CN106143535A - A kind of subway train optimization of operating parameters method based on immune algorithm - Google Patents

Abstract

A kind of subway train optimization of operating parameters method based on immune algorithm, initially set up total energy consumption equation and constraints thereof, then total energy consumption minimum operational factor is tried to achieve according to concentration immune algorithm, step is as follows: using total energy consumption equation and constraints thereof as the antigen of concentration immune algorithm, arrange calculating parameter；Randomly generate initial antibodies group A, the affinity of each antibody in calculating antibody group A, retain the antibody that in antibody population A, affinity is bigger, constitute antibody population B, introducing antibody Reproductive Strategy based on concentration, the antibody selecting expectation breeding potential high carries out replicating operation, produces antibody population C；The individuality of antagonist group C carries out intersecting and mutation operation, produces antibody population D；The affinity of each antibody in calculating antibody group D, chooses n the high antibody of affinity and replaces the antibody that in C, affinity is low, form antibody population E；Judge whether to meet end condition.It is fast that the present invention calculates speed, and solving precision is high, and optimum results effectively reduces train total energy consumption.

Description

A kind of subway train optimization of operating parameters method based on immune algorithm

Technical field

The present invention relates to a kind of subway many trains energy conservation optimizing method, be specifically related to a kind of subway based on immune algorithm row Car regenerating braking energy utilizes and timetable optimization method.

Background technology

Subway safety convenient and swift with it, comfortable, percent of punctuality advantages of higher, become the trip mode that people are important, to slow Solution urban traffic pressure plays an important role.Meanwhile, subway is power consumption unit maximum in urban public transport, research The energy saving optimizing of subway train, to saving operating cost, promotes that the healthy Green Development of urban transportation is significant.

Subway line distance between sites is short, in train travelling process, needs frequently traction and braking, when train regenerative braking Time, if adjacent train is in traction acceleration mode, then regenerating braking energy will feed back to contact net for adjacent train.Adjacent Two trains accelerate with the lock in time (overlapping time) braked by departure interval, dwell time, stand between the factor shadow such as operation time Ringing, if two following distances are excessive, at most only have a train operation in same power supply section, regenerating braking energy cannot be by adjacent column Car uses, and now for avoiding Traction networks to press through greatly, this portion of energy is by vehicle-mounted resistance absorption.At present, save for the many trains of subway The research of energy optimized handling method is concentrated mainly on train energy-saving and controls and regenerative braking field, and considers train regeneration system Energy utilizes less with the method for timetable design.

Summary of the invention

The present invention is directed to the deficiencies in the prior art, it is provided that a kind of subway train optimization of operating parameters side based on immune algorithm Method.

The present invention solves subway many trains energy saving optimizing problem by the following technical solutions:

A kind of subway train optimization of operating parameters method based on immune algorithm, comprises the following steps:

Step 1: divide two adjacent for the train run on subway line trains into a unit, including leading train with Following train, is divided into traction boost phase, constant velocity stage, coasting stage and regenerative braking rank by running between the station of train Section four-stage, according to the pull strength in actual track condition and train travelling process and resistance situation, sets up total energy consumption side Journey and constraints thereof；

Step 2: according to the operational factor of two trains in concentration immune algorithm adjustment unit, so that total energy consumption is minimum, Specifically comprise the following steps that

(1) using total energy consumption equation and constraints thereof as the antigen of concentration immune algorithm, and antibody scale P, most is set Big iterations N, crossover probability P_CWith mutation probability P_m；

(2) initial antibodies group A is randomly generated；

(3) affinity of each antibody in calculating antibody group A,

(4) retain the antibody that in antibody population A, affinity is bigger, constitute antibody population B,

(5) antibody Reproductive Strategy based on concentration, concentration N of each antibody in calculating antibody group B are introduced_kWith expectation breeding Rate E_k, whereinα is constant, and the antibody selecting expectation breeding potential high carries out replicating operation, produces antibody population C；

(6) individuality of antagonist group C carries out intersecting and mutation operation, produces antibody population D；

(7) affinity of each antibody in calculating antibody group D, chooses n the high antibody of affinity and replaces affinity in C low Antibody, formed antibody population E；

(8) judging whether to meet end condition, if meeting end condition, exporting current antibody, be energy consumption function Excellent solution, otherwise repeats step (3)～(7)；Described end condition is to reach the maximum of antibody in maximum iteration time or antibody population E Affinity tends to constant.

Described total energy consumption equation and constraints method for building up thereof are as follows:

1. train traction boost phase energy consumption Q_q:

$Q_q=\underset{i=1}{\overset{2}{\Σ}}\underset{j=1}{\overset{m-1}{\Σ}}{\∫}_0^{t_{i,j}^q}{M}_i\·a_{i,j}^q\·\frac{1}{2}\·a_{i,j}^q\·tdt=\frac{1}{2}\·\underset{i=1}{\overset{2}{\Σ}}\underset{j=1}{\overset{m-1}{\Σ}}{\∫}_0^{t_{i,j}^q}{M}_i\·{\left(a_{i,j}^q\right)}^2\·tdt$

$a_{i,j}^q=\frac{F_{i,j}^q-f_a-f_b}{M_i}$

f_a=2.43+0.0275v+0.0078v²

$f_b=M_{i}g(\mathrm{\θ}+\frac{0.6}{R})$

In formula,

For train i traction acceleration between j to (j+1) stands,

For the train i traction acceleration time between j to (j+1) stands,

M_iThe quality of train i,

For train Accelerating running pull strength,

f_aFor train basic resistance,

V is train speed,

f_bFor train additional drag,

G is acceleration of gravity,

θ is hill gradient,

R is turning radius,

2. train travels at the uniform speed stage energy consumption Q_y:

$Q_y=\underset{i=1}{\overset{2}{\Σ}}\underset{j=1}{\overset{m-1}{\Σ}}{\∫}_0^{t_{i,j}^y}{F}_{i,j}^y\·v_{i,j}dt$

In formula,

v_{I, j}For the train i speed that travels at the uniform speed between j to (j+1) stands,

For the train i pull strength that travels at the uniform speed between j to (j+1) stands,

Travel at the uniform speed the time between j to (j+1) stands for train i；

3. coasting stage train freewheeling, only by drag effects, the most vehicle-mounted auxiliary equipment electricity consumption, train energy consumption is neglected Slightly disregard, if train i coasting time between j to (j+1) stands is

4. the energy Q that the train regenerative braking stage reclaims_z:

$Q_z={\∫}_0^{T_c}{M}_i\·a_z\·(v_z-\frac{1}{2}\·a_z\·t)dt$

In formula,

v_zInitial speed of braking during for accelerating overlapping with braking,

a_zBraking acceleration during for accelerating overlapping with braking,

According to planning timetable and combining the manipulation operating mode run between train station, obtain adjacent two in same power supply section Train regenerative braking total time T overlapping with Acceleration of starting_c, adjacent two train tractions accelerate the feelings of generation Tong Bu with regenerative braking Condition is: leading train braking-following train draws, and following train braking-leading train draws, and is respectively T overlapping time_c1With T_c2, then two trains accelerate total time T overlapping with regenerative braking in whole piece circuit_c:

M is station sum,

ρ (j, j+1) is j stands whether be positioned at the coefficient of determination that same power supply is interval with (j+1),

5. total energy consumption Q equation:

$\begin{array}{c}\min Q=f(t_{i,j}^q,t_{i,j}^y,t_{i,j}^d,t_{i,j}^z,T_p,T_f,a_{i,j}^q,a_{i,j}^z,v_{i,j},a_z,v_z,F_{i,j}^q,F_{i,j}^y)\\ =Q_q+Q_y-Q_z\end{array}$

Constraints:

In formula,

T_pFor the dwell time,

T_{p min}For the minimum dwell time,

T_{p max}For the maximum dwell time,

T is the train i operation time between j to (j+1) stands,

β is time margin,

T_fFor headway,

T_{z min}For minimum tracking interval.

