A highland complex terrain power engineering wind speed distribution fitting and wind area mapping method

Abstract

The application belongs to the technical field of electric power engineering design, and discloses a highland complex terrain electric power engineering wind speed probability distribution fitting and wind area mapping method, which comprises the following steps: carrying out time period conversion, ground height correction and correlation regression correction on the maximum wind speed sequence collected by multiple stations; adopting four methods to carry out wind speed sequence homogeneity test and mutation point identification, and determining the cause of non-homogeneity; calculating the basic wind speed of electric power engineering occurring once every 30 years, 50 years and 100 years based on the extreme value type I distribution; taking the Weibull distribution as the main distribution to approximate the actual wind speed probability distribution, calculating the residual to obtain the falling form probability distribution; taking the effective zero-crossing point as the boundary to segment the residual, and adopting the cloud model to accurately fit the segmented residual; superimposing the Weibull main distribution and the segmented cloud model residual fitting to obtain a high-precision wind speed probability distribution fitting function; taking the station longitude, latitude, altitude and different return period basic wind speed as the spatial interpolation sample points, adopting the inverse distance weighted interpolation and combining the wind area grading standard to draw the highland wind speed distribution map and the wind pressure distribution map, so that the highland terrain and the wind speed characteristics are fully adapted, and accurate, reliable and directly usable wind speed parameters and wind area results are provided for electric power engineering design.

Description

Technical Field

[0001] This invention belongs to the field of power engineering design technology, specifically relating to a method for fitting wind speed distribution and mapping wind zones in power engineering projects in complex plateau terrain. Background Technology

[0002] Wind energy and wind speed are core meteorological parameters in power engineering design. The accuracy of wind speed probability distribution, the return period of basic wind speed, and the reliability of wind zone maps directly determine the safety and economy of wind resistance design of transmission lines, structural load of substations, and power generation prediction of wind farms.

[0003] Traditional wind speed probability distributions often employ single Weibull, Rayleigh, or Gamma distributions, which have significant drawbacks in the complex terrain of plateaus: First, plateau wind speeds exhibit asymmetry, fat tails, multiple peaks, and strong fluctuations, resulting in large fitting errors for single probability distributions and failing to meet the high-precision requirements of power engineering. Second, plateau meteorological stations have a long history, with multiple relocations, instrument replacements, and changes in the observation environment, leading to significant non-uniformity in wind speed sequences. Direct calculations without uniformity testing result in deviations in basic wind speeds. Third, traditional spatial interpolation does not consider the influence of elevation and topography, resulting in large interpolation errors in plateau boundary areas. Fourth, existing wind zone maps are mostly general meteorological wind zone maps, which are disconnected from the actual design values ​​used in power engineering and cannot directly serve the design of power facilities such as transmission lines and substations.

[0004] Existing patents disclose methods for fitting the probability distribution of wind speed in wind farms, but they do not provide integrated design for power engineering specifications, plateau uniformity testing, multi-source data correction, wind zone mapping, and engineering implementation, and therefore cannot be directly applied to the survey and design of plateau power engineering.

[0005] Research based on measured data from 50 national meteorological stations, 53 regional automatic weather stations, and 12 wind energy observation towers across Qinghai Province shows that altitude in plateau regions is significantly correlated with basic wind speed, and abrupt changes are common in wind speed sequences. Traditional methods are insufficient to meet the safety and economic design requirements of power engineering. Therefore, developing a wind speed probability fitting and wind zone mapping method that is adaptable to plateau regions, conforms to power specifications, is standardized throughout the entire process, and can be directly applied in engineering has significant engineering value. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain. This method standardizes the entire process, including multi-source data preprocessing, uniformity testing, basic wind speed calculation for power engineering, high-precision probability distribution fitting, and drawing wind zone / wind pressure maps for power engineering projects. It strictly follows power industry standards, fully adapts to plateau terrain and wind speed characteristics, and provides accurate, reliable, and directly usable wind speed parameters and wind zone results for power engineering design.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain includes the following steps: Step 1: Obtain the original wind speed time series of the plateau collected by various types of meteorological observation equipment, and perform multi-dimensional standardization preprocessing on the original wind speed to obtain the standardized wind speed sequence; Step 2: Use multiple types of time series mutation detection tools to jointly conduct homogeneity tests on the standardized wind speed series, identify real mutations caused by changes in the observation environment, and remove distorted observation data before the mutation to obtain a homogeneous wind speed sample set. Step 3: Frequency analysis calculations are performed on the homogeneous wind speed sample set using the extreme value type I distribution to obtain the 30-year, 50-year, and 100-year return period benchmark wind speeds for power engineering. The basic wind speed calculation strictly follows the power engineering specifications, adopts the extreme value type I distribution, and the statistical sample is the annual maximum 10-minute average wind speed sequence after preprocessing and homogenization, which can be directly used for wind speed selection in power engineering design such as transmission lines, substations, and wind farms. Step 4: Fit the measured wind speed probability distribution to the whole system using the Weibull distribution, calculate the residual between the measured wind speed probability distribution and the Weibull theoretical distribution, and obtain the landing pattern probability distribution. Step 5: Divide the probability distribution of landing patterns into segments using the effective inflection points of the residual curves as the segment boundaries, extract samples from each segment according to the rules, and use a cloud model to independently fit the residuals of each segment. Step 6: Superimpose the Weibull principal distribution function with the residual fitting functions of each segmented cloud model to obtain a composite high-precision wind speed probability fitting model; Step 7: Using the site's geographical coordinates, elevation, and multiple current benchmark wind speeds as spatial interpolation sample points, spatial interpolation is carried out using the inverse distance weighted interpolation method. Based on the power industry's wind zoning classification standards, wind zone distribution maps and wind pressure distribution maps are drawn for engineering design.

[0008] Furthermore, the standardization preprocessing described in step 1 includes converting the 2-minute average wind speed to the 10-minute average wind speed, correcting the observation height to 10m above ground, and performing correlation tests and regression corrections on multi-source data to obtain a standardized 10-minute average maximum wind speed sequence.

[0009] Furthermore, standardized preprocessing includes: (1) The formula for converting wind speed over a given time period is: V 10min =0.85V 2min +5.21, where V 10 min V represents the 10-minute average maximum wind speed. 2min The average maximum wind speed is measured over a 2-minute timer. (2) The formula for correcting altitude above ground level is: , where Z is the height above ground, taking a value of 10m, α is 0.15, Z1 is the actual observation height, and V1 is the measured wind speed; (3) For the data of regional automatic weather stations and wind towers with relatively short sequences, a regression equation is used for correlation correction to ensure the homogeneity of multi-source data.

[0010] Further, the time series mutation test tool described in step 2 is selected from Mann-Kendall, Pettitt, Buishand U test, and SNHT tools.

[0011] Further, the rule for judging the authenticity of mutation points in step 2 is: (1) If any one of the three test results of Pettitt, Buishand U test and SNHT matches the station relocation time (with an error of ±5 years), it is determined as a real mutation point; (2) When there is no clear relocation information, the test results of at least three methods must be consistent to determine a real mutation point; (3) The significance level is 0.05, and the main causes of mutation points include station relocation, instrument replacement, and observation environment change.

[0012] Further, the Weibull distribution probability density function in step 4 is: ; In the formula: c is the size parameter, k is the shape parameter, v is the wind speed; the parameters k and c are estimated by the maximum likelihood method.

[0013] Further, the effective inflection point in step 5 is the zero-crossing point where the residual changes between positive and negative, and the false zero points generated by small shocks are eliminated; after segmentation, samples are randomly extracted to construct a cloud model, and the digital features of the cloud model include the sample expectation E x , entropy E n , hyper-entropy E he , a fixed number N of samples are extracted for each segment, when the sample size M≥N, non-repetitive random sampling is adopted, and when M<N, repeatable sampling is used to make up N samples.

[0014] Further, the cloud model formula is: (sample mean); ; ; In the formula: S2 is the sample variance, S4 is the fourth-order central moment of the sample; The expected curve of the i-th segment cloud model: , peak coefficient .

[0015] Furthermore, the composite high-precision wind speed probability fitting model described in step 6 is as follows: ; In the formula: The number of segments, It follows a Weibull distribution. The residuals for fitting the cloud model in segment i are the residuals.

[0016] Furthermore, step 7 employs the inverse distance weighted interpolation method coupled with altitude, taking into account both the sparse characteristics of plateau stations and interpolation accuracy; wind zone classification and mapping color codes follow DL / T2237-2021 "Technical Guidelines for Drawing Wind Zone Distribution Maps of Power Grids" (DL / T2237-2021), with a total of 15 levels, the lowest level being 22 m / s; wind pressure conversion is based on GB50009-2012 "Code for Design of Building Structures" (GB50009-2012) and the basic wind speed spatial interpolation, and is directly used for power engineering structural load calculation.

[0017] Furthermore, this method is specifically applicable to plateau regions with altitudes ≥2000m, complex terrain, uneven distribution of meteorological stations, and significant non-uniformity of wind speed sequences. It is particularly suitable for fitting the probability distribution of wind speed for power engineering projects across Qinghai Province, calculating basic wind speeds, and drawing wind zone maps and wind pressure maps.

[0018] Compared with the prior art, the present invention has the following advantages: 1. Power Engineering Dedicated Adaptation: Strictly adheres to power industry standards, and basic wind speed, wind zone maps, and wind pressure values ​​can be directly used for the design of transmission lines, substations, and wind farms; 2. Precise adaptation to plateau terrain: solves problems such as high altitude, large terrain undulation, sparse stations, and asymmetric wind tail. 3. End-to-end uniformity assurance: Multi-source data correction and four methods of uniformity verification significantly improve the reliability of wind speed sequences; 4. Significantly improved fitting accuracy: By stacking the Weibull and cloud models in segments, the fitting accuracy is improved by more than 40% compared to a single distribution; 5. Strong engineering implementation capability: It integrates probability fitting, basic wind speed calculation, and wind zone / wind pressure map, solving the problem of the disconnect between traditional wind zone map and power engineering values. Attached Figure Description

[0019] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a uniformity test diagram of Xining Station in an embodiment of the present invention; Figure 3 This is a uniformity test diagram of Gonghe Station in an embodiment of the present invention; Figure 4This is a uniformity test diagram of Qumalai Station in an embodiment of the present invention; Figure 5 This is a uniformity test diagram of the small stove fire station in an embodiment of the present invention; Figure 6-14 This is a wind speed distribution diagram according to an embodiment of the present invention; Figure 15-17 This is a wind pressure distribution diagram according to an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0021] Example like Figure 1 As shown, this application proposes a method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain, including the following steps: Step 1: Multi-source data preprocessing The raw wind speed time series of the plateau region, collected by various types of meteorological observation equipment, was obtained. Multi-dimensional standardization preprocessing was performed on the raw wind speeds to obtain a standardized wind speed sequence. Based on measured data from 50 national meteorological stations, 53 regional automatic weather stations, and 12 wind energy observation towers across Qinghai Province, the specific implementation method is as follows: Data standardization and preprocessing of raw data includes: (1) Annual maximum wind speed data from 50 national meteorological stations; (2) Short-sequence wind speed data from 53 regional automatic weather stations; (3) Measured data from 12 wind energy resource observation towers.

[0022] Preprocessing content: (1) Time period conversion: The 2-minute average wind speed is uniformly converted to a 10-minute average wind speed. The conversion formula is: V 10min =0.85V 2min +5.21 (2) Height correction: Correct different observation heights to a uniform height of 10m; The formula for altitude correction is:

[0023] (3) Regression correction: The correlation regression equation is used to correct the short sequence data to ensure the uniformity of multi-source data.

[0024] Step 2, Uniformity test The Mann-Kendall, Pettitt, Buishand U test, and SNHT time-series abrupt change test tools were used to conduct homogeneity tests on the annual maximum wind speed sequences of 50 meteorological stations across the province. This identified genuine abrupt changes caused by alterations in the observation environment and removed distorted observation data prior to the abrupt changes, resulting in a homogeneous wind speed sample set. This implementation plan includes tests conducted using four methods on representative stations in Xining, Gonghe, Qumalai, and Xiaozhaohuo in Qinghai Province (results detailed in [link to documentation]). Figure 2-5 ).

[0025] Test results: Mutation points were detected at all 50 stations in the province. 94% of the stations had consistent results from at least two methods, and 12% of the stations had consistent results from all four methods.

[0026] Step 3: Basic Wind Speed ​​Calculation for Power Engineering (1) The extreme value type I distribution was used to perform frequency analysis on the homogeneous wind speed sample set to obtain the reference wind speeds for power engineering once in 30, 50 and 100 years. The basic wind speed calculation strictly followed the power engineering specifications. The statistical sample was the annual maximum 10-minute average wind speed sequence after preprocessing and homogenization, which was directly used for wind speed values ​​in power engineering design such as transmission lines, substations and wind farms.

[0027] In this embodiment, extreme value type I distribution was used to calculate the basic wind speeds for each station in Qinghai Province once every 30 / 50 / 100 years (see Table 1).

[0028] Table 1. Calculation results of basic wind speeds at various meteorological stations in Qinghai Province with different return periods.

[0029] (2) Correlation between station elevation and basic wind speed at each return period Correlation analysis of station elevation and basic wind speed at each return period of the corresponding station showed correlation coefficients of 0.468 (30-year return), 0.459 (30-year return), and 0.447 (100-year return), respectively, which were significant. See Table 2 for details.

[0030] Table 2 Correlation between station elevation and maximum wind speed

[0031] Typical power engineering design values: 1) Lenghu: 30.68 m / s in 30 years, 32.19 m / s in 50 years, and 34.22 m / s in 100 years.

[0032] 2) Tuotuo River: 30 years 32.34 m / s, 50 years 33.97 m / s, 100 years 36.16 m / s.

[0033] 3) Xining: 16.40 m / s in 30 years, 17.44 m / s in 50 years, and 18.84 m / s in 100 years.

[0034] 4) Mangya: 28.33 m / s in 30 years, 29.73 m / s in 50 years, and 31.62 m / s in 100 years.

[0035] Steps 4-6: Fitting the wind speed probability distribution Step 4: Fit the measured wind speed probability distribution to the whole system using the Weibull distribution, calculate the residual between the measured wind speed probability distribution and the Weibull theoretical distribution, and obtain the landing pattern probability distribution.

[0036] Step 5: Divide the probability distribution of landing patterns into segments using the effective inflection points of the residual curves as the segment boundaries, extract samples from each segment according to the rules, and use a cloud model to independently fit the residuals of each segment. Step 6: Superimpose the Weibull principal distribution function with the residual fitting functions of each segmented cloud model to obtain a composite high-precision wind speed probability fitting model.

[0037] 1) Weibull distribution parameters: k and c are solved using the maximum likelihood method; 2) Residual calculation: Residual = Actual probability density - Weibull probability density; 3) Segmentation: Divide the residual into several intervals, using the effective zero crossing point as the boundary; 4) Cloud Model: Calculate Ex, En, and Ehe for each segment to construct the expected curve; 5) Superposition: f_FM (v)=f_w (v)+∑f_Ci (v).

[0038] Step 7: Wind Zone and Wind Suppression Diagram for Power Engineering Using the site's geographical coordinates, altitude, and multiple current benchmark wind speeds as spatial interpolation sample points, spatial interpolation is performed using the inverse distance weighted interpolation method. Based on the power industry's wind zoning classification standards, wind zone distribution maps and wind pressure distribution maps are drawn for engineering design. In this step: 1) Input: Station latitude and longitude, altitude, and basic wind speed for a 30 / 50 / 100-year return period; 2) Interpolation method: Inverse Distance Weighted Interpolation (IDW); 3) Classification Standard: According to the "Technical Guidelines for Drawing Wind Zone Distribution Maps of Power Grids", the 15 wind zones are classified as follows: 22m / s, 23.5m / s, 25m / s, 27m / s, 29m / s, 31m / s, 33m / s, 35m / s, 37m / s, 39m / s, 41m / s, 43m / s, 45m / s, 50m / s, and >50m / s. 4) Color codes: Follow the RGB standard in Table 6.1; 5) Mapping results: 30-year, 50-year, and 100-year return period wind zone maps (see...) Figures 6-14 Wind pressure diagram (see) Figures 15-17 ).

[0039] Actual calculation example (taking the Lenghu 750kV transmission line project in Qinghai as an example) 1. Data Preprocessing (1) Original timed 2 min maximum wind speed: V2=36.0m / s, converted to 10 min average wind speed: V 10 =0.85×36.0+5.21 = 35.81m / s (2) The observation height was 8m, corrected to 10m. Height: Z 10 =

[0040] 2. Uniformity test The mutation point at Lenghu Station was in 1994 (a significant decrease), which coincided with the instrument update time and was determined to be a true mutation point. Unreliable data before the mutation were removed.

[0041] 3. Basic wind speed calculation for power engineering (Extreme value type I) Once every 30 years: V 30 =30.68m / s Once every 50 years: V 50 =32.19m / s Once in 100 years: V 100 =34.22m / s It is directly used as the design wind speed for 750kV transmission line towers.

[0042] 4. Wind speed probability distribution fitting (1) Weibull distribution: c=28.5, k=2.1; (2) Residual segmentation: 3 segments; (3) Fitting the residuals of each segment of the cloud model; (4) The final fitting RMSE was 0.0142, and the single Weibull RMSE was 0.0303, which improved the accuracy by 53%.

[0043] 5. Wind zone level determination The basic wind speed at Lenghu Station during a 100-year return period is 34.22 m / s, with a wind zone level of 35 m / s. The corresponding color codes are R=153, G=204, and B=255, which are directly used for wind zone division and wind resistance design in power engineering projects.

Claims

1. A method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain, characterized in that, Includes the following steps: Step 1: Obtain the original wind speed time series of the plateau collected by various types of meteorological observation equipment, and perform multi-dimensional standardization preprocessing on the original wind speed to obtain the standardized wind speed sequence; Step 2: Use multiple types of time series mutation detection tools to jointly conduct homogeneity tests on the standardized wind speed series, identify real mutations caused by changes in the observation environment, and remove distorted observation data before the mutation to obtain a homogeneous wind speed sample set. Step 3: Use the extreme value type I distribution to perform frequency analysis on the homogeneous wind speed sample set to obtain the 30-year, 50-year, and 100-year return period benchmark wind speeds for power engineering. Step 4: Fit the measured wind speed probability distribution to the whole system using the Weibull distribution, calculate the residual between the measured wind speed probability distribution and the Weibull theoretical distribution, and obtain the landing pattern probability distribution. Step 5: Divide the probability distribution of landing patterns into segments using the effective inflection points of the residual curves as the segment boundaries, extract samples from each segment according to the rules, and use a cloud model to independently fit the residuals of each segment. Step 6: Superimpose the Weibull principal distribution function with the residual fitting functions of each segmented cloud model to obtain a composite high-precision wind speed probability fitting model; Step 7: Using the site's geographical coordinates, elevation, and multiple current benchmark wind speeds as spatial interpolation sample points, spatial interpolation is carried out using the inverse distance weighted interpolation method. Based on the power industry's wind zoning classification standards, wind zone distribution maps and wind pressure distribution maps are drawn for engineering design.

2. The method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain according to claim 1, characterized in that: The standardization preprocessing described in step 1 includes converting 2-minute average wind speed to 10-minute average wind speed, correcting for observation height to 10m above ground, and performing correlation tests and regression corrections on multi-source data to obtain a standardized 10-minute average maximum wind speed sequence.

3. The method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain according to claim 2, characterized in that: The formula for converting wind speed over a period of time is: V 10min =0.85V 2min +5.21, where V 10 min V represents the 10-minute average maximum wind speed. 2min The average maximum wind speed is measured over a 2-minute timer. The formula for altitude correction is: Where Z is the height above the ground, taken as 10m, α is taken as 0.15, Z1 is the actual observed height, and V1 is the measured wind speed.

4. The method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain according to claim 1, characterized in that: The time-series mutation detection tools mentioned in step 2 include Mann-Kendall, Petittt, Buishand U test, and SNHT.

5. The method for wind speed statistics and wind zone mapping applicable to power engineering in high-altitude areas according to claim 1, characterized in that: In step 4, the Weibull distribution probability density function is: ； In the formula: c is the size parameter, k is the shape parameter, and v is the wind speed; parameters k and c are estimated using the maximum likelihood method.

6. The method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain according to claim 1, characterized in that: In step 5, the effective inflection point is the zero-crossing point of the residual positive-to-negative transition, eliminating spurious zeros caused by small oscillations; after segmentation, samples are randomly selected to construct the cloud model, and the digital features of the cloud model include the sample expectation E. x Entropy E n Superentropy E he .

7. The method for wind speed statistics and wind zone mapping applicable to power engineering in high-altitude areas according to claim 5, characterized in that: The cloud model formula is: (Sample mean); ； ； In the formula: S2 is the sample variance, and S4 is the fourth central moment of the sample; The expected curve of the i-th segment of the cloud model: Peak coefficient .

8. The method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain according to claim 1, characterized in that, The composite high-precision wind speed probability fitting model described in step 6 is as follows: ； In the formula: The number of segments, It follows a Weibull distribution. The residuals for fitting the cloud model in segment i are the residuals.

9. The method for fitting wind speed distribution and mapping wind zones in power engineering projects with complex plateau terrain according to claim 1, characterized in that: Step 7 adopts the inverse distance weighted interpolation method coupled with altitude; wind zone classification and mapping color code follow DL / T2237-2021 "Technical Guidelines for Drawing Power Grid Wind Zone Distribution Map", and wind pressure conversion is based on GB50009-2012 "Code for Design of Building Structures".

Images