CN102880987B - A kind of novel gas turbine Strategies of Maintenance formulating method - Google Patents

Abstract

The invention discloses a kind of novel gas turbine Strategies of Maintenance formulating method, comprise the following steps: (1) determines the service condition of gas turbine unit; (2) top overhaul plan is formulated; (3) set up gas turbine Strategies of Maintenance mathematic optimal model, the foundation of mathematic optimal model comprises the following steps: (3-1) simplifies influence factor; (3-2) the primary mold objective function of constitution optimization model is the objective function of Optimized model time minimum with the recondition expense in setting-up time; (3-3) constraint condition of mathematic optimal model is set up; (3-4) in primary mold objective function, penalty function is introduced, the final goal function of the model that is optimized; (4) apply genetic algorithm to solve above-mentioned mathematic optimal model, thus obtain the Optimal Maintenance strategy of gas turbine.The maintenance cost that method of the present invention realizes Gas Turbine Plant minimizes and has the advantage compared with high operating rate.

Description

A kind of novel gas turbine Strategies of Maintenance formulating method

Technical field

The present invention relates to gas turbine service technique field, particularly a kind of novel gas turbine Strategies of Maintenance formulating method.

Background technology

Before the nineties in 20th century, the electrical production of China is in planned economy mode, and the maintenance principle advocating gas turbine is " putting prevention first, scheduled overhaul ", " should repair required, it is required good to repair ".What generating plant was carried out is periodic plan maintenance, namely " expires required ", if it is exactly in bad repair for not carrying out maintenance, and to corresponding responsibility be born, unit or equipment one are to the time between overhauls(TBO), and status of equipment quality all must arrange maintenance, so very easily causes the mistake of equipment to repair or owe to repair.At present, the Strategies of Maintenance generally adopted is using the dependability parameter of equipment as foundation, and the working experience in conjunction with operating maintenance personnel carrys out guided maintenance.The gas-steam combined cycle set of unit style, each unit independent operating, whether run with working time is all adjust on the basis of instruction to be formulated by working experience in meeting, although such method of operation can meet middle tune instruction at that time, but do not consider Strategies of Maintenance, therefore the turn(a)round that likely each unit is respective after running a period of time is in the same period, the same period is caused to only have a unit even not have unit energy to run.

Design, the manufacturing technology of current China gas turbine are also relatively weak compared with developed countries, the gas turbine unit of large-scale Natural Gas Power Plant mainly relies on external import, the hot-end component particularly keeped in repair is expensive, can be increased the service life by effective Strategies of Maintenance, directly affect cost and the market competitiveness of Gas Turbine Plant.The production domesticization ground zero of current spare part, and it is too high to buy the price of spare unit to genuine man, therefore how to carry out the Strategies of Maintenance of gas turbine, to the safety of Gas Turbine Plant and economical operation has very important meaning.

Summary of the invention

The shortcoming that the object of the invention is to overcome prior art, with not enough, provides a kind of upkeep cost to minimize and the high novel gas turbine Strategies of Maintenance formulating method of operating efficiency.

Object of the present invention is achieved through the following technical solutions: a kind of novel gas turbine Strategies of Maintenance formulating method, comprises the following steps:

(1) set up gas turbine Strategies of Maintenance mathematic optimal model, wherein the foundation of mathematic optimal model comprises the following steps:

(1-1) influence factor is simplified;

(1-2) the primary mold objective function of structure mathematics Optimized model: be the primary mold objective function of mathematic optimal model time minimum with the recondition expense in setting-up time;

(1-3) set up the constraint condition of mathematic optimal model, comprising:

1) days running constraint;

2) stagger repair time of each unit;

3) total natural gas supply amount is met;

4) do not overhaul peak period;

(1-4) in primary mold objective function, penalty function is introduced, the final goal function of the model that is optimized;

(2) apply genetic algorithm and obtain Optimal Maintenance strategy: non-negative conversion has been carried out to the final goal function that step (1-4) obtains, be the fitness function of target minimizing that optimization object function is transformed to maximal value, search out the Strategies of Maintenance that recondition expense is minimum, thus obtain the Optimal Maintenance strategy of gas turbine.

Preferably, the primary mold objective function of step (1-2) is:

$P_{\min }=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\underset{j=1}{\overset{m}{\mathrm{\Σ}}}\underset{k=1}{\overset{z}{\mathrm{\Σ}}}(U_{\mathrm{jk}}^i\×v_k),i=\mathrm{1,2}...n;j=\mathrm{1,2}...m;k=\mathrm{1,2}...z;$

Wherein n is the number of units of unit, m be formulate Strategies of Maintenance time in run peak period and off-peak period total number, z is the quantity that every platform unit needs the parts of maintenance; I represents the number of units number of unit, and j represents the time hop count that unit is in, the type of the parts of every platform unit needs maintenance that what k represented is; variable represents the running status of the kth base part of jth period i-th gas turbine, v _kfor required expense when kth base part overhauls; Wherein this Optimized model comprises n × m × z variable

time, represent that i-th unit kth base part does not need maintenance when running to the jth period;

time, represent that i-th unit kth base part needs maintenance when running to the jth period; After unit maintenance, calculate from the j+1 period.

represent that i-th unit is the days running of period; represent the number of days that i-th unit altogether runs when the 1st period ran to the jth period, be the serviceable life of i-th unit kth base part, serviceable life represents with number of days.

Preferably, the method for operation that the simplification influence factor of step (1-1) refers to unit is that day start and stop run, and the start and stop of unit operation day characterize one-shot in one day; Natural gas supply amount is plan or reply gas; The constraint condition of step (1-3) comprising:

1) days running constraint:

$D_j(1-5\%)\≤\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i\≤D_j;i=\mathrm{1,2,3}...n;j=\mathrm{1,2,3}...m;$

Wherein argument table is shown in the days running of jth period i-th gas turbine; D _jwhat represent is all groups of total days running sums in the jth period.

2) stagger repair time of each unit:

$\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{U}_{\mathrm{jk}}^i\≤y;$

Wherein y represents the unit number of units that the same period allows to overhaul, variable represents the running status of the kth base part of jth period i-th gas turbine; represent that i-th unit kth base part does not need maintenance when running to the jth period; represent that i-th unit kth base part needs maintenance when running to the jth period.

3) total natural gas supply amount is met:

90%Q′ _j≤Q _j≤110%Q′ _j；

Wherein Q _jrepresent the gas consumption of all units in this period of jth; Q ' _jrepresent at the reply of jth period or plan gas consumption;

$Q_j=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i\×Q_j^{*},i=\mathrm{1,2,3}...n;j=\mathrm{1,2,3}...m$

represent the day air consumption of single unit, argument table is shown in the days running of jth period i-th gas turbine.

4) do not overhaul peak period: when namely j is in peak period,

Preferably, the final goal function of step (1-4) is:

$\mathrm{Min}:P_{\min }^{\′}=P_{\min }+g\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i\right)+\underset{i=1}{\overset{n}{\mathrm{\Σ}}}h\left[s(X_1^i,...,X_m^i)\right],i=\mathrm{1,2,3}...n;j=\mathrm{1,2,3}...m;$

Wherein function g and h is the penalty function introduced;

Described penalty function g is:

Wherein r _jfor penalty factor, and be defined as:

$t\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i\right)=\left\{\begin{array}{cc}{D}_j(1-5\%)-\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i& (\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i\≤D_j(1-5\%))\\ \underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i-D_j& (D_j\≤\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{X}_j^i)\end{array}\right.,i=\mathrm{1,2,3}...n;j=\mathrm{1,2,3}...m;$

D _jrepresent the total days running sum of all units of jth period;

Described penalty function h is:

$h\left(s(X_1^i,...,X_m^i)\right)=\left\{\begin{array}{cc}0& (s(X_1^i,...,X_m^i)=0)\\ \mathrm{Max}& (s(X_1^i,...,X_m^i)=1)\end{array}\right.,i=\mathrm{1,2,3}...n;j=\mathrm{1,2,3}...m;$

Function s is expressed as variable drawing after carrying out accumulation calculating needs the situation of carrying out overhauling in the j period, if the j period is peak period, then s value is 1, otherwise s is then 0;

Wherein, Max is a constant.

Preferably, the value of described Max is: Max=(90 ~ 110) P _min.

Preferably, the fitness function F (X) in described step (2) is:

F(X)=(C _max-P′ _min)；

Wherein C _maxbe a constant, for carrying out non-negative change.

Preferably, described C _maxvalue be: C _max=(90 ~ 110) P _min.

For convenience of intelligent algorithm optimizing in above-mentioned steps (1-3), prevent whole model rigidity comparatively large, the binding character of constraint condition is too strong, therefore, changes equality constraint into inequality constrain, by equality constraint: be revised as it and become inequality constrain:

The present invention has following advantage and effect relative to prior art:

(1) Strategies of Maintenance formulating method of the present invention by gas turbine Strategies of Maintenance formulate be reduced to a multi-constraint condition under ask maintenance cost minimum optimized mathematical model, determine its constraint condition, optimized variable and optimization aim thereof, and adopt the genetic algorithm with powerful global optimization search capability to obtain gas turbine Optimal Maintenance strategy, good applicability is had to the formulation of Strategies of Maintenance again due to genetic algorithm, its intelligent behaviour obtains the relative optimum solution close with optimum solution within the extremely short time, greatly can shorten the time that whole Strategies of Maintenance is formulated, therefore the maintenance cost that Strategies of Maintenance formulating method of the present invention can realize Gas Turbine Plant minimizes, and the advantage had compared with high operating rate.

(2) by introducing penalty function, restricted problem being converted into unconfinement problem solving in Strategies of Maintenance formulating method of the present invention, to solve the constraint condition can not carrying out peak period overhauling, and improve the speed solved.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of Strategies of Maintenance formulating method of the present invention.

Fig. 2 is genetic algorithm calculation flow chart in Strategies of Maintenance formulating method of the present invention.

Embodiment

Below in conjunction with embodiment and accompanying drawing, the present invention is described in further detail, but embodiments of the present invention are not limited thereto.

Embodiment

Be illustrated in figure 1 the process flow diagram of a kind of novel gas turbine Strategies of Maintenance formulating method of the present embodiment, the present embodiment is the process that in following 12 years of 3, certain variable load plant gas turbine unit 2013 to 2024, Strategies of Maintenance is formulated, and step is as described below:

(1) set up gas turbine Strategies of Maintenance mathematic optimal model, wherein the foundation of mathematic optimal model comprises the following steps:

(1-1) influence factor is simplified: the unit operation mode of the present embodiment is that day start and stop run; Do not consider that the unexpected shutdown that the unexpected accident of unit causes is restarted, namely think that the start and stop of unit operation day just characterize one-shot in one day; Natural gas supply amount only considers plan or reply gas, is not included within rock gas consumption by stock gas.

(1-2) the primary mold objective function of constitution optimization model is the primary mold objective function of Optimized model time minimum with the recondition expense in setting-up time; Every platform unit is added up one by one at the days running of each period, if when in process cumulative one by one, accumulation result is less than the serviceable life of parts, then represent and do not need maintenance, proceed to add up, otherwise, if accumulation result is greater than the serviceable life of a certain parts, then represent that these parts need maintenance, record this newly to change parts be newly purchase or newly repair simultaneously.Certain parts so known keeped in repair in certain period, and were newly purchase or newly repair to carry out pricing according to these parts, thus calculated maintenance total expenses.Concrete primary mold objective function is:

$P_{\min }=\underset{i=1}{\overset{3}{\mathrm{\Σ}}}\underset{j=1}{\overset{25}{\mathrm{\Σ}}}\underset{k=1}{\overset{8}{\mathrm{\Σ}}}(U_{\mathrm{jk}}^i\×v_k),i=\mathrm{1,2},3;j=\mathrm{1,2}...25;k=\mathrm{1,2}...8;$

I represents the number of units number of unit, and this enforcement includes 3 units altogether; J represents the time of the formulating Strategies of Maintenance setting interior time hop count run, the time interior number summation running peak period and off-peak period that the present embodiment formulates Strategies of Maintenance setting is 25,1 off-peak period in March, 2013 to June, 12 peak periods in July to October then during 2013 to 2024 years, 11 off-peak periods in the June in November to coming year then during 2013 to 2024,1 off-peak period in November to Dec of 2024; Wherein peak period and off-peak period interval appearance, when j is even number, appearance be peak period.The type of the parts of every platform unit needs maintenance that what the present embodiment k represented is, the every platform unit of the present embodiment includes the parts that 8 class needs overhaul.

variable represents the running status of the kth base part of jth period i-th gas turbine, v _kfor required expense when kth base part overhauls; This Optimized model comprises 3 × 8 × 25=600 variable

time, represent that i-th unit kth base part does not need maintenance when running to the jth period.

time, represent that i-th unit kth base part needs maintenance when running to the jth period; After unit maintenance, calculate from the j+1 period.

represent that i-th unit is the days running of this period; represent the number of days that i-th unit altogether runs when the 1st period ran to j period, be the serviceable life of i-th unit kth base part, serviceable life represents with number of days.

(1-3) set up the constraint condition of mathematic optimal model, comprising:

1) days running is in design time, and days running constraint condition is:

$\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i=D_j;i=\mathrm{1,2,3};j=\mathrm{1,2}...25;$

For convenience of intelligent algorithm optimizing, prevent whole model rigidity comparatively large, the binding character of constraint condition is too strong, and therefore, change equality constraint into inequality constrain, it becomes:

$D_j(1-5\%)\≤\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i\≤D_j,i=\mathrm{1,2,3};j=\mathrm{1,2}...25;$

Wherein variable represents the days running of jth period i-th gas turbine, D _jwhat represent is all groups of total days running sums of jth period.

2) stagger repair time of each unit: this power plant is totally 3 units, if occur that the unit of two or more overhauls with the period, can cause only having one or do not have unit energy to run like this, thus of that month middle tune requirement can not be met, therefore the same period can not occur that the unit of two or more keeps in repair simultaneously, namely $\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{U}_{\mathrm{jk}}^i\≤1.$

3) total natural gas supply amount is met: 90%Q ' _j≤ Q _j≤ 110%Q ' _j;

Wherein Q _jrepresent the gas consumption of all units in this period of jth; Q ' _jrepresent at the reply of jth period or plan gas consumption;

$Q_j=\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i\×Q_j^{*},i=\mathrm{1,2,3};j=\mathrm{1,2}...25$

represent the day air consumption of single unit, argument table is shown in the days running of jth period i-th gas turbine.

4) do not overhaul peak period: the main application of this power plant is used as peak-clipping and valley-filling, if overhaul peak period, must can not meet middle tune instruction load, therefore, can not be carried out overhauling one of constraint condition being set to Optimized model peak period.July to the October of the present embodiment is the peak period of unit, namely when j is even number, then

(1-4) in primary mold objective function, introduce penalty function, thus restricted problem be converted into unconfinement problem and solve, to solve the constraint condition can not carrying out peak period overhauling, the final goal function of the model that is optimized:

$\mathrm{Min}:P_{\min }^{\′}=P_{\min }+g\left(\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i\right)+\underset{i=1}{\overset{3}{\mathrm{\Σ}}}h\left[s(X_1^i,...,X_{25}^i)\right],i=\mathrm{1,2,3};j=\mathrm{1,2}...25;$

Wherein function g and h is the penalty function introduced, variable represents the days running of jth period i-th gas turbine.

Described penalty function g is:

Wherein r _jfor penalty factor, and be defined as:

$t\left(\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i\right)=\left\{\begin{array}{cc}{D}_j(1-5\%)-\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i& (\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i\≤D_j(1-5\%))\\ \underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i-D_j& (D_j\≤\underset{i=1}{\overset{3}{\mathrm{\Σ}}}{X}_j^i)\end{array}\right.,i=\mathrm{1,2,3};j=\mathrm{1,2}...25;$

D _jwhat represent is the total days running sum of all units of jth period.

Described penalty function h is:

$h\left(s(X_1^i,...,X_{25}^i)\right)=\left\{\begin{array}{cc}0& (s(X_1^i,...,X_{25}^i)=0)\\ \mathrm{Max}& (s(X_1^i,...,X_{25}^i)=1)\end{array}\right.,i=\mathrm{1,2,3};j=\mathrm{1,2}...25;$

Wherein Max=(90 ~ 110) P _min, the Max=100P of the present embodiment _min; S is expressed as these variablees drawing after carrying out accumulation calculating needs the situation of carrying out overhauling in the j period, if the j period is peak period, then s value is 1, otherwise s is then 0.If draw and need the period j carrying out overhauling to be peak period, so just by penalty function h, the solution obtained is rejected, then go to seek other excellent solution.

(2) apply genetic algorithm and obtain Optimal Maintenance strategy: non-negative conversion has been carried out to the final goal function that step (1-4) obtains, be the fitness function of target minimizing that optimization object function is transformed to maximal value, utilize genetic algorithm to search out the minimum Strategies of Maintenance of recondition expense, thus obtain the Optimal Maintenance strategy of gas turbine; Wherein fitness function is:

F(X)=C _max-P′ _min)；

Wherein C _max=(90 ~ 110) P _min, for carrying out non-negative change, the C of the present embodiment _max=100P _min.

As shown in Figure 2, this algorithm of genetic algorithm of the present embodiment adopts rotating disc type selection strategy, and crossover operator adopts single-point to intersect in conjunction with even arithmetic crossover, first determines the parent carrying out intersecting at random according to crossover probability, and match between two, then in individual UVR exposure string, point of crossing is set at random.If crossover location, at first 4, represents and intersects to often kind of numbering, now, the concrete operations intersected according to single-point make a variation.If crossover location, at latter 4, represents and intersects to often kind of ratio, then to 2 rear 4 individualities that generation 2 is new as follows of individuality, new individual production process is as follows:

A′=aB+（1-a）A，B′=aA+（1-a）B；

Wherein A, B represent individual, A ', B ' for individual A, B make a variation after the new individuality that produces.

Controling parameters is as follows:

Population size: N=100; Probability of crossover: P _c=0.60

Mutation probability: P _m=0.005; Evolutionary generation: H=50

For convenience of intelligent algorithm optimizing in the present embodiment step (1-3), prevent whole model rigidity comparatively large, the binding character of constraint condition is too strong, therefore, changes equality constraint into inequality constrain, by equality constraint: be revised as it and become inequality constrain:

In the prioritization scheme of the present embodiment, the constraint such as there is load and can not overhaul peak period, therefore in intelligent algorithm, also need the validity verifying sample, but the difficulty finding an effective sample is individual not second to optimizing, therefore introduces penalty function and restricted problem is converted into unconfinement problem solving.

The formulation of the genetic algorithm that this enforcement adopts to Strategies of Maintenance has good applicability, its intelligent behaviour obtains the relative optimum solution close with optimum solution within the extremely short time, greatly can shorten the time that whole Strategies of Maintenance is formulated, substantially increase the operating rate of power plant.

Above-described embodiment is the present invention's preferably embodiment; but embodiments of the present invention are not restricted to the described embodiments; change, the modification done under other any does not deviate from Spirit Essence of the present invention and principle, substitute, combine, simplify; all should be the substitute mode of equivalence, be included within protection scope of the present invention.

Claims (6)

1. a novel gas turbine Strategies of Maintenance formulating method, is characterized in that, comprise the following steps:

(1) set up gas turbine Strategies of Maintenance mathematic optimal model, wherein the foundation of mathematic optimal model comprises the following steps:

(1-1) influence factor is simplified;

(1-2) the primary mold objective function of structure mathematics Optimized model: be the primary mold objective function of mathematic optimal model time minimum with the recondition expense in setting-up time;

(1-3) set up the constraint condition of mathematic optimal model, comprising:

1) days running constraint;

2) stagger repair time of each unit;

3) total natural gas supply amount is met;

4) do not overhaul peak period;

(1-4) in primary mold objective function, penalty function is introduced, the final goal function of the model that is optimized;

(2) apply genetic algorithm and obtain Optimal Maintenance strategy: non-negative conversion has been carried out to the final goal function that step (1-4) obtains, be the fitness function of target minimizing that optimization object function is transformed to maximal value, search out the Strategies of Maintenance that recondition expense is minimum, thus obtain the Optimal Maintenance strategy of gas turbine;

The primary mold objective function of step (1-2) is:

$P_{\min }=\underset{i=1}{\overset{n}{\Σ}}\underset{j=1}{\overset{m}{\Σ}}\underset{k=1}{\overset{z}{\Σ}}(U_{jk}^i\×v_k),$ i＝1,2...n；j＝1、2...m；k＝1、2...z；

Wherein n is the number of units of unit, m be formulate Strategies of Maintenance time in run peak period and off-peak period total number, z is the quantity that every platform unit needs the parts of maintenance; I represents the number of units number of unit, and j represents the time hop count that unit is in, the type of the parts of every platform unit needs maintenance that what k represented is; variable represents the running status of the kth base part of jth period i-th gas turbine, v _kfor required expense when kth base part overhauls; Wherein this Optimized model comprises n × m × z variable

time, represent that i-th unit kth base part does not need maintenance when running to the jth period;

time, represent that i-th unit kth base part needs maintenance when running to the jth period; After unit maintenance, calculate from the j+1 period;

represent that i-th unit is the days running of period; represent the number of days that i-th unit altogether runs when the 1st period ran to the jth period, be the serviceable life of i-th unit kth base part, serviceable life represents with number of days.

2. novel gas turbine Strategies of Maintenance formulating method according to claim 1, it is characterized in that, the method of operation that the simplification influence factor of step (1-1) refers to unit is that day start and stop run, and the start and stop of unit operation day characterize one-shot in one day; Natural gas supply amount is plan or reply gas; The constraint condition of step (1-3) comprising:

1) days running constraint:

$D_j(1-5\%)\&le;\underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i<D_j;$ i＝1,2,3...n；j＝1、2、3...m；

Wherein argument table is shown in the days running of jth period i-th gas turbine; D _jwhat represent is in the total days running sum of all units of jth period;

2) stagger repair time of each unit:

$\underset{i=1}{\overset{n}{\&Sigma;}}{U}_{jk}^i\&le;y;$

Wherein y represents the unit number of units that the same period allows to overhaul, variable represents the running status of the kth base part of jth period i-th gas turbine; represent that i-th unit kth base part does not need maintenance when running to the jth period; represent that i-th unit kth base part needs maintenance when running to the jth period;

3) total natural gas supply amount is met:

90％Q' _j≤Q _j≤110％Q' _j；

Wherein Q _jrepresent the gas consumption of all units in this period of jth; Q ' _jrepresent at the reply of jth period or plan gas consumption;

$Q_j=\underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i\&times;Q_j^{*},$ i＝1,2,3...n；j＝1、2、3...m；

represent the day air consumption of single unit, argument table is shown in the days running of jth period i-th gas turbine;

4) do not overhaul peak period: when namely j is in peak period,

3. novel gas turbine Strategies of Maintenance formulating method according to claim 2, it is characterized in that, the final goal function of step (1-4) is:

$Min:P_{min}^{\&prime;}=P_{\min }+g\left(\underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i\right)+\underset{i=1}{\overset{n}{\&Sigma;}}h\&lsqb;s(X_1^i,...,X_m^i)\&rsqb;,$ i＝1,2,3...n；j＝1、2、3...m；

Wherein function g and h is the penalty function introduced;

Described penalty function g is:

i＝1,2,3...n；j＝1、2、3...m；

Wherein r _jfor penalty factor, and be defined as:

$t\left(\underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i\right)=\left\{\begin{array}{cc}{D}_j\left(1-5\%\right)-\underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i& \left(\underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i\&le;D_j\left(1-5\%\right)\right)\\ \underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i-D_j& \left(D_j\&le;\underset{i=1}{\overset{n}{\&Sigma;}}{X}_j^i\right)\end{array}\right.,$ i＝1,2,3...n；j＝1、2、3...m；

D _jrepresent the total days running sum of all units of jth period;

Described penalty function h is:

$h\left(s\right(X_1^i,...,X_m^i\left)\right)=\left\{\begin{array}{cc}0& \left(s\right(X_1^i,...,X_m^i)=0)\\ Max& \left(s\right(X_1^i,...,X_m^i)=1)\end{array}\right.,$ i＝1,2,3...n；j＝1、2、3...m；

Function s is expressed as to variable drawing after carrying out accumulation calculating needs the situation of carrying out overhauling in the j period, if the j period is peak period, then s value is 1, otherwise s is then 0;

Wherein, Max is a constant.

4. novel gas turbine Strategies of Maintenance formulating method according to claim 3, it is characterized in that, the value of described Max is: Max=(90 ~ 110) P _min.

5. novel gas turbine Strategies of Maintenance formulating method according to claim 3, is characterized in that, the fitness function F (X) in described step (2) is:

F(X)＝(C _max-P′ _min)；

Wherein C _maxbe a constant, for carrying out non-negative change.

6. novel gas turbine Strategies of Maintenance formulating method according to claim 5, is characterized in that, described C _maxvalue be: C _max=(90 ~ 110) P _min.