CN113300380B - Load curve segmentation-based power distribution network reactive power optimization compensation method - Google Patents
Load curve segmentation-based power distribution network reactive power optimization compensation method Download PDFInfo
- Publication number
- CN113300380B CN113300380B CN202110674283.8A CN202110674283A CN113300380B CN 113300380 B CN113300380 B CN 113300380B CN 202110674283 A CN202110674283 A CN 202110674283A CN 113300380 B CN113300380 B CN 113300380B
- Authority
- CN
- China
- Prior art keywords
- optimization
- load
- distribution network
- compensation
- power distribution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 126
- 238000009826 distribution Methods 0.000 title claims abstract description 88
- 238000000034 method Methods 0.000 title claims abstract description 80
- 230000011218 segmentation Effects 0.000 title claims abstract description 40
- 230000035945 sensitivity Effects 0.000 claims abstract description 26
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 13
- 239000013598 vector Substances 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 11
- 239000006185 dispersion Substances 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000003672 processing method Methods 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 claims description 2
- 230000035772 mutation Effects 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- -1 hydrogen Chemical class 0.000 claims 1
- 239000001257 hydrogen Substances 0.000 claims 1
- 238000013178 mathematical model Methods 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 6
- 238000011217 control strategy Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000012916 structural analysis Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
- H02J3/16—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load by adjustment of reactive power
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
- H02J3/50—Controlling the sharing of the out-of-phase component
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/30—Reactive power compensation
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a reactive power optimization compensation method for a power distribution network based on load curve segmentation. And secondly, providing a reactive compensation comprehensive sensitivity index to screen reactive compensation points based on each load section, and establishing an optimized mathematical model for solving a reactive compensation scheme in each section by taking the voltage average deviation and the running network loss of a power grid node as optimization indexes. And respectively solving the established optimization model by adopting a multi-objective optimization algorithm based on decomposition to obtain an optimal solution set, and evaluating the optimal solution set by using subjective and objective combination weighting of an improved analytic hierarchy process and an inverse entropy weight method to obtain a load segmentation result and an optimal compensation scheme corresponding to each segment. The reactive compensation optimization method provided by the invention has the advantages of complete system, reasonable and comprehensive consideration factors, easiness in engineering realization and capability of effectively realizing voltage self-adaptive control aiming at random load fluctuation of a power distribution network.
Description
Technical Field
The invention belongs to a reactive power optimization compensation method for a power distribution network, and particularly relates to a reactive power compensation optimization method considering random fluctuation characteristics of loads.
Background
The power distribution network is an important link for directly supplying power to users in a power system, but the structure and the operation characteristics of the power distribution network determine the node voltage quality and the line loss of the power distribution network, and particularly the influence of load fluctuation is considered, so that the requirements on the goodness and the economy of the operation of the power distribution network system are often difficult to achieve. Therefore, an effective reactive power optimization compensation strategy is necessary to be established, the voltage quality of the power distribution network system is improved, the operation economy is improved, and the operation voltage quality requirement of the power distribution network system can be still met under the characteristic of random fluctuation of loads.
Reactive compensation is a measure for regulating voltage of a distribution network, and the current reactive optimization compensation research of the distribution network mainly focuses on adopting a series compensation or parallel compensation mode to control reactive voltage of the distribution network under a single section of a fixed load, and less considers a reactive optimization strategy under the load fluctuation characteristic. For the load fluctuation characteristic, although a control strategy of historical scene matching is proposed through data generated by operation based on a scene analysis method to perform reactive power control and voltage management, the load fluctuation processing method based on scene generation has a large calculation amount and does not have good expansibility. The idea of adopting load curve segmentation is also researched, a time-sharing reactive power optimization control strategy of the power distribution network is provided, the practicability is good, the researches on the aspects of selection of reactive power compensation points, reasonable empowerment of optimization indexes and the like are not involved, and the established method strategy does not have good integrity and applicability.
Disclosure of Invention
The invention aims to provide a load curve segmentation-based reactive power optimization compensation method for a power distribution network, which is used for establishing a reactive power compensation optimization method for controlling reactive voltage of the power distribution network in a whole day period based on load curve optimization segmentation and a reactive power compensation point screening strategy.
In order to realize the purpose, the reactive power optimization compensation method for the power distribution network based on the load curve segmentation, which is provided by the invention, comprises the following steps of:
and 7, obtaining a reactive power optimization compensation strategy at each time all day based on the reactive power optimization compensation scheme obtained under each load segment.
The invention has the beneficial effects that:
1. based on the idea of time-phased staticizing of a load curve, the method can better process the random fluctuation characteristic of the load, and overcome the problem that the traditional reactive power optimization compensation scheme based on a fixed load single section cannot meet the voltage quality requirement in real time;
2. The optimal reactive compensation point is determined based on the feeder line partition and by combining the node voltage offset sensitivity and the network loss sensitivity, and compared with a method for randomly determining the reactive compensation point, the method has the advantages that a better reactive compensation effect can be obtained, and the reasonability is realized;
3. according to the method, subjective and objective factors are considered, subjective and objective weighting is carried out on the indexes based on an improved chromatographic analysis method and an anti-entropy weight method, the problem that the traditional method lacks reasonable assignment of index weights is solved and supplemented, and the established load curve segmentation-based reactive power optimization compensation method for the power distribution network has better rationality and applicability.
The system establishes a power distribution network reactive power optimization compensation strategy based on load curve segmentation. Firstly, an optimized segmented model of a typical daily load curve is established based on the idea of time-segment staticizing of the load curve. And secondly, providing a reactive compensation comprehensive sensitivity index for screening reactive compensation points based on each load segment, and establishing an optimized mathematical model for solving a reactive compensation scheme in each segment by taking the voltage average deviation and the running network loss of the power grid node as optimization indexes. And respectively solving the established optimization model by adopting a decomposition-based multi-objective optimization (MOEA/D) algorithm to obtain a load segmentation result and an optimal compensation scheme corresponding to each segment. And finally, performing simulation analysis by using an IEEE 33 node power distribution network system example, and verifying the feasibility and the effectiveness of the established power distribution network reactive power optimization method.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a topological diagram of an IEEE 33 node power distribution network system;
FIG. 3 is a typical daily load curve and the results of load optimization segmentation;
FIG. 4 is a voltage distribution of the distribution grid system under each load segment before reactive compensation;
FIG. 5 is a comparison of average node voltage offsets of a power distribution network system before and after reactive compensation;
FIG. 6 is a comparison of the running network loss of the power distribution network system before and after reactive compensation;
FIG. 7 is distribution network system voltage distribution under each load segment after reactive compensation;
FIG. 8 is a comparison of voltage offsets of the grid at various times throughout the day before and after the reactive compensation strategy is applied;
fig. 9 is a comparison of the network loss of the power grid running at each time all day before and after the reactive compensation strategy is applied.
Detailed Description
The invention is described in further detail below with reference to the following figures and specific examples:
a load curve segmentation-based reactive power optimization compensation method for a power distribution network comprises the following steps as shown in figure 1:
and 7, obtaining a reactive power optimization compensation strategy at each time all day based on the reactive power optimization compensation scheme obtained under each load segment.
In the above technical solution, the typical daily load curve of the power distribution network in step 1 is a curve of the change of the load of the power distribution network system with time in one day in the power distribution network system, and is a basis for scheduling a power generation plan and determining operation modes such as power grid voltage regulation and reactive power compensation.
In the above technical solution, the specific implementation method of step 1 is as follows:
the optimization target of the load curve optimization segmented model is an index M representing the dispersion between the average values of loads in different time periods1Maximum, representing the load dispersion within the same time interval2And (3) minimum, considering the positive and negative attributes of the indexes, and establishing a load curve optimization segmented model as follows:
wherein f iss1Is an objective function representing the dispersion between the mean values; f. ofs2An objective function representing the load dispersion in the same time period; n is a radical ofLThe total number of segments of the load curve; paveThe average value of the loads at all the moments is obtained; pave,iThe average value of the load in the ith segment is taken; pLijThe load size of the jth load point in the ith load segment is obtained; h iIs the total number of load points in the ith segment.
In the above technical solution, the average node voltage offset sensitivity and the network loss sensitivity index define an expression as follows:
wherein, EUkRepresenting the average node voltage deviation of the power distribution network system when the node k has no reactive disturbance; ePkRepresents the amount of change in the network loss caused after the reactive disturbance Δ Q ═ Q is injected at node k, where Δ Uave,kAnd Ploss,kThe computational expression of (a) is:
wherein, Delta Uave,kAveraging node voltage deviation for the power distribution network system; n is the total number of the nodes of the power distribution network system; u shapejIs the actual voltage of the jth node; u shapeN,jIs the rated voltage of the jth node; ploss,kOperating the total network loss for the power distribution network system; n is a radical oflThe total number of the branch circuits of the power distribution network system is; pt_lossThe active loss of the t branch of the power distribution network system is shown.
In the above technical solution, the method for implementing the weighting method specifically includes:
the weighting method is to standardize the negative or positive indicators under the jth scheme in the optimization process aiming at the problem that the load optimization segmentation indicators in the step 2, the voltage offset sensitivity and the network loss sensitivity indicators in the step 3 and the voltage offset and running network loss indicators in the step 5 are inconsistent in magnitude or direction, and to perform subjective and objective combined weighting on the indicators by adopting an improved analytic hierarchy process and an anti-entropy weight method, and the specific processing method is as follows:
Wherein, I1Representing a negative index set; i is2Representing a set of forward indicators;the negative direction index after the standardization treatment is shown;indicating the normalized forward indicator.
In the above technical solution, the specific implementation method of the subjective and objective combination empowerment includes:
step 2.1, improved analytic hierarchy process
Assuming that m indexes to be weighted are total, according to the principle that the importance degree is not reduced, setting the relative importance degree of each index to be ranked from large to small by experts:
xI1>=xI2>=...>=xIm
wherein x isIj(j-1, …, m) denotes the j-th index after sorting, and represents xIiAnd xIi+1Is denoted as siAnd the value of the relative important relation has a set value standard, so that the obtained judgment matrix W has the following elements:
accordingly, the subjective weight of each index is:
wherein the content of the first and second substances,is the product of all elements in the jth row in the matrix W;
step 2.2, inverse entropy weight method
If the number of the index samples is n, the information entropy R of the index j isjComprises the following steps:
wherein the content of the first and second substances,is calculated asWherein y iskjThe element values in the matrix after the normalization operation is carried out on the original decision matrix; from this, the objective weight b of each index can be determinedjComprises the following steps:
step 2.3, comprehensive weight determination
On the basis of obtaining the subjective weight and the objective weight, the comprehensive weight w of the index j is calculated by adopting multiplication weighting abjComprises the following steps:
in the above technical solution, the relative importance value standard is:
when the index importance is particularly important, the relative importance value is 1.8; when the index importance is more important, the relative importance value is 1.4; when the index importance is slightly important, the relative importance value is 1.2; when the index importance is equally important, the relative importance value is 1.0.
In the above technical solution, the constraint conditions established in the reactive compensation multi-objective optimization model in step 5 are as follows:
the method comprises the following steps of node power balance constraint, generator active and reactive power output constraint, node voltage upper and lower limit constraint, line maximum transmission power constraint and constraint condition addition of control variables, wherein the control variables are parallel compensation capacity upper and lower limit values, and the expression of the constraint condition is as follows:
wherein, PkActive power is injected into a node k; qkInjecting reactive power into the node k; gkjIs the mutual conductance between nodes k and j; b iskjIs the transadmittance between nodes k and j; deltakjIs the voltage phase difference between nodes k and j; (ii) a U shapekIs the voltage of node k; u shapejIs the voltage at node j; pGkActive power output is provided for the kth generator;is the active output lower limit of the generator k;is the active power output upper limit of the generator k; q GkReactive power output of a kth generator is achieved;is the reactive power lower limit of the generator k;is the upper limit of reactive power output of the generator k; n is a radical ofGThe total number of the generator nodes of the power distribution network system is calculated;is the upper limit of the node voltage;is the node voltage lower limit; n is the total number of the nodes of the power distribution network system; slTransmitting power for the first branch;maximum allowed transmission power for branch l; n is a radical oflThe total number of the branch circuits of the power distribution network system is; qckCompensating the capacity for the kth reactive power compensation device; qc,maxIs the upper limit of the reactive compensation capacity; qc,minIs the lower limit of the reactive compensation capacity; n is a radical ofvThe number of the reactive compensation is.
In the above technical solution, the improved decomposition-based multi-objective optimization algorithm is implemented by:
step 3.1, inputting algorithm parameters
Inputting improved multi-objective optimization algorithm parameters, wherein the parameters comprise population number NPThe number T of weight vectors in the field and the maximum iteration number genmax;
Step 3.2, initialize weight vector matrix
Defining the pareto optimal solution set to be empty initially, and decomposing the multi-objective optimization problem into N based on a Tchebycheff decomposition methodPThe single target sub-problems are optimized simultaneously, wherein each sub-problem can be expressed as:
in the formula, e represents an optimization objective; q is an optimization sub-problem, wherein q is 1,2, …, Np; f. of e(x) Representing the target value of the optimization target e corresponding to the optimization variable x; z is an initialized target value;representing the weight corresponding to the optimization target e; g is a radical of formulateFor the decomposed single-target optimization problem, the corresponding weight vectors are respectively [ lambda ]1,…,λNp]And is made ofWhereinCalculating Euclidean distance matrixes formed among the weight vectors, sorting according to Euclidean distance, and selecting T top distance ranks as weight vectors lambdaqE (q) is equal to (q)1,…,qT) Wherein E (q) represents a domain vector of the optimization sub-problem q, q1,…,qTRespectively represents elements in the field, and the total number of the elements is T;
step 3.3, random production of initial populations
Uniform generation of random in optimization spaceInitial population { x1,…,xNpAnd calculating an optimization objective function value f corresponding to each populatione(x1),…,fe(xNp) The initial optimization target value z ═ z (z)1,z2,…,zm) Initializing a reference point, taking the minimum value in each sub-standard function value of all individuals as the reference point, ze=min(fe(x1),…,fe(xNp));
Step 3.4, evolution
For each population, in the field E (q), a random method is used to generate the sequence number k1、k2Through xk1And xk2Obtaining a new optimization variable solution y ', and obtaining an improved solution y from y' based on a heuristic method;
step 3.5, Gauss variation
Each element y in the solution y obtained after evolution iPerforming Gaussian mutation operation, wherein the operation equation is as follows:
wherein Gussian (t) represents a Gaussian function with an argument of t; delta. for the preparation of a coatingpiGiven standard deviation of gaussian variation;standard variance for a given gaussian variation; rand1For generating random number operations satisfying independent variable upper and lower limit constraints, C is a set of generated random numbers, and the set is N in totalvThe number of the cells; ciIs the ith element in the set C; piFor a given rate of variation; rand2To generate random numbers that obey a gaussian distribution;
step 3.6, update reference points
Updating the reference point z if the reference point z corresponding to the optimization target eeAnd its optimized target value feThe size relationship of (y) is as follows: z is a radical ofe>fe(y) then let ze=fe(y);
Step 3.7, updating the domain solution and Pareto optimal solution set
If for any of the weight elements in E (q) < lambda >rG exists between the new optimization variable solution y and the solution xte(y∣λr,z)≤gte(x∣λrZ), then let x ═ y, and the optimization target value f (x) corresponding to x is updated to f (x) ═ f (y), and at the same time, the Pareto optimal solution set is updated;
step 3.8, judging termination conditions
Judging whether the maximum value g of the iteration times is reacheden=genmaxIf so, stopping iteration and entering the next step, otherwise, iterating for a time gen=gen+1 and return to step 3.4;
step 3.9, output
And outputting the Pareto optimal solution set obtained after optimization, and ending the optimization process.
In the embodiment, an IEEE 33 node distribution network system is used as an example simulation analysis, the topology of the distribution network system is shown in figure 2, the distribution network system comprises 4 feeders, and the numbers of the feeders are shown in figure 2. The voltage rated value of a node of the power distribution network system is 1.0pu, the power reference value is set to be 100MW, structural parameters of the power distribution network system are kept unchanged, and a typical daily load curve of the power distribution network system is simulated according to the change rule of the typical daily load curve of a certain area and in combination with the load scale of an original power distribution network system and is shown in figure 3. In addition, the maximum iteration number of the multi-target optimization algorithm based on decomposition is set to be 100, the variation rate is set to be 0.8, and the Gaussian variation variance is set to be 0.075. Optimizing segment index M when weighted using improved analytic hierarchy process in accordance with the present invention1Relative M2The importance of (2) is 1.0, and the importance of the average voltage deviation index relative to the grid loss index is also 1.0. Setting the load segment number NLTo 5, the load segmentation curve obtained by the established load optimization segmentation method is shown in FIG. 3, wherein the weight factor ε is evaluated for the optimal solution set of load segmentation1、ε20.533 and 0.477, respectively, the target value M is optimized1And M20.4659 and 0.5284, respectively.
As can be seen from FIG. 3, the 24h day is divided into five time intervals: the time period is 1 h-6 h, 7 h-10 h, 11 h-16 h, 17 h-21 h and 22 h-24 h. Based on the obtained load curve segmentation results, when no reactive compensation is available, the voltage distribution of each node of the power distribution network system corresponding to each segment is as shown in fig. 4, and it can be seen that the voltage level of each load segment is relatively low, especially the lowest voltage node of the line terminal exceeds the lower limit of allowable voltage deviation of the power distribution network by 0.93pu, so that it is necessary to perform reactive compensation to improve the voltage quality of the power grid.
And (3) capacity optimization determination of two groups of reactive compensation devices is set, and reactive compensation points are respectively selected from feeder 1 and feeder 2 areas by combining structural analysis of a power distribution network system shown in figure 2. And when the reactive power disturbance delta Q is 0.005pu, selecting the top 4 nodes and target values according to the comprehensive sensitivity sequencing results on the feeder line 1 and the feeder line 2 under each load subsection, wherein the weighting factor of each subsection is shown in the table 1.
TABLE 1 comprehensive sensitivity ranking under each load segment
From the analysis of the results in table 1, the optimal reactive compensation point on the feeder 1 is the node 32 under each segment. The optimal compensation points of the feeder 2 in the sections 4 and 5 are respectively the node 17 and the node 16, and the optimal compensation points of the other sections are all the nodes 15, so the reactive compensation points selected by the comprehensive comparison analysis are respectively the node 32 of the feeder 1 and the node 15 of the feeder 2.
In the reactive power optimization compensation process, setting the upper and lower voltage allowable limits to be 1.07pu and 0.93pu respectively; the maximum transmission power of the branch is 2 MW; the upper limit and the lower limit of the reactive compensation capacity are respectively 2Mvar and 0.5 Mvar. The optimal compensation scheme and the optimal target value under each section obtained by solving the optimization model based on MOEA/D algorithm are shown in Table 2
TABLE 2 reactive compensation parameters and corresponding optimization objectives and weights
Based on the optimization results in table 2, the average voltage offset comparison and the grid loss comparison before and after compensation in each load segment are shown in fig. 5 and 6, respectively. Therefore, compared with the situation without reactive compensation, the node voltage average deviation and the network loss after reactive optimization compensation of each load section are reduced to different degrees, so that the voltage quality of the power grid is better, and the running economy is improved. In addition, the voltage distribution of each node of the distribution network system after compensation in each load segment is shown in fig. 7. Compared with the analysis in fig. 4, it can be seen that the overall voltage of the power distribution network system is effectively improved after the reactive power optimization compensation, and the lowest point voltage of each segment is more than 0.93pu, so that the voltage deviation requirement of the power distribution network is met.
After reactive power optimization compensation is performed on the basis of load subsections, a good compensation effect is achieved under the determined load of each subsection. In order to verify the control effect of the optimization strategy on the grid voltage and the grid loss in the process of load change all day, the load of the power distribution network system is changed in a period of 24h all day as shown in fig. 3, a reactive compensation scheme determined based on a load segmentation method is adopted for compensation, and the average node voltage deviation and the running grid loss of the power distribution network system without reactive compensation in the process are compared with those shown in fig. 8 and 9. In the figure, the reactive power optimization compensation scheme determined based on the load segmentation method is adopted, compared with the situation that the voltage average deviation and the running network loss of a power distribution network system are smaller in the whole day without reactive power compensation, a better whole-day compensation effect can be obtained only by setting five groups of reactive power compensation schemes with different capacity sizes, and the established reactive power optimization strategy is verified to have better rationality and feasibility.
Those not described in detail in this specification are well within the skill of the art.
Claims (8)
1. A distribution network reactive power optimization compensation method based on load curve segmentation is characterized by comprising the following steps: the method comprises the following steps:
step 1, inputting the number N of load sections based on a typical daily load curve of a power distribution networkLEstablishing a multi-target load curve optimization segmented model for time-interval division of the load curve;
step 2, solving the load curve optimization segmentation model established in the step 1 by adopting an improved multi-objective optimization algorithm based on decomposition to obtain a pareto optimal solution set of load segmentation points, carrying out weighted evaluation on the optimal solution set by using a weighting method to obtain an optimal load segmentation scheme, wherein the load scales of all moments in each segment after the optimization segmentation are the same, and the load value of all moments in each segment is the load average value in the segment;
step 3, based on the structural characteristics of the power distribution network, feeder line partitioning is carried out on nodes of the power distribution network, based on average node voltage deviation sensitivity and network loss sensitivity indexes, comprehensive sensitivity indexes are defined by adopting a weighting method, then, the nodes on each feeder line are screened and sorted for the nodes with N reactive compensation pointsLEach load is segmented, and each feeder line obtains N LSelecting alternative compensation point scheme, selecting N on each feeder lineLThe node with the highest comprehensive sensitivity ranking for the first time under each section is used as the all-day optimal compensation point of the feeder line area;
step 4, establishing a reactive compensation multi-objective optimization model based on the node voltage average deviation and the power grid operation grid loss index, and defining the current load subsection number as i being equal to 1;
step 5, based on the load value under the ith load section determined in the step 2 and the all-day optimal compensation point of the feeder line area determined in the step 3, solving the reactive compensation multi-objective optimization model obtained in the step 4 by adopting an improved multi-objective optimization algorithm based on decomposition to obtain an optimal solution set, and evaluating the optimal solution set by using a weighting method to obtain an optimal reactive compensation capacity scheme under the ith load section;
step 6, executing i to i +1, and judging i and NLIf i is less than or equal to NLIf not, entering step 4, otherwise, entering step 7;
and 7, obtaining a reactive power optimization compensation strategy at each moment all day based on the reactive power optimization compensation scheme obtained under each load segment.
2. The load curve segmentation-based reactive power optimization compensation method for the power distribution network according to claim 1, wherein the load curve segmentation-based reactive power optimization compensation method comprises the following steps: the specific implementation method of the step 1 comprises the following steps:
The optimization target of the load curve optimization segmented model is an index M representing the dispersion among the average values of loads in different time periods1Maximum index M representing the load dispersion in the same time interval2And (3) considering the positive and negative attributes of the indexes, and establishing a load curve optimization segmented model as follows:
wherein, fs1Is an objective function representing the dispersion between the mean values; f. ofs2An objective function representing the load dispersion in the same time period; n is a radical ofLThe total number of segments of the load curve; paveThe average value of the loads at all the moments is obtained; pave,iThe average value of the load in the ith segment is taken; pLijThe load size of the jth load point in the ith load segment is obtained; hiIs the total number of load points in the ith segment.
3. The load curve segmentation-based reactive power optimization compensation method for the power distribution network, according to claim 1, is characterized in that: the average node voltage offset sensitivity and the network loss sensitivity index define an expression as follows:
wherein E isUkRepresenting the average node voltage deviation of the power distribution network system when the node k has no reactive disturbance; ePkRepresenting the effect caused after injecting a reactive disturbance Δ Q at node kAmount of change in loss of network, wherein Δ Uave,kAnd Ploss,kThe calculation expression of (a) is:
wherein, Delta Uave,kAveraging node voltage deviation for the power distribution network system; n is the total number of the nodes of the power distribution network system; u shape jIs the actual voltage of the jth node; u shapeN,jIs the rated voltage of the jth node; ploss,kOperating the total network loss for the power distribution network system; n is a radical oflThe total number of the branch circuits of the power distribution network system is; pt_lossThe active loss of the t branch of the power distribution network system is shown.
4. The load curve segmentation-based reactive power optimization compensation method for the power distribution network, according to claim 1, is characterized in that: the specific implementation method of the empowerment method comprises the following steps:
the weighting method is to standardize the negative or positive indicators under the jth scheme in the optimization process aiming at the problem that the load optimization segmentation indicators in the step 2, the voltage offset sensitivity and the network loss sensitivity indicators in the step 3 and the voltage offset and running network loss indicators in the step 5 are inconsistent in magnitude or direction, and to perform subjective and objective combined weighting on the indicators by adopting an improved analytic hierarchy process and an anti-entropy weight method, and the specific processing method is as follows:
5. The load curve segmentation-based reactive power optimization compensation method for the power distribution network, according to claim 4, is characterized in that: the specific implementation method of the subjective and objective combination empowerment comprises the following steps:
Step 2.1, improved analytic hierarchy Process
Assuming that m indexes to be weighted are total, according to the principle that the importance degree is not reduced, setting experts to sort the relative importance degrees of all indexes from large to small:
xI1>=xI2>=...>=xIm
wherein x isIj(j-1, …, m) represents the j-th index after sorting, and represents xIiAnd xIi+1Is denoted as siAnd the value of the relative important relation has a set value standard, so that the obtained judgment matrix W has the following elements:
accordingly, the subjective weight of each index is:
wherein, the first and the second end of the pipe are connected with each other,is the product of all elements in the jth row in the matrix W;
step 2.2, inverse entropy weight method
If the number of the index samples is n, the information entropy R of the index j is:
wherein the content of the first and second substances,is calculated asWherein y iskjThe element values in the matrix after the normalization operation is carried out on the original decision matrix; from this, the objective weight b of each index can be determinedjComprises the following steps:
step 2.3, comprehensive weight determination
On the basis of obtaining the subjective weight and the objective weight, the comprehensive weight w of the index j is calculated by adopting multiplication weightingabjComprises the following steps:
6. the load curve segmentation-based reactive power optimization compensation method for the power distribution network, according to claim 5, is characterized in that: the relative importance value standard is as follows:
when the index importance is particularly important, the relative importance value is 1.8; when the index importance is more important, the relative importance value is 1.4; when the index importance is slightly important, the relative importance value is 1.2; when the index importance is equally important, the relative importance value is 1.0.
7. The load curve segmentation-based reactive power optimization compensation method for the power distribution network, according to claim 1, is characterized in that: the constraint conditions established in the reactive power optimization compensation model in the step 5 are as follows:
the method comprises the following steps of node power balance constraint, generator active and reactive power output constraint, node voltage upper and lower limit constraint, line maximum transmission power constraint and constraint condition addition of control variables, wherein the control variables are parallel compensation capacity upper and lower limit values, and the expression of the constraint condition is as follows:
s.t.
wherein, PkActive power is injected into a node k; qkInjecting reactive power into the node k; gkjIs the mutual conductance between nodes k and j; b iskjIs the transadmittance between nodes k and j; deltakjIs the voltage phase difference between nodes k and j; u shapekIs the voltage of node k; u shapejIs the voltage at node j; pGkActive power output is provided for the kth generator;is the active output lower limit of the generator k;is the active power output upper limit of the generator k; qGkReactive power output of a kth generator is achieved;is the lower reactive power output limit of the generator k;is the upper limit of reactive power output of the generator k; n is a radical ofGThe total number of the generator nodes of the power distribution network system is calculated;is the upper limit of the node voltage;is the node voltage lower limit; n is the total number of the nodes of the power distribution network system; slTransmitting power for the first branch; Maximum allowed transmission power for branch l; n is a radical of hydrogenlThe total number of the branch circuits of the power distribution network system is; qckCompensating the capacity for the kth reactive power compensation device; qc,maxIs the upper limit of the reactive compensation capacity; qc,minIs the lower limit of the reactive compensation capacity; n is a radical ofvThe number of the reactive compensation is.
8. The load curve segmentation-based reactive power optimization compensation method for the power distribution network according to claim 1, wherein the improved decomposition-based multi-objective optimization algorithm is realized by the following steps:
step 3.1, inputting algorithm parameters
Inputting improved multi-objective optimization algorithm parameters, wherein the parameters comprise population number NPThe number T of weight vectors in the field and the maximum iteration number genmax;
Step 3.2, initialize weight vector matrix
Defining the pareto optimal solution set to be empty initially, and decomposing the multi-objective optimization problem into N based on a Tchebycheff decomposition methodPThe single target sub-problems are optimized simultaneously, wherein each sub-problem can be expressed as:
in the formula, e represents an optimization objective; q is an optimization sub-problem, wherein q is 1,2, …, Np; f. ofe(x) Representing the target value of the optimization target e corresponding to the optimization variable x; z is an initialized target value;representing the weight corresponding to the optimization target e; gteFor the decomposed single-target optimization problem, the corresponding weight vectors are respectively [ lambda ] 1,…,λNp]And is made ofWhereinCalculating Euclidean distance matrix formed among the weight vectors, sorting according to Euclidean distance, and selecting T top distance ranks as weight vectors LambdaqI.e., e (q) ═ q (q)1,…,qT) Wherein E (q) represents a domain vector of the optimization sub-problem q, q1,…,qTRespectively represents elements in the field, and the total number of the elements is T;
step 3.3, random production of initial populations
Random uniform generation of initial population { x in optimization space1,…,xNpAnd calculating an optimization objective function value f corresponding to each populatione(x1),…,fe(xNp) The initial optimization target value z ═ z (z)1,z2,…,zm) Initializing a reference point, taking the minimum value in each sub-standard function value of all individuals as the reference point, ze=min(fe(x1),…,fe(xNp));
Step 3.4, evolution
For each population, in the field E (q), a random method is used to generate the sequence number k1、k2Through xk1And xk2Obtaining a new optimization variable solution y ', and obtaining an improved solution y from y' based on a heuristic method;
step 3.5, Gauss variation
Each element y in the solution y obtained after evolutioniPerforming Gaussian mutation operation, wherein the operation equation is as follows:
in the formula, Gussian (t) represents a Gaussian function with an argument of t; deltapiGiven standard deviation of gaussian variation;standard variance for a given gaussian variation; rand 1For generating random number operation satisfying the constraint of upper and lower limits of independent variable, C is the generated random number set, and N is totalvThe number of the cells; ciIs the ith element in the set C; p isiFor a given rate of variation; rand2Operating to generate random numbers subject to a Gaussian distribution;
step 3.6, updating the reference point
Updating the reference point z if the reference point z corresponding to the optimization target eeAnd its optimized target value feThe size relationship of (y) is as follows: z is a radical ofe>fe(y) then let ze=fe(y);
Step 3.7, updating the domain solution and Pareto optimal solution set
If for any of the weight elements in E (q) < lambda >rG exists between the new optimization variable solution y and the solution xte(y∣λr,z)≤gte(x∣λrZ), then let x ═ y, and the optimization target value f (x) corresponding to x is updated to f (x) ═ f (y), and at the same time, the Pareto optimal solution set is updated;
step 3.8, judging termination conditions
Judging whether the maximum value g of the iteration times is reacheden=genmaxIf so, stopping iteration and entering the next step, otherwise, iterating for a time gen=gen+1 and return to step 3.4;
step 3.9, output
And outputting the Pareto optimal solution set obtained after optimization, and ending the optimization process.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110674283.8A CN113300380B (en) | 2021-06-17 | 2021-06-17 | Load curve segmentation-based power distribution network reactive power optimization compensation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110674283.8A CN113300380B (en) | 2021-06-17 | 2021-06-17 | Load curve segmentation-based power distribution network reactive power optimization compensation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113300380A CN113300380A (en) | 2021-08-24 |
CN113300380B true CN113300380B (en) | 2022-06-28 |
Family
ID=77328589
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110674283.8A Active CN113300380B (en) | 2021-06-17 | 2021-06-17 | Load curve segmentation-based power distribution network reactive power optimization compensation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113300380B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114221352B (en) * | 2021-12-09 | 2023-06-27 | 国网内蒙古东部电力有限公司通辽供电公司 | Voltage coordination control method based on reactive power-active network loss partition aggregation area |
CN115047271A (en) * | 2022-06-10 | 2022-09-13 | 福州大学 | Diagnosis method for health condition of reactive compensation equipment |
CN115207938B (en) * | 2022-08-12 | 2023-02-07 | 费莱(浙江)科技有限公司 | Hybrid power reactive compensation method and system |
CN115589000B (en) * | 2022-12-12 | 2023-04-07 | 广东电网有限责任公司阳江供电局 | Area voltage adjustment optimization method and system based on typical load classification |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104836243A (en) * | 2015-05-27 | 2015-08-12 | 华北电力大学(保定) | Reactive compensation method capable of computing temporal characteristics of loads for power distribution network |
CN108110752A (en) * | 2017-12-06 | 2018-06-01 | 中国南方电网有限责任公司 | A kind of planned manner idle work optimization method a few days ago of adaptive network load variation |
CN109768573A (en) * | 2019-01-29 | 2019-05-17 | 三峡大学 | Var Optimization Method in Network Distribution based on multiple target difference grey wolf algorithm |
CN110518575A (en) * | 2019-08-02 | 2019-11-29 | 南京理工大学 | Multiple Time Scales active distribution network voltage optimization control method based on region division |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130066480A1 (en) * | 2011-09-07 | 2013-03-14 | Quanta Associates, L.P | Real-time monitoring of electric power system voltage stability margins |
-
2021
- 2021-06-17 CN CN202110674283.8A patent/CN113300380B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104836243A (en) * | 2015-05-27 | 2015-08-12 | 华北电力大学(保定) | Reactive compensation method capable of computing temporal characteristics of loads for power distribution network |
CN108110752A (en) * | 2017-12-06 | 2018-06-01 | 中国南方电网有限责任公司 | A kind of planned manner idle work optimization method a few days ago of adaptive network load variation |
CN109768573A (en) * | 2019-01-29 | 2019-05-17 | 三峡大学 | Var Optimization Method in Network Distribution based on multiple target difference grey wolf algorithm |
CN110518575A (en) * | 2019-08-02 | 2019-11-29 | 南京理工大学 | Multiple Time Scales active distribution network voltage optimization control method based on region division |
Non-Patent Citations (2)
Title |
---|
杨君军.基于PSO-DP算法的配电***动态无功优化研究.《中国优秀硕士学位论文全文数据库》.2012,全文. * |
王磊 等.基于典型负荷曲线的配电网线损计算方法研究.《配电网技术》.2020,第48卷(第3期),全文. * |
Also Published As
Publication number | Publication date |
---|---|
CN113300380A (en) | 2021-08-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113300380B (en) | Load curve segmentation-based power distribution network reactive power optimization compensation method | |
CN110535146B (en) | Electric power system reactive power optimization method based on depth determination strategy gradient reinforcement learning | |
CN107437813B (en) | Power distribution network reactive power optimization method based on cuckoo-particle swarm | |
CN110417015A (en) | Micro-capacitance sensor Multiobjective Optimal Operation method and system based on Model Predictive Control | |
CN107591844A (en) | Consider the probabilistic active distribution network robust reconstructing method of node injecting power | |
CN110212551B (en) | Micro-grid reactive power automatic control method based on convolutional neural network | |
CN107516892A (en) | The method that the quality of power supply is improved based on processing active optimization constraints | |
CN116914751B (en) | Intelligent power distribution control system | |
CN113378100B (en) | Power distribution network source network load storage collaborative optimization scheduling model and method considering carbon emission | |
CN115986728A (en) | Power grid situation prediction method considering uncertainty factors and terminal | |
CN115313403A (en) | Real-time voltage regulation and control method based on deep reinforcement learning algorithm | |
CN109066728A (en) | The online damp coordinated control method of the multiple inter-area oscillation modes of extra-high voltage grid | |
CN110460043B (en) | Power distribution network frame reconstruction method based on multi-target improved particle swarm algorithm | |
CN115588998A (en) | Graph reinforcement learning-based power distribution network voltage reactive power optimization method | |
CN115795992A (en) | Park energy Internet online scheduling method based on virtual deduction of operation situation | |
Zhao et al. | An improved particle swarm optimization algorithm for optimal reactive power dispatch | |
CN114429248A (en) | Transformer apparent power prediction method | |
Chen et al. | A Novel Approach Based on Modified and Hybrid Flower Pollination Algorithm to Solve Multi-objective Optimal Power Flow. | |
Azadani et al. | Optimal placement of multiple STATCOM | |
CN117674263A (en) | Distribution network photovoltaic site selection and volume determination and bearing capacity evaluation method considering photovoltaic absorption rate | |
CN105896547B (en) | A kind of bulk power grid hierarchical voltage control method under wind power integration | |
Zhang et al. | A multi-objective optimization method for power system reactive power dispatch | |
CN114400675B (en) | Active power distribution network voltage control method based on weight mean value deep double-Q network | |
CN111293687A (en) | Three-dimensional particle swarm algorithm-based distributed power supply location and volume determination method | |
CN113690930B (en) | NSGA-III algorithm-based medium and long term locating and sizing method for distributed photovoltaic power supply |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |