CN113390495A - Scene recognition-based urban light rail vehicle load online estimation method - Google Patents
Scene recognition-based urban light rail vehicle load online estimation method Download PDFInfo
- Publication number
- CN113390495A CN113390495A CN202110644148.9A CN202110644148A CN113390495A CN 113390495 A CN113390495 A CN 113390495A CN 202110644148 A CN202110644148 A CN 202110644148A CN 113390495 A CN113390495 A CN 113390495A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- train
- traction
- weight
- scene
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G19/00—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups
- G01G19/02—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for weighing wheeled or rolling bodies, e.g. vehicles
- G01G19/04—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for weighing wheeled or rolling bodies, e.g. vehicles for weighing railway vehicles
- G01G19/045—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for weighing wheeled or rolling bodies, e.g. vehicles for weighing railway vehicles for weighing railway vehicles in motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Theoretical Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Algebra (AREA)
- Operations Research (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The embodiment of the invention relates to an urban light rail vehicle load online estimation method based on scene recognition, which comprises the following steps: the vehicle network controller CCU acquires vehicle weight information stored in a FLASH memory FLASH of a vehicle and sends the vehicle weight information to a vehicle traction controller TCU and a vehicle braking system controller BCU; the CCU detects scene identification information in real time, and determines the current scene information of the vehicle according to the detected scene identification information; when the current scene information of the vehicle is a normal operation scene, executing first estimation processing to obtain vehicle weight updating data; when the current scene information of the vehicle is a special operation scene, executing second estimation processing to obtain vehicle weight updating data; and writing the vehicle weight updating data into a FLASH memory for updating the vehicle weight information stored in the FLASH memory.
Description
Technical Field
The invention relates to the technical field of rail transit, in particular to an urban light rail vehicle load online estimation method based on scene recognition.
Background
The light rail is an important form of urban rail construction, is also the rail traffic form which develops most rapidly in the world at present, has the advantages of large traffic volume, high speed, low pollution, low energy consumption, punctual operation, high safety and the like, is an effective carrier for improving the current situation of urban traffic, and becomes an important choice for modernizing the metropolis.
Urban rail vehicle passenger transportation has the characteristics of high pedestrian volume density, large passenger-carrying rate fluctuation, uneven track infrastructure conditions and the like, and the accurate control of traction force and braking force of a vehicle traction system and a vehicle braking system is required to be realized according to vehicle load (equal weight). Through accurate control, can effectively control the vehicle and start the stage, control level transition stage, the driving impact rate under the operation scenes such as electric braking and mechanical braking interactive coordination stage, improve passenger's comfort rate, special passenger such as the sick, disabled and pregnant of maximum protection seatless and old and young, through accurate control, make vehicle traction force and brake force be linear relation with between the vehicle load, the dynamic behavior of vehicle keeps unanimous under each load, accurate control operation picture when being convenient for unmanned, train automatic driving (ATO) or manual operation. Through accurate control, the rail adhesion condition can be utilized to the maximum extent to improve the vehicle power performance, and the risks of triggering idling and sliding of the vehicle wheel pair are reduced as far as possible.
Therefore, the load of the vehicle plays an important role in rail transit, and most rail vehicles are provided with load sensors on a bogie or converted into vehicle load by measuring the height of an air spring, so that the vehicle weight calculation precision is high, and the vehicle weight can be monitored. However, light rail vehicles belonging to urban rail transit systems, especially low-floor vehicles, are subject to floor height, cost pressure, maintenance difficulty and other factors, and most vehicles are not provided with weight sensors, such as most vehicle models in europe and vehicle models in several vehicle host factories such as down mountain in China.
After the weight sensor is cancelled, most traction and braking systems do not consider the influence of vehicle load on the power performance any more, the magnitude of traction braking force is controlled only by the main handle level of a driver controller, and the maximum handle level corresponds to the maximum traction and braking force which can be exerted by a vehicle and is similar to an electric vehicle. This control method is simple, but has a number of disadvantages for systems such as rail transit:
1. when the seating rate is changed, the vehicle power performance fluctuates greatly along with the actual load, and the vehicle is difficult to control to follow an operation circuit diagram during an ATO driving system or manual operation, so that punctuality and accurate parking are difficult, and faults of late time, mark flushing and mark shortage frequently occur;
2. the change slope of the traction braking force cannot be adjusted according to the load, the vehicle impact rate exceeds the limit, the comfort level of passengers is reduced, and the personal safety of the passengers is threatened;
3. the adhesion condition of the track cannot be adaptively utilized, and the idling and sliding faults are easily and frequently triggered by the larger traction and braking force corresponding to the smaller load of the vehicle, even the wheel is rubbed, so that the power performance of the vehicle is attenuated.
The existing rail vehicle weight estimation method without the load sensor is less, the error is larger by a method for calculating the passenger riding rate, the only method for estimating the weight of the passenger vehicle is not suitable for special requirements of the rail vehicle, and the calculation error is larger due to the difference of dynamic characteristics.
Therefore, in order to solve the above difficulties, an online estimation method for vehicle load meeting the requirements of the rail transit vehicle running scene becomes more urgent.
Disclosure of Invention
The invention aims to provide an urban light rail vehicle load online estimation method based on scene recognition, which simulates the vehicle running under the condition of a load sensor by adopting different processing strategies corresponding to different scenes through deeply analyzing the vehicle application scenes and extracting scene identification information according to the cooperative relationship between the vehicle running and the vehicle load in different scenes, basically covers the daily working conditions of a rail transit system, has high accuracy of obtained results and provides powerful guarantee for the safe and stable running of a train.
Therefore, the embodiment of the invention provides an online estimation method for urban light rail vehicle load based on scene recognition, which comprises the following steps:
the method comprises the steps that a vehicle network controller CCU acquires vehicle weight information stored in a FLASH memory FLASH memory of a train vehicle, and issues the vehicle weight information to a vehicle traction controller TCU and a vehicle braking system controller BCU;
the CCU detects scene identification information in real time, and determines the current scene information of the vehicle according to the detected scene identification information;
when the current scene information of the vehicle is a normal operation scene, executing first estimation processing to obtain vehicle weight updating data;
when the current scene information of the vehicle is a special operation scene, executing second estimation processing to obtain vehicle weight updating data;
and writing the vehicle weight updating data into a FLASH memory for updating the vehicle weight information stored in the FLASH memory.
Preferably, before the CCU acquires the vehicle weight information stored in the FLASH memory of the vehicle FLASH memory and issues the vehicle weight information to the TCU and the BCU, the method further includes:
receiving a vehicle power-on activation signal;
and carrying out power-on initialization on an ATC (automatic train control), a CCU (central control unit), a TCU (remote control unit) and a BCU (binary-coded control unit) according to the vehicle power-on activation signal, and initializing the vehicle weight information stored in the FLASH memory into the vehicle weight information stored in the FLASH memory before the last power failure.
Preferably, the first estimation process includes:
determining a door closing state and a traction instruction signal receiving state of the vehicle;
when the door of the vehicle is closed and a traction command signal is received, the traction level parameter of the CCU is set to be 100% traction level, and the load parameter is set to be rated load AW 2;
detecting whether the current vehicle speed exceeds a first set threshold value;
when the current vehicle speed exceeds a first set threshold value, the CCU acquires train traction according to a set sampling frequency, and records the time when the vehicle speed reaches a first vehicle speed as first sampling time;
when the current vehicle speed exceeds a second set threshold value, ending the collection of the train traction force, and recording the time when the vehicle speed reaches a second vehicle speed as second sampling time;
calculating the acceleration of the vehicle in a time interval from the first sampling time to the second sampling time according to the first sampling time, the second sampling time, the first vehicle speed and the second vehicle speed, and taking the average value of the acquired train traction as the value of the train traction;
calculating theoretical estimated value of vehicle weight according to static parameters of the train, train speed, acceleration and ramp angle, and fitting function f by using least square method(m1)And compensating the theoretical estimated vehicle weight value to obtain the vehicle weight.
Further preferably, the calculating the theoretical estimated value of the weight of the vehicle according to the static parameters, the speed, the acceleration and the ramp angle of the train specifically comprises:
m1＝(F-130*n-([0.046+0.0065*(N-1)]*A*V2))/(k1*a+6.4+0.14*V+g*θ)
wherein m1 is the theoretical estimated value of the weight of the vehicle, F is the traction of the train, N is the number of axles, V is the speed of the train, N is the number of carriages of the train, A is the cross-sectional area of the train, a is the acceleration, k1 is the coefficient of revolution mass, g is the acceleration of gravity, and theta is the ramp angle.
Further preferably, the compensating the theoretical estimated vehicle weight value by a least square fitting function to obtain the vehicle weight specifically comprises:
m＝m1+f(m1)；
wherein m is the vehicle weight, f(m1)Fitting a function for a least square method under a normal operation scene.
Further preferably, the least square fitting function is used for matching the formula m-m 1+ f according to actual measured data under different loads of normal operation scenes AW0-AW3 before the vehicle leaves the factory(m1)And (5) carrying out calibration acquisition.
Preferably, the second estimation process includes:
determining a door closing state and a traction instruction signal receiving state of the vehicle;
when the door of the vehicle is closed and a traction command signal is received, the traction level parameter of the CCU is set to be 100% traction level, and the load parameter is set to be rated load AW 2;
detecting whether the current vehicle speed exceeds a first set threshold value;
when the current vehicle speed exceeds a first set threshold value, the CCU acquires train traction according to a set sampling frequency, and records the time when the vehicle speed reaches a first vehicle speed as first sampling time;
when the current vehicle speed exceeds a second set threshold value, ending the collection of the train traction force, and recording the time when the vehicle speed reaches a second vehicle speed as second sampling time;
calculating the acceleration of the vehicle in a time interval from the first sampling time to the second sampling time according to the first sampling time, the second sampling time, the first vehicle speed and the second vehicle speed, and taking the average value of the acquired train traction as the value of the train traction;
calculating a theoretical estimated value of the weight of the vehicle according to the static parameters of the train, the speed and the acceleration of the train and a preset maximum ramp angle, and fitting a function f by a least square method(m2)And compensating the theoretical estimated vehicle weight value to obtain the vehicle weight.
Further preferably, the calculating the theoretical estimated value of the weight of the vehicle according to the static parameters, the speed, the acceleration and the preset maximum ramp angle of the train specifically comprises:
m2＝(F-130*n-([0.046+0.0065*(N-1)]*A*V2))/(k1*a+6.4+0.14*V-θ0*g)
wherein m1 is the theoretical estimated value of the weight of the vehicle, F is the traction force of the train, N is the number of axles, V is the speed of the train, N is the number of carriages of the train, A is the section area of the train, a is the acceleration, k1 is the coefficient of revolution mass, g is the acceleration of gravity, theta0Is a preset maximum ramp angle.
Further preferably, the compensating the theoretical estimated vehicle weight value by a least square fitting function to obtain the vehicle weight specifically comprises:
m＝m2+f(m2)；
wherein m is the vehicle weight, f(m2)Fitting a function for a least square method under a special operation scene.
Further preferably, the least square fitting function is used for matching the formula m-m 1+ f through actual measured data under different load conditions according to normal operation scenes AW0-AW3 before the vehicle leaves a factory(m1)And (5) carrying out calibration acquisition.
According to the method for estimating the urban light rail vehicle load on line based on scene recognition, provided by the embodiment of the invention, the vehicle operation scene is deeply analyzed, the vehicle operation under the condition of a load sensor is simulated by adopting different processing strategies corresponding to different scenes through extracting the scene identification information according to the cooperative relationship between the vehicle operation and the vehicle load in different scenes, the daily working condition of a rail transit system is basically covered, the accuracy of the obtained result is high, and the powerful guarantee is provided for the safe and stable operation of a train.
Drawings
FIG. 1 is a flowchart of an online estimation method for urban light rail vehicle load based on scene recognition according to an embodiment of the present invention;
fig. 2 is a schematic view of an operation process of a normal operation scenario provided in an embodiment of the present invention;
fig. 3 is a schematic view of an operation process of a normal operation scenario provided in the embodiment of the present invention;
fig. 4 is a flowchart of a method for performing a first estimation process in a normal operation scenario according to an embodiment of the present invention;
FIG. 5 is a graph of train speed versus specific drag provided by an embodiment of the present invention;
FIG. 6 is a graph of train tractive effort provided by an embodiment of the present invention;
fig. 7 is a flowchart of a method for performing a second estimation process in a special operation scenario according to an embodiment of the present invention.
Detailed Description
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
The embodiment of the invention provides an urban light rail vehicle load online estimation method based on scene recognition, which can be applied to the online estimation of the weight of a rail vehicle without a load sensor and provides powerful guarantee for the safe and stable operation of a train.
The vehicle weight estimation is closely related to the vehicle running state, and is related to an entry point, an implementation strategy, a calculation method, a compensation strategy, a special working condition disposal strategy and the like in the estimation process, so that the vehicle running under the condition of the weight sensor can be simulated to the maximum extent through the load estimation based on scene identification.
By deeply analyzing the operation scene of the urban rail vehicle, the operation process, the in-site test scene and the fault rescue scene of the urban light rail vehicle can be divided according to the working conditions, and the combination of the three scenes basically covers the daily work content of the rail transit system.
The process of the urban light rail vehicle during operation can be as shown in fig. 1, and the method adopting the feature classification can be summarized into 3 scenes: firstly, vehicles carry passengers to normally enter and exit; the vehicle access section after the operation is started and finished; and thirdly, the vehicle is turned back from an ascending (descending) line to a descending (ascending) line.
The scene that the vehicle carries passengers to normally enter and exit is the most important working condition in load calculation, the vehicle load is updated, and a traction and braking system needs to adapt to the new vehicle weight again, so that the method is of great importance to the professional accurate control of an operation chart and the parking accuracy of operation scheduling, and therefore, the improvement of the measurement accuracy of the vehicle weight is an important task in the stage. Fig. 2 shows an exploded schematic diagram of this process, and as can be seen from fig. 2, there are several important scene identification information at this stage: the system comprises arrival information (including an arrival ID number), terminal information (a terminal ID number), a zero-speed signal and a door opening signal, wherein the 4 signals are important sign signals for the arrival and the stop of the vehicle, whether the train arrives at the station and stops can be accurately judged by identification, and the arrival information and the terminal information provide slope compensation basis for the following compensation. The method comprises the steps that a vehicle door closing signal, a traction instruction and a traction level mark indicates that the vehicle is about to operate, the weight of the vehicle is updated, the vehicle starts to operate under the control of an initial level, the starting process is also a deduction stage of subsequent vehicle weight estimation, the actual vehicle weight is estimated by using the virtual vehicle weight and the vehicle dynamic characteristics under the initial traction force, and the scene is finished after the estimation is finished.
The normal vehicle-carried inbound and outbound scenes are the normal operation scenes of the invention.
The vehicle entrance and exit scene refers to a scene that a vehicle enters and exits the parking garage, starts the vehicle when being powered on or stops the vehicle when being powered off. Under the scene that a vehicle enters and exits the parking garage and is powered on to start the vehicle, after the vehicle is powered on to activate, a driver opens and closes a door, and the vehicle sends out a zero-speed signal, namely scene identification information of the scene, because no station information exists in the scene, ramp compensation cannot be performed on weight estimation, and no passenger exists in the scene, the requirement on weight estimation precision of the scene is not high, and the condition that the vehicle does not slide on a large ramp and slides in a severe idling mode is ensured. And the power-off parking scene is special, and only a zero-speed signal can be identified. The present scenario does not require updating the weight.
In the vehicle turning-back scene of the uplink (downlink) line to the downlink (uplink) line, ramp compensation cannot be performed on weight estimation due to the fact that no station information exists in the scene, and no passenger exists in the scene, so that the requirement on weight estimation precision is not high in the scene. The zero-speed signal and the gate opening and closing signal are the scene identification information.
The in-site test scene comprises a vehicle delivery test, a repair period test, a fault repair test and the like, generally AW0 is unloaded, but the possibility of loading to AW2 or AW3 for heavy load is not eliminated, so that the scene needs weight estimation, and the occurrence of vehicle sliding and serious idle sliding faults during ramp test in the site is prevented. The scene identification information is a zero-speed parking signal, but no requirement is made for opening and closing the door, the weight estimation program is not avoided from being triggered by mistake, a driver is required to forcibly open and close the door after the parking is tested and powered on in the yard, and the slope cannot be compensated due to no station information, so that the vehicle cannot slide under a large slope in the yard. The zero-speed signal and the gate opening and closing signal are the scene identification information.
Under the condition of fault rescue, the weight of the vehicle is updated such as midway passenger clearing, rescue personnel and equipment getting on and off the vehicle, and the like. The zero-speed signal, the door opening and closing signal and the fault detection instruction signal are the scene identification information.
Through the analysis, the scene identification information of the vehicle entrance and exit section scene, the vehicle turning-back scene of the uplink (downlink) to downlink (uplink) line, the in-site test scene and the fault rescue scene is basically consistent, the requirement on the precision of the weight is not high, and only the successful ramp starting and the serious idle sliding do not occur, so that the ramp can be generally called as a special operation scene, and the processing modes are consistent.
Under the condition that the scene is explained above, the online estimation method for the urban light rail vehicle load based on scene identification provided by the embodiment of the invention is shown in fig. 3, and the main steps include:
if the initial state of the vehicle is a non-powered state, after receiving a vehicle power-on activation signal, firstly, a vehicle-mounted signal system (ATC), a CCU, a TCU and a BCU are powered on and initialized according to the vehicle power-on activation signal, and the vehicle weight information stored in the FLASH memory is initialized to the vehicle weight information stored in the FLASH memory before the last power failure.
The train TCU and BCU perform vehicle system operation control based on the received vehicle weight information.
specifically, the scene identification information of each scene has already been described in the above description of the scene, and is not described herein again.
The method judges whether the working scene of the vehicle is a normal operation scene or a special operation scene by identifying the scene identification information, and then adopts different estimation processing mechanisms aiming at different scenes.
The estimation processing of the invention can be carried out in the CCU or the TCU, and the finally obtained vehicle weight updating data is stored in the FLASH memory of the CCU or the TCU which is nonvolatile in power failure, so that the vehicle can obtain the vehicle weight data stored for the last time in the first time after being electrified, and the weight deviation is prevented from being overlarge after the vehicle is restarted in the midway of power failure.
the first estimation process is as shown in fig. 4, and specifically includes:
The static parameters of the train include, but are not limited to, the number of axles, the number of cars of the train, the cross-sectional area of the train, the coefficient of gyration mass, and the like.
For ease of understanding, a specific implementation is described. It should be noted that the parameters used in the estimation process are only data used in a preferred embodiment of the present invention, and are not intended to limit the scope of the present invention. Those skilled in the art can change various specific parameters according to actual working conditions and test conditions.
The method comprises the following steps of estimating the vehicle weight by using the dynamic characteristics of the rail vehicle, and calculating the vehicle load by using Newton mechanics knowledge:
m x a ═ F-W (formula 1)
Wherein M is the inertia mass of the vehicle containing the moment of inertia of the bogie running gear, a is the acceleration of the vehicle, F is the traction of the train, and W is the resistance of the vehicle.
Inertial mass is the dynamic mass of the vehicle when in motion: m ═ k 1M (equation 2)
Where k1 is the rotating mass coefficient and m is the static mass (kg) of the vehicle. The gyration mass coefficient is basically a definite value after the vehicle is shaped, and is determined through a dynamics related test, and if a vehicle manufacturer can not provide related parameters, an empirical value k1 is replaced by 1.1.
The vehicle resistance comprises basic resistance, starting resistance and ramp resistance.
The basic resistance adopts a standard Davis resistance formula:
W0＝6.4*m0+130*n+0.14m*V+[0.046+0.0065*(N-1)]*A*V2(formula 3)
Wherein m is the static mass (1000 kg) of the vehicle, N is the number of axles, V is the current speed (km/h) of the vehicle, N is the number of vehicles, A is the cross section area (m) of the train2)。
Train starting resistance:
w1 ═ k2 × m (V ═ 2 km/h to 3km/h) (formula 4)
Wherein, k2 is the train starting resistance coefficient, and generally k2 is 5 × 9.8 × 10-3kN/t. The starting resistance is the resistance for overcoming the static friction force and generally disappears when the vehicle speed is 2-3 km/h.
Ramp resistance:
w2 ═ m × g × sin θ (equation 5)
Where θ is the ramp angle, and because the angle is small, equation 5 can be approximated as:
w2 ═ m × g × θ (equation 6)
The maximum slope of a common urban rail line is 35 per thousand, and the maximum slope of a low-floor line can reach 60 per thousand.
The following can be solved from equation 1 to equation 6:
m a ═ F-W0-W1-W2 (formula 7)
Further simplification can be achieved:
m＝(F-130*n-([0.046+0.0065*(N-1)]*A*V2) /(k1 a +6.4+ 0.14V + k2+ g θ) (equation 8)
Equation 8 is an ideal equation for vehicle weight estimation, but due to non-linearity of vehicle dynamics and parameter setting errors, the accuracy is affected. Therefore, the problems of vehicle nonlinearity and parameter errors, the problem that starting resistance cannot be accurately calculated and the problem that the influence of the traction force establishment stage of the vehicle need to be solved.
The starting resistance cannot be accurately calculated and the vehicle traction build-up phase affects. The resistance curves of the basic resistance and the starting resistance are shown in figure 5, and the train traction is shown in figure 6:
as can be seen from FIG. 5, the vehicle resistance is mainly influenced by the starting resistance (2-3 km/h) in the low-speed section, the starting resistance disappearance in the high-speed section is basically linear with the vehicle speed, and the starting time of 0-4 km/h is about 3 s. In addition, as can be seen from fig. 6, the traction force non-establishment phase is gradually increased, and the traction force complete establishment time t1 is about 1.5 s. In order to eliminate the influence of the starting resistance and the unstable traction stage, the calculated starting point of the train weight is V >3 km/h. Equation 8 is modified as:
m＝(F-130*n-([0.046+0.0065*(N-1)]*A*V2))/(k1*a+6.4+0.14*V+g*θ)
m1 (equation 9)
Wherein m1 is a theoretical estimated value of the weight of the vehicle in a normal operation scene, F is the traction of the train, N is the number of axles, V is the speed of the train, N is the number of carriages of the train, A is the cross section of the train, a is the acceleration, k1 is the rotation mass coefficient, g is the gravity acceleration, and theta is the ramp angle.
Considering the effect due to non-linear deviations, the formula is further expressed as:
m＝m1+f(m1)；
where m is the last estimated vehicle weight, m, f, which will be used to replace the vehicle weight recorded in FLASH memory before(m1)Fitting a function for a least square method under a normal operation scene.
Because the nonlinearity and the parameter accuracy of the vehicle are deviated, a least square method is adopted to fit the function f(m1)Carry out compensation of(m1)In order to calculate the deviation function between the weight m1 and the actual weighing, when the vehicle leaves a factory and is tested, the deviation function is respectively loaded to AW0-AW3 loads to calibrate a calculation formula, so that a compensation function f is obtained(m1)In order to avoid occupying too much CPU time, the least square method does not exceed 3 orders at most.
The acceleration a in the above formula is calculated according to the first sampling time, the second sampling time, the first vehicle speed and the second vehicle speed. That is, according to the time difference between the CCU samples t1 and t2, by equation 11: a ═ t2-t1)/(V(t2)-V(t1)。
In practical implementation, the acceleration is calculated by selecting between 3Km/h and 5Km/h because the phase of static friction resistance needing to be overcome is considered as the vehicle speed from rest to 3Km during the acceleration of the vehicle. I.e. V(t1)＝3Km/h，V(t2)5 Km/h. Starting timing when the vehicle enters 3Km/h, and stopping timing when the vehicle reaches 5Km/h, and obtaining the acceleration of the vehicle according to the time difference and the speed difference.
Through the processes, vehicle weight updating data under a normal operation scene can be obtained.
the process of the second estimation process is as the steps in fig. 7, and specifically includes:
In step 142, the traction level parameter of the CCU is set to 100% traction level and the load parameter is set to the rated load AW 2.
The method executed in the steps 141-146 is the same as that executed in the steps 131-136, and will not be further described.
Different from the normal operation scene, under the special operation scene, the estimation system can not obtain the ramp information so as to ensure that the vehicle is inThe calculated weight at the maximum downhill meets the requirement of the maximum uphill starting or rescue, and the preset maximum ramp angle theta is set0-35% o, add virtual weight without identifying current ramp information, so equation 9 above is transformed to:
m＝(F-130*n-([0.046+0.0065*(N-1)]*A*V2))/(k1*a+6.4+0.14*V-0.035*g)
m2 (equation 12)
Considering the effect due to non-linear deviations, the formula is further expressed as:
m＝m2+f(m2)
wherein m2 is a theoretical estimated value of vehicle weight under a special operation scene, f(m2)Fitting a function for a least square method under a special operation scene. Least squares fitting function f(m2)The formula m is m2+ f through actual measured data under different load conditions according to normal operation scenes AW0-AW3 before the vehicle leaves the factory(m2)And (5) carrying out calibration acquisition.
And 150, writing the vehicle weight updating data into the FLASH memory to update the vehicle weight information stored in the FLASH memory.
Through the processes, the vehicle application scene is deeply analyzed, a load calculation method based on scene recognition is creatively provided according to the cooperative relationship between the vehicle operation and the vehicle load in different scenes, the scene identification information is extracted, and the vehicle operation under the condition of a load sensor is maximally simulated; the dynamic characteristics of the light rail vehicle are fully researched, the characteristics, the influence range and the influence depth of the nonlinear resistance are analyzed, and a weight calculation formula which is reversely deduced according to the dynamic characteristics is provided under the condition that various influence factors are comprehensively considered; in the calculation process, according to the vehicle resistance characteristic and the traction force establishment characteristic, an optimal entry point of data acquisition and vehicle weight calculation is provided, a measurement interval is located in a stable linear interval, the measurement precision is reduced by considering vehicle nonlinearity and parameter deviation, a least square method nonlinear error compensation method based on actual load calibration is provided, vehicle operation under the condition that different processing strategies are adopted to simulate a load sensor is realized corresponding to different scenes, the daily working condition of a rail transit system is basically covered, the accuracy of the obtained result is high, and powerful guarantee is provided for safe and stable operation of a train.
Those of skill would further appreciate that the various illustrative components and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, a software module executed by a processor, or a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. An urban light rail vehicle load online estimation method based on scene recognition is characterized by comprising the following steps:
the method comprises the steps that a vehicle network controller CCU acquires vehicle weight information stored in a FLASH memory FLASH memory of a train vehicle, and issues the vehicle weight information to a vehicle traction controller TCU and a vehicle braking system controller BCU;
the CCU detects scene identification information in real time, and determines the current scene information of the vehicle according to the detected scene identification information;
when the current scene information of the vehicle is a normal operation scene, executing first estimation processing to obtain vehicle weight updating data;
when the current scene information of the vehicle is a special operation scene, executing second estimation processing to obtain vehicle weight updating data;
and writing the vehicle weight updating data into a FLASH memory for updating the vehicle weight information stored in the FLASH memory.
2. The on-line urban light rail vehicle load estimation method according to claim 1, wherein the CCU obtains vehicle weight information stored in a FLASH memory of the vehicle and issues the vehicle weight information to the TCU and the BCU, and the method further comprises:
receiving a vehicle power-on activation signal;
and carrying out power-on initialization on an ATC (automatic train control), a CCU (central control unit), a TCU (remote control unit) and a BCU (binary-coded control unit) according to the vehicle power-on activation signal, and initializing the vehicle weight information stored in the FLASH memory into the vehicle weight information stored in the FLASH memory before the last power failure.
3. The on-line urban light rail vehicle load estimation method according to claim 1, wherein the first estimation process comprises:
determining a door closing state and a traction instruction signal receiving state of the vehicle;
when the door of the vehicle is closed and a traction command signal is received, the traction level parameter of the CCU is set to be 100% traction level, and the load parameter is set to be rated load AW 2;
detecting whether the current vehicle speed exceeds a first set threshold value;
when the current vehicle speed exceeds a first set threshold value, the CCU acquires train traction according to a set sampling frequency, and records the time when the vehicle speed reaches a first vehicle speed as first sampling time;
when the current vehicle speed exceeds a second set threshold value, ending the collection of the train traction force, and recording the time when the vehicle speed reaches a second vehicle speed as second sampling time;
calculating the acceleration of the vehicle in a time interval from the first sampling time to the second sampling time according to the first sampling time, the second sampling time, the first vehicle speed and the second vehicle speed, and taking the average value of the acquired train traction as the value of the train traction;
calculating theoretical estimated value of vehicle weight according to static parameters of the train, train speed, acceleration and ramp angle, and fitting function f by using least square method(m1)And compensating the theoretical estimated vehicle weight value to obtain the vehicle weight.
4. The on-line urban light rail vehicle load estimation method according to claim 3, wherein the theoretical estimation of vehicle weight calculated from train static parameters, speed, acceleration and ramp angle is specifically:
m1＝(F-130*n-([0.046+0.0065*(N-1)]*A*V2))/(k1*a+6.4+0.14*V+g*θ)
wherein m1 is the theoretical estimated value of the weight of the vehicle, F is the traction of the train, N is the number of axles, V is the speed of the train, N is the number of carriages of the train, A is the cross-sectional area of the train, a is the acceleration, k1 is the coefficient of revolution mass, g is the acceleration of gravity, and theta is the ramp angle.
5. The on-line estimation method for the urban light rail vehicle load according to claim 4, wherein the theoretical estimated value of the vehicle weight is compensated by a least square fitting function, and the obtained vehicle weight specifically comprises:
m＝m1+f(m1)；
wherein m is the vehicle weight, f(m1)Fitting a function for a least square method under a normal operation scene.
6. The on-line urban light rail vehicle load estimation method according to claim 5, characterized in thatThen, the least square fitting function is used for matching the formula m-m 1+ f through actual measured data under different loads according to normal operation scenes AW0-AW3 before the vehicle leaves the factory(m1)And (5) carrying out calibration acquisition.
7. The on-line urban light rail vehicle load estimation method according to claim 1, wherein the second estimation process comprises:
determining a door closing state and a traction instruction signal receiving state of the vehicle;
when the door of the vehicle is closed and a traction command signal is received, the traction level parameter of the CCU is set to be 100% traction level, and the load parameter is set to be rated load AW 2;
detecting whether the current vehicle speed exceeds a first set threshold value;
when the current vehicle speed exceeds a first set threshold value, the CCU acquires train traction according to a set sampling frequency, and records the time when the vehicle speed reaches a first vehicle speed as first sampling time;
when the current vehicle speed exceeds a second set threshold value, ending the collection of the train traction force, and recording the time when the vehicle speed reaches a second vehicle speed as second sampling time;
calculating the acceleration of the vehicle in a time interval from the first sampling time to the second sampling time according to the first sampling time, the second sampling time, the first vehicle speed and the second vehicle speed, and taking the average value of the acquired train traction as the value of the train traction;
calculating a theoretical estimated value of the weight of the vehicle according to the static parameters of the train, the speed and the acceleration of the train and a preset maximum ramp angle, and fitting a function f by a least square method(m2)And compensating the theoretical estimated vehicle weight value to obtain the vehicle weight.
8. The on-line urban light rail vehicle load estimation method according to claim 7, wherein the theoretical estimation of vehicle weight calculated from the train static parameters, speed, acceleration and the preset maximum ramp angle is specifically:
m2＝(F-130*n-([0.046+0.0065*(N-1)]*A*V2))/(k1*a+6.4+0.14*V-θ0*g)
wherein m1 is the theoretical estimated value of the weight of the vehicle, F is the traction force of the train, N is the number of axles, V is the speed of the train, N is the number of carriages of the train, A is the section area of the train, a is the acceleration, k1 is the coefficient of revolution mass, g is the acceleration of gravity, theta0Is a preset maximum ramp angle.
9. The on-line estimation method for the urban light rail vehicle load according to claim 8, wherein the theoretical estimated value of the vehicle weight is compensated by a least square fitting function, and the obtained vehicle weight specifically comprises:
m＝m2+f(m2)；
wherein m is the vehicle weight, f(m2)Fitting a function for a least square method under a special operation scene.
10. The on-line urban light rail vehicle load estimation method according to claim 9, wherein the least square fitting function f(m2)The formula m is m2+ f through actual measured data under different load conditions according to normal operation scenes AW0-AW3 before the vehicle leaves the factory(m2)And (5) carrying out calibration acquisition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110644148.9A CN113390495B (en) | 2021-06-09 | 2021-06-09 | Scene recognition-based urban light rail vehicle load online estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110644148.9A CN113390495B (en) | 2021-06-09 | 2021-06-09 | Scene recognition-based urban light rail vehicle load online estimation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113390495A true CN113390495A (en) | 2021-09-14 |
| CN113390495B CN113390495B (en) | 2022-04-29 |
Family
ID=77620087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110644148.9A Active CN113390495B (en) | 2021-06-09 | 2021-06-09 | Scene recognition-based urban light rail vehicle load online estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113390495B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113859308A (en) * | 2021-09-23 | 2021-12-31 | 浙江众合科技股份有限公司 | Framework and information fusion system of vehicle-mounted ATO (automatic transfer optical) module and BCU (binary coded decimal) controller |
| CN113990514A (en) * | 2021-10-28 | 2022-01-28 | 平安医疗健康管理股份有限公司 | Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0374960A2 (en) * | 1988-12-23 | 1990-06-27 | Hitachi, Ltd. | Control equipment of electric rolling stock |
| US20160332633A1 (en) * | 2013-07-11 | 2016-11-17 | C.R.F. Societa' Consortile Per Azioni | Automotive control unit programmed to estimate road slope and vehicle mass, vehicle with such a control unit and corresponding program product |
| CN106444421A (en) * | 2016-09-29 | 2017-02-22 | 南京理工大学 | Train traction-brake controller system of urban rail transit and working method of system |
| CN108944935A (en) * | 2018-05-31 | 2018-12-07 | 重庆大学 | A kind of car mass and road grade estimation method considering parameter coupled relation |
| CN112507459A (en) * | 2020-12-11 | 2021-03-16 | 交控科技股份有限公司 | Indoor test method and system for rail transit |
-
2021
- 2021-06-09 CN CN202110644148.9A patent/CN113390495B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0374960A2 (en) * | 1988-12-23 | 1990-06-27 | Hitachi, Ltd. | Control equipment of electric rolling stock |
| US20160332633A1 (en) * | 2013-07-11 | 2016-11-17 | C.R.F. Societa' Consortile Per Azioni | Automotive control unit programmed to estimate road slope and vehicle mass, vehicle with such a control unit and corresponding program product |
| CN106444421A (en) * | 2016-09-29 | 2017-02-22 | 南京理工大学 | Train traction-brake controller system of urban rail transit and working method of system |
| CN108944935A (en) * | 2018-05-31 | 2018-12-07 | 重庆大学 | A kind of car mass and road grade estimation method considering parameter coupled relation |
| CN112507459A (en) * | 2020-12-11 | 2021-03-16 | 交控科技股份有限公司 | Indoor test method and system for rail transit |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113859308A (en) * | 2021-09-23 | 2021-12-31 | 浙江众合科技股份有限公司 | Framework and information fusion system of vehicle-mounted ATO (automatic transfer optical) module and BCU (binary coded decimal) controller |
| CN113990514A (en) * | 2021-10-28 | 2022-01-28 | 平安医疗健康管理股份有限公司 | Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113390495B (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113390495B (en) | Scene recognition-based urban light rail vehicle load online estimation method | |
| CN111994129B (en) | Antiskid control method and system for wheel-rail train | |
| US9150209B2 (en) | System and method for monitoring braking effort | |
| MX2010011321A (en) | Method and system for determining brake shoe effectiveness. | |
| CN109000935B (en) | Method for judging performance of new energy automobile brake system | |
| US9580052B2 (en) | Method for decelerating a vehicle | |
| US10933896B2 (en) | Method and apparatus for determining changes in the longitudinal dynamic behavior of a railway vehicle | |
| CN109649436B (en) | On-line evaluation method and device for comfort index of high-speed railway automatic driving system | |
| CN112334375A (en) | Method for predicting an adhesion coefficient in a contact point between a wheel of a rail vehicle and a rail, method for improving an adhesion coefficient in a contact point between a wheel of a rail vehicle and a rail, and device for carrying out said method | |
| KR20130013133A (en) | Prediction methods for derailment of the wheels using the external force acted on the wheelset | |
| CN109308748A (en) | A kind of method and system saving car accident reason | |
| AU2014204416A1 (en) | System and method for monitoring braking effort | |
| CA3118032A1 (en) | Method and device for detecting a derailed state of a rail vehicle | |
| CN104228886B (en) | Urban rail transit vehicles day-to-day operation state measurement strategy | |
| Saraiev et al. | A Mathematical Model of the Braking Dynamics of a Car | |
| JP2000211487A (en) | Skid control device for rolling stock | |
| WO2006136648A1 (en) | A method and an apparatus for collecting information on the mass of load of a vehicle in heavy road traffic | |
| CN105946578A (en) | Accelerator pedal control method and device and vehicle | |
| Kim et al. | Precise stop control and experimental validation for metro train overcoming delays and nonlinearities | |
| KR20110124054A (en) | Railway vehicle load measuring device and method | |
| Pichlík et al. | Extended Kalman filter utilization for a railway traction vehicle slip control | |
| CN116215600B (en) | Method and device for controlling standard alignment and parking of railway vehicle and computer storage medium | |
| CN113870580A (en) | Overspeed detection method and device for truck, truck vehicle and truck system | |
| CN113453968B (en) | Method for detecting braking action of railway vehicle and emergency braking method of railway vehicle | |
| CN109238752B (en) | Low-floor vehicle train speed diagnosis method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |