CN112686404A - Power distribution network fault first-aid repair-based collaborative optimization method - Google Patents
Power distribution network fault first-aid repair-based collaborative optimization method Download PDFInfo
- Publication number
- CN112686404A CN112686404A CN202011602253.8A CN202011602253A CN112686404A CN 112686404 A CN112686404 A CN 112686404A CN 202011602253 A CN202011602253 A CN 202011602253A CN 112686404 A CN112686404 A CN 112686404A
- Authority
- CN
- China
- Prior art keywords
- training
- fault
- distribution network
- power distribution
- classification
- 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.)
- Granted
Links
- 230000008439 repair process Effects 0.000 title claims abstract description 71
- 238000005457 optimization Methods 0.000 title claims abstract description 59
- 238000009826 distribution Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012549 training Methods 0.000 claims abstract description 71
- 238000012360 testing method Methods 0.000 claims abstract description 10
- 238000012843 least square support vector machine Methods 0.000 claims abstract description 4
- 239000007787 solid Substances 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 239000002245 particle Substances 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 238000012804 iterative process Methods 0.000 claims 1
- 239000000203 mixture Substances 0.000 claims 1
- 239000002994 raw material Substances 0.000 claims 1
- 238000004422 calculation algorithm Methods 0.000 description 24
- 238000012423 maintenance Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 108090000623 proteins and genes Proteins 0.000 description 2
- 238000002922 simulated annealing Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003016 pheromone Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a cooperative optimization method based on power distribution network fault first-aid repair, which comprises the steps of collecting first-aid repair resource data of power distribution network faults to extract characteristics, and respectively obtaining training sample data, test sample data and corresponding fault first-aid repair category data; constructing a training model based on a weighted least square support vector machine by using the training sample data and the distance between the classification planes; training the training model, controlling the weak classifier to generate an identification error rate by using the filter factor, stopping training until the weak classifier meeting the conditions cannot be found, and outputting a fault first-aid repair identification model; importing the test sample data into the fault first-aid repair identification model, and finishing optimization if the actual fault first-aid repair type is correctly identified; and identifying the power distribution network fault emergency repair resource data by using the optimized fault emergency repair identification model and outputting an optimized identification result. The method improves the accuracy and reliability of fault first-aid repair classification.
Description
Technical Field
The invention relates to the technical field of power distribution network fault first-aid repair resources and weighted optimization, in particular to a collaborative optimization method based on power distribution network fault first-aid repair.
Background
The natural disasters can damage the power system, and severe faults such as pole collapse, line breakage, permanent short circuit, transformer collision damage and the like can be caused by more than 10 grades of hurricanes brought by typhoons and other foreign matters blown by the hurricanes; flood can loosen the foundation of the power tower, induce pole-reversing accidents and cause serious accidents such as water inlet, shutdown and the like of a transformer substation; the ice and snow disasters can lead the lead and the tower to be covered with ice, so that the electric tower collapses due to self weight and is damaged by leading nearby power transformation equipment through the electric wire; geological disasters such as earthquake and the like can destroy various electric power buildings including towers and substations, and even completely destroy the power supply network of disaster areas.
When formulating the distribution network trouble rush repair strategy under the calamity condition, two points should be noted: firstly, under the disaster condition, the number of faults is large, so that the scale of the optimization problem is large; second, emergency repair under a disaster condition easily occurs, so that initial parameters such as faults and external environments are frequently updated in optimization problems, and the optimization algorithm is required to perform frequent repeated operation and timely update emergency repair strategies.
Compared with the modern intelligent optimization algorithm, the traditional optimization algorithm has a long history and has a strict theoretical basis, the optimal solution can be guaranteed to be solved as long as the problem is suitable for the algorithm, but the traditional optimization algorithm is only suitable for the optimization problem with feasible solution and continuous space, and when the related optimization problem is a combination optimization problem, namely, the permutation and combination of numbers are optimized, the traditional optimization algorithm can hardly be used for the problem.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The present invention has been made in view of the above-mentioned conventional problems.
