CN115912331A - Power adjusting method of intelligent wind-solar hybrid power generation system - Google Patents
Power adjusting method of intelligent wind-solar hybrid power generation system Download PDFInfo
- Publication number
- CN115912331A CN115912331A CN202211352772.2A CN202211352772A CN115912331A CN 115912331 A CN115912331 A CN 115912331A CN 202211352772 A CN202211352772 A CN 202211352772A CN 115912331 A CN115912331 A CN 115912331A
- Authority
- CN
- China
- Prior art keywords
- wind
- group
- model
- follows
- formula
- 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
- 238000010248 power generation Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 27
- 239000010926 waste battery Substances 0.000 claims abstract description 23
- 230000001276 controlling effect Effects 0.000 claims abstract description 4
- 238000007599 discharging Methods 0.000 claims abstract description 4
- 230000001105 regulatory effect Effects 0.000 claims abstract description 4
- 210000002569 neuron Anatomy 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 15
- 210000004027 cell Anatomy 0.000 claims description 8
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 6
- 230000007246 mechanism Effects 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims description 3
- 238000009304 pastoral farming Methods 0.000 claims description 3
- 230000005855 radiation Effects 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 abstract description 3
- 230000000295 complement effect Effects 0.000 abstract description 2
- 230000008901 benefit Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000010845 search algorithm Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 2
- 241001600609 Equus ferus Species 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- ZXQYGBMAQZUVMI-GCMPRSNUSA-N gamma-cyhalothrin Chemical compound CC1(C)[C@@H](\C=C(/Cl)C(F)(F)F)[C@H]1C(=O)O[C@H](C#N)C1=CC=CC(OC=2C=CC=CC=2)=C1 ZXQYGBMAQZUVMI-GCMPRSNUSA-N 0.000 description 1
- 229910001385 heavy metal Inorganic materials 0.000 description 1
- QSHDDOUJBYECFT-UHFFFAOYSA-N mercury Chemical compound [Hg] QSHDDOUJBYECFT-UHFFFAOYSA-N 0.000 description 1
- 229910052753 mercury Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Landscapes
- Photovoltaic Devices (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention relates to the technical field of wind and light prediction and power regulation, and discloses a power regulation method of an intelligent wind and light complementary power generation system, which comprises the following steps: acquiring historical power generation data, and establishing an RBF neural network prediction model aiming at short-term wind-solar prediction; establishing a model of distributed energy and a waste battery charging and discharging model; optimizing the RBF neural network prediction model by utilizing a wild Ma Sousuo algorithm; carrying out short-term wind and light prediction according to the optimized model to obtain a wind and light prediction result; and intelligently regulating and controlling the output power of the distributed energy system according to the wind and light prediction result. Compared with the prior art, the method uses the wild Ma Sousuo algorithm to optimize the RBF neural network prediction model, so that the model error is smaller, wind and light clean energy and waste batteries are fully utilized, and the economic cost and the environmental cost are greatly saved.
Description
Technical Field
The invention relates to the technical field of wind and light prediction and power regulation, in particular to a power regulation method of an intelligent wind and light complementary power generation system.
Background
Under the double pressure of energy shortage and environmental pollution, the power generation technology based on renewable energy is rapidly developed, and photovoltaic power and wind power are widely applied with the advantages of safety, cleanness, abundance and the like. As representative of clean energy sources, they have incomparable advantages with other energy sources, but also have outstanding limitations such as low energy density, instability, regional variability. Due to the difference of regions and terrains, the wind speed, the wind energy capacity, the illumination time and the illumination intensity are different. The power generation and the power supply in two energy forms have instability and intermittence, and direct grid connection can affect the quality of electric energy and bring threats to safe and stable operation.
In the traditional short-term wind and light prediction, meteorological information conditions (air pressure and the like) and physical quantities (unit states and the like) of the surrounding environment and the self environment are mostly adopted as input parameters, and a neural network is trained and optimized by combining historical generated energy, so that the aim of realizing accurate prediction is fulfilled. However, conventional neural networks are prone to fall into local optima leading to failure of predictions. Moreover, the hidden layer and hidden layer node number of the neural network used by the traditional prediction are not easy to determine, which causes the training period to be too long and the precision to be low.
The traditional batteries with the capacity lower than 80 percent are regarded as scrapped, but the batteries still have larger utilization space, and the traditional battery recycling industry is to disassemble and recycle the batteries, so that heavy metals such as mercury and the like are generated in the process, and the concept of green development is not met.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the problems in the prior art, the invention provides a power regulation method of an intelligent wind-solar hybrid power generation system, which optimizes an RBF neural network prediction model by using a wild Ma Sousuo algorithm, so that the model error is smaller, wind-solar clean energy and waste batteries are fully utilized, and the economic cost and the environmental cost are greatly saved.
The technical scheme is as follows: the invention provides a power regulation method of an intelligent wind-solar hybrid power generation system, which comprises the following steps:
step 1: acquiring historical power generation data and influence parameters thereof, and establishing an RBF neural network prediction model aiming at short-term wind-solar prediction;
step 2: establishing a distributed energy model and a waste battery charging and discharging model;
and 3, step 3: optimizing the RBF neural network prediction model by utilizing a wild Ma Sousuo algorithm;
and 4, step 4: performing short-term wind and light prediction according to the model optimized in the step 3 to obtain a wind and light prediction result;
and 5: and (4) intelligently regulating and controlling the output power of the distributed energy system according to the wind and light prediction result in the step (4).
Further, the modeling of the RBF neural network prediction model in the step 1 is divided into the following steps:
step 1.1: determining 8 nodes, x, of the input layer 1 、x 2 、x 3 、x 4 、x 5 、x 6 、x 7 、x 8 Respectively corresponding to the solar radiation quantity, the inclination angle,Component shielding, wind vector, wind density, air pressure, humidity and unit state; determining 1 node of an output layer, wherein f (x) is predicted power generation amount; the RBF neural network prediction model formula is as follows:
wherein q is the number of cryptic neurons, w j Is the weight corresponding to the jth hidden layer neuron, c j Is the center corresponding to the jth hidden layer neuron,
is a radial basis function, c i Is the central point of the ith neuron, σ is the width of the Gaussian kernel, | | x i -c i | | is the sample point x i To a central point c i In the Euclidean distance of +>
And the radial basis function is the radial basis function of the corresponding position of the center corresponding to the jth hidden layer neuron of all the input parameters.
Step 1.2: according to the step 1.1, establishing a mean square error of the wind-solar prediction model, wherein the formula is as follows:
when E is minimum, the prediction model result is optimal; wherein e is i F (x) is the predicted power generation amount, and y is the actual power generation amount;
step 1.3: the neural linear weight formula for establishing the output layer is as follows:
wherein eta is the learning efficiency, E is the mean square error,
is a radial basis function, and w is a hidden layer total weight;
step 1.4: establishing a hidden layer neuron central point formula as follows:
step 1.5: the Gaussian kernel width formula for establishing the hidden layer is as follows:
further, the step 2 of modeling the distributed energy source model and the waste battery charge and discharge model is divided into the following steps:
step 2.1: establishing a photovoltaic cell model, wherein the formula is as follows:
wherein ,ISC Is short-circuit current, V oc Is an open circuit voltage, I m Is the output current at the maximum power point, V m Is the output voltage at the maximum power point;
step 2.2: establishing a fan battery model, wherein the formula is as follows:
wherein ,Pm Is the mechanical output power of the turbine, ρ is the air density, R is the blade radius, C p Is the coefficient of performance of the turbine, beta blade pitch angle, V is the wind speed, the ratio of tip speed to wind speed of the lambda rotor blade;
step 2.3: establishing a waste battery model, wherein the formula is as follows:
V T ＝E m -I batt R int
V out ＝N s V T
wherein ,VT Is a single module voltage, E m Is an open circuit voltage, I batt Current of battery, R int Internal resistance of battery, I in Current of battery pack, N p Number of parallel batteries, V out The voltage of the battery pack, the number of the Ns series-connected batteries, SOC is the state of charge of the battery, cap batt The battery capacity.
Further, the specific steps of optimizing the RBF neural network model by using the wild Ma Sousuo algorithm in the step 3 are as follows:
step 3.1: initializing parameters, inputting quantities affecting the accuracy of the prediction model: weight w, center c of basis function, gaussian kernel width σ;
step 3.2: initializing a population: setting the population number as N, the grouping number as G and the maximum iteration number as maximum;
step 3.3: creating a group and randomly selecting a leader, finding all members to carry out fitness, wherein an iterative formula is as follows:
wherein iter is the current iteration number, maxim is the maximum iteration number of the algorithm, and TDR is a self-adaptive parameter which is gradually reduced from 1 to 0 along with the increase of the iteration number;
step 3.4: the adaptive mechanism Z formula of the horse group is as follows:
wherein P is a vector consisting of 0 and 1,
is [0,1]An internal random number, IDX is a vector = = 0) that satisfies the condition (P = = 0)>
An index value of (d);
step 3.5: if the random number is less than the crossing percentage, continuing to operate; otherwise, go to step 3.3; the average crossover operator formula is as follows:
Crossover＝Mean
wherein ,
the table shows the individual positions of the group k which are entered again after an individual p in the group k has been outlier, which are then present in the group k>
Table indicates the individual position in group i at which an individual q has departed from the group and reenters group i, and/or in combination with a group number q>
The table represents the individual positions of the group j again after the individual z in the group j is outlier;
the leader is the center of the pasture area, and the group members search around the center (pasture), with the search formula as follows:
wherein ,
is the current position of a member in the group, stallion j Is the location of the leader, Z is the adaptive structure, R is [ -2,2]A uniform random number of ranges ″, based on the number of cells in the range>
New positions of the group members at grazing;
step 3.7: the leader mainly takes the members to go to a more suitable habitat, and if the current group is dominant, the region is used; if another group is dominant and they must leave the site, the location update formula is as follows:
wherein ,
is the next position of the leader of group i, WH isPosition of the sump (habitat), based on the water level in the sump>
Is the current location of the leader of group i;
step 3.8: in the later stage of the algorithm, a leader is selected according to the fitness; if the fitness of one member is better than that of the group leader, the group leader and the position of the corresponding member are changed, and the formula is as follows:
step 3.9: if the termination criteria are met, go to step 3.10; otherwise, go to step 3.3;
step 3.10: and outputting the obtained optimal solution.
Further, in the step 5, the wind and light prediction data in the step 4 are used for obtaining the generated energy trend and power prediction of the wind and light in a short period, the battery parameters of the waste battery pack and the wind and light prediction data are monitored in real time, and when the generation trend in the short period is predicted to be increased, the distributed energy system starts to charge the waste battery pack to absorb the power excess of the power grid; and when the power generation trend is predicted to rise and fall in a short period, the waste battery pack starts to discharge, and the power of the power grid is balanced.
Has the advantages that:
1. by using the RBF neural network prediction model, the wind-solar power generation in the short term in the future can be predicted by using the local historical power generation data of the chemical plant, and accurate and reasonable data support is provided for the distributed energy power regulation and control; compared with the traditional neural network model, the RBF neural network model has the advantages of better generalization capability, stronger optimization capability and no falling into local optimization. Optimizing the model by using a wild horse search algorithm, so that the model error is smaller; the accurate wind and light prediction data enables the control system to adjust the power of the distributed power supply more accurately, the use efficiency of wind and light is improved, and the economic cost is reduced.
2. In addition, the waste battery pack is used for assisting the distributed energy system to adjust power, and the waste battery pack discharges when the distributed energy system is charged at the peak, so that redundant energy is well consumed. Has better economic benefit and accords with the design concept of green and environmental protection.
Drawings
FIG. 1 is a block diagram of the present invention;
FIG. 2 is a flow chart of RBF neural network prediction according to the present invention;
FIG. 3 is a topology diagram of the RBF neural network of the present invention;
FIG. 4 is a flow chart of the field Ma Sousuo algorithm of the present invention;
FIG. 5 is a flow chart of the wild Ma Sousuo algorithm optimized RBF neural network model of the present invention
FIG. 6 is a flow chart of a wind-solar control system of the present invention
FIG. 7 is a diagram of the prediction error of the RBF neural network of the present invention
FIG. 8 is a diagram of the prediction error of the wild Ma Sousuo algorithm optimized RBF neural network of the present invention
FIG. 9 is a comparison error graph of RBF neural network prediction and optimized RBF neural network prediction by the Trojan horse search algorithm of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The invention discloses a power regulation method of an intelligent wind-solar hybrid power generation system, which is arranged on power regulation equipment shown in figure 1, wherein the power regulation equipment comprises an RBF neural network prediction model, a control system, a distributed energy system, a waste battery pack and the like; the control system may be a central processing unit or a server terminal.
The power regulation method of the intelligent wind-solar hybrid power generation system comprises the following steps:
step 1: historical power generation data are acquired, an RBF neural network prediction model for short-term wind and light prediction is established, the specific flow is shown in figure 2, and the RBF neural network topological structure is shown in figure 3.
Step 1.1: determining 8 nodes, x, of the input layer 1 、x 2 、x 3 、x 4 、x 5 、x 6 、x 7 、x 8 Respectively corresponding to the solar radiation amount, the inclination angle, the component shielding, the wind speed, the wind direction, the wind density, the air pressure and the humidity; determining 1 node of an output layer, wherein f (x) is predicted power generation amount; the RBF neural network prediction model formula is as follows:
wherein q is the number of cryptic neurons, w j Is the weight corresponding to the jth hidden layer neuron, c j Is the center corresponding to the jth hidden layer neuron,
is a radial basis function, c i Is the center point of the ith neuron and σ is the width of the gaussian kernel. | x i -c i I is a sample point x i To the central point c i Is in the Euclidean range of->
And the radial basis function is the radial basis function of the corresponding position of the center corresponding to the jth hidden layer neuron of all the input parameters.
Step 1.2: according to the step 1.1, establishing a mean square error of the wind-solar prediction model, wherein the formula is as follows:
when E is minimum, the prediction model result is optimal. Wherein e is i To input the error at the time of the ith sample, f (x) is the predicted power generation amount, and y is the actual power generation amount.
Step 1.3: the neural linear weight formula for establishing the output layer is as follows:
wherein eta is learning efficiency, E is mean square error,
for radial basis functions, w is the hidden layer total weight.
Step 1.4: establishing a hidden layer neuron central point formula as follows:
step 1.5: the Gaussian kernel width formula for the hidden layer is established as follows:
step 2: and establishing a distributed energy model and a waste battery charging and discharging model.
Step 2.1: establishing a photovoltaic cell model, wherein the formula is as follows:
wherein ,ISC Is short-circuit current, V oc Is an open circuit voltage, I m Is the output current at the maximum power point, V m Is the output voltage at the maximum power point;
step 2.2: establishing a fan battery model, wherein the formula is as follows:
wherein ,Pm Is the mechanical output power of the turbine, ρ is the air density, R is the blade radius, C p Is the coefficient of performance of the turbine, beta blade pitch angle, V is the wind speed, the ratio of tip speed to wind speed of the lambda rotor blade;
step 2.3: establishing a waste battery model, wherein the formula is as follows:
V T ＝E m -I batt R int
V out ＝N s V T
wherein ,VT Is a single module voltage, E m Is an open circuit voltage, I batt Current of battery, R int Internal resistance of battery, I in Current of battery pack, N p Number of parallel batteries, V out The voltage of the battery pack, the number of the Ns series-connected batteries, SOC is the state of charge of the battery, cap batt The battery capacity.
And step 3: and (3) optimizing the RBF neural network prediction model by using a wild Ma Sousuo algorithm, wherein the process is shown in an attached figure 4.
Step 3.1: initializing parameters, inputting quantities affecting the accuracy of the prediction model: weight w, center c of basis function, gaussian kernel width sigma;
step 3.2: initializing a population: setting the population number as N, the grouping number as G and the maximum iteration number as maximum;
step 3.3: creating a group and randomly selecting a leader, finding all members to carry out fitness, wherein an iterative formula is as follows:
wherein iter is the current iteration number, maximer is the maximum iteration number of the algorithm, and TDR is a self-adaptive parameter which is gradually reduced from 1 to 0 along with the increase of the iteration number;
step 3.4: the adaptive mechanism Z formula of the horse group is as follows:
wherein P is a vector consisting of 0 and 1,
is [0,1]An internal random number, IDX is a vector = = 0) that satisfies the condition (P = = 0)>
An index value of (d);
step 3.5: if the random number is less than the cross percentage, continuing to operate; otherwise, go to step 3.3; the average crossover operator formula is as follows:
Crossover＝Mean
wherein ,
the table shows the individual positions of the group k which are entered again after an individual p in the group k has been outlier, which are then present in the group k>
The table shows the individual position of the group i again after an individual q in the group i has been outlier, is present in the group i again and is present in the group q>
The table represents the individual positions of the group j again after the individual z in the group j is outlier;
the leader is the center of the pasture area, and the group members search around the center (pasture), with the search formula as follows:
wherein ,
is the current position of a member in the group, stallion j Is the location of the leader, Z is the adaptive structure, R is [ -2,2]Range of uniform random numbers, <' > based on the number of cells in the frame>
New positions of the group members at grazing;
step 3.7: the leader mainly takes the members to go to a more suitable habitat, and if the current group is dominant, the region is used; if another team is dominant and they must leave the site, the location update formula is as follows:
wherein ,
is the next position of the leader of group i, WH is the position of the puddle (habitat), and/or is located in the vicinity of the puddle>
Is the current location of the leader of group i;
step 3.8: in the later stage of the algorithm, a leader is selected according to the fitness; if the fitness of one member is better than that of the group leader, the group leader and the position of the corresponding member are changed according to the following formula: :
step 3.9: if the termination criteria are met, go to step 3.10; otherwise, go to step 3.3;
step 3.10: and outputting the obtained optimal solution.
And 4, step 4: and (4) carrying out short-term wind and light prediction according to the model optimized in the step (3) to obtain a wind and light prediction result.
And 5: and (4) intelligently regulating and controlling the output power of the distributed energy system according to the wind and light prediction result in the step (4), wherein the regulation and control process is shown in an attached figure 6.
Through wind and light prediction data in the step 4, generating capacity trend and power prediction of wind and light in a short period are obtained, battery parameters of the waste battery pack and wind and light prediction data are monitored in real time, and when the generation trend in the short period is predicted to be increased, the distributed energy system starts to charge the waste battery pack to absorb power excess by a power grid; and when the power generation trend is predicted to rise and fall in a short period, the waste battery pack starts to discharge, and the power of the power grid is balanced.
As shown in fig. 7 and 8, the RBF neural network model without using the wilderness search algorithm has higher accuracy for prediction, and after the optimization is performed by using the wilderness Ma Sousuo algorithm, the fitting degree of the model prediction value to the true value is improved, so that the model is more accurate. As shown in fig. 9, after the marquee algorithm optimizes the RBF model, the error value in a short period is lower than 0.02, and the matching degree to the predicted value is high.
The above embodiments are merely illustrative of the technical concepts and features of the present invention, and the purpose of the embodiments is to enable those skilled in the art to understand the contents of the present invention and implement the present invention, and not to limit the protection scope of the present invention. All equivalent changes and modifications made according to the spirit of the present invention should be covered in the protection scope of the present invention.
Claims (5)
1. A power regulation method of an intelligent wind-solar hybrid power generation system is characterized by comprising the following steps:
step 1: acquiring historical power generation data and influence parameters thereof, and establishing an RBF neural network prediction model aiming at short-term wind and light prediction;
step 2: establishing a model of distributed energy and a waste battery charging and discharging model;
and step 3: optimizing the RBF neural network prediction model by utilizing a wild Ma Sousuo algorithm;
and 4, step 4: carrying out short-term wind-solar prediction according to the model optimized in the step 3 to obtain a wind-solar prediction result;
and 5: and (5) intelligently regulating and controlling the output power of the distributed energy system according to the wind and light prediction result in the step (4).
2. The power regulation method of the intelligent wind-solar hybrid power generation system according to claim 1, wherein the modeling of the RBF neural network prediction model in the step 1 is divided into the following steps:
step 1.1: determining 8 nodes, x, of the input layer 1 、x 2 、x 3 、x 4 、x 5 、x 6 、x 7 、x 8 Respectively corresponding to the solar radiation amount, the inclination angle, the component shielding, the wind vector, the wind density, the air pressure, the humidity and the unit state; determining 1 node of an output layer, wherein f (x) is predicted power generation amount; the RBF neural network prediction model formula is as follows:
wherein q is the number of cryptic neurons, w j Is the weight corresponding to the jth hidden layer neuron, c j Is the center corresponding to the jth hidden layer neuron,
is a radial basis function, c i Is the central point of the ith neuron, σ is the width of the Gaussian kernel, | | x i -c i I is a sample point x i To the central point c i Is in the Euclidean range of->
And the radial basis function is the radial basis function of the corresponding position of the center corresponding to the jth hidden layer neuron of all the input parameters.
Step 1.2: according to the step 1.1, establishing a mean square error of the wind-solar prediction model, wherein a formula is as follows:
when E is minimum, the prediction model result is optimal; wherein e is i F (x) is the predicted power generation amount, and y is the actual power generation amount;
step 1.3: the neural linear weight formula for establishing the output layer is as follows:
wherein eta is learning efficiency, E is mean square error,
is a radial basis function, and w is a hidden layer total weight;
step 1.4: establishing a hidden layer neuron central point formula as follows:
step 1.5: the Gaussian kernel width formula for establishing the hidden layer is as follows:
3. the power regulation method of the intelligent wind-solar hybrid power generation system according to claim 1, wherein the modeling of the distributed energy source model and the waste battery charge-discharge model in the step 2 is divided into the following steps:
step 2.1: establishing a photovoltaic cell model, wherein the formula is as follows:
wherein ,ISC Is short-circuit current, V oc Is an open circuit voltage, I m Is the output current at the maximum power point, V m Is the output voltage at the maximum power point;
step 2.2: establishing a fan battery model, wherein the formula is as follows:
wherein ,Pm Is the mechanical output power of the turbine, ρ is the air density, R is the blade radius, C p Is the coefficient of performance of the turbine, beta blade pitch angle deg.C, V is the wind speed, the ratio of the tip speed of the lambda rotor blade to the wind speed;
step 2.3: establishing a waste battery model, wherein the formula is as follows:
V T ＝E m -I batt R int
V out ＝N s V T
wherein ,VT Is a single module voltage, E m Is an open circuit voltage, I batt Current of battery, R int Internal resistance of battery, I in Current of battery pack, N p Number of parallel batteries, V out Voltage of battery pack, N s Number of series connected cells, SOC being the state of charge of the cell, cap batt The battery capacity.
4. The power regulation method of the intelligent wind-solar hybrid power generation system according to claim 2, wherein the specific step of optimizing the RBF neural network model by using the wild Ma Sousuo algorithm in the step 3 is as follows:
step 3.1: initializing parameters, inputting quantities affecting the accuracy of the prediction model: weight w, center c of basis function, gaussian kernel width sigma;
step 3.2: initializing a population: setting the population quantity as N, the grouping quantity as G, and the maximum iteration number as maximum;
step 3.3: creating a group and randomly selecting a leader, finding all members to carry out fitness, wherein an iterative formula is as follows:
wherein iter is the current iteration number, maxim is the maximum iteration number of the algorithm, and TDR is a self-adaptive parameter which is gradually reduced from 1 to 0 along with the increase of the iteration number;
step 3.4: the adaptive mechanism Z formula of the horse group is as follows:
wherein P is a vector consisting of 0 and 1,
is [0,1]An internal random number, IDX is a vector satisfying a condition (P = = 0)
The index value of (a);
step 3.5: if the random number is less than the crossing percentage, continuing to operate; otherwise, go to step 3.3; the average crossover operator formula is as follows:
Crossover＝Mean
wherein ,
the table shows the positions of individuals in group k who have escaped from group p and reentered group k, and/or>
The table shows the individual position of the group i again after an individual q in the group i has been outlier, is present in the group i again and is present in the group q>
The table represents the individual positions of the group j again after the individual z in the group j is outlier;
the leader is the center of the pasture area, and the group members search around the center (pasture), with the search formula as follows:
wherein ,
is the current position of a member in the group, stallion j Is the location of the leader, Z is the adaptive structure, R is [ -2,2]A uniform random number of ranges ″, based on the number of cells in the range>
New positions of the group members at grazing;
step 3.7: the leader mainly takes the members to go to a more suitable habitat, and if the current group is dominant, the region is used; if another team is dominant and they must leave the site, the location update formula is as follows:
wherein ,
is the next position of the leader of group i, WH is the position of the puddle (habitat), and/or is located in the vicinity of the puddle>
Is the current location of the group i leader;
step 3.8: in the later stage of the algorithm, a leader is selected according to the fitness; if the fitness of one member is better than the group leader, the group leader and the position of the corresponding member are changed, and the formula is as follows:
step 3.9: if the termination criteria are met, go to step 3.10; otherwise, go to step 3.3;
step 3.10: and outputting the obtained optimal solution.
5. The power regulation method of the intelligent wind-solar hybrid power generation system according to any one of claims 1 to 4, characterized in that in the step 5, wind-light prediction data in the step 4 is used to obtain the power generation amount trend and power prediction of wind-light in a short period, battery parameters of the waste battery pack and the wind-light prediction data are monitored in real time, and when the power generation trend in the short period is predicted to be increased, the distributed energy system starts to charge the waste battery pack to absorb the excess power of the power grid; and when the power generation trend is predicted to rise and fall in a short period, the waste battery pack starts to discharge, and the power of the power grid is balanced.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211352772.2A CN115912331B (en) | 2022-10-30 | 2022-10-30 | Power adjusting method of intelligent wind-solar complementary power generation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211352772.2A CN115912331B (en) | 2022-10-30 | 2022-10-30 | Power adjusting method of intelligent wind-solar complementary power generation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115912331A true CN115912331A (en) | 2023-04-04 |
| CN115912331B CN115912331B (en) | 2023-09-29 |
Family
ID=86480669
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211352772.2A Active CN115912331B (en) | 2022-10-30 | 2022-10-30 | Power adjusting method of intelligent wind-solar complementary power generation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115912331B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118113805A (en) * | 2024-04-29 | 2024-05-31 | 山东省国土测绘院 | Geographic information survey calibration method and system based on deep learning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013074695A (en) * | 2011-09-27 | 2013-04-22 | Meiji Univ | Device, method and program for predicting photovoltaic generation |
| CN106227915A (en) * | 2016-07-07 | 2016-12-14 | 南京工程学院 | Disc type solar energy heat collector exit temperature prediction method based on GA RBF |
| CN108960491A (en) * | 2018-06-15 | 2018-12-07 | 常州瑞信电子科技有限公司 | Method for forecasting photovoltaic power generation quantity based on RBF neural |
| CN112537215A (en) * | 2020-12-19 | 2021-03-23 | 郑州新开元科技有限公司 | Semi-off-grid type wind-solar complementary intelligent charging station |
| CN113555899A (en) * | 2021-07-20 | 2021-10-26 | 南京工程学院 | Coordination control method for wind-solar energy storage power generation system |
| CN113673786A (en) * | 2021-09-06 | 2021-11-19 | 上海海事大学 | Effective wind speed estimation system and method of wind driven generator based on RBF and LSTM |
-
2022
- 2022-10-30 CN CN202211352772.2A patent/CN115912331B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013074695A (en) * | 2011-09-27 | 2013-04-22 | Meiji Univ | Device, method and program for predicting photovoltaic generation |
| CN106227915A (en) * | 2016-07-07 | 2016-12-14 | 南京工程学院 | Disc type solar energy heat collector exit temperature prediction method based on GA RBF |
| CN108960491A (en) * | 2018-06-15 | 2018-12-07 | 常州瑞信电子科技有限公司 | Method for forecasting photovoltaic power generation quantity based on RBF neural |
| CN112537215A (en) * | 2020-12-19 | 2021-03-23 | 郑州新开元科技有限公司 | Semi-off-grid type wind-solar complementary intelligent charging station |
| CN113555899A (en) * | 2021-07-20 | 2021-10-26 | 南京工程学院 | Coordination control method for wind-solar energy storage power generation system |
| CN113673786A (en) * | 2021-09-06 | 2021-11-19 | 上海海事大学 | Effective wind speed estimation system and method of wind driven generator based on RBF and LSTM |
Non-Patent Citations (3)
| Title |
|---|
| HUSAM A. RAMADAN ET AL.: "Accurate Parameters Estimation of Three Diode Model of Photovoltaic Modules Using Hunter–Prey and Wild Horse Optimizers", 《IEEE ACCESS》, pages 87435 * |
| 姚传安: "小型风光互补发电系统控制和能量预测技术研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》, no. 03, pages 042 - 26 * |
| 胡治辉: "基于云服务平台智能微网控制策略研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》, no. 8, pages 042 - 412 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118113805A (en) * | 2024-04-29 | 2024-05-31 | 山东省国土测绘院 | Geographic information survey calibration method and system based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115912331B (en) | 2023-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mirzapour et al. | A new prediction model of battery and wind-solar output in hybrid power system | |
| CN109888840B (en) | Scheduling optimization method and device of wind-solar-garbage power generation energy comprehensive utilization system | |
| CN110956324B (en) | Day-ahead high-dimensional target optimization scheduling method for active power distribution network based on improved MOEA/D | |
| CN114400712B (en) | Micro-grid group optimization scheduling method based on improved second-order particle swarm algorithm | |
| CN109560574A (en) | A kind of intelligent distribution network space truss project method considering uncertain factor | |
| CN113783224A (en) | Power distribution network double-layer optimization planning method considering operation of various distributed energy sources | |
| CN114243791A (en) | Multi-objective optimization configuration method, system and storage medium for wind-solar-hydrogen storage system | |
| CN115526401A (en) | Novel power supply optimal planning method for electric power system based on digital twinning | |
| CN112418496B (en) | Power distribution station energy storage configuration method based on deep learning | |
| CN116667325B (en) | Micro-grid-connected operation optimization scheduling method based on improved cuckoo algorithm | |
| CN115912331A (en) | Power adjusting method of intelligent wind-solar hybrid power generation system | |
| CN115622101A (en) | Energy storage optimal configuration double-layer planning method for promoting renewable energy consumption | |
| CN109950933B (en) | Wind-solar-storage combined peak regulation optimization method based on improved particle swarm optimization | |
| CN111311032B (en) | Micro-grid system capacity optimal configuration method based on sector radar map model | |
| CN116995740A (en) | Distributed wind power and energy storage optimal configuration method and system for power distribution network | |
| CN112036735A (en) | Energy storage capacity planning method and system for energy storage system of photovoltaic power station | |
| Nagaballi et al. | A hybrid fuzzy and PSO based optimum placement and sizing of DG in radial distribution system | |
| Raj et al. | Numerical Simulation and Comparative Assessment of Improved Cuckoo Search and PSO based MPPT System for Solar Photovoltaic System Under Partial Shading Condition | |
| CN114759616B (en) | Micro-grid robust optimization scheduling method considering characteristics of power electronic devices | |
| CN112952914B (en) | Multi-objective operation optimization method for multi-energy complementary power system | |
| CN115473306A (en) | Hybrid energy storage system recycling regulation and control method based on intelligent algorithm | |
| CN110729759B (en) | Method and device for determining distributed power supply configuration scheme in micro-grid | |
| Wang et al. | A novel framework for optimal sizing of a dc microgrid considering energy management and battery degradation | |
| CN113937767A (en) | Micro-grid capacity optimization configuration method based on improved whale algorithm | |
| CN112117772A (en) | Energy storage fast optimization configuration method for considering output resistor plug under new energy access |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |