CN109886560A - Distribution network transform measure and rate of qualified voltage index relevance method for digging and device - Google Patents
Distribution network transform measure and rate of qualified voltage index relevance method for digging and device Download PDFInfo
- Publication number
- CN109886560A CN109886560A CN201910077166.6A CN201910077166A CN109886560A CN 109886560 A CN109886560 A CN 109886560A CN 201910077166 A CN201910077166 A CN 201910077166A CN 109886560 A CN109886560 A CN 109886560A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- rate
- neural network
- output
- network
- 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.)
- Pending
Links
- 238000009826 distribution Methods 0.000 title claims abstract description 189
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000013528 artificial neural network Methods 0.000 claims abstract description 124
- 238000012797 qualification Methods 0.000 claims abstract description 77
- 238000012986 modification Methods 0.000 claims abstract description 42
- 230000004048 modification Effects 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 38
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 34
- 238000005457 optimization Methods 0.000 claims abstract description 32
- 230000002068 genetic effect Effects 0.000 claims abstract description 29
- 238000013507 mapping Methods 0.000 claims abstract description 25
- 238000013139 quantization Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 35
- 230000006870 function Effects 0.000 claims description 28
- 238000004088 simulation Methods 0.000 claims description 24
- 238000012545 processing Methods 0.000 claims description 19
- 239000000243 solution Substances 0.000 claims description 16
- 238000004146 energy storage Methods 0.000 claims description 15
- 239000003637 basic solution Substances 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 6
- 238000009434 installation Methods 0.000 claims description 6
- 230000007774 longterm Effects 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000013178 mathematical model Methods 0.000 claims description 5
- 239000004744 fabric Substances 0.000 claims description 4
- 210000004218 nerve net Anatomy 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 239000002585 base Substances 0.000 description 4
- 238000000547 structure data Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 238000000342 Monte Carlo simulation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000010429 evolutionary process Effects 0.000 description 1
- 210000004209 hair Anatomy 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000009418 renovation Methods 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Abstract
The present invention relates to a kind of distribution network transform measures and rate of qualified voltage index relevance method for digging and device, this method comprises: all kinds of factor layer indexs of analog quantization, obtain the sample data of active distribution network configurable resource modification measures and the incidence relation of corresponding runnability, the direct mapping between distribution network transform measure and rate of qualified voltage is constructed using the non-linear mapping capability of BP neural network, using sample data as training sample, BP neural network is trained, obtain the distribution network voltage qualification rate assessment models under the different modification measures based on BP neural network, parameter optimization is finally carried out to BP neural network using genetic algorithm, obtain more accurate distribution network voltage qualification rate assessment models.This method and device are conducive to find the potential relevance between distribution network transform measure and rate of qualified voltage index, and improve computational efficiency.
Description
Technical field
The present invention relates to power grid construction renovation technique field, especially a kind of distribution network transform measure refers to rate of qualified voltage
Mark relevance method for digging and device.
Background technique
It is the distributed energy of representative as green energy resource using photovoltaic power generation as the continuous propulsion of renewable energy develops
Main body of development largely dispersedly accesses low and medium voltage distribution network.For the consumption for effectively solving the problems, such as large-scale distributed power grid,
Active distribution network will become the main development form of the following smart grid.Simultaneously more and more distributed energies, energy-storage system,
Flexible load accesses in power distribution network, it will there are a large amount of non-linear elements and equipment, all these is all that voltage and current generates
The potential source of harmonic wave, while the uncertainty of distributed energy member processing, can all adversely affect electric power netting safe running.With
Upper problem has been difficult to adapt to for conventional electrical distribution net.For power quality, the high reliability operation, Yi Jifen for guaranteeing power grid
It is very urgent to accelerate distribution network construction for the high permeability of the cloth energy.Rate of qualified voltage in distribution network transform measure as wanting
One of major issue of solution is the critical issue in distribution network construction, is proposed to the optimization of investment decision of power distribution network higher
Requirement and challenge.
For the optimization of investment decision problem for solving power distribution network, active distribution network modification measures and network voltage qualification rate are utilized
It is most important to establish accurate investment decision model for relationship between index.However, the voltage in traditional investment decision modeling
Yield analysis is related to electric network swim calculating, and the analytic process of physical model is extremely complex, is unfavorable for current distribution network planning
Draw investment decision.And existing big data analysis method is applied to the solution that the two association analysis is more advantageous to the problem.
Commonly relevant regular (Apriori) method of big data relevance mining analysis method, frequent mode increase (FP-Growth)
The methods of method, post-class processing method, neural network.Apriori method is a kind of most influential Mining Boolean association rule
The then algorithm of frequent item set, but the combination that circulation generates when each step generates candidate Item Sets is excessive, and not excluding should not
Involved element.FP-Growth method is a kind of Mining Algorithms of Frequent Patterns increased based on mode, avoids a large amount of times
The generation of set of choices, but its time and space efficiency are not high enough.Post-class processing method can be effectively reduced index system
Scale, quickly compactly attribute is selected, obtain its importance degree sequence, but when post-class processing draw split hairs
When, over-fitting effect can be generated to noise data.
Summary of the invention
The purpose of the present invention is to provide a kind of distribution network transform measures and rate of qualified voltage index relevance method for digging
And device, this method and device are conducive to find the potential relevance between distribution network transform measure and rate of qualified voltage index,
And improve computational efficiency.
To implement above-mentioned purpose, the technical scheme is that a kind of distribution network transform measure and rate of qualified voltage index
Relevance method for digging, comprising: obtain active distribution network configurable resource modification measures and corresponding rate of qualified voltage index it
Between corresponding relationship sample data, it is right using sample data as training sample using the non-linear mapping capability of BP neural network
BP neural network is trained, and constructs the direct mapping between distribution network transform measure and rate of qualified voltage, is obtained based on BP mind
Distribution network voltage qualification rate assessment models under different modification measures through network.
Further, the acquisition active distribution network configurable resource modification measures and corresponding rate of qualified voltage index it
Between corresponding relationship sample data, wherein configurable resource modification measures be using can integrate distributed generation resource, energy storage,
The distribution network transform technical solution that flexible load resource is realized；Pass through the medium-term and long-term simulation side simulated based on sequential Monte Carlo
Method obtains being added the timing simulation of each active member in power distribution network after various modification measures as a result, and calculating and deducing each period
Under distribution network voltage qualification rate index, to obtain sample data.
Further, BP neural network is trained, is constructed straight between distribution network transform measure and rate of qualified voltage
Connect mapping, comprising the following steps:
(1) it obtains in power distribution network and runs timing simulation data for a long time, initial training determines the base of each parameter of BP neural network
This solution space；
(2) power supply power output and input vector as BP neural network of position, node load data in a distributed manner, voltage,
Power data determines BP neural network model and mode of learning as output vector；
(3) timing simulation data power distribution network early period are inputted, calculating BP neural network hidden layer is defeated with output layer each unit
Enter and export, is i.e. the information forward-propagating process of completion BP neural network study；
(4) correction error for calculating each neuron of output layer, completes error back propagation process；
(5) it is adjusted using weight of the genetic algorithm to control BP neural network performance with threshold value, optimizes different resource
The convergence rate of allocation plan and rate of qualified voltage correlation rule；
(6) mode of learning and study number are constantly updated；
(7) step (6) are repeated, constantly training BP neural network is until meeting cut-off condition；
(8) for trained BP neural network, node load, voltage, the power, distribution under different modification schemes are inputted
Formula power supply power output and telemechanical apparatus installation data, are calculated corresponding rate of qualified voltage index result.
Further, by the distributed generation resource power output and electric load and corresponding distribution network voltage under existing Net Frame of Electric Network
Path is transformed according to investment in sample data of the qualification rate as training BP neural network, and analog quantization distribution network structure structure is divided
Cloth plant-grid connection position, capacity of energy storing device configure all kinds of factor layer indexs, deduce all kinds of configurable resource sides of overall merit
Active distribution network technical-economic index under case obtains distribution network voltage qualification rate and various anticipation configuration sides by learning training
BP neural network between case constructs the direct mapping between distribution network transform measure and rate of qualified voltage；
Relationship between the input and output of BP neural network is as follows:
Wherein, xiFor the electric load and distributed generation resource power output, energy storage power output and node load under known grid structure
Data, hjFor hidden layer output, ykFor the distribution network voltage qualification rate of output, wijAnd θijRespectively power of the input layer to hidden layer
Value and threshold value, νjkAnd rjkRespectively weight and threshold value of the hidden layer to output layer；N indicates input data xiNumber, p indicate it is hidden
Number containing layer, m indicate output data ykNumber；f1[] indicates the input data x of settingiH is exported with hidden layerjBetween
Relation function, f2Indicate [] that the hidden layer of setting exports hjWith distribution network voltage qualification rate ykBetween relation function；
Error back propagation process is as follows:
Wherein, e is the distribution network voltage qualification rate of output layer output and the difference of actual numerical value, dkFor allowable range of error
The distribution network voltage qualification rate of interior output, ▽ w and ▽ b are respectively the adjustment amount of weight and threshold value, and η is learning rate.
Further, parameter optimization is carried out to BP neural network using genetic algorithm, obtains more accurate distribution network voltage
Qualification rate assessment models；The mathematical model for carrying out parameter optimization to BP neural network using genetic algorithm is as follows:
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Wherein, E be output distribution network voltage qualification rate and actual numerical value difference,It is defeated in allowable range of error
Distribution network voltage qualification rate out, ykFor the distribution network voltage qualification rate of neural network output layer output；
Quadratic nonlinearity optimization problem is solved using genetic algorithm, obtains each parameter value of BP neural network, and same
When meet output error minimum, process is as follows:
Step 1: obtaining sample data, initial training determines the basic solution space of each parameter of network；
Step 2: it is defined as follows fitness function:
Using its maximum value as the objective function in optimization process, then have:
maxF(w,v,θ,r)
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Step 3: basic solution space being encoded, the sequence for encoding generation contains the control of control hidden layer node number
Code part processed, and the weight coefficient code part of control network weight and threshold value；
Step 4: generating initial population, each individual is made of above-mentioned two parts coding in group；
Step 5: calculating the fitness value of each individual, and the highest individual of fitness in group is directly hereditary to next
Generation, other individuals are then selected using roulette wheel selection；
Step 6: using the means evolution current group for intersecting, making a variation, generating new progeny population；
Step 7: repeating step 5,6, the new group that constantly evolves is until meeting cut-off condition；
Step 8: the highest individual of fitness in final generation being decoded, corresponding node in hidden layer and network are obtained
Connection weight utilizes the generalization ability of test set Sample neural network.
The present invention also provides a kind of distribution network transform measures and rate of qualified voltage index relevance excavating gear, comprising:
Acquiring unit, for obtain active distribution network configurable resource modification measures and corresponding rate of qualified voltage index it
Between corresponding relationship sample data；And
Processing unit, it is right using sample data as training sample for the non-linear mapping capability using BP neural network
BP neural network is trained, and constructs the direct mapping between distribution network transform measure and rate of qualified voltage, is obtained based on BP mind
Distribution network voltage qualification rate assessment models under different modification measures through network.
Further, it is qualified with corresponding voltage to obtain active distribution network configurable resource modification measures for the acquiring unit
The sample data of corresponding relationship between rate index, wherein configurable resource modification measures are to utilize the distributed electrical that can be integrated
The distribution network transform technical solution that source, energy storage, flexible load resource are realized；It is medium-term and long-term by being simulated based on sequential Monte Carlo
Analogy method obtains being added the timing simulation of each active member in power distribution network after various modification measures as a result, and to calculate deduction every
Distribution network voltage qualification rate index under one period, to obtain sample data.