According to affinity function F_kThe affinity of=-Q+c calculating antibody, in formula, Q is total energy consumption equation, and c is constant.

Arbitrary antibody in described antibody population

Compared with prior art, the beneficial effect that the present invention possesses:

Existing achievement in research fails when formulating time-table to consider to transport between departure interval, dwell time and station comprehensively The factors such as row time, or timetable design is not combined with the optimization of Handling Strategy between station, cause train regenerative braking energy Amount utilizes the highest, and train energy consumption is bigger.The present invention precisely identifies feasible solution by affinity and concentration double mechanism, with tradition group Collection intelligent algorithm is compared, and calculates speed fast, and solving precision is high, is meeting train safe, comfortableness, running on time and train On the premise of speed limit, effectively reduce train total energy consumption.From many cars Energy Angle, analyze regenerating braking energy utilization power, build Vertical many cars energy consumption model is to reduce energy consumption in train journey to greatest extent and to improve regenerating braking energy utilization rate as target, logical Cross introducing concentration immune algorithm, and binding time nargin thought, adjust Train Schedule, dwell time and departure interval etc. Element of time, appropriateness optimizes train handling operating mode, designs the energy-conservation timetable with robustness, thus increase adjacent train and add Speed and the overlapping time of braking, make full use of regenerating braking energy, effectively reduce train operation total energy consumption, for the driving of train Scheduling provides theoretical direction with energy-saving driving.

Accompanying drawing explanation

Fig. 1 is that immune algorithm solves train energy consumption flow chart.

Fig. 2 is route map of train before optimizing.

Fig. 3 is route map of train after optimizing.

Detailed description of the invention

Below in conjunction with detailed description of the invention, the present invention is described in detail.

Embodiment 1

(1) duty parameter

The present embodiment uses Subway Line 1, Nanning service data, and in simulation uplink, railway station is to station, Place de la Nation 4 Stand 3 interval in two train operation situations of actual tracks.For phase commuter rush hour, railway station-Chaoyang Plaza station, Chaoyang Plaza Stand-new people way station, new people way station-station, Place de la Nation respectively with AW3, AW3, AW2 operating mode be simulated, AW3 operating mode correspondence train Mass M₁=344380kg, AW2 operating mode correspondence train mass M₂=307300kg.Under AW3 Yu AW2 operating mode, train is respectively with constant Maximum drawbar pull 386kN, 346kN Acceleration of starting, in process of regenerative braking, maximum electric braking force is the most constant for 335kN, train Speed limit is 80km/h.

Before optimization, headway T_fFor 150s, when two trains are according to plan shown in duty parameter described in table 1 and table 2 Quarter, table ran, and leading train is with following train speed-time curve as shown in Figure 2.

Table 1

Table 2

(2) optimization method

A kind of subway train optimization of operating parameters method based on immune algorithm, comprises the following steps:

Step 1: divide two adjacent for the train run on subway line trains into a unit, including leading train with Following train, is divided into traction boost phase, constant velocity stage, coasting stage and regenerative braking rank by running between the station of train Section four-stage, according to the pull strength in actual track condition and train travelling process and resistance situation, sets up total energy consumption side Journey and constraints thereof；

Described total energy consumption equation and constraints method for building up thereof are as follows:

1. train traction boost phase energy consumption Q_q:

$Q_q=\underset{i=1}{\overset{2}{\Σ}}\underset{j=1}{\overset{m-1}{\Σ}}{\∫}_0^{t_{i,j}^q}{M}_i\·a_{i,j}^q\·\frac{1}{2}\·a_{i,j}^q\·tdt=\frac{1}{2}\·\underset{i=1}{\overset{2}{\Σ}}\underset{j=1}{\overset{m-1}{\Σ}}{\∫}_0^{t_{i,j}^q}{M}_i\·{\left(a_{i,j}^q\right)}^2\·tdt$

$a_{i,j}^q=\frac{F_{i,j}^q-f_a-f_b}{M_i}$

f_a=2.43+0.0275v+0.0078v²

$f_b=M_{i}g(\mathrm{\θ}+\frac{0.6}{R})$

In formula,

For train i traction acceleration between j to (j+1) stands,

For the train i traction acceleration time between j to (j+1) stands,

M_iThe quality of train i,

For train Accelerating running pull strength,

f_aFor train basic resistance,

V is train speed,

f_bFor train additional drag,

G is acceleration of gravity,

θ is hill gradient,

R is turning radius,

2. train travels at the uniform speed stage energy consumption Q_y:

$Q_y=\underset{i=1}{\overset{2}{\Σ}}\underset{j=1}{\overset{m-1}{\Σ}}{\∫}_0^{t_{i,j}^y}{F}_{i,j}^y\·v_{i,j}dt$

In formula,

v_{I, j}For the train i speed that travels at the uniform speed between j to (j+1) stands,

For the train i pull strength that travels at the uniform speed between j to (j+1) stands,

Travel at the uniform speed the time between j to (j+1) stands for train i；

3. coasting stage train freewheeling, only by drag effects, the most vehicle-mounted auxiliary equipment electricity consumption, train energy consumption is neglected Slightly disregard, if train i coasting time between j to (j+1) stands is

4. the energy Q that the train regenerative braking stage reclaims_z:

$Q_z={\∫}_0^{T_c}{M}_i\·a_z\·(v_z-\frac{1}{2}\·a_z\·t)dt$

In formula,

v_zInitial speed of braking during for accelerating overlapping with braking,

a_zBraking acceleration during for accelerating overlapping with braking,

According to planning timetable and combining the manipulation operating mode run between train station, obtain adjacent two in same power supply section Train regenerative braking total time T overlapping with Acceleration of starting_c, adjacent two train tractions accelerate the feelings of generation Tong Bu with regenerative braking Condition is: leading train braking-following train draws, and following train braking-leading train draws, and is respectively T overlapping time_c1With T_c2, then two trains accelerate total time T overlapping with regenerative braking in whole piece circuit_c:

M is station sum,

ρ (j, j+1) is j stands whether be positioned at the coefficient of determination that same power supply is interval with (j+1),

5. total energy consumption Q equation:

$\begin{array}{c}\min Q=f(t_{i,j}^q,t_{i,j}^y,t_{i,j}^d,t_{i,j}^z,T_p,T_f,a_{i,j}^q,a_{i,j}^z,v_{i,j},a_z,v_z,F_{i,j}^q,F_{i,j}^y)\\ =Q_q+Q_y-Q_z\end{array}$

Constraints:

In formula,

T_pFor the dwell time,

T_{p min}For the minimum dwell time,

T_{p max}For maximum dwell time, phase commuter rush hour T_pEmpirical value is [20,50] s,

T is the train i operation time between j to (j+1) stands, and β is time margin (taking 5%),

T_fFor headway,

T_{z min}For minimum tracking interval.

Step 2: overlapping time is reached time parameter corresponding during maximum and speed parameter and the traction of train maximum The parameter such as power, maximum braking force, train weight substitutes into energy consumption equation, tries to achieve two trains operation minimum in same power supply is interval Energy consumption.

According to the operational factor of two trains in concentration immune algorithm adjustment unit, adjust the departure interval in time-table T_f, dwell time T_p, make obtain maximum overlapping time, so that total energy consumption is minimum, specifically comprise the following steps that

(1) using total energy consumption equation and constraints thereof as the antigen of concentration immune algorithm, and antibody scale P=is set 50, maximum iteration time N=200, crossover probability P_C=0.8, mutation probability P_m=0.2；

(2) initial antibodies group A is randomly generated；

(3) affinity of each antibody in calculating antibody group A,

(4) retain the antibody that in antibody population A, affinity is bigger, constitute antibody population B,

(5) antibody Reproductive Strategy based on concentration, concentration N of each antibody in calculating antibody group B are introduced_kWith expectation breeding Rate E_k, whereinα is constant, and the antibody selecting expectation breeding potential high carries out replicating operation, produces antibody population C；

(6) individuality of antagonist group C carries out intersecting and mutation operation, produces antibody population D；

(7) affinity of each antibody in calculating antibody group D, chooses n the high antibody of affinity and replaces affinity in C low Antibody, formed antibody population E；

(8) judging whether to meet end condition, if meeting end condition, exporting current antibody, be energy consumption function Excellent solution, otherwise repeats step (3)～(7)；Described end condition is to reach the maximum of antibody in maximum iteration time or antibody population E Affinity tends to constant.

The antibody i.e. feasible solution of energy consumption equation, the energy consumption equation functions value that antibody affinity is the most to be optimized, antibody concentration N_k I.e. similar to antibody antibody levels.

(3) optimum results

Use concentration immune algorithm, in the range of constraints limits, search for the feasible solution of many cars energy consumption model, finally seek Find out regenerating braking energy and utilize maximized energy-conservation timetable, as shown in table 2, and Handling Strategy between station, and draw first ranks Car and following train energy-saving run curve, as shown in Figure 3.After optimization, headway T_fIt is adjusted to 143s.

Table 3

Table 4

Contrast table 3 and table 4 data, railway station, Chaoyang Plaza station, new people way station stand in Place de la Nation and meet train work On the basis of time the dwell time round after be adjusted to 35s, 27s, 26s, 28s.The actual arrival time of train meet setting time Between nargin scope, meet percent of punctuality requirement.From the data in table 4, it can be seen that regenerating braking energy utilization rate improves 11.4%, run between standing Total energy consumption reduces by 16.9%.

For other concrete cases, if adjusting departure interval T_f, dwell time T_pCan not make reach maximum overlapping time, Adjust the time parameters such as train traction acceleration time, at the uniform velocity time, coasting time and regenerative braking time the most further, and accordingly Adjust train uniform velocity's uniform velocity parameter, make the object function of overlapping time obtain maximum.

Claims (4)

1. a subway train optimization of operating parameters method based on immune algorithm, it is characterised in that comprise the following steps:

Step 1: divide two adjacent for the train run on subway line trains into a unit, including leading train and tracking Train, is divided into traction boost phase, constant velocity stage, coasting stage and regenerative braking stage four by running between the station of train The individual stage, according to the pull strength in actual track condition and train travelling process and resistance situation, set up total energy consumption equation and Its constraints；

Step 2: according to the operational factor of two trains in concentration immune algorithm adjustment unit, so that total energy consumption is minimum, specifically Step is as follows:

(1) using total energy consumption equation and constraints thereof as the antigen of concentration immune algorithm, and arrange antibody scale P, maximum repeatedly For times N, crossover probability P_CWith mutation probability P_m；

(2) initial antibodies group A is randomly generated；

(3) affinity of each antibody in calculating antibody group A,

(4) retain the antibody that in antibody population A, affinity is bigger, constitute antibody population B,

(5) antibody Reproductive Strategy based on concentration, concentration N of each antibody in calculating antibody group B are introduced_kWith expectation breeding potential E_k, Whereinα is constant, and the antibody selecting expectation breeding potential high carries out replicating operation, produces antibody population C；

(6) individuality of antagonist group C carries out intersecting and mutation operation, produces antibody population D；

(7) affinity of each antibody in calculating antibody group D, chooses n the high antibody of affinity and replaces low the resisting of affinity in C Body, forms antibody population E；

(8) judge whether to meet end condition, if meeting end condition, exporting current antibody, being the optimal solution of energy consumption function, Otherwise repeat step (3)～(7)；Described end condition is that to reach the maximum of antibody in maximum iteration time or antibody population E affine Degree tends to constant.

2. the method for claim 1, it is characterised in that described total energy consumption equation and constraints method for building up thereof are such as Under:

1. train traction boost phase energy consumption Q_q:

$Q_q=\underset{i=1}{\overset{2}{\mathrm{\Σ}}}\underset{j=1}{\overset{m-1}{\mathrm{\Σ}}}{\∫}_0^{t_{i,j}^q}{M}_i\·a_{i,j}^q\·\frac{1}{2}\·a_{i,j}^q\·tdt=\frac{1}{2}\·\underset{i=1}{\overset{2}{\mathrm{\Σ}}}\underset{j=1}{\overset{m-1}{\mathrm{\Σ}}}{\∫}_0^{t_{i,j}^q}{M}_i\·{\left(a_{i,j}^q\right)}^2\·tdt$

$a_{i,j}^q=\frac{F_{i,j}^q-f_a-f_b}{M_i}$

f_a=2.43+0.0275v+0.0078v²

$f_b=M_{i}g(\mathrm{\θ}+\frac{0.6}{R})$

In formula,

For train i traction acceleration between j to (j+1) stands,

For the train i traction acceleration time between j to (j+1) stands,

M_iThe quality of train i,

For train Accelerating running pull strength,

f_aFor train basic resistance,

V is train speed,

f_bFor train additional drag,

G is acceleration of gravity,

θ is hill gradient,

R is turning radius,

2. train travels at the uniform speed stage energy consumption Q_y:

$Q_y=\underset{i=1}{\overset{2}{\Σ}}\underset{j=1}{\overset{m-1}{\Σ}}{\∫}_0^{t_{i,j}^y}{F}_{i,j}^y\·v_{i,j}dt$

In formula,

v_{I, j}For the train i speed that travels at the uniform speed between j to (j+1) stands,

For the train i pull strength that travels at the uniform speed between j to (j+1) stands,

Travel at the uniform speed the time between j to (j+1) stands for train i；

3. coasting stage train freewheeling, only by drag effects, the most vehicle-mounted auxiliary equipment electricity consumption, train energy consumption is ignored not Count, if train i coasting time between j to (j+1) stands is

4. the energy Q that the train regenerative braking stage reclaims_z:

$Q_z={\∫}_0^{T_c}{M}_i\·a_z\·(v_z-\frac{1}{2}\·a_z\·t)dt$

In formula,

v_zInitial speed of braking during for accelerating overlapping with braking,

a_zBraking acceleration during for accelerating overlapping with braking,

According to planning timetable and combining the manipulation operating mode run between train station, obtain adjacent two trains in same power supply section Regenerative braking total time T overlapping with Acceleration of starting_c, adjacent two train tractions accelerate the situation of generation Tong Bu with regenerative braking For: leading train braking-following train draws, and following train braking-leading train draws, and is respectively T overlapping time_c1With T_c2, Then two trains accelerate total time T overlapping with regenerative braking in whole piece circuit_c:

M is station sum,

ρ (j, j+1) is j stands whether be positioned at the coefficient of determination that same power supply is interval with (j+1),

5. total energy consumption Q equation:

$\begin{array}{c}minQ=f(t_{i,j}^q,t_{i,j}^y,t_{i,j}^d,t_{i,j}^z,T_p,T_f,a_{i,j}^q,a_{i,j}^z,v_{i,j},a_z,v_z,F_{i,j}^q,F_{i,j}^y)\\ =Q_q+Q_y-Q_z\end{array}$

Constraints:

In formula,

T_pFor the dwell time,

T_{p min}For the minimum dwell time,

T_{p max}For the maximum dwell time,

T is the train i operation time between j to (j+1) stands,

β is time margin,

T_fFor headway,

T_{z min}For minimum tracking interval.

3. the method for claim 1, it is characterised in that according to affinity function F_kThe affinity of=-Q+c calculating antibody, In formula, Q is total energy consumption equation, and c is constant.

4. the method for claim 1, it is characterised in that arbitrary antibody in described antibody population