CN113076696A - Load short-term prediction method and prediction system based on IPSO-chaotic BP network - Google Patents
Load short-term prediction method and prediction system based on IPSO-chaotic BP network Download PDFInfo
- Publication number
- CN113076696A CN113076696A CN202110392956.0A CN202110392956A CN113076696A CN 113076696 A CN113076696 A CN 113076696A CN 202110392956 A CN202110392956 A CN 202110392956A CN 113076696 A CN113076696 A CN 113076696A
- Authority
- CN
- China
- Prior art keywords
- chaotic
- network
- ipso
- power load
- term prediction
- 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 43
- 239000002245 particle Substances 0.000 claims abstract description 53
- 238000005457 optimization Methods 0.000 claims abstract description 29
- 230000000739 chaotic effect Effects 0.000 claims abstract description 28
- 230000001133 acceleration Effects 0.000 claims abstract description 27
- 238000013528 artificial neural network Methods 0.000 claims abstract description 13
- 238000005312 nonlinear dynamic Methods 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 18
- 238000012937 correction Methods 0.000 claims description 8
- 230000001105 regulatory effect Effects 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 description 8
- 230000003044 adaptive effect Effects 0.000 description 7
- 230000007547 defect Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000013213 extrapolation Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Water Supply & Treatment (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a load short-term prediction method and a prediction system based on an IPSO-chaotic BP network, wherein the method comprises the steps of firstly obtaining actual historical power load data of a to-be-detected place, then inputting the power load data into a power load short-term prediction model of the IPSO-chaotic BP network, and obtaining power load prediction results of all integral point moments of a place prediction day, wherein the power load short-term prediction model is based on the chaotic BP neural network, a weight value and a threshold value in the chaotic BP neural network are obtained through improving particle swarm optimization, and the improved particle swarm adopts a nonlinear dynamic improved inertia weight strategy and an acceleration coefficient linear adjustment strategy which introduce an inverse sine function. The method can effectively improve the optimization speed and the search performance of the algorithm, reduce the operation time of the algorithm, and improve the accuracy of short-term power load prediction, thereby reducing the operation cost of the power grid.
Description
Technical Field
The invention relates to the technical field of electric power information processing, in particular to a load short-term prediction method and a load short-term prediction system based on an IPSO-chaotic BP network.
Background
In recent years, the demand for electric energy has been increasing, and higher demands have been made on the quality of electric energy. With increasing challenges, power system load prediction becomes an important part of power system planning and power equipment selection work. Under the current situation, because the electric energy cannot be stored in a large amount, the electric energy generated by the system must be dynamically balanced with the load change of the system, otherwise, the power supply quality is affected, and the safe and stable operation of the system is threatened. If the predicted value of the load is low, the power supply is short, the requirement of a user cannot be met, and the power supply reliability is greatly reduced; if the predicted value of the load is higher, the operation efficiency of the power generation, transmission and transformation equipment after being put into the system is reduced, and the economic index of the system is seriously influenced. Therefore, the more accurate load prediction can reasonably arrange the start and stop of the power plant unit, ensure the safe and stable operation of the power system, reduce the occurrence of accidents, improve the energy utilization rate, effectively reduce the power generation cost and improve the economic benefit and the social benefit.
The power load prediction may be classified into long-term, medium-term, and short-term load prediction. The results of the long and medium term load forecasting can determine the construction of a future power plant, the installation of equipment, the size, the place and the time of unit capacity, the capacity expansion and reconstruction of a power grid and the construction and development of the power grid. The short-term load prediction generally refers to predicting the load condition of one day to one week in the future, and the purpose of the short-term load prediction is to meet the accuracy requirement of the power system production on the demand prediction value to the maximum extent. With the continued sophistication of the electricity market, low-error short-term power load forecasting reduces waste.
The research on the short-term load prediction of the power system has a long history, and a plurality of experts and scholars at home and abroad have not been interrupted and have made many progress through continuous research on the load prediction. More mature traditional methods include time series methods, regression analysis methods, trend extrapolation methods, and the like; until the end of the 80 s in the 20 th century, the expert system method is applied to the short-term load prediction of the power system, and a brand-new field is opened up for the prediction method; since the nineties, artificial networks have attracted extensive attention of researchers in the application of power system load prediction, and are continuously developed to achieve satisfactory effects; later, with the development of many methods for load prediction, such as: fuzzy theory, power system short-term load prediction based on point pattern matching and a support vector machine. However, although there are many prediction methods for the power load, and each has its own advantages and disadvantages and different applicable occasions, the accuracy of the overall prediction is not high.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a load short-term prediction method and a load short-term prediction system based on an IPSO-chaotic BP network, which can effectively improve the optimization speed and the search performance, reduce the operation time and improve the accuracy of short-term power load prediction so as to reduce the operation cost of a power grid.
The purpose of the invention can be realized by the following technical scheme:
the invention provides a load short-term prediction method based on an IPSO-chaotic BP network, which comprises the following steps:
step S1: acquiring actual historical power load data of a to-be-detected place;
step S2: inputting the power load data into a power load short-term prediction model based on an IPSO-chaotic BP network, and acquiring power load prediction results of each integral point moment of a prediction day of the to-be-detected place;
the power load short-term prediction model is based on a chaotic BP neural network, a weight and a threshold in the chaotic BP neural network are obtained through improved particle swarm optimization, and the improved particle swarm adopts a nonlinear dynamic improved inertia weight strategy and an acceleration coefficient linear adjustment strategy which introduce an arcsine function.
Preferably, in the nonlinear dynamic modified inertia weight strategy for introducing the arcsine function, the inertia weight ω (t) is expressed as:
in the formula, ω1And ω2The inertial weight at the beginning and end, respectively, t representing the current iteration number, tmaxRepresents the maximum iteration number of the population, gamma is a regulating factor, and rand () is a random value function between (0, 1).
Preferably, the range of the regulating factor gamma is 1.1-1.3.
Preferably, in the linear adjustment strategy for acceleration coefficient, the acceleration coefficient c1And c2The adjustment formula of (c) is expressed as:
in the formula, c11、c12Respectively representing the acceleration coefficients c1Values at the beginning and end, c21,c22Respectively representing the acceleration coefficients c2Values at the beginning and end, NiterRepresenting the current iteration number, and k is a correction constant.
Preferably, the correction constant k is in a range of 2.78 to 2.86.
The invention also provides a load short-term prediction system based on the IPSO-chaotic BP network, which comprises:
the historical data acquisition module is used for acquiring actual historical power load data of a to-be-detected place;
the automatic prediction module is used for inputting the power load data into a power load short-term prediction model based on an IPSO-chaotic BP network and acquiring power load prediction results of all integral points of the prediction day of the to-be-detected place;
the power load short-term prediction model is based on a chaotic BP neural network, a weight and a threshold in the chaotic BP neural network are obtained through improved particle swarm optimization, and the improved particle swarm adopts a nonlinear dynamic improved inertia weight strategy and an acceleration coefficient linear adjustment strategy which introduce an arcsine function.
Preferably, in the nonlinear dynamic modified inertia weight strategy for introducing the arcsine function, the inertia weight ω (t) is expressed as:
in the formula, ω1And ω2The inertial weight at the beginning and end, respectively, t representing the current iteration number, tmaxRepresents the maximum iteration number of the population, gamma is a regulating factor, and rand () is a random value function between (0, 1).
Preferably, in the linear adjustment strategy for acceleration coefficient, the acceleration coefficient c1And c2The adjustment formula of (c) is expressed as:
in the formula, c11、c12Respectively representing the acceleration coefficients c1Values at the beginning and end, c21,c22Respectively representing the acceleration coefficientsc2Values at the beginning and end, NiterRepresenting the current iteration number, and k is a correction constant.
The present invention also provides an electronic device comprising:
one or more processors, a memory, and one or more programs stored in the memory, the one or more programs including instructions for performing the load short-term prediction method based on the IPSO-chaotic BP network as described above.
The present invention also provides a computer readable storage medium containing one or more programs for execution by one or more processors of an electronic device, the one or more programs including instructions for performing the IPSO-chaotic BP network based load short term prediction method as described above.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention provides a chaos BP network short-term prediction method based on improved particle swarm optimization, which overcomes the defects that the overall search performance is poor and the true overall optimal solution cannot be searched in the iteration process of the standard particle swarm algorithm and the chaos BP network algorithm, keeps the advantages of simple parameter setting, easy operation and the like of the particle swarm algorithm, and has satisfactory fitness;
2. by improving the particle swarm optimization algorithm, the invention overcomes the defects of local optimization, low iterative convergence speed and the like of the existing algorithm, effectively improves the optimization speed and the search performance of the algorithm, and reduces the operation time of the algorithm;
3. the method effectively improves the accuracy of short-term power load prediction, thereby reducing the operation cost of the power grid.
4. The hybrid algorithm of the improved particle swarm optimization algorithm and the chaotic BP network, provided by the invention, overcomes the defects of low convergence rate and easiness in dispersion of the traditional standard particle swarm algorithm, and also overcomes the defects of low prediction precision and difficulty in data acquisition of the chaotic learning algorithm.
Drawings
FIG. 1 shows E of three algorithmsgComparing values;
FIG. 2 is a comparison of an actual load curve with the load curves of three algorithms;
FIG. 3 is a schematic flow chart of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
The embodiment provides a load short-term prediction method based on an IPSO-chaotic BP network, which comprises the following steps:
step S1: acquiring actual historical power load data of a to-be-detected place;
step S2: inputting the power load data into a power load short-term prediction model based on an IPSO-chaotic BP network, and obtaining a power load prediction result of each integral point moment of the site prediction day, wherein the power load short-term prediction model is based on the chaotic BP neural network, a weight and a threshold value in the chaotic BP neural network are obtained by improving particle swarm optimization, and the improved particle swarm adopts a nonlinear dynamic improved inertia weight strategy and an acceleration coefficient linear adjustment strategy which introduce an arcsine function.
Specifically, as shown in fig. 3, the training step of the power load prediction model includes:
step S01: initializing particle swarm population information and maximum allowable iteration number TmaxThe particle swarm information comprises the size of the particle swarm and the initial position X of the determined particlei,j(0) And speed and number of particles M;
step S02: respectively determining the global optimal position and the local search position of each particle by adopting an improved particle swarm optimization algorithm and through balanced global search and local search, and solving the weight and the threshold of the chaotic BP network;
step S03: calculating an individual history optimal adaptive value corresponding to each particle by using a chaotic BP network according to the weight and the threshold, and selecting the best individual history optimal adaptive value in the particle swarm as a population history optimal adaptive value;
step S04: calculating the adaptive value of each particle according to the fitness function of the particle;
step S05: judging whether the calculated particle adaptive value is larger than the individual historical optimal adaptive value and the population historical optimal adaptive value, if not, updating the position and the speed of the particle, and if so, returning to the step S02;
step S06: and judging whether a process ending condition is met, if so, stopping iteration and outputting a result, otherwise, returning to the step S02, wherein the process ending condition is that the difference between the hidden layer output value at the current moment and the hidden layer output value at the previous moment is smaller than the floating point relative precision.
The training process can effectively improve the optimization speed and the search performance of the algorithm, reduce the operation time of the algorithm, and improve the accuracy of short-term power load prediction, thereby reducing the operation cost of the power grid.
Specifically, in the improved particle swarm optimization algorithm, a nonlinear dynamic improved inertia weight strategy of an arcsine function is introduced, and the inertia weight ω (t) is:
in the formula, ω1And ω2The inertial weight at the beginning and end, respectively, t representing the current iteration number, tmaxRepresenting the maximum iteration number of the population; gamma is an adjusting factor, 1.1-1.3 can be selected, and rand () is a function randomly taking a value between (0, 1).
The improved inertia weight can ensure that the algorithm has good global optimization capability in the initial stage, can quickly find a range close to global optimization, has good local optimization performance in the fast search ending stage, and can more reliably solve a global optimal solution.
In the improved particle swarm optimization algorithm, an acceleration coefficient c is adjusted by adopting an acceleration coefficient linear adjustment strategy1And c2To ensure that the particles should search the entire solution space as much as possible during the initial stage of the iterative search, and to require that the particles be able to jump out during the final stage of the iterative searchA local extremum. The adjustment formula is expressed as:
in the formula, c11,c12Values at the beginning and end of the acceleration factor C1, C21,c22Values at the beginning and end of the acceleration factor C2, N, respectivelyiterRepresenting the current iteration number; k is a correction constant, and is preferably 2.78 to 2.86.
Except for the parameters of inertia weight omega (t) and acceleration coefficient c of the particle swarm optimization algorithm1And c2Besides the correction, the improved particle swarm optimization algorithm is fused with the chaotic BP neural network to make up for the deficiencies, so that the global optimization is realized, and the efficiency and the solving precision of the algorithm are improved. Output function x of chaotic neuron in chaotic BP networkj(t +1) is:
where t is the number of iterations, yi(t +1) is the internal state of the chaotic neuron.
The initial parameters of the model set in this embodiment are: c. C1=1.516,c21.798, number of groups N80, maximum number of iterations T max5000. Setting a 3-layer network, inputting 3-layer neurons in the layer, setting a hidden layer as 5 neurons, setting an output layer as 1 neuron, and outputting the load output of the prediction day. In order to verify the condition of the improved particle swarm optimized chaotic BP network model in short-term load prediction practical application, certain place load data is selected for test verification. Firstly, the data is normalized, the load data of the previous 30 days is used as a network training sample, the trained network is used as a verification model, and the load data of the next 30 days is used as a test numberAccording to the method, the effect of the model in the short-term load prediction process is verified. The power load of the whole 24h point in a certain day is respectively predicted by adopting a standard particle swarm algorithm, a chaotic BP network and the chaotic BP network algorithm for improving particle swarm optimization, the comparison condition of evaluation indexes is shown in table 1, and the load prediction result is shown in table 2.
TABLE 1 evaluation index E of three algorithmsgComparison table
As can be seen from Table 1, the chaos BP network algorithm for improving particle swarm optimization is trained for about 500 times EgE with value smaller than 3000 times of training of standard particle swarm algorithm and chaotic BP network algorithmgThe value is obtained. E of improving chaos BP network algorithm of particle swarm optimization in 1500 times of traininggThe value is also less than E when the standard particle swarm algorithm and the chaotic BP network algorithm are trained for 5000 timesgThe values of the time used for the former 75s and the latter 373s and 306s, respectively. Therefore, the hybrid algorithm for improving particle swarm optimization adopted by the method has obvious advantages in convergence and training speed. The average percent absolute error ratio of the predicted results of the three algorithms to the actual values is shown in fig. 1.
Comparing the prediction results and the relative errors in the table 2, the accuracy of the load prediction results obtained by training the chaos BP network algorithm optimized by the improved particle swarm for 400 times is obviously better than that obtained by training the chaos BP network algorithm by the standard particle swarm algorithm (400 times) and the chaos BP network algorithm (400 times), and the error fluctuation is small, so that the prediction method and the model adopted by the invention are obviously better than those of the other two algorithms in the aspects of prediction accuracy and speed.
TABLE 2 comparison of predicted results for three algorithms
Referring to fig. 2, in the present embodiment, the actual load curve and the predicted load curve are compared by using a standard particle swarm algorithm, a chaotic BP network algorithm, and a chaotic BP network algorithm optimized by improving a particle swarm, respectively. As can be seen from the figure, the load curve predicted by the chaos BP network algorithm for improving particle swarm optimization is closer to the actual load curve, which shows that the prediction effect of the hybrid algorithm is superior to that of the standard particle swarm algorithm and the chaos BP network algorithm. As can be seen from fig. 1 and 2, the relative error of the prediction result of the chaos BP network algorithm for improving particle swarm optimization is controlled within 1.0%, and the error fluctuation is small.
The method provided by the invention overcomes the defects of low convergence speed and easiness in dispersion of the traditional standard particle swarm algorithm, overcomes the defects of low prediction precision and difficulty in data acquisition of the chaotic learning algorithm, and can establish different improved hybrid prediction models according to different daily load characteristics of working days and rest days. Practical application shows that compared with the traditional prediction method, the prediction method is an effective load prediction method.
The above functions, if implemented in the form of software functional units and sold or used as a separate product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another embodiment, an electronic device is provided that includes one or more processors, memory, and one or more programs stored in the memory, the one or more programs including instructions for performing the IPSO-chaotic BP network based load short term prediction method.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (10)
1. A load short-term prediction method based on an IPSO-chaotic BP network is characterized by comprising the following steps:
step S1: acquiring actual historical power load data of a to-be-detected place;
step S2: inputting the power load data into a power load short-term prediction model based on an IPSO-chaotic BP network, and acquiring power load prediction results of each integral point moment of a prediction day of the to-be-detected place;
the power load short-term prediction model is based on a chaotic BP neural network, a weight and a threshold in the chaotic BP neural network are obtained through improved particle swarm optimization, and the improved particle swarm adopts a nonlinear dynamic improved inertia weight strategy and an acceleration coefficient linear adjustment strategy which introduce an arcsine function.
2. The IPSO-chaotic BP network-based load short-term prediction method according to claim 1, wherein in the nonlinear dynamic modified inertia weight strategy introducing the arcsine function, an inertia weight ω (t) is expressed as:
in the formula, ω1And ω2The inertial weight at the beginning and end, respectively, t representing the current iteration number, tmaxRepresents the maximum iteration number of the population, gamma is a regulating factor, and rand () is a random value function between (0, 1).
3. The IPSO-chaotic BP network-based load short-term prediction method according to claim 2, wherein the range of the adjustment factor gamma is 1.1-1.3.
4. The IPSO-chaotic BP network-based load short-term prediction method according to claim 1, wherein in the acceleration coefficient linear adjustment strategy, an acceleration coefficient c1And c2The adjustment formula of (c) is expressed as:
in the formula, c11、c12Respectively representing the acceleration coefficients c1Values at the beginning and end, c21,c22Respectively representing the acceleration coefficients c2Values at the beginning and end, NiterRepresenting the current iteration number, and k is a correction constant.
5. The load short-term prediction method based on the IPSO-chaotic BP network according to claim 4, wherein the range of the correction constant k is 2.78-2.86.
6. A load short-term prediction system based on an IPSO-chaotic BP network is characterized by comprising the following components:
the historical data acquisition module is used for acquiring actual historical power load data of a to-be-detected place;
the automatic prediction module is used for inputting the power load data into a power load short-term prediction model based on an IPSO-chaotic BP network and acquiring power load prediction results of all integral points of the prediction day of the to-be-detected place;
the power load short-term prediction model is based on a chaotic BP neural network, a weight and a threshold in the chaotic BP neural network are obtained through improved particle swarm optimization, and the improved particle swarm adopts a nonlinear dynamic improved inertia weight strategy and an acceleration coefficient linear adjustment strategy which introduce an arcsine function.
7. The IPSO-chaotic BP network-based load short-term prediction system according to claim 6, wherein in the nonlinear dynamic modified inertia weight strategy introducing the arcsine function, the inertia weight ω (t) is expressed as:
in the formula, ω1And ω2The inertial weight at the beginning and end, respectively, t representing the current iteration number, tmaxRepresents the maximum iteration number of the population, gamma is a regulating factor, and rand () is a random value function between (0, 1).
8. The IPSO-chaotic BP network-based load short-term prediction system of claim 6, wherein in the acceleration coefficient linear adjustment strategy, an acceleration coefficient c1And c2The adjustment formula of (c) is expressed as:
in the formula, c11、c12Respectively representing the acceleration coefficients c1Values at the beginning and end, c21,c22Respectively representing the acceleration coefficients c2Values at the beginning and end, NiterRepresenting the current iteration number, and k is a correction constant.
9. An electronic device, comprising:
one or more processors, a memory, and one or more programs stored in the memory, the one or more programs including instructions for performing the IPSO-chaotic BP network-based load short-term prediction method of any one of claims 1-5.
10. A computer-readable storage medium comprising one or more programs for execution by one or more processors of an electronic device, the one or more programs comprising instructions for performing the IPSO-chaotic BP network based load short term prediction method of any of claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110392956.0A CN113076696A (en) | 2021-04-13 | 2021-04-13 | Load short-term prediction method and prediction system based on IPSO-chaotic BP network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110392956.0A CN113076696A (en) | 2021-04-13 | 2021-04-13 | Load short-term prediction method and prediction system based on IPSO-chaotic BP network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113076696A true CN113076696A (en) | 2021-07-06 |
Family
ID=76617363
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110392956.0A Pending CN113076696A (en) | 2021-04-13 | 2021-04-13 | Load short-term prediction method and prediction system based on IPSO-chaotic BP network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113076696A (en) |
-
2021
- 2021-04-13 CN CN202110392956.0A patent/CN113076696A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200161867A1 (en) | Method, system and storage medium for load dispatch optimization for residential microgrid | |
CN105846461B (en) | Control method and system for large-scale energy storage power station self-adaptive dynamic planning | |
Ge et al. | A hybrid model for short-term PV output forecasting based on PCA-GWO-GRNN | |
CN111738512A (en) | Short-term power load prediction method based on CNN-IPSO-GRU hybrid model | |
CN105896575B (en) | Hundred megawatt energy storage power control method and system based on self-adaptive dynamic programming | |
CN110866633B (en) | Micro-grid ultra-short-term load prediction method based on SVR support vector regression | |
CN112418496B (en) | Power distribution station energy storage configuration method based on deep learning | |
CN109978268A (en) | A kind of short-term load forecasting method, system and relevant apparatus | |
Zhang et al. | Deep reinforcement learning for load shedding against short-term voltage instability in large power systems | |
CN114400712A (en) | Improved second-order particle swarm algorithm-based micro-grid cluster optimization scheduling method | |
CN112072643A (en) | Light-storage system online scheduling method based on depth certainty gradient strategy | |
CN115795992A (en) | Park energy Internet online scheduling method based on virtual deduction of operation situation | |
CN117057228A (en) | Inverter multi-objective optimization method based on deep reinforcement learning | |
CN115345380A (en) | New energy consumption electric power scheduling method based on artificial intelligence | |
CN105119285B (en) | Multiobjective optimization control method is coordinated in wind storage based on dynamic weight index | |
CN113887794A (en) | Method and device for reactive power optimization of power distribution network | |
CN111799820A (en) | Double-layer intelligent hybrid zero-star cloud energy storage countermeasure regulation and control method for power system | |
CN113076696A (en) | Load short-term prediction method and prediction system based on IPSO-chaotic BP network | |
Pang et al. | A photovoltaic power predicting model using the differential evolution algorithm and multi-task learning | |
CN114298133A (en) | Short-term wind speed hybrid prediction method and device | |
Zhou et al. | Voltage regulation based on deep reinforcement learning algorithm in distribution network with energy storage system | |
CN111859780A (en) | Micro-grid operation optimization method and system | |
Yu et al. | Dynamic grouping control of electric vehicles based on improved k-means algorithm for wind power fluctuations suppression | |
CN113705067B (en) | Microgrid optimization operation strategy generation method, system, equipment and storage medium | |
CN115528687B (en) | Power system flexible response capability optimization method under limited cost constraint |
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 |