CN117787109A - Wind speed conversion reliability assessment method and system - Google Patents
Wind speed conversion reliability assessment method and system Download PDFInfo
- Publication number
- CN117787109A CN117787109A CN202410200562.4A CN202410200562A CN117787109A CN 117787109 A CN117787109 A CN 117787109A CN 202410200562 A CN202410200562 A CN 202410200562A CN 117787109 A CN117787109 A CN 117787109A
- Authority
- CN
- China
- Prior art keywords
- wind speed
- information
- speed conversion
- result
- model
- 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
- 238000006243 chemical reaction Methods 0.000 title claims abstract description 181
- 238000000034 method Methods 0.000 title claims description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000000342 Monte Carlo simulation Methods 0.000 claims abstract description 13
- 238000012098 association analyses Methods 0.000 claims abstract description 9
- 230000006870 function Effects 0.000 claims description 38
- 230000004927 fusion Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 4
- 238000013136 deep learning model Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims 3
- 238000011156 evaluation Methods 0.000 abstract description 12
- 210000002569 neuron Anatomy 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000002123 temporal effect Effects 0.000 description 3
- 238000012876 topography Methods 0.000 description 3
- 238000013076 uncertainty analysis Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000012097 association analysis method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a wind speed conversion reliability evaluation method and a system, which relate to the technical field of reliability evaluation and comprise the steps of obtaining wind speed information and space information of a target area, carrying out association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information; inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model; evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm, and obtaining a comparison result according to the predicted wind speed conversion result and the uncertainty result; and if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition.
Description
Technical Field
The invention relates to a reliability evaluation technology, in particular to a wind speed conversion reliability evaluation method and system.
Background
The wind power is closely related to the wind speed, and the wind power and the wind speed have obvious nonlinear relations. In conventional wind energy and wind power statistics, all wind energy calculations are based on average wind speed over a period of time. This conventional practice may introduce a non-negligible error into the wind energy statistics. One obvious adverse effect is that wind energy is quite different in the same period of time, which is calculated from the average wind speeds of different time scales such as seconds, minutes and hours, and cannot be kept consistent.
Disclosure of Invention
The embodiment of the invention provides a wind speed conversion reliability evaluation method and a wind speed conversion reliability evaluation system, which at least can solve part of problems in the prior art.
In a first aspect of an embodiment of the present invention,
provided is a wind speed conversion reliability evaluation method, including:
acquiring wind speed information and space information of a target area, performing association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information;
inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model;
evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm, wherein the uncertainty result comprises an uncertainty range and an uncertainty probability, and comparing the actual wind speed conversion result corresponding to the target area according to the predicted wind speed conversion result and the uncertainty result to obtain a comparison result;
and if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition.
In an alternative embodiment of the present invention,
acquiring wind speed information and space information of a target area, performing association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information comprises the following steps:
performing interpolation calculation on the spatial information corresponding to each time point, and regarding the spatial distance between each point and the adjacent point in the target area;
randomly selecting a target point for the corresponding wind speed information distributed to each piece of space information, determining wind speed difference information according to the target wind speed information corresponding to the target point and the space wind speed information corresponding to any space point in the target area, and taking the wind speed difference information as the input of a radial basis function;
setting a wind speed difference weight corresponding to the wind speed difference information, and determining a wind speed influence factor of the space information on the wind speed information by combining the radial basis function.
In an alternative embodiment of the present invention,
setting a wind speed difference weight corresponding to the wind speed difference information, and determining a wind speed influence factor of the space information on the wind speed information by combining the radial basis function comprises the following steps:
；
wherein,f(x)representing the wind speed influencing factor,nrepresenting the number of spatial points, phi () represents the radial basis function,x、x i respectively represent the target wind speed information and the firstiSpatial wind speed information corresponding to the individual spatial points,w i represent the firstiAnd (5) weighting the wind speed difference.
In an alternative embodiment of the present invention,
inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model comprises the following steps:
the wind speed conversion model is constructed based on a deep learning model, and comprises an input layer, a characteristic fusion layer, a hidden layer and an output layer;
the wind speed information, the space information and the wind speed influence factors enter the wind speed conversion model through an input layer, and a characteristic fusion layer of the wind speed conversion model fuses the wind speed information, the space information and the wind speed influence factors to obtain comprehensive characteristics;
and introducing a nonlinear factor into the integrated feature through an activation function of a hidden layer, mapping the integrated feature to a high-dimensional space, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model through an output layer of the wind speed conversion model.
In an alternative embodiment of the present invention,
if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition comprises the following steps:
iteratively optimizing a loss function of the wind speed conversion model, and adjusting model parameters of the wind speed conversion model until a comparison result of a predicted wind speed conversion result and an actual wind speed conversion result of the wind speed conversion model meets a preset threshold condition;
the loss function of the wind speed conversion model is shown in the following formula:
；
wherein,Lthe loss value is indicated as such,Nrepresenting the number of samples to be taken,y j 、respectively represent the firstjPredicted wind speed conversion results and the thjAnd the actual wind speed conversion result, lambda represents the regularization coefficient,Mrepresenting the number of parameters of the model,r k represent the firstkAnd model parameters.
In an alternative embodiment of the present invention,
evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm includes:
determining probability distribution corresponding to model parameters of the wind speed conversion model, sampling the probability distribution, generating a plurality of groups of random parameter values, and operating the wind speed conversion model according to the plurality of groups of random parameter values to obtain a plurality of groups of predicted wind speed conversion results;
and calculating at least one statistic in a mean value, a variance and a confidence interval of the wind speed conversion prediction based on the multiple groups of predicted wind speed conversion results, and determining a central trend and an uncertainty range of the wind speed conversion prediction results.
In an alternative embodiment of the present invention,
the method further comprises modifying wind speed information:
；
wherein,V corrected in order to correct the value of the wind speed,V measured in order to measure the value of the wind speed obtained,△hfor altitude differences, β and γ are correction coefficients.
In a second aspect of an embodiment of the present invention,
provided is a wind speed conversion reliability evaluation system including:
the first unit is used for acquiring wind speed information and space information of a target area, carrying out association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information;
the second unit is used for inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model and determining a predicted wind speed conversion result corresponding to the wind speed conversion model;
the third unit is used for evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm, wherein the uncertainty result comprises an uncertainty range and an uncertainty probability, and comparing with an actual wind speed conversion result corresponding to the target area to obtain a comparison result according to the predicted wind speed conversion result and the uncertainty result;
and if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition.
In a third aspect of an embodiment of the present invention,
there is provided an electronic device including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the instructions stored in the memory to perform the method described previously.
In a fourth aspect of an embodiment of the present invention,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method as described above.
The beneficial effects of the embodiments of the present invention may refer to the effects corresponding to technical features in the specific embodiments, and are not described herein.
Drawings
FIG. 1 is a flow chart of a wind speed conversion reliability evaluation method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a wind speed conversion reliability evaluation system according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, 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.
The technical scheme of the invention is described in detail below by specific examples. The following embodiments may be combined with each other, and some embodiments may not be repeated for the same or similar concepts or processes.
FIG. 1 is a flow chart of a wind speed conversion reliability evaluation method according to an embodiment of the invention, as shown in FIG. 1, the method includes:
s101, acquiring wind speed information and space information of a target area, performing association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information;
illustratively, wind speed information and associated spatial information data for a target area are acquired. The quality of the data is ensured, the missing value and the abnormal value are processed, and the consistency of the time and the space coordinates of the data is ensured.
And carrying out preliminary exploration and visual analysis on the wind speed and space information data to know the distribution and characteristics of the data. The relationships between the data can be viewed using tools such as scatter plots, box plots, correlation matrices, and the like.
And carrying out feature engineering on the spatial information, and extracting spatial features related to wind speed. This may include terrain, grade, vegetation coverage, elevation, and the like. Temporal characteristics, such as seasons, time periods, etc., are considered in order to better capture changes in the impact factors. And selecting a proper association analysis method, such as correlation analysis, regression analysis, decision tree regression, random forest and the like, for analyzing the relationship between the wind speed and the spatial characteristics. The appropriate method is selected based on the nature of the problem and the distribution of the data. And establishing a relation model between the wind speed and the space information by using the selected association analysis method. If a regression method is used, a regression model is built to estimate wind speed.
The relational model is evaluated, and various performance indicators such as Root Mean Square Error (RMSE), decision coefficients (R), etc. are used to evaluate the performance of the model. Cross-validation or the like is used to ensure generalization capability of the model.
Wherein the wind speed influencing factor may comprise a geographical location: wind speed is typically related to geographical location. Factors such as topography, altitude, longitude and latitude at different sites can have a significant effect on wind speed. Topography and topography: the flow and speed of wind can be affected by the topographical features of mountains, hills, plains, etc. Mountains may cause wind speeds to increase or decrease, creating gusts. Season and time: seasonal and temporal factors typically have an impact on wind speed. Seasonal variations, day-night temperature differences, monsoon, etc. can change the pattern of wind speeds.
In an alternative embodiment of the present invention,
acquiring wind speed information and space information of a target area, performing association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information comprises the following steps:
performing interpolation calculation on the spatial information corresponding to each time point, and regarding the spatial distance between each point and the adjacent point in the target area;
randomly selecting a target point for the corresponding wind speed information distributed to each piece of space information, determining wind speed difference information according to the target wind speed information corresponding to the target point and the space wind speed information corresponding to any space point in the target area, and taking the wind speed difference information as the input of a radial basis function;
setting a wind speed difference weight corresponding to the wind speed difference information, and determining a wind speed influence factor of the space information on the wind speed information by combining the radial basis function.
Illustratively, wind speed information and associated spatial information data for a target area are collected. And ensuring that the data has a corresponding relationship between time and space coordinates. And processing missing values and abnormal values in the data to ensure the quality of the data. For each point in time, wind speed information is calculated for each point in the target area using a spatial interpolation method (e.g., radial basis function interpolation). For points of unknown wind speed, interpolation calculation is performed by using information of known points, and a continuous spatial distribution diagram is generated. The spatial distance between each point and adjacent points within the target area is calculated, and Euclidean distance or other suitable distance measurement methods may be used. In performing interpolation computation on spatial information corresponding to each time point, it is generally required to calculate spatial distances between each point and adjacent points in a target area, and one common method for calculating the spatial distances is to use euclidean distances.
And randomly selecting a target point from the target area, wherein the target point corresponds to known target wind speed information, and calculating wind speed difference information between the known target point and any spatial point, wherein the wind speed difference information can be the difference between actual wind speed values. Each wind differential information is assigned a weight, which may be determined based on different factors, such as distance, characteristics of the spatial information, etc., which may be determined empirically, model training, or otherwise.
Taking the wind speed difference information as an input of a radial basis function, the radial basis function can calculate an interpolation value of the wind speed information according to the space distance and the weight. The resulting interpolated wind speed value and the actual observed wind speed information are combined, and the influence factor of each piece of spatial information on the wind speed information is calculated, which can be determined by comparing the difference between the interpolated wind speed and the actual wind speed.
Wherein a target point is randomly selected, which is known to have target wind speed information. Let the coordinates of the target point be (Tx, ty). Calculating wind speed difference information: for any spatial point (coordinates Sx, sy) in the target area, wind speed difference information between the spatial point and the target point is calculated. The wind speed difference information may be defined as the difference between the target wind speed information of the target point and the spatial wind speed information of the spatial point, i.e.: wind speed difference information = target wind speed information-spatial wind speed information;
and taking the wind speed difference information between each spatial point and the target point as the input of the radial basis function. These wind speed difference information will be used for interpolation calculations. Selecting a radial basis function: an appropriate radial basis function is selected, typically a gaussian radial basis function or a multi-radial basis function. The choice of radial basis function depends on the nature of the problem and the distribution of the data.
In an alternative embodiment of the present invention,
setting a wind speed difference weight corresponding to the wind speed difference information, and determining a wind speed influence factor of the space information on the wind speed information by combining the radial basis function comprises the following steps:
；
wherein,f(x)representing the wind speed influencing factor,nrepresenting the number of spatial points, phi () represents the radial basis function,x、x i respectively represent the target wind speed information and the firstiSpatial wind speed information corresponding to the individual spatial points,w i represent the firstiAnd (5) weighting the wind speed difference.
For each spatial point, the interpolation calculation will generate estimated wind speed information.
By means of the correlation analysis of the wind speed information and the spatial information, the method can determine which spatial information factors have significant influence on wind speed, which is helpful for understanding driving factors of wind speed change, and provides valuable information in the establishment of a wind speed prediction model or wind speed field analysis; the wind speed information can be expanded from the known position to other positions in the target area through spatial interpolation calculation, and a continuous wind speed distribution map is generated; by calculating the wind speed difference information between the randomly selected target point and other spatial points, the wind speed difference between different points can be quantified. The radial basis function is utilized to conduct interpolation calculation according to the wind speed difference information, the wind speed of each point in the target area is estimated, the radial basis function allows interpolation according to distance and weight distribution, and the spatial distribution characteristics of the wind speed are better captured.
S102, inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model;
in an alternative embodiment of the present invention,
inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model comprises the following steps:
the wind speed conversion model is constructed based on a deep learning model, and comprises an input layer, a characteristic fusion layer, a hidden layer and an output layer;
the wind speed information, the space information and the wind speed influence factors enter the wind speed conversion model through an input layer, and a characteristic fusion layer of the wind speed conversion model fuses the wind speed information, the space information and the wind speed influence factors to obtain comprehensive characteristics;
and introducing a nonlinear factor into the integrated feature through an activation function of a hidden layer, mapping the integrated feature to a high-dimensional space, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model through an output layer of the wind speed conversion model.
Illustratively, a dataset of wind speed information, spatial information, and wind speed influencing factors is obtained, ensuring that the data has corresponding temporal and spatial coordinate correspondence. A deep learning model is constructed, and the model comprises an input layer, a feature fusion layer, a hidden layer and an output layer. In the feature fusion layer, the wind speed information, the space information and the wind speed influence factors are subjected to feature fusion to generate comprehensive features, the feature fusion can be performed in various modes, such as connection (establishment) operation, weighted summation and the like, and a specific fusion method is required to be selected according to the problems and the data characteristics. The comprehensive features are mapped to a high-dimensional space through an activation function by introducing nonlinear factors through a hidden layer, the hidden layer can comprise a plurality of neurons and a plurality of hidden layers, and the specific structure needs to be selected according to the complexity of the problem.
The output layer determines the predicted wind speed conversion result of the model, and the configuration of the output layer depends on the nature of the task, and can be a regression problem (outputting continuous values) or a classification problem (outputting discrete categories). For regression problems, the output layer typically includes one neuron, outputting the predicted wind speed value, and for classification problems, the output layer may include a plurality of neurons, each neuron representing a class.
The model is trained using a data set, during which the model continuously adjusts parameters via a back-propagation algorithm to reduce the gap between predicted and actual values (loss function minimization).
In an alternative embodiment of the present invention,
if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition comprises the following steps:
iteratively optimizing a loss function of the wind speed conversion model, and adjusting model parameters of the wind speed conversion model until a comparison result of a predicted wind speed conversion result and an actual wind speed conversion result of the wind speed conversion model meets a preset threshold condition;
the loss function of the wind speed conversion model is shown in the following formula:
；
wherein,Lthe loss value is indicated as such,Nrepresenting the number of samples to be taken,y j 、respectively represent the firstjPredicted wind speed conversion results and the thjAnd the actual wind speed conversion result, lambda represents the regularization coefficient,Mrepresenting the number of parameters of the model,r k represent the firstkAnd model parameters.
Illustratively, using a given loss function formula, a loss function L is defined, which includes two parts: a first part: a Mean Square Error (MSE) term for measuring an error between the predicted wind speed conversion result and the actual wind speed conversion result; a second part: regularization term for controlling complexity of model parameters to prevent overfitting. The regularization coefficient λ may be adjusted as desired.
Starting training a model, and adjusting model parameters through iterative optimization of a loss function so as to reduce a loss value; gradients of the loss function to the model parameters are calculated using gradient descent or other optimization algorithms, and the parameters are then updated. The training is iterated until a preset stopping condition is met, typically until a preset number of training rounds or a loss value is reached that is sufficiently small, or until a preset threshold condition is met.
S103, evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm, wherein the uncertainty result comprises an uncertainty range and an uncertainty probability, and comparing the actual wind speed conversion result corresponding to the target area with the actual wind speed conversion result according to the predicted wind speed conversion result and the uncertainty result to obtain a comparison result;
and if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition.
In an alternative embodiment of the present invention,
evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm includes:
determining probability distribution corresponding to model parameters of the wind speed conversion model, sampling the probability distribution, generating a plurality of groups of random parameter values, and operating the wind speed conversion model according to the plurality of groups of random parameter values to obtain a plurality of groups of predicted wind speed conversion results;
and calculating at least one statistic in a mean value, a variance and a confidence interval of the wind speed conversion prediction based on the multiple groups of predicted wind speed conversion results, and determining a central trend and an uncertainty range of the wind speed conversion prediction results.
In an alternative embodiment of the present invention,
the method further comprises modifying wind speed information:
；
wherein,V corrected in order to correct the value of the wind speed,V measured in order to measure the value of the wind speed obtained,△hfor altitude differences, β and γ are correction coefficients.
Illustratively, a Monte Carlo simulation algorithm that will be used to estimate the uncertainty range and uncertainty probability of the predicted wind speed conversion result; for each time point and spatial point combination, a sample of the multiple wind speed conversion results is generated using a Monte Carlo simulation algorithm, which may be accomplished by introducing randomness to the model parameters, input data, or other relevant variables. Wind speed conversion results are calculated for each sample, thereby obtaining a set of a plurality of predicted wind speed conversion results. Based on the generated plurality of wind speed conversion results, uncertainty analysis is performed to estimate an uncertainty range and an uncertainty probability.
Taking the estimated predicted wind speed conversion result into account along with an uncertainty range and an uncertainty probability; the estimated result is compared with the actual wind speed conversion result corresponding to the target area, and the comparison result is calculated, for example, a difference value or other suitable index is calculated. Whether the comparison result meets a preset threshold condition is checked, and the threshold condition can comprise an error range, a confidence level or other performance indexes, and the threshold condition depends on the requirement of the problem. And if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model according to the current uncertainty analysis result and the comparison result. Adjusting parameters may include retraining the model, adjusting parameters of the uncertainty analysis algorithm, increasing the number of samples, and so forth.
The method can estimate the uncertainty range and the uncertainty probability of the predicted wind speed conversion result by using a Monte Carlo simulation algorithm. This helps to determine the confidence level of the wind speed predictions. By generating sets of random parameter values, uncertainty in model parameters is taken into account. The method can more comprehensively reflect the influence of the change of the model parameters on the prediction result. Based on the multiple groups of predicted wind speed conversion results, statistics such as mean value, variance, confidence interval and the like of wind speed conversion predictions are calculated to determine a central trend and an uncertainty range of the wind speed conversion prediction results. By considering the uncertainty, the method improves the reliability and the robustness of the wind speed prediction model. This is important for applications requiring highly reliable predictions, such as wind energy yield estimation, weather prediction, etc.
FIG. 2 is a schematic structural diagram of a wind speed conversion reliability evaluation system according to an embodiment of the present invention, as shown in FIG. 2, the system includes:
the first unit is used for acquiring wind speed information and space information of a target area, carrying out association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information;
the second unit is used for inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model and determining a predicted wind speed conversion result corresponding to the wind speed conversion model;
the third unit is used for evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm, wherein the uncertainty result comprises an uncertainty range and an uncertainty probability, and comparing with an actual wind speed conversion result corresponding to the target area to obtain a comparison result according to the predicted wind speed conversion result and the uncertainty result;
and if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition.
The present invention may be a method, apparatus, system, and/or computer program product. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for performing various aspects of the present invention.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (10)
1. A wind speed transition reliability assessment method, comprising:
acquiring wind speed information and space information of a target area, performing association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information;
inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model;
evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm, wherein the uncertainty result comprises an uncertainty range and an uncertainty probability, and comparing the actual wind speed conversion result corresponding to the target area according to the predicted wind speed conversion result and the uncertainty result to obtain a comparison result;
and if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition.
2. The method of claim 1, wherein obtaining wind speed information and spatial information of a target area, performing correlation analysis on the wind speed information and the spatial information, and determining a wind speed influence factor of the spatial information on the wind speed information comprises:
performing interpolation calculation on the spatial information corresponding to each time point, and regarding the spatial distance between each point and the adjacent point in the target area;
randomly selecting a target point for the corresponding wind speed information distributed to each piece of space information, determining wind speed difference information according to the target wind speed information corresponding to the target point and the space wind speed information corresponding to any space point in the target area, and taking the wind speed difference information as the input of a radial basis function;
setting a wind speed difference weight corresponding to the wind speed difference information, and determining a wind speed influence factor of the space information on the wind speed information by combining the radial basis function.
3. The method according to claim 2, wherein setting a wind speed difference weight corresponding to the wind speed difference information, and determining a wind speed influence factor of the spatial information on the wind speed information in combination with the radial basis function comprises:
；
wherein,f(x)representing the wind speed influencing factor,nrepresenting the number of spatial points, phi () represents the radial basis function,x、x i respectively represent the target wind speed information and the firstiSpatial wind speed information corresponding to the individual spatial points,w i represent the firstiAnd (5) weighting the wind speed difference.
4. The method of claim 1, wherein inputting the wind speed information, the spatial information, and the wind speed influencing factor into a pre-constructed wind speed conversion model, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model comprises:
the wind speed conversion model is constructed based on a deep learning model, and comprises an input layer, a characteristic fusion layer, a hidden layer and an output layer;
the wind speed information, the space information and the wind speed influence factors enter the wind speed conversion model through an input layer, and a characteristic fusion layer of the wind speed conversion model fuses the wind speed information, the space information and the wind speed influence factors to obtain comprehensive characteristics;
and introducing a nonlinear factor into the integrated feature through an activation function of a hidden layer, mapping the integrated feature to a high-dimensional space, and determining a predicted wind speed conversion result corresponding to the wind speed conversion model through an output layer of the wind speed conversion model.
5. The method of claim 4, wherein readjusting model parameters of the wind speed conversion model if the comparison result does not satisfy a preset threshold condition until a comparison result of a predicted wind speed conversion result and an actual wind speed conversion result of the wind speed conversion model satisfies a preset threshold condition comprises:
iteratively optimizing a loss function of the wind speed conversion model, and adjusting model parameters of the wind speed conversion model until a comparison result of a predicted wind speed conversion result and an actual wind speed conversion result of the wind speed conversion model meets a preset threshold condition;
the loss function of the wind speed conversion model is shown in the following formula:
；
wherein,Lthe loss value is indicated as such,Nrepresenting the number of samples to be taken,y j 、respectively represent the firstjPredicted wind speed conversion results and the thjAnd the actual wind speed conversion result, lambda represents the regularization coefficient,Mrepresenting the number of parameters of the model,r k represent the firstkAnd model parameters.
6. The method of claim 1, wherein evaluating an uncertainty result of the predicted wind speed conversion result based on a monte carlo simulation algorithm comprises:
determining probability distribution corresponding to model parameters of the wind speed conversion model, sampling the probability distribution, generating a plurality of groups of random parameter values, and operating the wind speed conversion model according to the plurality of groups of random parameter values to obtain a plurality of groups of predicted wind speed conversion results;
and calculating at least one statistic in a mean value, a variance and a confidence interval of the wind speed conversion prediction based on the multiple groups of predicted wind speed conversion results, and determining a central trend and an uncertainty range of the wind speed conversion prediction results.
7. The method of claim 1, further comprising modifying wind speed information:
；
wherein,for the corrected wind speed value +.>For measuring the wind speed value>For altitude difference, < >>And->Is a correction coefficient.
8. A wind speed transition reliability assessment system for implementing a wind speed transition reliability assessment method according to any one of the preceding claims 1-7, comprising:
the first unit is used for acquiring wind speed information and space information of a target area, carrying out association analysis on the wind speed information and the space information, and determining a wind speed influence factor of the space information on the wind speed information;
the second unit is used for inputting the wind speed information, the space information and the wind speed influence factors into a pre-constructed wind speed conversion model and determining a predicted wind speed conversion result corresponding to the wind speed conversion model;
the third unit is used for evaluating an uncertainty result of the predicted wind speed conversion result based on a Monte Carlo simulation algorithm, wherein the uncertainty result comprises an uncertainty range and an uncertainty probability, and comparing with an actual wind speed conversion result corresponding to the target area to obtain a comparison result according to the predicted wind speed conversion result and the uncertainty result;
and if the comparison result does not meet the preset threshold condition, readjusting the model parameters of the wind speed conversion model until the comparison result of the predicted wind speed conversion result and the actual wind speed conversion result of the wind speed conversion model meets the preset threshold condition.
9. An electronic device, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the instructions stored in the memory to perform the method of any of claims 1 to 7.
10. A computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410200562.4A CN117787109B (en) | 2024-02-23 | 2024-02-23 | Wind speed conversion reliability assessment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410200562.4A CN117787109B (en) | 2024-02-23 | 2024-02-23 | Wind speed conversion reliability assessment method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117787109A true CN117787109A (en) | 2024-03-29 |
| CN117787109B CN117787109B (en) | 2024-05-03 |
Family
ID=90394750
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410200562.4A Active CN117787109B (en) | 2024-02-23 | 2024-02-23 | Wind speed conversion reliability assessment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117787109B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130110558A1 (en) * | 2011-10-26 | 2013-05-02 | Sean Maher | Method and system for estimating economic losses from wind storms |
| KR20140018497A (en) * | 2012-08-01 | 2014-02-13 | 한국전력공사 | Prediction method of short-term wind speed and wind power and power supply line voltage prediction method therefore |
| CN110009037A (en) * | 2019-04-03 | 2019-07-12 | 中南大学 | A kind of engineering wind speed Forecasting Approach for Short-term and system based on physical message coupling |
| US20210056246A1 (en) * | 2019-08-21 | 2021-02-25 | Northwestern Polytechnical University | Method for evaluating reliability of a sealing structure in a multi-failure mode based on an adaboost algorithm |
| CN114578087A (en) * | 2022-01-24 | 2022-06-03 | 华北电力大学（保定） | Wind speed uncertainty measurement method based on non-dominated sorting and random simulation algorithm |
| JP7114162B1 (en) * | 2021-12-09 | 2022-08-08 | 株式会社日立パワーソリューションズ | Wind condition estimation device and wind condition estimation method |
| CN115494521A (en) * | 2022-09-22 | 2022-12-20 | 中国民航大学 | Airport runway low-altitude wind shear early warning method based on laser radar |
-
2024
- 2024-02-23 CN CN202410200562.4A patent/CN117787109B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130110558A1 (en) * | 2011-10-26 | 2013-05-02 | Sean Maher | Method and system for estimating economic losses from wind storms |
| KR20140018497A (en) * | 2012-08-01 | 2014-02-13 | 한국전력공사 | Prediction method of short-term wind speed and wind power and power supply line voltage prediction method therefore |
| CN110009037A (en) * | 2019-04-03 | 2019-07-12 | 中南大学 | A kind of engineering wind speed Forecasting Approach for Short-term and system based on physical message coupling |
| US20210056246A1 (en) * | 2019-08-21 | 2021-02-25 | Northwestern Polytechnical University | Method for evaluating reliability of a sealing structure in a multi-failure mode based on an adaboost algorithm |
| JP7114162B1 (en) * | 2021-12-09 | 2022-08-08 | 株式会社日立パワーソリューションズ | Wind condition estimation device and wind condition estimation method |
| CN114578087A (en) * | 2022-01-24 | 2022-06-03 | 华北电力大学（保定） | Wind speed uncertainty measurement method based on non-dominated sorting and random simulation algorithm |
| CN115494521A (en) * | 2022-09-22 | 2022-12-20 | 中国民航大学 | Airport runway low-altitude wind shear early warning method based on laser radar |
Non-Patent Citations (3)
| Title |
|---|
| 李春祥;刘晨哲;: "基于径向基神经网络的谐波叠加法", 振动与冲击, no. 01, 25 January 2010 (2010-01-25) * |
| 潘雄;周明;孔晓民;吴;刘文霞;廖平;: "风速相关性对最优潮流的影响", 电力系统自动化, no. 06, 25 March 2013 (2013-03-25) * |
| 田启东;刘雪飞;林欣慰;李俊;何蓝图;郭创新;: "考虑风速相关性和出力不确定性的风电场运行风险评估", 机电工程, no. 03, 20 March 2016 (2016-03-20) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117787109B (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639787A (en) | Spatio-temporal data prediction method based on graph convolution network | |
| JP4719893B2 (en) | Control device, control method, and program thereof | |
| CN110689179A (en) | Water bloom prediction method based on space-time sequence mixed model | |
| CN113496104A (en) | Rainfall forecast correction method and system based on deep learning | |
| CN112200362B (en) | Landslide prediction method, landslide prediction device, landslide prediction equipment and storage medium | |
| CN110456026B (en) | Soil moisture content monitoring method and device | |
| CN113723757A (en) | Decision generation model training method, decision generation method and device | |
| Singh et al. | Unsupervised change detection from remote sensing images using hybrid genetic FCM | |
| CN117114913A (en) | Intelligent agricultural data acquisition system based on big data | |
| CN117423003B (en) | Winter wheat seedling condition grading remote sensing monitoring method in overwintering period | |
| CN117787109B (en) | Wind speed conversion reliability assessment method and system | |
| CN108510097B (en) | Power transmission line corridor snow disaster monitoring method and system based on satellite remote sensing | |
| CN117522117A (en) | Ecological risk assessment method and early warning system based on ecological protection red line demarcation | |
| CN117271979A (en) | Deep learning-based equatorial Indian ocean surface ocean current velocity prediction method | |
| CN113343783A (en) | Intelligent crop identification and growth prediction method and system | |
| CN117057490A (en) | Prediction method and system for wet stress heat wave-flood composite disaster and electronic equipment | |
| CN113221447A (en) | Soil humidity prediction method for optimizing BP neural network based on improved genetic algorithm | |
| CN117131654A (en) | Target observation method based on nonlinear optimal disturbance of pre-analysis initial guess condition | |
| Ranjan et al. | ANN and GRNN-based coupled model for flood inundation mapping of the Punpun River Basin | |
| Masngut et al. | Comparison of daily rainfall forecasting using multilayer perceptron neural network model | |
| Bhoyar et al. | Comparative Analysis of Peak Current Prediction based on Random Forest and MLP Neural Network Algorithms | |
| CN117079416B (en) | Multi-person 5D radar falling detection method and system based on artificial intelligence algorithm | |
| CN117909719B (en) | Terminal internet of things equipment monitoring and early warning method and system based on autonomous learning algorithm | |
| CN117935081B (en) | Cultivated land change monitoring method and system based on remote sensing satellite data | |
| CN111523258B (en) | Microseism effective signal first arrival pickup method and system based on MS-Net network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |