CN102779322A - Thermal power plant comprehensive scheduling method based on environment protection and economical benefit - Google Patents

Abstract

The invention discloses a thermal power plant comprehensive scheduling method based on environment protection and economical benefit. The method includes the following steps: building a mathematical model of thermal power plant environment economical scheduling problems; obtaining various types of parameters in the model; initiating a practicable scheduling generating process and encoding processing; evaluating practicable scheduling; updating the practicable scheduling; updating a non-bad scheduling solution set in an external file set; determining three important non-domination scheduling sequences affecting an iterative process by circulating crowding distance; judging if iteration t reaches a maximum value, on yes judgment, outputting practicable scheduling in the current non-bad solution set, and on no judgment, setting the iteration t=t+1, and returning to step 5 to conduct updating again. Due to the fact that increasing of electric energy production can inevitably cause increasing of pollution discharge, namely increasing of cost for treating environment pollution, the thermal power plant comprehensive scheduling method adjusts generating efficiency of all thermal power plant units in real time and achieves optimization of comprehensive benefit.

Description

Thermal power plant's integrated dispatch method based on environmental protection and economic benefit

Technical field

The invention belongs to the Comprehensive Control Technology field, be specifically related to a kind of thermal power plant's integrated dispatch method based on environmental protection and economic benefit.

Background technology

In recent years, along with rapid growth of economy, environmental pressure increases continuing.The harmful gases such as oxysulfide, oxides of nitrogen and carbon dioxide that power plant emission is a large amount of; Not only direct pollutant atmosphere; And be the key factor that causes greenhouse effect, therefore, take all factors into consideration the research of the power system environment economic load dispatching problem of environmental protection and economic benefit; Not only have important significance for theories, and be the Real Choice of China's energy and the electric power strategy of sustainable development.But consider that discharge amount of pollution can become multi-objective optimization question with original single goal optimization problem, not only increased the complexity of problem, brought difficulty and challenge also for the enforcement of scheduling.

In power system environment economic load dispatching problem, have a plurality of indexs of fuel cost and toxic emission usually, and be conflicting relation between each target, therefore formulating rational operation plan is the emphasis of studying at present.

Summary of the invention

The purpose of this invention is to provide a kind of thermal power plant's integrated dispatch method, solved existing dispatching method and between the economic benefit of electrical network and environmental benefit, be difficult to regulate, be difficult for obtaining the problem of optimum efficiency based on environmental protection and economic benefit.

The technical scheme that the present invention adopted is that a kind of thermal power plant's integrated dispatch method based on environmental protection and economic benefit may further comprise the steps:

Step 1, set up the mathematical model of thermal power plant's environmental economy scheduling problem

The expression formula of its mathematical model is:

In the formula (1), i=1,2 ..., N _GRepresent i platform generator, N _GBe generator sum in the system;

The fuel used to generate electricity total burn-off of

expression thermal power plant is the fuel total expenses, and calculation expression is:

$F_i\left(P_i\right)=a_i+b_i+c_i{P}_i^2---\left(2\right)$

In the formula (2), P _iThe active power of representing i platform generator; F _i(P _i) fuel consumption of expression i platform generator, a _i, b _i, c _iBeing systematic parameter, representing the constant term of i platform genset fuel consumption characteristic, once coefficient and quadratic term coefficient respectively, is known parameters in system;

In addition;

is the total release of dusty gas, and calculation expression is:

$E_i\left(P_i\right)={\mathrm{\α}}_i+{\mathrm{\β}}_i+{\mathrm{\γ}}_i{P}_i^2+{\mathrm{\ξ}}_i\exp \left({\mathrm{\λ}}_i{P}_i\right),---\left(3\right)$

In the formula (3), E _i(P _i) discharge amount of pollution of expression i platform generator; α _i, β _i, γ _i, ξ _i, λ _iBe systematic parameter, α _i, β _i, γ _iRepresent the constant term of i platform genset disposal of pollutants flow characteristic, once coefficient and quadratic term coefficient respectively, ξ _i, λ _iThe correlation parameter of expression exponential term is a constant value with respect to concrete system;

Suffered restrictive condition comprises in the environmental economy optimizing scheduling process:

A) the minimax capacity limit of generator, expression formula is:

wherein is respectively the minimum and maximum active power output of i platform thermoelectric generator;

B) balance constraint; Be that each genset generated output sum of system should equal total demand power of load and via net loss sum, expression formula is:

In the formula,

Total active power for power plant units; P _DRated power for system; P _LossBe active power loss in the electrical network, net decreases to calculate and adopts the B Y-factor method Y, and expression formula is: Wherein claim B _Ij, B _0i, B ₀₀Be the B coefficient, the concrete setting according to unit obtains respectively, is known parameters;

Step 2, obtain all kinds of parameters in the model

2.1) load valley of confirming electrical network according to grid load curve and load peak period, confirm the power demand of day part load distribution

From the grid dispatching center statistics, obtain the data of current time system, comprise the rated power P of system _D, active power loss P _LossParameter value B _Ij, B _0i, B ₀₀

2.2) running performance parameters of unit, the discharge of harmful gases amounts such as coal consumption amount, flue dust and carbon dioxide when comprising operation are according to the soot emissions price, reduce discharging the supplemental characteristic α that price, CO2 emission exercise price obtain discharge amount of pollution _i, β _i, γ _i, ξ _i, λ _iCoal consumption price in producing according to electric system assistant service price, unit of electrical energy, cost of electricity-generating when obtaining raw coal price, diesel-fuel price thermal power unit operation obtains the parameter alpha of fuel total expenses _i, b _i, c _i

Step 3, initial feasible schedule production process and encoding process

In above-mentioned mathematical model, the decision variable that needs to confirm is N _GThe active power of individual unit

Employing realizes based on the particle cluster algorithm of cluster thought, need produce initial feasible schedule for this reason, and the concrete steps that produce to initial feasible schedule are:

To preceding N _G-1 genset in the scope that satisfies maximum generation active power and minimum generating active power, produces the active power P of each unit at random _i, i=1,2 ..., N _G-1, utilize equality constraint then, try to achieve N _GThe active power of individual unit, computational grid loss P _Loss, and guarantee

Satisfy maximum, the restriction of minimum genset active power;

Produce N feasible schedule sequence according to same process;

Step 4, initial feasible schedule is estimated

N the feasible schedule sequence that above-mentioned steps 3 produces distinguished aforesaid formula of substitution (2) and formula (3), feasible schedule is estimated, and carried out non-bad ordering;

Because therefore all corresponding two objective functions of each feasible schedule need carry out non-bad ordering to N feasible schedule, the feasible schedule of not arranging mutually is kept in the outside set, be referred to as outside filing collection;

Step 5, feasible schedule is upgraded

Before beginning, iterations initial value t=0 need be set, and maximum iteration time T _Max,

After producing N initial feasible schedule and estimating; Next step will realize the renewal to this N feasible schedule, before upgrading, at first adopt the k-means clustering method; N feasible schedule carried out cluster; And the information that the fresh information that will obtain through cluster and particle cluster algorithm provide combines, and comes the genset active power of each feasible schedule is carried out the iteration update mechanism, specifically implements according to following steps:

Be located in the t time iteration, the active power of i unit is P in j feasible schedule _Ji(t), the rate of change of active power is V _Ji(t), before j feasible schedule in t iteration history during optimum economic benefit i unit active power be Pbest _JiThe active power of i unit is Pgbest when (t), obtaining optimum economic benefit in preceding t the iteration in all feasible schedule _i(t), the pairing genset active power of optimum economic benefit is C in the class of j feasible schedule place class _Ji(t), then when iterations is t+1, the active power P of i unit in j feasible schedule _Ji(t+1) iterative formula is following:

P _ji(t+1)＝P _ji(t)+V _ji(t+1) （4）

V _ji(t+1)＝ω·V _ji(t)+c ₁·r ₁·(Pbest _ji(t)-P _ji(t))+c ₂r ₂(P _gbest(t)-P _ji(t))+c ₃·r ₃·(C _ji(t)-P _ji(t))

（5）

In following formula (4) and the formula (5), three study factor c ₁, c ₂And c ₃Be constant, r ₁, r ₂, r ₃Be the random number between [0,1]; W is decremented to minimum value 0.4, V from maximal value 0.9 according to iterations _Ji(t) be the 10%-20% of genset active power variation range,

Through above two iterative formulas, the generating active power of the rate of change of each genset and genset is brought in constant renewal in, so that seek out N genset generated output feasible schedule set that more meets objective function under the integrated environment factor;

Step 6, the non-bad scheduling disaggregation that outside filing is concentrated are upgraded

At first to this N feasible schedule sequence difference aforesaid formula of substitution (2) and formula (3); All feasible schedule are estimated; And carry out non-bad ordering, adopt circulation crowding distance ranking method to come outside filing collection is safeguarded that the size of feasible outside filing collection is in the scope of its initial setting;

The renewal of outside filing collection is above-mentionedly also selected the active power value of the genset of non-bad optimum economic benefit of η % and discharge amount of pollution from the non-domination scheduling in the population except adopting from the filing set, the span of η is 5 ~ 20; Near the active power value of each genset, search for for radius, so that can produce better feasible schedule with SR ∈ [0,1];

Step 7, calculate outside filing and concentrate the crowding distance between each non-domination scheduling, and confirm the Pbest that makes new advances according to crowding distance _Ji(t), Pgbest _i(t) and C _Ji(t);

Step 8, judge whether iterations t reaches maximum of T _MaxIf,, export the feasible schedule that current noninferior solution is concentrated, if not, iterations t=t+1 is set, forward step 5 to and upgrade again.

The invention has the beneficial effects as follows; With thermal power plant as electric power system; Utilization is to the collection analysis of electric system and thermal power plant's data, in conjunction with the colony intelligence optimized Algorithm based on cluster thought, realizes finding the solution thermal power plant's economic load dispatching problem of integrated environment protection demand.Through exerting oneself of adjustment each genset of thermal power plant; Satisfying under the conditions such as line balancing and each unit output constraint; Rationally arrange the generated energy of each fired power generating unit of day part; Make and discharge amount of pollution and the economic benefit COMPREHENSIVE OPTIMAL that this cycle is interior when reducing disposal of pollutants, increase economic benefit.

Embodiment

The present invention is a kind of thermal power plant's integrated dispatch method based on environmental protection and economic benefit, may further comprise the steps:

Step 1, set up the mathematical model of thermal power plant's environmental economy scheduling problem

This scheduling problem is tangible two objective optimization problems, and its mathematical model is described as:

The expression formula of environmental economy scheduling problem is:

In the formula (1), i=1,2 ..., N _GRepresent i platform generator, N _GBe generator sum in the system;

The fuel used to generate electricity total burn-off of

expression thermal power plant is the fuel total expenses, and calculation expression is:

$F_i\left(P_i\right)=a_i+b_i+c_i{P}_i^2---\left(2\right)$

In the formula (2), P _iThe active power of representing i platform generator; F _i(P _i) fuel consumption of expression i platform generator, a _i, b _i, c _iBeing systematic parameter, representing the constant term of i platform genset fuel consumption characteristic, once coefficient and quadratic term coefficient respectively, is known parameters in system;

In addition;

is the total release of dusty gas, and calculation expression is:

$E_i\left(P_i\right)={\mathrm{\α}}_i+{\mathrm{\β}}_i+{\mathrm{\γ}}_i{P}_i^2+{\mathrm{\ξ}}_i\exp \left({\mathrm{\λ}}_i{P}_i\right),---\left(3\right)$

In the formula (3), E _i(P _i) discharge amount of pollution of expression i platform generator; α _i, β _i, γ _i, ξ _i, λ _iBe systematic parameter, α _i, β _i, γ _iRepresent the constant term of i platform genset disposal of pollutants flow characteristic, once coefficient and quadratic term coefficient respectively, ξ _i, λ _iThe correlation parameter of expression exponential term is a constant value with respect to concrete system.