Therefore, the invention provides a distribution network fault first-aid repair-based collaborative optimization method, which can solve the problem that the digital permutation and combination cannot be optimized by the existing distribution network fault first-aid repair optimization algorithm.
In order to solve the technical problems, the invention provides the following technical scheme: collecting emergency repair resource data of a power distribution network fault to perform feature extraction, and respectively obtaining training sample data, test sample data and corresponding fault emergency repair category data; constructing a training model based on a weighted least square support vector machine by using the training sample data and the distance between the classification planes; training the training model, controlling the weak classifier to generate an identification error rate by using the filter factor, stopping training until the weak classifier meeting the conditions cannot be found, and outputting a fault first-aid repair identification model; importing the test sample data into the fault first-aid repair identification model, and finishing optimization if the actual fault first-aid repair type is correctly identified; and identifying the power distribution network fault emergency repair resource data by using the optimized fault emergency repair identification model and outputting an optimized identification result.
As an optimal scheme of the cooperative optimization method based on the distribution network fault first-aid repair, the method comprises the following steps: the characteristic extraction comprises the steps of defining the initial state of the fault solid particles of the power distribution network as i and the energy as EiAt a temperature of TiAnd randomly disturbing the initial state to obtain a new state j, wherein the corresponding energy state of the new state j is Ej(ii) a If Ej<EiIf not, judging whether to accept the new state according to the probability of the solid in the new state; the probability ratio formula of the solid in the states i and j is
Wherein E isi、EjThe energies of the molecules in the i-th and j-th states, T is the temperature in degrees Kelvin, kbBoltzmann constant.
As an optimal scheme of the cooperative optimization method based on the distribution network fault first-aid repair, the method comprises the following steps: the weighted least squares support vector machine obtains different classification results for the same training sample data by using three types of division, including,
wherein v isi: weight of the sample, c1、c2: parameter, di: the distance between the ith sample and the classification plane.
As an optimal scheme of the cooperative optimization method based on the distribution network fault first-aid repair, the method comprises the following steps: the training model divides the training sample data into the three types based on the distance between the training sample data and the classification plane, wherein the three types comprise type A data, type B data and type C data; the type A data comprises sample data which is endowed with larger weight and smaller distance; said type B data comprises said sample data being given intermediate-size weight and intermediate distance; the type C data includes the sample data given the smallest weight and the largest distance.
As an optimal scheme of the cooperative optimization method based on the distribution network fault first-aid repair, the method comprises the following steps: the filtering factor includes defining a ratio of the recognition error rate generated by the current weak classifier to the recognition error rate of the first weak classifier as the filtering factor; when the classification is carried out for the second time, and the conditions that the recognition error rate is less than 0.5 and less than the product of the first weak classifier and the filter factor are simultaneously met, generating a new weak classifier; repeating the classification step of generating the weak classifiers until the weak classifiers meeting the conditions can not be found, and stopping classification; the weak classifier respectively generates classification opinions to the training sample data during classification; and integrating the classification opinions of all the generated weak classifiers, and respectively marking the classification opinions as the same classification opinion and different classification opinions.
The invention is based on the power distribution networkAn optimized scheme of a cooperative optimization method for emergency repair of barriers, wherein: training the training models includes inputting labeled training samples { (x) for the training models, respectively1,y1),…,(xN,yN) The penalty factor C and the kernel width parameter initial value sigma in the weighted LSSVM are calculatediniSearch step σstepMinimum value σminA filter factor k and a weight model parameter c1、c2(ii) a Initializing input values, training sample initial weights
Weighted LSSVM initial weight
Carrying out iterative processing on the training model, wherein the iteration time T is 1, 2, … and T; iteration is completed, the fault first-aid repair identification model is output,
wherein x isi: the training sample data, yi: training sample labels, ht: the weak classifier, αt: and (4) weighting.
As an optimal scheme of the cooperative optimization method based on the distribution network fault first-aid repair, the method comprises the following steps: the iterative processing comprises classifying the training sample data by using a weighted LSSVM under the distribution of the current sample weight to obtain the weak classifier; the training model calculates the recognition error rate of the weak classifier,
wherein epsilont: the recognition error rate, ωi t: a current sample weight distribution.
