CN104627168A - Plug-in hybrid power bus dynamic logic threshold energy management method based on road condition model - Google Patents

Abstract

The invention provides a plug-in hybrid power bus dynamic logic threshold energy management method based on a road condition model. The method specifically includes the steps that first, the vehicle travelling road condition model is established according to historical traffic information of a bus and a neural network; second, the changing curve of the driving speed in the future and the road slope can be obtained according to departure time, a line and the road condition model; third, energy needed for travelling of the bus is calculated on the basis of the curve and vehicle parameters, and a vehicle energy management logic threshold parameter is dynamically adjusted according to the difference value between the energy needed for travelling of the bus and energy capable of being provided by a battery; fourth, a vehicle controller selects a work mode according to the travelling parameters of the bus and the logic threshold parameter. The logic threshold parameter can be dynamically adjusted on the basis of the road condition model, it can be guaranteed that an energy management strategy can adapt to different travelling road conditions, and fuel economy of the plug-in hybrid power bus is further improved.

Description

A kind of plug-in hybrid bus dynamic logic thresholding energy management method based on road conditions model

Technical field

The present invention relates to the energy management method of plug-in hybrid bus, particularly relating to a kind of plug-in hybrid bus multi power source optimal control based on travelling road conditions model, belonging to hybrid-power bus control technology field.

Background technology

Plug-in hybrid bus is the novel energy-saving environment-friendly vehicle recently derived from conventional hybrid bus basis, have that use cost is low, discharge less, the plurality of advantages such as electrical network degree of utilization is high, battery can not only be relied on separately to travel longer distance, and can work as traditional full hybrid vehicle when needed, become one of the preferred plan to final clean energy vehicle transition.

Plug-in hybrid-power automobile is the perfect adaptation of conventional fuel oil automobile and pure electric automobile, in conjunction with tie be then energy management strategies.The energy management strategies proposed mainly comprises static logic thresholding strategy, fuzzy logic profile, instantaneous optimization strategy and global optimization strategy four class.Wherein, fuzzy logic profile, instantaneous optimization strategy and global optimization strategy due to operand large, its practical application of the problems affect such as usage condition is harsh.At present, can real vehicle run plug-in hybrid-power automobile often adopt static logic thresholding energy management strategies, it realizes comparatively simple, but the threshold parameter related in its energy management mostly is static value, can not play the optimal fuel economy of vehicle when road condition change.Especially plug-in hybrid bus often operates in the bustling section in urban district, for making full use of the cheap energy obtained from electrical network, ensure the power savings advantages that all can play multi power source in whole travelled distance, emissions reduction and fuel oil consumption, must consider the energy distribution problem under different traveling road conditions.

For bus, its circulation line is fixed, and is easy to the road conditions model setting up vehicle operating.Therefore, in the multi power source energy distribution of plug-in hybrid bus, based on road conditions model dynamic conditioning logic threshold parameter, be conducive to energy management strategies and adapt to different traveling road conditions, further lifting vehicle fuel economy.

Summary of the invention

The object of the invention is the energy management method proposing a kind of plug-in hybrid bus, the shortcoming travelling road condition change cannot be adapted to for static logic thresholding strategy, traveling road conditions model is set up according to the vehicle running history data that GPS gathers, and adjust logic threshold parameter accordingly, realize the energy-optimised distribution to hybrid power system, the fuel economy of further lifting vehicle.

The plug-in hybrid bus dynamic logic thresholding energy management method based on road conditions model that the present invention proposes comprises the steps:

1, for certain circuit bus, obtain bus running history road condition data and set up this public bus network road conditions model

(1) vehicle-mounted GPS equipment is utilized to obtain the time of departure (what day/hour/point) of certain day, with one second for the data such as longitude and latitude, sea level elevation in the sampling frequency real time recording vehicle service time on the same day;

(2) longitude and latitude is converted into the Gauss plane coordinate of 84-WGS ellipsoid, time m-latitude and longitude information be converted into time m-speed information,--velocity curve and time--the sea level elevation curve that obtains the time of every day, according to vehicle acceleration restriction, exception handling is carried out to this curve, reject or revise abnormal data, identification add obliterated data.

(3) for vehicle position during consideration road conditions modeling is on the impact of the speed of a motor vehicle, based on the time--velocity curve calculates the corresponding moment speed of adjacent moment Distance geometry and obtains Vehicle-Miles of Travel-speed curves, utilizes the time--and velocity curve calculates the corresponding moment sea level elevation of adjacent moment Distance geometry and obtains Vehicle-Miles of Travel-vehicle sea level elevation curve.

(4) based on step (1)--(3) obtain the Vehicle-Miles of Travel-speed curves of nearly one month every day of bus, Vehicle-Miles of Travel-vehicle sea level elevation curve, calculate the average travel of every day;

(5) based on the curve data described in the time of departure and (4), neural metwork training is utilized to obtain the road conditions model of this circuit, this network is the 3 layers of BP neural network comprising a hidden layer, input layer has 2 neurons, the corresponding time of departure, vehicle travelled distance respectively, output layer is 2 neurons, respectively corresponding car speed and vehicle sea level elevation, and the neuron number in hidden layer adopts try and cut method to determine.

Set up the road conditions model of other public bus network according to the method described above.

2, by public bus network road conditions model storage in plug-in hybrid-power automobile vehicle control device unit, when vehicle is dispatched a car, the road conditions model corresponding according to route choosing of dispatching a car, by the time of departure and average travel, the measurable vehicle travelled distance-speed curves obtained in this vehicle this working day, vehicle travelled distance-vehicle sea level elevation curve.

3, institute energy requirement is travelled for ease of calculating vehicle, vehicle travelled distance-speed curves and vehicle travelled distance-vehicle sea level elevation curve is converted to the time respectively--velocity curve and time--intensity gradient curve.

Calculate vehicle and travel institute's energy requirement, the formula of employing is:

$Q_{\mathrm{dem}}=\underset{0}{\overset{T}{\∫}}\frac{V_{\mathrm{vel}}}{3600{\mathrm{\η}}_T}\·(m\·g\·f\·\cos \mathrm{\α}+m\·g\·\sin \mathrm{\α}+\frac{C_D\·A\·V_{\mathrm{vel}}^2}{21.225}+\mathrm{\δ}\·m\·\frac{{\mathrm{dV}}_{\mathrm{vel}}}{\mathrm{dt}})$

Wherein, Q _demfor vehicle travels institute's energy requirement; T is for travelling total time; V _velfor the speed of a motor vehicle; M is complete vehicle quality; G is acceleration due to gravity; F is coefficient of rolling resistance; C _dfor air resistance coefficient; A is wind area; α is road grade; δ is vehicle correction coefficient of rotating mass; for vehicle acceleration; η _tfor vehicle drive system mechanical efficiency.

Calculating battery can for the energy exported, and the formula of employing is:

Q _out=(S _c，h-S _c，l)·U _b·C·η _b·η _m

Wherein, Q _outfor battery can provide the electric energy driving vehicle; U _bfor battery rated voltage; S _{c, h}for the power battery charged state permission work upper limit; S _{c, l}for power battery charged state allows lower work threshold; η _bfor cell discharge efficiency; η _mfor motor working efficiency.

4, the difference of energy can be provided to upgrade logic threshold relevant to hybrid power system energy management in vehicle control device according to institute's energy requirement and battery, power threshold is set to dynamic logic thresholding parameter herein, i.e. the Reasonable adjustment engine operation lower limit P when energy demand can provide energy more than battery _{e_min}with engine operation power upper limit P _{e_max}, suitably widen the work area of driving engine, otherwise reduce the work area of driving engine.Method is as follows:

Work as Q _out>=Q _demtime, do not adjust;

Work as Q _out<Q _demtime,

$P_{e\_\min }^{\&prime;}=\frac{k_1}{Q_{\mathrm{dem}}-Q_{\mathrm{out}}}{P}_{e\_\min }$

$P_{e\_\max }^{\&prime;}=\frac{Q_{\mathrm{dem}}-Q_{\mathrm{out}}}{k_2}{P}_{e\_\max }$

K ₁and k ₂for adjustment P _{e_min}and P _{e_max}the proportionality coefficient of suitable increase or minimizing; P ' _{e_min}with P ' _{e_max}for the engine operation lower limit after adjustment and engine operation power upper limit.

5, in vehicle real time execution, energy management strategies will speed up the demand power P that pedal and brake pedal signal interpretation are hybrid power system _r, and according to monitor vehicle moving velocity V, battery charge state S _cand the engine power P of setting _{e_min}and P _{e_max}, vehicle velocity V _ewith battery charge state S _{c, h}, S _{c, l}select mode of operation Deng threshold parameter, concrete grammar is as follows:

(1) when Vehicle Speed is lower than minimum vehicle velocity V _eor vehicle needs power P _rlower than P _{e_min}, and SOC S _clower work threshold S is allowed higher than state-of-charge _{c, l}time, driven separately by motor;

(2) when vehicle needs power is between engine optimization region [P _{e_min}, P _{e_max}] or as battery charge state value S _clower than S _{c, l}time, driving engine drives separately;

(3) when demand power is more than P _{e_max}and battery pack SOC S _chigher than S _{c, l}time, carry out combination drive, now engine control is operated in optimum efficiency curve, and residue driving torque is provided by motor;

(4) during braking, if battery charge state value S _cbe less than S _{c, h}, motor is reclaiming braking energy as much as possible, and remainder is consumed by mechanical brake, namely enters energy feedback pattern, if battery charge state value S _cbe greater than battery charge state upper limit S _{c, h}, then engaging friction braking mode.

The energy management method that the present invention proposes can control the horsepower output of hybrid power system, this control method realizes comparatively simple, and the change of road conditions can be adapted to, the cheap electric energy of battery storage can be made full use of in the whole travelled distance of bus, ensure that the fuel economy that plug-in hybrid bus is good.

Accompanying drawing explanation

Fig. 1: a kind of dynamic assembly configuration picture of plug-in hybrid bus

Fig. 2: schematic flow sheet of the invention process

Fig. 3: BP neural network structure figure

Fig. 4: engine efficiency diagram of curves

Fig. 5: mode of operation switching flow figure

Specific implementation method

Content of the present invention is further illustrated below in conjunction with accompanying drawing

As shown in Figure 1, plug-in hybrid bus adopts twin axle parallel connection type structure, battery can external charge, and there is accelerator pedal position sensor, brake pedal position sensor, car speed sensor, GPS module, entire car controller HCU, engine controller ICU, electric machine controller MCU, battery management unit BMU, engine controller ICU, electric machine controller MCU are connected with entire car controller HCU by CAN with battery management unit BMU.

As shown in Figure 2, a kind of plug-in hybrid bus of embodiment of the present invention energy management method, implementation step is as follows:

I, obtain bus running history road condition data based on the equipment such as GPS and set up this public bus network road conditions model.

Step 1 specifically can be divided into a few sub-steps below

(1) vehicle-mounted GPS equipment is utilized to obtain the time of departure (what day/hour/point) of certain day, with one second for the data such as longitude and latitude, sea level elevation in the sampling frequency real time recording vehicle service time on the same day;

(2) longitude and latitude is converted into the Gauss plane coordinate of 84-WGS ellipsoid, time m-latitude and longitude information be converted into time m-speed information,--velocity curve and time--the sea level elevation curve that obtains the time of every day, according to vehicle acceleration restriction, exception handling is carried out to this curve, reject or revise abnormal data, identification add obliterated data.

(3) for vehicle position during consideration road conditions modeling is on the impact of the speed of a motor vehicle, based on the time--velocity curve calculates the corresponding moment speed of adjacent moment Distance geometry and obtains Vehicle-Miles of Travel-speed curves, utilizes the time--and velocity curve calculates the corresponding moment sea level elevation of adjacent moment Distance geometry and obtains Vehicle-Miles of Travel-vehicle sea level elevation curve.

(4) based on step (1)--(3) obtain the Vehicle-Miles of Travel-speed curves of nearly one month every day of bus, Vehicle-Miles of Travel-vehicle sea level elevation curve, calculate the average travel of every day;

(5) based on the curve data described in the time of departure and (4), neural metwork training is utilized to obtain the road conditions model of this circuit, this network is comprise-3 layers of BP neural network of individual hidden layer, input layer has 2 neurons, the corresponding time of departure, vehicle travelled distance respectively, output layer is 2 neurons, respectively corresponding car speed and vehicle sea level elevation, and the neuron number in hidden layer adopts try and cut method to determine.

Set up the road conditions model of other public bus network according to the method described above.

2, by public bus network road conditions model storage in plug-in hybrid-power automobile vehicle control device unit, when vehicle is dispatched a car, when vehicle operating, HCU can automatically confirm current line according to GPS position information and use corresponding road conditions model.By the time of departure and average travel, the measurable vehicle travelled distance-speed curves obtained in this vehicle this working day, vehicle travelled distance-vehicle sea level elevation curve.

3, the curve data obtained according to step 2 upgrades logic threshold parameter.

3.1 travel institute energy requirement for ease of calculating vehicle, vehicle travelled distance-speed curves and vehicle travelled distance-vehicle sea level elevation curve is converted to the time respectively--velocity curve and time--intensity gradient curve.

Calculate vehicle and travel institute's energy requirement, the formula of employing is:

$Q_{\mathrm{dem}}=\underset{0}{\overset{T}{\&Integral;}}\frac{V_{\mathrm{vel}}}{3600{\mathrm{\&eta;}}_T}\&CenterDot;(m\&CenterDot;g\&CenterDot;f\&CenterDot;\cos \mathrm{\&alpha;}+m\&CenterDot;g\&CenterDot;\sin \mathrm{\&alpha;}+\frac{C_D\&CenterDot;A\&CenterDot;V_{\mathrm{vel}}^2}{21.225}+\mathrm{\&delta;}\&CenterDot;m\&CenterDot;\frac{{\mathrm{dV}}_{\mathrm{vel}}}{\mathrm{dt}})$

Wherein, Q _demfor vehicle travels institute's energy requirement; T is for travelling total time; V _velfor the speed of a motor vehicle; M is complete vehicle quality; G is acceleration due to gravity; F is coefficient of rolling resistance; C _dfor air resistance coefficient; A is wind area; α is road grade; δ is vehicle correction coefficient of rotating mass; for vehicle acceleration; η _tfor vehicle drive system mechanical efficiency.

3.2 calculate battery can for the energy exported, and the formula of employing is:

Q _out=(S _c，h-S _c，l)·U _b·C·η _b·η _m

Wherein, Q _outfor battery can provide the electric energy driving vehicle; U _bfor battery rated voltage; S _{c, h}for the power battery charged state permission work upper limit; S _{c, l}for power battery charged state allows lower work threshold; η _bfor cell discharge efficiency; η _mfor motor working efficiency.

3.3 can provide the difference of energy to upgrade logic threshold relevant to hybrid power system energy management in vehicle control device according to institute's energy requirement and battery, power threshold is set to dynamic logic thresholding parameter herein, i.e. the Reasonable adjustment engine operation lower limit P as shown in Figure 4 when energy demand can provide energy more than battery _{e_min}with engine operation power upper limit P _{e_max}, suitably widen the work area of driving engine, otherwise reduce the work area of driving engine.Method is as follows:

Work as Q _out>=Q _demtime, do not adjust;

Work as Q _out<Q _demtime,

$P_{e\_\min }^{\&prime;}=\frac{k_1}{Q_{\mathrm{dem}}-Q_{\mathrm{out}}}{P}_{e\_\min }$

$P_{e\_\max }^{\&prime;}=\frac{Q_{\mathrm{dem}}-Q_{\mathrm{out}}}{k_2}{P}_{e\_\max }$

K ₁and k ₂for adjustment P _{e_min}and P _{e_max}the proportionality coefficient of suitable increase or minimizing; P ' _{e_min}with P ' _{e_max}for the engine operation lower limit after adjustment and engine operation power upper limit.

4, in vehicle real time execution, energy management strategies will speed up the demand power P that pedal and brake pedal signal interpretation are hybrid power system _r, and according to the Vehicle Speed V of car speed sensor monitoring, the battery charge state S of battery management unit calculating _cand the engine power P of setting _{e_min}and P _{e_max}, vehicle velocity V _ewith battery charge state S _{c, h}, S _{c, l}select mode of operation Deng threshold parameter, as shown in Figure 5, concrete grammar is as follows:

(1) when Vehicle Speed is lower than minimum vehicle velocity V _eor vehicle needs power P _rlower than P _{e_min}, and SOC S _clower work threshold S is allowed higher than state-of-charge _{c, l}time, driven separately by motor;

(2) when vehicle needs power is between engine optimization region [P _{e_min}, P _{e_max}] or as battery charge state value S _clower than S _{c, l}time, driving engine drives separately;

(3) when demand power is more than P _{e_max}and battery pack SOC S _chigher than S _{c, l}time, carry out combination drive, now engine control is operated in optimum efficiency curve, and residue driving torque is provided by motor;

(4) during braking, if battery charge state value S _cbe less than S _{c, h}, motor is reclaiming braking energy as much as possible, and remainder is consumed by mechanical brake, namely enters energy feedback pattern, if battery charge state value S _cbe greater than battery charge state upper limit S _{c, h}, then engaging friction braking mode.

5, based on switch mode, HCU exports corresponding control information to the corresponding control unit such as ICU, MCU, BMU by CAN, and ICU controls the variablees such as throttle opening, regulates engine output; MCU adjusts output power of motor; BMU detects the parameters such as battery charge state.

Claims (4)

1. A control method for plug-in hybrid bus, is characterized in that the method comprises the steps:

   1., for certain circuit bus, obtain bus running history road condition data and set up this public bus network road conditions model

   (1) vehicle-mounted GPS equipment is utilized to obtain in nearest one month the time of departure (what day/hour/point), the data such as longitude and latitude, sea level elevation of each sampling instant;

   (2) pretreatment is carried out to acquisition data, reject or revise abnormal data, identification add obliterated data.

   (3) calculate the corresponding moment speed of adjacent moment Distance geometry based on longitude and latitude data, obtain Vehicle-Miles of Travel-speed curves, Vehicle-Miles of Travel-intensity gradient curve.

   (4) based on the curve data described in the time of departure and (3), utilize BP neural metwork training to obtain the road conditions model of this circuit, its network is input as the time of departure, vehicle travelled distance, and network exports as car speed and road grade.

   (5) the road conditions model of other public bus network is set up according to the method described above.
2. 2. by public bus network road conditions model storage in plug-in hybrid-power automobile vehicle control device unit, when vehicle is dispatched a car, the road conditions model corresponding according to route choosing of dispatching a car, by the time of departure, measurable Vehicle-Miles of Travel-the speed curves obtained in this vehicle this working day, Vehicle-Miles of Travel-vehicle sea level elevation curve.
3. 3. the dynamic conditioning of plug-in hybrid bus hybrid power system logic threshold parameter, comprises the following steps:

   (1) according to vehicle parameter and 2, curve calculation vehicle travels institute's energy requirement, and wherein adopted vehicle dynamics formula is:

   $P_{\mathrm{dem}}=\frac{V_{\mathrm{vel}}}{3600{\mathrm{\&eta;}}_T}\&CenterDot;(m\&CenterDot;g\&CenterDot;f\&CenterDot;\cos \mathrm{\&alpha;}+m\&CenterDot;g\&CenterDot;\sin \mathrm{\&alpha;}+\frac{C_D\&CenterDot;A\&CenterDot;V_{\mathrm{vel}}^2}{21.225}+\mathrm{\&delta;}\&CenterDot;m\&CenterDot;\frac{{\mathrm{dV}}_{\mathrm{vel}}}{\mathrm{dt}})$

   Wherein, P _demfor vehicle running power; V _velfor the speed of a motor vehicle; M is complete vehicle quality; G is acceleration due to gravity; F is coefficient of rolling resistance; C _dfor air resistance coefficient; A is wind area; α is road grade; δ is vehicle correction coefficient of rotating mass; for vehicle acceleration; η _tfor vehicle drive system mechanical efficiency.

   (2) calculating battery can for the energy exported, and the formula of employing is:

   Q _out=(S _c，h-S _c，l)·U _b·C·η _b·η _m

   Wherein, Q _outfor battery can provide the electric energy driving vehicle; U _bfor battery rated voltage; S _{c, h}for the power battery charged state permission work upper limit;

   S _{c, l}for power battery charged state allows lower work threshold; η _bfor cell discharge efficiency; η _mfor motor working efficiency.

   (3) difference of energy can be provided to upgrade logic threshold relevant to hybrid power system energy management in vehicle control device, i.e. the minimum operating power P of driving engine according to institute's energy requirement and battery _{e_min}with driving engine maximum service rating P _{e_max}.
4. 4., in vehicle real time execution, energy management strategies will speed up the demand power P that pedal and brake pedal signal interpretation are hybrid power system _r, and according to the Vehicle Speed V, the battery charge state S that monitor _cand the logic threshold Selecting parameter mode of operation set up.Mode switching method is as follows:

   (1) when Vehicle Speed is lower than minimum vehicle velocity V _eor vehicle needs power P _rlower than P _{e_min}, and SOC S _clower work threshold S is allowed higher than state-of-charge _{c, l}time, driven separately by motor;

   (2) when vehicle needs power is between engine optimization region [P _{e_min}, P _{e_max}] or as battery charge state value S _clower than S _{c, l}time, driving engine drives separately;

   (3) when demand power is more than P _{e_max}and battery pack SOC S _chigher than S _{c, l}time, carry out combination drive, now engine control is operated in optimum efficiency curve, and residue driving torque is provided by motor;

   (4) during braking, if battery charge state value S _cbe less than S _{c, h}, motor is reclaiming braking energy as much as possible, and remainder is consumed by mechanical brake, namely enters energy feedback pattern, if battery charge state value S _cbe greater than battery charge state upper limit S _{c, h}, then engaging friction braking mode.