CN101602364B - Quick control method applied to PHEV - Google Patents

Abstract

The invention relates to a quick DP control method applied to PHEV, which adopts a method based on energy to define an on-vehicle storage cell SOC, and derived equations of an SOC state transition equation, a relation equation of the optimal specific fuel consumption and the power of an engine, a quadratic performance index function, a quadratic optimal performance index of the specific fuel consumption, a multiple-message merged equation for calculating the total power required for vehicle running, and the like, wherein the on-vehicle storage cell SOC is defined by the method based on energy; the SOC state transition equation has a linear form; the optimal specific fuel consumption and the power of the engine have a quadratic function form; the performance index function is in a 1*1 dimensional quadratic function form; and the optimal performance index of the specific fuel consumption is in the iteration form of the quadratic analytical function of the SOC. The quick DP control method can quickly calculate the optimal power distribution ratio of an engine/motor and the optimal speed ratio of the speed threshold of a speed changer, realize the economical global optimal control of fuel, and ensure that the on-vehicle storage cell SOC is maintained in the expected operation interval.

Description

Be applied to the fast control method of plug-in hybrid electronlmobil

Technical field

The present invention relates to a kind of Plug-in Hybrid ElectricVehicle (plug-in hybrid electronlmobil of considering complicated road and running information; Be PHEV) the quick Dynamic Programming of fuel economy optimization (dynamic programming, i.e. DP) control method.

Background technology

The energy and environment are the necessary conditions that realizes sustainable development, and reducing with the dependence of eliminating oil is the urgent task of relevant global economy safety and energy security.Progress along with society; Orthodox car (being fuel with gasoline and diesel oil mainly) recoverable amount increases year by year; Make three aspect subject matters such as the energy, greenhouse gases, air quality be absorbed in vicious circle, and become the three big problems that influence automotive technology development from now on.Nowadays along with the development of electrokinetic cell, motor and Power Electronic Technique, electronlmobil has the possibility that effectively addresses the above problem.Following 10 years, the developing direction of electronlmobil was PHEV.

PHEV is meant can the electrification net, and electrokinetic cell is carried out electrically-charged mixed power electric car.PHEV has the function of the longer distance of pure motor driving; But still can be when needing with full mixed mode work; Its maximum characteristics are that hybrid electric drive system and pure electric drive system are combined; Can improve pernicious gas, greenhouse gas emission and the fuel economy of HEV greatly, improve the tractive performance and the continual mileage of pure electric automobile.Therefore PHEV is a kind of mixed power electric car drive pattern of tool development prospect, also is to one of preferred plan of final clean energy resource transition vehicle.

The technology of PHEV mainly comprises car load power system coupling and control policy, electrokinetic cell, charging Infrastructure and motor, and wherein the control policy of power system is one of gordian technique.The control policy of PHEV generally is divided into: two kinds of instantaneous optimization control policy and global optimization control policies.The instantaneous optimization control policy is not owing to receive the restriction of concrete driving cycle, and is easy to realize, therefore be widely used in the working control of PHEV, and the Rule Based control policy that proposes such as Japanese TOYOTA company, and based on the optimization strategy of LQ.Yet these control policies can't obtain the global optimization of fuel consumption under whole section path.So, there is the scholar to propose global optimization's strategy of PHEV, the total amount of fuel minimum that promptly consumes in certain state of cyclic operation with vehicle ' is the method that target is optimized.Global optimization method commonly used mainly contains the DP method.French PSA company at first is applied to HEV control with the DP method, but does not provide concrete computation process.U.S. Michigan university has proposed application DP method and has improved its rule-based HEV control policy; But the model of being set up needs simultaneously co-operative control to be carried out in the torque of driving engine (or electrical motor) and the gear transmitting ratio of change speed gear box, belongs to the DP control method (shown in Figure 7) of two dimension.Therefore to obtain the optimisation strategy of one section driving path, need expend the surprising iterative computation time, be difficult to realize the real-time application of this method, generally can only be used for method control Evaluation on effect.U.S. Ohio state university has proposed the Equivalent Consumption Minimization strategy based on the DP method; With the electric energy equivalence is fuel consumption; And, still do not solve the big problem of DP method calculated amount as optimizing the performance figure CONTROLLER DESIGN.French PSA company has proposed the Pantryagin control policy for the real-time of implementation method, yet in the time of the Hanmilton of method for solving equation and judgement boundary condition, needs certain experience and spend a large amount of computing times in earlier stage.U.S. Clemson university has proposed Model Predictive PowerManagement strategy, has realized the conversion between multiple optimisation strategy, but has not had the real-time problem of solution equally.U.S. Wisconsin university-Milwaukee branch school has proposed Trip ModelBased control policy; Through the driving path section is handled; Make certain minimizing has been arranged computing time; But its optimal control results can and produce bigger error along with the section processing, and work of treatment in the early stage amount of section is huge, and commonality is poor.

Existing achievement in research shows that global optimization and real-time are to weigh two good and bad key indexs of PHEV power system control policy.In view of above-mentioned problem, patent of the present invention proposes the quick PHEV control policy with global optimization's performance around the thought of " according to the road state of cyclic operation, optimizing power division ".

Summary of the invention

The objective of the invention is to; Develop a kind of fast control method that is applied to the plug-in hybrid electronlmobil; Fuel economy global optimum to realize driving engine among the PHEV turns to target; Utilize the optimum fuel consumption rate figure of driving engine, propose the functional relation of engine optimum fuel consumption rate and power, provide Vehicular accumulator cell SOC define method on this basis based on energy; Finally derive the engine/motor power-division ratios optimal policy with analytic function form and the alternative manner of fuel consumption rate optimal performance index; And further utilize the optimum fuel consumption rate figure of driving engine, obtain the relation curve of engine optimum working speed and power, the counter thus optimum speed ratio of releasing the speed changer gear.That is,

The present invention adopts following technological means to realize::

A kind of quick DP control method that is applied to PHEV, PHEV power drive system possess driving engine, electrical motor, Vehicular accumulator cell and have the automatic transmission with hydraulic torque converter of a plurality of gears; This control method adopts the method definition Vehicular accumulator cell SOC based on energy; Derive the relation equation that comprises SOC state transition equation, engine optimum fuel consumption rate and power, the required gross horsepower accounting equation of performance index function, quadratic form fuel consumption rate optimal performance index and many information fusion running car of quadratic form; Described optimal policy is that the linearity of SOC is resolved the function iteration form:

u ^*(k)＝η ₁(k)·SOC(k)+η ₂(k)

Wherein, parameter η ₁(k), η ₂(k) iterative relation is following:

