CN117767430A - Distributed power supply system and method based on intelligent energy management platform - Google Patents
Distributed power supply system and method based on intelligent energy management platform Download PDFInfo
- Publication number
- CN117767430A CN117767430A CN202311724348.0A CN202311724348A CN117767430A CN 117767430 A CN117767430 A CN 117767430A CN 202311724348 A CN202311724348 A CN 202311724348A CN 117767430 A CN117767430 A CN 117767430A
- Authority
- CN
- China
- Prior art keywords
- generator
- generator set
- power
- module
- loss
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000002068 genetic effect Effects 0.000 claims abstract description 45
- 238000012417 linear regression Methods 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 40
- 238000013528 artificial neural network Methods 0.000 claims description 18
- 210000000349 chromosome Anatomy 0.000 claims description 14
- 238000012795 verification Methods 0.000 claims description 13
- 238000004146 energy storage Methods 0.000 claims description 7
- 230000035772 mutation Effects 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- ZZUFCTLCJUWOSV-UHFFFAOYSA-N furosemide Chemical compound C1=C(Cl)C(S(=O)(=O)N)=CC(C(O)=O)=C1NCC1=CC=CO1 ZZUFCTLCJUWOSV-UHFFFAOYSA-N 0.000 claims 1
- 238000004364 calculation method Methods 0.000 description 23
- 238000010248 power generation Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The utility model provides a distributed power supply system and method based on wisdom energy management platform, its core scheme includes two at least generating sets and energy management platform, and energy management platform includes: the global power prediction module is used for calculating expected use power of the power supply area in the first time period based on a linear regression equation; the unit power prediction module is used for calculating the maximum unit output power of the generator unit in the first time period based on the first prediction model; the weight distribution module is used for setting the loss weight of the generator set; the judging module is used for selecting a first genetic algorithm as a global scheduling algorithm; and the scheduling module is used for calculating the expected power of the generator set according to the first genetic algorithm. The distributed power supply system and the method can reasonably schedule the output power of different generator sets, ensure the stability of the complex distributed power supply system, and improve the flexibility of system scheduling.
Description
Technical Field
The application relates to a distributed power supply system, in particular to a distributed power supply system and method based on an intelligent energy management platform.
Background
With the advancement of renewable energy technology and the demand for modernization of power grids, distributed power supply systems have become a key trend in the power industry. Distributed power systems often include a variety of generator sets, such as wind, thermal, hydroelectric, solar. Under complex power supply systems, an efficient power distribution strategy is critical to ensure the stability of the grid. Meanwhile, considering the power cost factor, how to schedule the output power of the generator set based on the stability and cost factors becomes a problem to be solved.
In addition, since the distributed power generation system is complex, how to improve the flexibility of system scheduling is also a concern under such complex systems.
In the prior art, such as CN111668882A, CN114329317a, a scheduling scheme of a hybrid generator set is proposed, but stability and cost factors of a power grid are not comprehensively considered, and the provided scheduling algorithm has limited accuracy, lacks flexibility, and is difficult to meet the requirements of a complex power grid.
Disclosure of Invention
The distributed power supply system and the distributed power supply method based on the intelligent energy management platform can reasonably schedule the output power of different generator sets, ensure the stability of a complex distributed power supply system and improve the flexibility of system scheduling.
In one aspect, the application provides a distributed power supply system based on wisdom energy management platform, including two at least generating sets and energy management platform, above-mentioned two at least generating sets include first generating set, second generating set, and the generator type of above-mentioned two at least generating sets is different, and above-mentioned energy management platform includes:
the power prediction module comprises a global power prediction module and a unit power prediction module;
the global power prediction module is configured to calculate a first time of the power supply area based on a linear regression equationExpected usage power P of a bay Total (S) ；
The unit power prediction module is configured to calculate a maximum unit output power P of the first generator unit in the first period based on a first prediction model 1max Maximum unit output power P of second generator unit in the first time period 2max The method comprises the steps of carrying out a first treatment on the surface of the The algorithm according to the first prediction model is a neural network algorithm;
the weight distribution module is used for setting the loss weight w1 of the first generator set and the loss weight w2 of the second generator set;
the judging module is used for selecting a first genetic algorithm as a global scheduling algorithm when the fact that the at least two generator sets do not comprise a third generator set is recognized; the type of the generator of the third generator set is different from that of the first generator set and the second generator set;
a scheduling module for calculating the expected power P of the first generator set in a first time period according to the first genetic algorithm, wherein the adaptability function according to the first genetic algorithm is at a minimum value 1 anticipation of And the expected power P of the second generator set in the first time period 2 anticipation of ；
The fitness function according to the first genetic algorithm is F (P) = (w1×c) 1 (P 1 )+w2×C 2 (P 2 )+|w0(P 1 +P 2 -P Total (S) ) I), C above 1 (P 1 )、C 2 (P 2 ) The loss functions of the first generator set and the second generator set are respectively, P 1 、P 2 The output power of the first generator set and the second generator set is the output power of the first generator set and the second generator set; the w0 is a power fluctuation loss weight and is used for adjusting the loss of power fluctuation to the system;
the constraint conditions of the fitness function F (P) are: p is more than or equal to 0 1 ≤P 1max ，0≤P 1 ≤P 2max 。
The distributed power supply system has the following advantages: the judging module determines genetic algorithms with different complexity based on the attributes of the generator set, so that operation results caused by excessively complex parameters can be avoidedPoor or host system crashes, ensuring scheduling stability. Different power sets are provided with different loss weights, and the output power of the different power sets can be adjusted according to the requirements of a decision maker in a weight setting mode, so that the scheduling flexibility is ensured. The fitness function includes a loss function of the generator set and a power fluctuation loss w0 (P 1 +P 2 -P Total (S) ) The power fluctuation loss is used for evaluating the loss caused by unbalanced power supply and demand, and the fitness function can enable the power generation loss and the loss caused by unbalanced power supply and demand to be fully considered in the system scheduling process, so that the scheduling stability and flexibility are ensured. The maximum unit output power represents the maximum power of the generator unit, the maximum unit output power is predicted, the expected power can be limited within the maximum power, and the accuracy and the stability of scheduling are improved. The global prediction module adopts a linear regression equation to operate, because the power demand change of the power supply area is easier to calculate from historical data, and the system calculation force can be saved by adopting a simpler linear regression equation. The scheme fully realizes reasonable scheduling of the output power of different generator sets, ensures the stability of a complex distributed power supply system, and improves the scheduling flexibility of the system.
In one embodiment, the generator type of the first generator set is a thermal generator, the generator type of the second generator set is a wind generator, the generator type of the third generator set is an energy storage generator or a hydroelectric generator, and the loss function of the first generator setLoss function of the second generator set>The above-mentioned a1, a2, a3, b1, b2, b3 are loss constants of the generator.
The chromosome matrix adopted by the first genetic algorithm is P x ＝[P 1 code ，P 2 coding ]Wherein, P is as described above 1 code 、P 2 coding Respectively P 1 、P 2 Is a vectorized representation of (2);
the crossover operator according to the first genetic algorithm is P x (t+1)＝P great +c1×P great -c2×P x (t); wherein t is the iteration number and P is the number of iterations great For a chromosome matrix that minimizes the F (P) value in t iterations, c1 and c2 are random vectors.
In one possible embodiment, the mutation operator according to the first genetic algorithm is P x (t+1)＝P x (t) + (∈θ), where e is a random vector,γ＝2F(P x (t))-F(P x (t-1))。
the improved crossover operator and mutation operator are adopted to calculate the scheduling scheme, so that the calculation accuracy can be greatly improved, and the stability of system scheduling is ensured.
In one embodiment, the first prediction model includes a first neural network algorithm and a second neural network algorithm, and the unit power prediction module is specifically configured to:
inputting the historical data of the same date into the first neural network to obtain a first predicted maximum output power P 1a 、P 2a Inputting historical data of a first time period with different dates into the second neural network to obtain a second predicted maximum output power P 1b 、P 2b ；
Weighting the first and second predicted maximum output powers to obtain P 1 anticipation of 、P 2 anticipation of 。
The method adopts a mode of combining transverse calculation and longitudinal calculation, and the maximum output power of the generator set is calculated through the historical data of the same day and the historical data of the same time period of different days, so that the calculation precision can be greatly improved, and the stability of system scheduling is ensured.
The loss weight w1 of the first generator set is greater than the loss weight w2 of the second generator set.
Because the first generator set is a thermal generator set and the second generator set is a wind generator set, the loss weight w1 of the first generator set is larger than the loss weight w2 of the second generator set in consideration of environmental protection factors, and more power generation is distributed to the wind generator sets.
In one embodiment, upon identifying that the at least two gensets include a third genset, selecting a second genetic algorithm as the global scheduling algorithm; wherein the complexity of the second genetic algorithm is less than the complexity of the first genetic algorithm, the complexity being indicative of the time taken for the operation.
When various generator sets exist in the system, a global scheduling algorithm with lower complexity is adopted in time, so that system breakdown or excessively long scheduling time consumption is avoided.
In one embodiment, the weight allocation module includes: the monitoring module is used for monitoring the weight setting instruction; the verification module is used for reading the instruction verification information when the weight setting instruction is obtained, and verifying the weight setting instruction based on the instruction verification information; the analyzing module is used for analyzing the weight setting instruction to obtain the loss weight w1 of the first generator set and the loss weight w2 of the second generator set when the weight setting instruction passes the verification.
According to the method and the device, the monitoring module is arranged, the weight setting instruction can be timely obtained, and the scheduling strategy can be timely adjusted. Because the weight distribution is the decision item with the highest authority, if the decision item is intruded improperly, the stability of the dispatching system is seriously damaged, and therefore, the application specially verifies the weight setting instruction so as to ensure the dispatching safety.
The distributed power supply system provided by the application can reasonably schedule the output power of different generator sets, ensure the stability of the complex distributed power supply system and improve the flexibility of system scheduling.
Drawings
In order to more clearly describe the technical solutions in the embodiments or the background of the present application, the following description will describe the drawings that are required to be used in the embodiments or the background of the present application.
FIG. 1 is a distributed power supply system based on an intelligent energy management platform;
FIG. 2 is a schematic diagram of a distributed power supply method based on an intelligent energy management platform;
fig. 3 is a schematic diagram of a distributed power supply device based on an intelligent energy management platform according to the present application.
Detailed Description
The terms first, second and the like in the description and in the claims of the present application and in the above-described figures, are used for distinguishing between different objects and not for describing a particular sequential order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, or apparatus.
Fig. 1 is a schematic diagram of a distributed power supply system based on an intelligent energy management platform.
The distributed power supply system includes at least two generator sets 100 and an energy management platform; the at least two generator sets comprise a first generator set 110 and a second generator set 120, and the generator types of the at least two generator sets are different; the energy management platform comprises a power prediction module 210, a weight distribution module 220, a discrimination module 230 and a scheduling module 240, wherein the power prediction module comprises a global power prediction module 211 and a unit power prediction module 212.
The generator types of the at least two generator sets may include a thermal generator, a wind generator, a solar generator, a hydro-discharge machine, an energy storage generator, etc., the generator types of the at least two generator sets being different.
The global power prediction module 211 is configured to be based on a linear regression equationCalculating the expected usage power P of the first period of the power supply region Total (S) . The linear regression equation may be a unitary linear regression model; the input parameters of the linear regression equation are the use power of the power supply area in the same time period with different dates. The power supply area is a target power supply area of the distributed power supply system.
In the distributed power supply system, a certain power supply area is often powered by a plurality of power supply units with different categories, and the power supply area is smaller. The power supply area is a target power supply area of the distributed power supply system. The first period is the period to be predicted, often the peak electricity consumption period, during which the output power of the generator set is typically required to be increased.
The unit power prediction module 212 is configured to calculate a maximum unit output power P of the first generator unit during the first period based on a first prediction model 1max Maximum unit output power P of second generator unit in the first time period 2max The method comprises the steps of carrying out a first treatment on the surface of the The algorithm according to the first prediction model is a neural network algorithm. Wherein the maximum unit output power is related to the power generation factor. For a wind generator, the maximum unit output power is related to the wind force of the first period of time; for a thermal generator, the maximum unit output power is related to the input fuel in the first time period; for a hydro-generator, the maximum unit output power is related to the water flow rate for the first period of time.
Because the new energy power generation element is difficult to predict, a neural network algorithm is specially arranged to predict the maximum unit output power so as to improve the accuracy. Further, in order to further improve the accuracy of the prediction, the unit output power of the generator can be calculated by combining the lateral calculation and the longitudinal calculation.
Specifically, the first prediction model includes a first neural network algorithm and a second neural network algorithm, and the lateral calculation refers to: inputting historical data of different time periods of the same day into the first neural network to obtain a first predicted maximum output power P 1a 、P 2a . The longitudinal calculation refers to: the date to which it belongs is notThe historical data of the first time period of the same day is input into the second neural network to obtain a second predicted maximum output power P 1b 、P 2b . Finally, weighting the first predicted maximum output power and the second predicted maximum output power to obtain the P 1 anticipation of 、P 2 anticipation of . The weighting of the weighting operation may be distributed according to the accuracy of the lateral calculation result and the longitudinal calculation result determined by the test result, for example, if the maximum output power calculated by the lateral calculation result is determined to be more matched with the actual situation by the test, the lateral calculation result is distributed with a higher weight.
It should be noted that the purpose of calculating the maximum unit output is to determine the constraint of the fitness function, i.e. the power variable of the fitness function cannot exceed the maximum unit output.
The weight distribution module 220 is configured to set the loss weight w1 of the first generator set and the loss weight w2 of the second generator set. The loss weights described above are used to construct a fitness function of the genetic algorithm, wherein, if the loss weights are increased, the output power that the engine block can ultimately distribute will decrease. In an actual application scenario, if the first generator set is a thermal generator and the second generator set is a wind driven generator, w1> w2 can be set based on environmental protection, so that the output power of the thermal generator is lower and the output power of the wind driven generator is higher.
A discriminating module 230, configured to select a first genetic algorithm as a global scheduling algorithm when it is recognized that the at least two generating sets do not include a third generating set; the type of the generator of the third generator set is different from that of the first generator set and the second generator set. The generator type of the first generator set is a thermal generator, the generator type of the second generator set is a wind driven generator, and the generator type of the third generator set is an energy storage generator or a hydroelectric generator. Because the first genetic algorithm is a scheduling algorithm with higher complexity, the calculation accuracy is relatively higher, but the first genetic algorithm is not suitable for a system with excessively complex parameters. When the number of generator sets is high, and particularly the hydroelectric generator with more complex parameters is included, the genetic algorithm with lower complexity should be adjusted. Therefore, the discrimination module 230 is also configured to: when the at least two generator sets are identified to comprise a third generator set, selecting a second genetic algorithm as a global scheduling algorithm; wherein the complexity of the second genetic algorithm is less than the complexity of the first genetic algorithm, the complexity being indicative of the time taken for the operation.
A scheduling module 240 for calculating an expected power P of the first generator set in a first period of time according to the first genetic algorithm such that an fitness function according to the first genetic algorithm is at a minimum 1 anticipation of And the expected power P of the second generator set in the first time period 2 anticipation of . The expected power is a planned output power of the generator set in a first period of time.
The fitness function according to the first genetic algorithm is F (P) = (w1×c) 1 (P 1 )+w2×C 2 (P 2 )+|w0(P 1 +P 2 -P Total (S) ) I), C above 1 (P 1 )、C 2 (P 2 ) The loss functions of the first generator set and the second generator set are respectively, P 1 、P 2 The output power of the first generator set and the second generator set is the output power of the first generator set and the second generator set; and w0 is a power fluctuation loss weight and is used for adjusting the weight occupied by the loss of the power fluctuation to the system.
As described above, the fitness function includes two major parts, namely the loss of the generator set, which includes the sum of the losses of the generators of each set, and the power fluctuation loss. W0 for power fluctuation loss (P) 1 +P 2 -P Total (S) ) W0 is the power fluctuation loss weight, the larger w0 means that the larger the influence of the power fluctuation factor on the calculation result is, the smaller the loss influence of the generator set is relative, and the calculation result will cause the power fluctuation (P 1 +P 2 -P Total (S) ) Is smaller.
The above C 1 (P 1 )、C 2 (P 2 ) The loss functions of the first generator set and the second generator set are respectively that of the thermal generatorMotor setLoss function of wind generating setThe above-mentioned a1, a2, a3, b1, b2, b3 are loss constants of the generator. The loss constant is related to the properties of the unit equipment itself, often obtained by testing the unit equipment.
If the system further comprises a third generator set (such as an energy storage generator set or a hydroelectric generator set), the fitness function can also comprise a loss function C of the third generator set 3 (P 3 ). The loss function of the energy storage generator can be defined as a linear function, and can be defined as a function which is positively correlated with the source of electric quantity according to the source of electric quantity (such as hydraulic power, firepower and wind power) of the energy storage generator. The loss function of a hydroelectric generator is generally related to factors such as equipment investment cost, operation and maintenance cost, water source cost, environment and supervision cost, and is more complex than other generator sets, and one expression form is as follows: wherein c1, c2, c3 are loss constants. In order to make the calculation more accurate, the loss function often needs to be measured by comprehensively considering the equipment and environmental conditions through field investigation and adopting complex expressions such as indexes, logarithms and the like.
The method adopts a genetic algorithm as a global scheduling algorithm, wherein the basic steps of the genetic algorithm comprise: basic steps of population initialization, fitness evaluation, chromosome selection, chromosome crossing, chromosome variation, termination condition examination and the like. There are a number of expressions associated with each basic step in the art, not specifically recited herein. In chromosome initialization, variables are encoded into a chromosome matrix, such as P x ＝[P 1 code ，P 2 coding ]Wherein, P is as described above 1 code 、P 2 coding Respectively P 1 、P 2 Vectorized or matrixed representation of (c).
In order to improve the calculation accuracy, the application is particularly aimed at improving the crossover operator in the chromosome crossover step and the mutation operator in the chromosome mutation. Wherein the crossover operator is P x (t+1)＝P great +c1×P great -c2×P x (t). Wherein t is the iteration number and P is the number of iterations great Is a chromosome matrix that minimizes the F (P) value in t iterations. The above c1 and c2 are random vectors, and in one embodiment, each element can be randomly selected from 0 and 1, and the length and width and P x The same applies. The global optimal value is brought into the cross calculation formula by the cross operator, so that the calculation accuracy can be greatly improved.
In one embodiment, the mutation operator is P x (t+1)＝P x (t) + (∈θ), where e is a random vector,γ＝2F(P x (t))-F(P x (t-1)). The mutation operator fuses the operation results of the previous two times, so that the operation precision can be further improved.
The above list only one preferred calculation method, and in fact, other calculation methods in the prior art may be selected, which may only have a difference in calculation accuracy. For example, the chromosome selection step may be performed by using conventional formulas such as roulette and tournament selection, and the method for encoding the chromosome may be a conventional binary encoding method.
In one possible implementation, the weight allocation module 220 includes: a listening module 221, configured to listen for a weight setting instruction; a verification module 222, configured to read instruction verification information when a weight setting instruction is obtained, and verify the weight setting instruction based on the instruction verification information; the analyzing module 223 is configured to analyze the weight setting instruction to obtain the loss weight w1 of the first generator set and the loss weight w2 of the second generator set when the weight setting instruction passes the verification.
In fact, the loss weight of the generator and the loss weight of the power fluctuation can determine the power distribution of the generator set to a great extent, and often can represent the intention of a decision maker, so that the monitoring and verification of the weight setting instruction are limited, and the real-time performance and the stability of the power distribution of the power supply system are fully ensured.
In one possible implementation, the scheduling module 240 is further configured to: transmitting a scheduling instruction to the at least two generator sets, wherein the scheduling instruction is used for setting the output power of the first generator set to P in the first time period 1 anticipation of Setting the output power of the second generator set to P 2 anticipation of . After the expected power of each generator set is calculated, the expected power is required to be sent to each generator set in time, so that each generator set sets the output power of the generator set to the expected power in a first time period.
The following beneficial effects are achieved with respect to the present application: the genetic algorithm with different complexity is determined based on the attributes of the generator set, so that poor operation results or host system breakdown caused by too complex parameters can be avoided, and the scheduling stability is ensured. Different power sets are provided with different loss weights, and the output power of the different power sets can be adjusted according to the requirements of a decision maker in a weight setting mode, so that the scheduling flexibility is ensured. The fitness function comprises a loss function of the generator set and power fluctuation loss, so that the power generation loss and the loss caused by unbalanced power supply and demand can be fully considered in the system scheduling process, and the scheduling stability and flexibility are ensured. And the maximum unit output power is predicted, the expected power can be limited within the maximum generated power, and the accuracy and stability of scheduling are improved. The method and the device realize reasonable scheduling of the output power of different generator sets, ensure the stability of the complex distributed power supply system, and improve the scheduling flexibility of the system.
As shown in fig. 2, the present application further provides a distributed power supply method based on an intelligent energy management platform, which is used for a distributed power supply system, where the distributed power supply system includes at least two generator sets and an energy management platform; the at least two generator sets comprise a first generator set and a second generator set, and the generator types of the at least two generator sets are different; the energy management platform comprises a power prediction module, a weight distribution module, a judging module and a scheduling module, wherein the power prediction module comprises a global power prediction module and a unit power prediction module, and the method is characterized by comprising the following steps:
101. calculating the expected use power P of the power supply area in the first time period based on a linear regression equation by the global power prediction module Total (S) ；
102. Calculating the maximum unit output power P of the first generator set in the first time period based on the first prediction model through a unit power prediction module 1max Maximum unit output power P of second generator unit in the first time period 2max The method comprises the steps of carrying out a first treatment on the surface of the The algorithm according to the first prediction model is a neural network algorithm;
103. the loss weight w1 of the first generator set and the loss weight w2 of the second generator set are set through a weight distribution module;
104. when the at least two generator sets are identified to not comprise a third generator set through the judging module, a first genetic algorithm is selected as a global scheduling algorithm; the type of the generator of the third generator set is different from that of the first generator set and the second generator set;
105. calculating, by a scheduling module, an expected power P of the first generator set in a first period of time according to the first genetic algorithm, such that an fitness function according to the first genetic algorithm is at a minimum 1 anticipation of And the expected power P of the second generator set in the first time period 2 anticipation of ；
The fitness function according to the first genetic algorithm is F (P) = (w1×c) 1 (P 1 )+w2×C 2 (P 2 )+|w0(P 1 +P 2 -P Total (S) ) I), C above 1 (P 1 )、C 2 (P 2 ) The loss functions of the first generator set and the second generator set are respectively, P 1 、P 2 The output power of the first generator set and the second generator set is the output power of the first generator set and the second generator set; the w0 is a power fluctuation loss weight and is used for adjusting the weight occupied by the loss of the power fluctuation to the system;
the constraint conditions of the fitness function F (P) are: p is more than or equal to 0 1 ≤P 1max ，0≤P 2 ≤P 2max 。
For the extension part and the detail part of the foregoing distributed power supply method based on the intelligent energy management platform, please refer to an embodiment of the distributed power supply system part based on the intelligent energy management platform, which is not described herein again.
Fig. 3 is a schematic structural diagram of a distributed power supply device based on an intelligent energy management platform. The device comprises: at least one processor 310, such as a central processing unit (central processing unit, CPU), at least one memory 320, and at least one bus 330. The memory 320 may store program instructions, and the processor 310 may be configured to invoke the program instructions to perform a distributed power method based on an intelligent energy management platform.
Claims (10)
1. Distributed power supply system based on wisdom energy management platform, including two at least generating sets and energy management platform, its characterized in that, two at least generating sets include first generating set, second generating set, two at least generating set's generator type is different, energy management platform includes:
the power prediction module comprises a global power prediction module and a unit power prediction module;
the global power prediction module is used for calculating the expected use power P of the first time period of the power supply area based on a linear regression equation Total (S) ；
The unit power prediction module is used for calculating the maximum unit output power P of the first generator unit in the first time period based on a first prediction model 1max Maximum unit output power P of second generator unit in the first time period 2max The method comprises the steps of carrying out a first treatment on the surface of the The first prediction model is based onThe algorithm is a neural network algorithm;
the weight distribution module is used for setting the loss weight w1 of the first generator set and the loss weight w2 of the second generator set;
the judging module is used for selecting a first genetic algorithm as a global scheduling algorithm when the at least two generator sets do not comprise a third generator set; the type of the generator of the third generator set is different from that of the first generator set and the second generator set;
a scheduling module for calculating, according to the first genetic algorithm, an expected power P of the first generator set in a first period of time, the fitness function according to which the first genetic algorithm is made to be at a minimum 1 anticipation of And the expected power P of the second generator set in the first time period 2 anticipation of ；
The fitness function according to which the first genetic algorithm is based is F (P) = (w1×c) 1 (P 1 )+w2×C 2 (P 2 )+|w0(P 1 +P 2 -P Total (S) ) I), said C 1 (P 1 )、C 2 (P 2 ) The loss functions of the first generator set and the second generator set are respectively P 1 、P 2 The output power of the first generator set and the output power of the second generator set are respectively; the w0 is a power fluctuation loss weight and is used for adjusting the weight occupied by the loss of the power fluctuation to the system;
constraint conditions of the fitness function F (P) are as follows: p is more than or equal to 0 1 ≤P 1max ，0≤P 2 ≤P 2max 。
2. The system of claim 1, wherein the generator type of the first generator set is a thermal generator, the generator type of the second generator set is a wind generator, the generator type of the third generator set is an energy storage generator or a hydro generator, and the loss function of the first generator set The loss function of the second generator set +.> The a1, a2, a3, b1, b2 and b3 are loss constants of the generator.
3. The system of claim 2, wherein the first genetic algorithm employs a chromosome matrix of P x ＝[P 1 code ，P 2 coding ]Wherein the P is 1 code 、P 2 coding Respectively P 1 、P 2 Is a vectorized representation of (2);
the crossover operator according to the first genetic algorithm is P x (t+1)＝P great +c1×P great -c2×P x (t); wherein t is the iteration number and P is the number of iterations great For a chromosome matrix that minimizes the F (P) value in t iterations, c1, c2 are random vectors.
4. The system of claim 3, wherein the mutation operator according to the first genetic algorithm is P x (t+1)＝P x (t) + (∈θ), where e is a random vector,γ＝2F(P x (t))-F(P x (t-1))。
5. the system of claim 4, wherein the first prediction model comprises a first neural network algorithm and a second neural network algorithm, and wherein the unit power prediction module is specifically configured to:
inputting historical data of different time periods of the same day into the first neural network to obtainFirst predicted maximum output power P 1a 、P 2a Inputting historical data of first time periods with different dates into the second neural network to obtain second predicted maximum output power P 1b 、P 2b ；
Weighting the first predicted maximum output power and the second predicted maximum output power to obtain the P 1max 、P 2max 。
6. The system of claim 5, wherein the loss weight w1 of the first genset is greater than the loss weight w2 of the second genset.
7. The system of claim 1, wherein the discrimination module is further configured to:
selecting a second genetic algorithm as a global scheduling algorithm when the at least two generator sets are identified to include a third generator set; wherein the complexity of the second genetic algorithm is less than the complexity of the first genetic algorithm, the complexity being indicative of the time taken for the operation.
8. The system of claim 7, wherein the weight distribution module comprises:
the monitoring module is used for monitoring the weight setting instruction;
the verification module is used for reading instruction verification information when the weight setting instruction is obtained, and verifying the weight setting instruction based on the instruction verification information;
the analyzing module is used for analyzing the weight setting instruction when the weight setting instruction passes the verification, and obtaining the loss weight w1 of the first generator set and the loss weight w2 of the second generator set.
9. The system of claim 5, wherein the scheduling module is further configured to:
transmitting a scheduling instruction to the at least two generator sets, wherein the scheduling instruction is used for the first generator setSetting the output power of the first generator set to P for a period of time 1 anticipation of Setting the output power of the second generator set to P 2 anticipation of 。
10. A distributed power supply method based on an intelligent energy management platform is used for a distributed power supply system, and the distributed power supply system comprises at least two generator sets and an energy management platform; the at least two generator sets comprise a first generator set and a second generator set, and the generator types of the at least two generator sets are different; the energy management platform comprises a power prediction module, a weight distribution module, a discrimination module and a scheduling module, wherein the power prediction module comprises a global power prediction module and a unit power prediction module, and the method is characterized by comprising the following steps:
calculating, by the global power prediction module, expected usage power P for a first period of time of the power supply region based on a linear regression equation Total (S) ；
Calculating, by a unit power prediction module, a maximum unit output power P of a first generator unit during the first period of time based on a first prediction model 1max Maximum unit output power P of second generator unit in the first time period 2max The method comprises the steps of carrying out a first treatment on the surface of the The algorithm according to which the first prediction model is based is a neural network algorithm;
the loss weight w1 of the first generator set and the loss weight w2 of the second generator set are set through a weight distribution module;
when the at least two generator sets are identified to not comprise a third generator set, a first genetic algorithm is selected as a global scheduling algorithm through a judging module; the type of the generator of the third generator set is different from that of the first generator set and the second generator set;
calculating, by a scheduling module, an expected power P of the first generator set over a first period of time, in accordance with the first genetic algorithm, such that an fitness function according to the first genetic algorithm is at a minimum 1 anticipation of And the expected power P of the second generator set in the first time period 2 anticipation of ；
The saidThe fitness function according to the first genetic algorithm is F (P) = (w1×C) 1 (P 1 )+w2×C 2 (P 2 )+|w0(P 1 +P 2 -P Total (S) ) I), said C 1 (P 1 )、C 2 (P 2 ) The loss functions of the first generator set and the second generator set are respectively P 1 、P 2 The output power of the first generator set and the output power of the second generator set are respectively; the w0 is a power fluctuation loss weight and is used for adjusting the weight occupied by the loss of the power fluctuation to the system;
constraint conditions of the fitness function F (P) are as follows: p is more than or equal to 0 1 ≤P 1max ，0≤P 2 ≤P 2max 。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311724348.0A CN117767430A (en) | 2023-12-15 | 2023-12-15 | Distributed power supply system and method based on intelligent energy management platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311724348.0A CN117767430A (en) | 2023-12-15 | 2023-12-15 | Distributed power supply system and method based on intelligent energy management platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117767430A true CN117767430A (en) | 2024-03-26 |
Family
ID=90319119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311724348.0A Pending CN117767430A (en) | 2023-12-15 | 2023-12-15 | Distributed power supply system and method based on intelligent energy management platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117767430A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003199252A (en) * | 2001-12-27 | 2003-07-11 | Kansai Electric Power Co Inc:The | Distributed power supply system and control method therefor |
| CN102509175A (en) * | 2011-11-07 | 2012-06-20 | 上海电力学院 | Reliability optimization method of distributed power supply system |
| US20130176751A1 (en) * | 2012-01-05 | 2013-07-11 | Ingeteam Power Technology, S.A. | Method and apparatus for controlling a frequency converter |
| CN104218578A (en) * | 2014-09-15 | 2014-12-17 | 华北电力大学 | Method and device for planning distributed type power supply system |
| CN205544325U (en) * | 2016-02-02 | 2016-08-31 | 国网上海市电力公司 | Distributing type energy access system |
-
2023
- 2023-12-15 CN CN202311724348.0A patent/CN117767430A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003199252A (en) * | 2001-12-27 | 2003-07-11 | Kansai Electric Power Co Inc:The | Distributed power supply system and control method therefor |
| CN102509175A (en) * | 2011-11-07 | 2012-06-20 | 上海电力学院 | Reliability optimization method of distributed power supply system |
| US20130176751A1 (en) * | 2012-01-05 | 2013-07-11 | Ingeteam Power Technology, S.A. | Method and apparatus for controlling a frequency converter |
| CN104218578A (en) * | 2014-09-15 | 2014-12-17 | 华北电力大学 | Method and device for planning distributed type power supply system |
| CN205544325U (en) * | 2016-02-02 | 2016-08-31 | 国网上海市电力公司 | Distributing type energy access system |
Non-Patent Citations (1)
| Title |
|---|
| 朱灿元等: "考虑清洁能源与储能的分布式数据中心低碳调度策略", 《智慧电力》, vol. 51, no. 2, 28 February 2023 (2023-02-28), pages 16 - 23 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shafie-Khah et al. | An innovative two-level model for electric vehicle parking lots in distribution systems with renewable energy | |
| Soares et al. | Application-Specific Modified Particle Swarm Optimization for energy resource scheduling considering vehicle-to-grid | |
| Mei et al. | Game approaches for hybrid power system planning | |
| Hosseini et al. | A novel straightforward unit commitment method for large-scale power systems | |
| Kebriaei et al. | Short-term load forecasting with a new nonsymmetric penalty function | |
| CN112308427A (en) | New energy consumption restriction factor evaluation method and system based on combined empowerment-grey correlation | |
| Tao et al. | Joint optimization of wind turbine micrositing and cabling in an offshore wind farm | |
| CN113191575B (en) | Optimization method, device, terminal and storage medium for power distribution network maintenance power failure mode | |
| Su et al. | An integrated, systematic data-driven supply-demand side management method for smart integrated energy systems | |
| Javidsharifi et al. | Probabilistic model for microgrids optimal energy management considering AC network constraints | |
| Li et al. | Day-ahead risk averse market clearing considering demand response with data-driven load uncertainty representation: A Singapore electricity market study | |
| CN106505624A (en) | Determine regulator control system and the method for power distribution network distributed power source optimum ability to arrange jobs | |
| Zhang et al. | Multi‐objective economic dispatch of power system with wind farm considering flexible load response | |
| Memari et al. | Clustering-based reliability assessment of smart grids by fuzzy c-means algorithm considering direct cyber–physical interdependencies and system uncertainties | |
| Lu et al. | An interval prediction method for day-ahead electricity price in wholesale market considering weather factors | |
| CN113887809A (en) | Power distribution network supply and demand balance method, system, medium and computing equipment under double-carbon target | |
| Abdulwhab et al. | Maintenance scheduling optimization using a genetic algorithm (GA) with a probabilistic fitness function | |
| CN117767430A (en) | Distributed power supply system and method based on intelligent energy management platform | |
| Zhang et al. | Time-of-use pricing model considering wind power uncertainty | |
| Lu et al. | Effects of wind energy supplied by independent power producers on the generation dispatch of electric power utilities | |
| Coria et al. | Computational efficient approach to compute a prediction-of-use tariff for coordinating charging of plug-in electric vehicles under uncertainty | |
| Soroudi et al. | Multi-objective planning model for integration of distributed generations in deregulated power systems | |
| CN113112296B (en) | Typical scene library construction method and device for centralized spot market | |
| Zheng et al. | Data-Adaptive Robust Transmission Network Planning Incorporating Post-Contingency Demand Response | |
| Okachi et al. | LOAD FLOW EVALUATION OF NIGERIAN TRANSMISSION SYSTEM FOR IMPROVED STABILITY/PERFORMANCE USING GLOBAL-BEST ARTIFICIAL BEE COLONY (GABC) ALGORITHM; A CASE STUDY OF THE CALABAR 132/33KV TRANSMISSION/DISTRIBUTION NETWORK |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |