CN104573844A - Quarterly power transmission and transformation integrated maintenance optimization method based on genetic algorithm - Google Patents

Abstract

The invention discloses a quarterly power transmission and transformation integrated maintenance optimization method based on a genetic algorithm, belongs to the field of optimal operation of power systems, and solves the problems of high and unreasonable workload, and the like of a maintenance mode of a power grid. The method comprises the following steps: determining flow sequences of the power grid during the target time period according to quarterly load prediction of the power grid, startup mode arrangement and topological data of the power grid; determining flow sequences of the power grid in a maintenance mode; setting a fault set; checking whether system flow in the maintenance mode meets the thermal stability requirements or not, and working out overload quantity of an overload element; creating a quarterly power transmission and transformation integrated maintenance optimization function of the power grid, and substituting the overload quantity of the element into the function for solving; updating a maintenance mode group by the genetic algorithm, and recombining the flow sequences of the power grid in the maintenance mode; performing loop iteration and calculation to obtain an optimal quarterly power transmission and transformation integrated maintenance scheme of the power grid. Therefore, the working efficiency of power grid operation personnel is improved, the obtained maintenance scheme of the power grid is reasonable and reliable, and the social and economic benefits are remarkable.

Description

Based on the season power transmission and transformation integration optimized maintenance method of genetic algorithm

Technical field

The invention belongs to electric power system optimization and run field, particularly a kind of season power transmission and transformation integration optimized maintenance method based on genetic algorithm.

Background technology

Run field in electric power system dispatching, the overhaul of the equipments of electric system is the important foundation ensureing power grid security, stable operation.The present stage arrangement of maintenance scheduling for power systems is carried out in ultra-short term planning, the maintenance grade that the method reports according to each grass-roots unit and time, carries out the turnaround plan arrangement of next stage.Because the method concept is simple and clear, principle simple, thus in the whole nation, electrical network at different levels is widely used.But along with the continuous expansion of electric system scale and the raising of complicacy, the method need carry out a large amount of off-line manual analyses to grid maintenance demand information, amount of calculation is large, and maintenance mode arrangement is reasonable not, has a progressive space of improving lifting.Therefore in the urgent need to the problem having new technology, new method innovation and application is optimized in maintenance scheduling for power systems, explore, put into practice the possible developing direction of following electrical network.

The Maintenance Schedule Optimization of electric system refers in certain special time period, guarantees that the repair apparatus declared completes maintenance smoothly, meanwhile reduces the impact of the safety and stability that overhaul of the equipments brings to electrical network as far as possible.At present, in literature review, propose the concept of medium-term and long-term of electrical network transmission of electricity optimized maintenance abroad, achieve the coordination optimization of medium-term and long-term transmission of electricity turnaround plan and Short Term Generation Schedules.In the scheduling turnaround plan of China's actual electric network arranges, due in there is many uncertain factors in long-term turnaround plan arrangement, the security and stability that electrical network is real-time can not be ensured.Therefore, the present invention takes into full account electrical network actual demand, be the time scale that grid maintenance mode arranges with season, propose the season power transmission and transformation integration optimized maintenance method based on genetic algorithm, solve that the grid maintenance mode amount of arranging work is large and scheme arranges the problems such as unreasonable.

Summary of the invention

Goal of the invention:

Large and scheme arranges irrational problem in order to solve the electrical network season maintenance mode amount of arranging work, overcome traditional simple dependence manually to calculate to arrange the ineffective drawback of grid maintenance mode, the present invention proposes a kind of season power transmission and transformation integration optimized maintenance method based on genetic algorithm, by building season power transmission and transformation integration maintenance wish function, continuous iteration, calculating, finally draw optimum grid maintenance scheme, the basis ensureing optimality greatly improves work efficiency, there is great popularizing application prospect in actual electric network.

Technical scheme:

The present invention is achieved through the following technical solutions:

Based on a season power transmission and transformation integration optimized maintenance method for genetic algorithm, by building season power transmission and transformation integration maintenance wish function, iteration, calculating optimum grid maintenance scheme, it is characterized in that: the method step is as follows:

S1. according to the load prediction of electrical network season, the arrangement of electrical network start-up mode, grid topology data, object time section sequence of current is determined;

S2. according to turnaround plan and the relevant information of intending report, the sequence of current under grid maintenance mode is determined;

S3. power grid fault set is set, comprises arbitrary element in N-1 electrical network, parallel erected on same tower circuit in N-2 electrical network;

S4. under checking maintenance mode, whether system load flow meets power grid heat stability provisioning request, and obtains the overload quantity μ of overload element _i, wherein:

μ _i＝P _i-P _lim(1)

In formula (1), μ _ifor the overload quantity of element, P _ifor the trend value of element i under this power system operating mode, P _limfor the tidal current limit value of this element;

S5. the objective function building season power transmission and transformation integration maintenance wish is as follows:

$W_{i,t}=\underset{j=1}{\overset{\mathrm{Nt}}{\mathrm{\Σ}}}\left[{\∫}_t^{t+\mathrm{\Δt}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{u}_i\left(x_{i,t}\right)\mathrm{dt}\right]+\underset{j=1}{\overset{\mathrm{Nl}}{\mathrm{\Σ}}}\left[{\∫}_t^{t+\mathrm{\Δt}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{u}_i\left(x_{i,t}\right)\mathrm{dt}\right]---\left(2\right)$

In formula (2), W _i,tfor the objective function of season power transmission and transformation integration maintenance wish, characterize the maintenance overload wish of element i, Nt is the number transformer that electrical network intends the maintenance of report the coming season, and Nl is the number of, lines that electrical network intends the maintenance of report the coming season, u _ifor the overload quantity of element i, x _i,tcharacterize the 0-1 variable that element i is overhauled in t, if element i is starting maintenance the coming season a day, and continue T time, then x _i,a=1, x _{i, a+1}=1...x _{i, a+T}=1, all the other periods are 0, X _i,tfor element inspecting state incidence matrix, be specifically expressed as:

X _i,t＝[X _i·T _i][α _i] (3)

In formula (3), X _irepresent i-th element maintenance, T _irepresent i-th element maintenance duration, X _iand T _ibe constant matrix and known, α _ibattle array is the initial time matrix of element i maintenance;

S6. by u _i(x _i,t) substitute into the objective function that wish is overhauled in season power transmission and transformation integration, and solve;

S7. the first round calculates and terminates, and adopts genetic algorithm to upgrade maintenance mode population x _i,t, electric network swim sequence under restructuring maintenance mode;

1) initialization: arrange evolutionary generation counter t=0, arranges maximum evolutionary generation K, and stochastic generation M individual as initial population X _i,t(k);

2) individual evaluation: calculate initial population X _i,tthe fitness of each individuality in (k);

3) Selecting operation: selection opertor is acted on population, the object of selection the individuality optimized is genetic directly to the next generation or produces new individuality by pairing intersection be genetic to the next generation again.Selection operation is based upon on the assessment basis of population at individual;

4) crossing operation: crossover operator is acted on population;

5) mutation operator: mutation operator is acted on population, namely changes the genic value on some locus of string individual in population;

6) population X _i,tk () obtains population X of future generation after selection, intersection, mutation operator _i,t(k+1);

7) end condition judges: if k=K, then the maximum adaptation degree individuality that has obtained in evolutionary process exports as having most to separate, and stops calculating.

S8. according to the maintenance mode population x upgraded in step S7 _i,t, redefine electric network swim sequence under maintenance mode, and repeat step S2-S7;

If S9. final Output rusults meets following formula:

ΔW _i,t≤ξ (4)

Then Output rusults W _i,t.

Advantage and effect:

The present invention relates to a kind of season power transmission and transformation integration optimized maintenance method based on genetic algorithm, adopt method proposed by the invention compared with prior art, there is following beneficial effect: contemplated by the invention the practical problemss such as service element between actual electric network turn(a)round is numerous, maintenance mode arrangement is unreasonable, by setting up the objective function of season power transmission and transformation integration maintenance wish, genetic algorithm is utilized to upgrade maintenance mode population, electric network swim sequence under restructuring maintenance mode, obtains optimum grid maintenance scheme with this.Security and stability during greatly improving grid maintenance, traditional artificial method calculated that only relies on of comparing has more application value.Greatly improve the work efficiency of operation of power networks personnel, the grid maintenance scheme obtained rationally, reliably, Social benefit and economic benefit is remarkable.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the season power transmission and transformation integration optimized maintenance method based on genetic algorithm of the present invention.

Embodiment

The present invention relates to a kind of season power transmission and transformation integration optimized maintenance method based on genetic algorithm, by setting up the objective function of season power transmission and transformation integration maintenance wish, the maintenance solution likely occurred in electrical network is checked, adopt the integration of the power transmission and transformation based on the genetic algorithm optimized maintenance method with inventive concept, realize the optimization selection of grid maintenance scheme, greatly improve work efficiency and the security and stability of electrical network between turn(a)round of operation of power networks personnel.

Below in conjunction with the drawings and specific embodiments, the present invention is described further:

In present embodiment, for the grid structure of certain actual electric network, according to this electrical network season load prediction, start-up mode arrangement, grid topology data, determine the sequence of current of this electrical network between turn(a)round in season; The relevant informations such as the service element reported according to this electrical network each subordinate electric company plan and repair time, determine the sequence of current of electrical network under maintenance mode; Fault set is set, i.e. arbitrary element in N-1 electrical network, N-2 parallel erected on same tower circuit; Under checking maintenance mode, whether system load flow meets thermally-stabilised requirement, and obtains the overload quantity of overload element; Build electrical network season power transmission and transformation integration optimized maintenance function, and the substitution of element overload quantity is solved; Genetic algorithm is adopted to upgrade maintenance mode population, electric network swim sequence under restructuring maintenance mode; Loop iteration, calculating, obtain optimum electrical network season power transmission and transformation integration maintenance solution.The above-mentioned season power transmission and transformation integration optimized maintenance method based on genetic algorithm, comprises the steps:

Step 1: according to the load prediction of electrical network season, the arrangement of electrical network start-up mode, grid topology data, determine object time section sequence of current.

Step 2: according to turnaround plan and the relevant information of intending report, determine the sequence of current under grid maintenance mode.

Step 3: arrange power grid fault set, comprises arbitrary element in N-1 electrical network, parallel erected on same tower circuit in N-2 electrical network.

Step 4: under checking maintenance mode, whether system load flow meets power grid heat stability provisioning request, and obtain the overload quantity μ of overload element _i, wherein:

μ _i＝P _i-P _lim(1)，

In formula (1), μ _ifor the overload quantity of element, P _ifor the trend value of element i under this power system operating mode, P _limfor the tidal current limit value of this element.

Step 5: the objective function building season power transmission and transformation integration maintenance wish is as follows:

$W_{i,t}=\underset{j=1}{\overset{\mathrm{Nt}}{\mathrm{\Σ}}}\left[{\∫}_t^{t+\mathrm{\Δt}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{u}_i\left(x_{i,t}\right)\mathrm{dt}\right]+\underset{j=1}{\overset{\mathrm{Nl}}{\mathrm{\Σ}}}\left[{\∫}_t^{t+\mathrm{\Δt}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{u}_i\left(x_{i,t}\right)\mathrm{dt}\right]---\left(2\right),$

In formula (2), W _i,tfor the objective function of season power transmission and transformation integration maintenance wish, characterize the maintenance overload wish of element i, Nt is the number transformer that electrical network intends the maintenance of report the coming season, and Nl is the number of, lines that electrical network intends the maintenance of report the coming season, u _ifor the overload quantity of element i, x _i,tcharacterize the 0-1 variable that element i is overhauled in t, if element i is starting maintenance the coming season a day, and continue T time, then x _i,a=1, x _{i, a+1}=1...x _{i, a+T}=1, all the other periods are 0, X _i,tfor element inspecting state incidence matrix, be specifically expressed as:

X _i,t＝[X _i·T _i][α _i] (3)，

In formula (3), X _irepresent i-th element maintenance, T _irepresent i-th element maintenance duration, X _iand T _ibe constant matrix and known, a _ibattle array is the initial time matrix of element i maintenance.

Step 6: by u _i(x _i,t) substitute into the objective function that wish is overhauled in season power transmission and transformation integration, and solve.

Step 7: the first round calculates and terminates, adopts genetic algorithm to upgrade maintenance mode population x _i,t, electric network swim sequence under restructuring maintenance mode; Process is as follows:

1) initialization: arrange evolutionary generation counter t=0, arranges maximum evolutionary generation K, and stochastic generation M individual as initial population X _i,t(k);

2) individual evaluation: calculate initial population X _i,tthe fitness of each individuality in (k);

3) Selecting operation: selection opertor is acted on population, the object of selection the individuality optimized is genetic directly to the next generation or produces new individuality by pairing intersection be genetic to the next generation again; Selection operation is based upon on the assessment basis of population at individual;

4) crossing operation: crossover operator is acted on population;

5) mutation operator: mutation operator is acted on population, namely changes the genic value on some locus of string individual in population;

6) population X _i,tk () obtains population X of future generation after selection, intersection, mutation operator _i,t(k+1);

7) end condition judges: if k=K, then the maximum adaptation degree individuality that has obtained in evolutionary process exports as having most to separate, and stops calculating.

Step 8: according to the maintenance mode population x upgraded in step 7 _i,t, redefine electric network swim sequence under maintenance mode, and repeat step 2-7.

Step 9: if final Output rusults meets following formula:

ΔW _i,t≤ξ (4)，

Then Output rusults W _i,t.

The concrete maintenance solution of this electrical network is as shown in table 1:

Certain actual electric network maintenance mode of table 1 arranges scheme

Note: the number of days of numeral postponement from maintenance starts the same day of each element " maintenance initial time ".

Claims (2)

1. based on a season power transmission and transformation integration optimized maintenance method for genetic algorithm, by building season power transmission and transformation integration maintenance wish function, iteration, calculating optimum grid maintenance scheme, it is characterized in that: the method step is as follows:

S1. according to the load prediction of electrical network season, the arrangement of electrical network start-up mode, grid topology data, object time section sequence of current is determined;

S2. according to turnaround plan and the relevant information of intending report, the sequence of current under grid maintenance mode is determined;

S3. power grid fault set is set, comprises arbitrary element in N-1 electrical network, parallel erected on same tower circuit in N-2 electrical network;

S4. under checking maintenance mode, whether system load flow meets power grid heat stability provisioning request, and obtains the overload quantity μ of overload element _i, wherein:

μ _i＝P _i-P _lim(1)

In formula (1), μ _ifor the overload quantity of element, P _ifor the trend value of element i under this power system operating mode, P _limfor the tidal current limit value of this element;

S5. the objective function building season power transmission and transformation integration maintenance wish is as follows:

$W_{i,t}=\underset{j=1}{\overset{\mathrm{Nt}}{\mathrm{\Σ}}}\left[{\∫}_t^{t+\mathrm{\Δt}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{u}_i\left(x_{i,t}\right)\mathrm{dt}\right]+\underset{j=1}{\overset{\mathrm{Nl}}{\mathrm{\Σ}}}\left[{\∫}_t^{t+\mathrm{\Δt}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{u}_i\left(x_{i,t}\right)\mathrm{dt}\right]---\left(2\right)$

In formula (2), W _i,tfor the objective function of season power transmission and transformation integration maintenance wish, characterize the maintenance overload wish of element i, Nt is the number transformer that electrical network intends the maintenance of report the coming season, and Nl is the number of, lines that electrical network intends the maintenance of report the coming season, u _ifor the overload quantity of element i, x _i,tcharacterize the 0-1 variable that element i is overhauled in t, if element i is starting maintenance the coming season a day, and continue T time, then x _i,a=1, x _{i, a+1}=1...x _{i, a+T}=1, all the other periods are 0, X _i,tfor element inspecting state incidence matrix, be specifically expressed as:

$X_{i,t}=\left[\begin{array}{cc}{X}_i& T_i\end{array}\right]\left[a_i\right]---\left(3\right)$

In formula (3), X _irepresent i-th element maintenance, T _irepresent i-th element maintenance duration, X _iand T _ibe constant matrix and known, a _ibattle array is the initial time matrix of element i maintenance;

S6. by u _i(x _i,t) substitute into the objective function that wish is overhauled in season power transmission and transformation integration, and solve;

S7. the first round calculates and terminates, and adopts genetic algorithm to upgrade maintenance mode population x _i,t, electric network swim sequence under restructuring maintenance mode;

S8. according to the maintenance mode population x upgraded in step S7 _i,t, redefine electric network swim sequence under maintenance mode, and repeat step S2-S7;

If S9. final Output rusults meets following formula:

ΔW _i,t≤ξ (4)

Then Output rusults W _i,t.

2. the season power transmission and transformation integration optimized maintenance method based on genetic algorithm according to claim 1, is characterized in that: step S7 adopts genetic algorithm to upgrade maintenance mode population x _i,t, electric network swim sequence under restructuring maintenance mode, process is as follows:

1) initialization: arrange evolutionary generation counter t=0, arranges maximum evolutionary generation K, and stochastic generation M individual as initial population X _i,t(k);

2) individual evaluation: calculate initial population X _i,tthe fitness of each individuality in (k);

3) Selecting operation: selection opertor is acted on population, the object of selection the individuality optimized is genetic directly to the next generation or produces new individuality by pairing intersection be genetic to the next generation again; Selection operation is based upon on the assessment basis of population at individual;

4) crossing operation: crossover operator is acted on population;

5) mutation operator: mutation operator is acted on population, namely changes the genic value on some locus of string individual in population;

6) population X _i,tk () obtains population X of future generation after selection, intersection, mutation operator _i,t(k+1);

7) end condition judges: if k=K, then the maximum adaptation degree individuality that has obtained in evolutionary process exports as having most to separate, and stops calculating.