${\mathrm{\η}}_1\left(k\right)=-\frac{{\mathrm{\ζ}}_1(k+1){\mathrm{\μ}}_2}{{\mathrm{\δ}}_1+{\mathrm{\ζ}}_1(k+1){\mathrm{\μ}}_2^2}$

${\mathrm{\η}}_2\left(k\right)=-\frac{{\mathrm{\δ}}_2+{2\mathrm{\ζ}}_1(k+1){\mathrm{\μ}}_1{\mathrm{\μ}}_2+{\mathrm{\ζ}}_2(k+1){\mathrm{\μ}}_2}{{\mathrm{\δ}}_1+{\mathrm{\ζ}}_1(k+1){\mathrm{\μ}}_2^2}$

${\mathrm{\μ}}_1=-\frac{\mathrm{\Δt}}{E_p}{P}_{\mathrm{req}}\left(k\right)$

${\mathrm{\μ}}_2=\frac{\mathrm{\Δt}}{E_p}{P}_{\mathrm{req}}\left(k\right)$

${\mathrm{\ζ}}_1\left(k\right)={\mathrm{\δ}}_1\·{\mathrm{\η}}_1^2\left(k\right)+{\mathrm{\ζ}}_1(k+1)\·{(1+{\mathrm{\μ}}_2\·{\mathrm{\η}}_1\left(k\right))}^2$

ζ ₂(k)＝2δ ₁·η ₁(k)·η ₂(k)+δ ₂·η ₁(k)+ζ ₂(k+1)·(1+μ ₂·η ₁(k))+2ζ ₁(k+1)·(1+μ ₂·η ₁(k))·(μ ₁+μ ₂·η ₂(k))

${\mathrm{\ζ}}_3\left(k\right)={\mathrm{\δ}}_1\·{\mathrm{\η}}_2^2\left(k\right)+{\mathrm{\δ}}_2\·{\mathrm{\η}}_2\left(k\right)+{\mathrm{\δ}}_3+{\mathrm{\ζ}}_1(k+1)\·{({\mathrm{\μ}}_1+{\mathrm{\μ}}_2\·{\mathrm{\η}}_2\left(k\right))}^2$

$+{\mathrm{\ζ}}_2(k+1)\·({\mathrm{\μ}}_1+{\mathrm{\μ}}_2\·{\mathrm{\η}}_2\left(k\right))+{\mathrm{\ζ}}_3(k+1)$

That is δ, ₁=α ₁Δ t, δ ₂=α ₂Δ t, δ ₃=α ₃Δ t; α ₁, α ₂, α ₃Be the match constant coefficient; Δ t is the sampling time; E _pTotal electric energy when being full of for storage battery; 1≤k≤N; Function is input as the required gross horsepower P of running car _ReqAnd the initial sum end of a period state of battery SOC (k); The optimal power allocation that is output as engine/motor compares u ^*(k).

Aforesaid Vehicular accumulator cell SOC adopts the define method based on energy:

$\mathrm{SOC}\left(k\right)=\frac{E\left(k\right)}{E_p}$

Wherein, E (k) is a battery k dump energy constantly;

Provide based on the SOC of energy and SOC based on electric weight _QBetween transformational relation:

$\mathrm{SOC}\left(k\right)=\frac{{Q_p}^2}{E_p\·C}{\mathrm{SOC}}_Q^2\left(k\right)$

Wherein, C is the electric capacity of representative capacity of cell; Q _pTotal electric weight when being full of for storage battery.

Aforesaid derived equation comprises:

SOC state transition equation with linear forms;

Engine optimum fuel consumption rate and power relation equation with Quadratic Function Optimization form;

Performance index function with Quadratic Function Optimization form of 1 * 1 dimension;

Fuel consumption rate optimal performance index with SOC secondary analytic function iteration form;

Consider the required gross horsepower method of calculating of running car of many information fusion;

Wherein, said SOC state transition equation has linear forms:

SOC(k+1)-SOC(k)＝μ ₁+μ ₂·u(k)

Wherein, the input control variable does

P _Req(k) be the required gross horsepower of running car; P _Ice(k) be engine power;

Said engine optimum fuel consumption rate and power have the Quadratic Function Optimization form:

fuel(k)＝δ ₁·P _ice(k) ²+δ ₂·P _ice(k)+δ ₃

Wherein, input

is an engine power; Output fuel (k) is optimum fuel consumption rate; Function is the engine optimum fuel consumption rate that proposed and the approximate match of power relation curve;

Said performance index function is the Quadratic Function Optimization form of 1 * 1 dimension:

$J_k(\mathrm{SOC}\left(k\right),u\left(k\right))$

$=\underset{i=k}{\overset{N}{\mathrm{\Σ}}}\left[\mathrm{fuel}(P_{\mathrm{req}}\left(i\right)\·u\left(i\right))\right]+\mathrm{\β}{(\mathrm{SOC}(N+1)-{\mathrm{SOC}}_{\mathrm{des}})}^2$

Wherein, SOC converges on SOC to β for constraint ends constantly _DesThe performance figure weighted value; Having with power-division ratios u (i) is the one dimension control variable; With SOC is the one dimension state variable; With fuel consumption rate fuel (*) and SOC variable quantity two parts serves as to optimize performance figure;

The secondary analytic function iteration form that said fuel consumption rate optimal performance index is SOC:

$J_k^{*}\left(\mathrm{SOC}\left(k\right)\right)={\mathrm{\ζ}}_1\left(k\right)\·{\mathrm{SOC}}^2\left(k\right)+{\mathrm{\ζ}}_2\left(k\right)\·\mathrm{SOC}\left(k\right)+{\mathrm{\ζ}}_3\left(k\right)$

$={\mathrm{\ζ}}_1\left(k\right)\·{[\mathrm{SOC}(k+1)-\frac{\mathrm{\Δt}}{E_p}{P}_{\mathrm{req}}\left(k\right)+\frac{\mathrm{\Δt}}{E_p}{P}_{\mathrm{req}}\left(k\right)\·u\left(k\right)]}^2$

$+{\mathrm{\ζ}}_2\left(k\right)\·[\mathrm{SOC}(k+1)-\frac{\mathrm{\Δt}}{E_p}{P}_{\mathrm{req}}\left(k\right)+\frac{\mathrm{\Δt}}{E_p}{P}_{\mathrm{req}}\left(k\right)\·u\left(k\right)]+{\mathrm{\ζ}}_3\left(k\right)$

Wherein, be input as the required gross horsepower P of running car _ReqAnd the initial sum end of a period state of battery SOC (k); Be output as the fuel consumption rate optimal value of driving engine

The optimum speed ratio of speed changer gear; Be to utilize the engine optimum working speed proposed and the relation curve of power; And combine the optimum horsepower output of present engine, reverse interpolation is obtained final the comparing with current vehicle speed v (k) of engine optimum working speed

and is obtained optimum speed ratio:

