CN115146936A - Dynamic adjustment learning factor algorithm for solving cascade water-light storage complementary scheduling model - Google Patents
Dynamic adjustment learning factor algorithm for solving cascade water-light storage complementary scheduling model Download PDFInfo
- Publication number
- CN115146936A CN115146936A CN202210716402.6A CN202210716402A CN115146936A CN 115146936 A CN115146936 A CN 115146936A CN 202210716402 A CN202210716402 A CN 202210716402A CN 115146936 A CN115146936 A CN 115146936A
- Authority
- CN
- China
- Prior art keywords
- population
- individual
- learning factor
- dynamic adjustment
- optimal
- 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
- 230000013016 learning Effects 0.000 title claims abstract description 59
- 230000000295 complement effect Effects 0.000 title claims abstract description 53
- 238000005457 optimization Methods 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 8
- 238000004364 calculation method Methods 0.000 claims description 36
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 16
- 230000009286 beneficial effect Effects 0.000 claims description 5
- 150000001875 compounds Chemical class 0.000 claims description 3
- 230000009326 social learning Effects 0.000 claims description 3
- 230000002028 premature Effects 0.000 abstract description 3
- 238000010248 power generation Methods 0.000 description 7
- 229910052799 carbon Inorganic materials 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Feedback Control In General (AREA)
Abstract
The invention relates to a dynamic adjustment learning factor algorithm for solving a cascade water-light storage complementary scheduling model, which belongs to the technical field of hydropower and photovoltaic complementary operation, combines the difficult problem of high-dimensional nonlinear solving of the cascade water-light storage complementary scheduling model, and generates a one-dimensional array of decision variables through the dynamic adjustment learning factor intelligent algorithm to realize dimension reduction processing of complex high-dimensional problems of different power stations. The dynamic adjustment learning factor is introduced to effectively control the optimization process of the intelligent algorithm, and when the dynamic adjustment learning factor exceeds a certain range, variation factors with different variation probabilities are added to adjust, so that the premature convergence phenomenon is improved, the global optimal solution is effectively judged, and the individual self-adaptive global situation is realized. And performing coordination control on local searching capability and global convergence capability by calculating the global optimal fitness value of the population and updating all individual positions with different iteration times and combining with a dynamic adjustment learning factor, thereby realizing efficient solution of the cascade water-light storage complementary scheduling model.
Description
Technical Field
The invention belongs to the technical field of hydropower and photovoltaic complementary operation, and particularly relates to a dynamic adjustment learning factor algorithm for solving a cascade water-light storage complementary scheduling model.
Background
With the proposition of a vision target of carbon neutralization and carbon peak reaching, new energy sources such as wind power and photovoltaic in China develop rapidly, but the new energy source power generation is influenced by meteorological factors and has obvious randomness, volatility and intermittency, so that the large-scale grid-connected operation of the new energy source power generation is restricted. In order to promote grid-connected consumption of intermittent energy sources such as photovoltaic power generation, cascade hydropower stations and pumped storage power stations which are flexible in starting and high in adjusting speed are utilized to perform complementary adjustment on photovoltaic power generation, cascade water-light storage complementary coordinated operation is achieved, the grid-connected consumption level of clean water-light energy resources is improved, and the cascade water-light storage complementary coordinated operation becomes an important gripper for energy structure transformation and upgrading and novel power system construction in China. However, due to the complementary scheduling of the cascade hydropower stations, the pumped storage power stations and the photovoltaic power generation, the number of involved power stations is large, the uncertainty is strong, the influence factors are many and complex, the problem belongs to a high-dimensional nonlinear discontinuous feasible domain optimization solving problem, and the requirements on the calculation accuracy and the solving efficiency of a model solving algorithm are high. Therefore, in order to realize the cascade water-light storage multi-energy complementary cooperative scheduling operation, promote the large-scale development and grid-connected consumption of new energy, promote the construction of a novel electric power system and realize the double-carbon target, a dynamic adjustment learning factor intelligent algorithm for solving a cascade water-light storage complementary scheduling model is urgently required to be researched, the efficient intelligent solution of the cascade water-light storage complementary scheduling model is realized, and an important technical support is provided for the cascade water-light storage complementary scheduling and the integrated operation of a clean energy base.
Disclosure of Invention
The invention aims to provide a dynamic adjustment learning factor algorithm for solving a cascade water-light storage complementation scheduling model, which is used for solving the technical problems in the prior art, aiming at solving the high-dimensional nonlinear problem faced by the cascade water-electricity station group, a pumped storage power station and a photovoltaic power generation complementation scheduling model, the dynamic adjustment learning factor is introduced to coordinate the local searching capability and the global convergence capability of the solution algorithm, the convergence speed and the solution efficiency of the algorithm are improved, and the efficient solution of the cascade water-light storage complementation scheduling model is realized.
In order to realize the purpose, the technical scheme of the invention is as follows:
the dynamic adjustment learning factor algorithm for solving the cascade water-light storage complementary scheduling model comprises the following steps of:
s1, initializing basic parameters of an intelligent algorithm optimized population according to a cascade water light storage complementary scheduling model, and executing a step S2;
s2, determining initial optimal individuals and initial population optimal fitness values in the population, and executing the step S4;
s3, calculating a local optimal fitness value and a global optimal fitness value of the population in an iterative loop manner, and executing the step S4;
s4, calculating a dynamic adjustment learning factor, updating individual variation based on the dynamic adjustment learning factor, and executing the step S5;
and S5, updating the positions of all individuals in the population, returning to the step S3 again until the circulation ending condition is met, and outputting the calculation result of the cascade water light storage complementary scheduling optimization.
Further, in step S1, initializing basic parameters of an intelligent algorithm optimized population according to the step water light storage complementary scheduling model, including the following steps:
s11, determining the size of the population, determining the number of variables of each individual in the population, and executing the step S12;
s12, setting values of all constraint conditions and initial values of an objective function in the cascade water light storage complementary scheduling model, setting initial values of local optimal fitness and global optimal fitness, setting a cycle ending condition of an intelligent algorithm, and executing the step S13;
and S13, randomly generating an initial population within a variable threshold, namely generating a plurality of individuals.
Further, in step S2, determining the initial optimal individual and the initial population optimal fitness value in the population includes the following steps:
s21, determining an objective function of each individual according to an objective function calculation formula, and executing the step S22;
s22, determining the optimal fitness value of the initial population by sequentially comparing the fitness values of all individuals, and determining the initial optimal individuals in the population.
Further, in step S3, the iterative loop calculates the local optimal fitness value and the global optimal fitness value of the population, and includes the following steps:
s31, determining an updated target function of each individual position according to the updated position information of each individual and a target function calculation formula, and executing the step S32;
and S32, determining the optimal fitness value of the current population by sequentially comparing the fitness values of all the current individuals, and determining the optimal individuals in the current population.
Further, in the step S4, calculating a dynamic adjustment learning factor, and updating the individual variation based on the dynamic adjustment learning factor, includes the following steps:
s41, calculating a dynamic adjustment learning factor of the current population according to the fitness values of all individuals of the current population, and executing the step S42;
s42, judging according to the dynamic adjustment learning factor of the current population, if the dynamic adjustment learning factor is larger than 1, increasing a variation factor, adding random disturbance to the position of the globally optimal individual of the current population to form a new individual, calculating the fitness value of the formed new individual after disturbance, and executing the step S43;
s43, comparing the fitness value of the new individual with the global optimal individual fitness value before disturbance, and updating the global optimal individual and the global optimal fitness value if the value of the fitness value is superior to the global optimal individual fitness value before disturbance.
Further, in step S5, the positions of all individuals in the population are updated, and step S3 is returned again until the cycle end condition is satisfied, and a calculation result of the cascade water light storage complementary scheduling optimization is output, including the following steps:
s51, determining position updating parameters of the population, updating the position information of all individuals, and executing the step S52;
s52, judging whether a cycle ending condition is met or not according to global optimal solution comparison or cycle number calculation of two times of cycle calculation, if so, ending the cycle calculation and outputting a cascade water light storage complementary scheduling optimization calculation result; if not, the step S3 is continued.
Further, in step S41, the calculation formula of the dynamically adjusted learning factor of the current population is as follows:
in the formula, σ 2 Dynamically adjusting the learning factor; i is the serial number of the individual; m is the number of population individuals;
fitness value for each individual;
the average fitness value of all individuals in the population; f. of k And normalizing the factors for the fitness value of each individual in the population.
Further, in step S42, a variation factor is added, and random disturbance is added to the position of the globally optimal individual of the current population, so as to form a calculation formula of a new individual, where the calculation formula is:
θ～N(0，1)
in the formula (I), the compound is shown in the specification,
in the k iteration, the j dimension variable of the new individual is taken value after random disturbance updating; j is the variable dimension serial number of the individual;
taking the value of the j dimension variable of the global optimal individual during the k iteration; θ is a random perturbation factor, obeying the standard normal distribution N (0, 1).
Further, in step S51, a location update parameter of the population is determined, and a calculation formula for updating the location information of all the individuals is as follows:
in the formula:
when the iteration is the (k + 1) th iteration, the j dimension variable value of the ith new individual is updated; c. C 1 The self-learning factor reflects the influence of self memory of an individual, mainly influences local search and improves the search precision; c. C 2 The social learning factor reflects group historical experience of cooperative cooperation and knowledge sharing among individuals and is beneficial to global search; r is 1 And r 2 Are all [0,1 ]]Uniformly distributed random numbers are subjected to the interval;
is k times of overlappingIn the generation process, the j dimension variable value of the history optimal position of the ith individual is taken.
Compared with the prior art, the invention has the following beneficial effects:
one of the beneficial effects of the scheme is that the problem of solving the high-dimensional nonlinearity of the cascade water-light storage complementary scheduling model is combined, and the one-dimensional array of the decision variables is generated by dynamically adjusting the learning factor intelligent algorithm, so that the dimension reduction processing of the complex high-dimensional problems of different power stations is realized. The dynamic adjustment learning factors are introduced to effectively control the optimization process of the intelligent algorithm, and when the dynamic adjustment learning factors exceed a certain range, variation factors with different variation probabilities are added to adjust, so that the premature convergence phenomenon is improved, the global optimal solution is effectively judged, and the individual self-adaption global situation is realized. And performing coordination control on local searching capability and global convergence capability by calculating the global optimal fitness value of the population and updating all individual positions with different iteration times and combining with a dynamic adjustment learning factor, thereby realizing efficient solution of the cascade water-light storage complementary scheduling model.
Drawings
Fig. 1 is a calculation flow chart of the intelligent algorithm for dynamically adjusting the learning factor to solve the cascade water light storage complementation scheduling model.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below with reference to fig. 1 of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Due to the fact that cascade hydropower stations, pumped storage power stations and photovoltaic power generation are complementarily scheduled, the number of involved power stations is large, uncertainty is strong, influence factors are large and complex, the problem belongs to a high-dimensional nonlinear discontinuous feasible domain optimization solving problem, and the requirements on the calculation accuracy and the solving efficiency of a model solving algorithm are high. Therefore, in order to realize the cascade water-light storage multi-energy complementary cooperative scheduling operation, promote the large-scale development and grid-connected consumption of new energy, boost the construction of a novel power system and realize the double-carbon target, a dynamic adjustment learning factor intelligent algorithm for solving a cascade water-light storage complementary scheduling model is urgently needed to be researched, the efficient intelligent solution of the cascade water-light storage complementary scheduling model is realized, and an important technical support is provided for the cascade water-light storage complementary scheduling and the integrated operation of a clean energy base.
Example (b):
as shown in fig. 1, an intelligent algorithm for dynamically adjusting learning factor for solving a cascade water-light-storage complementary scheduling model is provided, which comprises the following steps:
s1, initializing basic parameters of an intelligent algorithm optimized population according to a cascade water light storage complementary scheduling model, and executing the step S2:
s2, determining initial optimal individuals and initial population optimal fitness values in the population, and executing the step S4;
s3, calculating the local optimal fitness value and the global optimal fitness value of the population in an iterative loop manner, and executing the step S4;
s4, calculating a dynamic adjustment learning factor, updating individual variation based on the dynamic adjustment learning factor, and executing the step S5;
and S5, updating the positions of all individuals in the population, returning to the step S3 again until the circulation ending condition is met, and outputting the calculation result of the cascade water light storage complementary scheduling optimization.
Further, in step S1, initializing basic parameters of an intelligent algorithm optimized population according to the step water light storage complementary scheduling model, including the following steps:
s11, determining the size of the population (namely the number of population individuals), determining the variable number of each individual in the population, and executing the step S12;
s12, setting values of all constraint conditions and initial values of an objective function in the cascade water light storage complementary scheduling model, setting initial values of local optimal fitness and global optimal fitness, setting a cycle ending condition of an intelligent algorithm, and executing the step S13;
and S13, randomly generating an initial population (initial position) within the variable threshold, namely generating a plurality of individuals (namely a one-dimensional array formed by decision variable values at different moments).
Further, in step S2, determining the initial optimal individuals and the initial population optimal fitness value in the population includes the following steps:
s21, determining an objective function (namely, the fitness value of each individual) of each individual according to an objective function calculation formula, and executing the step S22;
s22, determining the optimal fitness value of the initial population by sequentially comparing the fitness values of all individuals, and determining the initial optimal individuals in the population.
Further, in step S3, the iterative loop calculates the local optimal fitness value and the global optimal fitness value of the population, and includes the following steps:
s31, determining an updated target function (namely the fitness value of the individual) of each individual position according to the updated position information of each individual and a target function calculation formula, and executing the step S32;
and S32, determining the optimal fitness value of the current population by sequentially comparing the fitness values of all the current individuals, and determining the optimal individuals in the current population.
Further, in the step S4, calculating a dynamic adjustment learning factor, and updating the individual variation based on the dynamic adjustment learning factor, includes the following steps:
s41, calculating a dynamic adjustment learning factor of the current population according to the fitness values of all individuals of the current population, and executing the step S42;
s42, judging according to the dynamic adjustment learning factor of the current population, if the dynamic adjustment learning factor is larger than 1, increasing a variation factor, adding random disturbance to the position of the globally optimal individual of the current population to form a new individual, calculating the fitness value of the formed new individual after disturbance, and executing the step S43;
s43, the fitness value of the new individual is compared with the global optimal individual fitness value before disturbance, and if the value of the fitness value is superior to the global optimal individual fitness value before disturbance, the global optimal individual and the global optimal fitness value are updated.
Further, in step S5, the positions of all individuals in the population are updated, and the step S3 is returned again until the cycle end condition is satisfied, and then the calculation result of the cascade water-light storage complementary scheduling optimization is output, including the following steps:
s51, determining position updating parameters of the population, updating the position information of all individuals, and executing the step S52;
s52, judging whether a cycle ending condition is met or not according to global optimal solution comparison or cycle number calculation of two times of cycle calculation, if so, ending the cycle calculation and outputting a cascade water light storage complementary scheduling optimization calculation result; if not, the step S3 is continued.
Further, in step S41, the calculation formula of the dynamically adjusted learning factor of the current population is as follows:
in the formula, σ 2 Dynamically adjusting the learning factor; i is the serial number of the individual; m is the number of population individuals;
fitness values (i.e., objective function values) for individual individuals;
the average fitness value of all individuals in the population; f. of k And normalizing the factors for the fitness value of each individual in the population.
Further, in step S42, a variation factor is added, and random disturbance is added to the position of the globally optimal individual of the current population, so as to form a new individual according to the following calculation formula:
θ～N(0，1)
in the formula (I), the compound is shown in the specification,
in the k iteration, the j dimension variable of the new individual is taken value after random disturbance updating; j is the variable dimension serial number of the individual;
taking the value of the j dimension variable of the global optimal individual during the k iteration; θ is a random perturbation factor, obeying the standard normal distribution N (0, 1).
Further, in step S51, a location update parameter of the population is determined, and a calculation formula for updating the location information of all the individuals is as follows:
in the formula:
when the iteration is the (k + 1) th iteration, the j dimension variable value of the ith new individual is updated; c. C 1 The self-learning factor reflects the influence of self memory of an individual, mainly influences local search and improves the search precision; c. C 2 The social learning factor reflects group historical experience of cooperative cooperation and knowledge sharing among individuals and is beneficial to global search; r is 1 And r 2 Are all [0,1]Uniformly distributed random numbers are subjected to the interval;
the value of the jth dimension variable of the history optimal position of the ith individual in the k iteration processes。
In conclusion, the problem is solved in a high-dimensional nonlinear way by combining the cascade water-light storage complementary scheduling model, and the one-dimensional array of decision variables is generated by dynamically adjusting the learning factor intelligent algorithm, so that the dimension reduction processing of the complex high-dimensional problems of different power stations is realized. The dynamic adjustment learning factor is introduced to effectively control the optimization process of the intelligent algorithm, and when the dynamic adjustment learning factor exceeds a certain range, variation factors with different variation probabilities are added to adjust, so that the premature convergence phenomenon is improved, the global optimal solution is effectively judged, and the individual self-adaptive global situation is realized. And performing coordination control on local searching capability and global convergence capability by calculating the global optimal fitness value of the population and updating all individual positions with different iteration times and combining with a dynamic adjustment learning factor, thereby realizing efficient solution of the cascade water-light storage complementary scheduling model.
The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.
Claims (9)
1. The dynamic adjustment learning factor algorithm for solving the cascade water-light storage complementary scheduling model is characterized by comprising the following steps of:
s1, initializing basic parameters of an intelligent algorithm optimized population according to a cascade water light storage complementary scheduling model, and executing a step S2;
s2, determining initial optimal individuals and initial population optimal fitness values in the population, and executing the step S4;
s3, calculating a local optimal fitness value and a global optimal fitness value of the population in an iterative loop manner, and executing the step S4;
s4, calculating a dynamic adjustment learning factor, updating individual variation based on the dynamic adjustment learning factor, and executing the step S5;
and S5, updating the positions of all individuals in the population, returning to the step S3 again until the circulation ending condition is met, and outputting the calculation result of the cascade water light storage complementary scheduling optimization.
2. The algorithm for dynamically adjusting learning factors solved by the cascaded water-light-storage complementary scheduling model according to claim 1, wherein in the step S1, the method for initializing the basic parameters of the intelligent algorithm optimized population according to the cascaded water-light-storage complementary scheduling model comprises the following steps:
s11, determining the size of the population, determining the variable number of each individual in the population, and executing the step S12;
s12, setting values of all constraint conditions and initial values of an objective function in the cascade water light storage complementary scheduling model, setting initial values of local optimal fitness and global optimal fitness, setting a cycle ending condition of an intelligent algorithm, and executing the step S13;
and S13, randomly generating an initial population within a variable threshold, namely generating a plurality of individuals.
3. The dynamic adjustment learning factor algorithm for the stepped water light storage complementary scheduling model solution according to claim 2, wherein in the step S2, the initial optimal individual and the initial population optimal fitness value in the population are determined, and the method comprises the following steps:
s21, determining an objective function of each individual according to an objective function calculation formula, and executing the step S22;
s22, determining the optimal fitness value of the initial population by sequentially comparing the fitness values of all individuals, and determining the initial optimal individuals in the population.
4. The algorithm of the learning factor for dynamic adjustment of the gradient water light storage complementary scheduling model solution according to claim 3, wherein in the step S3, the iterative loop calculates the local optimal fitness value and the global optimal fitness value of the population, and comprises the following steps:
s31, determining an updated target function of the position of each individual according to the updated position information of each individual and a target function calculation formula, and executing the step S32;
and S32, determining the optimal fitness value of the current population by sequentially comparing the fitness values of all the current individuals, and determining the optimal individuals in the current population.
5. The algorithm of the dynamically adjusted learning factor for solving the cascaded water-light-storage complementary scheduling model according to claim 4, wherein in the step S4, the dynamically adjusted learning factor is calculated, and the individual variation is updated based on the dynamically adjusted learning factor, and the algorithm comprises the following steps:
s41, calculating a dynamic adjustment learning factor of the current population according to the fitness values of all individuals of the current population, and executing the step S42;
s42, judging according to the dynamic adjustment learning factor of the current population, if the dynamic adjustment learning factor is larger than 1, increasing a variation factor, adding random disturbance to the position of the globally optimal individual of the current population to form a new individual, calculating the fitness value of the formed new individual after disturbance, and executing the step S43;
s43, comparing the fitness value of the new individual with the global optimal individual fitness value before disturbance, and updating the global optimal individual and the global optimal fitness value if the value of the fitness value is superior to the global optimal individual fitness value before disturbance.
6. The algorithm of learning factor for dynamic adjustment of the stepped water light storage complementary scheduling model solution according to claim 5, wherein in the step S5, the positions of all individuals in the population are updated, and the step S3 is returned again until a cycle end condition is satisfied, and then a calculation result of the stepped water light storage complementary scheduling optimization is output, including the following steps:
s51, determining position updating parameters of the population, updating the position information of all individuals, and executing the step S52;
s52, judging whether a cycle ending condition is met or not according to global optimal solution comparison or cycle number calculation of two times of cycle calculation, if so, ending the cycle calculation and outputting a cascade water light storage complementary scheduling optimization calculation result; if not, the step S3 is continued.
7. The algorithm of the learning factor for dynamic adjustment of the cascaded water-light-storage complementary scheduling model according to claim 6, wherein in step S41, the formula for the learning factor for dynamic adjustment of the current population is:
in the formula, σ 2 Dynamically adjusting the learning factor; i is the serial number of the individual; m is the number of population individuals;
fitness value for each individual;
the average fitness value of all individuals in the population is obtained; f. of k And normalizing the factors for the fitness value of each individual in the population.
8. The algorithm of the learning factor for dynamic adjustment of the cascade water light storage complementary scheduling model solution according to claim 7, wherein in step S42, a variation factor is added, and random disturbance is added to the position of the globally optimal individual of the current population, so as to form a new individual according to a calculation formula:
θ～N(0，1)
in the formula (I), the compound is shown in the specification,
in the k iteration, the j dimension variable of the new individual is taken value after random disturbance updating; j is the variable dimension serial number of the individual;
taking the value of the j dimension variable of the global optimal individual during the k iteration; θ is a random perturbation factor, obeying the standard normal distribution N (0, 1).
9. The algorithm of learning factor for dynamic adjustment of cascaded water-light-storage complementary scheduling model solution according to claim 8, wherein in step S51, a location update parameter of a population is determined, and a calculation formula for updating location information of all individuals is:
in the formula:
when the iteration is the (k + 1) th iteration, the j dimension variable value of the ith new individual is updated; c. C 1 The self-learning factor reflects the influence of self-memory of an individual, mainly influences local search and improves the search precision; c. C 2 The social learning factor reflects group historical experience of cooperative cooperation and knowledge sharing among individuals and is beneficial to global search; r is 1 And r 2 Are all [0,1]Uniformly distributed random numbers are subjected to the interval;
and taking the value of the jth dimension variable of the history optimal position of the ith individual in the k iteration processes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210716402.6A CN115146936A (en) | 2022-06-22 | 2022-06-22 | Dynamic adjustment learning factor algorithm for solving cascade water-light storage complementary scheduling model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210716402.6A CN115146936A (en) | 2022-06-22 | 2022-06-22 | Dynamic adjustment learning factor algorithm for solving cascade water-light storage complementary scheduling model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115146936A true CN115146936A (en) | 2022-10-04 |
Family
ID=83408287
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210716402.6A Pending CN115146936A (en) | 2022-06-22 | 2022-06-22 | Dynamic adjustment learning factor algorithm for solving cascade water-light storage complementary scheduling model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115146936A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117877749A (en) * | 2024-03-11 | 2024-04-12 | 西安交通大学医学院第一附属医院 | Atrial fibrillation patient anticoagulation treatment effect evaluation and optimization method based on big data |
-
2022
- 2022-06-22 CN CN202210716402.6A patent/CN115146936A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117877749A (en) * | 2024-03-11 | 2024-04-12 | 西安交通大学医学院第一附属医院 | Atrial fibrillation patient anticoagulation treatment effect evaluation and optimization method based on big data |
| CN117877749B (en) * | 2024-03-11 | 2024-05-28 | 西安交通大学医学院第一附属医院 | Atrial fibrillation patient anticoagulation treatment effect evaluation and optimization method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108365608B (en) | Uncertain optimization scheduling method and system for regional energy Internet | |
| CN109615146B (en) | Ultra-short-term wind power prediction method based on deep learning | |
| CN113342124B (en) | Photovoltaic MPPT method based on improved wolf optimization algorithm | |
| CN108491983B (en) | Large-scale unit combination optimization method for multi-stack fuel cell power generation system | |
| CN104299173A (en) | Robust optimization day-ahead scheduling method suitable for multi-energy-source connection | |
| CN108493992A (en) | A kind of wind power plant Optimization Scheduling of the controller containing Distributed Power Flow | |
| CN113052389A (en) | Distributed photovoltaic power station ultra-short-term power prediction method and system based on multiple tasks | |
| CN115146936A (en) | Dynamic adjustment learning factor algorithm for solving cascade water-light storage complementary scheduling model | |
| CN110336285B (en) | Optimal economic power flow calculation method for power system | |
| CN113255138B (en) | Load distribution optimization method for power system | |
| CN113471989B (en) | Intelligent micro-grid secondary frequency control method based on wolf optimization method | |
| CN117638939A (en) | Water-light complementary optimization scheduling method based on Adam algorithm consideration | |
| CN113205228A (en) | Method for predicting short-term wind power generation output power | |
| CN116485139A (en) | Short-term photovoltaic power generation amount prediction method based on multi-feature fusion | |
| CN115271254A (en) | Short-term wind power prediction method for optimizing extreme learning machine based on gull algorithm | |
| CN111061331A (en) | Photovoltaic maximum power control system and method | |
| CN116128211A (en) | Wind-light-water combined short-term optimization scheduling method based on wind-light uncertainty prediction scene | |
| CN109217357A (en) | A kind of grid-connected photovoltaic system MPPT method based on Markov model | |
| CN115543005A (en) | Photovoltaic maximum power tracking control method based on differential evolution slime mold algorithm | |
| CN111553398B (en) | Wind power scene uncertain continuous interval obtaining method based on multidimensional normal distribution | |
| CN114626622A (en) | Photovoltaic power prediction method, system, device and storage medium | |
| CN114400660A (en) | Method and system for determining proportion of wind power photovoltaic hydrogen production system | |
| CN110717694B (en) | Energy storage configuration random decision method and device based on new energy consumption expected value | |
| CN114580298A (en) | RBF-ISBO-based microgrid optimization scheduling method and system | |
| CN112152267B (en) | Power grid random reactive power optimization scheduling method considering source load uncertainty |
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 |