Suffered restrictive condition comprises in the environmental economy optimizing scheduling process:

A) the minimax capacity limit of generator, expression formula is:

wherein is respectively the minimum and maximum active power output of i platform thermoelectric generator;

B) balance constraint; Be that each genset generated output sum of system should equal total demand power of load and via net loss sum, expression formula is:

In the formula, Total active power for power plant units; P _DRated power for system; P _LossBe active power loss in the electrical network, net decreases to calculate and adopts the B Y-factor method Y, and expression formula is: Wherein claim B _Ij, B _0i, B ₀₀Be the B coefficient, the concrete setting according to unit obtains respectively, is known parameters.

Step 2, obtain all kinds of parameters in the model

In the above-mentioned environmental economy scheduling problem numerous parameters is arranged, main obtain manner is:

2.1) load valley of confirming electrical network according to grid load curve and load peak period, confirm the power demand of day part load distribution

From the grid dispatching center statistics, obtain the data of current time system, mainly comprise the rated power P of system _D, active power loss P _LossParameter value B _Ij, B _0i, B ₀₀

2.2) running performance parameters of unit, the discharge of harmful gases amounts such as coal consumption amount, flue dust and carbon dioxide when mainly comprising operation are according to the soot emissions price, reduce discharging the supplemental characteristic α that price, CO2 emission exercise price obtain discharge amount of pollution _i, β _i, γ _i, ξ _i, λ _i, the coal consumption price in producing according to electric system assistant service price, unit of electrical energy, cost of electricity-generating when obtaining raw coal price, diesel-fuel price thermal power unit operation obtains the parameter alpha of fuel total expenses _i, b _i, c _i

The initial feasible schedule production process and the encoding process of step 3, economic environment scheduling problem

In above-mentioned mathematical model, the decision variable that needs to confirm is N _GThe active power of individual unit

The present invention adopts a kind of colony intelligence optimized Algorithm of innovation---and the particle cluster algorithm based on cluster thought is realized, need produce initial feasible schedule for this reason, and the concrete steps that produce to initial feasible schedule among the present invention are:

To preceding N _G-1 genset in the scope that satisfies maximum generation active power and minimum generating active power, produces the active power P of each unit at random _i, i=1,2 ..., N _G-1, utilize equality constraint then, try to achieve N _GThe active power of individual unit, computational grid loss P _Loss, and guarantee

Satisfy maximum, the restriction of minimum genset active power;

Produce N feasible schedule sequence according to same process.

Step 4, initial feasible schedule is estimated

N the feasible schedule sequence that above-mentioned steps 3 produces distinguished aforesaid target function type of substitution (2) and formula (3), just can realize feasible schedule is estimated,

Because therefore all corresponding two objective functions of each feasible schedule need carry out non-bad ordering to N feasible schedule, the feasible schedule series that complementation is arranged is kept in the outside set, be referred to as outside filing and collect,

Step 5, feasible schedule is upgraded

From the intelligent searching process of step 5 beginning, before beginning, iterations initial value t=0 need be set, and maximum iteration time T _Max

After producing N initial feasible schedule and estimating, next step will realize the renewal to this N feasible schedule, before renewal; The inventive method at first adopts the k-means clustering method; N feasible schedule carried out cluster, and the information that the fresh information that will obtain through cluster and particle cluster algorithm provide combines, come the genset active power of each feasible schedule is carried out the iteration update mechanism; This update mechanism is the most important innovative point of the present invention, specifically implements according to following steps:

Be located in the t time iteration, the active power of i unit is P in j feasible schedule _Ji(t), the rate of change of active power is V _Ji(t), before j feasible schedule in t iteration history during optimum economic benefit i unit active power be Pbest _JiThe active power of i unit is Pgbest when (t), obtaining optimum economic benefit in preceding t the iteration in all feasible schedule _i(t), the pairing genset active power of optimum economic benefit is C in the class of j feasible schedule place class _Ji(t), then when iterations is t+1, the active power P of i unit in j feasible schedule _Ji(t+1) iterative formula is following:

P _ji(t+1)＝P _ji(t)+V _ji(t+1) （4）

V _ji(t+1)＝w·V _ji(t)+c ₁·r ₁·(Pbest _ji(t)-P _ji(t))+c ₂r ₂(P _gbest(t)-P _ji(t))+c ₃·r ₃·(C _ji(t)-P _ji(t))

（5）

In following formula (4) and the formula (5), three study factor c ₁, c ₂And c ₃Be constant, preferred fixedly value c is adopted in suggestion ₁=c ₂=c ₃=4/3; r ₁, r ₂, r ₃Be the random number between [0,1]; The w suggestion is decremented to minimum value 0.4, v from maximal value 0.9 according to iterations _Ji(t) be generally the 10%-20% of genset active power variation range,

Through above two iterative formulas, the generating active power of the rate of change of each genset and genset is brought in constant renewal in, and is convenient to seek out N genset generated output feasible schedule set that more meets objective function under the integrated environment factor like this.

Innovative point of the present invention mainly is: through all feasible schedule being carried out cluster operation, the genset active power C of correspondence when in the renewal process of feasible schedule, introducing the optimal objective that obtains in each type _Ji(t), the introducing of this variable has changed the heading of each feasible schedule, makes dispatching method have more deep part to explore, and has strengthened the local optimum ability.

Step 6, the non-bad scheduling disaggregation that outside filing is concentrated are upgraded

The inventive method is utilized the extraction to relevant information in the electric system, the method that analysis combines the comprehensive particle group optimizing of thought of cluster, realizes dispatching the optimization of sequence.

Because it is optimum simultaneously with disposal of pollutants that target is an economic benefit, therefore this N feasible schedule sequence difference aforesaid formula of substitution (2) and the formula (3) to obtaining in the step 5 at first estimated all new feasible schedule, obtains non-bad feasible schedule sequence.Through non-bad ordering, adopt circulation crowding distance ranking method to come outside filing collection is safeguarded that the size of feasible outside filing collection is in the scope of its initial setting;

The renewal of outside filing collection is above-mentionedly also selected the active power value of the genset of non-bad optimum economic benefit of η % and discharge amount of pollution from the non-domination scheduling in the population except adopting from the filing set, the span of η is 5 ~ 20, and preferred value gets 8; Near the active power value of each genset, search for for radius, so that can produce better feasible schedule with SR ∈ [0,1].

Step 7, calculate outside filing and concentrate the crowding distance between each non-domination scheduling, and confirm the Pbest that makes new advances according to crowding distance _Ji(t), Pgbest _i(t) and C _Ji(t), so that the carrying out smoothly of iterative process next time;

Step 8, judge whether iterations t reaches maximum of T _MaxIf,, export the feasible schedule that current noninferior solution is concentrated, if not, iterations t=t+1 is set, forward step 5 to and upgrade again.

The advantage of the inventive method is:

Real-time has good global convergence performance and speed of convergence faster, uses it for the power system environment Economic Load Dispatch problem of finding the solution, and can effectively realize the scheduling real-time of electrical network.

Practicality; Under the electricity market condition; The power-management centre considers that according to the indexs such as coal consumption amount characteristic, dusty gas emission performance and quotation of each thermal power plant each generating plant of coordinating each other makes up, and the generated energy of making overall planning under the energy, environmental protection, the market each side interests is considered in unification.Be generating Fang Eryan like this, can practice thrift cost and increase the benefit power plant.As far as the user side, except obtaining satisfied electric weight, can also protect environment, energy savings, reach the effect that many-sided combined factors is considered.Meet the strategy of sustainable development more than surfing the Net at a competitive price according to each power plant merely, also more meet the requirement of the electric system marketization than the Optimization Dispatching that merely makes overall arrangements with energy savings.

Do not consider unit one-time investment and depreciation cost; It is the main target function that the inventive method obtains maximum economic benefit with electrical network; The increase of considering generated energy will inevitably cause the increase of discharge amount of pollution, i.e. the expense of environment decontamination increases, in order to realize economic maximization; Adjust the generated output of each power plant units in real time, thereby realize optimal scheduling.

Thermal power plant of the present invention integrated dispatch method; To integrated environment protection and economic benefit double factor, propose to utilize cluster multi-target particle crowd algorithm, the constraint that has scheduling with the peak and the low ebb of electricity consumption based on the knowledge extraction; The service requirement that can reflect system more conscientiously; Realization realizes the Economic Load Dispatch to thermal power plant to the environmental economy scheduling of thermal power plant, realizes the comprehensive benefit maximization.

Claims (3)

1. the thermal power plant's integrated dispatch method based on environmental protection and economic benefit is characterized in that, may further comprise the steps:

Step 1, set up the mathematical model of thermal power plant's environmental economy scheduling problem

The expression formula of its mathematical model is:

In the formula (1), i=1,2 ..., N _GRepresent i platform generator, N _GBe generator sum in the system;

The fuel used to generate electricity total burn-off of

expression thermal power plant is the fuel total expenses, and calculation expression is:

$F_i\left(P_i\right)=a_i+b_i+c_i{P}_i^2---\left(2\right)$

In the formula (2), P _iThe active power of representing i platform generator; E _i(P _i) fuel consumption of expression i platform generator, a _i, b _i, c _iBeing systematic parameter, representing the constant term of i platform genset fuel consumption characteristic, once coefficient and quadratic term coefficient respectively, is known parameters in system;

In addition;

is the total release of dusty gas, and calculation expression is:

$E_i\left(P_i\right)={\mathrm{\α}}_i+{\mathrm{\β}}_i+{\mathrm{\γ}}_i{P}_i^2+{\mathrm{\ξ}}_i\exp \left({\mathrm{\λ}}_i{P}_i\right),---\left(3\right)$

In the formula (3), E _i(P _i) discharge amount of pollution of expression i platform generator; α _i, β _i, γ _i, ξ _i, λ _iBe systematic parameter, α _i, β _i, γ _iRepresent the constant term of i platform genset disposal of pollutants flow characteristic, once coefficient and quadratic term coefficient respectively, ξ _i, λ _iThe correlation parameter of expression exponential term is a constant value with respect to concrete system;

Suffered restrictive condition comprises in the environmental economy optimizing scheduling process:

A) the minimax capacity limit of generator, expression formula is:

wherein is respectively the minimum and maximum active power output of i platform thermoelectric generator;

B) balance constraint; Be that each genset generated output sum of system should equal total demand power of load and via net loss sum, expression formula is:

In the formula,

Total active power for power plant units; P _DRated power for system; P _LossBe active power loss in the electrical network,

Net decreases to calculate and adopts the B Y-factor method Y, and expression formula is:

Wherein claim B _Ij, B _0i, B ₀₀Be the B coefficient, the concrete setting according to unit obtains respectively, is known parameters;

Step 2, obtain all kinds of parameters in the model

2.1) load valley of confirming electrical network according to grid load curve and load peak period, confirm the power demand of day part load distribution

From the grid dispatching center statistics, obtain the data of current time system, comprise the rated power P of system _D, active power loss P _LossParameter value B _Ij, B _0i, B ₀₀

2.2) running performance parameters of unit, coal consumption amount, flue dust and discharge of harmful gases amount when comprising operation are according to the soot emissions price, reduce discharging the supplemental characteristic α that price, CO2 emission exercise price obtain discharge amount of pollution _i, β _i, γ _i, ξ _i, λ _iCoal consumption price in producing according to electric system assistant service price, unit of electrical energy, cost of electricity-generating when obtaining raw coal price, diesel-fuel price thermal power unit operation obtains the parameter a of fuel total expenses _i, b _i, c _i

Step 3, initial feasible schedule production process and encoding process

In above-mentioned mathematical model, the decision variable that needs to confirm is N _GThe active power of individual unit

Employing realizes based on the particle cluster algorithm of cluster thought, need produce initial feasible schedule for this reason, and the concrete steps that produce to initial feasible schedule are:

To preceding N _G-1 genset in the scope that satisfies maximum generation active power and minimum generating active power, produces the active power P of each unit at random _i, i=1,2 ..., N _G-1, utilize equality constraint then, try to achieve N _GThe active power of individual unit, computational grid loss P _Loss, and guarantee

Satisfy maximum, the restriction of minimum genset active power;

Produce N feasible schedule sequence according to same process;

Step 4, initial feasible schedule is estimated

N the feasible schedule sequence that above-mentioned steps 3 produces distinguished aforesaid formula of substitution (2) and formula (3), feasible schedule is estimated, and carried out non-bad ordering;

Because therefore all corresponding two objective functions of each feasible schedule need carry out non-bad ordering to N feasible schedule, the feasible schedule of not arranging mutually is kept in the outside set, be referred to as outside filing collection;

Step 5, feasible schedule is upgraded

Before beginning, iterations initial value t=0 need be set, and maximum iteration time T _Max,

After producing N initial feasible schedule and estimating; Next step will realize the renewal to this N feasible schedule, before upgrading, at first adopt the k-means clustering method; N feasible schedule carried out cluster; And the information that the fresh information that will obtain through cluster and particle cluster algorithm provide combines, and comes the genset active power of each feasible schedule is carried out the iteration update mechanism, specifically implements according to following steps:

Be located in the t time iteration, the active power of i unit is P in j feasible schedule _Ji(t), the rate of change of active power is V _Ji(t), before j feasible schedule in t iteration history during optimum economic benefit i unit active power be Pbest _JiThe active power of i unit is Pgbest when (t), obtaining optimum economic benefit in preceding t the iteration in all feasible schedule _i(t), the pairing genset active power of optimum economic benefit is C in the class of j feasible schedule place class _Ji(t), then when iterations is t+1, the active power P of i unit in j feasible schedule _Ji(t+1) iterative formula is following:

P _ji(t+1)＝P _ji(t)+V _ji(t+1) （4）

V _ji(t+1)＝ω·V _ji(t)+c ₁·r ₁·(Pbest _ji(t)-P _ji(t))+c ₂r ₂(P _gbest(t)-P _ji(t))+c ₃·r ₃·(C _ji(t)-P _ji(t))

（5）

In following formula (4) and the formula (5), three study factor c ₁, c ₂And c ₃Be constant, r ₁, r ₂, r ₃Be the random number between [0,1]; W is decremented to minimum value 0.4, V from maximal value 0.9 according to iterations _Ji(t) be the 10%-20% of genset active power variation range,

Through above two iterative formulas, the generating active power of the rate of change of each genset and genset is brought in constant renewal in, and seeks out N genset generated output feasible schedule set that more meets objective function under the integrated environment factor;

Step 6, the non-bad scheduling disaggregation that outside filing is concentrated are upgraded

At first to this N feasible schedule sequence difference aforesaid formula of substitution (2) and formula (3); All feasible schedule are estimated; And carry out non-bad ordering, adopt circulation crowding distance ranking method to come outside filing collection is safeguarded that the size of feasible outside filing collection is in the scope of its initial setting;

The renewal of outside filing collection is above-mentionedly also selected the active power value of the genset of non-bad optimum economic benefit of η % and discharge amount of pollution from the non-domination scheduling in the population except adopting from the filing set, the span of η is 5 ~ 20; Near the active power value of each genset, search for for radius, so that produce better feasible schedule with SR ∈ [0,1];

Step 7, calculate outside filing and concentrate the crowding distance between each non-domination scheduling, and confirm the Pbest that makes new advances according to crowding distance _Ji(t), Pgbest _i(t) and C _Ji(t);

Step 8, judge whether iterations t reaches maximum of T _MaxIf,, export the feasible schedule that current noninferior solution is concentrated, if not, iterations t=t+1 is set, forward step 5 to and upgrade again.

2. the thermal power plant's integrated dispatch method based on environmental protection and economic benefit according to claim 1 is characterized in that: in the described step 5, and preferred fixing value c ₁=c ₂=c ₃=4/3.

3. the thermal power plant's integrated dispatch method based on environmental protection and economic benefit according to claim 1 is characterized in that: in the described step 6, the value of preferred η is 8.