$i_g^{*}\left(k\right)=\frac{{\mathrm{\ω}}_{\mathrm{ice}}^{*}\left(k\right)\·r}{v\left(k\right)\·i_m}$

Wherein, r is a tire radius; i _mBe the main reduction gear transmitting ratio;

The required gross horsepower method of calculating of the running car of said many information fusion:

$P_{\mathrm{req}}\left(k\right)$

$=[\frac{\mathrm{\ρ}}{2}{A}_f{C}_{d}g\·v^3\left(k\right)+M\left(k\right)({\mathrm{\μ}}_r+\sin \mathrm{\Θ}\left(k\right))g\·v\left(k\right)]+\frac{M\left(k\right)}{g}(1+\frac{{\mathrm{\δ}}_{\mathrm{eqm}}}{g})\·\stackrel{\·}{v}\left(k\right)\·v\left(k\right)$

Wherein, method is integrated multiple road and running information comprises speed of a motor vehicle v (k), road grade θ (k) and load-carrying mass M (k); ρ is a density of air; A _fBe the vehicle wind area; C _dBe the Reynolds coefficient; G=9.8m/s ²μ _rBe coefficient of rolling resistance;

Be acceleration/accel; δ _EqmBe equivalent moment of inertia.

The present invention compared with prior art has following remarkable advantages and beneficial effect:

The present invention is applied to the fast control method of plug-in hybrid electronlmobil; Can calculate the optimal power allocation ratio of engine/motor fast; And the optimum speed ratio of the gear of change-speed box; Realize the fuel economy global optimization control of PHEV under known driving path, and guarantee that Vehicular accumulator cell SOC maintains the expectation operation interval.

Quick DP control method of the present invention has proposed the SOC define method based on energy, has obtained the SOC state transition equation of linear forms; The Quadratic Function Optimization relation of engine optimum fuel consumption rate and power has been proposed.

The optimal policy of utilizing said linear SOC state transition equation and dimensionality reduction quadratic performance index function to derive; The linearity that shows as state SOC is resolved the function iteration form; Only need the required gross horsepower of input running car; And the initial sum end of a period state of battery SOC, can be by each moment optimal power allocation ratio of the direct iteration acquisition of linear analytic function engine/motor.Avoided existing DP control method when the optimal policy of Converse solved each moment state point; Need quantize and interpolation optimization the two-dimentional control variable of driving engine (or electrical motor) torque and Automatic Transmission lifting/lowering shelves simultaneously, cause calculated amount " dimension disaster " problem with geometric growth thus.

The fuel consumption rate optimal performance index function that utilizes said linear SOC state transition equation and dimensionality reduction quadratic performance index function to derive; Show as the secondary analytic function iteration form of state SOC; Only need the required gross horsepower of input running car; And the initial sum end of a period state of battery SOC, can obtain the fuel consumption rate optimal value of driving engine by analytic function.Avoided existing DP control method; When the fuel consumption rate optimal value of Converse solved each moment state point; Need quantize and interpolation optimization the two-dimentional control variable of driving engine (or electrical motor) torque and Automatic Transmission lifting/lowering shelves simultaneously, cause calculated amount " dimension disaster " problem with geometric growth thus.

The optimum speed ratio of the speed changer gear of this DP control method; Be according to above-mentioned optimal power allocation ratio; Calculate the optimum horsepower output of driving engine, then according to the relation curve of engine optimum working speed and power, reverse interpolation is obtained the engine optimum working speed; Through comparing, determine optimum speed ratio at last with current vehicle speed.Whole process only relates to analytic function calculating and tables look-up, and has therefore further saved computing time.

In addition, the input of this DP control method, promptly the required gross horsepower of running car has merged multiple driving, road informations such as the speed of a motor vehicle in the vehicle running path, road grade and load-carrying quality, has improved the accuracy and the universality of methods and strategies.

Description of drawings

Fig. 1 is the vehicle ' control system of expression PHEV that the present invention is suitable for and the schematic diagram of power drive system;

Fig. 2 is an expression self-changing gearbox gear-shifting characteristic map of the present invention;

Fig. 3 is the best fuel consumption rate figure of driving engine among expression the present invention;

Fig. 4 is expression one section driving path scheme drawing that comprises complicated road and running information that the present invention considered;

Fig. 5 is the scheme drawing that the expression forward is asked for engine optimum fuel consumption rate and power relation curve;

Fig. 6 is that the expression forward is asked for engine optimum working speed and power relation curve and the reverse anti-scheme drawing that pushes away optimum working speed;

Fig. 7 is the iterative process figure of the existing DP control method of expression;

Fig. 8 is the emulation comparing result of existing DP control method of expression and quick DP control method.

Nomenclature

1 chaufeur, 2 power computation module, 3 power control modules, 4 transmitting ratio control modules

5 driving engines, 6 power drive systems, 7 storage batterys, 8 motors, 9 torsion couplers

10 main reduction gears, 11 wheels, 12 vehicles

The specific embodiment

Below, the quick DP control method of embodiment of the present invention is described with reference to accompanying drawing.Fig. 1 has represented to carry the PHEV vehicle ' control system of suitable control method of the present invention and the whole associated diagram of power-transmission system; The driving intention that is input as chaufeur 1 of PHEV and driving, road information; Control system is made up of power computation module 2, power control module 3 and transmitting ratio control module 4; Power drive system is made up of driving engine 5, electrical motor 8, Vehicular accumulator cell 7, automatic transmission with hydraulic torque converter 6, torsion coupler 9 and main reduction gear 10 etc.At first, chaufeur is imported certain driving intention through throttle gate and brake pedal, and combines information such as road grade, load-carrying quality, calculates the required gross horsepower of running car.Then, according to certain criterion (like optimum fuel consumption), calculate the optimum horsepower output allocation strategy of corresponding driving engine and electrical motor by power control module 3; Calculate optimum transmitting ratio by transmitting ratio control module 4.Optimum horsepower output allocation strategy will be used to control the horsepower output of driving engine and electrical motor, and optimum transmitting ratio then is used to control the gear change of automatic transmission with hydraulic torque converter.The ultimate aim of system is the co-ordination that realizes between two kinds of propulsions source, and assurance PHEV driving engine is accomplished going of set path with the fuel consumption of minimum, makes the SOC of storage battery change within the specific limits simultaneously.

Fig. 2 representes the shift schedule curve in the Automatic Transmission control setup.Fig. 3 is the MAP figure of driving engine, comprises best fuel oil consumption rate curve.

Quick DP control method is made up of following 3 steps:

(1) the required gross horsepower of running car calculates;

(2) optimal power allocation of engine/motor is calculated than strategy and fuel consumption rate optimal value;

(3) the optimum Transmission Ratio Control of speed changer gear.

In step (1); At first consider the known initial point shown in Figure 4 and the vehicle running path of terminal point; Merge wherein complicated road and running information; The expectation road speed v, road grade θ and the load-carrying mass M that comprise chaufeur 1 calculate the k required gross horsepower P of running car constantly by power computation module shown in Figure 12 then _Req(k), computing formula is:

$P_{\mathrm{req}}\left(k\right)=f_{P_{\mathrm{req}}}(v\left(k\right),\stackrel{\·}{v}\left(k\right),M\left(k\right),\mathrm{\Θ}\left(k\right))=(F_w\left(k\right)+F_r\left(k\right)+F_c\left(k\right)+F_a\left(k\right))\·v\left(k\right)$

$=[\frac{\mathrm{\ρ}}{2}{A}_f{C}_{d}g\·v^3\left(k\right)+M\left(k\right)({\mathrm{\μ}}_r+\sin \mathrm{\Θ}\left(k\right))g\·v\left(k\right)]$

$+\frac{M\left(k\right)}{g}(1+\frac{{\mathrm{\δ}}_{\mathrm{eqm}}}{g})\·\stackrel{\·}{v}\left(k\right)\·v\left(k\right)$

Wherein, F _w(k) be frontal resistance; F _r(k) be tire drag; F _c(k) be grade resistance; F _a(k) be resistance due to acceleration; ρ is a density of air; A _fBe the vehicle wind area; C _dBe the Reynolds coefficient; G=9.8m/s ²μ _rBe coefficient of rolling resistance; Be acceleration/accel.

In step (2), the required gross horsepower P of running car that step (1) is calculated _Req(k), and initial/end of a period state of SOC will be imported into power control module 3.This module can reverse iterative computation go out k constantly the optimal power allocation of engine/motor compare u ^*(k) (being optimal policy) and fuel consumption rate optimal value

(being optimal performance index), and forward calculates the optimal trajectory of SOC.Utilize optimal power allocation to compare u ^*(k), can finally calculate the optimum horsepower output of driving engine (or electrical motor)

(or

).

Step (2) is the main program of said quick DP control method, and wherein the derivation of power control module 3 is made up of following 5 sub-steps: (2.1) are based on the SOC define method of energy; (2.2) Quadratic Function Optimization of engine optimum fuel consumption rate and power relation is set up; (2.3) the DP math modeling of PHEV is set up; (2.4) optimal power allocation of engine/motor compares u ^*(k) derivation; (2.5) fuel consumption rate optimal performance index Derivation.

In substep (2.1), battery SOC of the present invention adopts the define method based on energy, and promptly k battery SOC constantly is defined as:

$\mathrm{SOC}\left(k\right)=\frac{E\left(k\right)}{E_p}$

Wherein, E (k) is a battery k dump energy constantly; E _pTotal electric energy when being full of for storage battery.

In addition, existing battery SOC based on electric weight _QDefine method is defined as:

${\mathrm{SOC}}_Q\left(k\right)=\frac{Q\left(k\right)}{Q_p}$

Wherein, Q (k) is a battery k dump energy constantly; Q _pTotal electric weight when being full of for storage battery.

Owing to have following transformational relation between electric energy and the electric weight:

$E=\frac{Q^2}{C}$

Wherein, C is the electric capacity of representative capacity of cell.Therefore, based on the SOC of energy with based on the SOC of electric weight _QBetween can change each other:

$E_p\&CenterDot;\mathrm{SOC}\left(k\right)=\frac{{{\mathrm{<figure-callout\; id="2"\; label="Q\; p"\; filenames="DEST\_PATH\_RE-HA20172818200810247017101E00011.png,DEST\_PATH\_RE-HA20172818200810247017101E00051.png"\; state="\{\{state\}\}">Q</figure-callout>}}_p}^2\&CenterDot;{\mathrm{SOC}}_Q^2\left(k\right)}{C}$

According to SOC based on electric weight _QBattery operating range commonly used:

0.3≤SOC _Q≤0.8

By above-mentioned conversion relational expression, be convertible into following cooresponding SOC operating range commonly used, that is: based on energy

0.5477≤SOC≤0.8944

The SOC definition that proposes in the substep (2.1) is with the SOC state transition equation in the DP math modeling that is used to set up PHEV.

In substep (2.2), the best fuel oil consumption rate curve of the driving engine among Fig. 3 is with being used to confirm the engine optimum operation point under the equipower.Definite step of optimal working point is as shown in Figure 5, and with intersecting of equipower curve (long and short dash line) and best fuel oil consumption rate curve (dotted line), intersection point is the optimal working point of driving engine under the equipower.By that analogy, can generate the optimal working point of a series of expression engine optimum fuel consumption rates and power relation.

Since driving engine to wait the oil consumption curve be closed curve, and along with the increase of engine output, its fuel consumption rate reduces gradually, reaches minimum during to medium horsepower output, fuel consumption rate increases gradually afterwards.Therefore, can be similar to the Quadratic Function Optimization relation that fits between the engine optimum fuel consumption rate in the k moment and the power:

fuel(k)＝f _fuel(P _ice(k))＝[α ₁·P _ice(k) ²+α ₂·P _ice(k)+α ₃]·Δt＝δ ₁·P _ice(k) ²+δ ₂·P _ice(k)+δ ₃

Wherein, P _Ice(k) be k engine output constantly; α ₁, α ₂, α ₃Be the match constant coefficient; Δ t is the sampling time; δ ₁=α ₁Δ t, δ ₂=α ₂Δ t, δ ₃=α ₃Δ t.Dotted line is the quadratic fit curve of this function among Fig. 5.

The engine optimum fuel consumption rate that proposes in the substep (2.2) and the Quadratic Function Optimization of power are with the performance index function in the DP math modeling that is used to set up PHEV.

In substep (2.3), with the DP math modeling of setting up PHEV.At first, utilize the SOC define method that proposes in the substep (2.1), can set up the state transition equation of DP math modeling based on energy:

$\mathrm{SOC}(k+1)-\mathrm{SOC}\left(k\right)=-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{em}}\left(k\right)=-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)[1-u\left(k\right)]$

$={\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;u\left(k\right),$ k＝1，…N

Wherein, k policy definition constantly is:

$u\left(k\right)=\frac{P_{\mathrm{ice}}\left(k\right)}{P_{\mathrm{req}}\left(k\right)}$

Be the ratio of engine output and required gross horsepower;

P _Em(k) be k output power of motor constantly

Employing is based on the SOC define method of energy, and its advantage is to make the state transition equation of setting up have linear forms.

Take second place, utilize engine optimum fuel consumption rate and the power Quadratic Function Optimization relation set up in the substep (2.2), can set up the performance index function of DP math modeling:

$J_k(\mathrm{SOC}\left(k\right);u\left(k\right))=\underset{i=k}{\overset{N}{\mathrm{\&Sigma;}}}\left[\mathrm{fuel}\left(P_{\mathrm{ice}}\left(i\right)\right)\right]+\mathrm{\&beta;}{(\mathrm{SOC}(N+1)-{\mathrm{SOC}}_{\mathrm{des}})}^2$

$=\underset{i=k}{\overset{N}{\mathrm{\&Sigma;}}}\left[\mathrm{fuel}(P_{\mathrm{req}}\left(i\right)\&CenterDot;u\left(i\right))\right]+\mathrm{\&beta;}{(\mathrm{SOC}(N+1)-{\mathrm{SOC}}_{\mathrm{des}})}^2,$ k＝1，…N

Performance index function is optimized index by minimum fuel consumption rate and SOC variable quantity two parts and is formed.

The optimization of decision process promptly is a minimum value of asking for the engine fuel rate of consumption, and makes the SOC variable quantity maintain the variation range of expectation, and therefore corresponding optimal performance index function is:

$J_k^{*}\left(\mathrm{SOC}\left(k\right)\right)=\underset{\left[u\left(k\right)\right]}{\min }{J}_k(\mathrm{SOC}\left(k\right);u\left(k\right))$

Obviously, end of a period condition is:

$J_{N+1}^{*}\left(\mathrm{SOC}(N+1)\right)=\mathrm{\&beta;}{(\mathrm{SOC}(N+1)-{\mathrm{SOC}}_{\mathrm{des}})}^2$

$={\mathrm{\&zeta;}}_1(N+1)\&CenterDot;{\mathrm{SOC}}^2(N+1)+{\mathrm{\&zeta;}}_2(N+1)\&CenterDot;\mathrm{SOC}(N+1)+{\mathrm{\&zeta;}}_3(N+1)$

Wherein, SOC _DesBe the expectation value of SOC constantly that ends; SOC converges on SOC to β for constraint ends constantly _DesThe performance figure weighted value; ζ ₁(N+1)=and β, ζ ₂(N+1)=-2 β SOC _Des,

${\mathrm{\&zeta;}}_3(N+1)=\mathrm{\&beta;}\&CenterDot;{\mathrm{SOC}}_{\mathrm{des}}^2.$

Said performance index function has quadratic form, and has taken all factors into consideration fuel consumption rate and SOC variable quantity two parts optimization index.

In addition, the running state of each constituent elements in the PHEV power-transmission system needs to satisfy following equality and inequality constrain:

ω _{em_min}≤ω _em(k)≤ω _{em_max}

T _{ice_min}(ω _ice(k)，SOC(k))≤T _ice(k)≤T _{ice_max}(ω _ice(k)，SOC(k))

T _{em_min}(ω _em(k))≤T _em(k)≤T _{em_max}(ω _em(k))

SOC _min≤SOC(k)≤SOC _max

T _ice(k)+T _em(k)＝T _req(k)

The DP math modeling of substep (2.3) PHEV that sets up, the state transition equation with linear forms; The performance index function of quadratic form; One dimension control variable-engine/motor power-division ratios; One dimension state variable-SOC.

In substep (2.4), the DP math modeling of substep (2.3) PHEV that sets up compares u with the optimal power allocation of the engine/motor that is used to derive ^*(k).At first, row write out the iterative relation formula of the optimal performance index function of n-hour:

$J_N^{*}\left(\mathrm{SOC}\left(N\right)\right)=\underset{\left[u\left(N\right)\right]}{\min }\underset{i=N}{\overset{N}{\mathrm{\&Sigma;}}}[\mathrm{fuel}\left(P_{\mathrm{ice}}\left(N\right)\right)+J_{N+1}^{*}\left(\mathrm{SOC}(N+1)\right)]$

$=\underset{\left[u\left(N\right)\right]}{\min }[\mathrm{fuel}(P_{\mathrm{req}}\left(N\right)\&CenterDot;u\left(N\right))+J_{N+1}^{*}({\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;u\left(N\right)+\mathrm{SOC}\left(N\right))]$

The iterative relation formula shows as Quadratic Function Optimization, uses the differential method can ask for its extreme value to be:

$\frac{{\&PartialD;J}_N^{*}\left(\mathrm{SOC}\left(N\right)\right)}{\&PartialD;u\left(N\right)}=2{\mathrm{\&delta;}}_{1}u\left(N\right)+{\mathrm{\&delta;}}_2+{\mathrm{\&zeta;}}_2(N+1){\mathrm{\&mu;}}_2$

$+2{\mathrm{\&zeta;}}_1(N+1)({\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_{2}u\left(N\right)+\mathrm{SOC}\left(N\right)){\mathrm{\&mu;}}_2=0$

The optimal policy of deriving n-hour thus is:

$u^{*}\left(N\right)=f_{u^{*}}(\mathrm{SOC}\left(N\right),{\mathrm{\&eta;}}_1\left(N\right),{\mathrm{\&eta;}}_2\left(N\right))={\mathrm{\&eta;}}_1\left(N\right)\&CenterDot;\mathrm{SOC}\left(N\right)+{\mathrm{\&eta;}}_2\left(N\right)$

Wherein,

${\mathrm{\&eta;}}_1\left(N\right)=f_{{\mathrm{\&eta;}}_1}\left({\mathrm{\&zeta;}}_1(N+1)\right)=-\frac{{\mathrm{\&zeta;}}_1(N+1){\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(N+1){\mathrm{\&mu;}}_2^2}$

${\mathrm{\&eta;}}_2\left(N\right)=f_{{\mathrm{\&eta;}}_2}({\mathrm{\&zeta;}}_1(N+1),{\mathrm{\&zeta;}}_2(N+1))$

$=-\frac{{\mathrm{\&delta;}}_2+[{2\mathrm{\&zeta;}}_1(N+1){\mathrm{\&mu;}}_1+{\mathrm{\&zeta;}}_2(N+1)]{\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(N+1){\mathrm{\&mu;}}_2^2}.$

$\mathrm{SOC}\left(N\right)-\mathrm{SOC}(N-1)=-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{em}}(N-1)$

$=-\frac{\mathrm{\&Delta;t}}{E_p}{T}_{\mathrm{em}}(N-1){\mathrm{\&omega;}}_{\mathrm{em}}(N-1)$

$=-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}(N-1)[1-u(N-1)]$

$=-\frac{\mathrm{\&Delta;t}}{E_p}[P_{\mathrm{req}}(N-1)-T_{\mathrm{ice}}(N-1){\mathrm{\&omega;}}_{\mathrm{ice}}(N-1)]$

When satisfying the constraint condition of following equality and inequality

ω _{em_min}≤ω _em(N-1)≤ω _{em_max}

T _{ice_min}(ω _ice(N-1)，SOC(N-1))≤T _ice(N-1)≤T _{ice_max}(ω _ice(N-1)，SOC(N-1))

T _{em_min}(ω _em(N-1))≤T _em(N-1)≤T _{em_max}(ω _em(N-1))

SOC _min≤SOC(N-1)≤SOC _max

T _ice(N-1)+T _em(N-1)＝T _req(N-1)

Order

$\frac{{\&PartialD;}^2{J}_N^{*}\left(\mathrm{SOC}\left(N\right)\right)}{{\&PartialD;u}^2\left(N\right)}>0,$ Promptly ${\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(N+1){\mathrm{\&mu;}}_2^2>0$

Can guarantee optimal policy u ^*(N) be smallest point.

Repeat said process, generally obtain k constantly optimal policy (1≤k≤N):

$u^{*}\left(k\right)=f_{u^{*}}(\mathrm{SOC}\left(k\right),{\mathrm{\&eta;}}_1\left(k\right),{\mathrm{\&eta;}}_2\left(k\right))={\mathrm{\&eta;}}_1\left(k\right)\&CenterDot;\mathrm{SOC}\left(k\right)+{\mathrm{\&eta;}}_2\left(k\right)$

Wherein, parameter η ₁(k), η ₂(k) have following iterative relation:

${\mathrm{\&eta;}}_1\left(k\right)=f_{{\mathrm{\&eta;}}_1}\left({\mathrm{\&zeta;}}_1(k+1)\right)=-\frac{{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2^2}$

${\mathrm{\&eta;}}_2\left(k\right)=f_{{\mathrm{\&eta;}}_2}({\mathrm{\&zeta;}}_1(k+1),{\mathrm{\&zeta;}}_2(k+1))=-\frac{{\mathrm{\&delta;}}_2+{2\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_1{\mathrm{\&mu;}}_2+{\mathrm{\&zeta;}}_2(k+1){\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2^2}$

Finally, can calculate the k optimum horsepower output of driving engine (or electrical motor) constantly:

$P_{\mathrm{ice}}^{*}\left(k\right)=P_{\mathrm{req}}\left(k\right)\&CenterDot;u^{*}\left(k\right)$

$P_{\mathrm{em}}^{*}\left(k\right)=P_{\mathrm{req}}\left(k\right)\&CenterDot;(1-u^{*}\left(k\right))$

Recursion according to this is for guaranteeing optimal policy u ^*(k) be optimal value, the running state of each constituent elements in the PHEV power-transmission system needs to satisfy following equality and inequality constrain:

ω _{em_min}≤ω _em(k)≤ω _{em_max}

T _{ice_min}(ω _ice(k)，SOC(k))≤T _ice(k)≤T _{ice_max}(ω _ice(k)，SOC(k))

T _{em_min}(ω _em(k))≤T _em(k)≤T _{em_max}(ω _em(k))

SOC _min≤SOC(k)≤SOC _max

T _ice(k)+T _em(k)＝T _req(k)，k＝1，…N

Based on above-mentioned u ^*(k), the optimal trajectory that can forward calculates SOC:

$\mathrm{SOC}(k+1)=f_{\mathrm{SOC}}(P_{\mathrm{req}}\left(k\right),u^{*}\left(k\right),\mathrm{SOC}\left(k\right))$

$={\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;u^{*}\left(k\right)+\mathrm{SOC}\left(k\right),$

$=-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)\&CenterDot;(1-u^{*}\left(k\right))+\mathrm{SOC}\left(k\right),$ k＝1，…N

In substep (2.5); The optimal policy of substep (2.4) fuel consumption rate optimal performance index

the derivation step that will be used to derive repeats no more, directly provide its k step optimal performance index function (1≤k≤N):

$J_k^{*}\left(\mathrm{SOC}\left(k\right)\right)=f_{J_k^{*}}({\mathrm{\&zeta;}}_1\left(k\right),{\mathrm{\&zeta;}}_2\left(k\right),{\mathrm{\&zeta;}}_3\left(k\right),\mathrm{SOC}\left(k\right))$

$={\mathrm{\&zeta;}}_1\left(k\right)\&CenterDot;{\mathrm{SOC}}^2\left(k\right)+{\mathrm{\&zeta;}}_2\left(k\right)\&CenterDot;\mathrm{SOC}\left(k\right)+{\mathrm{\&zeta;}}_3\left(k\right)$

Wherein, parameter ζ ₁(k), ζ ₂(k), ζ ₃(k) have following iterative relation:

${\mathrm{\&zeta;}}_1\left(k\right)=f_{{\mathrm{\&zeta;}}_1}({\mathrm{\&zeta;}}_1(k+1),{\mathrm{\&eta;}}_1\left(k\right))$

$={\mathrm{\&delta;}}_1\&CenterDot;{\mathrm{\&eta;}}_1^2\left(k\right)+{\mathrm{\&zeta;}}_1(k+1)\&CenterDot;{(1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_1\left(k\right))}^2$

${\mathrm{\&zeta;}}_2\left(k\right)=f_{{\mathrm{\&zeta;}}_2}({\mathrm{\&zeta;}}_1(k+1),{{\mathrm{\&zeta;}}_2(k+1),\mathrm{\&eta;}}_1\left(k\right),{\mathrm{\&eta;}}_2\left(k\right))$

$={2\mathrm{\&delta;}}_1\&CenterDot;{\mathrm{\&eta;}}_1\left(k\right)\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right)+{\mathrm{\&delta;}}_2\&CenterDot;{\mathrm{\&eta;}}_1\left(k\right)+{\mathrm{\&zeta;}}_2(k+1)\&CenterDot;(1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_1\left(k\right))$

${+2\mathrm{\&zeta;}}_1(k+1)\&CenterDot;(1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_1\left(k\right))\&CenterDot;({\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right))$

${\mathrm{\&zeta;}}_3\left(k\right)=f_{{\mathrm{\&zeta;}}_3}({\mathrm{\&zeta;}}_1(k+1),{{\mathrm{\&zeta;}}_2(k+1),\mathrm{\&zeta;}}_3(k+1),{\mathrm{\&eta;}}_2\left(k\right))$

$={\mathrm{\&delta;}}_1\&CenterDot;{\mathrm{\&eta;}}_2^2\left(k\right)+{\mathrm{\&delta;}}_2\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right)+{\mathrm{\&delta;}}_3+{\mathrm{\&zeta;}}_1(k+1)\&CenterDot;{{(\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right))}^2$

${+\mathrm{\&zeta;}}_2(k+1)\&CenterDot;({\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right))+{\mathrm{\&zeta;}}_3(k+1)$

In step (3); Vehicle is at k expectation moving velocity v constantly; And will be imported into transmitting ratio control module 4 by the engine optimum horsepower output

that substep (2.4) calculates, this module can reversely calculate the optimum speed ratio

of speed changer gear

The transmitting ratio control method derivation of module 4 is made up of following 2 sub-steps: the asking for of (3.1) engine optimum working speed and power relation curve; (3.2) the reverse calculating of the optimum speed ratio of speed changer gear

.

In substep (3.1), the best fuel oil consumption rate curve of the driving engine among Fig. 3 is with the optimum working speed that is used to confirm driving engine under the equipower.Definite step of optimum working speed is as shown in Figure 6, and with intersecting of equipower curve (long and short dash line) and best fuel oil consumption rate curve (dotted line), the cooresponding abscissa tachometer value of intersection point is the optimum working speed of driving engine under the equipower.By that analogy, can generate the relation curve of expression engine optimum working speed and power, like Fig. 6 dotted line.

In substep (3.2); Utilize engine optimum horsepower output that substep (2.4) calculates and engine optimum working speed and the power relation curve among Fig. 6; Can reverse interpolation obtain cooresponding engine optimum working speed

and compare with expectation moving velocity v then, calculate the optimum speed ratio of speed changer gear:

$i_g^{*}\left(k\right)=\frac{{\mathrm{\&omega;}}_{\mathrm{ice}}^{*}\left(k\right)\&CenterDot;r}{v\left(k\right)\&CenterDot;i_m}$

Wherein, r is a tire radius; i _mBe the main reduction gear transmitting ratio.

Yet,, therefore, when the gear of practical implementation change-speed box is regulated, should select and the most close gear of optimum speed ratio because the gear ratios of change-speed box is discontinuous discrete speed ratios.The influence that brings thus is that along with the gear ratios distribution of change-speed box is dense more, the control effect of quick DP control method is accurate more.

With one section driving path with starting point and terminal point shown in Figure 4 is example, and overall pathway length 27.2km, highest line sail speed of a motor vehicle 108km/h; 2183 seconds overall travel time; Road grade changes between ± 2 degree, because the demand that conveys goods, vehicle loading quality drops to 0kg stage by stage from 250kg.The quick DP control method of existing DP method and this patent proposition is applied to the PHEV of this model; Adopt 32 2.00GHz treaters of Pentium Dual Core; 1.00GB the iterative computation that the PC of internal memory carries out dynamic programming to whole section path is asked for corresponding optimal control policy.Simulation result shows, the optimisation strategy effect consistent basically (shown in Figure 8) of quick DP control method and existing DP method, but only need 0.066Second fast the computing time of DP method, and existing DP method needs the 11640Second (≈ 31Hour) that reaches consuming time.

As stated, quick DP control method of the present invention is applicable to that object is the PHEV that has carried driving engine, electrical motor, Vehicular accumulator cell and had the automatic transmission with hydraulic torque converter of a plurality of gears.Said method is through proposing the new SOC definition and the relation function of optimum fuel consumption rate and power; Set up performance index function with quadratic form; Success realizes the dimensionality reduction to the DP math modeling of PHEV; Derive optimal policy and fuel consumption rate optimal value thus, thereby make the computational efficiency of method obtain geometric series raising, the real-time application of implementation method with analytic function form.Said method has merged multiple driving, road informations such as the speed of a motor vehicle in the driving path, road grade and load-carrying quality when calculating the required gross horsepower of running car, further improved the accuracy and the universality of method.In addition, this method both had been applicable to electric quantity consumption pattern (Charge-Depleting), was applicable to that also electric weight keeps pattern (Charge-Sustaining); Be not limited to have the automatic transmission with hydraulic torque converter of a plurality of gears of discrete converter speed ratio, also can be used for having the CVT change-speed box of stepless change ratio.Can do corresponding adjustment to method according to different PHEV power drive systems.

Claims (3)

1. fast control method that is applied to the plug-in hybrid electronlmobil, wherein, the power transmission of plug-in hybrid electronlmobil system possesses driving engine, electrical motor, Vehicular accumulator cell and has the automatic transmission with hydraulic torque converter of a plurality of gears; This control method adopts the state-of-charge based on the method definition Vehicular accumulator cell of energy, i.e. SOC; On this basis, derive the relation equation that comprises SOC state transition equation, engine optimum fuel consumption rate and power, the performance index function of quadratic form, the fuel consumption rate optimal performance index of quadratic form and the required gross horsepower accounting equation of running car of many information fusion; It is characterized in that:

Described optimal policy is that the linearity of SOC is resolved the function iteration form:

$u^{*}\left(k\right)=-\frac{{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2^2}\&CenterDot;\mathrm{SOC}\left(k\right)-\frac{{\mathrm{\&delta;}}_2+{2\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_1{\mathrm{\&mu;}}_2+{\mathrm{\&zeta;}}_2(k+1){\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2^2}$

Wherein, the iterative relation of each parameter is following:

${\mathrm{\&eta;}}_1\left(k\right)=-\frac{{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2^2}$

${\mathrm{\&eta;}}_2\left(k\right)=-\frac{{\mathrm{\&delta;}}_2+{2\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_1{\mathrm{\&mu;}}_2+{\mathrm{\&zeta;}}_2(k+1){\mathrm{\&mu;}}_2}{{\mathrm{\&delta;}}_1+{\mathrm{\&zeta;}}_1(k+1){\mathrm{\&mu;}}_2^2}$

${\mathrm{\&mu;}}_1=-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)$

${\mathrm{\&mu;}}_2=\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)$

${\mathrm{\&zeta;}}_1\left(k\right)={\mathrm{\&delta;}}_1\&CenterDot;{\mathrm{\&eta;}}_1^2\left(k\right)+{\mathrm{\&zeta;}}_1(k+1)\&CenterDot;{(1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_1\left(k\right))}^2$

ζ ₂(k)＝2δ ₁·η ₁(k)·η ₂(k)+δ ₂·η ₁(k)+ζ ₂(k+1)·(1+μ ₂·η ₁(k))+2ζ ₁(k+1)·(1+μ ₂·η ₁(k))·(μ ₁+μ ₂·η ₂(k))

${\mathrm{\&zeta;}}_3\left(k\right)={\mathrm{\&delta;}}_1\&CenterDot;{\mathrm{\&eta;}}_2^2\left(k\right)+{\mathrm{\&delta;}}_2\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right)+{\mathrm{\&delta;}}_3+{\mathrm{\&zeta;}}_1(k+1)\&CenterDot;{({\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right))}^2$

$+{\mathrm{\&zeta;}}_2(k+1)\&CenterDot;({\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;{\mathrm{\&eta;}}_2\left(k\right))+{\mathrm{\&zeta;}}_3(k+1)$

That is δ, ₁=α ₁Δ t, δ ₂=α ₂Δ t, δ ₃=α ₃Δ t; α ₁, α ₂, α ₃Be the match constant coefficient; Δ t is the sampling time; E _pTotal electric energy when being full of for storage battery; 1≤k≤N; Function is input as the required gross horsepower P of running car _ReqAnd the initial sum end of a period state of battery SOC (k); The optimal power allocation that is output as engine/motor compares u ^*(k).

2. the fast control method that is applied to the plug-in hybrid electronlmobil according to claim 1 is characterized in that: described Vehicular accumulator cell SOC adopts the define method based on energy:

$\mathrm{SOC}\left(k\right)=\frac{E\left(k\right)}{E_p}$

Wherein, E (k) is a battery k dump energy constantly;

Provide based on the SOC of energy and SOC based on electric weight _QBetween transformational relation:

$\mathrm{SOC}\left(k\right)=\frac{{Q_p}^2}{E_p\&CenterDot;C}{\mathrm{SOC}}_Q^2\left(k\right)$

Wherein, C is the electric capacity of representative capacity of cell; Q _pTotal electric weight when being full of for storage battery.

3. the fast control method that is applied to the plug-in hybrid electronlmobil according to claim 1 is characterized in that described derived equation comprises:

Have the SOC state transition equation and have linear forms;

Engine optimum fuel consumption rate and power relation equation with Quadratic Function Optimization form;

Performance index function with Quadratic Function Optimization form of 1 * 1 dimension;

Fuel consumption rate optimal performance index with SOC secondary analytic function iteration form;

Consider the required gross horsepower method of calculating of running car of many information fusion;

Wherein, said SOC state transition equation has linear forms:

$\mathrm{SOC}(k+1)-\mathrm{SOC}\left(k\right)={\mathrm{\&mu;}}_1+{\mathrm{\&mu;}}_2\&CenterDot;u\left(k\right)$

$=-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)+\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)\&CenterDot;u\left(k\right)$

Wherein, the input control variable does

P _Req(k) be the required gross horsepower of running car; P _Ice(k) be engine power;

Said engine optimum fuel consumption rate and power have the Quadratic Function Optimization form:

fuel(k)＝δ ₁·P _ice(k) ²+δ ₂·P _ice(k)+δ ₃

Wherein, input P _Ice(k) be engine power; Output fuel (k) is optimum fuel consumption rate; Function is the engine optimum fuel consumption rate that proposed and the approximate match of power relation curve;

Said performance index function is the Quadratic Function Optimization form of 1 * 1 dimension:

$J_k(\mathrm{SOC}\left(k\right),u\left(k\right))$

$=\underset{i=k}{\overset{N}{\mathrm{\&Sigma;}}}\left[\mathrm{fuel}(P_{\mathrm{req}}\left(i\right)\&CenterDot;u\left(i\right))\right]+\mathrm{\&beta;}{(\mathrm{SOC}(N+1)-{\mathrm{SOC}}_{\mathrm{des}})}^2$

Wherein, SOC converges on SOC to β for constraint ends constantly _DesThe performance figure weighted value; Having with power-division ratios u (i) is the one dimension control variable; With SOC is the one dimension state variable; With fuel consumption rate fuel (*) and SOC variable quantity two parts serves as to optimize performance figure;

The secondary analytic function iteration form that said fuel consumption rate optimal performance index is SOC:

$J_k^{*}\left(\mathrm{SOC}\left(k\right)\right)={\mathrm{\&zeta;}}_1\left(k\right)\&CenterDot;{\mathrm{SOC}}^2\left(k\right)+{\mathrm{\&zeta;}}_2\left(k\right)\&CenterDot;\mathrm{SOC}\left(k\right)+{\mathrm{\&zeta;}}_3\left(k\right)$

$={\mathrm{\&zeta;}}_1\left(k\right)\&CenterDot;{[\mathrm{SOC}(k+1)-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)+\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)\&CenterDot;u\left(k\right)]}^2$

$+{\mathrm{\&zeta;}}_2\left(k\right)\&CenterDot;[\mathrm{SOC}(k+1)-\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)+\frac{\mathrm{\&Delta;t}}{E_p}{P}_{\mathrm{req}}\left(k\right)\&CenterDot;u\left(k\right)]+{\mathrm{\&zeta;}}_3\left(k\right)$

Wherein, be input as the required gross horsepower P of running car _Req(k), power-division ratios u (k), and the initial sum end of a period state of battery SOC; Be output as the fuel consumption rate optimal value of driving engine $J_k^{*}\left(\mathrm{SOC}\left(k\right)\right);$

The optimum speed ratio of speed changer gear; Be to utilize the engine optimum working speed proposed and the relation curve of power; And combine the optimum horsepower output of present engine, reverse interpolation is obtained final the comparing with current vehicle speed v (k) of engine optimum working speed

and is obtained optimum speed ratio:

$i_g^{*}\left(k\right)=\frac{{\mathrm{\&omega;}}_{\mathrm{ice}}^{*}\left(k\right)\&CenterDot;r}{v\left(k\right)\&CenterDot;i_m}$

Wherein, r is a tire radius; i _mBe the main reduction gear transmitting ratio;

The required gross horsepower method of calculating of the running car of said many information fusion:

$P_{\mathrm{req}}\left(k\right)$

$=[\frac{\mathrm{\&rho;}}{2}{A}_f{C}_{d}g\&CenterDot;v^3\left(k\right)+M\left(k\right)({\mathrm{\&mu;}}_r+\sin \mathrm{\&Theta;}\left(k\right))g\&CenterDot;v\left(k\right)]+\frac{M\left(k\right)}{g}(1+\frac{{\mathrm{\&delta;}}_{\mathrm{eqm}}}{g})\&CenterDot;\stackrel{\&CenterDot;}{v}\left(k\right)\&CenterDot;v\left(k\right)$

Wherein, method is integrated multiple road and running information comprises speed of a motor vehicle v (k), road grade θ (k) and load-carrying mass M (k); ρ is a density of air; A _fBe the vehicle wind area; C _dBe the Reynolds coefficient; G=9.8m/s ²μ _rBe coefficient of rolling resistance;

Be acceleration/accel; δ _EqmBe equivalent moment of inertia.

Images