As an optimal scheme of the cooperative optimization method based on the distribution network fault first-aid repair, the method comprises the following steps: if the actual fault first-aid repair type is correctly identified, the optimization is completed, and the formed objective function comprises,
wherein, tiIs the time for recovering power supply from the i th fault after the first-aid plan is executed, and the mathematical optimization aims to minimize the function value, omegaiCoefficient of importance representing the i-th load node, determined directly from its load class, PiThe average value of active power consumed by the i-th load node is shown, and the change of a system function F (t) of the power distribution network along with time is shown as toughness.
As an optimal scheme of the cooperative optimization method based on the distribution network fault first-aid repair, the method comprises the following steps: outputting the emergency repair optimization identification result comprises the following steps,
Rreal(t)＝R*inc(t)
where R is a path time matrix, inc (t) is a path increment coefficient, R is a fixed value, and inc (t) is a dynamic parameter that varies in hours, which is a positive real number.
The invention has the beneficial effects that: the method of the invention adopts the distance weight updating model, divides and updates the sample weight into three types according to the distance between the sample and the classification plane, and controls the generation of the weak classifiers by using the filter factors, so that a series of weak classifiers which have a small amount of classification differences and a large amount of classification consensus are generated.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
fig. 1 is a schematic flow chart of a cooperative optimization method based on power distribution network fault emergency repair according to an embodiment of the present invention;
fig. 2 is a schematic diagram of training sample data and a plane classification distance of a power distribution network fault first-aid repair-based collaborative optimization method according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a power grid structure and fault distribution of a power distribution network fault first-aid repair-based collaborative optimization method according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, specific embodiments accompanied with figures are described in detail below, and it is apparent that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present invention, shall fall within the protection scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
The present invention will be described in detail with reference to the drawings, wherein the cross-sectional views illustrating the structure of the device are not enlarged partially in general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Meanwhile, in the description of the present invention, it should be noted that the terms "upper, lower, inner and outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and operate, and thus, cannot be construed as limiting the present invention. Furthermore, the terms first, second, or third are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected and connected" in the present invention are to be understood broadly, unless otherwise explicitly specified or limited, for example: can be fixedly connected, detachably connected or integrally connected; they may be mechanically, electrically, or directly connected, or indirectly connected through intervening media, or may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
Referring to fig. 1 and 2, for a first embodiment of the present invention, a cooperative optimization method based on power distribution network fault emergency repair is provided, including:
s1: collecting emergency repair resource data of the power distribution network fault for feature extraction, and respectively obtaining training sample data, test sample data and corresponding fault emergency repair category data. It should be noted that the feature extraction includes:
defining the initial state of the fault solid particles of the power distribution network as i and the energy as EiAt a temperature of TiMaking a random disturbance to the initial state to obtain a new state j, whose corresponding energy state is Ej；
If Ej<EiIf not, judging whether to accept the new state according to the probability that the solid is in the new state;
the probability ratio of the solid in the states i and j is expressed as
Wherein E isi、EjThe energies of the molecules in the i-th and j-th states, T is the temperature in degrees Kelvin, kbBoltzmann constant.
S2: and constructing a training model based on a weighted least square support vector machine by using the distance between training sample data and the classification plane. It should be noted that, in this step, the training model divides the training sample data into three types based on the distance between the training sample data and the classification plane, including:
type a data, type B data, and type C data;
the type a data includes sample data given a larger weight and a smaller distance;
the type B data comprises sample data which is endowed with middle-size weight and middle distance;
the type C data includes sample data to which a minimum weight and a distance are given to be maximum;
the weighted least squares support vector machine utilizes the three types of division to obtain different classification results for the same training sample data, including,
wherein v isi: weight of the sample, c1、c2: parameter, di: the distance between the ith sample and the classification plane.
S3: training the training model, controlling the weak classifier to generate the recognition error rate by using the filter factor, stopping training until the weak classifier meeting the conditions cannot be found, and outputting a fault first-aid repair recognition model. The training model comprises the following steps:
respectively inputting labeled training samples (x) into the training model1,y1),…,(xN,yN) The penalty factor C and the kernel width parameter initial value sigma in the weighted LSSVM are calculatediniSearch step σstepMinimum value σminA filter factor k and a weight model parameter c1、c2；
Initializing input values, training sample initial weights
Weighted LSSVM initial weight
Carrying out iterative processing on the training model, wherein the iteration time T is 1, 2, … and T;
iteration is completed, a fault first-aid repair identification model is output,
wherein x isi: training sample data, yi: training sample labels, ht: weak classifier, αt: and (4) weighting.
Further, the filter factor includes:
defining the ratio of the recognition error rate generated by the current weak classifier to the recognition error rate of the first weak classifier as a filtering factor;
when the classification is carried out for the second time, the conditions that the recognition error rate is less than 0.5 and less than the product of the first weak classifier and the filter factor are simultaneously met, and a new weak classifier is generated;
repeating the classification step of generating the weak classifiers until the weak classifiers meeting the conditions can not be found, and stopping classification;
respectively generating classification opinions for training sample data by the weak classifier during classification;
and integrating all the generated classification opinions of the weak classifiers, and respectively marking the classification opinions as the same classification opinion and different classification opinions.
Specifically, the iterative processing includes:
classifying training sample data by using a weighted LSSVM under the current sample weight distribution to obtain a weak classifier;
the training model calculates the recognition error rate of the weak classifier,
wherein epsilont: identifying error rate, ωi t: a current sample weight distribution.
S4: and importing the test sample data into the fault first-aid repair identification model, and finishing optimization if the actual fault first-aid repair type is correctly identified. It should be further noted that, if the actual failure emergency repair type is correctly identified, the optimization is completed, and the formed objective function includes:
wherein, tiIs the time for recovering power supply from the i th fault after the first-aid plan is executed, and the mathematical optimization aims to minimize the function value, omegaiCoefficient of importance representing the i-th load node, determined directly from its load class, PiThe average value of active power consumed by the i-th load node is shown, and the change of a system function F (t) of the power distribution network along with time is shown as toughness.
S5: and identifying the power distribution network fault emergency repair resource data by using the optimized fault emergency repair identification model and outputting an optimized identification result. It should be further noted that the outputting of the emergency repair optimization identification result includes:
Rreal(t)＝R*inc(t)
where R is a path time matrix, inc (t) is a path increment coefficient, R is a fixed value, and inc (t) is a dynamic parameter that varies in hours, which is a positive real number.
Referring to fig. 2, the method of the present invention makes full use of the distance information between the sample and the classification plane, wherein the dots and the squares represent two different types of data respectively, the classification plane generated in the classification problem divides the two different types of data, the vertical distance from the sample to the classification plane is defined as the distance from the sample to the classification plane, the greater the difficulty of correctly distinguishing the sample is considered, the greater the required weight is, and the method of the present invention provides the limitation condition of the filter factor for the clear correlation between the weak classifiers, controls the difference between the weak classifiers while increasing the diversity of the weak classifiers in the AdaBoost algorithm, and has higher identification classification accuracy and stronger robustness.
Preferably, the method provided by the invention adopts the distance weight updating model, divides and updates the sample weight into three types according to the distance between the sample and the classification plane, and controls the generation of the weak classifiers by using the filter factors, so that a series of weak classifiers which have a small amount of classification differences and a large amount of classification consensus are generated.
Example 2
In order to better verify the technical effect of the method of the present invention, the embodiment selects the traditional simulated annealing algorithm, genetic algorithm, ant colony algorithm and the method of the present invention to perform a comparative test, and compares the experimental results by means of scientific demonstration to verify the real effect of the method of the present invention.
Referring to fig. 3, a fault occurring between load nodes is a line fault, a fault occurring on a load node is a fault of the load node itself, does not affect line operation, and includes 1 source node, 32 load nodes, 10 faults,The system comprises 5 emergency maintenance teams, wherein the 5 emergency maintenance teams comprise 2 line emergency maintenance teams, 1 cable emergency maintenance team and 2 large-scale engineering vehicles; only one starting point of the team is defined, the efficiency loss coefficient k of each fault is set to be 0.64, and the lowest efficiency coefficient a of the emergency maintenance team allowing emergency maintenancemin＝0.3。
The fault problem is solved by utilizing SAA, GA, COA and the method of the invention, and the respective solving results are compared, and the parameters of four algorithms participating in comparison are as follows:
(1) simulated Annealing Algorithm (SAA): initial temperature t0＝1030Final temperature tmin＝10-30，Lk1, α is 0.9 (1312 iterations);
(2) genetic Algorithm (GA): the population quantity is 20, the gene crossing probability is 70%, the gene variation probability is 10%, and the iterative computation is carried out for 300 times;
(3) ant colony algorithm (COA): the number of the ant colony individuals is 20, the pheromone reduction coefficient is 0.9975, the rise coefficient is 1.0025, and the iterative computation is carried out for 500 times;
(4) inventive method (distance weighted LSSVM): filtering factor is the distance weighted value of the dynamic influence factor of the external environment, the wavelet packet transformation is adopted to carry out 6-layer decomposition on all fault first-aid repair data, and the first 16-dimensional wavelet packet coefficient (marked as E) of the 6 th layer is extracted6 i)。
The basic principle of the four algorithms is random search, so that when the iteration times are not infinite, the final optimization result fluctuates, in order to reflect the respective actual performance, after a plurality of independent repeated tests are carried out, the research idea of analyzing data by using a statistical method is adopted, 500 independent repeated optimization calculations are carried out by using each of the four algorithms, and compared indexes comprise 5 items of program average running time, final optimization objective function mean value, optimization rate, 1/10000 optimized solution rate and 5/10000 optimized solution rate, wherein the optimization rate refers to the probability of searching the actual optimal solution of the optimization problem by the algorithm, and represents the capability of searching the optimal solution by the algorithm; 1/10000(5/10000) the optimal solution rate indicates the probability that the algorithm finds the optimal solution before 1/10000(5/10000) the optimization problem, the higher the value, the stronger the stability of the algorithm.
Table 1: data comparison table of 500 independent repeated experiments.
Referring to table 1, compared with three traditional algorithms, the method disclosed by the invention has advantages in operation speed, stability and optimizing capability, and the analysis and speculation of the advantages of the distance weighted weak classifier algorithm are verified, so that the method disclosed by the invention is more suitable for solving the problem of optimization of the distribution network fault emergency repair strategy under the disaster condition compared with other common group-based intelligent algorithms.
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (9)
1. A cooperative optimization method based on power distribution network fault first-aid repair is characterized by comprising the following steps: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
collecting emergency repair resource data of a power distribution network fault for feature extraction, and respectively obtaining training sample data, test sample data and corresponding fault emergency repair category data;
constructing a training model based on a weighted least square support vector machine by using the training sample data and the distance between the classification planes;
training the training model, controlling the weak classifier to generate an identification error rate by using the filter factor, stopping training until the weak classifier meeting the conditions cannot be found, and outputting a fault first-aid repair identification model;
importing the test sample data into the fault first-aid repair identification model, and finishing optimization if the actual fault first-aid repair type is correctly identified;
and identifying the power distribution network fault emergency repair resource data by using the optimized fault emergency repair identification model and outputting an optimized identification result.
2. The power distribution network fault emergency repair-based collaborative optimization method according to claim 1, characterized in that: the feature extraction includes the steps of,
defining the initial state of the fault solid particles of the power distribution network as i and the energy as EiAt a temperature of TiAnd randomly disturbing the initial state to obtain a new state j, wherein the corresponding energy state of the new state j is Ej；
If Ej<EiIf not, judging whether to accept the new state according to the probability of the solid in the new state;
the probability ratio formula of the solid in the states i and j is
Wherein E isi、EjThe energies of the molecules in the i-th and j-th states, T is the temperature in degrees Kelvin, kbBoltzmann constant.
3. The power distribution network fault emergency repair-based collaborative optimization method according to claim 1 or 2, characterized in that: the weighted least squares support vector machine obtains different classification results for the same training sample data by using three types of division, including,
wherein v isi: weight of the sample, c1、c2: parameter, di: the ithThe distance between the sample and the classification plane.
4. The power distribution network fault emergency repair-based collaborative optimization method according to claim 3, characterized in that: the training model divides the training sample data into the three types based on the distance between the training sample data and the classification plane, wherein the three types comprise type A data, type B data and type C data;
the type A data comprises sample data which is endowed with larger weight and smaller distance;
said type B data comprises said sample data being given intermediate-size weight and intermediate distance;
the type C data includes the sample data given the smallest weight and the largest distance.
5. The power distribution network fault emergency repair-based collaborative optimization method according to claim 4, characterized in that: the filter factors include, for example,
defining a ratio of the recognition error rate generated by the current weak classifier to the recognition error rate of the first weak classifier as the filtering factor;
when the classification is carried out for the second time, and the conditions that the recognition error rate is less than 0.5 and less than the product of the first weak classifier and the filter factor are simultaneously met, generating a new weak classifier;
repeating the classification step of generating the weak classifiers until the weak classifiers meeting the conditions can not be found, and stopping classification;
the weak classifier respectively generates classification opinions to the training sample data during classification;
and integrating the classification opinions of all the generated weak classifiers, and respectively marking the classification opinions as the same classification opinion and different classification opinions.
6. The power distribution network fault emergency repair-based collaborative optimization method according to claim 5, characterized in that: training the training model may include training the training model to include,
to what is neededThe training model respectively inputs training samples with labels { (x)1,y1),…,(xN,yN) The penalty factor C and the kernel width parameter initial value sigma in the weighted LSSVM are calculatediniSearch step σstepMinimum value σminA filter factor k and a weight model parameter c1、c2；
Initializing input values, training sample initial weights
Weighted LSSVM initial weight
＝1，i＝1,…,N；
Carrying out iterative processing on the training model, wherein the iteration time T is 1, 2, … and T;
iteration is completed, the fault first-aid repair identification model is output,
wherein x isi: the training sample data, yi: training sample labels, ht: the weak classifier, αt: and (4) weighting.
7. The power distribution network fault emergency repair-based collaborative optimization method according to claim 6, characterized in that: the iterative process may include the steps of,
classifying the training sample data by using a weighted LSSVM under the current sample weight distribution to obtain the weak classifier;
the training model calculates the recognition error rate of the weak classifier,
wherein epsilont: the recognition error rate, ωi t: a current sample weight distribution.
8. The power distribution network fault emergency repair-based collaborative optimization method according to claim 7, characterized in that: if the actual fault first-aid repair type is correctly identified, the optimization is completed, and the formed objective function comprises,
wherein, tiIs the time for recovering power supply from the i th fault after the first-aid plan is executed, and the mathematical optimization aims to minimize the function value, omegaiCoefficient of importance representing the i-th load node, determined directly from its load class, PiThe average value of active power consumed by the i-th load node is shown, and the change of a system function F (t) of the power distribution network along with time is shown as toughness.
9. The power distribution network fault emergency repair-based collaborative optimization method according to claim 8, wherein: outputting the emergency repair optimization identification result comprises the following steps,
Rreal(t)＝R*inc(t)
where R is a path time matrix, inc (t) is a path increment coefficient, R is a fixed value, and inc (t) is a dynamic parameter that varies in hours, which is a positive real number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011602253.8A CN112686404B (en) | 2020-12-29 | 2020-12-29 | Power distribution network fault first-aid repair-based collaborative optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011602253.8A CN112686404B (en) | 2020-12-29 | 2020-12-29 | Power distribution network fault first-aid repair-based collaborative optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112686404A true CN112686404A (en) | 2021-04-20 |
| CN112686404B CN112686404B (en) | 2022-05-06 |
Family
ID=75454422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011602253.8A Active CN112686404B (en) | 2020-12-29 | 2020-12-29 | Power distribution network fault first-aid repair-based collaborative optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112686404B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7240691B1 (en) | 2021-09-08 | 2023-03-16 | 山東大学 | Data drive active power distribution network abnormal state detection method and system |
| CN117522091A (en) * | 2024-01-08 | 2024-02-06 | 国网四川省电力公司电力科学研究院 | Intelligent scheduling system and method for emergency repair of post-earthquake power equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103136587A (en) * | 2013-03-07 | 2013-06-05 | 武汉大学 | Power distribution network operating state classification recognition method based on support vector machine |
| CN104915638A (en) * | 2015-05-08 | 2015-09-16 | 国家电网公司 | Least squares support vector machine electric shock current detection method based on parameter optimization |
| WO2016046744A1 (en) * | 2014-09-26 | 2016-03-31 | Thomson Reuters Global Resources | Pharmacovigilance systems and methods utilizing cascading filters and machine learning models to classify and discern pharmaceutical trends from social media posts |
| CN106815647A (en) * | 2016-12-28 | 2017-06-09 | 国家电网公司 | A kind of high efficiency distribution network failure repairing system and method based on data analysis |
| CN109919178A (en) * | 2019-01-23 | 2019-06-21 | 广西大学 | Failure prediction method based on characteristic quantity preferably with Wavelet Kernel Function LSSVM |
| CN110854907A (en) * | 2019-11-07 | 2020-02-28 | 华北电力大学 | Collaborative optimization operation method and system for power distribution network wind power plant under communication fault |
| CN111582350A (en) * | 2020-04-30 | 2020-08-25 | 上海电力大学 | Filtering factor optimization AdaBoost method and system based on distance weighted LSSVM |
| CN111654066A (en) * | 2020-05-20 | 2020-09-11 | 中国电力科学研究院有限公司 | Method and system for determining filter factor of water-fire-electricity frequency division control AGC system |
| CN112039073A (en) * | 2020-09-18 | 2020-12-04 | 上海交通大学烟台信息技术研究院 | Collaborative optimization method and system suitable for fault judgment of power distribution room equipment |
-
2020
- 2020-12-29 CN CN202011602253.8A patent/CN112686404B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103136587A (en) * | 2013-03-07 | 2013-06-05 | 武汉大学 | Power distribution network operating state classification recognition method based on support vector machine |
| WO2016046744A1 (en) * | 2014-09-26 | 2016-03-31 | Thomson Reuters Global Resources | Pharmacovigilance systems and methods utilizing cascading filters and machine learning models to classify and discern pharmaceutical trends from social media posts |
| CN104915638A (en) * | 2015-05-08 | 2015-09-16 | 国家电网公司 | Least squares support vector machine electric shock current detection method based on parameter optimization |
| CN106815647A (en) * | 2016-12-28 | 2017-06-09 | 国家电网公司 | A kind of high efficiency distribution network failure repairing system and method based on data analysis |
| CN109919178A (en) * | 2019-01-23 | 2019-06-21 | 广西大学 | Failure prediction method based on characteristic quantity preferably with Wavelet Kernel Function LSSVM |
| CN110854907A (en) * | 2019-11-07 | 2020-02-28 | 华北电力大学 | Collaborative optimization operation method and system for power distribution network wind power plant under communication fault |
| CN111582350A (en) * | 2020-04-30 | 2020-08-25 | 上海电力大学 | Filtering factor optimization AdaBoost method and system based on distance weighted LSSVM |
| CN111654066A (en) * | 2020-05-20 | 2020-09-11 | 中国电力科学研究院有限公司 | Method and system for determining filter factor of water-fire-electricity frequency division control AGC system |
| CN112039073A (en) * | 2020-09-18 | 2020-12-04 | 上海交通大学烟台信息技术研究院 | Collaborative optimization method and system suitable for fault judgment of power distribution room equipment |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7240691B1 (en) | 2021-09-08 | 2023-03-16 | 山東大学 | Data drive active power distribution network abnormal state detection method and system |
| JP2023042527A (en) * | 2021-09-08 | 2023-03-27 | 山東大学 | Data drive active power distribution network abnormal state sensing method and system |
| CN117522091A (en) * | 2024-01-08 | 2024-02-06 | 国网四川省电力公司电力科学研究院 | Intelligent scheduling system and method for emergency repair of post-earthquake power equipment |
| CN117522091B (en) * | 2024-01-08 | 2024-04-16 | 国网四川省电力公司电力科学研究院 | Intelligent scheduling system and method for emergency repair of post-earthquake power equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112686404B (en) | 2022-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110596492B (en) | Transformer fault diagnosis method based on particle swarm optimization random forest model | |
| US20230136352A1 (en) | Method and system for predicting a day-ahead wind power of wind farms | |
| CN105354646B (en) | Power load forecasting method for hybrid particle swarm optimization and extreme learning machine | |
| CN112686404B (en) | Power distribution network fault first-aid repair-based collaborative optimization method | |
| CN110942205B (en) | Short-term photovoltaic power generation power prediction method based on HIMVO-SVM | |
| CN106897826A (en) | A kind of street accidents risks appraisal procedure and system | |
| CN110598726A (en) | Transmission tower bird damage risk prediction method based on random forest | |
| CN107346459B (en) | Multi-mode pollutant integrated forecasting method based on genetic algorithm improvement | |
| CN104951834A (en) | LSSVM (least squares support vector machine) wind speed forecasting method based on integration of GA (genetic algorithm) and PSO (particle swarm optimization) | |
| CN110969262A (en) | Transformer fault diagnosis method | |
| CN110689069A (en) | Transformer fault type diagnosis method based on semi-supervised BP network | |
| CN105388402A (en) | Cross-linked cable partial discharge mode recognition method based on parameter optimization SVM (Support Vector Machine) algorithm | |
| CN109214604A (en) | The haze calamity risk forecast method and system of power transmission and transforming equipment based on Adaboost | |
| CN114330100A (en) | Short-term photovoltaic power probability interval prediction method | |
| CN104732067A (en) | Industrial process modeling forecasting method oriented at flow object | |
| CN116109178A (en) | High-proportion new energy access-oriented power system stability evaluation method | |
| CN111242266A (en) | Operation data management system | |
| CN110348489B (en) | Transformer partial discharge mode identification method based on self-coding network | |
| CN116663393A (en) | Random forest-based power distribution network continuous high-temperature fault risk level prediction method | |
| CN116826743A (en) | Power load prediction method based on federal graph neural network | |
| CN113837096B (en) | Rolling bearing fault diagnosis method based on GA random forest | |
| CN113466630B (en) | Power distribution network fault cause classification method based on RSSPN model | |
| CN113641990A (en) | Intrusion detection method based on multi-innovation extended Kalman filtering | |
| CN113420829A (en) | Weather-related fault prediction method based on Bayesian neural network | |
| CN107508279B (en) | Stability simulation method for power network |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220419 Address after: 250000 room 1-3133, building 1, No. 88, West Exhibition Road, high tech Zone, Jinan, Shandong Province Applicant after: Shandong Huake Information Technology Co.,Ltd. Applicant after: Beijing Huaqing Future Energy Technology Research Institute Co.,Ltd. Applicant after: Beijing Huaqing Zhihui Energy Technology Co.,Ltd. Address before: 210000 288 Qinhuai Road, Yung Yang Street, Lishui District, Nanjing, Jiangsu Applicant before: Nanjing housheng Yuanda Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A collaborative optimization method based on distribution network fault repair Effective date of registration: 20230106 Granted publication date: 20220506 Pledgee: Shandong science and technology finance Company limited by guarantee Pledgor: Shandong Huake Information Technology Co.,Ltd. Registration number: Y2023370000004 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Granted publication date: 20220506 Pledgee: Shandong science and technology finance Company limited by guarantee Pledgor: Shandong Huake Information Technology Co.,Ltd. Registration number: Y2023370000004 |