CN115186907A - Method, system, equipment and medium for predicting long-term power generation amount in wind power plant - Google Patents
Method, system, equipment and medium for predicting long-term power generation amount in wind power plant Download PDFInfo
- Publication number
- CN115186907A CN115186907A CN202210830162.2A CN202210830162A CN115186907A CN 115186907 A CN115186907 A CN 115186907A CN 202210830162 A CN202210830162 A CN 202210830162A CN 115186907 A CN115186907 A CN 115186907A
- Authority
- CN
- China
- Prior art keywords
- power
- medium
- data
- long
- term
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000007774 longterm Effects 0.000 title claims abstract description 120
- 238000010248 power generation Methods 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000004364 calculation method Methods 0.000 claims abstract description 20
- 238000004590 computer program Methods 0.000 claims description 14
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 2
- 230000005611 electricity Effects 0.000 claims 1
- 230000002349 favourable effect Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 230000003472 neutralizing effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
-
- 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
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Environmental & Geological Engineering (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Ecology (AREA)
- Atmospheric Sciences (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Environmental Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Biodiversity & Conservation Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method, a system, equipment and a medium for predicting medium-term and long-term power generation amount of a wind power plant. Acquiring historical actual data and meteorological forecast data of a wind power plant; constructing a data interpolation model, and training the data interpolation model according to the historical actual data to obtain the medium-term and long-term prediction power of the meteorological forecast; carrying out statistical calculation on the historical actual data according to the months and the moments to obtain the historical actual medium and long term predicted power; carrying out weighted calculation on the weather forecast medium and long term prediction power and the historical actual medium and long term prediction power to obtain comprehensive medium and long term prediction power; and carrying out integral summation on the comprehensive medium and long-term predicted power according to a time rule of power generation amount statistics and the time length to obtain the medium and long-term predicted power generation amount of the wind power plant. The method effectively improves the accuracy and robustness of the prediction of the medium-term power generation amount and the long-term power generation amount of the wind power plant, and is favorable for making a medium-term power generation plan and developing medium-term trading and long-term trading in the power market of the wind power plant.
Description
Technical Field
The invention relates to the technical field of wind power generation, in particular to a method, a system, equipment and a medium for predicting medium-term and long-term power generation amount of a wind power plant.
Background
Since a new round of electric power market reformation is implemented, annual and monthly electric power medium and long term trading in various regions is carried out in a normalized manner. The medium-long term prediction is an important basis for developing power generation index distribution and electric power market declared volume price trading, risks can be avoided in advance, trading contracts are large in proportion, and market profit proportion is highest. However, due to fluctuation and intermittency of wind speed, grid connection of a wind power plant can bring a lot of influences to a power system, and great uncertainty and even huge economic loss can be brought to power generation enterprises in power market transaction. A large amount of historical actual data are needed for long-term power generation amount prediction in a wind power plant at present, but most of historical data of the wind power plant have the problems of small data amount, missing, errors and the like, the overall quality of the data is poor, and the accuracy of a prediction model is affected.
At present, a method for predicting the medium-term and long-term power generation amount of a wind power plant generally counts a daily or monthly power generation amount result according to years of historical actual data of the wind power plant, and properly adjusts the historical power generation amount by combining a size proportional relation between years of historical average wind speed and the average wind speed of the medium-term and long-term in the future so as to obtain a future power generation amount prediction result. Because the atmospheric circulation and the local climate have uncertainty, the annual historical power generation amount of the wind power plant has larger fluctuation and deviation, and the future power generation amount situation cannot be accurately predicted according to historical actual data alone; in addition, because the output of the fan at each moment is directly related to wind conditions such as wind speed fluctuation degree and the like, the generated energy of the fan is related to wind speed distribution, the average wind speed is large, the generated energy of the fan is not necessarily high, and the rise and fall of the generated energy cannot be completely determined by the size of the average wind speed; the time period of the medium-long term prediction is long, which generally refers to annual and monthly predictions, in the prior art, the advance along with time is not considered, the credibility of the historical actual data and the future predicted wind speed changes, and the credibility change trends of the historical actual data and the future predicted wind speed are not fully and effectively utilized, so that high accuracy rate is difficult to guarantee in the early stage and the later stage of the medium-long term.
In order to fully utilize the periodic influence rule of earth revolution (one year) and rotation (one day) on the wind power plant, the periodicity and the commonality of years of historical actual data are mined to form a specific typical output curve of the wind power plant, and the predicted power based on the historical data is obtained and is used as a reference value of the final predicted power generation amount; the method comprises the steps of fully utilizing high-precision medium-term and long-term weather forecast data, obtaining prediction data of short time intervals by adopting a scientific frontier interpolation technology, training weather and power models, obtaining prediction power based on weather forecast, and taking the prediction power as an adjustment value of final predicted power generation amount; the high credibility of the prediction power based on historical data in a long time period is fully utilized, the high credibility of the prediction power based on weather forecast in a short time period is utilized, the uncertainty and the untrustworthiness of the prediction power in the long time period are avoided, a Gong Baici curve is adopted for neutralization, and the prediction power based on a Gong Baici curve with high overall precision is obtained; and finally, calculating to obtain the medium-term and long-term predicted power generation according to the time rule of power generation statistics, so that a medium-term and long-term power generation prediction method of the wind power plant is developed.
Disclosure of Invention
The invention provides a method, a system, equipment and a medium for predicting medium-term and long-term power generation amount of a wind power plant.
In a first aspect, an embodiment of the present invention provides a method for predicting long-term power generation amount in a wind farm, where the method includes the following steps:
acquiring historical actual data and meteorological forecast data of a wind power plant; the historical actual data comprises actual operation data including average fans and total power of the whole fans; the weather forecast data comprises medium and long term numerical weather forecast data including wind speed, wind direction, air temperature, air pressure and relative humidity; the time interval of the weather forecast data is a first time interval;
constructing a data interpolation model, training the data interpolation model according to the historical actual data, and inputting the weather forecast data of a first time interval into the trained data interpolation model to obtain the weather forecast data of a second time interval;
training the weather forecast data of the second time interval through a neural network model to obtain the medium and long term prediction power of the weather forecast;
performing statistical calculation on the historical actual data according to the months and the moments to obtain the actual medium-long term prediction power of the history;
carrying out weighted calculation on the weather forecast medium and long term prediction power and the historical actual medium and long term prediction power to obtain comprehensive medium and long term prediction power;
and carrying out integral summation on the comprehensive medium and long-term predicted power according to a time rule of power generation amount statistics and the time length to obtain the medium and long-term predicted power generation amount of the wind power plant.
Further, constructing a data interpolation model, and training the data interpolation model according to the historical actual data, wherein the method comprises the following steps:
interpolating the meteorological forecast data from a first time interval into a second time interval to obtain sample data; interpolating the weather forecast data according to the wind speed, the wind direction, the air temperature, the air pressure and the relative humidity, and interpolating from a first time interval to a second time interval to obtain corresponding five sample data;
carrying out normalization processing on the five sample data, and training the sample data respectively belonging to corresponding data interpolation models; the number of the data interpolation models is 5, and the data interpolation models respectively correspond to the wind speed, the wind direction, the air temperature, the air pressure and the relative humidity in the meteorological forecast data.
Further, training the weather forecast data of the second time interval through a neural network model to obtain the weather forecast medium and long term prediction power, including:
performing characteristic engineering on the interpolated historical meteorological forecast data, taking meteorological data at a specific moment and variation deviations of a plurality of moments before and after as an input sequence of a sample, and taking historical actual power at a corresponding moment as an output value of the sample to obtain a sample set of a neural network training model;
modeling and training weather forecast data corresponding to wind speed, wind direction, air temperature, air pressure and relative humidity by using a machine learning algorithm, inputting the interpolated historical weather forecast data into a trained neural network training model, and predicting to obtain historical future predicted power; the historical future predicted power includes historical predicted power and future predicted power.
Further, after the historical and future predicted power is predicted, the method further comprises the following steps:
calculating a first average value in a preset position in historical actual power in a specific period and a second average value in a preset position in historical predicted power of the historical future predicted power in the same period;
and taking the ratio of the first average value to the second average value as a correction coefficient, and carrying out integral stretching and scaling on the future predicted power of the historical future predicted power to obtain the weather forecast medium and long-term predicted power.
Further, the predetermined quantile is [90%,100% ].
Further, the statistical calculation is performed on the historical actual data according to the month and the moment to obtain the historical actual medium and long term predicted power, and the method comprises the following steps:
calculating the average value, the 25% quantile value and the 75% quantile value of each time of each month in the historical actual power according to the time of the month and the third time interval, and calculating the actual medium-long term predicted power according to the following formula:
(average value + (a 25% quantile value + (1-a) 75% quantile value)). 0.5
Wherein, the value range of a is (0.25,0.75).
Further, performing weighted calculation on the weather forecast medium and long term prediction power and the historical actual medium and long term prediction power to obtain comprehensive medium and long term prediction power, and the method comprises the following steps:
calculating the average value of the ratio of the annual root mean square error to the installed capacity of the wind power plant in months according to the historical actual medium-long term predicted power and the historical actual power, and taking the average value as the monthly credibility of the historical actual medium-long term predicted power; calculating the average value of the ratio of the annual root mean square error to the installed capacity of the wind power plant in months according to the historical predicted power and the historical actual power, taking the average value as the monthly credibility of the long-term predicted power in the meteorological forecast, and performing equal-proportion adjustment on the two groups of credibility in months to enable the proportion sum of the two groups of credibility to be equal to 1;
according to the monthly credibility sequence of the adjusted historical actual medium-long term predicted power, fitting by adopting a Gong Baici curve to obtain a fitting curve;
and performing weighted summation on the historical actual medium and long term prediction power and the weather forecast medium and long term prediction power by adopting a fitting curve, so that the sum of the weights of the two is 1, and obtaining the comprehensive medium and long term prediction power.
In a second aspect, an embodiment of the present invention provides a system for predicting long-term power generation in a wind farm, where the system includes:
the data acquisition module is used for acquiring historical actual data and meteorological forecast data of the wind power plant; the historical actual data comprises actual operation data including average fans and total power of the whole fans; the weather forecast data comprises medium and long term numerical weather forecast data including wind speed, wind direction, air temperature, air pressure and relative humidity; the time interval of the weather forecast data is a first time interval;
the data interpolation module is used for constructing a data interpolation model, training the data interpolation model according to the historical actual data, and inputting weather forecast data of a first time interval into the trained data interpolation model to obtain weather forecast data of a second time interval;
the weather forecast power prediction module is used for training the weather forecast data of the second time interval through a neural network model to obtain the weather forecast medium and long term prediction power;
the historical prediction power prediction module is used for carrying out statistical calculation on the historical actual data according to the months and the moments to obtain the historical actual medium and long term prediction power;
the comprehensive prediction module is used for carrying out weighted calculation on the weather forecast middle and long term prediction power and the historical actual middle and long term prediction power to obtain comprehensive middle and long term prediction power;
and the power generation capacity prediction module is used for carrying out integral summation on the comprehensive medium and long-term prediction power according to a time rule of power generation capacity statistics and obtaining the medium and long-term prediction power generation capacity of the wind power plant.
In a third aspect, an embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the above method when executing the computer program.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is configured to, when executed by a processor, implement the steps of the foregoing method.
The invention provides a method, a system, equipment and a medium for predicting medium-term and long-term power generation amount of a wind power plant. According to the method, historical actual data is used as a reference standard of future power generation, the periodicity and the commonality of the annual historical actual data of the wind power plant are fully utilized, weather prediction data is used as an adjustment basis of the future power generation, the high-precision correction effect of the medium-and-long-term weather prediction data is fully exerted, the weather prediction data is interpolated by a deep learning algorithm, the wind condition characteristics of the wind power plant are fully integrated, the prediction data of different sources are neutralized by a Gong Baici curve, the reliability change of a prediction result along with the time lapse is fully considered, and the accuracy of medium-and-long-term power generation prediction is improved through multiple technologies.
The method can improve the accuracy of the prediction of the medium-term and long-term generated energy of the wind power plant, and is scientific, precise, high in accuracy, stable, reliable and good in universality.
Drawings
FIG. 1 is a schematic diagram of application of a long-term power generation amount prediction method in a wind power plant in the embodiment of the invention;
FIG. 2 is a schematic flow chart of a method for predicting long-term power generation in a wind farm according to an embodiment of the invention;
FIG. 3 is a schematic flow chart of the step S12 of building a data interpolation model and obtaining weather forecast data of a second time interval according to the embodiment of the present invention;
FIG. 4 is a schematic flowchart of the step S13 of training the weather forecast data of the second time interval through the neural network model in the embodiment of the present invention;
FIG. 5 is a schematic flow chart of the step S15 to obtain the comprehensive long and medium term prediction power in the embodiment of the present invention;
FIG. 6 is a schematic structural diagram of a long-term power generation amount prediction system in a wind power plant according to an embodiment of the invention;
fig. 7 is an internal structural diagram of a computer device in the embodiment of the present invention.
Detailed Description
In order to make the purpose, technical solution and advantages of the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments, and it is obvious that the embodiments described below are part of the embodiments of the present invention, and are used for illustrating the present invention only, but not for limiting the scope of the present invention. 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.
The method for predicting the medium-term power generation amount and the long-term power generation amount of the wind power plant can be applied to a terminal or a server shown in figure 1. The terminal can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, and the server can be implemented by an independent server or a server cluster formed by a plurality of servers. The following examples will describe the method of long term power generation in a wind farm of the present invention in detail.
In one embodiment, as shown in FIG. 2, there is provided a method of long term power generation in a wind farm, comprising the steps of:
s11, acquiring historical actual data and weather forecast data of the wind power plant; the historical actual data comprises actual operation data including average fans and total power of the whole fan; the time interval of the weather forecast data of the medium-long term numerical weather forecast data including wind speed, wind direction, air temperature, air pressure and relative humidity is a first time interval.
Preferably, the time interval of the historical actual data is 15 minutes, and the time period is more than two years of history or the wind power plant is connected to the grid to the present; the weather forecast data time interval is 6 hours, and the time period is the past year and the next year. The time interval is selected to provide both a sufficient number of samples and to keep the cost of acquiring data within a certain range; the time period is set to ensure that the time period of the selected data is longer and the reliability is strong.
S12, constructing a data interpolation model, training the data interpolation model according to the historical actual data, and inputting the weather forecast data at the first time interval into the trained data interpolation model to obtain the weather forecast data at the second time interval, as shown in fig. 3, specifically including:
s121, interpolating the weather forecast data from a first time interval to a second time interval to obtain sample data; interpolating the weather forecast data according to the wind speed, the wind direction, the air temperature, the air pressure and the relative humidity, and interpolating from a first time interval to a second time interval to obtain corresponding five sample data;
s122, carrying out normalization processing on the five sample data, and training the sample data respectively belonging to corresponding data interpolation models; the number of the data interpolation models is 5, and the data interpolation models respectively correspond to the wind speed, the wind direction, the air temperature, the air pressure and the relative humidity in the meteorological forecast data.
Preferably, a 6 hour interval is selected to interpolate to a 1 hour interval. Sampling historical actual data according to 6 hours, taking two actual wind speeds of adjacent 6 hours, linearly interpolating to the 1 st, 2 nd, 3 rd, 4 th and 5 th hours in the middle of the 6 hours, and inputting the two wind speeds of the adjacent 6 hours and the wind speeds of the 1 st, 2 nd, 3 rd, 4 th and 5 th hours together. The method can more fully utilize the existing data modeling training and obtain the weather forecast data with smaller time interval by the interpolation mode without acquiring the actual weather data too frequently.
S13, training the weather forecast data of the second time interval through a neural network model to obtain the weather forecast medium and long term prediction power, as shown in FIG. 4, specifically comprising:
s131, performing characteristic engineering on the interpolated historical meteorological forecast data, taking meteorological data at a specific moment and variation deviations of a plurality of moments before and after as an input sequence of a sample, and taking historical actual power at a corresponding moment as an output value of the sample to obtain a sample set of a neural network training model;
s132, modeling and training meteorological forecast data corresponding to wind speed, wind direction, air temperature, air pressure and relative humidity respectively by using a machine learning algorithm, inputting the interpolated historical meteorological forecast data into a trained neural network training model, and predicting to obtain historical future prediction power; the historical future predicted power comprises historical predicted power and future predicted power;
s133, calculating a first average value in a preset position in historical actual power in a specific period and a second average value in a preset position in historical predicted power of the historical future predicted power at the same time period; and taking the ratio of the first average value to the second average value as a correction coefficient, and carrying out integral stretching and scaling on the future predicted power of the historical future predicted power to obtain the weather forecast medium and long term predicted power, wherein the preset quantiles are [90%,100% ].
The specific period may be selected to be a period of several months, such as a period of 2-3 months. In addition, the stretching or scaling depends on the specific value of the correction coefficient, if the correction coefficient is greater than 1, the future prediction data is stretched, and if the coefficient is less than 1, the future prediction data is scaled, so that the obtained future prediction data is more accurate.
S14, carrying out statistical calculation on the historical actual data according to the months and the moments to obtain the historical actual medium and long term predicted power; calculating the average value, the 25% quantile value and the 75% quantile value of each time of each month in the historical actual power according to the time of the month and the third time interval, and calculating the actual medium-long term predicted power according to the following formula:
(average value + (a 25% quantile value + (1-a) 75% quantile value)). 0.5
Wherein, the value range of a is (0.25,0.75).
Preferably, the time interval of 15 minutes is chosen to provide both a sufficient number of samples and to keep the cost of acquiring the data within a certain range. The value of a is adjusted according to the actual situation so as to better predict the historical actual data.
S15, carrying out weighted calculation on the weather forecast medium and long term prediction power and the historical actual medium and long term prediction power to obtain comprehensive medium and long term prediction power, wherein as shown in figure 5, the method specifically comprises the following steps:
s151, calculating the average value of the ratio of the annual root mean square error to the installed capacity of the wind power plant in months according to the historical actual medium-long term predicted power and the historical actual power, and taking the average value as the monthly credibility of the historical actual medium-long term predicted power;
s152, calculating the average value of the ratio of the annual root mean square error to the installed capacity of the wind power plant in months according to the historical predicted power and the historical actual power of the weather forecast, and taking the average value as the monthly credibility of the medium-term and long-term predicted power of the weather forecast;
s153, carrying out equal proportion adjustment on the two groups of credibility in months, and enabling the proportion sum of the two groups of credibility to be equal to 1; according to the monthly credibility sequence of the adjusted historical actual medium-long term predicted power, fitting by adopting a Gong Baici curve to obtain a fitting curve;
and S154, performing weighted summation on the historical actual medium and long term predicted power and the weather forecast medium and long term predicted power by adopting a fitting curve, and enabling the sum of the weights of the two to be 1 to obtain the comprehensive medium and long term predicted power.
The method comprises the steps of comprehensively considering weather forecast predicted power and historical actual predicted power, and adjusting the reliability according to the error of the weather forecast predicted power and the average value of the ratio of the error to the installed capacity of the wind power plant. The Gong Baici curve model is widely used for data prediction, and can be better fitted to obtain comprehensive medium and long-term prediction power. The Gong Baici curve can be replaced by other curves according to actual conditions, such as a positive logic curve and the like.
And S16, carrying out integral summation on the comprehensive medium and long-term predicted power according to a time rule of power generation amount statistics and obtaining medium and long-term predicted power generation amount of the wind power plant.
The time rule of the power generation amount statistics is selected according to the needs, such as integrating and summing hour by hour, or integrating and summing the peak and valley period of one day. Because people are difficult to evaluate and consider power data relative to power generation data, the comprehensive medium-and-long-term predicted power is calculated to obtain the medium-and-long-term predicted power generation amount of the wind power plant in a time integration mode, and medium-and-long-term power generation plans in wind power plant making and medium-and-long-term trading in the power market are facilitated.
The embodiment of the application provides a method for predicting medium-and-long-term power generation capacity of a wind power plant, historical actual data are used as a reference standard of future power generation capacity, periodicity and commonality of the historical actual data of the wind power plant over years are fully utilized, meteorological prediction data are used as an adjustment basis of the future power generation capacity, a high-precision correction effect of medium-and-long-term meteorological forecast data is fully exerted, a deep learning algorithm is used for interpolating the meteorological forecast data, wind condition characteristics of the wind power plant are fully integrated, gong Baici curves are used for neutralizing the forecast data from different sources, credibility changes of prediction results along with time lapse are fully considered, and accuracy of medium-and-long-term power generation capacity prediction is improved through multiple technologies. The method can improve the accuracy of the prediction of the medium-term and long-term generated energy of the wind power plant, and is scientific, precise, high in accuracy, stable, reliable and good in universality.
Based on the method for predicting the medium-term power generation amount and the long-term power generation amount of the wind farm, the embodiment of the invention also provides a system for predicting the medium-term power generation amount and the long-term power generation amount of the wind farm, and as shown in fig. 6, the system comprises:
the data acquisition module 1 is used for acquiring historical actual data and meteorological forecast data of the wind power plant; the historical actual data comprises actual operation data including average fans and total power of the whole fans; the weather forecast data comprises medium and long term numerical weather forecast data including wind speed, wind direction, air temperature, air pressure and relative humidity; the time interval of the weather forecast data is a first time interval;
the data interpolation module 2 is used for constructing a data interpolation model, training the data interpolation model according to the historical actual data, and inputting weather forecast data of a first time interval into the trained data interpolation model to obtain weather forecast data of a second time interval;
the weather forecast power prediction module 3 is used for training the weather forecast data of the second time interval through a neural network model to obtain the weather forecast medium and long term prediction power;
the historical prediction power prediction module 4 is used for carrying out statistical calculation on the historical actual data according to the months and the moments to obtain the historical actual medium and long term prediction power;
the comprehensive prediction module 5 is used for carrying out weighted calculation on the weather forecast medium and long term prediction power and the historical actual medium and long term prediction power to obtain comprehensive medium and long term prediction power;
and the power generation capacity prediction module 6 is used for carrying out integral summation on the comprehensive medium and long-term prediction power according to a time rule of power generation capacity statistics and obtaining the medium and long-term prediction power generation capacity of the wind power plant.
For specific limitations of the medium-and-long-term power generation amount prediction system of the wind farm, reference may be made to the above limitations of the medium-and-long-term power generation amount prediction method of the wind farm, and details are not repeated here. All or part of each module in the medium-and-long-term power generation amount prediction system of the wind power plant can be realized through software, hardware and combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Fig. 7 shows an internal structure diagram of a computer device in one embodiment, and the computer device may be specifically a terminal or a server. As shown in fig. 7, the computer apparatus includes a processor, a memory, a network interface, a display, and an input device, which are connected through a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of predicting long term power generation in a wind farm. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those of ordinary skill in the art that the architecture shown in FIG. 7 is a block diagram of only a portion of the architecture associated with the subject application, and is not intended to limit the computing devices to which the subject application may be applied, as a particular computing device may include more or less components than those shown, or may combine certain components, or have a similar arrangement of components.
In one embodiment, a computer device is provided, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the steps of the above method being performed when the computer program is executed by the processor.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the above-mentioned method.
In summary, the invention relates to a method, a system, equipment and a medium for predicting the medium-term and long-term power generation capacity of a wind power plant. Acquiring historical actual data and meteorological forecast data of a wind power plant; constructing a data interpolation model, and training the data interpolation model according to the historical actual data to obtain the medium-term and long-term prediction power of the meteorological forecast; carrying out statistical calculation on the historical actual data according to the months and the moments to obtain the historical actual medium and long term predicted power; carrying out weighted calculation on the weather forecast medium and long term prediction power and the historical actual medium and long term prediction power to obtain comprehensive medium and long term prediction power; and carrying out integral summation on the comprehensive medium and long-term predicted power according to a time rule of power generation amount statistics and a time length to obtain medium and long-term predicted power generation amount of the wind power plant. The method effectively improves the accuracy and robustness of the prediction of the long-term generated energy in the wind power plant, and is beneficial to the long-term power generation plan in the wind power plant manufacturing and the long-term trading in the power market.
The embodiments in this specification are described in a progressive manner, and all the same or similar parts of the embodiments are directly referred to each other, and each embodiment is described with emphasis on differences from other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment. It should be noted that, the technical features of the embodiments may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express some preferred embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for those skilled in the art, without departing from the technical principle of the present invention, several improvements and substitutions can be made, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the present patent shall be subject to the protection scope of the claims.
Claims (10)
1. A method for predicting long-term power generation in a wind power plant is characterized by comprising the following steps:
acquiring historical actual data and meteorological forecast data of a wind power plant; the historical actual data comprises actual operation data including average fans and total power of the whole fans; the weather forecast data comprises medium and long term numerical weather forecast data including wind speed, wind direction, air temperature, air pressure and relative humidity; the time interval of the weather forecast data is a first time interval;
constructing a data interpolation model, training the data interpolation model according to the historical actual data, and inputting the meteorological forecast data of a first time interval into the trained data interpolation model to obtain the meteorological forecast data of a second time interval;
training the weather forecast data of the second time interval through a neural network model to obtain the medium and long term prediction power of the weather forecast;
carrying out statistical calculation on the historical actual data according to the months and the moments to obtain the historical actual medium and long term predicted power;
carrying out weighted calculation on the weather forecast medium and long term prediction power and the historical actual medium and long term prediction power to obtain comprehensive medium and long term prediction power;
and carrying out integral summation on the comprehensive medium and long-term predicted power according to a time rule of power generation amount statistics and the time length to obtain the medium and long-term predicted power generation amount of the wind power plant.
2. The method for predicting the long-term power generation amount in the wind farm according to claim 1, wherein the constructing of the data interpolation model and the training of the data interpolation model according to the historical actual data comprise:
interpolating the meteorological forecast data from a first time interval into a second time interval to obtain sample data; interpolating the weather forecast data according to the wind speed, the wind direction, the air temperature, the air pressure and the relative humidity, and interpolating from a first time interval to a second time interval to obtain corresponding five sample data;
carrying out normalization processing on the five sample data, and training the sample data respectively belonging to corresponding data interpolation models; the number of the data interpolation models is 5, and the data interpolation models respectively correspond to wind speed, wind direction, air temperature, air pressure and relative humidity in meteorological forecast data.
3. The method for predicting the medium-term power generation amount and the long-term power generation amount of the wind farm according to claim 2, wherein the training of the weather forecast data of the second time interval through the neural network model to obtain the medium-term and long-term predicted power of the weather forecast comprises the following steps:
performing characteristic engineering on the interpolated historical meteorological forecast data, taking meteorological data at a specific moment and variation deviations of a plurality of moments before and after as an input sequence of a sample, and taking historical actual power at a corresponding moment as an output value of the sample to obtain a sample set of a neural network training model;
modeling and training weather forecast data corresponding to wind speed, wind direction, air temperature, air pressure and relative humidity by using a machine learning algorithm, inputting the interpolated historical weather forecast data into a trained neural network training model, and predicting to obtain historical future predicted power; the historical future predicted power includes historical predicted power and future predicted power.
4. The method for predicting long-term power generation amount in a wind farm according to claim 3, wherein after the predicting obtains the historical predicted power and the future predicted power, the method further comprises the following steps:
calculating a first average value in a preset position in historical actual power in a specific period and a second average value in a preset position in historical predicted power of the historical future predicted power in the same period;
and taking the ratio of the first average value to the second average value as a correction coefficient, and carrying out integral stretching and scaling on the future predicted power of the historical future predicted power to obtain the weather forecast medium and long-term predicted power.
5. Method for prediction of long term electricity production in a wind park according to claim 4, characterized in that the predetermined quantile is [90%,100% ].
6. The method for predicting the medium-and-long-term power generation amount of the wind farm according to claim 1, wherein the step of performing statistical calculation on the historical actual data according to months and moments to obtain the historical actual medium-and-long-term predicted power comprises the following steps:
calculating the average value, the 25% quantile value and the 75% quantile value of each time of each month in the historical actual power according to the time of the month and the third time interval, and calculating the actual medium-long term predicted power according to the following formula:
(average value + (a 25% quantile value + (1-a) 75% quantile value)). 0.5
Wherein, the value range of a is (0.25,0.75).
7. The method for predicting the medium-and-long-term power generation amount of the wind farm according to claim 3, wherein the step of performing weighted calculation on the weather forecast medium-and-long-term predicted power and the historical actual medium-and-long-term predicted power to obtain comprehensive medium-and-long-term predicted power comprises the following steps:
calculating the average value of the ratio of the annual root mean square error to the installed capacity of the wind power plant in months according to the historical actual medium-long term predicted power and the historical actual power, and taking the average value as the monthly credibility of the historical actual medium-long term predicted power; calculating the average value of the ratio of the annual root mean square error to the installed capacity of the wind power plant in months according to the historical predicted power and the historical actual power of the weather forecast, taking the average value as the monthly credibility of the long-term predicted power in the weather forecast, and performing equal-proportion adjustment on the two groups of credibility in months to enable the proportion sum of the two groups of credibility to be equal to 1;
according to the monthly credibility sequence of the adjusted historical actual medium-long term predicted power, fitting by adopting a Gong Baici curve to obtain a fitting curve;
and performing weighted summation on the historical actual medium and long term prediction power and the weather forecast medium and long term prediction power by adopting a fitting curve, so that the sum of the weights of the two is 1, and obtaining the comprehensive medium and long term prediction power.
8. A system for predicting long term power generation in a wind farm, the system comprising:
the data acquisition module is used for acquiring historical actual data and meteorological forecast data of the wind power plant; the historical actual data comprises actual operation data including average fans and total power of the whole fans; the weather forecast data comprises medium and long term numerical weather forecast data including wind speed, wind direction, air temperature, air pressure and relative humidity; the time interval of the weather forecast data is a first time interval;
the data interpolation module is used for constructing a data interpolation model, training the data interpolation model according to the historical actual data, and inputting weather forecast data of a first time interval into the trained data interpolation model to obtain weather forecast data of a second time interval;
the weather forecast power prediction module is used for training the weather forecast data of the second time interval through a neural network model to obtain the weather forecast medium and long term prediction power;
the historical prediction power prediction module is used for carrying out statistical calculation on the historical actual data according to the months and the moments to obtain the historical actual medium and long term prediction power;
the comprehensive prediction module is used for carrying out weighted calculation on the weather forecast middle and long term prediction power and the historical actual middle and long term prediction power to obtain comprehensive middle and long term prediction power;
and the power generation capacity prediction module is used for carrying out integral summation on the comprehensive medium and long-term prediction power according to a time rule of power generation capacity statistics and obtaining the medium and long-term prediction power generation capacity of the wind power plant.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented when the computer program is executed by the processor.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210830162.2A CN115186907B (en) | 2022-07-14 | 2022-07-14 | Method, system, equipment and medium for predicting long-term power generation capacity in wind power plant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210830162.2A CN115186907B (en) | 2022-07-14 | 2022-07-14 | Method, system, equipment and medium for predicting long-term power generation capacity in wind power plant |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115186907A true CN115186907A (en) | 2022-10-14 |
| CN115186907B CN115186907B (en) | 2024-06-18 |
Family
ID=83519849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210830162.2A Active CN115186907B (en) | 2022-07-14 | 2022-07-14 | Method, system, equipment and medium for predicting long-term power generation capacity in wind power plant |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115186907B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115936238A (en) * | 2022-12-23 | 2023-04-07 | 华润电力技术研究院有限公司 | Method, system, equipment and medium for predicting middle and long term output of global wind power |
| CN116109007A (en) * | 2023-03-23 | 2023-05-12 | 阿里巴巴达摩院(杭州)科技有限公司 | Power generation power determination method, server and storage medium |
| CN118052415A (en) * | 2024-04-15 | 2024-05-17 | 国网浙江省电力有限公司宁波供电公司 | Distribution optimization method, device, equipment and medium for month priority power generation plan |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461444A (en) * | 2020-04-07 | 2020-07-28 | 上海电气风电集团股份有限公司 | Prediction method, system, medium and electronic device for unit power of wind power plant |
| CN111667093A (en) * | 2020-04-22 | 2020-09-15 | 华北电力科学研究院有限责任公司 | Medium-and-long-term wind power generation calculation method and device |
| CN112613674A (en) * | 2020-12-29 | 2021-04-06 | 国能日新科技股份有限公司 | Medium-and-long-term wind power generation capacity prediction method and device, electronic equipment and storage medium |
| CN112836870A (en) * | 2021-01-25 | 2021-05-25 | 华北电力大学 | Method, device, equipment and computer readable storage medium for predicting wind power |
| WO2021259622A1 (en) * | 2020-06-26 | 2021-12-30 | Siemens Gamesa Renewable Energy Innovation & Technology S.L. | A method and an apparatus for computer-implemented prediction of power production of one or more wind turbines in a wind farm |
-
2022
- 2022-07-14 CN CN202210830162.2A patent/CN115186907B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461444A (en) * | 2020-04-07 | 2020-07-28 | 上海电气风电集团股份有限公司 | Prediction method, system, medium and electronic device for unit power of wind power plant |
| CN111667093A (en) * | 2020-04-22 | 2020-09-15 | 华北电力科学研究院有限责任公司 | Medium-and-long-term wind power generation calculation method and device |
| WO2021259622A1 (en) * | 2020-06-26 | 2021-12-30 | Siemens Gamesa Renewable Energy Innovation & Technology S.L. | A method and an apparatus for computer-implemented prediction of power production of one or more wind turbines in a wind farm |
| CN112613674A (en) * | 2020-12-29 | 2021-04-06 | 国能日新科技股份有限公司 | Medium-and-long-term wind power generation capacity prediction method and device, electronic equipment and storage medium |
| CN112836870A (en) * | 2021-01-25 | 2021-05-25 | 华北电力大学 | Method, device, equipment and computer readable storage medium for predicting wind power |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115936238A (en) * | 2022-12-23 | 2023-04-07 | 华润电力技术研究院有限公司 | Method, system, equipment and medium for predicting middle and long term output of global wind power |
| CN115936238B (en) * | 2022-12-23 | 2024-01-12 | 华润电力技术研究院有限公司 | Method, system, equipment and medium for predicting medium-long-term output of global wind power |
| CN116109007A (en) * | 2023-03-23 | 2023-05-12 | 阿里巴巴达摩院(杭州)科技有限公司 | Power generation power determination method, server and storage medium |
| CN118052415A (en) * | 2024-04-15 | 2024-05-17 | 国网浙江省电力有限公司宁波供电公司 | Distribution optimization method, device, equipment and medium for month priority power generation plan |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115186907B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115186907A (en) | Method, system, equipment and medium for predicting long-term power generation amount in wind power plant | |
| Han et al. | Mid-to-long term wind and photovoltaic power generation prediction based on copula function and long short term memory network | |
| Shin et al. | Operational planning and optimal sizing of microgrid considering multi-scale wind uncertainty | |
| CN113937763B (en) | Wind power prediction method, device, equipment and storage medium | |
| JP5940922B2 (en) | Renewable energy prediction device | |
| Wang et al. | Wind speed forecasting based on multi-objective grey wolf optimisation algorithm, weighted information criterion, and wind energy conversion system: A case study in Eastern China | |
| Park et al. | Predictive model for PV power generation using RNN (LSTM) | |
| CN111461444A (en) | Prediction method, system, medium and electronic device for unit power of wind power plant | |
| CN114493050B (en) | Multi-dimensional fusion new energy power parallel prediction method and device | |
| Pourhabib et al. | Short-term wind speed forecast using measurements from multiple turbines in a wind farm | |
| CN110705771A (en) | Method and device for predicting and optimizing power generation power of new energy of regional power grid | |
| CN112906757A (en) | Model training method, optical power prediction method, device, equipment and storage medium | |
| He et al. | Probabilistic solar irradiance forecasting via a deep learning‐based hybrid approach | |
| CN114330934A (en) | Model parameter self-adaptive GRU new energy short-term power generation power prediction method | |
| CN116091118A (en) | Electricity price prediction method, device, equipment, medium and product | |
| US20230327440A1 (en) | Power generation amount management system and power generation amount management method | |
| CN114819385A (en) | Wind power prediction method and device, electronic equipment and storage medium | |
| CN113095547B (en) | Short-term wind power prediction method based on GRA-LSTM-ICE model | |
| Amonkar et al. | A k-nearest neighbor space-time simulator with applications to large-scale wind and solar power modeling | |
| CN112232886B (en) | Electricity price probability prediction method, system, computer equipment and storage medium | |
| Mokhov et al. | Electrical energy consumption prediction of the federal district of russia on the based of the reccurent neural network | |
| Sarathkumar et al. | Uncertainty borne balancing cost modeling for wind power forecasting | |
| JP2016170468A (en) | Electric power transaction amount determination system, electric power transaction amount determination method and program | |
| Pan et al. | Modeling optimization method based on Gamma test and NSGA II for forecast of PV power output | |
| Liu et al. | A Self-Data-Driven Method for Lifetime Prediction of PV Arrays Considering the Uncertainty and Volatility |
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 |