CN115882455B - Distributed photovoltaic power generation prediction method, system and terminal - Google Patents
Distributed photovoltaic power generation prediction method, system and terminal Download PDFInfo
- Publication number
- CN115882455B CN115882455B CN202310133174.4A CN202310133174A CN115882455B CN 115882455 B CN115882455 B CN 115882455B CN 202310133174 A CN202310133174 A CN 202310133174A CN 115882455 B CN115882455 B CN 115882455B
- Authority
- CN
- China
- Prior art keywords
- power generation
- photovoltaic power
- prediction
- data
- distributed photovoltaic
- 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.)
- Active
Links
Images
Classifications
-
- 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
Abstract
The invention provides a distributed photovoltaic power generation prediction method, a distributed photovoltaic power generation prediction system and a distributed photovoltaic power generation prediction terminal, which relate to the technical field of power prediction of a new energy power generation system, acquire photovoltaic power generation historical data, classify the photovoltaic power generation data of different areas by adopting a density peak clustering algorithm, and acquire photovoltaic power generation power types; according to the classified photovoltaic power generation power, analyzing factors influencing the photovoltaic output in different clusters, extracting characteristic factors, performing linear transformation on distributed photovoltaic power generation power data, and performing numerical processing on the extracted characteristic factors; and establishing a power prediction model according to the characteristic factors and the photovoltaic output magnitudes in different spaces and different time periods, and determining a final photovoltaic power generation power prediction result so as to determine a final planned power generation amount. According to the invention, characteristic factors of photovoltaic power generation prediction data are established, and the prediction is carried out by adopting the time-division side weight according to the power generation output conditions of different time periods, so that the control accuracy of the power generation process is improved.
Description
Technical Field
The invention relates to the technical field of power prediction of new energy power generation systems, in particular to a distributed photovoltaic power generation prediction method, a system and a terminal.
Background
In view of environmental protection and carbon emission, renewable energy power generation is adopted to replace traditional energy sources for energy conversion. Renewable energy power generation represented by photovoltaic power generation is rapidly developed at present, and the duty ratio in the installed capacity of an electric power system is gradually increased. As photovoltaic power generation is integrated into a power grid on a large scale, the randomness and fluctuation of the photovoltaic power generation may impact a main power grid, and the stable operation of a power system is affected. The prediction of the photovoltaic power generation power is beneficial to the scheduling department to make a reasonable power generation plan, is beneficial to the frequency modulation, voltage regulation, standby and the like of a power grid, and further ensures the safety and the economical efficiency of power supply.
After photovoltaic power generation, the photovoltaic power generation can be on-line and integrated into a power grid for comprehensive use. And the distributed photovoltaic power generation system needs to be predicted, so that the stability and the safety of the whole regional power grid work are ensured. For example, in the prior art, the application number is CN201510043728.7, which discloses a power prediction system of a distributed photovoltaic power generation system, where the file includes a weather data processing terminal, configured to receive and refine weather data, and obtain downscaled weather data; the distributed photovoltaic monitoring device is used for collecting operation data of the photovoltaic power generation system; the power prediction server is used for receiving the downscaled weather data and the photovoltaic power generation system operation data and performing power prediction on the photovoltaic power generation system by taking the downscaled weather data and the photovoltaic power generation system operation data as parameters. The file realizes the power prediction of the photovoltaic power generation system in the area through the meteorological data processing terminal, the distributed photovoltaic monitoring device and the power prediction server, thereby achieving the purpose of guaranteeing the working stability of the power grid in the whole area.
The document does not consider the characteristics of randomness, volatility and the like of the change of the photovoltaic power generation curve, and does not comprehensively predict the power generation condition of different periods, so that the problem that the power grid is possibly impacted by surfing the net after the photovoltaic power generation and the stable operation of the power system is affected is caused.
Disclosure of Invention
The distributed photovoltaic power generation prediction method provided by the invention considers space factors, and aims to solve the problems that the randomness and the fluctuation of photovoltaic power generation possibly impact a main power grid and influence the stable operation of a power system. The method can be used for carrying out distributed photovoltaic power generation process treatment by adopting time-division edge weights according to the power generation output conditions of different time periods, and improves the control accuracy of the power generation process.
The method comprises the following steps:
step 1: acquiring photovoltaic power generation historical data, and classifying the photovoltaic power generation data of different areas by adopting a density peak clustering algorithm to obtain photovoltaic power generation power types;
step 2: according to the classified photovoltaic power generation power, analyzing factors influencing the photovoltaic output in different clusters, extracting characteristic factors, performing linear transformation on distributed photovoltaic power generation power data, and performing numerical processing on the extracted characteristic factors;
step 3: and establishing a power prediction model according to the characteristic factors and the photovoltaic output magnitudes in different spaces and different time periods, and determining a final photovoltaic power generation power prediction result so as to determine a final planned power generation amount.
In step 1, the specific step of classifying the distributed photovoltaic power generation power data by using the density peak clustering algorithm includes:
extracting the distributed photovoltaic power generation condition;
constructing a clustering model, taking the extracted distributed photovoltaic power generation condition as the input of the clustering model, setting the number of clusters and an iteration stop threshold value, and initializing a clustering prototype mode matrix and an iteration counter;
iteratively calculating a partition matrix and a clustering prototype mode matrix;
judging whether the values of two iterations before and after the clustering prototype mode matrix are smaller than a set iteration stop threshold, if yes, stopping calculating, outputting the partition matrix and the clustering prototype mode matrix, otherwise, adding 1 to an iteration counter, and returning to continue to iteratively calculate the partition matrix and the clustering prototype mode matrix.
It should be further noted that, for any sample point in the data set
The density peak clustering algorithm calculates two variables, namely sample point +.>
Is>
And relative distance->
；
Sample point
Is>
The definition is as shown in the formula:
wherein, the liquid crystal display device comprises a liquid crystal display device,
for sample dot->
And->
European distance,/, of->
Is the cut-off distance;
when (when)
When (I)>
Otherwise->
；
Relative distance
For sample dot->
Distance to a point higher than its local density and closest to it;
relative distance
The formula of the calculation of (2) is as follows:
relative distance
And local Density->
The product of (2) is the complex variable +.>
。
It should be further noted that the method further includes the following steps:
(1) Acquiring a photovoltaic power generation historical data set and configuring the photovoltaic power generation historical data set as a sample classification number k;
(2) Calculating a distance matrix of the data set, and calculating a cut-off distance according to the percentage
；
(3) Calculating local density
Distance->
And the complex variables +.>
；
(4) Labeling k cluster centers:
(5) Categorizing non-clustered central data points:
(6) And judging the noise points, calculating an average local density upper bound for each class, and judging the noise points if the boundary point density of the class is lower than the average local density upper bound.
In the step 1, according to the classified photovoltaic power generation power, analyzing factors affecting the photovoltaic output in different categories, and extracting characteristic factors;
and calculating the parameter correlation in the distributed photovoltaic power generation power, performing linear transformation on the distributed photovoltaic power generation power data, and performing numerical treatment on the extracted characteristic factors.
And linearly transforming the distributed photovoltaic power generation power data by adopting a dispersion standardization method.
In step 3, according to the illumination duration and illumination intensity in different spaces, a comprehensive prediction model with variable weight in different time periods is established, different prediction algorithms are adopted in different time periods, and finally a prediction result is obtained through weighting.
It should be further noted that, the step of establishing the comprehensive prediction model with variable weights in time intervals according to the illumination duration and illumination intensity in different spaces further includes:
the period of photovoltaic power generation per day is defined as 6:00-18:00;
according to the generated power, the first period is as follows: 6:00-10: 00. the second period is: 10:00-14: 00. the third period is: 14:00-18:00;
the method comprises the steps of adopting a gray prediction method in a first period, adopting a support vector machine prediction algorithm in a second period, and adopting a neural network prediction method in a third period.
The invention also provides a distributed photovoltaic power generation prediction system, which comprises: the system comprises a data classification module, an analysis and extraction module and a prediction planning module;
the data classification module is used for acquiring photovoltaic power generation historical data, classifying the photovoltaic power generation data of different areas by adopting a density peak clustering algorithm, and obtaining photovoltaic power generation power types;
the analysis and extraction module analyzes factors influencing the photovoltaic output in different clusters according to the classified photovoltaic power generation power, extracts characteristic factors, performs linear transformation on distributed photovoltaic power generation power data, and performs numerical processing on the extracted characteristic factors;
and the prediction planning module establishes a power prediction model according to the characteristic factors and the photovoltaic output magnitudes of different spaces and different time periods, and determines a final photovoltaic power generation power prediction result so as to determine a final planned power generation amount.
The invention also provides a terminal, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of the distributed photovoltaic power generation prediction method.
From the above technical scheme, the invention has the following advantages:
in the distributed photovoltaic power generation prediction method and system provided by the invention, a distributed photovoltaic power generation prediction model considering space factors is established, and the distributed photovoltaic data with different characteristics is classified by adopting a DPC clustering method, so that characteristic factors of the photovoltaic power generation prediction data are established in a targeted manner. The invention also considers the time sequence characteristic of the photovoltaic power generation, and adopts a comprehensive prediction model of the time-interval side weight aiming at the power generation output condition of different time intervals, so that the prediction accuracy is higher; the method integrates the advantages of the traditional gray level prediction method and the support vector machine time sequence prediction, and the weights can be dynamically distributed to ensure the prediction precision.
According to the invention, not only is the influence of meteorological feature factors in different areas on the prediction of the photovoltaic power generation power considered, but also the weight of the predicted value in the final power generation amount is dynamically adjusted by establishing different weight prediction models, the photovoltaic power generation power is accurately predicted, the weight of the photovoltaic power generation amount in different time periods is synthesized, and the adaptability of the photovoltaic power generation prediction is effectively met. The method can also analyze and predict the distributed photovoltaic power generation state, define the daily period of photovoltaic power generation, analyze the change trend of the distributed photovoltaic power generation, evaluate whether the distributed photovoltaic power generation meets the requirement, and whether the power generation and the power supply bring risk to the power grid, thereby having a positive effect on guaranteeing the safety and the economy of the distributed photovoltaic power generation.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are needed in the description will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a distributed photovoltaic power generation prediction method;
fig. 2 is a schematic diagram of a distributed photovoltaic power generation prediction system.
Detailed Description
The distributed photovoltaic power generation prediction method and the system provided by the invention are used for accurately predicting the photovoltaic power generation by considering the influence of meteorological characteristic factors of different areas on the photovoltaic power generation prediction. The technology of the software layer in the photovoltaic power generation prediction method and the system mainly comprises the technologies of machine learning/deep learning, artificial neural network, confidence network, reinforcement learning, transfer learning, induction learning, teaching learning and the like. The programming languages involved in the photovoltaic power generation prediction methods and systems include, but are not limited to, object-oriented programming languages such as Java, smalltalk, C ++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer.
The distributed photovoltaic power generation prediction method and system provided by the invention utilize photovoltaic power generation historical data, combine a clustering algorithm and a power prediction model technology, utilize technologies such as time intervals and data processing of photovoltaic power generation per day, accurately predict photovoltaic power generation power, and synthesize the weight of photovoltaic power generation capacity in different time intervals, so that the adaptability of photovoltaic power generation prediction is effectively met.
Fig. 1 is a flow chart of a preferred embodiment of the distributed photovoltaic power generation prediction method of the present invention. The distributed photovoltaic power generation prediction method is applied to one or more terminal machines, wherein the terminal machines are equipment capable of automatically performing numerical calculation and/or information processing according to preset or stored instructions, and the hardware of the terminal machines comprises, but is not limited to, microprocessors, application specific integrated circuits (ApplicationSpecific Integrated Circuit, ASICs), programmable gate arrays (FPGAs), digital processors (Digital Signal Processor, DSPs), embedded equipment and the like.
The terminal may be any electronic product that can interact with a user, such as a personal computer, a tablet, a smart phone, a personal digital assistant (Personal Digital Assistant, PDA), an interactive internet protocol television (Internet Protocol Television, IPTV), a smart wearable device, etc.
The terminal may also include network devices and/or user devices. Wherein the network device includes, but is not limited to, a single network server, a server group composed of a plurality of network servers, or a Cloud based Cloud Computing (Cloud Computing) composed of a large number of hosts or network servers.
The network in which the terminal is located includes, but is not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (Virtual Private Network, VPN), and the like.
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a method for predicting distributed photovoltaic power generation in an embodiment is shown, where the method includes:
s101: acquiring photovoltaic power generation historical data, and classifying the photovoltaic power generation data of different areas by adopting a density peak clustering algorithm (DPC) to obtain photovoltaic power generation power types;
the embodiment of the invention also relates to the following specific steps:
(1) Extracting the conditions of distributed photovoltaic power generation power, including power generation capacity, peak value of photovoltaic power generation power and power generation average value of different periods in different areas;
(2) And constructing a clustering model, wherein the clustering model can be realized based on a DPC density peak clustering algorithm. The DPC density peak clustering algorithm is a clustering algorithm (clusteringby fast search and find of density peaks, DPC) based on fast search and discovery of density peaks. The algorithm can automatically find the cluster center, and realize efficient clustering of data in any shape.
The method takes the extracted distributed photovoltaic power generation power condition as the input of a clustering model, sets the number of clusters and an iteration stop threshold value, and initializes a clustering prototype mode matrix and an iteration counter;
in this embodiment, the number k of clusters is set to 15, the iteration stop threshold is set to 0.001, and the initialized cluster prototype mode matrix is assumed to be
An iteration counter->
；
(3) Calculating a distance matrix of the data set, and calculating a cut-off distance according to the percentage
Calculating local density
Distance->
And the complex variables +.>
；
(4) Labeling k cluster centers:
(5) Categorizing non-clustered central data points:
and judging the noise points, calculating an average local density upper bound for each class, and judging the noise points if the boundary point density of the class is lower than the average local density upper bound.
In one embodiment of the present invention, based on step S101, a possible embodiment thereof will be given below for non-limiting illustration.
In the embodiment of the method, factors influencing the photovoltaic output in different clusters are analyzed according to the classified photovoltaic power generation power, and characteristic factors are extracted.
According to different clusters and characteristic influence factors, different types of photovoltaic power generation output prediction models are established, multiple prediction algorithms are adopted for increasing reliability, and a final photovoltaic power generation prediction result is determined according to space factors, so that final planned power generation capacity is determined.
The density peak clustering algorithm (DPC) adopted by the invention specifically comprises the following steps of:
extracting the distributed photovoltaic power generation condition;
constructing a clustering model, taking the extracted distributed photovoltaic power generation condition as the input of the clustering model, setting the number of clusters and an iteration stop threshold value, and initializing a clustering prototype mode matrix and an iteration counter;
iteratively calculating a partition matrix and a clustering prototype mode matrix;
judging whether the values of two iterations before and after the clustering prototype mode matrix are smaller than a set iteration stop threshold, if yes, stopping calculating, outputting the partition matrix and the clustering prototype mode matrix, otherwise, adding 1 to an iteration counter, and returning to continue to iteratively calculate the partition matrix and the clustering prototype mode matrix.
For any sample point in the dataset
The DPC algorithm requires the calculation of two variables, namely sample point +.>
Is of the local density of (2)
And relative distance->
. Sample dot->
Is>
The definition is shown in the formula.
Wherein, the liquid crystal display device comprises a liquid crystal display device,
for sample dot->
And->
European distance,/, of->
For the cut-off distance, when->
When (I)>
Otherwise
. For different data sets, it is necessary to empirically set different +.>
。
Relative distance
For sample dot->
The distance to the closest point, which is higher than its local density, is determined by the following formula:
relative distance
And local Density->
The product of (2) is +.>
(complex variables).
The method comprises the following specific steps:
(1) Inputting a photovoltaic power generation historical data set and a sample classification number k;
(2) Calculating a distance matrix of the data set, and calculating a cut-off distance according to the percentage
；
(3) Calculating local density
Distance->
And the complex variables +.>
；
(4) Labeling k cluster centers:
(5) Categorizing non-clustered central data points:
(6) And judging the noise points, calculating an average local density upper bound for each class, and judging the noise points if the boundary point density of the class is lower than the average local density upper bound.
As some possible implementations, according to the classified photovoltaic power generation power classes, according to the classified photovoltaic power generation power, factors affecting photovoltaic output in different classes are analyzed, and feature factors are extracted.
The extraction method comprises the following steps: the photovoltaic output is greatly influenced by meteorological factors, namely irradiance, cloud movement and temperature and humidity, so that the change trend of photovoltaic power generation power curves in different areas is different. And (3) calculating correlation in the photovoltaic power generation, performing linear transformation on distributed photovoltaic power generation data, and performing numerical treatment on the extracted characteristic factors.
S102: according to the classified photovoltaic power generation power, analyzing factors influencing the photovoltaic output in different clusters, extracting characteristic factors, performing linear transformation on distributed photovoltaic power generation power data, and performing numerical processing on the extracted characteristic factors;
according to the classified photovoltaic power generation power class, analyzing the correlation of different meteorological factors in the photovoltaic power generation power class, extracting characteristic factors, performing linear transformation on distributed photovoltaic power generation power data, and performing numerical treatment on the extracted characteristic factors.
The method for acquiring the correlation of the meteorological factors in the photovoltaic power generation power class can extract characteristic factors and adopts a correlation coefficient analysis algorithm.
The linear transformation of the photovoltaic power generation power data adopts a dispersion normalization method.
In this example, the illumination intensity, illumination duration, illumination, temperature, wind speed, pressure, humidity are aimed at among the meteorological factors. Adopting a correlation coefficient analysis algorithm to analyze the correlation of different meteorological factors in each photovoltaic power generation power class, and extracting factors with the absolute value of the correlation coefficient exceeding 0.4 as characteristic factors; and (3) performing linear transformation on the original distributed photovoltaic power generation power data by using dispersion standardization to enable the characteristic factors to be digitized and fall into the [0,1] interval. The normalization formula is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,
representing transformed data, ++>
Representing pre-transformation data, < >>
Representing the minimum value in the transformed data set,
representing the maximum value in the transformed dataset.
S103: and establishing a power prediction model according to the characteristic factors and the photovoltaic output magnitudes in different spaces and different time periods, and determining a final photovoltaic power generation power prediction result so as to determine a final planned power generation amount.
And (3) establishing a power prediction model according to the characteristic factors in the S102 and the photovoltaic output magnitudes of different spaces and different periods, and determining a final photovoltaic power generation power prediction result.
The period of photovoltaic power generation per day is from 6:00-18:00, according to the generated power, the first period is: 6:00-10: 00. the second period is: 10:00-14: 00. the third period is: 14:00-18:00, adopting different prediction methods.
The method comprises the steps of adopting a gray prediction method in a first period, adopting a support vector machine prediction algorithm in a second period, and adopting a neural network prediction method in a third period. Prediction results of different classifications in different regions:
Then the first time period of the first time period,
。
in this way, the influence of meteorological feature factors in different areas on the prediction of the photovoltaic power generation power is considered, the weight of the predicted value in the final power generation amount is dynamically adjusted by establishing prediction models with different weights, the photovoltaic power generation power is accurately predicted, the weight of the photovoltaic power generation amount in different time periods is synthesized, and the adaptability of the photovoltaic power generation prediction is effectively met.
The method provided by the invention also predicts the photovoltaic power generation power, is beneficial to the dispatching department to make a reasonable power generation plan, is beneficial to frequency modulation, voltage regulation, standby and the like of the power grid, and further ensures the safety and economy of power supply.
The method can also analyze and predict the distributed photovoltaic power generation state, define the daily period of photovoltaic power generation, analyze the change trend of the distributed photovoltaic power generation, evaluate whether the distributed photovoltaic power generation meets the requirement, and whether the power generation and the power supply bring risk to the power grid, thereby having a positive effect on guaranteeing the safety and the economy of the distributed photovoltaic power generation.
As shown in fig. 2, the following is an embodiment of a distributed photovoltaic power generation prediction system provided by an embodiment of the present disclosure, which belongs to the same inventive concept as the distributed photovoltaic power generation prediction method of the above embodiments, and details of the embodiment of the distributed photovoltaic power generation prediction system, which are not described in detail, may be referred to the embodiment of the above distributed photovoltaic power generation prediction method.
The system comprises: the system comprises a data classification module, an analysis and extraction module and a prediction planning module;
the data classification module is used for acquiring photovoltaic power generation historical data, classifying the photovoltaic power generation data of different areas by adopting a density peak clustering algorithm, and obtaining photovoltaic power generation power types;
the analysis and extraction module analyzes factors influencing the photovoltaic output in different clusters according to the classified photovoltaic power generation power, extracts characteristic factors, performs linear transformation on distributed photovoltaic power generation power data, and performs numerical processing on the extracted characteristic factors;
and the prediction planning module establishes a power prediction model according to the characteristic factors and the photovoltaic output magnitudes of different spaces and different time periods, and determines a final photovoltaic power generation power prediction result so as to determine a final planned power generation amount.
As an embodiment of the system, photovoltaic power generation history data can be archived and stored; so as to facilitate the extraction and use. The distributed photovoltaic power generation layout diagram can be constructed, and the distributed photovoltaic power generation state in the layout diagram is updated in real time. The distributed photovoltaic operation interface can be configured in the system, so that monitoring personnel can add photovoltaic power generation data which are not stored or configured in the system; or modifying or deleting the stored photovoltaic power generation data;
and acquiring the distributed photovoltaic power generation state information in real time based on the local area network or the wide area network, and comparing the acquired distributed photovoltaic power generation state information with a preset threshold value to acquire the current distributed photovoltaic power generation state information. The system can predict the current distributed photovoltaic power generation trend to form a histogram or a graph for reference of operators;
the distributed photovoltaic power generation state can be tracked and collected, data sharing is achieved, and trend display is carried out on each piece of distributed photovoltaic power generation state information every day, every week, every month and every quarter.
The distributed photovoltaic power generation prediction system can also collect and store the photovoltaic power generation power prediction result and the power generation amount with high efficiency, process the photovoltaic power generation power prediction result and the power generation amount, process monitoring can be realized based on the photovoltaic power generation state, and the whole photovoltaic power generation process is described by using a multidimensional space. The quality and efficiency of photovoltaic power generation are improved, potential safety hazards in the photovoltaic power generation process are timely found and early warning is carried out, so that the energy efficiency of the photovoltaic power generation process is improved, the risk of surfing the Internet of the photovoltaic power generation is controlled, and timeliness and scientificity of monitoring, managing and controlling the whole photovoltaic power generation process are realized.
The elements and algorithm steps of the examples described in the embodiments disclosed in the distributed photovoltaic power generation prediction method and system of the present invention can be implemented in electronic hardware, computer software, or a combination of both, and to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been generally described in terms of their functions in the above description. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. In addition, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices, or elements, or may be an electrical, mechanical, or other form of connection.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (4)
1. A distributed photovoltaic power generation prediction method, the method comprising:
step 1: acquiring photovoltaic power generation historical data, and classifying the photovoltaic power generation data of different areas by adopting a density peak clustering algorithm to obtain photovoltaic power generation power types;
according to the classified photovoltaic power generation power, analyzing factors affecting photovoltaic output in different categories, and extracting characteristic factors;
calculating the parameter correlation in the distributed photovoltaic power generation power, performing linear transformation on distributed photovoltaic power generation power data, and performing numerical treatment on the extracted characteristic factors;
the specific steps of classifying the distributed photovoltaic power generation power data by adopting a density peak clustering algorithm comprise the following steps:
extracting the distributed photovoltaic power generation condition;
constructing a clustering model, taking the extracted distributed photovoltaic power generation condition as the input of the clustering model, setting the number of clusters and an iteration stop threshold value, and initializing a clustering prototype mode matrix and an iteration counter;
iteratively calculating a partition matrix and a clustering prototype mode matrix;
judging whether the values of two iterations before and after the clustering prototype mode matrix are smaller than a set iteration stop threshold, if yes, stopping calculating, outputting a partition matrix and the clustering prototype mode matrix, otherwise, adding 1 to an iteration counter, and returning to continue to iteratively calculate the partition matrix and the clustering prototype mode matrix;
for any sample point in the dataset
The density peak clustering algorithm calculates two variables, namely sample point +.>
Is>
And relative distance->
；
Sample point
Is>
The definition is as shown in the formula:
wherein, the liquid crystal display device comprises a liquid crystal display device,
for sample dot->
And->
European distance,/, of->
Is the cut-off distance;
when (when)
When (I)>
Otherwise->
；
Relative distance
For sample dot->
To be denser than its local partThe distance to the nearest point with higher degree;
relative distance
The formula of the calculation of (2) is as follows:
relative distance
And local Density->
The product of (2) is the complex variable +.>
；
The method further comprises the following steps:
(1) Acquiring a photovoltaic power generation historical data set and configuring the photovoltaic power generation historical data set as a sample classification number k;
(2) Calculating a distance matrix of the data set, and calculating a cut-off distance according to the percentage
；
(3) Calculating local density
Distance->
And the complex variables +.>
；
(4) Labeling k cluster centers:
(5) Categorizing non-clustered central data points:
(6) Judging noise points, calculating an average local density upper bound for each class, and judging the noise points if the boundary point density of the class is lower than the average local density upper bound;
step 2: according to the classified photovoltaic power generation power, analyzing factors influencing the photovoltaic output in different clusters, extracting characteristic factors, performing linear transformation on distributed photovoltaic power generation power data, and performing numerical processing on the extracted characteristic factors;
step 3: according to the characteristic factors and the photovoltaic output magnitudes of different spaces and different time periods, a power prediction model is established, and a final photovoltaic power generation power prediction result is determined, so that a final planned power generation amount is determined;
according to illumination duration and illumination intensity on different spaces, a comprehensive prediction model with variable weight in time intervals is established, different prediction algorithms are adopted in different time intervals, and finally a prediction result is obtained through weighting;
the method for establishing the comprehensive prediction model with the time-interval variable weights further comprises the following steps:
the period of photovoltaic power generation per day is defined as 6:00-18:00;
according to the generated power, the first period is as follows: 6:00-10: 00. the second period is: 10:00-14: 00. the third period is: 14:00-18:00;
the method comprises the steps of adopting a gray prediction method in a first period, adopting a support vector machine prediction algorithm in a second period, and adopting a neural network prediction method in a third period.
2. The method for predicting distributed photovoltaic power generation according to claim 1, wherein the distributed photovoltaic power generation power data is linearly transformed by using a dispersion normalization method.
3. A distributed photovoltaic power generation prediction system, characterized in that the system adopts the distributed photovoltaic power generation prediction method according to any one of claims 1 to 2;
the system comprises: the system comprises a data classification module, an analysis and extraction module and a prediction planning module;
the data classification module is used for acquiring photovoltaic power generation historical data, classifying the photovoltaic power generation data of different areas by adopting a density peak clustering algorithm, and obtaining photovoltaic power generation power types;
the analysis and extraction module analyzes factors influencing the photovoltaic output in different clusters according to the classified photovoltaic power generation power, extracts characteristic factors, performs linear transformation on distributed photovoltaic power generation power data, and performs numerical processing on the extracted characteristic factors;
and the prediction planning module establishes a power prediction model according to the characteristic factors and the photovoltaic output magnitudes of different spaces and different time periods, and determines a final photovoltaic power generation power prediction result so as to determine a final planned power generation amount.
4. A terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the distributed photovoltaic power generation prediction method of any of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310133174.4A CN115882455B (en) | 2023-02-20 | 2023-02-20 | Distributed photovoltaic power generation prediction method, system and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310133174.4A CN115882455B (en) | 2023-02-20 | 2023-02-20 | Distributed photovoltaic power generation prediction method, system and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115882455A CN115882455A (en) | 2023-03-31 |
| CN115882455B true CN115882455B (en) | 2023-07-04 |
Family
ID=85761312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310133174.4A Active CN115882455B (en) | 2023-02-20 | 2023-02-20 | Distributed photovoltaic power generation prediction method, system and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115882455B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116191423B (en) * | 2023-04-11 | 2023-10-10 | 中南大学 | Power collection system and method of photovoltaic power generation system |
| CN116128167B (en) * | 2023-04-17 | 2023-06-16 | 江苏图南数字科技有限公司 | Distributed photovoltaic power generation analysis method based on cloud computing real-time monitoring |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329977A (en) * | 2020-09-10 | 2021-02-05 | 国家电网有限公司 | Wind power prediction system for extreme scene |
| CN114792156A (en) * | 2022-03-10 | 2022-07-26 | 国网甘肃省电力公司电力科学研究院 | Photovoltaic output power prediction method and system based on curve characteristic index clustering |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107169683A (en) * | 2017-06-30 | 2017-09-15 | 上海电力学院 | A kind of grid-connected photovoltaic power station short term power combination forecasting method of variable weight coefficient |
| US10651654B2 (en) * | 2017-11-07 | 2020-05-12 | State Grid Corporation Of China | Model predictive controller for autonomous hybrid microgrids |
| CN112215383A (en) * | 2019-07-10 | 2021-01-12 | 中国电力科学研究院有限公司 | Distributed photovoltaic power generation power prediction method and system |
| CN113920349A (en) * | 2021-10-14 | 2022-01-11 | 国网青海省电力公司清洁能源发展研究院 | Wind and light typical scene construction method containing meteorological data based on density peak value-FCM |
-
2023
- 2023-02-20 CN CN202310133174.4A patent/CN115882455B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329977A (en) * | 2020-09-10 | 2021-02-05 | 国家电网有限公司 | Wind power prediction system for extreme scene |
| CN114792156A (en) * | 2022-03-10 | 2022-07-26 | 国网甘肃省电力公司电力科学研究院 | Photovoltaic output power prediction method and system based on curve characteristic index clustering |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115882455A (en) | 2023-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115882455B (en) | Distributed photovoltaic power generation prediction method, system and terminal | |
| Fan et al. | A review on data preprocessing techniques toward efficient and reliable knowledge discovery from building operational data | |
| Majidpour et al. | Fast prediction for sparse time series: Demand forecast of EV charging stations for cell phone applications | |
| Dou et al. | Hybrid model for renewable energy and loads prediction based on data mining and variational mode decomposition | |
| CN109190890A (en) | A kind of user behavior analysis method based on custom power consumption data | |
| CN112330078B (en) | Power consumption prediction method and device, computer equipment and storage medium | |
| Ruan et al. | Workload time series prediction in storage systems: a deep learning based approach | |
| JP2017135964A (en) | Operation of photovoltaic power generation system | |
| CN112668611B (en) | Kmeans and CEEMD-PE-LSTM-based short-term photovoltaic power generation power prediction method | |
| CN114066073A (en) | Power grid load prediction method | |
| Wang et al. | Wind speed interval prediction based on multidimensional time series of Convolutional Neural Networks | |
| CN110162390A (en) | A kind of method for allocating tasks and system of mist computing system | |
| CN113780684A (en) | Intelligent building user energy consumption behavior prediction method based on LSTM neural network | |
| CN111179108A (en) | Method and device for predicting power consumption | |
| CN114912720A (en) | Memory network-based power load prediction method, device, terminal and storage medium | |
| Khodaverdian et al. | Combination of convolutional neural network and gated recurrent unit for energy aware resource allocation | |
| CN109344875B (en) | Method and device for generating solar wind power output time sequence based on cluster analysis | |
| CN109391515A (en) | Network failure prediction technique and system based on dove group's algorithm optimization support vector machines | |
| CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
| Alsafasfeh | An Efficient Algorithm for Power Prediction in PV Generation System. | |
| CN110659681B (en) | Time sequence data prediction system and method based on pattern recognition | |
| Pan et al. | Application of Parallel Clustering Algorithm Based on R in Power Customer Classification | |
| CN115392102A (en) | Method and device for establishing energy consumption prediction model and method and system for predicting energy consumption | |
| CN112949908A (en) | Electricity price probability prediction method and device | |
| Wang et al. | Application of domain adaptation approach for weather data mining |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |