an end of zone impulse FZ is applied to its input. This selector controls the selection of one of the three resistances comprised by each of 10 groups of resistances 3-12, each of these groups corresponding to a value to be indicated, the first resistance of each group corresponding to the zone Z1, the second to the zone Z2 and the third to the zone Z3. The device comprises a distance counter 13 with two outputs 13a and 13b, of which the first generates a current which is proportional to the number D'r of kilometers run from the starting station G1, and the second a current proportional to the number D"r of hectometers covered since the end of the last kilometer. The current generated by the output 13a is applied to the resistances of the first group 3 while the current generated by the output 13b is applied to the second group 4. The opposite terminals of these resistances being connected to a common negative voltage source, the negative terminal of which is indicated in 14, it will be seen that when the currents of the outputs 13a and 13b are on the increase, the voltage at these outputs referenced to ground passes through zero. These two outputs are also connected, the first 13a to the input of a voltage discriminator 15 which generates a constant output voltage when its input voltage is situated in the neighborhood of zero, and the second 13b to a discriminator 16 operating in the same fashion. The outputs of the two discriminators 15 and 16 are applied to the corresponding inputs of a logic circuit and 17, which generates a positive voltage when the two outputs of 15 and 16 are positive. The output signal of the logic circuit and 17 is applied to a monostable circuit 18 which generates a single pulse 0 at the moment of arrival of the forward flank of a signal emitted by the circuit 17. The set of circuits 15, 16, 17 and 18 thus constitutes a comparator circuit 19 which emits a pulse when the distance indicated by the counter 13 is equal to that indicated by the resistances of the groups 3 and 4 which are inversely proportional to a predetermined distance, e.g., 1 km. = 10 kΩ, 2 km. = 5 kΩ, 3 km. = 3.33 kΩ, etc.

The group of resistances 5 cooperates with an analog digital converter 20, which is initiated by the end of the pulse 0, and emits a train of pulses the number of which is inversely proportional to the input current defined by that resistance of the group 5 which is in active position, the resistances 5 representing, as will be seen later on, a time interval T'o in hundredths of a second.

The group of resistances 6 cooperates in the same way with an analog digital converter 21, which, at the end of the pulse 0, emits a train of pulses, representing a time interval T"o in tenths of a second.

The analog values T1, D1, T2, D2, T3, D3 respectively indicated by the groups of resistances 7, 8, 9, 10, 11, 12 are applied to one of the inputs of a corresponding analog gate 22, 23, 24, 25, 26, 27 which allow these values to pass through at their output, when a logic signal corresponding to the value 1 is applied to their second input. The said second inputs of the analog gates 22 and 23 are controlled by the output signal of a logic circuit OR 28 which emits a logic signal having the value 1 when its input signals FA and FB have the value 0. The said second inputs of the analog gates 24 and 25 are controlled by the output signal of a logic circuit AND 29 with an inverted input, AND 29 emits a logic signal having the value 1 when the input signal FA= 1 and the input signal FB=0.

The second inputs of the analog gates 26 and 27 are controlled by the output signal of a logic circuit AND 30 with an inverted input, AND 30 emits a logic signal having the value 1 when the logic input signal FB=1 and the logic input signal FA=0.

The output signals of the analog gates 22, 24 and 26 are applied to the inputs of an analog summing device 31, which emits an analog output signal which is equal to the sum of the analog input signals.

The output signals of the gates 23, 25 and 27 are applied in the same way to the inputs of an analog summing device 32, which emits an output analog signal which is equal to the sum of the analog input signals.

The analog output signal of the summing device 31 is applied to the input of an analog logic converter 33, which, at the moment the forward flank of a signal arrives emits a train of pulses the number of which is proportional to the analog input signal.

This train of pulses is applied to the input of a counter 34 comprising three decades 35, 36 and 37 which are set back to zero by the forward flank of the pulse 0 applied thereto. These three decades are each constituted in a well-known manner by four flip-flops connected in cascade, with a feedback circuit so that they count only up to 10. The output signal of the first decade is applied to one of the inputs of a logic circuit OR 38a the other input of which is fed by the output of the analog-logic converter 21. The output of the circuit OR 38a is applied to the input of the second decade 36 the output signal of which is applied to one of the inputs of a logic circuit OR 38b the other input of which is fed by the output of the analog-digital converter 20 and the output of which is applied to the last decade 37.

The device comprises in addition a counter 39 counting the real time and composed of a pulse generator 40 emitting a pulse every second, these pulses being applied to a chain of decades 41, 42, 43 each constituted like the decades 35, 36, 37. The outputs of the decades 35 and 41, 36 and 42, 37 and 43 are respectively compared in 44, 45, 46 where a logic signal is emitted when the values are the same. These three logic signals are applied to a logic circuit AND 47 at the output of which a pulse Pcc is emitted when the outputs of the two counters are the same, this pulse controlling the cutting off of the current. Counter 39 can be manually reset by pulse D.

The device also comprises in addition an analog-digital divider 48 which receives a pulse at input E1 for every 10 meters running distance and at input DM an analog signal issuing from the summing device 32 divider 48 emits a pulse P after it has received at E1 a number of pulses which is proportional to the analog signal at DM. These pulses P are applied to a decade 49 which emits one pulse P' for every 10 pulses P it receives.

The output pulses of the decade 49 are applied to a flip-flop 50 the output signal of which feeds a second flip-flop 51 as well as one of the inputs of a logic circuit AND 52 the other input signal of which is constituted by the output signal of the second flip-flop 51 and the output signal of which constitutes the end of zone signal FZ. The output signals FA and FB of the flip-flops 50 and 51 themselves constitute the input signals of the logic circuits 28, 29, 30.

The output Tcc as a function of distance of the device described with reference to FIG. 3 is illustrated in FIG. 2. It is seen there that the theoretical curve Tcc representing the moments when the current must be cut off is approximated by a polygonal curve formed by three segments of a straight line, this polygonal curve being different for each one of the zones Z1, Z2 and Z3. The zone in which the automatic control is effected begins at the distance Do, this distance being indicated in km. D'o by the resistances 3 and in hectometers D"o by the resistances 4. When the train passes at this distance Do, the comparator 19 emits a pulse 0 which signals the beginning of the zone. This pulse 0 indicates in the counter 34 the time To corresponding to the beginning of the zone, this time To being indicated analogically in hundredths of a second T'o by the resistances 5 to be converted into a digital value by the converter 21.

The analog-digital divider 48 generates pulses according to a metric step defined to begin with by the distance D1 indicated by the resistances 8. After 10 pulses indicating that the train has run the distance D1 from the point Do, the decade 49 actuates the two flip-flops 50 and 51 which substitute, under control of their output signals FA, FB, the distance D2 for D1, then D3 for D2. When the last pulse corresponding to the distance D3 is generated, the logic circuit 52 emits an end of zone pulse FZ which causes the selector 1 to advance one step and the resetting of the counters at zero.

For every pulse P the converter 33 causes the counter 34 to advance its count by as many seconds as indicated by the coefficient of increase of the curve Tcc, T being equal to T1 for the segment D1, then to T2 and T3 for the segments D2 and D3. The commutation of these times T will obviously be effected at the same time as that of the distances D, under control of the signals FA and FB.

A pulse Pcc causing the current to be cut off will only be generated when the real time signal To generated by the counter 39 becomes equal to the time signal Tcc generated by the counter 34.

FIG. 4, corresponding to FIG. 2, shows how the curve Tcc, instead of being approximated by a polygonal function, can be approximated by a continuous monotonic function Tb, generated in the second embodiment illustrated in FIG. 5.

The components 1-6, 13-21 and 33 to 47 of this second embodiment are identical with those corresponding to the embodiment according to FIG. 3 and will not again be described.

This embodiment comprises in addition three sets of resistances 100, 101 and 102 which can be commutated by the selector 1. Set 100 serves to determine the initial slope of the curve Tb, at To, Do. The set of resistances 101 serves to determine, together with the set of resistances 100, the final slope of the curve Tb, i.e., the beginning of the zone Z of the following station. The set of resistances 102 determines the function according to which the slope is changed.

The resistances 100 are connected to one of the inputs of an analog-summing device 103 with two inputs, by means of the corresponding commutator. In the same fashion, the resistances 101 are connected to the input of an analog gate 104, which multiplies the coefficient determined by the resistances 101 by a factor comprised between zero and one, and which is introduced into a second input S of the gate 104. The signal S is given in the shape of pulses the width of which is modulated according to the said coefficient, the gate 104 being only open when the signal S is applied. It may thus be seen that the mean value of the output signal of the gate 104 corresponds effectively to the product of the two input values. This mean value is applied to the second input of the summing device 103, the output of which is applied to the input of the analog-digital converter 33 described above.

The resistances 102 are connected to one of the inputs of an analog-logic divider 105, which at the other input E1, receives a pulse for every 10 meters run. This divider generates a pulse P after having received in E1 a number of pulses which is proportional to the analog value indicated by the resistances 102.

The pulses P are applied to a chain of four flip-flops capable of taking up 16 positions.

The outputs of these four flip-flops 106-109 are respectively applied to four inputs of a digital-to-analog converter 110, the analog output voltage of which is proportional to the number indicated by the chain of flip-flops 106-109. The analog output of the converter 110 controls a pulse generator 111, so as to modulate the width of the pulses S generated by the generator 111, the ratio of the duration of the pulses to their frequency of repetition varying from zero to one when the number indicated by the flip-flops passes from zero to 16.

The converter 110 also generates a pulse FZ when the number indicated by the chain of flip-flops reaches 16.

This device operates as follows: each time a distance corresponding to the step indicated by the resistances 102 has been run through, a pulse is generated in P. Each one of these pulses causes the chain of flip-flops 106-109 to advance by one unit and consequently increases the rate of modulation of the pulses S by one-sixteenth. At the start, when the number indicated by the chain of flip-flops 106-109 is zero, the duration of the pulses S is zero and the input signal of the converter 33 corresponds to the value indicated by the resistances 100, i.e., to the initial slope of the curve Tb.

After the 15th pulse P, the duration of the pulse S is equal to its period of repetition, the output signal of the gate 104 is equal to the value indicated by the resistances 101, and consequently the output signal of the summing device 103 corresponds to the sum of the values indicated by the resistances 100 and 101, i.e., to the final slope of the curve Tb. Similarly to resistance groups 3, 4, 5 and 6, resistances 100, 101 and 102 are inversely proportional to a predetermined distance.

Although an itinerary has been represented which only comprises three zones, i.e., two intermediate stations, any number of zones may be provided for, and it is only necessary to increase in corresponding fashion the number of resistances of each one of the sets of resistances.

The curve Tcc is approximated in the first embodiment by a curve comprising only three segments and although in the second embodiment only 16 steps have been provided; these numbers may be increased in order to achieve a better approximation.