Further, the processing unit is trained BP neural network, and building distribution network transform measure and voltage close
Direct mapping between lattice rate, comprising the following steps:
(1) it obtains in power distribution network and runs timing simulation data for a long time, initial training determines the base of each parameter of BP neural network
This solution space；
(2) power supply power output and input vector as BP neural network of position, node load data in a distributed manner, voltage,
Power data determines BP neural network model and mode of learning as output vector；
(3) timing simulation data power distribution network early period are inputted, calculating BP neural network hidden layer is defeated with output layer each unit
Enter and export, is i.e. the information forward-propagating process of completion BP neural network study；
(4) correction error for calculating each neuron of output layer, completes error back propagation process；
(5) it is adjusted using weight of the genetic algorithm to control BP neural network performance with threshold value, optimizes different resource
The convergence rate of allocation plan and rate of qualified voltage correlation rule；
(6) mode of learning and study number are constantly updated；
(7) step (6) are repeated, constantly training BP neural network is until meeting cut-off condition；
(8) for trained BP neural network, node load, voltage, the power, distribution under different modification schemes are inputted
Formula power supply power output and telemechanical apparatus installation data, are calculated corresponding rate of qualified voltage index result.
Further, the processing unit by under existing Net Frame of Electric Network distributed generation resource power output and electric load with it is corresponding
Distribution network voltage qualification rate as training BP neural network sample data, according to investment be transformed path, analog quantization distribution
Net grid structure, distributed generation resource on-position, capacity of energy storing device configure all kinds of factor layer indexs, and it is all kinds of to deduce overall merit
Active distribution network technical-economic index under configurable resource scheme obtains distribution network voltage qualification rate and each by learning training
BP neural network between kind anticipation allocation plan, constructs directly reflecting between distribution network transform measure and rate of qualified voltage
It penetrates；
Relationship between the input and output of BP neural network is as follows:
Wherein, xiFor the electric load and distributed generation resource power output, energy storage power output and node load under known grid structure
Data, hjFor hidden layer output, ykFor the distribution network voltage qualification rate of output, wijAnd θijRespectively power of the input layer to hidden layer
Value and threshold value, νjkAnd rjkRespectively weight and threshold value of the hidden layer to output layer；N indicates input data xiNumber, p indicate it is hidden
Number containing layer, m indicate output data ykNumber；f1[] indicates the input data x of settingiH is exported with hidden layerjBetween
Relation function, f2Indicate [] that the hidden layer of setting exports hjWith distribution network voltage qualification rate ykBetween relation function；
Error back propagation process is as follows:
Wherein, e is the distribution network voltage qualification rate of output layer output and the difference of actual numerical value, dkFor allowable range of error
The distribution network voltage qualification rate of interior output, ▽ w and ▽ b are respectively the adjustment amount of weight and threshold value, and η is learning rate.
Further, the processing unit is also used to carry out parameter optimization to BP neural network using genetic algorithm, obtains
More accurate distribution network voltage qualification rate assessment models；The mathematical modulo of parameter optimization is carried out to BP neural network using genetic algorithm
Type is as follows:
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Wherein, E be output distribution network voltage qualification rate and actual numerical value difference,It is defeated in allowable range of error
Distribution network voltage qualification rate out, ykFor the distribution network voltage qualification rate of neural network output layer output；
Quadratic nonlinearity optimization problem is solved using genetic algorithm, obtains each parameter value of BP neural network, and same
When meet output error minimum, process is as follows:
Step 1: obtaining sample data, initial training determines the basic solution space of each parameter of network；
Step 2: it is defined as follows fitness function:
Using its maximum value as the objective function in optimization process, then have:
maxF(w,v,θ,r)
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Step 3: basic solution space being encoded, the sequence for encoding generation contains the control of control hidden layer node number
Code part processed, and the weight coefficient code part of control network weight and threshold value；
Step 4: generating initial population, each individual is made of above-mentioned two parts coding in group；
Step 5: calculating the fitness value of each individual, and the highest individual of fitness in group is directly hereditary to next
Generation, other individuals are then selected using roulette wheel selection；
Step 6: using the means evolution current group for intersecting, making a variation, generating new progeny population；
Step 7: repeating step 5,6, the new group that constantly evolves is until meeting cut-off condition；
Step 8: the highest individual of fitness in final generation being decoded, corresponding node in hidden layer and network are obtained
Connection weight utilizes the generalization ability of test set Sample neural network.
Compared to the prior art, the beneficial effects of the present invention are: this method and device are arranged in discovery active distribution network transformation
Grant potential rule between rate of qualified voltage index and improve computational efficiency etc. has greater advantage, not only can be to avoid multiple
Miscellaneous Load flow calculation process, and can effectively promote computational efficiency.Instruction is constituted with rate of qualified voltage index and modification measures
Practice sample set and corresponding association relation model can be obtained by the off-line learning to sample data.In practical applications, when giving
When determining resource distribution index, neural network model can quickly provide relevant voltage qualification rate index as a result, as subsequent distribution
The constraint condition of net investment decision model.In addition, by BP neural network threshold value and right-value optimization based on genetic algorithm, BP mind
Convergence through network has also obtained larger promotion.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of BP neural network in the method for the embodiment of the present invention.
Fig. 2 is the distribution network voltage qualification rate assessment models signal in the method for the embodiment of the present invention based on BP neural network
Figure.
Fig. 3 is the training flow diagram of BP neural network in the method for the embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the invention will be further described.
The present invention provides a kind of distribution network transform measure and rate of qualified voltage index relevance method for digging, comprising: simulation
Quantify all kinds of factor layer indexs, obtains between active distribution network configurable resource modification measures and corresponding rate of qualified voltage index
The sample data of corresponding relationship, such as a large amount of distributed generation resource power output, Power system load data and corresponding distribution network voltage close
The data such as lattice rate index.Wherein, configurable resource modification measures are to utilize the distributed generation resource that can be integrated, energy storage, flexibility negative
The resources such as lotus realize distribution network transform technical solution, such as blower will be created in power distribution network, photovoltaic distributed power supply is set
The modification measures alternately such as standby, energy storage device, switchgear and two distant, three telecontrol equipments such as distant.By being based on sequential illiteracy
The medium-term and long-term analogy method of special Carlow simulation obtains being added the timing simulation of each active member in power distribution network after various modification measures
As a result, the fluctuation situation of the power output of such as distributed generation resource, the charge and discharge situation of energy storage and flexible load, and it is each to calculate deduction
Distribution network voltage qualification rate index under period, to obtain a large amount of sample data.
Then, as shown in Fig. 2, the non-linear mapping capability using BP neural network constructs distribution network transform measure and voltage
Direct mapping between qualification rate is trained BP neural network using sample data as training sample, obtains based on BP mind
Distribution network voltage qualification rate assessment models under different modification measures through network.
As shown in Fig. 2, the model can quickly estimate the modification measures when distribution network transform measure scene changes
Under rate of qualified voltage value, therefore, it is determined that influence degree of the different distribution network transform measure to distribution network voltage qualification rate index,
As the relevance constraint condition of later period power distribution network investment decision, the time loss of time-domain-simulation is saved.BP is based on using this
The distribution network voltage qualification rate assessment models of neural network can greatly promote the solving speed of model, thus quickly and efficiently
Formulate power distribution network investment tactics.
To improve the speed and effect that BP neural network finds related law between distribution network transform measure and rate of qualified voltage
Rate carries out parameter optimization to BP neural network using genetic algorithm, and initial weight is determining when solving building model and BP algorithm is received
The problems such as slow is held back, more accurate distribution network voltage qualification rate assessment models are obtained.
Fig. 3 is the training process of BP neural network in the embodiment of the present invention.As shown in figure 3, being instructed to BP neural network
Practice, construct the direct mapping between distribution network transform measure and rate of qualified voltage, to excavate different distribution network transforms using it
Relevance between measure and rate of qualified voltage, comprising the following steps:
(1) it obtains in power distribution network and runs timing simulation data for a long time, initial training determines the base of each parameter of BP neural network
This solution space；
(2) power supply power output and input vector as BP neural network of position, node load data in a distributed manner, voltage,
Power data determines BP neural network model and mode of learning as output vector；
(3) timing simulation data power distribution network early period are inputted, calculating BP neural network hidden layer is defeated with output layer each unit
Enter and export, is i.e. the information forward-propagating process of completion BP neural network study；
(4) correction error for calculating each neuron of output layer, completes error back propagation process；
(5) it is adjusted using weight of the genetic algorithm to control BP neural network performance with threshold value, optimizes different resource
The convergence rate of allocation plan and rate of qualified voltage correlation rule；
(6) mode of learning and study number are constantly updated；
(7) step (6) are repeated, constantly training BP neural network is until meeting cut-off condition；
(8) for trained BP neural network, node load, voltage, the power, distribution under different modification schemes are inputted
Formula power supply power output and telemechanical apparatus installation data, are calculated corresponding rate of qualified voltage index result.
In the present invention, by the distributed generation resource power output and electric load and corresponding distribution network voltage under existing Net Frame of Electric Network
Path is transformed according to investment in sample data of the qualification rate as training BP neural network, and analog quantization distribution network structure structure is divided
All kinds of factor layer indexs such as cloth plant-grid connection position, capacity of energy storing device configuration, the angle learnt from statistical analysis and data
The active distribution network technical-economic index under all kinds of configurable resource schemes of overall merit is deduced, distribution is obtained by learning training
BP neural network between net rate of qualified voltage and various anticipation allocation plans, constructs distribution network transform measure and voltage is qualified
Direct mapping between rate.
BP neural network is a kind of multilayer feedforward neural network based on error backpropagation algorithm, is had good non-thread
Property mapping ability, can learn with adaptive unknown message, structure is as shown in Figure 1.
As shown in Figure 1, the relationship between the input and output of BP neural network is as follows:
Wherein, xiFor the electric load and distributed generation resource power output, energy storage power output and node load under known grid structure
Data, hjFor hidden layer output, ykFor the distribution network voltage qualification rate of output, wijAnd θijRespectively power of the input layer to hidden layer
Value and threshold value, νjkAnd rjkRespectively weight and threshold value of the hidden layer to output layer；N indicates input data xiNumber, p indicate it is hidden
Number containing layer, m indicate output data ykNumber；f1[] indicates the input data x of settingiH is exported with hidden layerjBetween
Relation function, f2Indicate [] that the hidden layer of setting exports hjWith distribution network voltage qualification rate ykBetween relation function；
Error back propagation process is as follows:
Wherein, e is the distribution network voltage qualification rate of output layer output and the difference of actual numerical value, dkFor allowable range of error
The distribution network voltage qualification rate of interior output, ▽ w and ▽ b are respectively the adjustment amount of weight and threshold value, and η is learning rate, are to influence
One of an important factor for algorithm the convergence speed.
In the present invention, parameter optimization is carried out to BP neural network using genetic algorithm, to promote the study of neural network
Efficiency improves estimated accuracy, obtains more accurate distribution network voltage qualification rate assessment models.Genetic algorithm is a kind of simulation biology
The computation model of heredity selection and species survival of the fittest evolutionary process, is mainly characterized by collective search strategy and individual in population
Between information exchange, search has stronger problem resolving ability and extensive adaptability independent of gradient information.
The mathematical model for carrying out parameter optimization to BP neural network using genetic algorithm is as follows:
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Wherein, E be output distribution network voltage qualification rate and actual numerical value difference,It is defeated in allowable range of error
Distribution network voltage qualification rate out, ykFor the distribution network voltage qualification rate of neural network output layer output；Utilize genetic algorithm pair
Quadratic nonlinearity optimization problem is solved, and each parameter value of BP neural network is obtained, and meets output error minimum simultaneously.
Above-mentioned quadratic nonlinearity optimization problem is solved using genetic algorithm, each parameter value of neural network can be obtained,
Meet output error minimum simultaneously.Its detailed process is as follows:
Step 1: obtaining sample data, initial training determines the basic solution space of each parameter of network；
Step 2: it is defined as follows fitness function:
Using its maximum value as the objective function in optimization process, then have:
maxF(w,v,θ,r)
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Step 3: basic solution space being encoded, the sequence for encoding generation contains the control of control hidden layer node number
Code part processed, and the weight coefficient code part of control network weight and threshold value.
Step 4: generating initial population, each individual is made of above-mentioned two parts coding in group.
Step 5: calculating the fitness value of each individual, and the highest individual of fitness in group is directly hereditary to next
Generation, other individuals are then selected using roulette wheel selection.
Step 6: using the means evolution current group for intersecting, making a variation, generating new progeny population.
Step 7: repeating step 5,6, the new group that constantly evolves is until meeting cut-off condition (maximum evolutionary generation).
Step 8: the highest individual of fitness in final generation being decoded, corresponding node in hidden layer and network are obtained
Connection weight utilizes the generalization ability of test set Sample neural network.
The present invention also provides a kind of distribution network transform measures for realizing the above method and rate of qualified voltage index to close
Connection property excavating gear, comprising:
Acquiring unit, for obtain active distribution network configurable resource modification measures and corresponding rate of qualified voltage index it
Between corresponding relationship sample data；And
Processing unit, it is right using sample data as training sample for the non-linear mapping capability using BP neural network
BP neural network is trained, and constructs the direct mapping between distribution network transform measure and rate of qualified voltage, is obtained based on BP mind
Distribution network voltage qualification rate assessment models under different modification measures through network.
The acquiring unit obtain active distribution network configurable resource modification measures and corresponding rate of qualified voltage index it
Between corresponding relationship sample data, wherein configurable resource modification measures be using can integrate distributed generation resource, energy storage,
The distribution network transform technical solution that flexible load resource is realized；Pass through the medium-term and long-term simulation side simulated based on sequential Monte Carlo
Method obtains being added the timing simulation of each active member in power distribution network after various modification measures as a result, and calculating and deducing each period
Under distribution network voltage qualification rate index, to obtain sample data.
The processing unit is trained BP neural network, constructs between distribution network transform measure and rate of qualified voltage
Directly map, comprising the following steps:
(1) it obtains in power distribution network and runs timing simulation data for a long time, initial training determines the base of each parameter of BP neural network
This solution space；
(2) power supply power output and input vector as BP neural network of position, node load data in a distributed manner, voltage,
Power data determines BP neural network model and mode of learning as output vector；
(3) timing simulation data power distribution network early period are inputted, calculating BP neural network hidden layer is defeated with output layer each unit
Enter and export, is i.e. the information forward-propagating process of completion BP neural network study；
(4) correction error for calculating each neuron of output layer, completes error back propagation process；
(5) it is adjusted using weight of the genetic algorithm to control BP neural network performance with threshold value, optimizes different resource
The convergence rate of allocation plan and rate of qualified voltage correlation rule；
(6) mode of learning and study number are constantly updated；
(7) step (6) are repeated, constantly training BP neural network is until meeting cut-off condition；
(8) for trained BP neural network, node load, voltage, the power, distribution under different modification schemes are inputted
Formula power supply power output and telemechanical apparatus installation data, are calculated corresponding rate of qualified voltage index result.
The processing unit is by the distributed generation resource power output and electric load and corresponding power distribution network under existing Net Frame of Electric Network
Path, analog quantization distribution network structure knot is transformed according to investment in sample data of the rate of qualified voltage as training BP neural network
Structure, distributed generation resource on-position, capacity of energy storing device configure all kinds of factor layer indexs, deduce all kinds of configurable moneys of overall merit
Active distribution network technical-economic index under the scheme of source obtains distribution network voltage qualification rate by learning training and matches with various anticipations
The BP neural network between scheme is set, the direct mapping between distribution network transform measure and rate of qualified voltage is constructed；
Relationship between the input and output of BP neural network is as follows:
Wherein, xiFor the electric load and distributed generation resource power output, energy storage power output and node load under known grid structure
Data, hjFor hidden layer output, ykFor the distribution network voltage qualification rate of output, wijAnd θijRespectively power of the input layer to hidden layer
Value and threshold value, νjkAnd rjkRespectively weight and threshold value of the hidden layer to output layer；N indicates input data xiNumber, p indicate it is hidden
Number containing layer, m indicate output data ykNumber；f1[] indicates the input data x of settingiH is exported with hidden layerjBetween
Relation function, f2Indicate [] that the hidden layer of setting exports hjWith distribution network voltage qualification rate ykBetween relation function；
Error back propagation process is as follows:
Wherein, e is the distribution network voltage qualification rate of output layer output and the difference of actual numerical value, dkFor allowable range of error
The distribution network voltage qualification rate of interior output, ▽ w and ▽ b are respectively the adjustment amount of weight and threshold value, and η is learning rate.
The processing unit is also used to carry out parameter optimization to BP neural network using genetic algorithm, is more accurately matched
Network voltage qualification rate assessment models；The mathematical model for carrying out parameter optimization to BP neural network using genetic algorithm is as follows:
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Wherein, E be output distribution network voltage qualification rate and actual numerical value difference,It is defeated in allowable range of error
Distribution network voltage qualification rate out, ykFor the distribution network voltage qualification rate of neural network output layer output；
Quadratic nonlinearity optimization problem is solved using genetic algorithm, obtains each parameter value of BP neural network, and same
When meet output error minimum, process is as follows:
Step 1: obtaining sample data, initial training determines the basic solution space of each parameter of network；
Step 2: it is defined as follows fitness function:
Using its maximum value as the objective function in optimization process, then have:
maxF(w,v,θ,r)
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Step 3: basic solution space being encoded, the sequence for encoding generation contains the control of control hidden layer node number
Code part processed, and the weight coefficient code part of control network weight and threshold value；
Step 4: generating initial population, each individual is made of above-mentioned two parts coding in group；
Step 5: calculating the fitness value of each individual, and the highest individual of fitness in group is directly hereditary to next
Generation, other individuals are then selected using roulette wheel selection；
Step 6: using the means evolution current group for intersecting, making a variation, generating new progeny population；
Step 7: repeating step 5,6, the new group that constantly evolves is until meeting cut-off condition；
Step 8: the highest individual of fitness in final generation being decoded, corresponding node in hidden layer and network are obtained
Connection weight utilizes the generalization ability of test set Sample neural network.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The above are preferred embodiments of the present invention, all any changes made according to the technical solution of the present invention, and generated function is made
When with range without departing from technical solution of the present invention, all belong to the scope of protection of the present invention.
Claims (10)
1. a kind of distribution network transform measure and rate of qualified voltage index relevance method for digging characterized by comprising obtain master
The sample data of corresponding relationship, utilizes BP between dynamic power distribution network configurable resource modification measures and corresponding rate of qualified voltage index
The non-linear mapping capability of neural network is trained BP neural network using sample data as training sample, constructs distribution
Direct mapping between net modification measures and rate of qualified voltage obtains the distribution under the different modification measures based on BP neural network
Net rate of qualified voltage assessment models.
2. distribution network transform measure according to claim 1 and rate of qualified voltage index relevance method for digging, feature
It is, it is described to obtain corresponding relationship between active distribution network configurable resource modification measures and corresponding rate of qualified voltage index
Sample data, wherein configurable resource modification measures are to utilize the distributed generation resource that can be integrated, energy storage, flexible load resource
The distribution network transform technical solution of realization；By the medium-term and long-term analogy method simulated based on sequential Monte Carlo, obtain being added each
After kind of modification measures in power distribution network the timing simulation of each active member as a result, and calculating the distribution network voltage deduced under each period
Qualification rate index, to obtain sample data.
3. distribution network transform measure according to claim 1 and rate of qualified voltage index relevance method for digging, feature
It is, BP neural network is trained, the direct mapping between building distribution network transform measure and rate of qualified voltage, including with
Lower step:
(1) it obtains in power distribution network and runs timing simulation data for a long time, initial training determines the elementary solution of each parameter of BP neural network
Space；
(2) power supply power output and input vector as BP neural network of position, node load data in a distributed manner, voltage, power
Data determine BP neural network model and mode of learning as output vector；
(3) timing simulation data power distribution network early period are inputted, calculate the input of BP neural network hidden layer and output layer each unit with
Output, i.e. the information forward-propagating process of completion BP neural network study；
(4) correction error for calculating each neuron of output layer, completes error back propagation process；
(5) it is adjusted using weight of the genetic algorithm to control BP neural network performance with threshold value, optimization different resource configuration
The convergence rate of scheme and rate of qualified voltage correlation rule；
(6) mode of learning and study number are constantly updated；
(7) step (6) are repeated, constantly training BP neural network is until meeting cut-off condition；
(8) for trained BP neural network, node load, voltage, power, the distributed electrical under different modification schemes are inputted
Source power output and telemechanical apparatus installation data, are calculated corresponding rate of qualified voltage index result.
4. distribution network transform measure according to claim 1 and rate of qualified voltage index relevance method for digging, feature
It is, the distributed generation resource under existing Net Frame of Electric Network is contributed and electric load is with corresponding distribution network voltage qualification rate as instruction
Path, analog quantization distribution network structure structure, distributed generation resource access is transformed according to investment in the sample data for practicing BP neural network
Position, capacity of energy storing device configure all kinds of factor layer indexs, and the active deduced under all kinds of configurable resource schemes of overall merit is matched
Electric power network technique economic indicator obtains the BP mind between distribution network voltage qualification rate and various anticipation allocation plans by learning training
Through network, the direct mapping between distribution network transform measure and rate of qualified voltage is constructed；
Relationship between the input and output of BP neural network is as follows:
Wherein, xiFor the electric load and distributed generation resource power output, energy storage power output and node load data under known grid structure,
hjFor hidden layer output, ykFor the distribution network voltage qualification rate of output, wijAnd θijRespectively input layer to hidden layer weight and
Threshold value, νjkAnd rjkRespectively weight and threshold value of the hidden layer to output layer；N indicates input data xiNumber, p indicate hidden layer
Number, m indicate output data ykNumber；f1[] indicates the input data x of settingiH is exported with hidden layerjBetween relationship
Function, f2Indicate [] that the hidden layer of setting exports hjWith distribution network voltage qualification rate ykBetween relation function；
Error back propagation process is as follows:
Wherein, e is the distribution network voltage qualification rate of output layer output and the difference of actual numerical value, dkFor in allowable range of error
The distribution network voltage qualification rate of output, ▽ w and ▽ b are respectively the adjustment amount of weight and threshold value, and η is learning rate.
5. distribution network transform measure according to claim 4 and rate of qualified voltage index relevance method for digging, feature
It is, parameter optimization is carried out to BP neural network using genetic algorithm, obtains more accurate distribution network voltage qualification rate assessment mould
Type；The mathematical model for carrying out parameter optimization to BP neural network using genetic algorithm is as follows:
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Wherein, E be output distribution network voltage qualification rate and actual numerical value difference,For the output in allowable range of error
Distribution network voltage qualification rate, ykFor the distribution network voltage qualification rate of neural network output layer output；
Quadratic nonlinearity optimization problem is solved using genetic algorithm, obtains each parameter value of BP neural network, and is full simultaneously
Sufficient output error is minimum, and process is as follows:
Step 1: obtaining sample data, initial training determines the basic solution space of each parameter of network；
Step 2: it is defined as follows fitness function:
Using its maximum value as the objective function in optimization process, then have:
maxF(w,v,θ,r)
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Step 3: basic solution space being encoded, the sequence for encoding generation contains the control code of control hidden layer node number
Part, and the weight coefficient code part of control network weight and threshold value；
Step 4: generating initial population, each individual is made of above-mentioned two parts coding in group；
Step 5: the fitness value of each individual is calculated, and the highest individual of fitness in group is directly hereditary to the next generation,
Other individuals are then selected using roulette wheel selection；
Step 6: using the means evolution current group for intersecting, making a variation, generating new progeny population；
Step 7: repeating step 5,6, the new group that constantly evolves is until meeting cut-off condition；
Step 8: the highest individual of fitness in final generation being decoded, corresponding node in hidden layer and network connection are obtained
Weight utilizes the generalization ability of test set Sample neural network.
6. a kind of distribution network transform measure and rate of qualified voltage index relevance excavating gear characterized by comprising
Acquiring unit is right between active distribution network configurable resource modification measures and corresponding rate of qualified voltage index for obtaining
The sample data that should be related to；And
Processing unit, for the non-linear mapping capability using BP neural network, using sample data as training sample, to BP mind
It is trained through network, constructs the direct mapping between distribution network transform measure and rate of qualified voltage, obtained based on BP nerve net
Distribution network voltage qualification rate assessment models under the different modification measures of network.
7. distribution network transform measure according to claim 6 and rate of qualified voltage index relevance excavating gear, feature
It is, the acquiring unit obtains right between active distribution network configurable resource modification measures and corresponding rate of qualified voltage index
The sample data that should be related to, wherein configurable resource modification measures are to utilize the distributed generation resource that can be integrated, energy storage, flexibility
The distribution network transform technical solution that burdened resource is realized；By the medium-term and long-term analogy method simulated based on sequential Monte Carlo, obtain
To the timing simulation that each active member in power distribution network after various modification measures is added as a result, and calculating and deducing matching under each period
Network voltage qualification rate index, to obtain sample data.
8. distribution network transform measure according to claim 6 and rate of qualified voltage index relevance excavating gear, feature
It is, the processing unit is trained BP neural network, constructs straight between distribution network transform measure and rate of qualified voltage
Connect mapping, comprising the following steps:
(1) it obtains in power distribution network and runs timing simulation data for a long time, initial training determines the elementary solution of each parameter of BP neural network
Space；
(2) power supply power output and input vector as BP neural network of position, node load data in a distributed manner, voltage, power
Data determine BP neural network model and mode of learning as output vector；
(3) timing simulation data power distribution network early period are inputted, calculate the input of BP neural network hidden layer and output layer each unit with
Output, i.e. the information forward-propagating process of completion BP neural network study；
(4) correction error for calculating each neuron of output layer, completes error back propagation process；
(5) it is adjusted using weight of the genetic algorithm to control BP neural network performance with threshold value, optimization different resource configuration
The convergence rate of scheme and rate of qualified voltage correlation rule；
(6) mode of learning and study number are constantly updated；
(7) step (6) are repeated, constantly training BP neural network is until meeting cut-off condition；
(8) for trained BP neural network, node load, voltage, power, the distributed electrical under different modification schemes are inputted
Source power output and telemechanical apparatus installation data, are calculated corresponding rate of qualified voltage index result.
9. distribution network transform measure according to claim 6 and rate of qualified voltage index relevance excavating gear, feature
It is, the processing unit is by the distributed generation resource power output and electric load and corresponding distribution network voltage under existing Net Frame of Electric Network
Path is transformed according to investment in sample data of the qualification rate as training BP neural network, and analog quantization distribution network structure structure is divided
Cloth plant-grid connection position, capacity of energy storing device configure all kinds of factor layer indexs, deduce all kinds of configurable resource sides of overall merit
Active distribution network technical-economic index under case obtains distribution network voltage qualification rate and various anticipation configuration sides by learning training
BP neural network between case constructs the direct mapping between distribution network transform measure and rate of qualified voltage；
Relationship between the input and output of BP neural network is as follows:
Wherein, xiFor the electric load and distributed generation resource power output, energy storage power output and node load data under known grid structure,
hjFor hidden layer output, ykFor the distribution network voltage qualification rate of output, wijAnd θijRespectively input layer to hidden layer weight and
Threshold value, νjkAnd rjkRespectively weight and threshold value of the hidden layer to output layer；N indicates input data xiNumber, p indicate hidden layer
Number, m indicate output data ykNumber；f1[] indicates the input data x of settingiH is exported with hidden layerjBetween relationship
Function, f2Indicate [] that the hidden layer of setting exports hjWith distribution network voltage qualification rate ykBetween relation function；
Error back propagation process is as follows:
Wherein, e is the distribution network voltage qualification rate of output layer output and the difference of actual numerical value, dkFor in allowable range of error
The distribution network voltage qualification rate of output, ▽ w and ▽ b are respectively the adjustment amount of weight and threshold value, and η is learning rate.
10. distribution network transform measure according to claim 9 and rate of qualified voltage index relevance excavating gear, feature
It is, the processing unit is also used to carry out parameter optimization to BP neural network using genetic algorithm, obtains more accurate distribution
Net rate of qualified voltage assessment models；The mathematical model for carrying out parameter optimization to BP neural network using genetic algorithm is as follows:
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Wherein, E be output distribution network voltage qualification rate and actual numerical value difference,For the output in allowable range of error
Distribution network voltage qualification rate, ykFor the distribution network voltage qualification rate of neural network output layer output；
Quadratic nonlinearity optimization problem is solved using genetic algorithm, obtains each parameter value of BP neural network, and is full simultaneously
Sufficient output error is minimum, and process is as follows:
Step 1: obtaining sample data, initial training determines the basic solution space of each parameter of network；
Step 2: it is defined as follows fitness function:
Using its maximum value as the objective function in optimization process, then have:
maxF(w,v,θ,r)
s.t.w∈Rn×p,v∈Rp×m,θ∈Rn×p,r∈Rp×m
Step 3: basic solution space being encoded, the sequence for encoding generation contains the control code of control hidden layer node number
Part, and the weight coefficient code part of control network weight and threshold value；
Step 4: generating initial population, each individual is made of above-mentioned two parts coding in group；
Step 5: the fitness value of each individual is calculated, and the highest individual of fitness in group is directly hereditary to the next generation,
Other individuals are then selected using roulette wheel selection；
Step 6: using the means evolution current group for intersecting, making a variation, generating new progeny population；
Step 7: repeating step 5,6, the new group that constantly evolves is until meeting cut-off condition；
Step 8: the highest individual of fitness in final generation being decoded, corresponding node in hidden layer and network connection are obtained
Weight utilizes the generalization ability of test set Sample neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910077166.6A CN109886560A (en) | 2019-01-25 | 2019-01-25 | Distribution network transform measure and rate of qualified voltage index relevance method for digging and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910077166.6A CN109886560A (en) | 2019-01-25 | 2019-01-25 | Distribution network transform measure and rate of qualified voltage index relevance method for digging and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109886560A true CN109886560A (en) | 2019-06-14 |
Family
ID=66927103
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910077166.6A Pending CN109886560A (en) | 2019-01-25 | 2019-01-25 | Distribution network transform measure and rate of qualified voltage index relevance method for digging and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109886560A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009181A (en) * | 2019-01-25 | 2019-07-12 | 国网福建省电力有限公司 | Distribution network transform measure and mistake load figureofmerit relevance method for digging and device |
| CN110705873A (en) * | 2019-09-30 | 2020-01-17 | 国网福建省电力有限公司 | Novel power distribution network operation state portrait analysis method |
| CN111105100A (en) * | 2020-01-10 | 2020-05-05 | 昆明理工大学 | Neural network-based optimization method and system for multi-microgrid scheduling mechanism |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110054815A1 (en) * | 2007-04-20 | 2011-03-03 | Jianhong Xu | Professional diagnosis method of battery performance analysis |
| CN106991531A (en) * | 2017-03-30 | 2017-07-28 | 国家电网公司 | Power distribution network project construction effect Post-assessment Method based on BP neural network |
| CN108710975A (en) * | 2018-05-18 | 2018-10-26 | 国网江苏省电力有限公司徐州供电分公司 | A kind of electric automobile load prediction technique |
| CN110009181A (en) * | 2019-01-25 | 2019-07-12 | 国网福建省电力有限公司 | Distribution network transform measure and mistake load figureofmerit relevance method for digging and device |
-
2019
- 2019-01-25 CN CN201910077166.6A patent/CN109886560A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110054815A1 (en) * | 2007-04-20 | 2011-03-03 | Jianhong Xu | Professional diagnosis method of battery performance analysis |
| CN106991531A (en) * | 2017-03-30 | 2017-07-28 | 国家电网公司 | Power distribution network project construction effect Post-assessment Method based on BP neural network |
| CN108710975A (en) * | 2018-05-18 | 2018-10-26 | 国网江苏省电力有限公司徐州供电分公司 | A kind of electric automobile load prediction technique |
| CN110009181A (en) * | 2019-01-25 | 2019-07-12 | 国网福建省电力有限公司 | Distribution network transform measure and mistake load figureofmerit relevance method for digging and device |
Non-Patent Citations (2)
| Title |
|---|
| BIN CHEN ET AL.: "An Investment Decision Model of Distribution Network Planning Based on Correlation Mining of Reconstruction Measures and Performance Indexes", 《2018 CHINA INTERNATIONAL CONFERENCE ON ELECTRICITY DISTRIBUTION》 * |
| BIN CHEN ET AL.: "Correlation Mining for Reconstruction Measures and Performance Indexes of Distribution Network Planning Based on BP Neural Network", 《2018 INTERNATIONAL CONFERENCE ON POWER SYSTEM TECHNOLOGY》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009181A (en) * | 2019-01-25 | 2019-07-12 | 国网福建省电力有限公司 | Distribution network transform measure and mistake load figureofmerit relevance method for digging and device |
| CN110705873A (en) * | 2019-09-30 | 2020-01-17 | 国网福建省电力有限公司 | Novel power distribution network operation state portrait analysis method |
| CN110705873B (en) * | 2019-09-30 | 2022-06-03 | 国网福建省电力有限公司 | Power distribution network running state portrait analysis method |
| CN111105100A (en) * | 2020-01-10 | 2020-05-05 | 昆明理工大学 | Neural network-based optimization method and system for multi-microgrid scheduling mechanism |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110009181A (en) | Distribution network transform measure and mistake load figureofmerit relevance method for digging and device | |
| CN108985515B (en) | New energy output prediction method and system based on independent cyclic neural network | |
| CN105260786B (en) | A kind of simulation credibility of electric propulsion system assessment models comprehensive optimization method | |
| CN111027732B (en) | Method and system for generating multi-wind power plant output scene | |
| CN109858798B (en) | Power grid investment decision modeling method and device for correlating transformation measures with voltage indexes | |
| CN103489038A (en) | Photovoltaic ultra-short-term power prediction method based on LM-BP neural network | |
| CN105138717A (en) | Transformer state evaluation method by optimizing neural network with dynamic mutation particle swarm | |
| CN107133695A (en) | A kind of wind power forecasting method and system | |
| CN109886560A (en) | Distribution network transform measure and rate of qualified voltage index relevance method for digging and device | |
| CN104636985A (en) | Method for predicting radio disturbance of electric transmission line by using improved BP (back propagation) neural network | |
| CN110837915B (en) | Low-voltage load point prediction and probability prediction method for power system based on hybrid integrated deep learning | |
| CN105389623A (en) | Thermal process model parameter identification method adopting improved ant colony algorithm | |
| CN109978253B (en) | Electric power system short-term load prediction method based on incremental learning | |
| CN112163715A (en) | Training method and device of generative countermeasure network and power load prediction method | |
| Li et al. | An improved multiobjective estimation of distribution algorithm for environmental economic dispatch of hydrothermal power systems | |
| CN116169698A (en) | Distributed energy storage optimal configuration method and system for stable new energy consumption | |
| Zeng et al. | Short-term load forecasting of smart grid systems by combination of general regression neural network and least squares-support vector machine algorithm optimized by harmony search algorithm method | |
| CN114139377A (en) | Power system modeling method and system, readable storage medium and server | |
| CN110674460B (en) | E-Seq2Seq technology-based data driving type unit combination intelligent decision method | |
| CN116843083A (en) | Carbon emission prediction system and method based on hybrid neural network model | |
| CN106503793A (en) | A kind of neural network short-term wind speed forecasting method based on improvement difference algorithm | |
| Sulaiman et al. | Real and reactive power flow allocation in deregulated power system utilizing genetic-support vector machine technique | |
| CN113627655B (en) | Method and device for simulating and predicting pre-disaster fault scene of power distribution network | |
| Ferreira et al. | Evolutionary strategy for simultaneous optimization of parameters, topology and reservoir weights in echo state networks | |
| CN113158134B (en) | Method, device and storage medium for constructing non-invasive load identification model |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190614 |
|
| RJ01 | Rejection of invention patent application after publication |